Bispecific antibody construct that bonds DLL3 and CD3

Abstract

To provide an antibody construct for preventing and treating small cell lung cancer (SCLC) being progressive lung cancer.SOLUTION: A bispecific antibody construct includes: a first binding domain that bonds human DLL3 which is SCLC specific tumor antigen and is on a target cell surface; and a second binding domain that bonds human CD3 which is on a T-cell surface. Furthermore, a polynucleotide which codes the antibody construct, a vector including a polynucleotide, and a host cell transformed or transfected by the polynucleotide or the vector are provided. Furthermore, a method for producing the antibody construct, a medical usage of the antibody construct, and a kit including the antibody construct are provided.SELECTED DRAWING: None

Description

Technical Field

[0001] The present invention relates to a bispecific antibody construct comprising a first binding domain that binds to human DLL3 on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell. Further, the present invention provides a polynucleotide encoding the antibody construct, a vector comprising the polynucleotide, and a host cell transformed or transfected with the polynucleotide or the vector. Further, the present invention provides a method for producing the antibody construct of the present invention, a medical use of the antibody construct, and a kit comprising the antibody construct.

Background Art

[0002] Small cell lung cancer (SCLC) is an aggressive lung cancer with a poor prognosis and limited treatment options, accounting for approximately 15% of all newly diagnosed lung cancers, corresponding to approximately 25,000 new cases per year in the United States and 180,000 new cases worldwide. Survival rates have remained low for decades, with only 5% of SCLC patients surviving for 5 years. Most of the reason is due to the insufficient new therapies against this form of lung cancer. Most patients present with advanced disease, although approximately one-third of patients present with limited-stage disease defined by the tumor being present in only one side of the chest and fitting within a single radiation field. Such disease stages affect the available treatment plans, with limited-stage disease treated with chemotherapy and radiation and advanced disease treated with chemotherapy alone. Disseminated metastatic tumors with lymphoid-like characteristics are a prominent feature of SCLC. The first diagnosed SCLC patients were described as lymphatic diseases and were not recognized as lung cancers until 1926. This reveals one aspect of the properties unique to SCLC tumors compared to other solid tumors.

[0003] Patients usually respond well to current state-of-the-art therapies that include etoposide and cisplatin but present with chemotherapy-resistant disease for which no treatment options are currently available and invariably relapse soon after. The prognosis in the refractory situation after relapse is extremely poor, the disease progresses rapidly, and the median survival is short, less than six months. In addition, SCLC patients have a high prevalence of comorbidities, including hypertension, heart disease, diabetes, and tumor-associated syndromes. This, combined with the fact that SCLC patients are generally older, affects the patient's ability to tolerate an aggressive chemotherapy regimen and further limits treatment options.

[0004] The bispecific antibody format, which includes an scFv that recognizes CD3 expressed on T cells and another scFv that recognizes a tumor-associated antigen, has shown promising clinical efficacy due to the high response rate in hematological malignancies such as refractory B-ALL (Topp, M.S. et al. Blood, 2012. 120(26): p. 5185-5187 (Non-Patent Document 1)), and as a result, Blincyto has been approved. Although the efficacy of T cell-induced therapy has not yet been shown in solid tumor indications, considering the disseminated nature of this disease, SCLC may be a promising solid tumor indication for the bispecific antibody format targeting CD3x tumor. Therefore, bispecific T cell-induced therapy that induces T cells to specific tumor antigens presents new possibilities as a new treatment option in SCLC treatment.

[0005] By next-generation sequencing comparing the prevalence of DLL3 mRNA in a panel of primary patient tumors with a large population of normal tissues, DLL3 has now been identified as an SCLC-specific tumor antigen. Although the expression level of DLL3 in SCLC tumors is moderate, it is very commonly seen, and approximately 90% of the tumors analyzed showed evidence of DLL3 expression by RNA-seq. In contrast to SCLC tumors, in normal tissues, the expression of DLL3 transcripts was extremely low, detected only faintly in the testis, optic nerve, and cerebellum. Comparisons of SCLC cell lines and tumors by RNA-seq showed similar expression levels, but for cell surface quantification of DLL3 expression in SCLC cell lines, while the general expression level was less than 2000 DLL3 / cell, the expression level of DLL3 was shown to be less than 5000 per cell. When the expression of DLL3 protein was confirmed by IHC, 86% of SCLC tumors showed positive staining with a uniform membrane staining pattern for DLL3. Except for faint staining in the cerebellum, all other normal tissues were DLL3 staining negative.

[0006] DLL3 is a non-classical Notch ligand that functions cell-autonomously to inhibit Notch signaling, binds to Notch in cis configuration, thereby preventing the cell-cell interaction and internalization of Notch in target cells, which are prominent features of classical Notch signaling. The main role of DLL3 is in somitogenesis during embryonic development. Mice with DLL3 knocked out show regional defects in the development of the axial skeleton as well as the skull and nerves. Defects in somitogenesis pattern formation are also seen in humans with certain germline DLL3 mutations, giving rise to a condition called spondylocostal dysostosis.

[0007] DLL3 has already been proposed for diagnostic and treatment methods using antibody-drug conjugates (ADCs) in gliomas in addition to SCLC (WO2013 / 126746 (Patent Document 1)). Since DLL3 has a low expression level of the protein on the cell surface and considering that the performance of ADCs decreases for targets with low expression, there may be limitations in using the method with ADCs for DLL3. Furthermore, ADC molecules often exhibit toxicity associated with the released warhead drugs, which mostly occur due to the degradation of the linker, so there is a limit to the maximum tolerated dose and it may affect the efficacy unrelated to the target selected for the antibody. In a cell line expressing hundreds of target proteins per cell, since the sensitivity of T cells to the required target and very strong in vitro cytotoxicity have been demonstrated, this is less likely to be a problem for T cell-inducing bispecific molecules designed to simultaneously induce DLL3 and CD3. Furthermore, bispecific T cell-inducing antibody constructs, which are usually smaller in size compared to normal antibodies (full-length IgG), can promote tissue penetration and enhance efficacy in order to more efficiently induce DLL3 and CD3 targets, thereby strengthening synapse formation between T cells and target tumor cells.

Prior Art Documents

Patent Documents

[0008]

Patent Document 1

Non-Patent Documents

[0009]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0010] There are still unmet medical needs in SCLC, and new treatment options are needed to improve the prognosis of this significant patient population. The bispecific antibody format described above has been clinically proven, and antibody constructs targeting DLL3 and CD3 as such show new possibilities that are promising for the treatment of SCLC and the potential to increase the survival rate of patients suffering from this indication. Although additional options are still needed for treating tumors or cancer diseases associated with overexpression of DLL3, the present specification provides means and methods for solving this problem in the form of a bispecific antibody construct having a first binding domain directed to DLL3 and a second binding domain directed to CD3 on T cells. [Means for Solving the Problem]

[0011] Accordingly, in a first aspect, the present invention provides a bispecific antibody construct comprising a first binding domain that binds to human DLL3 on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells, wherein the first binding domain binds to an epitope of DLL3 contained within the region shown in SEQ ID NO: 260. [Invention 1001] A bispecific antibody construct comprising a first binding domain that binds to human DLL3 on the surface of target cells and a second binding domain that binds to human CD3 and at least cynomolgus CD3 on the surface of T cells, wherein the first binding domain binds to an epitope of DLL3 contained within the region shown in SEQ ID NO: 260. [Invention 1002] The antibody construct of Invention 1001, wherein the first binding domain further binds to cynomolgus DLL3, preferably Macaca fascicularis DLL3. [Invention 1003] An antibody construct of the present invention 1001 or 1002, wherein the second binding domain binds to human CD3 epsilon and CD3 epsilon of common marmoset (Callithrix jacchus), cotton-top tamarin (Saguinus Oedipus), or common squirrel monkey (Saimiri sciureus). [The present invention 1004] An antibody construct of any of the prior inventions, which is in a form selected from the group consisting of (scFv)2, scFv single domain mAb, diabody, and oligomers of said form. [The present invention 1005] An antibody construct of any of the prior inventions, wherein the first binding domain binds to an epitope of DLL3 contained within the region shown in SEQ ID NO: 258. [The present invention 1006] An antibody construct of the present invention 1005, wherein the first binding domain comprises a VH region containing CDR-H1, CDR-H2, and CDR-H3 selected from the group consisting of: and a VL region containing CDR-L1, CDR-L2, and CDR-L3: a) CDR-H1 shown in SEQ ID NO: 31, CDR-H2 shown in SEQ ID NO: 32, CDR-H3 shown in SEQ ID NO: 33, CDR-L1 shown in SEQ ID NO: 34, CDR-L2 shown in SEQ ID NO: 35, and CDR-L3 shown in SEQ ID NO: 36; b) CDR-H1 shown in SEQ ID NO: 41, CDR-H2 shown in SEQ ID NO: 42, CDR-H3 shown in SEQ ID NO: 43, CDR-L1 shown in SEQ ID NO: 44, CDR-L2 shown in SEQ ID NO: 45, and CDR-L3 shown in SEQ ID NO: 46; c) CDR-H1 shown in SEQ ID NO: 51, CDR-H2 shown in SEQ ID NO: 52, CDR-H3 shown in SEQ ID NO: 53, CDR-L1 shown in SEQ ID NO: 54, CDR-L2 shown in SEQ ID NO: 55, and CDR-L3 shown in SEQ ID NO: 56; d) CDR-H1 shown in SEQ ID NO: 61, CDR-H2 shown in SEQ ID NO: 62, CDR-H3 shown in SEQ ID NO: 63, CDR-L1 shown in SEQ ID NO: 64, CDR-L2 shown in SEQ ID NO: 65, and CDR-L3 shown in SEQ ID NO: 66; e) CDR-H1 shown in SEQ ID NO: 71, CDR-H2 shown in SEQ ID NO: 72, CDR-H3 shown in SEQ ID NO: 73, CDR-L1 shown in SEQ ID NO: 74, CDR-L2 shown in SEQ ID NO: 75, and CDR-L3 shown in SEQ ID NO: 76; f) CDR-H1 shown in SEQ ID NO: 81, CDR-H2 shown in SEQ ID NO: 82, CDR-H3 shown in SEQ ID NO: 83, CDR-L1 shown in SEQ ID NO: 84, CDR-L2 shown in SEQ ID NO: 85, and CDR-L3 shown in SEQ ID NO: 86; g) CDR-H1 shown in SEQ ID NO: 91, CDR-H2 shown in SEQ ID NO: 92, CDR-H3 shown in SEQ ID NO: 93, CDR-L1 shown in SEQ ID NO: 94, CDR-L2 shown in SEQ ID NO: 95, and CDR-L3 shown in SEQ ID NO: 96; h) CDR-H1 shown in SEQ ID NO: 101, CDR-H2 shown in SEQ ID NO: 102, CDR-H3 shown in SEQ ID NO: 103, CDR-L1 shown in SEQ ID NO: 104, CDR-L2 shown in SEQ ID NO: 105, and CDR-L3 shown in SEQ ID NO: 106; and i) CDR-H1 shown in SEQ ID NO: 111, CDR-H2 shown in SEQ ID NO: 112, CDR-H3 shown in SEQ ID NO: 113, CDR-L1 shown in SEQ ID NO: 114, CDR-L2 shown in SEQ ID NO: 115, and CDR-L3 shown in SEQ ID NO: 116. [Invention 1007] The antibody construct of Invention 1005 or 1006, wherein the first binding domain comprises a VH region selected from the group consisting of those shown in SEQ ID NO: 37, SEQ ID NO: 47, SEQ ID NO: 57, SEQ ID NO: 67, SEQ ID NO: 77, SEQ ID NO: 87, SEQ ID NO: 97, SEQ ID NO: 107, SEQ ID NO: 117, SEQ ID NO: 435, and SEQ ID NO: 529. [Invention 1008] The antibody construct of any one of Inventions 1005 to 1007, wherein the first binding domain comprises a VL region selected from the group consisting of those shown in SEQ ID NO: 38, SEQ ID NO: 48, SEQ ID NO: 58, SEQ ID NO: 68, SEQ ID NO: 78, SEQ ID NO: 88, SEQ ID NO: 98, SEQ ID NO: 108, SEQ ID NO: 118, SEQ ID NO: 436, and SEQ ID NO: 530. [Invention 1009] The antibody construct according to any one of Inventions 1005 to 1008, wherein the first binding domain comprises a VH region and a VL region selected from the group consisting of pairs of VH regions and VL regions shown in SEQ ID NO: 37+38, SEQ ID NO: 47+48, SEQ ID NO: 57+58, SEQ ID NO: 67+68, SEQ ID NO: 77+78, SEQ ID NO: 87+88, SEQ ID NO: 97+98, SEQ ID NO: 107+108, SEQ ID NO: 117+118, SEQ ID NO: 435+436, and SEQ ID NO: 529+530. [Invention 1010] The antibody construct according to any one of Inventions 1005 to 1009, wherein the first binding domain comprises a polypeptide selected from the group consisting of those shown in SEQ ID NO: 39, SEQ ID NO: 49, SEQ ID NO: 59, SEQ ID NO: 69, SEQ ID NO: 79, SEQ ID NO: 89, SEQ ID NO: 99, SEQ ID NO: 109, SEQ ID NO: 119, SEQ ID NO: 437, and SEQ ID NO: 531. [Invention 1011] The antibody construct according to any one of Inventions 1005 to 1010, comprising a polypeptide selected from the group consisting of those shown in SEQ ID NO: 40, SEQ ID NO: 50, SEQ ID NO: 60, SEQ ID NO: 70, SEQ ID NO: 80, SEQ ID NO: 90, SEQ ID NO: 100, SEQ ID NO: 110, SEQ ID NO: 120, SEQ ID NO: 211, SEQ ID NO: 212, SEQ ID NO: 213, SEQ ID NO: 214, SEQ ID NO: 215, SEQ ID NO: 216, SEQ ID NO: 217, SEQ ID NO: 438, and SEQ ID NO: 532. [Invention 1012] The antibody construct according to any one of Inventions 1001 to 1004, wherein the first binding domain binds to an epitope of DLL3 contained within the region shown in SEQ ID NO: 259. [Invention 1013] The antibody construct of Invention 1012, wherein the first binding domain comprises a VH region comprising CDR-H1, CDR-H2, and CDR-H3 and a VL region comprising CDR-L1, CDR-L2, and CDR-L3, selected from the following group: a) CDR-H1 shown in SEQ ID NO: 121, CDR-H2 shown in SEQ ID NO: 122, CDR-H3 shown in SEQ ID NO: 123, CDR-L1 shown in SEQ ID NO: 124, CDR-L2 shown in SEQ ID NO: 125, and CDR-L3 shown in SEQ ID NO: 126; b) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; c) CDR-H1 shown in SEQ ID NO: 141, CDR-H2 shown in SEQ ID NO: 142, CDR-H3 shown in SEQ ID NO: 143, CDR-L1 shown in SEQ ID NO: 144, CDR-L2 shown in SEQ ID NO: 145, and CDR-L3 shown in SEQ ID NO: 146; d) CDR-H1 shown in SEQ ID NO: 151, CDR-H2 shown in SEQ ID NO: 152, CDR-H3 shown in SEQ ID NO: 153, CDR-L1 shown in SEQ ID NO: 154, CDR-L2 shown in SEQ ID NO: 155, and CDR-L3 shown in SEQ ID NO: 156; e) CDR-H1 shown in SEQ ID NO: 161, CDR-H2 shown in SEQ ID NO: 162, CDR-H3 shown in SEQ ID NO: 163, CDR-L1 shown in SEQ ID NO: 164, CDR-L2 shown in SEQ ID NO: 165, and CDR-L3 shown in SEQ ID NO: 166; f) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 439, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; g) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 440, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; h) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 441, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; i) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 442, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; j) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 443, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; k) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 444, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; l) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 439, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 441, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; and m) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 440, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 442, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136. [Invention 1014] The antibody construct of the present invention 1012 or 1013, wherein the first binding domain comprises a VH region selected from the group consisting of those shown in SEQ ID NO: 127, SEQ ID NO: 137, SEQ ID NO: 147, SEQ ID NO: 157, SEQ ID NO: 167, SEQ ID NO: 445, SEQ ID NO: 446, SEQ ID NO: 447, SEQ ID NO: 448, SEQ ID NO: 449, SEQ ID NO: 450, SEQ ID NO: 451, SEQ ID NO: 452, SEQ ID NO: 453, SEQ ID NO: 454, and SEQ ID NO: 455. [The present invention 1015] The antibody construct of any one of the present inventions 1012 to 1014, wherein the first binding domain comprises a VL region selected from the group consisting of those shown in SEQ ID NO: 128, SEQ ID NO: 138, SEQ ID NO: 148, SEQ ID NO: 158, SEQ ID NO: 168, SEQ ID NO: 456, SEQ ID NO: 457, SEQ ID NO: 458, SEQ ID NO: 459, SEQ ID NO: 460, SEQ ID NO: 461, SEQ ID NO: 462, SEQ ID NO: 463, SEQ ID NO: 464, SEQ ID NO: 465, SEQ ID NO: 466, SEQ ID NO: 467, SEQ ID NO: 468, SEQ ID NO: 469, and SEQ ID NO: 470. [The present invention 1016] The antibody construct of any one of the present inventions 1012 to 1015, wherein the first binding domain comprises a VH region and a VL region selected from the group consisting of pairs of VH regions and VL regions shown in SEQ ID NO: 127+128, SEQ ID NO: 137+138, SEQ ID NO: 147+148, SEQ ID NO: 157+158, SEQ ID NO: 167+168, SEQ ID NO: 137+456, SEQ ID NO: 137+457, SEQ ID NO: 137+458, SEQ ID NO: 137+459, SEQ ID NO: 137+460, SEQ ID NO: 445+138, SEQ ID NO: 446+138, SEQ ID NO: 447+138, SEQ ID NO: 445+460, SEQ ID NO: 448+461, SEQ ID NO: 449+462, SEQ ID NO: 450+463, SEQ ID NO: 450+464, SEQ ID NO: 450+465, SEQ ID NO: 450+466, SEQ ID NO: 450+467, SEQ ID NO: 450+468, SEQ ID NO: 451+463, SEQ ID NO: 452+463, SEQ ID NO: 453+463, SEQ ID NO: 451+468, SEQ ID NO: 454+469, and SEQ ID NO: 455+470. [The present invention 1017] The antibody construct according to any one of aspects 1012 to 1016 of the present invention, wherein the first binding domain comprises a polypeptide selected from the group consisting of those set forth in SEQ ID NO: 129, SEQ ID NO: 139, SEQ ID NO: 149, SEQ ID NO: 159, SEQ ID NO: 169, SEQ ID NO: 471, SEQ ID NO: 472, SEQ ID NO: 473, SEQ ID NO: 474, SEQ ID NO: 475, SEQ ID NO: 476, SEQ ID NO: 477, SEQ ID NO: 478, SEQ ID NO: 479, SEQ ID NO: 480, SEQ ID NO: 481, SEQ ID NO: 482, SEQ ID NO: 483, SEQ ID NO: 484, SEQ ID NO: 485, SEQ ID NO: 486, SEQ ID NO: 487, SEQ ID NO: 488, SEQ ID NO: 489, SEQ ID NO: 490, SEQ ID NO: 491, SEQ ID NO: 492, and SEQ ID NO: 493. [Aspect 1018] The antibody construct according to any one of aspects 1012 to 1017 of the present invention, which comprises a polypeptide selected from the group consisting of those set forth in SEQ ID NO: 130, SEQ ID NO: 140, SEQ ID NO: 150, SEQ ID NO: 160, SEQ ID NO: 170; SEQ ID NO: 218, SEQ ID NO: 219, SEQ ID NO: 220, SEQ ID NO: 494, SEQ ID NO: 495, SEQ ID NO: 496, SEQ ID NO: 497, SEQ ID NO: 498, SEQ ID NO: 499, SEQ ID NO: 500, SEQ ID NO: 501, SEQ ID NO: 502, SEQ ID NO: 503, SEQ ID NO: 504, SEQ ID NO: 505, SEQ ID NO: 506, SEQ ID NO: 507, SEQ ID NO: 508, SEQ ID NO: 509, SEQ ID NO: 510, SEQ ID NO: 511, SEQ ID NO: 512, SEQ ID NO: 513, SEQ ID NO: 514, SEQ ID NO: 515, and SEQ ID NO: 516. [Aspect 1019] The antibody construct according to any one of aspects 1005 to 1011 of the present invention, which comprises or consists of a polypeptide selected from the group consisting of those set forth in SEQ ID NO: 517, SEQ ID NO: 518, SEQ ID NO: 519, SEQ ID NO: 520, SEQ ID NO: 521, SEQ ID NO: 522, SEQ ID NO: 523, and SEQ ID NO: 524. [Aspect 1020] A polynucleotide encoding any antibody construct of the prior invention. [Aspect 1021] A vector comprising the polynucleotide of aspect 1020 of the present invention. [Aspect 1022] A host cell transformed or transfected with the polynucleotide of the present invention 1020 or the vector of the present invention 1021. [The present invention 1023] A method for producing an antibody construct of any one of the present inventions 1001 to 1019, comprising culturing the host cell of the present invention 1022 under conditions that allow expression of the antibody construct, and recovering the produced antibody construct from the culture. [The present invention 1024] A pharmaceutical composition comprising an antibody construct of any one of the present inventions 1001 to 1018 or an antibody construct produced according to the method of the present invention 1023. [The present invention 1025] An antibody construct of any one of the present inventions 1001 to 1019 or an antibody construct produced according to the method of the present invention 1023, which is used for the prevention, treatment, or improvement of a tumor or cancer disease, or a metastatic cancer disease. [The present invention 1026] A method for the treatment or improvement of a tumor or cancer disease, or a metastatic cancer disease, comprising administering to a subject in need thereof an antibody construct of any one of the present inventions 1001 to 1019 or an antibody construct produced according to the method of the present invention 1023. [The present invention 1027] The method of the present invention 1026 or the antibody construct of the present invention 1024, wherein the tumor or cancer disease is selected from the group consisting of tumors or cancers of the lung, preferably SCLC, breast, cervix, colon, colorectal, endometrium, head and neck, liver, ovary, pancreas, prostate, skin, stomach, testis, thyroid, adrenal gland, kidney, bladder, uterus, esophagus, urothelium, and brain, lymphomas, carcinomas, and sarcomas, and metastatic cancer diseases derived from any of the foregoing. [The present invention 1028] A kit comprising an antibody construct of any one of the present inventions 1001 to 1019, an antibody construct produced according to the method of the present invention 1023, the polynucleotide of the present invention 1020, the vector of the present invention 1021, and / or the host cell of the present invention 1022.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3A

Figure 3B

Figure 3C

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Mode for Carrying Out the Invention

[0013] As used herein, it should be noted that the singular forms "a", "an", and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a reagent" includes one or more such different reagents, and reference to "a method" includes reference to equivalent steps and methods known to those skilled in the art that can modify or replace the methods described herein.

[0014] Unless otherwise indicated, the term "at least" preceding a series of elements should be understood to refer to all of the elements in the series. Those skilled in the art will recognize many equivalents to the particular embodiments of the invention described herein or will be able to ascertain such equivalents using only routine experimentation. Such equivalents are intended to be encompassed by the present invention.

[0015] As used herein, the term "and / or" includes the meanings of "and", "or", and "all or any other combination of the elements connected by said term".

[0016] As used herein, the term "about" means within ±20%, preferably within ±15%, more preferably within ±10%, and most preferably within ±5% of a given value or range.

[0017] Throughout this specification and the appended claims, unless the context requires otherwise, the words "comprise", "comprises" and "comprising" are to be interpreted as including the recited integer or step, or group of integers or steps, but not excluding any other integer or step, or group of integers or steps. As used herein, the term "comprising" is interchangeable with the term "containing" or "including", or alternatively, when used herein, may in some cases be interchangeable with the term "having".

[0018] As used herein, "consisting of" excludes any element, step, or ingredient not specified in the claim element. As used herein, "consisting essentially of" excludes materials or steps that do not substantially affect the basic and novel characteristics of the claim.

[0019] In each example herein, the terms "comprising", "consisting essentially of", and "consisting of" are interchangeable with either of the other two terms.

[0020] The term "antibody construct" refers to a molecule whose structure and / or function is based on that of an antibody, e.g., a full-length or whole immunoglobulin molecule. Thus, an antibody construct can bind to its specific target or antigen. Furthermore, the antibody constructs according to the invention include the minimal structural requirements of an antibody that enable target binding. This minimal requirement can be specified, for example, by the presence of at least three light-chain CDRs (i.e., CDR1, CDR2, and CDR3 of the VL region) and / or three heavy-chain CDRs (i.e., CDR1, CDR2, and CDR3 of the VH region), preferably all six CDRs. Antibodies based on the constructs according to the invention include, for example, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies, humanized antibodies, and human antibodies.

[0021] Within the definition of "antibody construct" according to the present invention are also full-length antibodies or whole antibodies, including camel antibodies and other immunoglobulin antibodies produced by methods or processes of biotechnology or protein engineering. These full-length antibodies can be, for example, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies, humanized antibodies, and human antibodies. Fragments of full-length antibodies, such as VH, VHH, VL, (s)dAb, Fv, Fd, Fab, Fab’, F(ab’)2, or "r IgG" ("half antibody"), are also within the definition of "antibody construct". Antibody constructs according to the present invention are also modified antibody fragments, also referred to as antibody variants, for example, 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, "minibodies" exemplified by the following structures: (VH-VL-CH3)2, (scFv-CH3)2, ((scFv)2-CH3)+CH3), ((scFv)2-CH3), or (scFv-CH3-scFv)2, multi-bodies, such as tribody or tetrabody, and single-domain antibodies, such as nanobody, or single variable domain antibodies that may contain only one variable domain that specifically binds independently to an antigen or epitope of another V region or domain, such as VHH, VH, or VL. Further preferred forms of antibody constructs according to the present invention are crossbody, maxibody, hetero-Fc construct, and mono-Fc construct. Examples of these forms are described below.

[0022] The binding domain may generally include the variable region of the antibody light chain (VL) and the variable region of the antibody heavy chain (VH), but it is not necessary to include both. The Fd fragment, for example, has two VH regions and often retains some of the antigen-binding function of the intact antigen-binding domain. Additional examples of antibody fragments, antibody variants, or forms of the binding domain include: (1) the Fab fragment, which is a monovalent fragment having the VL, VH, CL, and CH1 domains; (2) the F(ab’)2 fragment, which is a divalent fragment having two Fab fragments linked by a disulfide bridge in the hinge region; (3) the Fd fragment having two VH and CH1 domains; (4) the Fv fragment having the VL and VH domains of a single arm of the antibody; (5) the dAb fragment having the VH domain (Ward et al., (1989) Nature 341:544-546); (6) the isolated complementarity-determining region (CDR), and (7) the single-chain Fv (scFv), with the last being preferred (e.g., from an scFv library). Examples of embodiments of the antibody constructs according to the present invention are described, for example, in WO00 / 006605, WO2005 / 040220, WO2008 / 119567, WO2010 / 037838, WO2013 / 026837, WO2013 / 026833, US2014 / 0308285, US2014 / 0302037, WO2014 / 144722, WO2014 / 151910, and WO2015 / 048272.

[0023] Furthermore, the definition of the term “antibody construct” includes monovalent, divalent, and polyvalent constructs, and thus monospecific constructs that specifically bind to only one antigenic structure, as well as bispecific and polyspecific / multispecific constructs that specifically bind to two or more antigenic structures, e.g., two, three, or more, through different binding domains. Furthermore, the definition of the term “antibody construct” includes molecules consisting of only one polypeptide chain, as well as molecules consisting of two or more polypeptide chains that can be either the same chain (homo-dimer, homo-trimer, or homo-oligomer) or different chains (hetero-dimer, hetero-trimer, or hetero-oligomer). Examples of the antibodies and their variants or derivatives specified above are described in particular in Harlow and Lane, Antibodies a laboratory manual, CSHL Press (1988) and Using Antibodies: a laboratory manual, CSHL Press (1999), Kontermann and Dubel, Antibody Engineering, Springer, 2nd ed. 2010 and Little, Recombinant Antibodies for Immunotherapy, Cambridge University Press 2009.

[0024] The antibody constructs of the present invention are preferably “antibody constructs made in vitro”. This term refers to antibody constructs as defined above, in which all or part of the variable region (e.g., at least one CDR) is made by selection of non-immune cells, e.g., in vitro phage display, protein chips, or any other method by which the antigen-binding ability of a candidate sequence can be tested. Thus, this term preferably excludes sequences made only by genomic rearrangement in the immune cells of an animal. A “recombinant antibody” is an antibody made using recombinant DNA technology or genetic engineering.

[0025] As used herein, the term "monoclonal antibody" (mAb) or monoclonal antibody construct means an antibody obtained from a substantially homogeneous population of antibodies, i.e., individual antibodies that comprise a population that is identical except for possible natural mutations and / or post-translational modifications (e.g., isomerization, amidation) that may be present in minor amounts. Monoclonal antibodies are highly specific and are directed against a single antigenic site or determinant on an antigen, in contrast to conventional (polyclonal) antibody preparations that generally include different antibodies against different determinants (or epitopes). In addition to their specificity, monoclonal antibodies are advantageous in that they are synthesized by hybridoma culture and are thus free of contamination by other immunoglobulins. The modifier "monoclonal" indicates the characteristic of the antibody as being obtained from a substantially homogeneous population of antibodies and should not be construed as requiring the production of the antibody by any particular method.

[0026] Any technique that provides for the production of antibodies by culturing of continuous cell lines can be used for the preparation of monoclonal antibodies. For example, the monoclonal antibodies to be used can be made by the hybridoma method first described by Koehler et al., Nature 256:495 (1975), or by recombinant DNA methods (see, e.g., U.S. Patent No. 4,816,567). Examples of additional techniques for the production of human monoclonal antibodies include the trioma technique, the human B-cell hybridoma technique (Kozbor, Immunology Today 4 (1983), 72), and the EBV-hybridoma technique (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc. (1985), 77-96).

[0027] The hybridomas are then screened using standard methods such as enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance (BIACORE™) analysis to identify one or more hybridomas that produce antibodies that specifically bind to the designated antigen. For example, any form of the relevant antigen can be used as an immunogen, such as a recombinant antigen, a naturally occurring form, any variant or fragment thereof, and antigenic peptides thereof. The use of surface plasmon resonance employed in the BIAcore system can enhance the efficiency of binding of phage antibodies to epitopes of target antigens such as DLL3 or CD3 epsilon (Schier, Human Antibodies Hybridomas 7(1996),97-105; Malmborg, J. Immunol. Methods 183(1995),7-13).

[0028] Another exemplary method for generating monoclonal antibodies involves screening protein expression libraries, such as phage display or ribosome display libraries. Phage display is described, for example, in U.S. Patent 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).

[0029] In addition to the use of display libraries, the relevant antigen can be used for immunization of animals other than humans, such as rodents (such as mice, hamsters, rabbits, or rats). In one embodiment, the animal other than human contains at least a part of the human immunoglobulin gene. For example, it is possible to modify a mouse strain lacking mouse antibody production using a large fragment of the human Ig (immunoglobulin) locus. Monoclonal antibodies specific for antigens derived from genes having desired specificities can be produced and selected using hybridoma technology. See, for example, XENOMOUSE (trademark), Green et al. (1994) Nature Genetics 7:13-21, US2003-0070185, WO96 / 34096, and WO96 / 33735.

[0030] Monoclonal antibodies can also be modified, such as humanization, deimmunization, and chimerization, using recombinant DNA techniques known in the art after being obtained from animals other than humans. Examples of modified antibody constructs include humanized variants of antibodies other than human, "affinity matured" antibodies (see, for example, Hawkins et al., J. Mol. Biol. 254, 889-896 (1992) and Lowman et al., Biochemistry 30, 10832-10837 (1991)), and antibody variants with modified effector function(s) (see, for example, U.S. Patent No. 5,648,260, Kontermann and Dubel (2010) supra, and Little (2009) supra).

[0031] In immunology, affinity maturation is the process by which B cells produce antibodies with increased affinity for an antigen during an immune response. Upon repeated exposure to the same antigen, the host produces antibodies with continuously increasing affinity. Similar to natural prototypes, in vitro affinity maturation is based on the principles of mutation and selection. In vitro affinity maturation has been used successfully for the optimization of antibodies, antibody constructs, and antibody fragments. Random mutations within the CDRs are introduced using radiation, chemical mutagens, or error-prone PCR methods. Genetic diversity can also be increased by chain shuffling. Two or three rounds of mutation and selection using presentation methods such as phage display typically yield antibody fragments with affinities in the low nanomolar range.

[0032] Preferred amino acid substitution variants of antibody constructs include substitutions of one or more hypervariable region residues of a parent antibody (e.g., a humanized or human antibody). Generally, as a result of selection for further development, the resulting variant(s) have improved biological properties compared to the original parent antibody. A convenient method for generating such substitution variants involves affinity maturation using phage display. Briefly, several hypervariable region sites (e.g., 6 - 7 sites) are mutated such that all possible amino acid substitutions occur at each site. The antibody variants thus generated are presented in monovalent form from filamentous phage particles as fusions with the gene III product of M13 packaged within each particle. The phage-displayed variants are then screened for their biological activities (e.g., binding affinity) disclosed herein. To identify candidate hypervariable region sites to be modified, the alanine scan mutagenesis method can be performed to identify hypervariable region residues that significantly contribute to antigen binding. Alternatively or in addition, it may be beneficial to analyze the crystal structure of the antigen - antibody complex to identify the contact points between the binding domain and, for example, human DLL3. Such contact residues and adjacent residues are candidates for substitution by the techniques detailed herein. After generating such variants, the panel of variants can be subjected to the screening described herein, and antibodies that exhibit excellent properties in one or more relevant assays can be selected for further development.

[0033] The monoclonal antibodies and antibody constructs of the present invention are specifically "chimeric" antibodies (immunoglobulins) in which a portion of the heavy and / or light chain is identical or homologous to the corresponding sequence of an antibody derived from a particular species or belonging to a particular antibody class or subclass, while the remaining portion of the chain(s) is derived from a different species or is identical or homologous to the corresponding sequence of an antibody belonging to a different antibody class or subclass, and fragments of such antibodies as long as they exhibit the desired biological activity (U.S. Patent No. 4,816,567; Morrison et al., Proc. Natl. Acad. Sci. USA, 81:6851-6855 (1984)). Chimeric antibodies as contemplated herein include "primatized" antibodies that include variable domain antigen-binding sequences derived from non-human primates (e.g., Old World monkeys, apes, etc.) and human constant region sequences. A variety of methods for making chimeric antibodies have been described. See, for example, Morrison et al., Proc. Natl. Acad. Sci. U.S.A. 81:6851, 1985; Takeda et al., Nature 314:452, 1985, Cabilly et al., U.S. Patent No. 4,816,567; Boss et al., U.S. Patent No. 4,816,397; Tanaguchi et al., EP0171496; EP0173494; and GB 2177096.

[0034] Antibodies, antibody constructs, antibody fragments, or antibody variants can also be modified by specific deletion of human T cell epitopes (a method called "deimmunization") by the methods disclosed in WO98 / 52976 or WO00 / 34317. Briefly, for peptides that bind to MHC class II, the heavy and light chain variable domains of the antibody can be analyzed. This peptide becomes a potential T cell epitope (as defined in WO98 / 52976 and WO00 / 34317). For the detection of potential T cell epitopes, a computer modeling technique called "peptide threading" can be used as described in WO98 / 52976 and WO00 / 34317. In addition, motifs present in the VH and VL sequences can be searched in a database of human MHC class II binding peptides. These motifs can become potential T cell epitopes because they bind to any of the 18 major MHC class II DR allotypes. The detected potential T cell epitopes can be removed by substituting a small number of amino acid residues within the variable domain, or preferably by a single amino acid substitution. Usually, conservative substitutions are made. In many cases, but not all, amino acids common to positions within the human germline antibody sequences can be used. For human germline sequences, see, for example, Tomlinson, et al. (1992) J. Mol. Biol. 227:776-798; Cook, G.P. et al. (1995) Immunol. Today Vol. 16(5):237-242; and Tomlinson et al. (1995) EMBO J. 14:14:4628-4638. The V BASE directory provides a comprehensive directory of human immunoglobulin variable region sequences (compiled by Tomlinson, L.A. et al., MRC Centre for Protein Engineering, Cambridge, UK). These sequences can be used as a source of human sequences, for example, in the framework regions and CDRs. For example, the consensus human framework regions described in U.S. Patent No. 6,300,064 can also be used.

[0035] A "humanized" antibody, antibody construct, or variant or fragment thereof (Fv, Fab, Fab', F(ab')2, or other antigen-binding portion sequence of an antibody) is an antibody or immunoglobulin that contains minimal non-human immunoglobulin-derived sequence(s) and is mostly human sequence. Primarily, a humanized antibody is a human immunoglobulin (recipient antibody) in which residues from the hypervariable regions (or CDRs) of the recipient have been replaced by residues from the hypervariable regions of a non-human (e.g., rodent) species (donor antibody) that has the desired specificity, affinity, and capacity, e.g., mouse, rat, hamster, or rabbit. Optionally, residues in the Fv framework regions (FRs) of the human immunoglobulin are replaced with the corresponding non-human residues. Further, as used herein, a "humanized antibody" can also contain residues not found in either the recipient antibody or the donor antibody. Such modifications are made to further improve and optimize the performance of the antibody. A humanized antibody can also contain an immunoglobulin constant region (Fc), usually at least a portion of the constant region of a human immunoglobulin. See, e.g., Jones et al., Nature, 321:522-525 (1986); Reichmann et al., Nature, 332:323-329 (1988); and Presta, Curr. Op. Struct. Biol., 2:593-596 (1992).

[0036] A humanized antibody or fragment thereof can be produced by replacing the sequence of the Fv variable domain that is not directly involved in antigen binding with an equivalent sequence derived from a human Fv variable domain. Exemplary methods for producing a humanized antibody or fragment thereof are shown in Morrison (1985) Science 229:1202-1207, Oi et al. (1986) BioTechniques 4:214, and US5,585,089, US5,693,761, US5,693,762, US5,859,205, and US6,407,213. This method involves isolating, manipulating, and expressing a nucleic acid sequence encoding all or a portion of an immunoglobulin Fv variable domain derived from at least one of the heavy or light chains. Such nucleic acids can be obtained from sources other than hybridomas that produce antibodies against a given target, as described above. Recombinant DNA encoding the humanized antibody molecule can then be cloned into an appropriate expression vector.

[0037] Humanized antibodies can also be produced using transgenic animals, such as mice, that express human heavy and light chain genes but are unable to express endogenous murine immunoglobulin heavy and light chain genes. Winter has described an exemplary CDR grafting method that can be used in the preparation of the humanized antibodies described herein (U.S. Patent No. 5,225,539). All of the CDRs of a particular human antibody may be replaced with at least a portion of non-human CDRs, or only some of the CDRs may be replaced with non-human CDRs. It is only necessary to replace the number of CDRs required for binding of the humanized antibody to a given antigen.

[0038] Humanized antibodies can be optimized by conservative substitutions, consensus sequence substitutions, germline substitutions, and / or introduction of back mutations. Such modified immunoglobulin molecules can be made by any of a number of techniques known in the art (see, e.g., Teng et al., Proc. Natl. Acad. Sci. U.S.A., 80:7308-7312, 1983; Kozbor et al., Immunology Today, 4:72-79, 1983; Olsson et al., Meth. Enzymol., 92:3-16, 1982; and EP239400).

[0039] The terms "human antibody," "human antibody construct," and "human binding domain" include antibody regions, e.g., variable and constant regions or domains, having antibody constructs, and binding domains that substantially correspond to human germline immunoglobulin sequences known in the art, e.g., those described by Kabat et al. (1991) (supra). The human antibodies, antibody constructs, or binding domains of the present invention can contain amino acid residues (e.g., mutations introduced by random mutagenesis or site-directed mutagenesis in vitro or by somatic mutations in vivo) that are not encoded by human germline immunoglobulin sequences, e.g., in CDRs, particularly CDR3. This human antibody, antibody construct, or binding domain can have at least 1, 2, 3, 4, 5, or more positions substituted with amino acid residues that are not encoded by human germline immunoglobulin sequences. As used herein, the definitions of human antibodies, antibody constructs, and binding domains also contemplate fully human antibodies containing only the human sequences of non-human and / or genetically modified antibodies that can be obtained by using techniques or systems such as Xenomouse.

[0040] In some embodiments, the antibody constructs of the invention are "isolated" or "substantially pure" antibody constructs. As used herein to describe the antibody constructs disclosed herein, "isolated" or "substantially pure" means an antibody construct that has been identified, separated, and / or recovered from the components of its production environment. Preferably, the antibody construct is not associated, or is substantially not associated, with all other components from its production environment. Contaminating components of its production environment, such as components resulting from recombinant transfected cells, are usually substances that interfere with the diagnostic or therapeutic use of the polypeptide and can include enzymes, hormones, and other proteinaceous or non-proteinaceous solutes. The antibody construct can constitute, for example, at least about 5% by weight or at least about 50% by weight of the total protein in a given sample. It is understood that an isolated protein can constitute from 5% to 99.9% by weight of the total protein content, depending on the environment. The polypeptide can be produced at significantly high concentrations by using an inducible or highly expressing promoter such that the polypeptide is produced at a significantly high concentration. 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 purified to a sufficient extent to obtain at least 15 residues of the N-terminal or internal amino acid sequence using a spinning cup sequenator or to homogeneity by SDS-PAGE under non-reducing or reducing conditions using Coomassie blue or preferably silver staining. However, usually, an isolated antibody construct is prepared by at least one purification step.

[0041] The term "binding domain" is, in the context of the present invention, considered to be a domain that (specifically) binds to / interacts with / recognizes a given target epitope or a given target site on a target molecule (antigen), namely DLL3 and CD3 respectively herein. The structure and function of the first binding domain (recognizing DLL3), preferably the structure and / or function of the second binding domain (recognizing CD3), are also based on the structure and / or function of an antibody, for example a full-length immunoglobulin molecule or a whole immunoglobulin molecule. According to the present invention, the first binding domain is characterized by the presence of three light-chain CDRs (i.e., CDR1, CDR2, and CDR3 of the VL region) and / or three heavy-chain CDRs (i.e., CDR1, CDR2, and CDR3 of the VH region). The second binding domain also preferably includes the minimum structural requirements of an antibody that enable target binding. More preferably, the second binding domain includes at least three light-chain CDRs (i.e., CDR1, CDR2, and CDR3 of the VL region) and / or three heavy-chain CDRs (i.e., CDR1, CDR2, and CDR3 of the VH region). The first and / or second binding domain is envisioned to be produced or obtained by phage display or library screening methods, in addition to grafting CDR sequences derived from existing (monoclonal) antibodies onto a scaffold.

[0042] According to the present invention, the binding domain is in the form of one or more polypeptides. Such polypeptides may include a proteinaceous portion and a non-proteinaceous portion (e.g., a chemical linker or a chemical crosslinking agent such as glutaraldehyde). Proteins (including fragments thereof, preferably biologically active fragments, and peptides having usually less than 30 amino acids) include two or more amino acids linked to each other through covalent peptide bonds (forming a chain of amino acids by binding). As used herein, the term "polypeptide" refers to a group of molecules consisting of usually more than 30 amino acids. Polypeptides may further form multimers such as dimers, trimers, and higher-order oligomers, i.e., they may consist of multiple polypeptide molecules. The polypeptide molecules forming such dimers, trimers, etc. may or may not be identical. The corresponding higher-order structures of such multimers are thus referred to as homo- or hetero-dimers, homo- or hetero-trimers, etc. An example of a heteromultimer is, in its native form, an antibody molecule consisting of two identical polypeptide light chains and two identical polypeptide heavy chains. The terms "peptide", "polypeptide", and "protein" also refer to naturally modified peptides / polypeptides / proteins that have been modified by post-translational modifications such as, for example, glycosylation, acetylation, phosphorylation, etc. As referred to herein, "peptide", "polypeptide", or "protein" may also be chemically modified such as PEGylated. Such modifications are well known in the art and are described below herein.

[0043] Preferably, the binding domain that binds to DLL3 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 avoid some of the problems associated with antibodies or antibody constructs that have non-human variable and / or constant regions, such as rodents (e.g., mice, rats, hamsters, or rabbits). The presence of such rodent-derived proteins can lead to rapid clearance of the antibody or antibody construct, or can generate an immune response against the antibody or antibody construct by the patient. To avoid the use of rodent-derived antibodies or antibody constructs, human or fully human antibodies / antibody constructs can be generated by introducing human antibody functions into rodents such that the rodents produce fully human antibodies.

[0044] The ability to clone and reconstruct megabase-sized human loci in YACs and introduce them into the mouse germ line is a powerful approach for elucidating the functional elements of very large or coarsely mapped loci, as well as for generating useful models of human disease. Furthermore, utilization of such techniques to replace mouse loci with their human equivalents can provide unique insights into the expression and regulation of developmental human gene products, their communication with other systems, and their involvement in disease induction and progression.

[0045] An important practical example of such a strategy is the "humanization" of the murine humoral immune system. Introduction of the human Ig locus into mice with inactivated endogenous immunoglobulin (Ig) genes provides an opportunity to study the mechanisms underlying the programmed expression and assembly of antibodies and the role in B cell development. Furthermore, such a strategy can provide an ideal source for the production of fully human monoclonal antibodies (mAbs), which are important milestones in realizing the potential of antibody therapies in human diseases. Fully human antibodies or antibody constructs are expected to minimize the immunogenicity and allergic reactions inherent to murine mAbs or mAb-derived from mice, thereby improving the efficacy and safety of the administered antibody / antibody construct. The use of fully human antibodies or antibody constructs is expected to provide significant advantages in the treatment of chronic and relapsing human diseases that require repeated administration of compounds, such as inflammation, autoimmunity, and cancer.

[0046] One approach towards this goal is to generate, by recombinant manipulation, a mouse strain that is defective in the production of mouse antibodies and has a large fragment of the human Ig locus, which was predicated on the prediction that such a mouse would produce an extensive repertoire of human antibodies without producing mouse antibodies. The large human Ig fragment is thought to retain appropriate regulation of antibody production and expression in addition to the extensive diversity of variable genes. By using this mouse mechanism for antibody diversification and selection and for breaking immune tolerance to human proteins, the human antibody repertoire recapitulated in this mouse strain should produce antibodies with high affinity for any antigen of interest, including human antigens. It is thought that antigen-specific human mAbs with desirable specificities can be readily generated and selected using hybridoma technology. This general strategy was demonstrated in connection with the generation of the first XenoMouse strain (see Green et al. Nature Genetics 7:13-21 (1994)). This XenoMouse strain was engineered using yeast artificial chromosomes (YACs) containing germline-constitutive fragments of 245 kb and 190 kb in size of the human heavy-chain and kappa light-chain loci, respectively, including the core sequences of the variable and constant regions. The YACs containing this human Ig were demonstrated to be compatible with the mouse system with respect to both antibody rearrangement and expression and were able to replace the inactivated mouse Ig genes. 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 suggested that the introduction of the majority of the human Ig locus, including a large number of V genes, additional regulatory elements, and human Ig constant regions, could recapitulate a substantially complete repertoire characteristic of the human humoral response to infection and immune stimulation. More recently, building on the work of Green et al., the introduction of megabase-sized germline-constitutive YAC fragments of the human heavy-chain and kappa light-chain loci, respectively, led to the introduction of more than about 80% of the human antibody repertoire.See Mendez et al., Nature Genetics 15:146-156 (1997) and U.S. Patent Application No. 08 / 759,620.

[0047] The generation of XenoMouse mice is further discussed and reviewed in U.S. Patent Application Nos. 07 / 466,008, 07 / 610,515, 07 / 919,297, 07 / 922,649, 08 / 031,801, 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, and 08 / 759,620, and U.S. Patents Nos. 6,162,963, 6,150,584, 6,114,598, 6,075,181, and 5,939,598, and Japanese Patents Nos. 3068180B2, 3068506B2, and 3068507B2. See also 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 WO03 / 47336.

[0048] In another approach, other companies, including GenPharm International, Inc., utilize the "mini-locus" approach. In this mini-locus approach, an exogenous Ig locus is mimicked by including a portion (individual genes) from this 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) form a construct that is inserted into an animal. This approach is described in U.S. Patent No. 5,545,807 to Surani et al., as well as U.S. Patent Nos. 5,545,806, 5,625,825, 5,625,126, 5,633,425, 5,661,016, 5,770,429, 5,789,650, 5,814,318, 5,877,397, 5,874,299, and 6,255,458 to Lonberg and Kay respectively, U.S. Patent Nos. 5,591,669 and 6,023,010 to Krimpenfort and Berns, U.S. Patent Nos. 5,612,205, 5,721,367, and 5,789,215 to Berns et al., U.S. Patent No. 5,643,763 to Choi and Dunn, as well as GenPharm International patent applications Nos. 07 / 574,748, 07 / 575,962, 07 / 810,279, 07 / 853,408, 07 / 904,068, 07 / 990,860, 08 / 053,131, 08 / 096,762, 08 / 155,301, 08 / 161,739, 08 / 165,699, 08 / 209,741. See also EP0546073B1, WO92 / 03918, WO92 / 22645, WO92 / 22647, WO92 / 22670, WO93 / 12227, WO94 / 00569, WO94 / 25585, WO96 / 14436, WO97 / 13852, and WO98 / 24884, as well as U.S. Patent No. 5,981,175.See also Taylor et al. (1992), Chen et al. (1993), Tuaillon et al. (1993), Choi et al. (1993), Lonberg et al. (1994), Taylor et al. (1994), and Tuaillon et al. (1995), Fishwild et al. (1996).

[0049] Kirin has also demonstrated the production of human antibodies from mice into which large chromosomal fragments or entire chromosomes were introduced by microcell fusion. See European Patent Applications Nos. 773288 and 843961. Xenerex Biosciences is developing promising human antibody production technology. In this technology, SCID mice are reconstituted with human lymphocytes, such as B cells and / or T cells. Thereafter, the mice can be immunostimulated with an antigen to generate an immune response against the antigen. See U.S. Patent Nos. 5,476,996, 5,698,767, and 5,958,765.

[0050] Triggered by the human anti-mouse antibody (HAMA) reaction, chimeric antibodies or other humanized antibodies have been produced in the industry. However, a certain type of human anti-chimeric antibody (HACA) reaction is expected to be observed in the future, particularly in the use of this antibody in chronic or multi-dose settings. Therefore, it is considered desirable to provide an antibody construct comprising a human binding domain for DLL3 and / or a human binding domain for CD3 in order to eliminate the concerns and / or effects of the HAMA or HACA reaction.

[0051] According to the present invention, the terms “(specifically) bind,” “(specifically) recognize,” “(specifically) induced,” and “(specifically) react” mean that the binding domain interacts or specifically interacts with a given target epitope or a given target site on the target molecule (antigen), i.e., DLL3 and CD3 respectively in this specification.

[0052] The term "epitope" refers to the site on an antigen to which a binding domain, an antibody or immunoglobulin, or a derivative, fragment, or variant of an antibody or immunoglobulin binds specifically. An "epitope" is antigenic, and thus the term "epitope" may also be referred to herein as "antigenic structure" or "antigenic determinant". Thus, the binding domain is an "antigen interaction site". The said binding / interaction is also understood to define "specific recognition".

[0053] An "epitope" can be formed by both contiguous amino acids or non-contiguous amino acids that are brought into proximity by the three-dimensional folding of a protein. A "linear epitope" is an epitope in which the primary sequence of amino acids contains the recognized epitope. Linear epitopes typically contain at least 3 or at least 4, and more commonly at least 5 or at least 6 or at least 7, for example about 8 to about 10 amino acids within a unique sequence.

[0054] In contrast to a linear epitope, a "conformational epitope" is an epitope in which the primary sequence of amino acids containing the epitope is not the only element defining the recognized epitope (e.g., an epitope in which the primary sequence of amino acids is not necessarily recognized by the binding domain). Generally, conformational epitopes contain a greater number of amino acids than linear epitopes. With respect to the recognition of conformational epitopes, the binding domain recognizes the three-dimensional structure of the antigen, preferably a peptide or protein or a fragment thereof (in the case of the present invention, the antigenic structure for the first binding domain is contained within the DLL3 protein). For example, when a protein molecule is folded to form a three-dimensional structure, the specific amino acids and / or polypeptide backbone forming the conformational epitope are brought into proximity, thereby enabling an antibody to recognize the epitope. Methods for determining the three-dimensional structure of an epitope include, but are not limited to, X-ray crystallography, two-dimensional nuclear magnetic resonance (2D-NMR) spectroscopy, site-directed spin labeling, and electron paramagnetic resonance (EPR) spectroscopy.

[0055] A method of epitope mapping is described below. When a region (adjacent amino acid chain) of the human DLL3 protein is replaced with the corresponding region of a DLL3 antigen other than human and non-primate (for example, mouse DLL3, but others such as chicken, rat, hamster, rabbit, etc. are also conceivable), a decrease in the binding of the binding domain is expected to occur unless the binding domain is cross-reactive with the non-human, non-primate DLL3 being used. The decrease is preferably at least 10%, 20%, 30%, 40%, or 50%, more preferably at least 60%, 70%, or 80%, most preferably 90%, 95%, or even 100% compared to the binding to the corresponding region in the human DLL3 protein, assuming the binding to the corresponding region of the human DLL3 protein is 100%. The above chimeras of human DLL3 and non-human DLL3 are assumed to be expressed in CHO cells. It is also assumed that chimeras of human DLL3 and non-human DLL3 are fused with transmembrane domains and / or cytoplasmic domains of different membrane-bound proteins such as EpCAM.

[0056] In alternative or additional methods of epitope mapping, several truncated forms of the human DLL3 extracellular domain can be generated to determine the specific regions recognized by the binding domain. In these truncated forms, different extracellular DLL3 domains / subdomains or regions are sequentially deleted starting from the N-terminus. The truncated DLL3s generated and used in connection with the present invention are shown in FIG. 1. It is contemplated that the truncated DLL3s are expressed in CHO cells. It is also contemplated that the truncated DLL3s are fused to the transmembrane domain and / or cytoplasmic domain of different membrane-bound proteins such as EpCAM. It is also contemplated that the truncated DLL3s contain a signal peptide domain at their N-terminus, for example a signal peptide derived from the mouse IgG heavy chain signal peptide. Further, the truncated DLL3s are contemplated to contain a v5 domain at the N-terminus (behind the signal peptide) so that accurate expression on the cell surface can be confirmed. In truncated DLL3s that no longer contain the DLL3 region recognized by the binding domain, a decrease or loss of binding is expected to occur. The decrease in binding is preferably at least 10%, 20%, 30%, 40%, or 50%, more preferably at least 60%, 70%, 80%, and most preferably 90%, 95%, or even 100% when the binding to the entire human DLL3 protein (or its extracellular region or domain) is taken as 100%. A method for testing this loss of binding is described in Example 2.

[0057] A further method for determining the contribution of specific residues of a target antigen to its recognition by an antibody construct or binding domain is an alanine scan, in which each residue to be analyzed is substituted with alanine, for example, by site-directed mutagenesis (see, for example, Morrison KL & Weiss GA. Cur Opin Chem Biol. 2001 Jun;5(3):302-7). Alanine is used because, despite mimicking the canonical secondary structure of many other amino acids, it is not bulky, is chemically inert, and has a methyl functional group. If it is desirable to conserve the size of the residue being mutated, bulkier amino acids such as valine or leucine can sometimes be used. Alanine scanning is a well-established technique that has been used for a long time.

[0058] The interaction between a binding domain and an epitope or a region containing the epitope means that the binding domain exhibits a measurable affinity for the epitope / region containing the epitope on a specific protein or antigen (herein, DLL3 and CD3, respectively), and generally does not show significant reactivity with proteins or antigens other than DLL3 or CD3. "Measurable affinity" includes bindings with an affinity of about 10 -6 M (KD) or stronger. Preferably, the binding affinity is about 10 -12 ~10 -8 M, 10 -12 ~10 -9 M, 10 -12 ~10 -10 M, 10 -11 ~10 -8 M, preferably about 10 -11 ~10 -9When it is M, the binding is regarded as specific. Whether the binding domain reacts or binds specifically to the target can be easily tested, in particular, by comparing the reaction of the binding domain to the target protein or antigen with the reaction of the binding domain to a protein or antigen other than DLL3 or CD3. Preferably, the binding domain of the present invention does not bind essentially or substantially to a protein or antigen other than DLL3 or CD3 (i.e., the first binding domain cannot bind to a protein other than DLL3, and the second binding domain cannot bind to a protein other than CD3).

[0059] The terms "does not bind essentially / substantially" or "cannot bind" mean that the binding domain of the present invention does not bind to a protein or antigen other than DLL3 or CD3, that is, when the binding to each of DLL3 or CD3 is taken as 100%, it does not show a reactivity of more than 30%, preferably more than 20%, more preferably more than 10%, particularly preferably more than 9%, 8%, 7%, 6%, or 5% to a protein or antigen other than DLL3 or CD3.

[0060] In addition, the antibody construct of the present invention is also assumed to bind to a human DLL3 isoform having one or both of the DLL3 point mutations of F172C and L218P. See Example 5.

[0061] Specific binding is considered to be brought about by specific motifs within the amino acid sequences of the binding domain and the antigen. Thus, the binding occurs as a result of the primary, secondary, and / or tertiary structures of those motifs, as well as the secondary modifications of each of the structures. The specific interaction between the antigen interaction site and its specific antigen can result in simple binding of the site to the antigen. Furthermore, the specific interaction between the antigen interaction site and its specific antigen can alternatively or additionally initiate a signal, for example, by inducing a conformational change of the antigen or oligomerization of the antigen.

[0062] The antibody construct according to the present invention binds to an epitope of DLL3 contained within the region shown in SEQ ID NO: 260, corresponding to the amino acid chain containing the EGF-3 and EGF-4 regions. Other anti-DLL3 binding substances were also prepared, and their DLL3 binding specificities were identified during epitope mapping (see Example 2).

[0063] The largest group of the binding substances produced recognized epitopes within the DSL domain. However, none of these antibody constructs met the criteria for sufficient cytotoxic activity in a 18-hour 51 cytotoxicity assay using Cr, with stimulated human CD8+ T cells as effector cells and human DLL3-transfected CHO cells as target cells.

[0064] Furthermore, in a 48-hour cytotoxicity assay by FACS (using unstimulated human PBMC as effector cells and human DLL3-transfected CHO cells as target cells), the binding substances that recognized the DLL3 epitope within the N-terminus of the protein showed EC50 values of 1455 - 1580 pM. In the current situation, this EC50 value is considered inappropriate for bispecific antibodies provided for therapeutic use in inducing the cytotoxic activity of the patient's immune system, more specifically T cells, against cancer cells.

[0065] Finally, another group of binding substances was prepared and their cytotoxic activities were characterized in various assays. From the epitope mapping of these binding substances, specificity for the DLL3 epitope contained within the EGF-5 region was revealed, and also to some extent within the EGF-6 region (see Example 2 for details). From the schematic diagrams of the data generated in different cytotoxicity assays (see Examples 8.3, 8.5, 8.6, and 8.7) for these binding substances named DLL3-18, DLL3-19, DLL3-20, and DLL3-21, it became clear that although none of the binding substances were inappropriate with regard to the effectiveness of the EC50 value in all assays, the entire group was clearly lower when compared to the antibody constructs according to the present invention. This observation is highlighted by the shading in the results shown in Tables 6-9.

[0066] In summary, it can be confidently stated that the antibody constructs according to the present invention (which bind to the DLL3 epitope contained within the region shown in SEQ ID NO: 260) exhibit significantly better activity compared to various other groups of DLL3 binding substances having different epitope specificities. In other words, the antibody constructs according to the present invention support potent bispecific antibody construct-mediated cytotoxic activity by showing an advantageous epitope-activity relationship.

[0067] In another aspect, the present invention provides a bispecific antibody construct comprising a first binding domain that binds to human DLL3 on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain binds to the DLL3 epitope contained within the region shown in SEQ ID NO: 258.

[0068] Preferably, the first binding domain of the bispecific antibody construct of the present invention comprises a VH region comprising CDR-H1, CDR-H2, and CDR-H3, and a VL region comprising CDR-L1, CDR-L2, and CDR-L3, selected from the group consisting of: a) CDR-H1 shown in SEQ ID NO: 31, CDR-H2 shown in SEQ ID NO: 32, CDR-H3 shown in SEQ ID NO: 33, CDR-L1 shown in SEQ ID NO: 34, CDR-L2 shown in SEQ ID NO: 35, and CDR-L3 shown in SEQ ID NO: 36; b) CDR-H1 shown in SEQ ID NO: 41, CDR-H2 shown in SEQ ID NO: 42, CDR-H3 shown in SEQ ID NO: 43, CDR-L1 shown in SEQ ID NO: 44, CDR-L2 shown in SEQ ID NO: 45, and CDR-L3 shown in SEQ ID NO: 46; c) CDR-H1 shown in SEQ ID NO: 51, CDR-H2 shown in SEQ ID NO: 52, CDR-H3 shown in SEQ ID NO: 53, CDR-L1 shown in SEQ ID NO: 54, CDR-L2 shown in SEQ ID NO: 55, and CDR-L3 shown in SEQ ID NO: 56; d) CDR-H1 shown in SEQ ID NO: 61, CDR-H2 shown in SEQ ID NO: 62, CDR-H3 shown in SEQ ID NO: 63, CDR-L1 shown in SEQ ID NO: 64, CDR-L2 shown in SEQ ID NO: 65, and CDR-L3 shown in SEQ ID NO: 66; e) CDR-H1 shown in SEQ ID NO: 71, CDR-H2 shown in SEQ ID NO: 72, CDR-H3 shown in SEQ ID NO: 73, CDR-L1 shown in SEQ ID NO: 74, CDR-L2 shown in SEQ ID NO: 75, and CDR-L3 shown in SEQ ID NO: 76; f) CDR-H1 shown in SEQ ID NO: 81, CDR-H2 shown in SEQ ID NO: 82, CDR-H3 shown in SEQ ID NO: 83, CDR-L1 shown in SEQ ID NO: 84, CDR-L2 shown in SEQ ID NO: 85, and CDR-L3 shown in SEQ ID NO: 86; g) CDR-H1 shown in SEQ ID NO: 91, CDR-H2 shown in SEQ ID NO: 92, CDR-H3 shown in SEQ ID NO: 93, CDR-L1 shown in SEQ ID NO: 94, CDR-L2 shown in SEQ ID NO: 95, and CDR-L3 shown in SEQ ID NO: 96; h) CDR-H1 shown in SEQ ID NO: 101, CDR-H2 shown in SEQ ID NO: 102, CDR-H3 shown in SEQ ID NO: 103, CDR-L1 shown in SEQ ID NO: 104, CDR-L2 shown in SEQ ID NO: 105, and CDR-L3 shown in SEQ ID NO: 106; and i) CDR-H1 shown in SEQ ID NO: 111, CDR-H2 shown in SEQ ID NO: 112, CDR-H3 shown in SEQ ID NO: 113, CDR-L1 shown in SEQ ID NO: 114, CDR-L2 shown in SEQ ID NO: 115, and CDR-L3 shown in SEQ ID NO: 116.

[0069] The term "variable" refers to a portion of an antibody or immunoglobulin domain (i.e., a "variable domain(s)") that exhibits variability within the sequence and is involved in determining the specificity and binding affinity of a particular antibody. The pair of variable heavy chain (VH) and variable light chain (VL) forms a single antigen-binding site together.

[0070] The variability is not uniformly distributed throughout the entire variable domain of the antibody, but rather is concentrated in subdomains of the variable regions of each of the heavy and light chains. These subdomains are called "hypervariable regions" or "complementarity-determining regions" (CDRs). The more conserved (i.e., non-hypervariable) portions of the variable domain are called "framework" regions (FRM or FR), and the six CDRs serve as a scaffold when forming the antigen-binding surface in three-dimensional space. The variable domains of the naturally occurring heavy and light chains each contain four FRM regions (FR1, FR2, FR3, and FR4), most of which adopt a β-sheet conformation, and these are connected by three hypervariable regions that form loop connections and in some cases form part of the β-sheet structure. The hypervariable regions of each chain are held as a cluster in close proximity by the FRM and, together with the hypervariable regions of the other chain, contribute to the formation of the antigen-binding site (see Kabat et al., supra).

[0071] The terms "CDR" and its plural form "CDRs" refer to complementarity-determining regions, three of which constitute the binding properties of the light chain variable region (CDR-L1, CDR-L2, and CDR-L3), and three of which constitute the binding properties of the heavy chain variable region (CDR-H1, CDR-H2, and CDR-H3). The CDRs contain most of the residues responsible for the specific interaction between the antibody and the antigen, thereby contributing to the functional activity of the antibody molecule. That is, this is the main determinant of antigen specificity.

[0072] The precise definition of the CDR boundaries and lengths follows the classification and numbering scheme. Thus, the CDRs can be represented by the Kabat, Chothia, contact, or any other boundary definition, including the numbering scheme described herein. Even though the boundaries are different, each of these schemes has some overlap in the regions that make up the so-called "hypervariable regions" within the variable array. Thus, the definition of the CDRs by each of these schemes may differ in length and in the boundary regions with adjacent framework regions. See, for example, Kabat (method based on interspecies sequence variability), Chothia (method based on crystallographic studies of antigen-antibody complexes), and / or MacCallum (Kabat et al., supra; Chothia et al., J. Mol. Biol, 1987, 196:901-917; and MacCallum et al., J. Mol. Biol, 1996, 262:732). Another standard for defining the characteristics of the antigen-binding site is the AbM definition used by Oxford Molecular's AbM antibody modeling software. See, for example, Protein Sequence and Structure Analysis of Antibody Variable Domains. In: Antibody Engineering Lab Manual (Ed.: Duebel, S. and Kontermann, R., Springer-Verlag, Heidelberg). Both can be combined to define a hybrid CDR only if the two residue identification techniques define overlapping rather than identical regions. However, numbering according to the so-called Kabat scheme is preferred.

[0073] Typically, CDRs form loop structures that can be classified into canonical structures. The term "canonical structure" refers to the backbone conformation adopted by the antigen-binding (CDR) loops. Comparative structural studies have found that five of the six antigen-binding loops have only a limited repertoire of available conformations. Each canonical structure may be characterized by the torsion angles of the polypeptide backbone. Thus, corresponding loops between antibodies can have very similar three-dimensional structures, even though a high degree of amino acid sequence variability is seen in most of each loop (Chothia and Lesk, J. Mol. Biol., 1987, 196:901; Chothia et al., Nature, 1989, 342:877; Martin and Thornton, J. Mol. Biol, 1996, 263:800). Furthermore, there is a relationship between the loop structure selected and the amino acid sequence around it. The conformation of a particular canonical class is determined by the length of the loop and by the amino acid residues present at important positions within that loop as well as within the conserved framework (i.e., outside the loop). Thus, assignment to a particular canonical class can be made based on the presence of these important amino acid residues.

[0074] The term "canonical structure" may also include considerations regarding the linear sequence of an antibody, as classified, for example, by Kabat (Kabat et al., supra). The Kabat numbering scheme is a standard widely adopted for numbering the amino acid residues of antibody variable domains in a consistent manner and is a preferred scheme for application in the present invention as described elsewhere herein. Still other structural considerations can be used to determine the canonical structure of an antibody. For example, differences not fully reflected by the Kabat numbering method can be described by the numbering scheme of Chothia et al., and / or revealed by other techniques, such as crystallography and two-dimensional or three-dimensional computer modeling. Thus, a given antibody sequence can be classified into canonical classes for which it is possible to identify appropriate chassis sequences, particularly in response to a desire, for example, to include various canonical structures in a library. The Kabat numbering method for the amino acid sequence of an antibody, and the structural considerations described by Chothia et al., (supra), and their involvement in constructing the canonical appearance of the antibody structure are described in the literature. The subunit structures and three-dimensional configurations of various 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.

[0075] CDR3 of the light chain and particularly CDR3 of the heavy chain can be the most important determinants in antigen binding within the light and heavy chain variable regions. In some antibody constructs, the heavy chain CDR3 is thought to be the major contact area between the antigen and the antibody. In vitro selection schemes that vary only CDR3 can be used to vary the binding properties of an antibody and to determine which residues contribute to antigen binding. Thus, CDR3 is usually the greatest source of molecular diversity within the antibody binding site. For example, H3 can be as short as about two amino acid residues or more than 26 amino acids.

[0076] In classical full-length antibodies or immunoglobulins, each light (L) chain is linked to the heavy (H) chain by one covalent disulfide bond, and at the same time, two H chains are linked to each other by one or more disulfide bonds depending on the H chain isotype. The CH domain closest to VH is usually referred to as CH1. The constant ("C") domains do not directly participate in antigen binding but exhibit various effector functions such as antibody-dependent, cell-mediated cytotoxicity, and complement activation. The Fc region of an antibody is contained within the heavy chain constant domains and can interact with the cell surface located, for example, on Fc receptors.

[0077] After assembly and somatic hypermutation, the sequences of antibody genes are extremely diverse, and these diversified genes are estimated to encode 10 10 different antibody molecules (Immunoglobulin Genes, 2 nd ed., eds. Jonio et al., Academic Press, San Diego, CA, 1995). The immune system thus provides the repertoire of immunoglobulins. The term "repertoire" refers to at least one nucleotide sequence, in whole or in part, derived from at least one sequence encoding at least one immunoglobulin. This sequence(s) can be generated by the in vivo rearrangement of the V, D, and J segments of the heavy chain, as well as the V and J segments of the light chain. Alternatively, this sequence(s) can be generated from cells in response to rearrangement-inducing stimuli, such as in vitro stimulation. Alternatively, some or all of this sequence(s) may be obtained by DNA splicing, nucleotide synthesis, mutagenesis, and other methods. (See, for example, U.S. Patent No. 5,565,332) The repertoire may contain only one sequence or multiple sequences, including those within a genetically diverse collection.

[0078] Preferred antibody constructs according to the present invention also include a first (preferably human) binding domain that binds to human DLL3 on the target cell surface and a second binding domain that binds to human CD3 on the T cell surface, wherein the first binding domain binds to the same DLL3 epitope as an antibody selected from the group consisting of DLL3-4, DLL3-5, DLL3-6, DLL3-7, DLL3-8, DLL3-9, and DLL3-10, i.e., an antibody comprising a VH region comprising CDR-H1, CDR-H2, and CDR-H3 selected from the group consisting of the following, and a VL region comprising CDR-L1, CDR-L2, and CDR-L3 can also be defined as: a) CDR-H1 shown in SEQ ID NO: 31, CDR-H2 shown in SEQ ID NO: 32, CDR-H3 shown in SEQ ID NO: 33, CDR-L1 shown in SEQ ID NO: 34, CDR-L2 shown in SEQ ID NO: 35, and CDR-L3 shown in SEQ ID NO: 36; b) CDR-H1 shown in SEQ ID NO: 41, CDR-H2 shown in SEQ ID NO: 42, CDR-H3 shown in SEQ ID NO: 43, CDR-L1 shown in SEQ ID NO: 44, CDR-L2 shown in SEQ ID NO: 45, and CDR-L3 shown in SEQ ID NO: 46; c) CDR-H1 shown in SEQ ID NO: 51, CDR-H2 shown in SEQ ID NO: 52, CDR-H3 shown in SEQ ID NO: 53, CDR-L1 shown in SEQ ID NO: 54, CDR-L2 shown in SEQ ID NO: 55, and CDR-L3 shown in SEQ ID NO: 56; d) CDR-H1 shown in SEQ ID NO: 61, CDR-H2 shown in SEQ ID NO: 62, CDR-H3 shown in SEQ ID NO: 63, CDR-L1 shown in SEQ ID NO: 64, CDR-L2 shown in SEQ ID NO: 65, and CDR-L3 shown in SEQ ID NO: 66; e) CDR-H1 shown in SEQ ID NO: 71, CDR-H2 shown in SEQ ID NO: 72, CDR-H3 shown in SEQ ID NO: 73, CDR-L1 shown in SEQ ID NO: 74, CDR-L2 shown in SEQ ID NO: 75, and CDR-L3 shown in SEQ ID NO: 76; f) CDR-H1 shown in SEQ ID NO: 81, CDR-H2 shown in SEQ ID NO: 82, CDR-H3 shown in SEQ ID NO: 83, CDR-L1 shown in SEQ ID NO: 84, CDR-L2 shown in SEQ ID NO: 85, and CDR-L3 shown in SEQ ID NO: 86; and g) CDR-H1 shown in SEQ ID NO: 91, CDR-H2 shown in SEQ ID NO: 92, CDR-H3 shown in SEQ ID NO: 93, CDR-L1 shown in SEQ ID NO: 94, CDR-L2 shown in SEQ ID NO: 95, and CDR-L3 shown in SEQ ID NO: 96.

[0079] Whether an antibody construct binds to the same DLL3 epitope as another given antibody construct can be measured, for example, by epitope mapping with a chimeric or truncated target molecule as described above in this specification and in Example 2.

[0080] Preferred antibody constructs according to the present invention also include a first (preferably human) binding domain that binds to human DLL3 on the target cell surface and a second binding domain that binds to human CD3 on the T cell surface, wherein the first binding domain competes for binding with an antibody selected from the group consisting of DLL3-4, DLL3-5, DLL3-6, DLL3-7, DLL3-8, DLL3-9, and DLL3-10, i.e., an antibody comprising a VH region comprising CDR-H1, CDR-H2, and CDR-H3 selected from the group consisting of the above, and a VL region comprising CDR-L1, CDR-L2, and CDR-L3.

[0081] Whether an antibody construct competes with another given antibody construct for binding can be measured by a competition assay such as a competitive ELISA or a competition assay using cells. It is also possible to use microparticles (beads) conjugated with avidin. Similar to an ELISA plate coated with avidin, each of these beads can be used as a substrate when reacting with a biotinylated protein, and an assay can be performed there. Coating the beads with an antigen and then precoating with a primary antibody. Add a secondary antibody to confirm whether any binding occurs. Read the measurement values by flow cytometry.

[0082] In one embodiment of the present invention, the first binding domain of the antibody construct of the present invention comprises a VH region selected from the group consisting of those shown in SEQ ID NO: 37, SEQ ID NO: 47, SEQ ID NO: 57, SEQ ID NO: 67, SEQ ID NO: 77, SEQ ID NO: 87, SEQ ID NO: 97, SEQ ID NO: 107, SEQ ID NO: 117, SEQ ID NO: 435, and SEQ ID NO: 529.

[0083] In a further embodiment of the present invention, the first binding domain of the antibody construct of the present invention comprises a VL region selected from the group consisting of those shown in SEQ ID NO: 38, SEQ ID NO: 48, SEQ ID NO: 58, SEQ ID NO: 68, SEQ ID NO: 78, SEQ ID NO: 88, SEQ ID NO: 98, SEQ ID NO: 108, SEQ ID NO: 118, SEQ ID NO: 436, and SEQ ID NO: 530.

[0084] In another embodiment, the first binding domain of the antibody construct of the present invention comprises a VH region and a VL region selected from the group consisting of pairs of VH regions and VL regions shown in SEQ ID NO: 37+38, SEQ ID NO: 47+48, SEQ ID NO: 57+58, SEQ ID NO: 67+68, SEQ ID NO: 77+78, SEQ ID NO: 87+88, SEQ ID NO: 97+98, SEQ ID NO: 107+108, SEQ ID NO: 117+118, SEQ ID NO: 435+436, and SEQ ID NO: 529+530.

[0085] In yet another embodiment of the present invention, the first binding domain of the antibody construct of the present invention comprises a polypeptide selected from the group consisting of those shown in SEQ ID NO: 39, SEQ ID NO: 49, SEQ ID NO: 59, SEQ ID NO: 69, SEQ ID NO: 79, SEQ ID NO: 89, SEQ ID NO: 99, SEQ ID NO: 109, SEQ ID NO: 119, SEQ ID NO: 437, and SEQ ID NO: 531.

[0086] The above-mentioned first binding domain (identified by its CDR, VH region, and VL region, and combinations thereof) is a binding domain characterized by binding to an epitope of DLL3 contained within the region shown in SEQ ID NO: 258.

[0087] As used herein, the term "bispecific" refers to an antibody construct that is "at least bispecific", i.e., an antibody construct that includes at least a first binding domain and a second binding domain, wherein the first binding domain binds to one antigen or target (DLL3 herein), and the second binding domain binds to another antigen or target (CD3 herein). Thus, the antibody constructs according to the present invention have specificity for at least two different antigens or targets. The term "bispecific antibody construct" of the present invention also encompasses multispecific antibody constructs, for example, trispecific antibody constructs containing three binding domains, or constructs having more than three (e.g., four, five, etc.) specificities.

[0088] When the antibody construct according to the present invention is (at least) bispecific, it does not occur in nature and is 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 & Lachmann, Clin. Exp. Immunol. 79:315-321 (1990).

[0089] At least two binding domains and variable domains of the antibody construct of the present invention may or may not contain a peptide linker (spacer peptide). According to the present invention, the term "peptide linker" includes an amino acid sequence that interconnects the amino acid sequences of one (variable and / or binding) domain and the other (variable and / or binding) domain of the antibody construct of the present invention. The essential technical feature of such a peptide linker is that the peptide linker does not have any polymerization activity. Particularly preferred peptide linkers are those described in U.S. Patent Nos. 4,751,180 and 4,935,233 or WO88 / 09344. The peptide linker can also be used to bind other domains or modules or regions (such as a half-life extension domain) to the antibody construct of the present invention.

[0090] When using a linker, this linker preferably has a sufficient length and sequence such that each of the first and second domains can reliably maintain different binding specificities independent of each other. The peptide linker that links at least two binding domains (or two variable domains) in the antibody construct of the present invention preferably contains only a very small number of amino acid residues, for example, 12 or fewer amino acid residues. Therefore, peptide linkers of 12, 11, 10, 9, 8, 7, 6, or 5 amino acid residues are preferred. The putative peptide linker of less than 5 amino acids contains 4, 3, 2, or 1 amino acid, and a Gly-rich linker is preferred. A particularly preferred "single" amino acid in relation to the "peptide linker" is Gly. Therefore, the peptide linker may consist of a single amino acid Gly. Another preferred embodiment of the peptide linker is characterized in that the amino acid sequence is Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 286), or a polymer thereof, i.e., (Gly4Ser)x, where x is an integer of 1 or greater (e.g., 2 or 3). Preferred linkers are shown in SEQ ID NOs: 285 to 293. The feature of the peptide linker that does not promote secondary structure is known in the art, for example, as described in Dall’Acqua et al. (Biochem. (1998) 37, 9266-9273), Cheadle et al. (Mol Immunol (1992) 29, 21-30), and Raag and Whitlow (FASEB (1995) 9(1), 73-80). A peptide linker that does not promote any secondary structure is preferred. The linkage between the domains can be obtained, for example, by genetic engineering as described in the examples. Methods for preparing a fused and functionally linked bispecific single-chain construct and expressing it in mammalian cells or bacteria are well known in the art (e.g., WO99 / 54440, or Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2001).

[0091] As described above in this specification, the present invention provides a preferred embodiment in which the antibody construct is in a form selected from the group consisting of (scFv)2, scFv single domain mAb, diabody, and oligomers of any of the foregoing forms. The term "in a form" does not exclude that the construct can be further modified, for example, by ligation or fusion with other moieties, as described herein.

[0092] According to certain preferred embodiments, as described in the attached examples, the antibody construct of the present invention is a "bispecific single-chain antibody construct", more preferably a bispecific "single-chain Fv" (scFv). The two domains of the Fv fragment, VL and VH, are encoded by separate genes, but as described above, they can be linked together by a synthetic linker using recombinant methods to form a single protein chain that forms a monovalent molecule as a pair of VL and VH regions (see, for example, Huston et al. (1988) Proc. Natl. Acad. Sci USA 85:5879-5883). This antibody fragment is obtained using conventional techniques known to those skilled in the art, and the function of the fragment is evaluated in the same manner as that of a whole antibody or full-length antibody. Thus, a single-chain variable fragment (scFv) is a fusion protein of the variable region of the heavy chain (VH) and the variable region of the light chain (VL) of an immunoglobulin, which are usually linked by a short linker peptide of about 10 to about 25 amino acids, preferably about 15 to 20 amino acids. The linker is usually rich in glycine to obtain flexibility and rich in serine or threonine to obtain solubility, and can link the N-terminus of VH to the C-terminus of VL, or the C-terminus of VH to the N-terminus of VL. This protein maintains the specificity of the original immunoglobulin despite removing the constant region and introducing a linker.

[0093] Bispecific single-chain molecules are known in the art and are described in 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, J. Mol. Biol., (1999), 293, 41-56. Techniques described for the production of single-chain antibodies (see in particular U.S. Patent No. 4,946,778, Kontermann and Dubel (2010), supra, and Little (2009), supra) can be adapted to the production of single-chain antibody constructs that specifically recognize selected target(s).

[0094] A bivalent (also called divalent) or bispecific single-chain variable fragment (bi-scFv or di-scFv having the form (scFv)2) can be designed by linking two scFv molecules (e.g., using the linker described above). When these two scFv molecules have the same binding specificity, the resulting (scFv)2 molecule is preferably called bivalent (i.e., having a valence of 2 for the same target epitope). When these two scFv molecules have different binding specificities, the resulting (scFv)2 molecule is preferably called bispecific. The linking can be done by creating a single peptide chain having two VH regions and two VL regions to generate a tandem scFv (e.g., Kufer P. et al., (2004) Trends in Biotechnology 22(5):238-244). Another possibility is to create scFv molecules using a linker peptide that is too short (e.g., about 5 amino acids) to fold the two variable regions together and dimerize the scFv. This type is known as a diabody (see, e.g., Hollinger, Philipp et al., (July 1993) Proceedings of the National Academy of Sciences of the United States of America 90(14):6444-8).

[0095] According to a further preferred embodiment of the antibody construct of the invention, the heavy chain (VH) and light chain (VL) of the binding domain (which binds to either target antigen DLL3 or CD3) are not directly linked by a peptide linker as described above, but form the binding domain as described for diabodies. Thus, the VH of the CD3 binding domain can be fused to the VL of the DLL3 binding domain by a peptide linker, and the VH of the DLL3 binding domain is fused to the VL of the CD3 binding domain by such a peptide linker.

[0096] A single-domain antibody contains only one (monomeric) antibody variable domain that can selectively bind to a specific antigen independently of other V regions or domains. The first single-domain antibodies were designed from heavy-chain antibodies found in camels, which are called V H H fragments. Cartilaginous fish also have heavy-chain antibodies (IgNAR) from which single-domain antibodies called V NAR fragments can be obtained. An alternative approach is to split the dimeric variable domains of common immunoglobulins from, for example, humans or rodents into monomers, thereby obtaining VH or VL as single-domain Abs. Most research on single-domain antibodies is currently based on heavy-chain variable domains, but nanobodies derived from light chains have also been shown to specifically bind to target epitopes. Examples of single-domain antibodies are called sdAbs, nanobodies, or single variable domain antibodies.

[0097] Thus, (single-domain mAb)2 is a monoclonal antibody construct composed of (at least) two single-domain monoclonal antibodies independently selected from the group comprising VH, VL, V H H, and V NAR . The linker is preferably in the form of a peptide linker. Similarly, "scFv-single-domain mAb" is a monoclonal antibody construct composed of at least one of the above single-domain antibodies and one of the above scFv molecules. Again, the linker is preferably in the form of a peptide linker.

[0098] Furthermore, the present invention contemplates providing a bispecific antibody construct comprising a first binding domain that binds to human DLL3 on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain binds to an epitope of DLL3 contained within the region shown in SEQ ID NO: 259.

[0099] Accordingly, in a further aspect of the invention, the first binding domain of the bispecific antibody construct comprises a VH region comprising CDR-H1, CDR-H2, and CDR-H3 selected from the group consisting of: and a VL region comprising CDR-L1, CDR-L2, and CDR-L3: a) CDR-H1 shown in SEQ ID NO: 121, CDR-H2 shown in SEQ ID NO: 122, CDR-H3 shown in SEQ ID NO: 123, CDR-L1 shown in SEQ ID NO: 124, CDR-L2 shown in SEQ ID NO: 125, and CDR-L3 shown in SEQ ID NO: 126; b) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; c) CDR-H1 shown in SEQ ID NO: 141, CDR-H2 shown in SEQ ID NO: 142, CDR-H3 shown in SEQ ID NO: 143, CDR-L1 shown in SEQ ID NO: 144, CDR-L2 shown in SEQ ID NO: 145, and CDR-L3 shown in SEQ ID NO: 146; d) CDR-H1 shown in SEQ ID NO: 151, CDR-H2 shown in SEQ ID NO: 152, CDR-H3 shown in SEQ ID NO: 153, CDR-L1 shown in SEQ ID NO: 154, CDR-L2 shown in SEQ ID NO: 155, and CDR-L3 shown in SEQ ID NO: 156; and e) CDR-H1 shown in SEQ ID NO: 161, CDR-H2 shown in SEQ ID NO: 162, CDR-H3 shown in SEQ ID NO: 163, CDR-L1 shown in SEQ ID NO: 164, CDR-L2 shown in SEQ ID NO: 165, and CDR-L3 shown in SEQ ID NO: 166; f) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 439, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; g) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 440, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; h) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 441, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; i) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 442, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; j) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 443, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; k) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 444, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; l) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 439, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 441, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; and m) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 440, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 442, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136.

[0100] In one embodiment, the first binding domain of the antibody construct of the present invention comprises a VH region selected from the group consisting of those set forth in SEQ ID NO: 127, SEQ ID NO: 137, SEQ ID NO: 147, SEQ ID NO: 157, SEQ ID NO: 167, SEQ ID NO: 445, SEQ ID NO: 446, SEQ ID NO: 447, SEQ ID NO: 448, SEQ ID NO: 449, SEQ ID NO: 450, SEQ ID NO: 451, SEQ ID NO: 452, SEQ ID NO: 453, SEQ ID NO: 454, and SEQ ID NO: 455.

[0101] In yet another embodiment, the first binding domain of the antibody construct of the present invention comprises a VL region selected from the group consisting of those set forth in SEQ ID NO: 128, SEQ ID NO: 138, SEQ ID NO: 148, SEQ ID NO: 158, SEQ ID NO: 168, SEQ ID NO: 456, SEQ ID NO: 457, SEQ ID NO: 458, SEQ ID NO: 459, SEQ ID NO: 460, SEQ ID NO: 461, SEQ ID NO: 462, SEQ ID NO: 463, SEQ ID NO: 464, SEQ ID NO: 465, SEQ ID NO: 466, SEQ ID NO: 467, SEQ ID NO: 468, SEQ ID NO: 469, and SEQ ID NO: 470.

[0102] In another embodiment, the first binding domain of the antibody construct of the present invention comprises a VH region and a VL region selected from the group consisting of pairs of VH and VL regions set forth in SEQ ID NO: 127 + 128, SEQ ID NO: 137 + 138, SEQ ID NO: 147 + 148, SEQ ID NO: 157 + 158, SEQ ID NO: 167 + 168, SEQ ID NO: 137 + 456, SEQ ID NO: 137 + 457, SEQ ID NO: 137 + 458, SEQ ID NO: 137 + 459, SEQ ID NO: 137 + 460, SEQ ID NO: 445 + 138, SEQ ID NO: 446 + 138, SEQ ID NO: 447 + 138, SEQ ID NO: 445 + 460, SEQ ID NO: 448 + 461, SEQ ID NO: 449 + 462, SEQ ID NO: 450 + 463, SEQ ID NO: 450 + 464, SEQ ID NO: 450 + 465, SEQ ID NO: 450 + 466, SEQ ID NO: 450 + 467, SEQ ID NO: 450 + 468, SEQ ID NO: 451 + 463, SEQ ID NO: 452 + 463, SEQ ID NO: 453 + 463, SEQ ID NO: 451 + 468, SEQ ID NO: 454 + 469, and SEQ ID NO: 455 + 470.

[0103] In yet another embodiment, the first binding domain of the antibody construct of the present invention comprises a polypeptide selected from the group consisting of those set forth in SEQ ID NO: 129, SEQ ID NO: 139, SEQ ID NO: 149, SEQ ID NO: 159, SEQ ID NO: 169, SEQ ID NO: 471, SEQ ID NO: 472, SEQ ID NO: 473, SEQ ID NO: 474, SEQ ID NO: 475, SEQ ID NO: 476, SEQ ID NO: 477, SEQ ID NO: 478, SEQ ID NO: 479, SEQ ID NO: 480, SEQ ID NO: 481, SEQ ID NO: 482, SEQ ID NO: 483, SEQ ID NO: 484, SEQ ID NO: 485, SEQ ID NO: 486, SEQ ID NO: 487, SEQ ID NO: 488, SEQ ID NO: 489, SEQ ID NO: 490, SEQ ID NO: 491, SEQ ID NO: 492, and SEQ ID NO: 493.

[0104] The first binding domain (identified by its CDRs, VH region, and VL region, and combinations thereof) is a binding domain characterized by binding to an epitope of DLL3 contained within the region shown in SEQ ID NO: 259.

[0105] Another preferred antibody construct according to the present invention also comprises a first (preferably human) binding domain that binds to human DLL3 on the target cell surface and a second binding domain that binds to human CD3 on the T cell surface, wherein the first binding domain binds to the same DLL3 epitope as an antibody selected from the group consisting of DLL3-13, DLL3-14, and DLL3-15, i.e., an antibody comprising a VH region comprising CDR-H1, CDR-H2, and CDR-H3 and a VL region comprising CDR-L1, CDR-L2, and CDR-L3, which can also be defined as an antibody selected from the group consisting of: a) CDR-H1 shown in SEQ ID NO: 121, CDR-H2 shown in SEQ ID NO: 122, CDR-H3 shown in SEQ ID NO: 123, CDR-L1 shown in SEQ ID NO: 124, CDR-L2 shown in SEQ ID NO: 125, and CDR-L3 shown in SEQ ID NO: 126; b) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; and c) CDR-H1 shown in SEQ ID NO: 141, CDR-H2 shown in SEQ ID NO: 142, CDR-H3 shown in SEQ ID NO: 143, CDR-L1 shown in SEQ ID NO: 144, CDR-L2 shown in SEQ ID NO: 145, and CDR-L3 shown in SEQ ID NO: 146.

[0106] Another preferred antibody construct according to the present invention also includes a first (preferably human) binding domain that binds to human DLL3 on the target cell surface and a second binding domain that binds to human CD3 on the T cell surface, and the first binding domain competes for binding with an antibody selected from the group consisting of DLL3-13, DLL3-14, and DLL3-15, that is, a VH region comprising CDR-H1, CDR-H2, and CDR-H3 selected from the group consisting of the above, and a VL region comprising CDR-L1, CDR-L2, and CDR-L3. It can also be defined as an antibody.

[0107] In addition to the function of binding to the target molecules DLL3 and CD3, the antibody constructs of the present invention are also envisioned to have additional functions. The antibody construct in this form targets target cells by binding to DLL3, mediates the activity of cytotoxic T cells by binding to CD3, and has additional functions, such as a fully functional Fc constant domain that mediates antibody-dependent cell cytotoxicity by mobilizing effector cells such as NK cells, a label (such as fluorescence), a therapeutic agent such as a toxin or a radionuclide, and / or means to extend the serum half-life, etc. It is a trifunctional or multifunctional antibody construct.

[0108] Examples of means for extending the serum half-life of the antibody constructs of the present invention include peptides, proteins, or protein domains fused to or otherwise conjugated to the antibody constructs. The group of peptides, proteins, or protein domains includes peptides that bind to other proteins that exhibit favorable pharmacokinetic properties in the human body, such as serum albumin (see WO2009 / 127691). As an alternative concept to such half-life extending peptides, peptides that bind to the neonatal Fc receptor (FcRn, see WO2007 / 098420) are mentioned, which can also be used in the constructs of the present invention. As concepts for binding large domains of proteins or complete proteins, for example, fusion of human serum albumin, variants or mutants of human serum albumin (see WO2011 / 051489, WO2012 / 059486, WO2012 / 150319, WO2013 / 135896, WO2014 / 072481, WO2013 / 075066) or fusions of their domains, as well as fusions of the constant regions (Fc domains) of immunoglobulins and their variants are mentioned. Such variants of the Fc domain can be optimized / modified to remove Fc receptor binding (e.g., Fcγ receptor) or to enable desirable dimer or multimer pairing for other reasons. A further concept known in the art for extending the half-life of small protein compounds in the human body is the PEGylation of small protein compounds such as the antibody constructs of the present invention.

[0109] In a preferred embodiment, the bispecific antibody construct according to the present invention can be linked to a fusion partner (such as a protein or polypeptide or peptide, for example by peptide binding) for the purpose of, for example, extending the serum half-life of the construct. This fusion partner can be selected from human serum albumin ("HSA" or "HALB"), as well as sequence variants thereof, peptides that bind to HSA, peptides that bind to FcRn ("FcRn BP"), or constructs containing the (antibody-derived) Fc region. Exemplary sequences of this fusion partner are shown in SEQ ID NOs: 295 to 341. Generally, the fusion partner can be directly (e.g., by peptide binding), or (GGGGS)n (In the formula, "n" is an integer of 2 or more, for example, 2 or 3 or 4), and can be linked to the N-terminus or C-terminus of the bispecific antibody construct according to the present invention via a peptide linker such as this. Preferred peptide linkers are shown in SEQ ID NOs: 285 to 293.

[0110] Therefore, preferred antibody constructs according to the present invention include the following.

[0111] (a) A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 39, SEQ ID NO: 49, SEQ ID NO: 59, SEQ ID NO: 69, SEQ ID NO: 79, SEQ ID NO: 89, SEQ ID NO: 99, SEQ ID NO: 109, SEQ ID NO: 119, SEQ ID NO: 129, SEQ ID NO: 139, SEQ ID NO: 149, SEQ ID NO: 159, SEQ ID NO: 169, SEQ ID NO: 437, SEQ ID NO: 471, SEQ ID NO: 472, SEQ ID NO: 473, SEQ ID NO: 474, SEQ ID NO: 475, SEQ ID NO: 476, SEQ ID NO: 477, SEQ ID NO: 478, SEQ ID NO: 479, SEQ ID NO: 480, SEQ ID NO: 481, SEQ ID NO: 482, SEQ ID NO: 483, SEQ ID NO: 484, SEQ ID NO: 485, SEQ ID NO: 486, SEQ ID NO: 487, SEQ ID NO: 488, SEQ ID NO: 489, SEQ ID NO: 490, SEQ ID NO: 491, SEQ ID NO: 492, SEQ ID NO: 493, and SEQ ID NO: 531; · A peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 285 to 293; and · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 350, SEQ ID NO: 359, SEQ ID NO: 368, SEQ ID NO: 377, SEQ ID NO: 386, SEQ ID NO: 395, SEQ ID NO: 404, SEQ ID NO: 413, SEQ ID NO: 422, SEQ ID NO: 431, and SEQ ID NO: 434; and · Optionally, a His tag such as that described in SEQ ID NO: 294;

[0112] (b) A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 39, SEQ ID NO: 49, SEQ ID NO: 59, SEQ ID NO: 69, SEQ ID NO: 79, SEQ ID NO: 89, SEQ ID NO: 99, SEQ ID NO: 109, SEQ ID NO: 119, SEQ ID NO: 129, SEQ ID NO: 139, SEQ ID NO: 149, SEQ ID NO: 159, SEQ ID NO: 169, SEQ ID NO: 437, SEQ ID NO: 471, SEQ ID NO: 472, SEQ ID NO: 473, SEQ ID NO: 474, SEQ ID NO: 475, SEQ ID NO: 476, SEQ ID NO: 477, SEQ ID NO: 478, SEQ ID NO: 479, SEQ ID NO: 480, SEQ ID NO: 481, SEQ ID NO: 482, SEQ ID NO: 483, SEQ ID NO: 484, SEQ ID NO: 485, SEQ ID NO: 486, SEQ ID NO: 487, SEQ ID NO: 488, SEQ ID NO: 489, SEQ ID NO: 490, SEQ ID NO: 491, SEQ ID NO: 492, SEQ ID NO: 493, and SEQ ID NO: 531; · A peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 285 to 293; · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 350, SEQ ID NO: 359, SEQ ID NO: 368, SEQ ID NO: 377, SEQ ID NO: 386, SEQ ID NO: 395, SEQ ID NO: 404, SEQ ID NO: 413, SEQ ID NO: 422, SEQ ID NO: 431, and SEQ ID NO: 434; · Optionally, a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 285 to 293; · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 295 and SEQ ID NOs: 301 to 330; and · Optionally, a His tag such as that described in SEQ ID NO: 294;

[0113] (c) A polypeptide comprising the following in the following order from the N-terminus: · An amino acid sequence where X1 is Y or H A polypeptide having TIFF2025106535000001.tif5128; and · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 39, SEQ ID NO: 49, SEQ ID NO: 59, SEQ ID NO: 69, SEQ ID NO: 79, SEQ ID NO: 89, SEQ ID NO: 99, SEQ ID NO: 109, SEQ ID NO: 119, SEQ ID NO: 129, SEQ ID NO: 139, SEQ ID NO: 149, SEQ ID NO: 159, SEQ ID NO: 169, SEQ ID NO: 437, SEQ ID NO: 471, SEQ ID NO: 472, SEQ ID NO: 473, SEQ ID NO: 474, SEQ ID NO: 475, SEQ ID NO: 476, SEQ ID NO: 477, SEQ ID NO: 478, SEQ ID NO: 479, SEQ ID NO: 480, SEQ ID NO: 481, SEQ ID NO: 482, SEQ ID NO: 483, SEQ ID NO: 484, SEQ ID NO: 485, SEQ ID NO: 486, SEQ ID NO: 487, SEQ ID NO: 488, SEQ ID NO: 489, SEQ ID NO: 490, SEQ ID NO: 491, SEQ ID NO: 492, SEQ ID NO: 493, and SEQ ID NO: 531; · A peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 285 to 293; · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 350, SEQ ID NO: 359, SEQ ID NO: 368, SEQ ID NO: 377, SEQ ID NO: 386, SEQ ID NO: 395, SEQ ID NO: 404, SEQ ID NO: 413, SEQ ID NO: 422, SEQ ID NO: 431, and SEQ ID NO: 434; · An amino acid sequence · A polypeptide having TIFF2025106535000002.tif5158; and · Optionally, a His tag such as that described in SEQ ID NO: 294;

[0114] (d) A polypeptide comprising the following in the following order from the N-terminus · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 348, SEQ ID NO: 357, SEQ ID NO: 366, SEQ ID NO: 375, SEQ ID NO: 384, SEQ ID NO: 393, SEQ ID NO: 402, SEQ ID NO: 411, SEQ ID NO: 420, SEQ ID NO: 429, and SEQ ID NO: 432; · A peptide linker having the amino acid sequence described in SEQ ID NO: 292; · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 38, SEQ ID NO: 48, SEQ ID NO: 58, SEQ ID NO: 68, SEQ ID NO: 78, SEQ ID NO: 88, SEQ ID NO: 98, SEQ ID NO: 108, SEQ ID NO: 118, SEQ ID NO: 128, SEQ ID NO: 138, SEQ ID NO: 148, SEQ ID NO: 158, SEQ ID NO: 168, SEQ ID NO: 436, SEQ ID NO: 456, SEQ ID NO: 457, SEQ ID NO: 458, SEQ ID NO: 459, SEQ ID NO: 460, SEQ ID NO: 461, SEQ ID NO: 462, SEQ ID NO: 463, SEQ ID NO: 464, SEQ ID NO: 465, SEQ ID NO: 466, SEQ ID NO: 467, SEQ ID NO: 468, SEQ ID NO: 469, SEQ ID NO: 470, and SEQ ID NO: 530, with a serine residue following on the C-terminal side; · A polypeptide having the amino acid sequence set forth in SEQ ID NO: 331; and A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 37, SEQ ID NO: 47, SEQ ID NO: 57, SEQ ID NO: 67, SEQ ID NO: 77, SEQ ID NO: 87, SEQ ID NO: 97, SEQ ID NO: 107, SEQ ID NO: 117, SEQ ID NO: 127, SEQ ID NO: 137, SEQ ID NO: 147, SEQ ID NO: 157, SEQ ID NO: 167, SEQ ID NO: 435, SEQ ID NO: 445, SEQ ID NO: 446, SEQ ID NO: 447, SEQ ID NO: 448, SEQ ID NO: 449, SEQ ID NO: 450, SEQ ID NO: 451, SEQ ID NO: 452, SEQ ID NO: 453, SEQ ID NO: 454, and SEQ ID NO: 455, and SEQ ID NO: 529; · A peptide linker having the amino acid sequence set forth in SEQ ID NO: 292; · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 349, SEQ ID NO: 358, SEQ ID NO: 367, SEQ ID NO: 376, SEQ ID NO: 385, SEQ ID NO: 394, SEQ ID NO: 403, SEQ ID NO: 412, SEQ ID NO: 421, SEQ ID NO: 430, and SEQ ID NO: 433, with a serine residue following on the C-terminal side; and · A polypeptide having the amino acid sequence set forth in SEQ ID NO: 332;

[0115] (e) A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 348, SEQ ID NO: 357, SEQ ID NO: 366, SEQ ID NO: 375, SEQ ID NO: 384, SEQ ID NO: 393, SEQ ID NO: 402, SEQ ID NO: 411, SEQ ID NO: 420, SEQ ID NO: 429, and SEQ ID NO: 432; · A peptide linker having the amino acid sequence set forth in SEQ ID NO: 292; · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 38, SEQ ID NO: 48, SEQ ID NO: 58, SEQ ID NO: 68, SEQ ID NO: 78, SEQ ID NO: 88, SEQ ID NO: 98, SEQ ID NO: 108, SEQ ID NO: 118, SEQ ID NO: 128, SEQ ID NO: 138, SEQ ID NO: 148, SEQ ID NO: 158, SEQ ID NO: 168, SEQ ID NO: 436, SEQ ID NO: 456, SEQ ID NO: 457, SEQ ID NO: 458, SEQ ID NO: 459, SEQ ID NO: 460, SEQ ID NO: 461, SEQ ID NO: 462, SEQ ID NO: 463, SEQ ID NO: 464, SEQ ID NO: 465, SEQ ID NO: 466, SEQ ID NO: 467, SEQ ID NO: 468, SEQ ID NO: 469, SEQ ID NO: 470, and SEQ ID NO: 530; and · A polypeptide having the amino acid sequence set forth in SEQ ID NO: 333; and · A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 37, SEQ ID NO: 47, SEQ ID NO: 57, SEQ ID NO: 67, SEQ ID NO: 77, SEQ ID NO: 87, SEQ ID NO: 97, SEQ ID NO: 107, SEQ ID NO: 117, SEQ ID NO: 127, SEQ ID NO: 137, SEQ ID NO: 147, SEQ ID NO: 157, and SEQ ID NO: 167, SEQ ID NO: 435, SEQ ID NO: 445, SEQ ID NO: 446, SEQ ID NO: 447, SEQ ID NO: 448, SEQ ID NO: 449, SEQ ID NO: 450, SEQ ID NO: 451, SEQ ID NO: 452, SEQ ID NO: 453, SEQ ID NO: 454, SEQ ID NO: 455, and SEQ ID NO: 529; · A peptide linker having the amino acid sequence set forth in SEQ ID NO: 292; · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 349, SEQ ID NO: 358, SEQ ID NO: 367, SEQ ID NO: 376, SEQ ID NO: 385, SEQ ID NO: 394, SEQ ID NO: 403, SEQ ID NO: 412, SEQ ID NO: 421, SEQ ID NO: 430, and SEQ ID NO: 433, with a serine residue following on the C-terminal side; and · A polypeptide having the amino acid sequence set forth in SEQ ID NO: 334;

[0116] (f) A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 39, SEQ ID NO: 49, SEQ ID NO: 59, SEQ ID NO: 69, SEQ ID NO: 79, SEQ ID NO: 89, SEQ ID NO: 99, SEQ ID NO: 109, SEQ ID NO: 119, SEQ ID NO: 129, SEQ ID NO: 139, SEQ ID NO: 149, SEQ ID NO: 159, SEQ ID NO: 169, SEQ ID NO: 437, SEQ ID NO: 471, SEQ ID NO: 472, SEQ ID NO: 473, SEQ ID NO: 474, SEQ ID NO: 475, SEQ ID NO: 476, SEQ ID NO: 477, SEQ ID NO: 478, SEQ ID NO: 479, SEQ ID NO: 480, SEQ ID NO: 481, SEQ ID NO: 482, SEQ ID NO: 483, SEQ ID NO: 484, SEQ ID NO: 485, SEQ ID NO: 486, SEQ ID NO: 487, SEQ ID NO: 488, SEQ ID NO: 489, SEQ ID NO: 490, SEQ ID NO: 491, SEQ ID NO: 492, SEQ ID NO: 493, and SEQ ID NO: 531; · A peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 285 to 293; · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 350, SEQ ID NO: 359, SEQ ID NO: 368, SEQ ID NO: 377, SEQ ID NO: 386, SEQ ID NO: 395, SEQ ID NO: 404, SEQ ID NO: 413, SEQ ID NO: 422, SEQ ID NO: 431, and SEQ ID NO: 434; · A polypeptide having the amino acid sequence set forth in SEQ ID NO: 335; and A polypeptide having the amino acid sequence set forth in SEQ ID NO: 336;

[0117] (g) A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 39, SEQ ID NO: 49, SEQ ID NO: 59, SEQ ID NO: 69, SEQ ID NO: 79, SEQ ID NO: 89, SEQ ID NO: 99, SEQ ID NO: 109, SEQ ID NO: 119, SEQ ID NO: 129, SEQ ID NO: 139, SEQ ID NO: 149, SEQ ID NO: 159, SEQ ID NO: 169, SEQ ID NO: 437, SEQ ID NO: 471, SEQ ID NO: 472, SEQ ID NO: 473, SEQ ID NO: 474, SEQ ID NO: 475, SEQ ID NO: 476, SEQ ID NO: 477, SEQ ID NO: 478, SEQ ID NO: 479, SEQ ID NO: 480, SEQ ID NO: 481, SEQ ID NO: 482, SEQ ID NO: 483, SEQ ID NO: 484, SEQ ID NO: 485, SEQ ID NO: 486, SEQ ID NO: 487, SEQ ID NO: 488, SEQ ID NO: 489, SEQ ID NO: 490, SEQ ID NO: 491, SEQ ID NO: 492, SEQ ID NO: 493, and SEQ ID NO: 531; and · A polypeptide having the amino acid sequence set forth in SEQ ID NO: 337; and A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 350, SEQ ID NO: 359, SEQ ID NO: 368, SEQ ID NO: 377, SEQ ID NO: 386, SEQ ID NO: 395, SEQ ID NO: 404, SEQ ID NO: 413, SEQ ID NO: 422, SEQ ID NO: 431, and SEQ ID NO: 434; and · A polypeptide having the amino acid sequence set forth in SEQ ID NO: 338;

[0118] (h) A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 39, SEQ ID NO: 49, SEQ ID NO: 59, SEQ ID NO: 69, SEQ ID NO: 79, SEQ ID NO: 89, SEQ ID NO: 99, SEQ ID NO: 109, SEQ ID NO: 119, SEQ ID NO: 129, SEQ ID NO: 139, SEQ ID NO: 149, SEQ ID NO: 159, SEQ ID NO: 169, SEQ ID NO: 437, SEQ ID NO: 471, SEQ ID NO: 472, SEQ ID NO: 473, SEQ ID NO: 474, SEQ ID NO: 475, SEQ ID NO: 476, SEQ ID NO: 477, SEQ ID NO: 478, SEQ ID NO: 479, SEQ ID NO: 480, SEQ ID NO: 481, SEQ ID NO: 482, SEQ ID NO: 483, SEQ ID NO: 484, SEQ ID NO: 485, SEQ ID NO: 486, SEQ ID NO: 487, SEQ ID NO: 488, SEQ ID NO: 489, SEQ ID NO: 490, SEQ ID NO: 491, SEQ ID NO: 492, SEQ ID NO: 493, and SEQ ID NO: 531; and · A polypeptide having the amino acid sequence set forth in SEQ ID NO: 339; and A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 350, SEQ ID NO: 359, SEQ ID NO: 368, SEQ ID NO: 377, SEQ ID NO: 386, SEQ ID NO: 395, SEQ ID NO: 404, SEQ ID NO: 413, SEQ ID NO: 422, SEQ ID NO: 431, and SEQ ID NO: 434; and · A polypeptide having the amino acid sequence set forth in SEQ ID NO: 340; or

[0119] (i) A polypeptide comprising the following in the following order from the N-terminus: · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 39, SEQ ID NO: 49, SEQ ID NO: 59, SEQ ID NO: 69, SEQ ID NO: 79, SEQ ID NO: 89, SEQ ID NO: 99, SEQ ID NO: 109, SEQ ID NO: 119, SEQ ID NO: 129, SEQ ID NO: 139, SEQ ID NO: 149, SEQ ID NO: 159, SEQ ID NO: 169, SEQ ID NO: 437, SEQ ID NO: 471, SEQ ID NO: 472, SEQ ID NO: 473, SEQ ID NO: 474, SEQ ID NO: 475, SEQ ID NO: 476, SEQ ID NO: 477, SEQ ID NO: 478, SEQ ID NO: 479, SEQ ID NO: 480, SEQ ID NO: 481, SEQ ID NO: 482, SEQ ID NO: 483, SEQ ID NO: 484, SEQ ID NO: 485, SEQ ID NO: 486, SEQ ID NO: 487, SEQ ID NO: 488, SEQ ID NO: 489, SEQ ID NO: 490, SEQ ID NO: 491, SEQ ID NO: 492, SEQ ID NO: 493, and SEQ ID NO: 531; · A peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 285 to 293; · A polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NO: 350, SEQ ID NO: 359, SEQ ID NO: 368, SEQ ID NO: 377, SEQ ID NO: 386, SEQ ID NO: 395, SEQ ID NO: 404, SEQ ID NO: 413, SEQ ID NO: 422, SEQ ID NO: 431, and SEQ ID NO: 434; and · A polypeptide having the amino acid sequence set forth in SEQ ID NO: 341.

[0120] For example, a preferred bispecific antibody construct of the present invention comprises or consists of a polypeptide selected from the group consisting of the following: SEQ ID NO: 224, SEQ ID NO: 225, SEQ ID NO: 226, SEQ ID NO: 227, SEQ ID NO: 228, SEQ ID NO: 229, SEQ ID NO: 230, SEQ ID NO: 231, SEQ ID NO: 232, SEQ ID NO: 233, SEQ ID NO: 234, SEQ ID NO: 235, SEQ ID NO: 236, SEQ ID NO: 237, SEQ ID NO: 242, SEQ ID NO: 243, SEQ ID NO: 244, SEQ ID NO: 245, SEQ ID NO: 246, SEQ ID NO: 247, SEQ ID NO: 248, SEQ ID NO: 249, SEQ ID NO: 250, and SEQ ID NO: 251.

[0121] As described above, some preferred antibody constructs of the present invention are modified by fusion with another moiety such as albumin or an albumin variant. Those skilled in the art will understand that if the properties of such fusion constructs, such as target affinity or cytotoxic activity, are characterized, it can be predicted that similar fusion constructs or unmodified bispecific antibody constructs will have similar (or perhaps even better) properties. For example, if a bispecific antibody construct fused with albumin has measurable or desirable cytotoxic activity or target affinity, it can be predicted that similar constructs without albumin will also exhibit the same, similar, or even higher cytotoxic activity / target affinity.

[0122] According to another preferred embodiment, the bispecific antibody construct of the present invention includes a third domain (in addition to the two binding domains), which comprises two polypeptide monomers each containing a hinge, CH2, and CH3 domains, and the two polypeptides (or polypeptide monomers) are fused to each other via a peptide linker. Preferably, the third domain includes hinge-CH2-CH3-linker-hinge-CH2-CH3 in order from the N-terminus to the C-terminus. The preferred amino acid sequences of the third domain are shown in SEQ ID NOs: 541 to 548. Each of the polypeptide monomers preferably has an amino acid sequence selected from the group consisting of SEQ ID NOs: 533 to 540, or an amino acid sequence that is at least 90% identical to those sequences. In another preferred embodiment, the first and second binding domains of the bispecific antibody construct of the present invention are fused to the third domain via a peptide linker selected from the group consisting of, for example, SEQ ID NOs: 285, 286, 288, 289, 290, 292, and 293.

[0123] According to the present invention, the "hinge" is the IgG hinge region. This region can be identified by analogy using the Kabat numbering method (see Kabat positions 223-243). According to the above, the minimum requirement for the "hinge" is the amino acid residues corresponding to the IgG1 sequence chain of D231-P243 by the Kabat numbering method. The terms CH2 and CH3 mean the immunoglobulin heavy chain constant regions 2 and 3. These regions can be identified by analogy using the Kabat numbering method (see Kabat positions 244-360 for CH2 and Kabat positions 361-478 for CH3). It has been found that there are some variations among immunoglobulins with respect to 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 (see, for example, Padlan, Molecular Immunology, 31(3), 169-217 (1993)). The term Fc monomer refers to the last two heavy chain constant regions of IgA, IgD, and IgG, and the last three heavy chain constant regions of IgE and IgM. The Fc monomer can also contain a flexible hinge at the N-terminus of these domains. In the case of IgA and IgM, the Fc monomer can contain a J chain. In the case of IgG, the Fc portion contains the immunoglobulin domains CH2 and CH3 and a hinge between the first two domains and CH2. Although the boundaries of the Fc portion of immunoglobulins can vary, in the example of the human IgG heavy chain Fc portion containing the functional hinge, CH2, and CH3 domains, for example, in the case of IgG4, it can be defined as containing residues D231-(the C-terminus of the CH3 domain) P476, or D231-L476 (numbering according to Kabat).

[0124] Thus, the antibody construct of the present invention can comprise, in order from the N-terminus to the C-terminus: (a) a first binding domain; (b) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 286, 292, and 293; (c) a second binding domain; (d) A peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 285, 286, 288, 289, 290, 292, and 293; (e) The first polypeptide monomer of the third domain (including the hinge, CH2, and CH3 domains); (f) A peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 550, 551, 552, and 553; and (g) The second polypeptide monomer of the third domain (including the hinge, CH2, and CH3 domains).

[0125] The antibody construct of the present invention preferably further comprises the following in order from the N-terminus to the C-terminus: · A first binding domain having an amino acid sequence selected from the group consisting of SEQ ID NO: 39, 49, 59, 69, 79, 89, 99, 109, 119, 129, 139, 149, 159, 169, 437, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, and 531; · A peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 286, 292, and 293; · A second binding domain having an amino acid sequence selected from the group consisting of SEQ ID NO: 350, 359, 368, 377, 386, 395, 404, 413, 422, 431, and 434; · A peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 285, 286, 288, 289, 290, 292, and 293; · A third domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 541 to 548.

[0126] Thus, in a preferred embodiment, the antibody construct of the present invention comprises or consists of a polypeptide selected from the group consisting of those shown in SEQ ID NO: 517, SEQ ID NO: 518, SEQ ID NO: 519, SEQ ID NO: 520, SEQ ID NO: 521, SEQ ID NO: 522, SEQ ID NO: 523, SEQ ID NO: 524, SEQ ID NO: 525, SEQ ID NO: 526, SEQ ID NO: 527, and SEQ ID NO: 528.

[0127] The sequence listing (Table 18) also shows sequence variants of the binding substances named DLL3-4 and DLL3-14. The point mutations inserted into these sequence variants are identified according to the position of this mutation within the corresponding scFv molecule. Alternatively, it is understood that a method of identifying this position is also possible according to the reference polypeptide, which may also be a CDR region or a VH / VL region. For example, the variant named DLL3-4-001 has a double mutation of G44C-G243C within its scFv molecule (SEQ ID NO: 437). This corresponds to the G44C mutation of the corresponding VH chain (SEQ ID NO: 435) and the G101C mutation of the corresponding VL chain (SEQ ID NO: 436).

[0128] Covalent modifications of the antibody construct are also included within the scope of the present invention and are generally carried out post-translationally, although not always. For example, some types of covalent modifications of the antibody construct are introduced into the molecule by reacting specific amino acid residues of the antibody construct with an organic derivatizing agent capable of reacting with the selected side chain or N-terminal or C-terminal residue.

[0129] Cysteinyl residues are most commonly derivatized by reaction with α-haloacetates (and corresponding amines), such as chloroacetic acid or chloroacetamide, to yield carboxymethyl or carboxamidomethyl derivatives. Cysteinyl residues may also be derivatized by reaction with bromotrifluoroacetone, α-bromo-β-(5-imidazolyl)propionic acid, chloroacetyl phosphate, N-alkylmaleimide, 3-nitro-2-pyridyldisulfide, methyl 2-pyridyldisulfide, p-chloromercuribenzoic acid, 2-chloromercuri-4-nitrophenol, or chloro-7-nitrobenzo-2-oxa-1,3-diazole.

[0130] Histidyl residues are derivatized by reaction with diethylpyrocarbonate at pH 5.5 - 7.0 because this reagent is relatively specific for the histidyl side chain. Para-bromophenacyl bromide is also useful, and this reaction is preferably carried out in 0.1 M sodium cacodylate at pH 6.0. Lysinyl and amino-terminal residues react with succinic or other carboxylic acid anhydrides. Derivatization with these reagents has the effect of inverting the charge of the lysinyl residues. Other suitable reagents for derivatizing alpha-amino-containing residues include imidoesters such as methyl picolinimidate; pyridoxal phosphate; pyridoxal; chloroborohydride; trinitrobenzenesulfonic acid; O-methylisourea; 2,4-pentanedione; and transaminase-catalyzed reaction with glyoxylate.

[0131] Arginyl residues are modified by reaction with one or more conventional reagents, in particular phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin. Derivatization of arginine residues requires that this reaction be carried out under alkaline conditions because of the high pKa of the guanidine functional group. Furthermore, these reagents can react with the lysine groups as well as the arginine epsilon-amino group.

[0132] Specific modification of tyrosine residues may be carried out, especially for the purpose of introducing spectral labels into tyrosine residues by reaction with aromatic diazonium compounds or tetranitromethane. Most commonly, N-acetylimidizole and tetranitromethane can be used to form O-acetyltyrosyl species and 3-nitro derivatives, respectively. 125 I or 131 The above-mentioned chloramine T method for iodinating tyrosine residues using I or I is suitable for preparing labeled proteins used in radioimmunoassays.

[0133] Carboxyl side groups (aspartyl or glutamyl) are selectively modified by reaction with carbodiimide (R'-N=C=N-R'), where R and R' are optionally different alkyl groups, such as 1-cyclohexyl-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 glutaminyl residues by reaction with ammonium ions.

[0134] Derivatization using a bifunctional agent is useful when the antibody constructs of the present invention are cross-linked to a water-insoluble support matrix or support surface for use in various methods. Commonly used cross-linking agents include, for example, 1,1-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters such as esters with 4-azidosalicylic acid, homobifunctional imide esters including disuccinimidyl 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]propionimidate produce photoactivatable intermediates that can form cross-links in the presence of light. Alternatively, reactive water-insoluble matrices such as cyanogen bromide-activated carbohydrates, as well as reactive substrates described in U.S. Patent Nos. 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537; and 4,330,440 are used for protein immobilization.

[0135] Glutaminyl and asparaginyl residues are often deamidated to the corresponding glutamyl and aspartyl residues, respectively. Alternatively, these residues are deamidated under mild acidic conditions. Any form of these residues is included within the scope of the present invention.

[0136] Other modifications include hydroxylation of proline and lysine, phosphorylation of the hydroxyl group of a seryl or threonyl residue, methylation of the α-amino group of lysine, arginine, and histidine side chains (T.E. Creighton, Proteins: Structure and Molecular Properties, W.H. Freeman & Co., San Francisco, 1983, pp. 79-86), acetylation of the N-terminal amine, and amidation of any C-terminal carboxyl group.

[0137] Another type of covalent modification of the antibody constructs that are included within the scope of the present invention involves alteration of the glycosylation pattern of the protein. As is known in the art, the glycosylation pattern can depend on the sequence of the protein (e.g., the presence or absence of specific glycosylated amino acid residues described below), or both the host cell or organism that produces the protein. Individual expression systems are described below.

[0138] Glycosylation of polypeptides is usually either N-linked or O-linked. N-linked refers to the addition of a sugar moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine (where X is any amino acid other than proline) serve as recognition sequences during the enzymatic addition of a sugar moiety to the asparagine side chain. Thus, the presence of either of these tripeptide sequences in a polypeptide results in a glycosylation-capable site. O-linked glycosylation refers to the addition of one of the sugars N-acetylgalactosamine, galactose, or xylose to a hydroxyamino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine may also be used.

[0139] The addition of glycosylation sites to antibody constructs is advantageously done by modifying the amino acid sequence to include one or more of the above tripeptide sequences (in the case of N-linked glycosylation sites). The modification can also be done by the addition or substitution of one or more serine or threonine residues to the starting sequence (in the case of O-linked glycosylation sites). Briefly, it is preferred to modify the amino acid sequence of the antibody construct by mutating the DNA encoding the polypeptide at the DNA level, specifically with preselected bases such that codons that translate to the desired amino acids occur.

[0140] Another means of increasing the number of glycosyl moieties on an antibody construct is by chemical or enzymatic attachment of glycosides to the protein. Such procedures are advantageous in that they do not require the production of glycosylation-capable proteins in host cells during N-linked and O-linked glycosylation. Depending on the attachment mode used, the sugar(s) can be added to (a) arginine and histidine, (b) free carboxyl groups, (c) free sulfhydryl groups such as those of cysteine, (d) free hydroxyl groups such as those of serine, threonine or hydroxyproline, (e) aromatic residues such as those of phenylalanine, tyrosine, or tryptophan, or (f) the amide group of glutamine. Such methods are described in WO87 / 05330 and Aplin and Wriston, 1981, CRC Crit. Rev. Biochem., pp. 259-306.

[0141] Removal of glycosyl moieties present on the starting antibody construct can be carried out chemically or enzymatically. Chemical deglycosylation requires exposure of the protein to the compound trifluoromethanesulfonic acid or an equivalent compound. This treatment cleaves most or all of the sugars, leaving the polypeptide intact, except for the bound sugars (N-acetylglucosamine or N-acetylgalactosamine). Chemical deglycosylation is described by Hakimuddin et al., 1987, Arch. Biochem. Biophys. 259:52, and by Edge et al., 1981, Anal. Biochem. 118:131. Enzymatic cleavage of glycosyl moieties on the polypeptide can be carried out by the use of various endoglycosidases and exoglycosidases described in Thotakura et al., 1987, Meth. Enzymol. 138:350. Glycosylation at glycosylation-capable sites can be prevented by the use of the compound tunicamycin described in Duskin et al., 1982, J. Biol. Chem. 257:3105. Tunicamycin inhibits the formation of protein-N-glycoside linkages.

[0142] Other modifications of the antibody constructs are also contemplated herein. For example, other types of covalent modifications of the antibody constructs include crosslinking of the antibody constructs to various non-proteinaceous polymers, which polymers include, but are not limited to, various polyols in the manner shown in U.S. Patent Nos. 4,640,835; 4,496,689; 4,301,144; 4,670,417; 4,791,192 or 4,179,337, such as polyethylene glycol, polypropylene glycol, polyoxyalkylene, or copolymers of polyethylene glycol and polypropylene glycol. In addition, as is known in the art, amino acid substitutions can be made at various positions within the antibody construct to facilitate the addition of polymers such as PEG, for example.

[0143] In some embodiments, the covalent modification of the antibody constructs of the invention includes the addition of one or more labels. To reduce the potential for steric hindrance, label groups can be attached to the antibody constructs by spacer arms of various lengths. Various methods of labeling proteins are known in the art and can be used in practicing the present invention. The term "label" or "label group" refers to any detectable label. Generally, labels are classified into various classes depending on the assay in which the label is to be detected, and examples include, but are not limited to: a) Isotope labels, which can be radioactive isotopes or radionuclides (e.g., 3 H, 14 C, 15 N, 35 S, 89 Zr, 90 Y, 99 Tc, 111 In, 125 I, 131 I), etc., isotope labels b) Magnetic labels (e.g., magnetic particles) c) Redox-active moieties d) An optical dye (including but not limited to chromophores, phosphors, and fluorophores) such as a fluorophore that can be any of a fluorescent group (e.g., FITC, rhodamine, lanthanide phosphor), a chemiluminescent group, and a "low molecular weight" phosphor or a proteinaceous phosphor e) Enzyme groups (e.g., horseradish peroxidase, β - galactosidase, luciferase, alkaline phosphatase) f) Biotinylated group g) A predetermined polypeptide epitope recognized by a secondary reporter (e.g., leucine zipper pair sequence, binding site of a secondary antibody, metal binding domain, epitope tag, etc.).

[0144] "Fluorescent label" means any molecule that can be detected by its inherent fluorescent properties. Suitable fluorescent labels include fluorescein, rhodamine, tetramethylrhodamine, eosin, erythrosin, coumarin, methyl - coumarin, pyrene, malachite green, stilbene, lucifer yellow, cascade blue J, texas red, IAEDANS, EDANS, BODIPY FL, LC Red 640, Cy 5, Cy 5.5, LC Red 705, Oregon green, Alexa - Fluor dyes (Alexa Fluor 350, Alexa Fluor 430, Alexa Fluor 488, Alexa Fluor 546, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor 633, Alexa Fluor 660, Alexa Fluor 680), cascade blue, cascade yellow, and R - phycoerythrin (PE) (Molecular Probes, Eugene, OR), FITC, rhodamine, and texas red (Pierce, Rockford, IL), Cy5, Cy5.5, Cy7 (Amersham Life Science, Pittsburgh, PA), but are not limited thereto. Suitable optical dyes containing fluorophores are described in the Molecular Probes Handbook by Richard P. Haugland.

[0145] Also, suitable proteinaceous fluorescent labels include GFP of the Renilla, Ptilosarcus, or Aequorea species (Chalfie et al., 1994, Science 263:802-805), green fluorescent protein including EGFP (Clontech Laboratories, Inc., Genbank accession number U55762), blue fluorescent protein (BFP, Quantum Biotechnologies, Inc. 1801 de Maisonneuve Blvd. West, 8th Floor, Montreal, Quebec, Canada H3H 1J9; Stauber, 1998, Biotechniques 24:462-471; Heim et al., 1996, Curr. Biol. 6:178-182), enhanced yellow fluorescent protein (EYFP, Clontech Laboratories, Inc.), luciferase (Ichiki et al., 1993, J. Immunol. 150:5408-5417), β-galactosidase (Nolan et al., 1988, Proc. Natl. Acad. Sci. U.S.A. 85:2603-2607), and Renilla (WO92 / 15673, WO95 / 07463, WO98 / 14605, WO98 / 26277, WO99 / 49019, U.S. Patent Nos. 5,292,658, 5,418,155, 5,683,888, 5,741,668, 5,777,079, 5,804,387, 5,874,304, 5,876,995, 5,925,558), but are not limited thereto.

[0146] The leucine zipper domain is a peptide that promotes oligomerization of the protein in which it is found. Leucine zippers were initially identified in several DNA-binding proteins (Landschulz et al., 1988, Science 240:1759) and have since been found in a variety of different proteins. Particularly well-known leucine zippers are naturally occurring peptides and their derivatives that dimerize or trimerize. Examples of leucine zipper domains suitable for the production of soluble oligomeric proteins are described in PCT application WO94 / 10308, and a leucine zipper from pulmonary surfactant protein D (SPD) is described in Hoppe et al., 1994, FEBS Letters 344:191. The use of modified leucine zippers that allow stable trimerization of heterologous proteins fused thereto is described in Fanslow et al., 1994, Semin. Immunol. 6:267-78. In one approach, a recombinant fusion protein comprising a DLL3 antibody fragment or derivative fused to a leucine zipper peptide is expressed in a suitable host cell, and the resulting soluble oligomeric DLL3 antibody fragment or derivative is recovered from the culture supernatant.

[0147] The antibody constructs of the present invention may also include additional domains that, for example, aid in the isolation of the molecule or are relevant to the adaptation of the pharmacokinetic profile of the molecule. Domains that aid in the isolation of the antibody construct can be selected from isolation methods, such as peptide motifs that can be captured on an isolation column or secondarily introduced moieties. Non-limiting embodiments of such additional domains include Myc tag, HAT tag, HA tag, TAP tag, GST tag, chitin binding domain (CBD tag), maltose binding protein (MBP tag), Flag tag, Strep tag and variants thereof (e.g., StrepII tag), as well as peptide motifs known as His tag. All of the antibody constructs disclosed herein, characterized by the identification of CDRs, preferably include a His tag domain that is generally known as a repeat of consecutive His residues, preferably 5, more preferably 6 His residues (hexahistidine) in the amino acid sequence of the molecule. The His tag can be located, for example, at either the N-terminus or the C-terminus of the antibody construct, but is preferably located at the C-terminus. Most preferably, the hexahistidine tag (HHHHHH) is linked to the C-terminus of the antibody construct according to the present invention via a peptide bond.

[0148] The first binding domain of the antibody construct of the present invention binds to human DLL3 on the target cell surface. The preferred amino acid sequence of human DLL3 is represented by SEQ ID NO: 252. The term "on the surface" is understood in the context of the present invention to mean that the binding domain specifically binds to an epitope contained within the extracellular domain of DLL3 (DLL3 ECD). Thus, the first binding domain according to the present invention preferably binds to DLL3 when expressed by a naturally occurring cell or cell line and / or by a cell or cell line transformed or (stably / transiently) transfected with DLL3. In a preferred embodiment, the first binding domain also binds to DLL3 when using DLL3 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 DLL3 on its surface, but preferably the target cell is a cell that is part of the human or animal body, such as a cancer cell or tumor cell that specifically expresses DLL3.

[0149] The term "DLL3 ECD" refers to a form of DLL3 that essentially does not contain the transmembrane and cytoplasmic domains of DLL3. Those skilled in the art will understand that the identification of the transmembrane domain in the DLL3 polypeptide of the present invention is identified according to the criteria customarily used in the art for the identification of hydrophobic domains of that kind. The exact boundaries of the transmembrane domain may vary, but are likely to be up to about 5 amino acids from either end of any of the domains specifically described herein. The preferred human DLL3 ECD is shown in SEQ ID NO: 253.

[0150] The affinity of the first binding domain for human DLL3 is preferably ≤20 nM, more preferably ≤10 nM, even more preferably ≤5 nM, even more preferably ≤2 nM, even more preferably ≤1 nM, even more preferably ≤0.6 nM, even more preferably ≤0.5 nM, and most preferably ≤0.4 nM. For example, the affinity can be measured in a BIAcore assay or a Scatchard assay as described in the examples. Other methods for measuring affinity are well known to those skilled in the art.

[0151] T cells, or T lymphocytes, are a type of lymphocyte (which itself is a type of white blood cell) that play a central role in cell-mediated immunity. 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 a T cell receptor (TCR) on their cell surface. The TCR is responsible for recognizing antigens bound to major histocompatibility complex (MHC) molecules and is composed of two types of protein chains. In 95% of T cells, the TCR consists of an alpha (α) chain and a beta (β) chain. When the TCR contacts an antigen peptide and MHC (peptide / MHC complex), the T lymphocyte is activated through a series of biochemical events mediated by associated enzymes, co-receptors, specialized adapter molecules, and activated or released transcription factors.

[0152] The CD3 receptor complex is a protein complex composed of four chains. In mammals, this complex includes the CD3γ (gamma) chain, the CD3δ (delta) chain, and two CD3ε (epsilon) chains. These chains associate with the T cell receptor (TCR) and the so-called ζ (zeta) chain to form the T cell receptor CD3 complex, which generates activation signals in T lymphocytes. The CD3γ (gamma), CD3δ (delta), and CD3ε (epsilon) chains are related cell surface proteins of the immunoglobulin superfamily that contain a single extracellular immunoglobulin domain. The intracellular tail of the CD3 molecule contains a single conserved motif known as the immunoreceptor tyrosine activation motif or abbreviated ITAM, which is essential for the signaling ability of the TCR. The CD3 epsilon molecule is a polypeptide encoded by the CD3E gene present on human chromosome 11. The most preferred epitope of CD3 epsilon is contained in the range of amino acid residues 1 to 27 of the extracellular domain of human CD3 epsilon.

[0153] Lysis of target cells redirected through the recruitment of T cells by multispecific, at least bispecific antibody constructs involves the formation of a cytolytic synapse and the delivery of perforin and granzymes. The T cells involved are capable of continuous lysis of target cells and are not affected by immune evasion mechanisms that prevent the processing and presentation of peptide antigens or the differentiation of clonal T cells (see, for example, WO2007 / 042261).

[0154] The cytotoxicity mediated by the DLL3xCD3 bispecific antibody construct can be measured in various ways. See Examples 8.1 to 8.7. Effector cells can be, for example, stimulated enriched (human) CD8-positive T cells or unstimulated (human) peripheral blood mononuclear cells (PBMCs). When the target cells are of cynomolgus origin, or express cynomolgus DLL3 or are transfected with cynomolgus DLL3, the effector cells should also be of cynomolgus origin, such as a cynomolgus T cell line, for example 4119LnPx. The target cells should express DLL3, for example at least the extracellular domain of human or cynomolgus DLL3. The target cells can be cell lines (e.g., CHO) stably or transiently transfected with DLL3, such as human or cynomolgus DLL3. Alternatively, the target cells can be natural DLL3-positive expressing cell lines, such as the human lung cancer cell line SHP-77. Usually, the EC50 value is expected to be low in target cell lines that express high levels of DLL3 on the cell surface. The effector to target cell (E:T) ratio is usually about 10:1, but this can also be changed. The cytotoxic activity of the DLL3xCD3 bispecific antibody construct can be measured in a 51 chromium release assay (incubation time about 18 hours), or in a cytotoxicity assay using FACS (incubation time about 48 hours). It is also possible to change the incubation time of the assay (cytotoxic reaction). Other methods of measuring cytotoxicity are well known to those skilled in the art and include MTT or MTS assays, ATP-based assays including bioluminescence assays, sulforhodamine B (SRB) assays, WST assays, clonogenic assays, and ECIS technology.

[0155] The cytotoxic activity mediated by the DLL3xCD3 bispecific antibody construct of the present invention is preferably measured in a cytotoxicity assay using cells. It can also be measured in a 51 chromium release assay. The cytotoxic activity is EC 50represented by a value, which corresponds to the maximum half-maximal effective concentration (the concentration of the antibody construct that induces a cytotoxic response midway between the baseline and the maximum value). Preferably, the EC 50 value of the DLL3xCD3 bispecific antibody construct is ≤5000 pM or ≤4000 pM, more preferably ≤3000 pM or ≤2000 pM, even more preferably ≤1000 pM or ≤500 pM, even more preferably ≤400 pM or ≤300 pM, even more preferably ≤200 pM, even more preferably ≤100 pM, even more preferably ≤50 pM, even more preferably ≤20 pM or ≤10 pM, and most preferably ≤5 pM.

[0156] In various assays, the EC 50 value defined above can be measured. Those skilled in the art recognize that when using stimulated / concentrated CD8+ T cells as effector cells, the EC50 value can be expected to be lower compared to unstimulated PBMCs. Furthermore, the EC50 value can be predicted to be lower when the target cells express a large number of target antigens compared to rats with low target expression. For example, when using stimulated / concentrated human CD8+ T cells as effector cells (and using either DLL3-transfected cells such as CHO cells or the DLL3-positive human lung cancer cell line SHP-77 as target cells), the EC 50 value of the DLL3xCD3 bispecific antibody construct is preferably ≤1000 pM, more preferably ≤500 pM, even more preferably ≤250 pM, even more preferably ≤100 pM, even more preferably ≤50 pM, even more preferably ≤10 pM, and most preferably ≤5 pM. When using human PBMCs as effector cells, the EC 50The value is preferably ≤5000 pM or ≤4000 pM (particularly when the target cells are the DLL3-positive human lung cancer cell line SHP-77), more preferably ≤2000 pM (particularly when the target cells are DLL3-transfected cells such as CHO cells), still more preferably ≤1000 pM or ≤500 pM, even more preferably ≤200 pM, even more preferably ≤150 pM, even more preferably ≤100 pM, most preferably ≤50 pM, or less. When using a cynomolgus T cell line such as LnPx4119 as the effector cell and a cynomolgus DLL3-transfected cell line such as CHO cells as the target cell line, the EC of the DLL3xCD3 bispecific antibody construct 50 The value is preferably ≤2000 pM or ≤1500 pM, more preferably ≤1000 pM or ≤500 pM, still more preferably ≤300 pM or ≤250 pM, even more preferably ≤100 pM, and most preferably ≤50 pM.

[0157] Preferably, the DLL3xCD3 bispecific antibody construct of the present invention does not induce / mediate, or essentially does not induce / mediate, the lysis of DLL3-negative cells such as CHO cells. The terms "does not induce lysis", "essentially does not induce lysis", "does not mediate lysis", or "essentially does not mediate lysis" mean that when the DLL3-positive human lung cancer cell line SHP-77 (see above) is taken as 100%, the antibody construct of the present invention does not induce or mediate lysis of more than 30%, preferably more than 20%, more preferably more than 10%, particularly preferably more than 9%, 8%, 7%, 6%, or 5% of the DLL3-negative cells. This generally holds true for antibody construct concentrations up to 500 nM. A person skilled in the art understands the method for measuring cell lysis without particular effort. Further, the present specification teaches specific instructions for the method of measuring cell lysis.

[0158] The difference in cytotoxic activity between the monomeric isoform and the dimeric isoform of an individual DLL3xCD3 bispecific antibody construct is referred to as the "potency gap". This potency gap is, for example, the EC of the monomeric form of the molecule 50 value and the EC of the dimeric form50 It can be calculated as a ratio to a value (see Example 15). The potency difference of the DLL3xCD3 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.

[0159] The first and / or second (or any additional) binding domain(s) of the antibody construct of the present invention is preferably species-specific in members of the mammalian order of primates. Species-specific CD3 binding domains are described, for example, in WO2008 / 119567. According to one embodiment, the first and / or second binding domain binds to DLL3 / CD3 of primates (including, but not limited to, New World primates (such as common marmoset (Callithrix jacchus), cotton-top tamarin (Saguinus Oedipus), or common squirrel monkey (Saimiri sciureus)), Old World primates (such as baboons and macaques), gibbons, orangutans, and human subfamily members other than humans) in addition to binding to human DLL3 and human CD3. The first binding domain of the antibody construct of the present invention that binds to human DLL3 on the surface of target cells also binds to at least macaque DLL3, and / or the second binding domain that binds to human CD3 on the surface of T cells is also expected to bind to at least macaque CD3. A preferred macaque is cynomolgus macaque (Macaca fascicularis). Rhesus macaque (Macaca mulatta) is also contemplated.

[0160] A preferred bispecific antibody construct of the present invention comprises a first binding domain that binds to human DLL3 on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells and at least macaque CD3. In one aspect of this embodiment, the first binding domain binds to an epitope of DLL3 contained within the region shown in SEQ ID NO: 260.

[0161] In one aspect of the present invention, the first binding domain binds to human DLL3 and further binds to cynomolgus DLL3 such as cynomolgus macaque DLL3, more preferably cynomolgus macaque DLL3 ECD. The preferred cynomolgus macaque DLL3 is shown in SEQ ID NO: 271. The preferred cynomolgus macaque DLL3 ECD is shown in SEQ ID NO: 272. The affinity of the first binding domain for cynomolgus macaque DLL3 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, or even ≤ 0.01 nM.

[0162] Preferably, the difference in binding affinity of the antibody construct according to the present invention for cynomolgus macaque DLL3 to human DLL3 [ma DLL3: hu DLL3] is 0.1 to 10, more preferably 0.2 to 5, even more preferably 0.3 to 4, even more preferably 0.5 to 3 or 0.5 to 2.5, most preferably 0.5 to 2 or 0.6 to 2 (for example, when measured by BiaCore or Scatchard analysis). See Examples 3 and 4.

[0163] In one embodiment of the antibody construct of the present invention, the second binding domain binds to human CD3 epsilon and to CD3 epsilon of common marmoset, cottontop tamarin, or common squirrel monkey. Preferably, the second binding domain binds to the extracellular epitope of these CD3 epsilon chains. It is also contemplated that the second binding domain binds to the extracellular epitopes of human and cynomolgus macaque CD3 epsilon chains. The most preferred epitope of CD3 epsilon is contained in the range of amino acid residues 1 to 27 of the extracellular domain of human CD3 epsilon. More specifically, the epitope contains at least the amino acid sequence Gln-Asp-Gly-Asn-Glu. Both common marmoset and cottontop tamarin are New World primates belonging to the family Callitrichidae, while common squirrel monkey is a New World primate belonging to the family Cebidae.

[0164] For the antibody construct of the present invention, it is particularly preferred that the second binding domain that binds to human CD3 on the T cell surface comprises a VL region comprising CDR-L1, CDR-L2, and CDR-L3 selected from the following: (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 SEQ ID NO: 29 of WO2008 / 119567; (b) CDR-L1 shown in SEQ ID NO: 117 of WO2008 / 119567, CDR-L2 shown in SEQ ID NO: 118 of WO2008 / 119567, and CDR-L3 shown in SEQ ID NO: 119 of WO2008 / 119567; and (c) CDR-L1 shown in SEQ ID NO: 153 of WO2008 / 119567, CDR-L2 shown in SEQ ID NO: 154 of WO2008 / 119567, and CDR-L3 shown in SEQ ID NO: 155 of WO2008 / 119567.

[0165] In a preferred alternative embodiment of the antibody construct of the present invention, the second binding domain that binds to human CD3 on the T cell surface comprises a VH region comprising CDR-H1, CDR-H2, and CDR-H3 selected from the following: (a) CDR-H1 shown in SEQ ID NO: 12 of WO2008 / 119567, CDR-H2 shown in SEQ ID NO: 13 of WO2008 / 119567, and CDR-H3 shown in SEQ ID NO: 14 of WO2008 / 119567; (b) CDR-H1 shown in SEQ ID NO: 30 of WO2008 / 119567, CDR-H2 shown in SEQ ID NO: 31 of WO2008 / 119567, and CDR-H3 shown in SEQ ID NO: 32 of WO2008 / 119567; (c) CDR-H1 shown in SEQ ID NO: 48 of WO2008 / 119567, CDR-H2 shown in SEQ ID NO: 49 of WO2008 / 119567, and CDR-H3 shown in SEQ ID NO: 50 of WO2008 / 119567; (d) CDR-H1 shown in SEQ ID NO: 66 of WO2008 / 119567, CDR-H2 shown in SEQ ID NO: 67 of WO2008 / 119567, and CDR-H3 shown in SEQ ID NO: 68 of WO2008 / 119567; (e) CDR-H1 shown in SEQ ID NO: 84 of WO2008 / 119567, CDR-H2 shown in SEQ ID NO: 85 of WO2008 / 119567, and CDR-H3 shown in SEQ ID NO: 86 of WO2008 / 119567; (f) CDR-H1 shown in SEQ ID NO: 102 of WO2008 / 119567, CDR-H2 shown in SEQ ID NO: 103 of WO2008 / 119567, and CDR-H3 shown in SEQ ID NO: 104 of 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 CDR-H3 shown in SEQ ID NO: 122 of WO2008 / 119567; (h) CDR-H1 shown in SEQ ID NO: 138 of WO2008 / 119567, CDR-H2 shown in SEQ ID NO: 139 of WO2008 / 119567, and CDR-H3 shown in SEQ ID NO: 140 of WO2008 / 119567; (i) CDR-H1 shown in SEQ ID NO: 156 of WO2008 / 119567, CDR-H2 shown in SEQ ID NO: 157 of WO2008 / 119567, and CDR-H3 shown in SEQ ID NO: 158 of WO2008 / 119567; (j) CDR-H1 shown in SEQ ID NO: 174 of WO2008 / 119567, CDR-H2 shown in SEQ ID NO: 175 of WO2008 / 119567, and CDR-H3 shown in SEQ ID NO: 176 of WO2008 / 119567.

[0166] For the antibody construct of the present invention, it is more preferable that the second binding domain that binds to human CD3 on the T cell surface includes a VL region selected from the group consisting of the VL regions shown in SEQ ID NOs: 35, 39, 125, 129, 161, or 165 of WO2008 / 119567.

[0167] Alternatively, the second binding domain that binds to human CD3 on the T cell surface preferably comprises a VH region selected from the group consisting of the VH regions set forth in SEQ ID NOs: 15, 19, 33, 37, 51, 55, 69, 73, 87, 91, 105, 109, 123, 127, 141, 145, 159, 163, 177, or 181 of WO2008 / 119567.

[0168] More preferably, the antibody construct of the present invention is characterized by a second binding domain that binds to human CD3 on the T cell surface and comprises a VL region and a VH region selected from the group consisting of: (a) a VL region set forth in SEQ ID NO: 17 or 21 of WO2008 / 119567 and a VH region set forth in SEQ ID NO: 15 or 19 of WO2008 / 119567; (b) a VL region set forth in SEQ ID NO: 35 or 39 of WO2008 / 119567 and a VH region set forth in SEQ ID NO: 33 or 37 of WO2008 / 119567; (c) a VL region set forth in SEQ ID NO: 53 or 57 of WO2008 / 119567 and a VH region set forth in SEQ ID NO: 51 or 55 of WO2008 / 119567; (d) a VL region set forth in SEQ ID NO: 71 or 75 of WO2008 / 119567 and a VH region set forth in SEQ ID NO: 69 or 73 of WO2008 / 119567; (e) a VL region set forth in SEQ ID NO: 89 or 93 of WO2008 / 119567 and a VH region set forth in SEQ ID NO: 87 or 91 of WO2008 / 119567; (f) a VL region set forth in SEQ ID NO: 107 or 111 of WO2008 / 119567 and a VH region set forth in SEQ ID NO: 105 or 109 of WO2008 / 119567; (g) a VL region set forth in SEQ ID NO: 125 or 129 of WO2008 / 119567 and a VH region set forth in SEQ ID NO: 123 or 127 of WO2008 / 119567; (h) a VL region set forth in SEQ ID NO: 143 or 147 of WO2008 / 119567 and a VH region set forth in SEQ ID NO: 141 or 145 of WO2008 / 119567; (i) The VL region shown in SEQ ID NO: 161 or 165 of WO2008 / 119567 and the VH region shown in SEQ ID NO: 159 or 163 of WO2008 / 119567; and (j) The VL region shown in SEQ ID NO: 179 or 183 of WO2008 / 119567 and the VH region shown in SEQ ID NO: 177 or 181 of WO2008 / 119567.

[0169] 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 T cell surface), has a form in which the pair of VH region and VL region 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. Preferably, the VH region is located on the N-terminal side of the linker sequence and the VL region is located on the C-terminal side of the linker sequence.

[0170] A preferred embodiment of the above antibody construct of the present invention is characterized by a second binding domain that binds to human CD3 on the T cell surface and includes an amino acid sequence selected from the group consisting of SEQ ID NO: 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185, or 187 of WO2008 / 119567.

[0171] Thus, in one embodiment, the antibody construct of the present invention includes a polypeptide selected from the group consisting of those shown in SEQ ID NO: 40, SEQ ID NO: 50, SEQ ID NO: 60, SEQ ID NO: 70, SEQ ID NO: 80, SEQ ID NO: 90, SEQ ID NO: 100, SEQ ID NO: 110, SEQ ID NO: 120, SEQ ID NO: 211, SEQ ID NO: 212, SEQ ID NO: 213, SEQ ID NO: 214, SEQ ID NO: 215, SEQ ID NO: 216, SEQ ID NO: 217, SEQ ID NO: 438, and SEQ ID NO: 532. This antibody construct has a first binding domain that binds to an epitope of DLL3 contained within the region shown in SEQ ID NO: 258.

[0172] In an alternative embodiment, the antibody construct of the invention comprises a polypeptide selected from the group consisting of those set forth in SEQ ID NO: 130, SEQ ID NO: 140, SEQ ID NO: 150, SEQ ID NO: 160, SEQ ID NO: 170, SEQ ID NO: 218, SEQ ID NO: 219, SEQ ID NO: 220, SEQ ID NO: 494, SEQ ID NO: 495, SEQ ID NO: 496, SEQ ID NO: 497, SEQ ID NO: 498, SEQ ID NO: 499, SEQ ID NO: 500, SEQ ID NO: 501, SEQ ID NO: 502, SEQ ID NO: 503, SEQ ID NO: 504, SEQ ID NO: 505, SEQ ID NO: 506, SEQ ID NO: 507, SEQ ID NO: 508, SEQ ID NO: 509, SEQ ID NO: 510, SEQ ID NO: 511, SEQ ID NO: 512, SEQ ID NO: 513, SEQ ID NO: 514, SEQ ID NO: 515, and SEQ ID NO: 516. This antibody construct has a first binding domain that binds to an epitope of DLL3 contained within the region set forth in SEQ ID NO: 259.

[0173] Amino acid sequence modifications of the antibody constructs described herein are also contemplated. For example, when it is desirable to improve the binding affinity and / or other biological properties of the antibody construct. Amino acid sequence variants of the antibody construct are prepared by introducing appropriate nucleotide changes into the nucleic acid of the antibody construct or by peptide synthesis. All of the amino acid sequence modifications described below should generate antibody constructs that continue to retain the desired biological activity (binding to DLL3 and CD3) of the unmodified parent molecule.

[0174] The term "amino acid" or "amino acid residue" generally refers to an amino acid having a definition recognized in the art, such as alanine (Ala or A); arginine (Arg or R); asparagine (Asn or N); aspartic acid (Asp or D); cysteine (Cys or C); glutamine (Gln or Q); glutamic acid (Glu or E); glycine (Gly or G); histidine (His or H); isoleucine (Ile or I); leucine (Leu or L); lysine (Lys or K); methionine (Met or M); phenylalanine (Phe or F); proline (Pro or P); serine (Ser or S); threonine (Thr or T); tryptophan (Trp or W); tyrosine (Tyr or Y); and valine (Val or V). Optionally, modified amino acids, synthetic amino acids, or rare amino acids may be used. Generally, amino acids can be classified by the presence of nonpolar side chains (e.g., Ala, Cys, Ile, Leu, Met, Phe, Pro, Val); negatively charged side chains (e.g., Asp, Glu); positively charged side chains (e.g., Arg, His, Lys); or uncharged polar side chains (e.g., Asn, Cys, Gln, Gly, His, Met, Phe, Ser, Thr, Trp, and Tyr).

[0175] Amino acid modifications include, for example, deletions from residues within the amino acid sequence of an antibody construct, and / or insertions into residues, and / or substitutions of residues. Any combination of deletions, insertions, and substitutions may be made to reach the final construct, as long as the final construct retains the desired characteristics. Amino acid changes can also alter post-translational processes of the antibody construct, such as changes in the number or position of glycosylation sites.

[0176] For example, in each CDR, 1, 2, 3, 4, 5, or 6 amino acids (depending on the length of the CDR, of course) may be inserted or deleted, and in each corresponding FR, 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. Preferably, amino acid sequence insertions include amino-terminal fusions and / or carboxyl-terminal fusions in the range of lengths from 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 residues to polypeptides containing 100 or more residues, as well as in-sequence insertions of single or multiple amino acid residues. Insertion variants of the antibody constructs of the present invention include fusions of enzymes to the N-terminus or C-terminus of the antibody construct, or fusions to polypeptides that increase the serum half-life of the antibody construct.

[0177] The most important target sites for substitution mutagenesis include the CDRs of the heavy and / or light chains, particularly those including the hypervariable regions, although alterations in the FRs in the heavy and / or light chains are also envisioned. Substitutions are preferably conservative substitutions as described herein. Preferably, depending on the length of the CDR or FR, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids may be substituted in the CDR, and 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 corresponding framework region (FR). For example, if a CDR sequence contains 6 amino acids, it is envisioned that 1, 2, or 3 of these amino acids may be substituted. Similarly, if a CDR sequence contains 15 amino acids, it is envisioned that 1, 2, 3, 4, 5, or 6 of these amino acids may be substituted.

[0178] A useful method for identifying specific residues or regions of an antibody construct that are preferred positions for mutagenesis is what is called "alanine scan mutagenesis" described by Cunningham and Wells in Science, 244:1081-1085 (1989). In this method, residues or groups of target residues within the antibody construct are identified (e.g., charged residues such as arg, asp, his, lys, and glu), and replaced with neutral or negatively charged amino acids (most preferably, alanine or polyalanine) that affect the interaction between the amino acid and the epitope.

[0179] Then, by introducing additional variants or other variants at or instead of the substitution site, the amino acid positions that show functional sensitivity to the substitution are stringently selected. Thus, the site or region where the amino acid sequence variant is introduced is predetermined, but it is not necessary to predetermine the nature of the mutation itself. For example, to analyze or optimize the function of a mutation at a given site, alanine scan or random mutagenesis can be performed on the target codon or target region, and variants of the expressed antibody construct are screened for the optimal combination for the desired activity. Techniques for performing substitution mutations at a predetermined site within DNA having a known sequence are well known, such as M13 primer mutagenesis and PCR mutagenesis. Screening of mutants is performed using an assay of antigen-binding activity such as DLL3 or CD3 binding.

[0180] Generally, when one or more or all of the CDRs of the heavy and / or light chains have amino acid substitutions, the resulting "substituted" sequence is preferably at least 60% or 65%, more preferably 70% or 75%, even more preferably 80% or 85%, particularly preferably 90% or 95% identical to the "original" CDR sequence. This means that the degree of identity between the CDR and the "substituted" sequence depends on the length of the CDR. For example, for a CDR having 5 amino acids, it is preferably 80% identical to the substituted amino acid sequence in order to have at least one substituted amino acid. Thus, the CDRs of the antibody construct may have different degrees of identity to their substituted sequences, for example, CDRL1 may have 80% and CDRL3 may have 90%.

[0181] Preferred substitutions are conservative substitutions. However, any substitution (including non-conservative substitutions or one or more of the "exemplary substitutions" listed in Table 1 below) is envisioned as long as the antibody construct retains its ability to bind to DLL3 via the first binding domain and to CD3 or CD3 epsilon via the second binding domain, and / or the CDRs of the antibody construct have identity to the sequence after substitution (at least 60% or 65%, more preferably 70% or 75%, even more preferably 80% or 85%, particularly preferably 90% or 95% identical to the "original" CDR sequence).

[0182] Conservative substitutions are shown under the heading "Preferred Substitutions" in Table 1. If such substitutions alter the biological activity, more substantial changes, referred to as "exemplary substitutions" in Table 1 or further described below in relation to amino acid classes, can be introduced and the product screened for the desired properties.

[0183] (Table 1) Amino Acid Substitutions TIFF2025106535000003.tif145168

[0184] Substantial alterations in the biological properties of the antibody constructs of the present invention are achieved by selecting substitutions that differ significantly in their effect on (a) the structure of the polypeptide backbone of the substitution region, which is, for example, a sheet-like or helical three-dimensional structure, (b) the charge or hydrophobicity of the molecule at the target site, or (c) the maintenance of the bulkiness of the side chains. Naturally occurring residues are classified into the following groups based on common side-chain properties: (1) hydrophobic: norleucine, met, ala, val, leu, ile; (2) neutral hydrophilic: cys, ser, thr; (3) acidic: asp, glu; (4) basic: asn, gin, his, lys, arg; (5) residues affecting chain orientation: gly, pro; and (6) aromatic: trp, tyr, phe.

[0185] Non-conservative substitutions will involve the exchange of one member of any of these classes with another class. Substituting any cysteine residues that are not involved in maintaining the proper three-dimensional structure of the antibody construct generally with serine can improve the oxidative stability of the molecule and avoid abnormal cross-linking. Conversely, adding cysteine bond(s) to the antibody can improve stability (especially when the antibody is an antibody fragment such as an Fv fragment).

[0186] Regarding amino acid sequences, sequence identity and / or similarity are determined by techniques known in the art, e.g., but not limited to, the subsequence identity algorithm of Smith and Waterman, 1981, Adv. Appl. Math. 2:482, the sequence identity alignment algorithm of Needleman and Wunsch, 1970, J. Mol. Biol. 48:443, the similarity search method of Pearson and Lipman, 1988, Proc. Nat. Acad. Sci. U.S.A. 85:2444, computer implementations of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group, 575 Science Dr., Madison, Wis), using, preferably, the Best Fit sequence program described by Devereux et al., 1984, Nucl. Acid Res. 12:387-395, with default settings, or by visual inspection. Preferably, percent identity is calculated by FastDB based on the following parameters: mismatch penalty of 1; gap penalty of 1; gap size penalty of 0.33; and join penalty of 30, "Current Methods in Sequence Comparison and Analysis," Macromolecule Sequencing and Synthesis, Selected Methods and Applications, pp 127-149 (1988), Alan R. Liss, Inc.

[0187] An example of a useful algorithm is PILEUP. PILEUP generates multiple sequence alignments from a group of related sequences using progressive pairwise alignment. It can also plot a tree showing the clustering relationships used to generate the alignment. PILEUP uses a simplified version of the progressive alignment method of Feng & Doolittle, 1987, J. Mol. Evol. 35:351-360. This method is similar to that described by Higgins and Sharp, 1989, CABIOS 5:151-153. Useful PILEUP parameters are a default gap weight of 3.00, a default gap length weight of 0.10, and include a weighted end gap.

[0188] Another example of a useful algorithm is the BLAST algorithm described in 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. U.S.A. 90:5873-5787. 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 to the following values: overlap span = 1, overlap fraction = 0.125, word threshold (T) = 11. The HSP S and HSP S2 parameters are dynamic values that are determined programmatically depending on the composition of the individual sequences and the composition of the specific database against which the target sequences are being searched, but the sensitivity can be increased by adjusting these values.

[0189] Another useful algorithm is gapped BLAST reported by Altschul et al., 1993, Nucl. Acids Res. 25: 3389-3402. Gapped BLAST uses the BLOSUM-62 substitution score, the threshold T parameter is set to 9, the two-hit method that performs gapless extension has the cost of gap length k as 10 + k, Xu is set to 16, Xg is set to 40 at the database search stage and 67 at the output stage of the algorithm. The gapped alignment is performed by a score corresponding to approximately 22 bits.

[0190] Generally, the amino acid homology, similarity, or identity between individual variant CDRs is at least 60% relative to the sequences shown herein, more generally the homology or identity is at least 65% or 70%, more preferably at least 75% or 80%, even more preferably at least 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, and almost 100%. Similarly, the "percent nucleic acid sequence identity (%)" with respect to the nucleic acid sequence of the binding protein identified herein is defined as the percentage of nucleotide residues in a candidate sequence that are identical to the nucleotide residues in the coding sequence of the antibody construct. In a specific method, the BLASTN module of WU-BLAST-2 with default parameters of overlap span and overlap fraction set to 1 and 0.125 respectively is utilized.

[0191] Generally, 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 the homology or identity is 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 increasing to almost 100%. Thus, a "variant CDR" has the homology, similarity, or identity specified for the parental CDR of the invention and shares a biological function that includes, but is not limited to, at least 60%, 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the specificity and / or activity of the parental CDR.

[0192] In one embodiment, the percent identity of the antibody construct according to the present invention to the human germline is ≧70% or ≧75%, more preferably ≧80% or ≧85%, even more preferably ≧90%, most preferably ≧91%, ≧92%, ≧93%, ≧94%, ≧95%, or even ≧96%. See Example 7. Identity to the human antibody germline gene product is thought to be an important function in reducing the risk that a therapeutic protein will induce an immune response to the drug in the patient being treated. Hwang & Foote (“Immunogenicity of engineered antibodies”; Methods 36 (2005) 3-10) have shown that reducing the non-human portion of a drug antibody construct leads to a decrease in the risk of inducing anti-drug antibodies in the patient being treated. By comparing a vast number of clinically evaluated antibody drugs and the corresponding immunogenicity data, humanization of the V region of an antibody has been shown to result in less immunogenicity of the protein (average 5.1% in patients) compared to antibodies bearing unmodified non-human V regions (average 23.59% in patients). Therefore, for protein therapeutics in the form of antibody constructs using the V region, it is desirable to have a high degree of identity to the human sequence. For the purpose of determining this germline identity, the V region of VL is aligned with the amino acid sequences of the human germline V and J segments (http: / / vbase.mrc-cpe.cam.ac.uk / ) using Vector NTI software, and the amino acid sequence can be calculated by dividing the number of identical amino acid residues by the total number of amino acid residues in VL to obtain a percentage. The same method is possible for the VH segment (http: / / vbase.mrc-cpe.cam.ac.uk / ), but VH CDR3 is highly diverse and lacks alignment partners among existing human germline VH CDR3s, so it is excluded. Subsequently, recombinant techniques can be used to increase the sequence identity to the human antibody germline gene.

[0193] In a further embodiment, the bispecific antibody construct of the invention exhibits a high monomer yield under standard research-scale conditions, for example, in a standard two-step purification method. 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, and most preferably ≧3 mg / L supernatant.

[0194] Similarly, the yield of the dimeric antibody construct isoform of this antibody construct, and thereby the monomer ratio (i.e., monomer: (monomer + dimer)) can be determined. The productivity of the monomer and dimer antibody constructs, as well as the calculated monomer ratio, can be obtained, for example, in the SEC purification step of the culture supernatant obtained from standardized research-scale production in roller bottles. In one embodiment, the monomer ratio of the antibody construct is ≧80%, more preferably ≧85%, even more preferably ≧90%, and most preferably ≧95%.

[0195] In one embodiment, the antibody construct preferably has a plasma stability (ratio of EC50 in the presence of plasma to EC50 in the absence of plasma) of ≦5 or ≦4, more preferably ≦3.5 or ≦3, even more preferably ≦2.5 or ≦2, and most preferably ≦1.5 or ≦1. The plasma stability of the antibody construct can be tested by incubating the construct in human plasma at 37° C. for 24 hours, followed by determining the EC50 in a 51 chromium release cytotoxicity assay. In the cytotoxicity assay, effector cells can be stimulated with enriched human CD8-positive T cells. The target cells can be, for example, CHO cells transfected with human DLL3. The effector to target cell (E:T) ratio can be selected as 10:1. The human plasma pool used for this purpose is obtained from the blood of healthy donors collected by EDTA-coated syringes. The cellular components are removed by centrifugation, and the upper plasma phase is recovered and then pooled. As a control, the antibody construct is diluted with RPMI-1640 medium immediately before the cytotoxicity assay. Plasma stability is calculated as the ratio of EC50 (after plasma incubation) to EC50 (control). See Example 11.

[0196] Furthermore, the conversion rate of the monomer of the antibody construct of the present invention to the dimer is preferably low. The conversion rate can be measured under different conditions and analyzed by high-performance size exclusion chromatography. See Example 9. For example, the incubation of the monomer isoform of the antibody construct can be carried out in an incubator at a concentration of, for example, 100 μg / ml or 250 μg / ml at 37° C. for 7 days. Under these conditions, the antibody construct of the present invention preferably exhibits a dimerization rate of ≦5%, more preferably ≦4%, even more preferably ≦3%, even more preferably ≦2.5%, even more preferably ≦2%, even more preferably ≦1.5%, most preferably ≦1% or ≦0.5%, or even 0%.

[0197] It is also preferable that the bispecific antibody construct of the present invention exhibits a very low dimer conversion rate after several freeze / thaw cycles. For example, the monomer of the antibody construct is adjusted to a concentration of 250 μg / ml in, for example, a general formulation buffer, subjected to 3 freeze / thaw cycles (freezing at -80° C. for 30 minutes and then thawing at room temperature for 30 minutes), and then subjected to high-speed SEC to determine the proportion of the initial monomer antibody construct converted to the dimeric antibody construct. Preferably, the dimerization rate of the bispecific antibody construct is, for example, ≦5%, more preferably ≦4%, even more preferably ≦3%, even more preferably ≦2.5%, even more preferably ≦2%, even more preferably ≦1.5%, most preferably ≦1%, or even ≦0.5% after 3 freeze / thaw cycles.

[0198] The bispecific antibody construct of the present invention preferably exhibits good thermal stability at an aggregation temperature of ≧45° C. or ≧50° C., more preferably ≧52° C. or ≧54° C., even more preferably ≧56° C. or ≧57° C., and most preferably ≧58° C. or ≧59° C. The thermal stability parameters can be determined as follows from the perspective of the aggregation temperature of the antibody. Transfer an antibody solution with a concentration of 250 μg / ml to a single-use cuvette and place it in a dynamic light scattering (DLS) apparatus. Heat the sample from 40° C. to 70° C. at a heating rate of 0.5° C. / min and continuously acquire the measured radius. Utilize the increase in radius indicating protein melting and aggregation to calculate the aggregation temperature of the antibody. Refer to Example 10.

[0199] Alternatively, to determine the intrinsic biophysical protein stability of the antibody construct, the melting temperature curve can be determined by differential scanning calorimetry (DSC). This experiment is performed using a MicroCal LLC (Northampton, MA, U.S.A) VP-DSC apparatus. Record the energy uptake of the sample containing the antibody construct from 20° C. to 90° C. and compare it with a sample containing only the formulation buffer. Adjust the antibody construct to a final concentration of 250 μg / ml, for example, in SEC running buffer. Gradually increase the overall temperature of the sample to record each melting curve. Record the energy uptake of the sample and the formulation buffer standard at each temperature T. Plot the difference in energy uptake Cp (kcal / mole / ° C.) obtained by subtracting the standard from the sample for each temperature. The melting temperature is defined as the temperature at which the energy uptake first reaches a maximum.

[0200] Furthermore, it is assumed that the DLL3xCD3 bispecific antibody construct of the present invention does not cross-react (i.e., essentially does not bind) with human DLL3 paralogs DLL1 and / or DLL4. Furthermore, it is assumed that the DLL3xCD3 bispecific antibody construct of the present invention does not cross-react (i.e., essentially does not bind) with cynomolgus / rhesus macaque DLL3 paralogs DLL1 and / or DLL4. Refer to Example 6.

[0201] The DLL3xCD3 bispecific antibody construct of the present invention also has a turbidity of ≤0.2, preferably ≤0.15, more preferably ≤0.12, even more preferably ≤0.1, and most preferably ≤0.08 (measured by OD340 after concentrating the purified monomeric antibody construct to 2.5 mg / ml and incubating overnight). See Example 12.

[0202] The DLL3xCD3 bispecific antibody construct of the present invention is also expected not to internalize or undergo significant internalization by target cells. For example, the rate of internalization can be analyzed as described in Example 16. Preferably, the rate of internalization (measured, for example, as a decrease in cytotoxicity) after incubating the antibody construct with target cells for 2 hours (pre) is ≤20%, more preferably ≤15%, even more preferably ≤10%, and most preferably ≤5%.

[0203] Furthermore, it is assumed that the secreted or soluble DLL3 does not significantly impair the efficacy or biological activity of the DLL3xCD3 bispecific antibody construct of the present invention. This can be measured, for example, in a cytotoxicity assay by adding soluble DLL3 to the assay while increasing its concentration, for example, to 0 nM - 0.3 nM - 0.7 nM - 1 nM - 3 nM - 7 nM - 12 nM. The EC50 value of the tested antibody construct should not increase significantly in the presence of soluble DLL3. See Example 17.

[0204] In a further embodiment, the antibody construct according to the invention is stable at acidic pH. The higher the resistance shown by the antibody construct under non-physiological pH, for example pH 5.5 (for example, the pH required for performing cation exchange chromatography), the higher the recovery rate of the antibody construct eluted from the ion exchange column with respect to the total amount of the loaded protein. The recovery rate of the antibody construct from an ion (e.g., cation) exchange column under pH 5.5 is preferably ≥ 30%, more preferably ≥ 40%, more preferably ≥ 50%, even more preferably ≥ 60%, even more preferably ≥ 70%, even more preferably ≥ 80%, even more preferably ≥ 90%, even more preferably ≥ 95%, and most preferably ≥ 99%. See Example 13.

[0205] It is further contemplated that the bispecific antibody construct of the present invention exhibits therapeutic efficacy or antitumor activity. This can be evaluated, for example, in tests disclosed in the examples of the following human tumor xenograft models at various progressive stages.

[0206] On the first day of the study, 5x10 6 cells of a human DLL3-positive cancer cell line (e.g., SHP-77) are subcutaneously injected into the right dorsal flank of female NOD / SCID mice. When the average tumor volume reaches about 100 mm 3 , human CD3-positive T cells grown in vitro are transplanted into the mice by injection of about 2x10 7 cells into the abdominal cavity of the animals. Mice in vehicle control group 1 are not given effector cells and are used as an untransplanted control for comparison with vehicle control group 2 (which is given effector cells) to monitor the effect of T cells alone on tumor growth. Antibody treatment is initiated when the average tumor volume reaches about 200 mm 3It starts when it reaches [the specified condition]. The average tumor size of each treatment group on the treatment start date should have no statistical difference from any other group (analysis of variance). The mice are treated by intravenous bolus injection with a DLL3xCD3 bispecific antibody construct at 0.5 mg / kg / day for about 15 to 20 days. During the test, the tumors are measured with calipers, and the progression is evaluated by comparing the tumor volume (TV) between groups. The tumor growth inhibition T / C [%] is determined by calculating the TV as T / C% = 100 x (median TV of the analysis group) / (median TV of control group 2).

[0207] Those skilled in the art understand how to obtain significant and reproducible results by changing or adapting specific parameters of this test, such as the number of tumor cells to be injected, the injection site, the number of human T cells to be transplanted, the amount of the bispecific antibody construct to be administered, and the timeline. Preferably, the tumor growth inhibition T / C [%] is ≤ 70 or ≤ 60, more preferably ≤ 50 or ≤ 40, even more preferably ≤ 30 or ≤ 20, most preferably ≤ 10 or ≤ 5, or even ≤ 2.5.

[0208] The present invention further provides a polynucleotide / nucleic acid molecule encoding the antibody construct of the present invention.

[0209] A polynucleotide is a biopolymer composed of 13 or more nucleotide monomers covalently bonded within a chain. DNA (e.g., cDNA) and RNA (e.g., mRNA) are examples of polynucleotides with different biological functions. A nucleotide is an organic molecule that functions as a monomer or subunit of a nucleic acid molecule such as DNA or RNA. The nucleic acid molecule or polynucleotide can be double-stranded and single-stranded, linear and circular. Preferably, it is contained within a vector contained in a host cell. The host cell can express the antibody construct, for example, after transformation or transfection using the vector or polynucleotide of the present invention. For that purpose, the polynucleotide or nucleic acid molecule is operably linked to a control sequence.

[0210] The genetic code is a set of rules for translating the coded information within genetic material (nucleic acids) into proteins. Biological decoding within living cells is performed by ribosomes that use transfer RNA (tRNA) molecules to carry amino acids and read three nucleotides in mRNA at a time, joining the amino acids in the order specified by the mRNA. This code defines how a sequence of three nucleotides, called a codon, specifies the amino acid to be added next during protein synthesis. With some exceptions, a codon of three nucleotides within a nucleic acid sequence specifies one amino acid. Since most genes are encoded by exactly the same code, this particular code is often referred to as the standard genetic code or reference genetic code. The genetic code determines the protein sequence of a particular coding region, but when and where these proteins are produced may be affected by other genomic regions.

[0211] Furthermore, the present invention provides a vector comprising the polynucleotide / nucleic acid molecule of the present invention.

[0212] A vector is a nucleic acid molecule used as a means of transmission for carrying (foreign) genetic material into cells. The term "vector" includes, but is not limited to, plasmids, viruses, cosmids, and artificial chromosomes. Generally, a genetically engineered vector contains an origin of replication, a multiple cloning site, and a selectable marker. The vector itself is generally a nucleotide sequence, usually a DNA sequence, that contains a larger sequence that serves as the "backbone" of the insert (transgene) and the vector. Recent vectors may contain additional features such as promoters, genetic markers, antibiotic resistance, reporter genes, targeting sequences, and protein purification tags in addition to the transgene insert and backbone. A vector called an expression vector (expression construct) is specifically for expressing a transgene in a target cell and generally has regulatory sequences.

[0213] The term "control array" refers to the DNA array necessary to express a functionally linked code array in a specific host organism. Control arrays suitable for prokaryotes include, for example, a promoter, optionally an operator array and a ribosome binding site. Eukaryotic cells are known to utilize a promoter, a polyadenylation signal, and an enhancer.

[0214] Nucleic acids are "functionally linked" when they are in a functional relationship with another nucleic acid array. For example, when the DNA of a pre-sequence or a secretion leader is expressed as a pre-protein involved in the secretion of a polypeptide, the DNA is functionally linked to the DNA of the polypeptide, and when a promoter or an enhancer affects the transcription of a coding array, they are functionally linked to the coding array, or when a ribosome binding site is arranged to facilitate translation, it is functionally linked to the coding array. Generally, "functionally linked" means that the DNA arrays to be linked are continuous, and in the case of a secretion leader, continuous and within the reading frame. However, an enhancer does not need to be continuous. Ligation is accomplished by ligation at convenient restriction sites. If such sites do not exist, synthetic oligonucleotide adapters or linkers are used according to conventional practice.

[0215] "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 when referring to the virus-mediated transfer of a nucleic acid molecule or polynucleotide. Transfection of animal cells usually involves creating transient pores or "holes" in the cell membrane to allow the uptake of substances. Transfection can be carried out using calcium phosphate, by electroporation, by cell squeezing, or by mixing a cationic lipid with a substance that produces liposomes, fusing with the cell membrane, and accumulating the internal cargo.

[0216] The term "transformation" is used when referring to the non-viral transfer of nucleic acid molecules or polynucleotides (including vectors) into bacteria and into eukaryotic cells other than animals, including plant cells. Thus, transformation is a genetic modification of bacteria or eukaryotic cells other than animals that results from the direct uptake from its surroundings through the cell membrane(s) and subsequent incorporation of foreign genetic material (nucleic acid molecules). Transformation can be effected by artificial means. To effect transformation, the cells or bacteria must be in a competent state in which transformation can occur as a time-limited response to environmental conditions such as starvation and cell density.

[0217] Furthermore, the present invention provides host cells transformed or transfected with the polynucleotide / nucleic acid molecule or vector of the present invention.

[0218] As used herein, the term "host cell" or "recipient cell" can be, or already is, a recipient of a vector encoding an antibody construct of the present invention, a foreign nucleic acid molecule, and a polynucleotide, and / or any individual cell or cell culture that includes a recipient of such antibody construct. The introduction of each substance into the cell is effected by transformation, transfection, etc. The term "host cell" is also intended to include the progeny or potential progeny of a single cell. Such progeny may not actually be identical to the parental cell (either morphologically or with respect to the genome or total DNA complement) because certain changes may occur due to any of natural, accidental, or intentional mutations, or due to environmental influences, but nonetheless are included within the scope of this term as used herein. Suitable host cells include prokaryotic or eukaryotic cells, and include, but are not limited to, bacteria, yeast cells, fungal cells, plant cells, and animal cells, such as insect cells and mammalian cells, such as mouse, rat, macaque, or human.

[0219] The antibody constructs of the present invention can be produced in bacteria. After expression, the antibody constructs of the present invention can be isolated from E. coli cell paste in the soluble fraction and purified, for example, by affinity chromatography and / or size exclusion. Final purification can be carried out in the same manner as the purification method of the antibody expressed in CHO cells, for example.

[0220] In addition to prokaryotes, eukaryotic microorganisms such as filamentous fungi or yeasts are suitable cloning or expression hosts for the antibody constructs of the present invention. Saccharomyces cerevisiae, i.e., common baker's yeast, is the most commonly used among lower eukaryotic host microorganisms. However, many other genera, species, and strains are generally available and useful in the present invention, for example, Kluyveromyces hosts such as Schizosaccharomyces pombe, 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; Yarrowia (EP 402 226); Pichia pastoris (EP 183 070); Candida; Trichoderma reesia (EP 244 234); Neurospora crassa; Schwanniomyces hosts such as Schwanniomyces occidentalis; and filamentous fungi such as Neurospora, Penicillium, Tolypocladium, and Aspergillus hosts such as A. nidulans and A. niger.

[0221] Host cells suitable for the expression of the glycosylated antibody constructs of the invention are obtained from multicellular organisms. Examples of invertebrate cells include plant cells and insect cells. Many baculovirus strains and variants, as well as corresponding permissive insect host cells derived from hosts such as Spodoptera frugiperda (caterpillar), Aedes aegypti (mosquito), Aedes albopictus (mosquito), Drosophila melanogaster (fruit fly), and Bombyx mori (silkworm) have been identified. Various virus strains for transfection, such as the L-1 variant of Autographa californica NPV and the Bm-5 strain of Bombyx mori NPV, are publicly available and such viruses can be used for the transfection of, in particular, Spodoptera frugiperda cells, as the virus of the present specification according to the invention.

[0222] Plant cell cultures of cotton, corn, potato, soybean, petunia, tomato, Arabidopsis thaliana, and tobacco can also be used as hosts. Cloning and expression vectors useful for protein production in plant cell cultures are known to those skilled in the art. See, for example, Hiatt et al., Nature (1989) 342:76-78, Owen et al. (1992) Bio / Technology 10:790-794, Artsaenko et al. (1995) The Plant J 8:745-750, and Fecker et al. (1996) Plant Mol Biol 32:979-986.

[0223] However, there is the highest interest in vertebrate cells, and the growth of vertebrate cells under culture (tissue culture) has become a conventional procedure. Examples of useful mammalian host cell lines are monkey kidney CV1 line transformed by SV40 (COS-7, ATCC CRL 1651); human fetal kidney line (293 cells or 293 cells subcloned for growth in suspension culture, Graham et al., J. Gen Virol. 36:59 (1977)); baby hamster kidney cells (BHK, ATCC CCL 10); Chinese hamster ovary cells / -DHFR (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA 77:4216 (1980)); mouse Sertoli cells (TM4, Mather, Biol. Reprod. 23:243-251 (1980)); monkey kidney cells (CVI ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL 1587); human cervical cancer cells (HELA, ATCC CCL 2); dog kidney cells (MDCK, ATCC CCL 34); buffalo rat liver cells (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human liver cells (Hep G2, 1413 8065); mouse mammary tumor (MMT 060562, ATCC CCL5 1); TRI cells (Mather et al., Annals N.Y Acad. Sci. (1982) 383:44-68); MRC 5 cells; FS4 cells; and human hepatoma line (Hep G2).

[0224] In a further embodiment, the present invention provides a method for producing an antibody construct of the present invention, the method comprising culturing a host cell of the present invention under conditions that allow expression of the antibody construct of the present invention, and recovering the produced antibody construct from the culture.

[0225] As used herein, the term "culturing" refers to the in vitro maintenance, differentiation, growth, proliferation, and / or propagation of cells under suitable conditions in a medium. The term "expression" includes any step involved in the production of the antibody constructs of the present invention, including, but not limited to, transcription, post-transcriptional modification, translation, post-translational modification, and secretion.

[0226] When using recombinant techniques, antibody constructs can be produced intracellularly, i.e., within the periplasmic space of the cell, or secreted directly into the medium. When producing antibody constructs intracellularly, as a first step, for example, particulate debris of host cells or lysed fragments is removed by centrifugation or ultrafiltration. Carter et al., Bio / Technology 10:163 - 167 (1992) describes a procedure for isolating antibodies secreted into the periplasmic space of E. coli. Briefly, cell paste is thawed over about 30 minutes in the presence of sodium acetate (pH 3.5), EDTA, and phenylmethylsulfonyl fluoride (PMSF). Cell debris can be removed by centrifugation. When secreting antibodies into the medium, generally the supernatant from such an expression system is first concentrated using a commercially available protein concentration filter, such as an Amicon or Millipore Pellicon ultrafiltration unit. To inhibit proteolysis, a protease inhibitor such as PMSF may be added at any of the foregoing steps, and an antibiotic may be added to prevent the growth of adventitious impurities.

[0227] The antibody constructs of the present invention prepared from host cells can be recovered or purified using, for example, hydroxyapatite chromatography, gel electrophoresis, dialysis, and affinity chromatography. Depending on the antibody to be recovered, other protein purification techniques such as fractionation on an ion exchange column, ethanol precipitation, reverse phase HPLC, chromatography on silica, chromatography on heparin SEPHAROSE™, chromatography on an anion or cation exchange resin (e.g., polyaspartic acid column), chromatofocusing, SDS-PAGE, and ammonium sulfate precipitation are also available. When the antibody constructs of the present invention contain the CH3 domain, Bakerbond ABX resin (J.T. Baker, Phillipsburg, NJ) is useful for purification.

[0228] Affinity chromatography is a 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 processing times than can be achieved with agarose.

[0229] Furthermore, the present invention provides a pharmaceutical composition comprising the antibody constructs of the present invention or antibody constructs made according to the method of the present invention.

[0230] As used herein, the term "pharmaceutical composition" relates to a composition suitable for administration to a patient, preferably a human patient. Particularly preferred pharmaceutical compositions of the present invention contain one or more antibody constructs (s) of the present invention, preferably in a therapeutically effective amount. Preferably, the pharmaceutical composition further comprises a suitable formulation of one or more (pharmaceutically effective) carriers, stabilizers, excipients, diluents, solubilizers, surfactants, emulsifiers, preservatives, and / or adjuvants. The components of the acceptable composition are preferably non-toxic to the recipient at the dosage and concentrations used. The pharmaceutical compositions of the present invention include, but are not limited to, liquid, frozen, and lyophilized compositions.

[0231] The composition of the present invention may contain a pharmaceutically acceptable carrier. Generally, as used herein, "pharmaceutically acceptable carrier" means any aqueous and non-aqueous solution, sterile solution, solvent, buffer, such as phosphate buffered saline (PBS) solution, water, suspension, emulsion such as oil / water emulsion, various types of wetting agents, liposomes, dispersion medium, and coating that are compatible with pharmaceutical administration, particularly 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.

[0232] Certain embodiments provide a pharmaceutical composition comprising the antibody construct of the present invention and further one or more excipients, such as those exemplified in this section and elsewhere in this specification. Excipients can be used in the present invention, for example, to adjust the physical, chemical, or biological properties of the formulation, such as adjusting viscosity, and / or to improve the effect and / or stabilize such a formulation, and also to address a wide range of purposes, such as countermeasures against degradation and damage caused by the loads occurring during manufacturing, transportation, storage, preparation before use, administration, and subsequent periods.

[0233] In certain embodiments, the pharmaceutical composition may contain formulation materials for the purpose of modifying, sustaining, or protecting, for example, the pH, osmolality, viscosity, clarity, color, isotonicity, odor, sterility, stability, dissolution or release rate, absorbability, or permeability of the composition (see REMINGTON’S PHARMACEUTICAL SCIENCES, 18” Edition, (A.R.Genrmo, ed.), 1990, Mack Publishing Company). In such embodiments, suitable formulation materials can include, but are not limited to: · Charged amino acids, preferably amino acids containing lysine, lysine acetate, arginine, glutamate, and / or histidine, such as glycine, alanine, glutamine, asparagine, threonine, proline, 2-phenylalanine; ·Antimicrobial agents such as antibacterial and antifungal agents; ·Antioxidants such as ascorbic acid, methionine, sodium sulfite, or sodium bisulfite; ·Buffers, buffer systems, and buffering agents used to maintain the composition within a physiological pH, or slightly lower pH, usually in the pH range of about 5 to about 8 or 9; Examples of buffers are borate, bicarbonate, Tris-HCl, citrate, phosphate, or other organic acids, succinate, phosphate, histidine, and acetate; for example, Tris buffer at about pH 7.0 - 8.5, or acetate buffer at about pH 4.0 - 5.5; ·Non-aqueous solvents such as propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate; ·Aqueous carriers including water, alcohol / aqueous solutions, emulsions or suspensions, including physiological saline and buffer media; ·Biodegradable polymers such as polyesters; ·Fillers such as mannitol or glycine; ·Chelating agents such as ethylenediaminetetraacetic acid (EDTA); ·Isotonic agents and absorption retardants; ·Complexing agents such as caffeine, polyvinylpyrrolidone, beta-cyclodextrin, or hydroxypropyl-beta-cyclodextrin; ·Bulking agents; ·Monosaccharides; disaccharides; and other carbohydrates (such as glucose, mannose, or dextrin); The carbohydrate can be a non-reducing sugar, preferably trehalose, sucrose, octasulfate, sorbitol or xylitol; ·(Low molecular weight) proteins, polypeptides, or proteinaceous carriers such as human or bovine serum albumin, gelatin, or preferably immunoglobulins of human origin; ·Coloring and flavoring agents; ·Sulfur-containing reducing agents such as glutathione, thioctic acid, sodium thioglycolate, thioglycerol, [alpha]-monothioglycerol, and sodium thiosulfate ·Diluents; ·Emulsifying agents; · Hydrophilic polymers such as polyvinylpyrrolidone; · Salt-forming counterions such as sodium; · Preservatives such as antimicrobial agents, antioxidants, chelating agents, inert gases (examples are benzalkonium chloride, benzoic acid, salicylic acid, thimerosal, phenethyl alcohol, methylparaben, propylparaben, chlorhexidine, sorbic acid, or hydrogen peroxide); · Metal complexes such as Zn-protein complexes; · Solvents and co-solvents (such as glycerin, propylene glycol, or polyethylene glycol); · Sugars and sugar alcohols, such as trehalose, sucrose, octasulfate, mannitol, sorbitol or xylitol stachyose, mannose, sorbose, xylose, ribose, myo-inositol (myoinisitose), galactose, lactitol, ribitol, myo-inositol, galactitol, glycerol, cyclitol (e.g., inositol), polyethylene glycol; and polyhydric sugar alcohols; · Suspending agents; · Surfactants or wetting agents, such as pluronics, PEG, sorbitan esters, polysorbates, such as polysorbate 20, polysorbate, triton, tromethamine, lecithin, cholesterol, tyloxapal; the surfactant may preferably be a cleaning solution having a molecular weight > 1.2 KD, and / or a polyether preferably having a molecular weight > 3 KD; non-limiting examples of preferred cleaning solutions are Tween 20, Tween 40, Tween 60, Tween 80, and Tween 85; non-limiting examples of preferred polyethers are PEG 3000, PEG 3350, PEG 4000, and PEG 5000; · Stability enhancers such as sucrose or sorbitol; · Isotonicity enhancers, such as alkali metal halides, preferably sodium chloride or potassium chloride, mannitol sorbitol; ·A parenteral delivery vehicle comprising a sodium chloride solution, Ringer's dextrose, dextrose and sodium chloride, lactated Ringer's solution, or a non-volatile oil; ·An intravenous delivery vehicle comprising a body fluid and nutritional supplement solution, an electrolyte replacement solution (such as one based on Ringer's dextrose).

[0234] It is obvious to those skilled in the art that different components of the pharmaceutical composition (for example, those listed above) can have different effects. For example, an amino acid can act as a buffer, a stabilizer, and / or an antioxidant, mannitol can act as a filler and / or an isotonicity enhancer, and sodium chloride can act as a delivery vehicle and / or an isotonicity enhancer, etc.

[0235] It is envisioned that the composition of the present invention may, depending on the intended use of the composition, in addition to the polypeptide of the present invention as defined herein, further contain additional biologically active agents. Such agents can be drugs acting on the gastrointestinal system, drugs acting as cell growth inhibitors, drugs preventing hyperurikemia, drugs inhibiting the immune response (such as corticosteroids), drugs regulating the inflammatory response, drugs acting on the circulatory system, and / or agents such as cytokines known in the art. It is also envisioned that the antibody construct of the present invention is applied in combination therapy, that is, in combination with another anticancer drug.

[0236] In certain embodiments, the optimal pharmaceutical composition is determined by one of ordinary skill in the art, e.g., according to the intended route of administration, delivery format, and desired dosage. See, e.g., REMINGTON’S PHARMACEUTICAL SCIENCES, supra. In certain embodiments, such compositions may affect the physical state, stability, in vivo release rate, and in vivo clearance rate of the antibody constructs of the invention. In certain embodiments, the major vehicle or carrier in the pharmaceutical composition may be either essentially aqueous or non-aqueous. For example, suitable vehicles or carriers can be water for injection, aqueous saline solution, or artificial cerebrospinal fluid, optionally supplemented with other substances common in parenteral compositions. Neutral buffered saline or saline mixed with serum albumin are further exemplary vehicles. In certain embodiments, the antibody constructs of the compositions of the invention can be prepared for storage in the form of a lyophilized cake or aqueous solution by mixing a selected composition having the desired degree of purity with any excipient (Remington’s Pharmaceutical Sciences, supra). Further, in certain embodiments, the antibody constructs of the invention can be formulated as a lyophilized product using suitable excipients such as sucrose.

[0237] When parenteral administration is contemplated, the therapeutic composition used in the present invention can be provided in the form of a pyrogen-free parenterally acceptable aqueous solution containing the antibody construct of the present invention in a pharmaceutically acceptable vehicle. A vehicle particularly suitable for parenteral injection is sterile distilled water, and the antibody construct of the present invention is formulated as a sterile isotonic solution properly stored therein. In certain embodiments, the preparation may include a formulation of the agent, such as injectable microspheres, biodegradable particles, polymeric compounds (polylactic acid or polyglycolic acid), beads, or liposomes, which can provide controlled or sustained release of the product deliverable via depot injection, and the molecule of interest. In certain embodiments, hyaluronic acid, which has the effect of improving the duration in the blood, may also be used. In certain embodiments, a implantable drug delivery device can be used to introduce the antibody construct of interest.

[0238] Additional pharmaceutical compositions will be apparent to those skilled in the art and include formulations containing the antibody constructs of the invention in sustained or controlled release delivery systems. A variety of other sustained or controlled release means, such as liposome carriers, biodegradable microparticles, or porous beads and depot injections, are also known to those skilled in the art. See, for example, International Patent Application No. PCT / US93 / 00829, which describes the controlled release of porous polymer microparticles for the delivery of pharmaceutical compositions. Sustained release preparations may include a semipermeable polymer matrix in the form of a shaped article, such as a film or a microcapsule. Sustained release matrices may include polyesters, hydrogels, polylactides (described in U.S. Patent No. 3,773,919, European Patent Application Publication No. EP058481), copolymers of L-glutamic acid and gamma ethyl-L-glutamate (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, supra), or poly-D(-)-3-hydroxybutyric acid (European Patent Application Publication No. EP133,988). Sustained release compositions may also include liposomes which can be prepared by any of several methods known in the art. See, for example, Eppstein et al., 1985, Proc. Natl. Acad. Sci. U.S.A. 82:3688-3692; European Patent Application Publication Nos. EP036,676; EP088,046 and EP143,949.

[0239] The antibody construct may also be encapsulated in, for example, microcapsules prepared by coacervation techniques or interfacial polymerization (e.g., hydroxymethylcellulose or gelatin microcapsules and poly(methylmethacrylate) microcapsules, respectively), in colloidal drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles, and nanocapsules), or in macroemulsions. Such techniques are disclosed in Remington’s Pharmaceutical Sciences, 16th edition, Oslo, A. Ed. (1980).

[0240] Pharmaceutical compositions for in vivo administration are usually provided as sterile preparations. Sterilization can be carried out by filtration through a sterile filtration membrane. When the composition is lyophilized, sterilization using this method may be carried out either before or after lyophilization and reconstitution. Compositions for parenteral administration can be stored in lyophilized form or in solution. Parenteral compositions are generally placed in a container having a sterile access port, e.g., an intravenous fluid bag or vial having a stopper pierceable by a hypodermic needle.

[0241] Another aspect of the invention includes a self-buffering antibody construct of a formulation of the invention that can be used as a pharmaceutical composition as described in International Patent Application WO06138181A2 (PCT / US2006 / 022599). Various explanations are available regarding the stabilization of proteins and useful formulation materials and methods in this regard, for example, 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), especially with regard to protein pharmaceuticals and processes for veterinary and / or human medical use, refer particularly to the parts regarding the same excipients and processes as the self-buffering protein formulations according to the present invention.

[0242] Salts can be used, according to certain embodiments of the present invention, for example, to adjust the ionic strength and / or isotonicity of a formulation and / or to improve the solubility and / or physical stability of a protein or other component of a composition according to the present invention. As is well known, ions can stabilize native proteins by binding to charged residues on the surface of the protein and by shielding charged and polar groups in the protein, thereby reducing the strength of its electrostatic interactions, attractive and repulsive interactions. Ions can also stabilize denatured proteins, particularly by binding to the denatured peptide bonds (--CONH) of the protein. Furthermore, ionic interactions with charged and polar groups in the protein can also reduce intermolecular electrostatic interactions, thereby preventing or reducing protein aggregation and insolubilization.

[0243] Ionic species have significantly different effects on proteins. Many rankings of ions and their effects on proteins by category have been elucidated and can be utilized in the formulation of pharmaceutical compositions according to the present invention. One example is the Hofmeister series, which ranks ionic and polar non-ionic solutes according to their effect on the conformational stability of proteins in solution. Solutes that stabilize are termed "cosmotropic". Solutes that destabilize are termed "chaotropic". Cosmotropes are generally used at high concentrations (e.g., >1 M ammonium sulfate) to precipitate proteins from solution ("salting out"). Chaotropes are generally used to denature and / or solubilize proteins ("salting in"). The position of an ion in the Hofmeister series is defined by its relative effects on "salting in" and "salting out".

[0244] Free amino acids can be used in the antibody constructs of the formulations of the present invention according to various embodiments of the present invention as fillers, stabilizers, and antioxidants, as well as in other standard applications. Lysine, proline, serine, and alanine can be used to stabilize proteins in the formulation. Glycine is useful in ensuring the proper structure and properties of the cake in lyophilization. Arginine can be useful in preventing protein aggregation in both liquid and lyophilized formulations. Methionine is useful as an antioxidant.

[0245] Polyols include sugars such as mannitol, sucrose, and sorbitol, as well as polyhydric alcohols such as glycerol and propylene glycol, and polyethylene glycol (PEG) and related substances for the purposes of discussion herein. Polyols are cosmotropic. Polyols are useful stabilizers for protecting proteins from physical and chemical degradation in both liquid and lyophilized formulations. Polyols are also useful in adjusting the isotonicity of the formulation. A particularly useful polyol in selected embodiments of the present invention is mannitol, which is widely used to ensure the structural stability of the cake in lyophilized formulations. Mannitol ensures the structural stability of the cake. Generally, this is used in combination with a lyoprotectant such as sucrose. Sorbitol and sucrose are particularly preferred as agents for adjusting isotonicity and as stabilizers for protecting against freeze-thaw stress during transportation or during bulk preparation in the manufacturing process. Reducing sugars (containing free aldehyde or ketone groups), such as glucose and lactose, can glycosylate surface lysine and arginine residues. Thus, generally it is not a particularly preferred polyol for use according to the present invention. In addition, sugars that form such reactive species, such as sucrose, are also not particularly preferred polyols of the present invention in that they are hydrolyzed to fructose and glucose under acidic conditions, resulting in fructose glycosylation. PEG is useful in stabilizing proteins and as a cryoprotectant and can be used in the present invention from this perspective.

[0246] Embodiments of the antibody constructs of the formulations of the present invention further comprise a surfactant. Protein molecules can be prone to adsorption to surfaces and to denaturation and consequent aggregation at gas - liquid, solid - liquid, and liquid - liquid interfaces. These effects generally vary inversely with protein concentration. These detrimental interactions generally vary inversely with protein concentration and are exacerbated mainly by physical agitation, for example, during product transportation and handling. Surfactants are conventionally 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 widely used to control the conformational stability of proteins. The use of surfactants in this regard is protein - specific, as any given surfactant will typically stabilize some proteins and destabilize others.

[0247] Polysorbates are prone to oxidative degradation and, when supplied, contain sufficient amounts of peroxide to cause oxidation of the side chains of protein residues, particularly methionine. Therefore, polysorbates should be used with caution and, when used, need to be used at the lowest effective concentration. In this regard, polysorbates are an example of the general rule that excipients should be used at the lowest effective concentration.

[0248] Embodiments of the antibody constructs of the formulations of the present invention further include one or more antioxidants. By maintaining appropriate levels of ambient oxygen and ambient temperature, as well as by avoiding exposure to light, harmful oxidation of proteins in pharmaceutical formulations can be prevented to some extent. Antioxidant excipients can be used similarly to prevent oxidative degradation of proteins. Antioxidants that are particularly useful in this regard are reducing agents, oxygen / free radical scavengers, and chelating agents. The antioxidants used in the therapeutic protein formulations according to the present invention are preferably water-soluble and maintain their activity over the shelf life of the product. In this regard, EDTA is a preferred antioxidant according to the present invention. Antioxidants can potentially damage proteins. For example, reducing agents, such as glutathione, can particularly break intramolecular disulfide linkages. Therefore, the antioxidants used in the present invention are particularly selected such that they themselves are unlikely or have a sufficiently low likelihood of damaging the proteins in the formulation.

[0249] The formulations according to the present invention can include metal ions that are protein cofactors and are required to form protein coordination complexes, for example, zinc required to form certain insulin suspensions. Metal ions can also inhibit some of the processes that degrade proteins. However, metal ions can also catalyze the physical and chemical processes that degrade proteins. Magnesium ions (10 - 120 mM) can be used to inhibit the isomerization of aspartic acid to isoaspartic acid. Ca +2 ions (up to 100 mM) can improve the stability of human deoxyribonuclease. However, Mg +2 , Mn +2 , and Zn +2 can destabilize rhDNase. Similarly, Ca +2 and Sr +2 can stabilize factor VIII, which can be destabilized by Mg +2 , Mn +2 , and Zn +2 , Cu +2 and Fe +2 and its aggregation can be caused by Al +3It can be increased by ions.

[0250] Embodiments of the antibody constructs of the formulations of the present invention further comprise one or more preservatives. Preservatives are required when developing multi-dose parenteral formulations involving multiple collections from the same container. Their main function is to inhibit the growth of microorganisms over the shelf life or period of use of the pharmaceutical product and ensure the sterility of the product. Widely used preservatives include benzyl alcohol, phenol, and m-cresol. Preservatives have a long history of use in low molecular weight parenteral drugs, but 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 multi-dose protein formulations. To date, most protein drugs have been formulated only for single use. However, if multi-dose formulations were possible, the advantages of patient convenience and high marketability would be added. The development of a formulation with a storage process led to the commercialization of a highly convenient multi-use injection pen. Human growth hormone (hGH) is a good example. At least four such pen devices containing a stored formulation of hGH are currently available on the market. Norditropin (liquid, Novo Nordisk), Nutropin AQ (liquid, Genentech), and Genotropin (lyophilized - dual chamber cartridge, Pharmacia & Upjohn) contain phenol, and Somatrope (Eli Lilly) is formulated with m-cresol. Several aspects need to be considered during the formulation and development of the stored dosage form. The effective preservative concentration in the pharmaceutical product must be optimized. To do so, it is necessary to test a given preservative in the dosage form within a concentration range that confers an antimicrobial effect without compromising the stability of the protein.

[0251] As can be expected, the development of liquid formulations containing preservatives is more difficult than that of lyophilized formulations. Lyophilized products are lyophilized without preservatives and can be reconstituted with a diluent containing preservatives at the time of use. This shortens the contact time of the preservative with the protein and significantly minimizes the associated stability risks. In the case of liquid formulations, the effectiveness and stability of the preservative should be maintained over the entire product shelf life (about 18 to 24 months). As an important point to note, the effectiveness of the preservative needs to be demonstrated in the final formulation containing the active drug and all excipient components.

[0252] The antibody constructs disclosed herein can also be formulated as immunoliposomes. A "liposome" is a small vesicle composed of various types of lipids, phospholipids, and / or surfactants useful for drug delivery to mammals. The components of liposomes are generally arranged in a bilayer format, similar to the lipid arrangement of biological membranes. Liposomes containing antibody constructs can be prepared by methods known in the art, such as those described in Epstein et al., Proc. Natl. Acad. Sci. USA, 82:3688 (1985); Hwang et al., Proc. Natl Acad. Sci. USA, 77:4030 (1980); U.S. Patent Nos. 4,485,045 and 4,544,545; and WO97 / 38731. Liposomes that remain in the blood for an extended period are disclosed in U.S. Patent No. 5,013,556. Particularly useful liposomes can be prepared by the reverse-phase evaporation method using a lipid composition containing phosphatidylcholine, cholesterol, and PEG-derivatized phosphatidylethanolamine (PEG-PE). To obtain liposomes of a desired diameter, the liposomes are extruded through a filter with a predetermined pore size. The Fab' fragment of the antibody construct of the present invention can be conjugated to liposomes by a disulfide exchange reaction, as described in Martin et al. J. Biol. Chem. 257:286-288 (1982). Optionally, a chemotherapeutic agent is encapsulated within the liposome. See Gabizon et al. J. National Cancer Inst. 81 (19)1484(1989).

[0253] After formulating the pharmaceutical composition, it can be stored in a sterile vial as a solution, suspension, gel, emulsion, solid, crystal, or as a dehydrated powder or lyophilized powder. Such formulations can be stored either in an immediate-use form or in a form that is reconstituted (e.g., lyophilized) prior to administration.

[0254] The biological activity of the pharmaceutical composition as defined herein can be determined, for example, by the cytotoxicity assays described in the following examples, in WO99 / 54440, or by Schlereth et al. (Cancer Immunol. Immunother. 20 (2005), 1-12). As used herein, the term "efficacy" or "in vivo efficacy" refers to the responsiveness to treatment with the pharmaceutical composition of the invention, for example using the standardized NCI response criteria. The success or in vivo efficacy of a therapy using the pharmaceutical composition of the invention refers to the efficacy of the composition against its intended purpose, i.e., its ability to cause the desired effect, i.e., a decrease in diseased cells, such as tumor cells. In vivo efficacy can be monitored by established standard methods for each disease entity, including but not limited to white blood cell counts, differentials, fluorescence-activated cell sorting, bone marrow aspiration. In addition, various disease-specific clinical chemistry parameters and other established standard methods can be used.Furthermore, computed tomography, X-rays, magnetic resonance imaging (e.g., response assessment based on the National 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, white blood cell count, differential, fluorescence-activated cell sorting, bone marrow aspiration, lymph node biopsy / histology, and clinical chemistry parameters specific to various lymphomas (e.g., lactate dehydrogenase), as well as other established standard methods can be used.

[0255] Another major challenge in the development of drugs such as the pharmaceutical compositions of the present invention is the predictable modulation of pharmacokinetic properties. For that purpose, it is possible to demonstrate the pharmacokinetic profile of a candidate drug, i.e., the profile of pharmacokinetic parameters that affect the ability of a particular drug to treat a given pathological condition. Pharmacokinetic parameters of a drug that affect the ability of the drug to treat a particular disease include, but are not limited to, half-life, volume of distribution, first-pass metabolism in the liver, and blood serum binding. The efficacy of a given drug can be affected by each of the above parameters.

[0256] "Half-life" means the time it takes for 50% of the administered drug to be eliminated through biological processes such as metabolism, excretion, etc. "First-pass metabolism in the liver" means the tendency of a drug to be metabolized upon first contact with the liver, i.e., during the first passage 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, as well as the degree of distribution of the drug within these compartments. "Blood serum binding" means the tendency of a drug to interact and bind with blood serum proteins such as albumin, resulting in a decrease or loss of the biological activity of the drug.

[0257] Pharmacokinetic parameters also include bioavailability, lag time (Tlag), Tmax, absorption rate, more onset, and / or Cmax for a given dose of the administered drug. "Bioavailability" means the amount of drug in the blood compartment. "Lag time" means the delay time from the administration of the drug to the detectable and measurable stage of the drug in the blood or plasma. "Tmax" is the time at which the maximum blood concentration of the drug is achieved thereafter, and "Cmax" is the maximum blood concentration obtained by a given drug. The time until the drug reaches the blood concentration or tissue concentration required for a biological effect is affected by all parameters. The pharmacokinetic parameters of bispecific antibody constructs showing species specificity that can be determined in preclinical animal studies in primates other than chimpanzees, as outlined above, are also defined in publications such as those by Schlereth et al. (Cancer Immunol. Immunother. 20(2005), 1-12).

[0258] In one embodiment, there is provided an antibody construct of the present invention or an antibody construct prepared according to the method of the present invention for use in the prevention, treatment, or amelioration of a tumor or cancer disease, or a metastatic cancer disease.

[0259] The formulations described herein are useful as pharmaceutical compositions for treating, ameliorating, and / or preventing the pathological medical conditions described herein in a patient in need thereof. The term "treatment" refers to both therapeutic treatment and prophylactic or palliative means. Treatment includes the application or administration of a formulation to a patient, isolated tissue, or cell having a disease / disorder, symptoms of a disease / disorder, or causative factor of a disease / disorder, with the goal of curing, alleviating / relieving, altering, restoring, ameliorating / improving the disease, symptoms of the disease, or causative factor of the disease / disorder, or having an effect thereon.

[0260] As used herein, the term "amelioration" refers to the improvement of the disease state of a patient having a tumor or cancer, or metastatic cancer as set forth below in the present specification, by administration of an antibody construct according to the present invention to a subject in need thereof. Such improvement may also be seen as a slowing or stopping of the progression of the patient's tumor or cancer, or metastatic cancer. As used herein, the term "prevention" means the avoidance of the onset or recurrence of a tumor or cancer, or metastatic cancer as set forth below in the present specification, in a patient having the same, by administration of an antibody construct according to the present invention to a subject in need thereof.

[0261] The term "disease" refers to any medical condition for which benefit may be obtained from treatment with the antibody constructs or pharmaceutical compositions described herein. This includes chronic and acute disorders or diseases, including pathological conditions that predispose a mammal to susceptibility to the disease.

[0262] A "neoplasm" is an abnormal growth of tissue that usually, but not always, forms a tumor. When a neoplasm forms a tumor, it is generally called a "tumor". A neoplasm or tumor can be benign, potentially malignant (precancerous), or malignant. Malignant neoplasms are generally called cancers. Malignant neoplasms usually invade and destroy surrounding tissue and can form metastases, i.e., spread to other parts, tissues, or organs of the body. Thus, the term "metastatic cancer" encompasses metastases to tissues or organs other than those of the primary tumor. Lymphomas and leukemias are lymphoid neoplasms. For the purposes of the present invention, these are also included within the terms "tumor" or "cancer".

[0263] In a preferred embodiment of the present invention, the tumor or cancer disease is selected from the group consisting of, but not limited to, lung cancer, preferably SCLC, breast, cervical, colon, colorectal, endometrial, head and neck, liver, ovary, pancreas, prostate, skin, stomach, testis, thyroid, adrenal, kidney, bladder, uterus, esophagus, urothelium, and brain tumors or cancers, lymphomas, carcinomas, and sarcomas, and metastatic cancer diseases derived from any of the foregoing.

[0264] More specifically, the tumor or cancer disease can be selected from the group consisting of small cell lung cancer (SCLC), non-small cell lung cancer (NSCLC), glioma, glioblastoma, melanoma, neuroendocrine prostate cancer, neuroendocrine pancreatic cancer, hepatoblastoma, and hepatocellular carcinoma. The metastatic cancer disease can be derived from any of the foregoing.

[0265] The present invention also provides a method for treating or ameliorating a tumor or cancer disease, or a metastatic cancer disease, comprising administering to a subject in need thereof an antibody construct of the present invention or an antibody construct made according to the method of the present invention.

[0266] The terms "subject in need thereof" or "in need of treatment" include subjects already having the disorder as well as subjects seeking to prevent the disorder. The subject in need thereof or "patient" includes human and other mammalian subjects undergoing either prophylactic or therapeutic treatment.

[0267] The antibody constructs of the present invention are generally designed in accordance with a particular route of administration and method of administration, a particular dosage and dosing frequency, and the treatment of a particular disease, particularly in terms of bioavailability and duration, and in particular. The components of the composition are preferably formulated at a concentration acceptable at the site of administration.

[0268] Thus, the formulations and compositions can be designed in accordance with the present invention for delivery by any suitable route of administration. In the context of the present invention, the routes of administration are · topical routes (e.g., skin, inhalation, nasal, ocular, auricular / aural, vaginal, mucosal); · enteral routes (e.g., oral, gastrointestinal, sublingual, sublabial, buccal, rectal); and · parenteral routes (e.g., intravenous, intraarterial, intramedullary, intramuscular, intracerebral, intraventricular, epidural, subarachnoid, subcutaneous, intraperitoneal, extraamniotic, intraarticular, intracardiac, intradermal, intralesional, intrauterine, intravesical, intravitreal, transdermal, intranasal, transmucosal, intrasynovial, intraluminal), including but not limited to these.

[0269] The pharmaceutical compositions and antibody constructs of the present invention are particularly useful for parenteral administration, such as subcutaneous or intravenous delivery, by injection, such as a bolus injection, or by infusion, such as a continuous infusion. The pharmaceutical composition can be administered using a medical device. Examples of medical devices for administering the pharmaceutical composition are described in U.S. Patent Nos. 4,475,196; 4,439,196; 4,447,224; 4,447,233; 4,486,194; 4,487,603; 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.

[0270] In particular, the present invention achieves continuous administration of a suitable composition. As a non-limiting example, continuous or substantially continuous administration, i.e., uninterrupted administration, can be achieved by a small pump system worn by the patient that measures the inflow of the therapeutic agent into the patient's body. The pharmaceutical composition containing the antibody construct of the present invention can be administered by using the pump system. Such pump systems are generally known in the art and usually require regular replacement of a cartridge containing the therapeutic agent to be infused. In such pump systems, when replacing the cartridge, there may be a temporary interruption in the inflow of the therapeutic agent into the patient's body that is not interrupted except during the replacement. Even in such cases, it is considered within the scope of the pharmaceutical means and methods of the present invention in the sense that one "continuous administration" of such a therapeutic agent is constituted both in the administration stage before cartridge replacement and in the administration stage after cartridge replacement.

[0271] Continuous or uninterrupted administration of the antibody construct of the present invention can be intravenous or subcutaneous administration by a fluid delivery device or a small pump system that includes a fluid delivery mechanism for delivering fluid from a reservoir and an actuation mechanism for actuating the delivery mechanism. The pump system for subcutaneous administration may include a needle or cannula for puncturing the patient's skin and delivering a suitable composition into the patient's body. The pump system can be directly fixed or worn on the patient's skin, directly contacting the pump system and the patient's skin, regardless of whether it is a vein, artery, or blood vessel. This pump system can be worn on the patient's skin for 24 hours to several days. There may also be a small pump system with a small reservoir volume. As a non-limiting example, the volume of the reservoir suitable for the pharmaceutical composition to be administered can be 0.1 to 50 ml.

[0272] Continuous administration may also be by transdermal administration using a patch that is applied to the skin and replaced at regular intervals. Those skilled in the art are familiar with patch systems for drug delivery suitable for this purpose. An advantage of transdermal administration is that, for example, a new second patch can be applied to the skin surface immediately adjacent to the first used patch immediately before removing the first used patch, and at the same time the replacement of the first used patch can be completed, making it particularly suitable for uninterrupted administration. There are no problems of interruption of influx or battery depletion.

[0273] If the pharmaceutical composition is lyophilized, the lyophilized material is first reconstituted with a suitable liquid before administration. The lyophilized material can be reconstituted, for example, with bacteriostatic water for injection (BWFI), physiological saline, phosphate-buffered saline (PBS), or the same formulation in which the protein was present before lyophilization.

[0274] The compositions of the present invention can be administered to a subject at a suitable dose that can be determined, for example, by a dose escalation study in which increasing doses of the antibody constructs of the present invention that exhibit the interspecies specificity described herein are administered to non-human primates other than chimpanzees, such as macaques. As described above, the antibody constructs of the present invention that exhibit the interspecies specificity described herein have the advantage that they can be used in preclinical studies in non-human primates other than chimpanzees and as drugs in humans in the same form. The dosing schedule will be determined by the attending physician according to clinical factors. As is well known in the medical field, the dosage for a particular patient varies depending on many factors, including the patient's physique, body surface area, age, the individual compound being administered, sex, time and route of administration, general health status, and other drugs being administered simultaneously.

[0275] The terms "effective amount" or "effective dosage" are defined as an amount sufficient to achieve or at least partially achieve the desired effect. The term "therapeutically effective amount" is defined as an amount sufficient to cure or at least partially arrest the disease and its complications in a patient already suffering from the disease. The amount or dosage effective for this use will vary depending on the condition being treated (indication), the antibody construct being delivered, the nature and purpose of the treatment, the severity of the disease, the treatment history, the patient's medical history and responsiveness to the therapeutic agent, the route of administration, the physical build (weight, body surface area, or organ size), and / or the patient's condition (age and general health), as well as the overall state of the patient's immune system. The appropriate dosage can be adjusted according to the judgment of the attending physician so as to be administered to the patient in a single dose or multiple doses and to obtain the optimal therapeutic effect.

[0276] Typical dosages can range from about 0.1 μg / kg to up to about 30 mg / kg or more, depending on the above factors. In certain embodiments, the dosage can range from 1.0 μg / kg to about 20 mg / kg, optionally from 10 μg / kg to about 10 mg / kg, or from 100 μg / kg to about 5 mg / kg.

[0277] A therapeutically effective amount of the antibody construct of the present invention preferably reduces the severity of disease symptoms, increases the frequency or duration of the disease-free period, or prevents the functional or disability impairment resulting from the disease pain. In the case of treating DLL3-expressing tumors, a therapeutically effective amount of the antibody construct of the present invention, such as an anti-DLL3 / anti-CD3 antibody construct, preferably inhibits cell growth or tumor growth by at least about 20%, at least about 40%, at least about 50%, at least about 60%, at least about 70%, at least about 80%, or at least about 90% compared to untreated patients. The ability of a compound to inhibit tumor growth can be evaluated in animal models that predict efficacy in human tumors.

[0278] The pharmaceutical composition can be administered in a single treatment or, if necessary, in additional treatments such as anticancer therapy, for example, in combination with other proteinaceous and non-proteinaceous drugs. These drugs can be administered simultaneously with the composition containing the antibody construct of the present invention as defined herein, or separately at a defined time interval and dose before or after administration of the antibody construct.

[0279] As used herein, the term "effective and non-toxic dose" refers to an acceptable dose of the antibody construct of the present invention that is sufficient to cause a reduction in diseased cells, removal of a tumor, regression of a tumor, or stabilization of a disease without causing or essentially causing significant toxic effects. Such effective and non-toxic doses can be determined, for example, by dose escalation studies described in the art, and the dose should be below the dose that induces serious adverse side events (dose-limiting toxicity, DLT).

[0280] As used herein, the term "toxicity" refers to the toxic effects of a drug that appear as adverse events or serious adverse events. These side events may refer to a lack of systemic drug tolerance and / or a lack of local tolerance after administration. Toxicity may also include teratogenic or carcinogenic effects caused by the drug.

[0281] As used herein, the terms "safety", "in vivo safety", or "tolerance" are defined as the administration of a drug that does not induce severe adverse events immediately after administration (local tolerance) and during a longer drug application period. "Safety", "in vivo safety" or "tolerance" can be evaluated regularly, for example, during treatment and follow-up. Measurements include clinical evaluations, such as signs of organs, and screening for abnormal test values. Clinical evaluations can be performed and deviations from normal findings can be recorded / coded according to the NCI-CTC and / or MedDRA standards. Signs of organs can include criteria such as those shown in the Common Terminology Criteria for adverse events v3.0 (CTCAE), such as allergy / immunology, blood / bone marrow, cardiac arrhythmia, coagulation, etc. Test parameters that can be tested include, for example, hematology, clinical chemistry, coagulation profile, and urine tests, as well as tests on other body fluids, such as serum, plasma, lymph fluid, or spinal fluid, cerebrospinal fluid, etc. Thus, safety can be evaluated, for example, by physical examination, imaging techniques (i.e., ultrasound, x-ray, CT scan, magnetic resonance imaging (MRI), other measurements using industrial equipment (i.e., electrocardiogram), vital signs, by measuring test parameters and recording adverse events. For example, in the uses and methods according to the present invention, adverse events in primates other than chimpanzees can be tested by histopathological methods and / or histochemical methods.

[0282] The above terms are also referred to, for example, in Preclinical safety evaluation of biotechnology-derived pharmaceuticals S6;ICH Harmonised Tripartite Guideline;ICH Steering Committee meeting on July 16,1997.

[0283] In a further embodiment, the present invention provides a kit comprising an antibody construct of the present invention, an antibody construct produced according to the method of the present invention, a polypeptide of the present invention, a vector of the present invention, and / or a host cell of the present invention.

[0284] In the context of the present invention, the term "kit" means that two or more components are packaged together in a container, vessel, or otherwise, and one of the components corresponds to an antibody construct, pharmaceutical composition, vector, or host cell of the present invention. Thus, a kit can be described as a set of products and / or instruments sufficient to achieve a particular purpose that can be sold as a single item.

[0285] The kit may include one or more containers (e.g., vials, ampoules, containers, syringes, bottles, bags) of any suitable shape, size, and material (preferably waterproof, e.g., plastic or glass) containing a dosage (see above) of the antibody construct or pharmaceutical composition of the present invention suitable for administration. The kit may further include instructions for use (e.g., in the form of a leaflet or instruction manual), means for administering the antibody construct of the present invention, such as a syringe, pump, infuser, etc., means for reconstituting the antibody construct of the present invention, and / or means for diluting the antibody construct of the present invention.

[0286] The present invention also provides a kit for single-dose administration units. The kit of the present invention may also include a first container containing a dry / freeze-dried antibody construct and a second container containing an aqueous formulation. In certain embodiments of the present invention, kits are provided that include single-chamber and multi-chamber pre-filled syringes (e.g., liquid syringes and lyosyringes).

Examples

[0287] The present invention is illustrated by the following examples. Each example should not be considered as limiting the scope of the present invention. The present invention is limited only by the claims.

[0288] Example 1 Preparation of CHO Cells Expressing Wild-Type and Truncated DLL3 The extracellular domain of the DLL3 antigen can be subdivided into individual subdomains or regions defined by the following amino acid positions for Examples 1 and 2.

[0289] Signal peptide: 1 - 26 N-terminus: 27 - 175 DSL: 176 - 215 EGF-1: 216 - 249 EGF-2: 274 - 310 EGF-3: 312 - 351 EGF-4: 353 - 389 EGF-5: 391 - 427 EGF-6: 429 - 465

[0290] To construct truncated DLL3 molecules for epitope mapping, the corresponding sequences of eight extracellular domains of human DLL3 (signal peptide + N-terminus, DSL, EGF1, EGF2, EGF3, EGF4, EGF5, and EGF6) were deleted stepwise from the antigen starting from the N-terminus. The following molecules were prepared. See also Figure 1. · Hu DLL3 ECD / whole ECD SEQ ID NO: 263 · Hu DLL3 ECD / up to DSL SEQ ID NO: 264 · Hu DLL3 ECD / up to EGF-1 SEQ ID NO: 265 · Hu DLL3 ECD / up to EGF-2 SEQ ID NO: 266 · Hu DLL3 ECD / up to EGF-3 SEQ ID NO: 267 · Hu DLL3 ECD / up to EGF-4 SEQ ID NO: 268 · Hu DLL3 ECD / up to EGF-5 SEQ ID NO: 269 · Hu DLL3 ECD / EGF-6 only SEQ ID NO: 270

[0291] To generate CHO cells expressing human DLL3, cynomolgus monkey ("cyno") DLL3, and truncated human N-terminal V5-tagged DLL3, human DLL3-ECD (see also SEQ ID NO: 253, GeneBank accession number NM_016941), cynomolgus monkey DLL3-ECD (SEQ ID NO: 272), and seven truncated human N-terminal V5-tagged DLL3-ECD forms (see above) were cloned into a plasmid named pEF-DHFR (for pEF-DHFR, see Raum et al. Cancer Immunol Immunother 50 (2001) 141-150). For cell surface expression of human and cynomolgus monkey DLL3, the native signal peptide was used, and for cell surface expression of truncated human N-terminal DLL3, the mouse IgG heavy chain signal peptide was used, with a V5 tag placed behind it. An in-frame artificial Ser / Gly linker and the coding sequence of a domain derived from the transmembrane / intracellular domain of human EpCAM (amino acids 266-314 of the sequence published in GenBank accession number NM_002354) are located behind all the DLL3-ECD sequences described. All cloning procedures were performed according to standard protocols (Sambrook, Molecular Cloning; A Laboratory Manual, 3rd edition, Cold Spring Harbour Laboratory Press, Cold Spring Harbour, New York (2001)). As described by Kaufman R.J. (1990) Methods Enzymol. 185, 537-566, the corresponding plasmids were transfected into DHFR-deficient CHO cells for eukaryotic expression to obtain each construct.

[0292] Expression of human and cynomolgus monkey DLL3 in CHO cells was confirmed by FACS assay using monoclonal mouse IgG2b anti-human DLL3 antibody. Expression of seven truncated human DLL3-ECDs was confirmed using monoclonal mouse IgG2a anti-v5 tag antibody. Bound monoclonal antibodies were detected using anti-mouse IgG Fc gamma-PE. As a negative control, cells were incubated with an isotype control antibody instead of the primary antibody. Each sample was measured by flow cytometry.

[0293] Example 2 Epitope mapping of anti-DLL3 antibody constructs Cells transfected with human DLL3 and truncated human DLL3 molecules (see Example 1) were stained in PBS / 1.5% FCS with a crude, undiluted periplasmic extract containing a bispecific DLL3xCD3 antibody construct (bearing a CD3 binding domain designated I2C) fused to human albumin (variant 1). Bound molecules were detected with an in-house produced mouse monoclonal anti-CD3 binding domain antibody (50 μl), followed by anti-mouse IgG Fc gamma-PE (1:100, 50 μl; Jackson Immunoresearch #115-116-071). All antibodies were diluted in PBS / 1.5% FCS. As a negative control, cells were incubated with PBS / 2% FCS instead of the periplasmic extract. Each sample was measured by flow cytometry.

[0294] The regions recognized by each DLL3 binding domain are shown in the sequence listing (Table 18). The binding substances were specifically mapped to epitopes located within the N-terminus of DLL3, within the DSL domain, and within individual EGF domains.

[0295] Figure 2 shows two exemplary binding substances that bind to the DLL3 epitope contained within the EGF-3 region (no FACS signal in the corresponding truncated DLL3 construct that does not contain EGF-3). Figure 2 also shows an exemplary binding substance that binds to the DLL3 epitope contained within the EGF-3 region (no FACS signal in the corresponding truncated DLL3 construct that does not contain EGF-4).

[0296] Some of the binding substances have been shown to be specific for epitopes located in a region designated EGF-5 / [6]. The square brackets mean that in the truncated DLL3 construct in which only the EGF-6 domain remains (the last construct in Figure 1), there is a decrease in the FACS signal of the binding substance (i.e., it is neither completely present nor completely lost).

[0297] The bispecific DLL3xCD3 constructs used in the following examples are selected from constructs having "I2C" as the CD3 binding domain and constructs having a C-terminal fusion to wild-type human serum albumin (e.g., SEQ ID NOs: 224 - 230, 233 - 235, 238 - 241).

[0298] Example 3 Measurement of Antibody Affinity for Human and Cynomolgus DLL3 by Biacore Biacore analysis experiments were performed using recombinant human / cynomolgus DLL3-ECD fusion proteins with chicken albumin to measure the target binding of the antibody constructs of the present invention.

[0299] Specifically, according to the manufacturer's instructions, each recombinant antigen was immobilized on a CM5 sensor chip (GE Healthcare) at approximately 600 - 800 RU using acetate buffer pH 4.5. Samples of the DLL3xCD3 bispecific antibody construct were loaded into dilution series diluted to concentrations of 50 nM, 25 nM, 12.5 nM, 6.25 nM, and 3.13 nM with HBS-EP running buffer (GE Healthcare). The flow rate was set to 30 μl / min for 3 minutes, and then HBS-EP running buffer was used again at a flow rate of 30 μl / ml for 8 - 20 minutes. Chip regeneration was performed using a solution of 10 mM glycine and 10 mM NaCl (pH 1.5). The dataset was analyzed using BiaEval software. Two independent experiments were performed throughout.

[0300] The DLL3xCD3 bispecific antibody construct according to the present invention showed extremely high affinity below nanomolar for human DLL3 (with the exception that the KD value of DLL3-13 was very low and in the single-digit nanomolar range). When the binding to cynomolgus DLL3 was used as a control, it also showed affinity in a similar range. The affinity values and the calculated differences in affinity are shown in Table 2.

[0301] (Table 2) Affinity of the DLL3xCD3 bispecific antibody construct for human and cynomolgus DLL3 determined by Biacore analysis and the calculated interspecies affinity differences TIFF2025106535000004.tif75161

[0302] Furthermore, when the binding of the bispecific antibody construct to both human CD3 and cynomolgus CD3 was confirmed by Biacore assay, it was low in the double-digit nanomolar range (data not shown).

[0303] Example 4 Scatchard analysis of the affinity of the DLL3xCD3 bispecific antibody construct for human and cynomolgus DLL3 in target antigen-positive cells, and measurement of the interspecies affinity difference Also, the affinity of the DLL3xCD3 bispecific antibody construct for CHO cells transfected with human or cynomolgus DLL3 was measured by Scatchard analysis, which is the most reliable method for measuring the possible difference in affinity between human and cynomolgus DLL3. To prepare for Scatchard analysis, a monovalent detection system was used to perform a saturation binding experiment to accurately measure the monovalent binding of the DLL3xCD3 bispecific antibody construct to each cell line.

[0304] 2x10 of each cell line (recombinant human DLL3-expressing CHO cell line, recombinant cynomolgus DLL3-expressing CHO cell line) 4 cells were incubated with 50 μl of a three-fold dilution series (12-fold dilution at 1:2) of the corresponding DLL3xCD3 bispecific antibody construct (starting concentration 10-20 nM) each (until saturation was reached), followed by incubation at 4°C for 16 hours with stirring, and one residue washing step. Next, the cells were incubated with 30 μl of CD3xALEXA488 conjugate solution for an additional 1 hour. After one washing step, the cells were resuspended in 150 μl of FACS buffer containing 3.5% formaldehyde, incubated for an additional 15 minutes, centrifuged, resuspended in FACS buffer, and analyzed using a FACS CantoII instrument and FACS Diva software. The data were generated from two independent experiments with triplicates each. Each Scatchard analysis was calculated and the maximum binding capacity (Bmax) was estimated. The concentration of the DLL3xCD3 bispecific antibody construct at half-maximum binding was determined based on each KD. The triplicate measurements were plotted to obtain a hyperbolic curve and a sigmoidal curve showing the appropriate concentration range from minimum to maximum binding.

[0305] The values shown in Table 3 were obtained from two independent experiments for each DLL3xCD3 bispecific antibody construct. By cell-based Scatchard analysis, it was confirmed that the DLL3xCD3 bispecific antibody construct of the present invention has sub-nanomolar affinity for human DLL3 and cynomolgus DLL3, and there is a slight difference in affinity of approximately 1 between the cynomolgus / human species.

[0306] (Table 3) Affinity (KD) of the DLL3xCD3 bispecific antibody construct determined by cell-based scratchard analysis and the difference in affinity calculated by KD of cynomolgus DLL3 / KD of human DLL3. The antibody constructs were measured in two independent experiments in triplicate. TIFF2025106535000005.tif81156

[0307] Example 5 Bispecific binding and cross-species reactivity To confirm binding to human DLL3 and CD3, as well as cynomolgus DLL3 and CD3, the bispecific antibody constructs of the present invention were tested by flow cytometry using the following. · CHO cells transfected with human DLL3, artificial human DLL3 isoform (characterized by point mutations F172C and L218P), and cynomolgus DLL3, respectively, · Human DLL3-positive human lung cancer cell line SHP-77, · Human T cell leukemia cell line HPB-all (DSMZ, Braunschweig, ACC483) expressing CD3, and · T cell line LnPx 4119 expressing cynomolgus CD3

[0308] For flow cytometry, 200,000 cells of each cell line were incubated with 50 μl of purified bispecific antibody construct at a concentration of 5 μg / ml for 60 minutes at 4°C. The cells were washed twice with PBS / 2% FCS and then incubated with an in-house mouse antibody (2 μg / ml) specific for the CD3-binding portion of the bispecific antibody construct for 30 minutes at 4°C. After washing, the bound mouse antibody was detected with goat anti-mouse Fcγ-PE (1:100) for 30 minutes at 4°C. Each sample was measured by flow cytometry. Non-transfected CHO cells were used as a negative control.

[0309] The results are shown in Figure 3. The DLL3xCD3 bispecific antibody construct of the present invention stained CHO cells transfected with human DLL3, artificial DLL3 isoform, and cynomolgus DLL3, and also stained the human DLL3-positive human lung cancer cell line SHP-77 (native form). Human and cynomolgus T cell lines expressing CD3 were also recognized by the bispecific antibody construct. Also, staining of negative control cells (untransfected CHO, data shown in Example 6) was not observed.

[0310] Example 6 Confirmation of non-binding to human paralogs Human DLL3 paralogs DLL1 and DLL4 were stably transfected into CHO cells. The sequences of the paralogs used in this example are shown in the sequence listing (SEQ ID NOs: 283 and 284). Protein expression was confirmed by FACS analysis using antibodies specific for each paralog. The antibodies were anti-human DLL1 MAB1818 (R&D; 5 μg / ml) and anti-human DLL4 MAB1506 (R&D, 5 μg / ml) against DLL4.

[0311] The flow cytometry assay was performed as described in Example 5 except that the bound mouse antibody was detected using goat anti-mouse FITC (1:100). The results are shown in Figure 4. From this analysis, it was confirmed that none of the DLL3xCD3 bispecific antibody constructs of the present invention cross-react with the human DLL3 paralogs DLL1 and DLL4.

[0312] Example 7 Identity with human germline To analyze the identity / similarity of the sequences of the antibody constructs to human antibody germline genes, the DLL3-binding domains of the present invention were aligned as follows. The complete VL containing all CDRs and the complete VH containing CDR1 and 2 but excluding CDR3 were aligned against human antibody germline genes (Vbase). Details can be referred to in the specification of the present application. The results are shown in Table 4 below.

[0313] (Table 4) Identity with human germline sequences of VH and VL TIFF2025106535000006.tif81128

[0314] Example 8 Cytotoxic activity The ability of the DLL3xCD3 bispecific antibody construct of the present invention to redirect effector T cells to DLL3-expressing target cells was analyzed in the following five types of in vitro cytotoxicity assays. · The ability of the DLL3xCD3 bispecific antibody construct to redirect stimulated human CD8+ effector T cells to human DLL3-transfected CHO cells was measured by a 18-hour 51-chromium release assay. · The ability of the DLL3xCD3 bispecific antibody construct to redirect stimulated human CD8+ effector T cells to the DLL3-positive human lung cancer cell line SHP-77 was measured by a 18-hour 51-chromium release assay. · The ability of the DLL3xCD3 bispecific antibody construct to redirect T cells in unstimulated human PBMCs to human DLL3-transfected CHO cells was measured by a 48-hour FACS-based cytotoxicity assay. · The ability of the DLL3xCD3 bispecific antibody construct to redirect T cells in unstimulated human PBMCs to the DLL3-positive human cell line SHP-77 was measured by a 48-hour FACS-based cytotoxicity assay. · To confirm that the cross-reactive DLL3xCD3 bispecific antibody construct can redirect cynomolgus T cells to cynomolgus DLL3-transfected CHO cells, a 48-hour FACS-based cytotoxicity assay was performed using a cynomolgus T cell line as effector T cells.

[0315] Example 8.1 Chromium release assay using unstimulated human T cells CD8 +Stimulated T cells enriched with T cells were obtained as described below. A Petri dish (diameter 145 mm, Greiner bio-one GmbH, Kremsmunster) was coated with a commercially available anti-CD3 specific antibody (OKT3, Orthoclone) at a final concentration of 1 μg / ml at 37 °C for 1 hour. Unbound proteins were removed by a single washing step with PBS. 3 - 5 x 10 7 human PBMCs were added to a pre-coated Petri dish containing 120 ml of RPMI 1640 supplemented with stabilized glutamine / 10% FCS / IL-2 20 U / ml (Proleukin®, Chiron) and stimulated for 2 days. On day 3, the cells were harvested and washed once with RPMI 1640. IL-2 was added to a final concentration of 20 U / ml and the cells were cultured again for 1 day in the same cell culture medium as above. According to the manufacturer's protocol, Dynal beads were used to remove CD4 + T cells and CD56 + NK cells, and CD8 + cytotoxic T lymphocytes (CTLs) were enriched.

[0316] Cynomolgus monkey DLL3 or human DLL3 transfected CHO target cells were washed twice with PBS and resuspended in 100 μl of RPMI with 50% FCS to a final volume containing 11.1 MBq of 51Labeling was performed at 37 °C for 60 minutes using Cr. Subsequently, the labeled target cells were washed three times with 5 ml of RPMI and then used for the cytotoxicity assay. This assay was performed in 96-well plates supplemented with RPMI at a total volume of 200 μl at an E:T ratio of 10:1. Purified bispecific antibody constructs at starting concentrations of 0.01 - 1 μg / ml and three-fold dilutions thereof were used. The incubation time during the assay was 18 hours. Cytotoxicity was determined from the relative value of chromium released into the supernatant relative to the difference between maximum lysis (addition of Triton-X) and spontaneous lysis (without effector cells). All measurements were performed in quadruplicate. Measurement of chromium activity in the supernatant was performed using a Wizard 3’’ gamma counter (Perkin Elmer Life Sciences GmbH, Koln, Germany). Analysis of the results was performed using Prism 5 for Windows (version 5.0, GraphPad Software Inc., San Diego, California, USA). The EC50 value calculated by the analysis program from the sigmoid dose-response curve was used for comparison of cytotoxic activities.

[0317] Example 8.2 Ability to redirect stimulated human effector T cells to human DLL3-transfected CHO cells The cytotoxic activity of the DLL3xCD3 bispecific antibody construct according to the present invention was analyzed by a 51 chromium ( 51 Cr) release cytotoxicity assay using CHO cells transfected with human DLL3 as target cells and stimulated human CD8+ T cells as effector cells. This experiment was performed as described in Example 8.1.

[0318] The results are shown in Table 5. The DLL3xCD3 bispecific antibody construct showed extremely potent cytotoxic activity in the single-digit picomolar range against CHO cells transfected with human DLL3.

[0319] (Table 5) Using CHO cells transfected with human DLL3 as target cells and stimulated human CD8 T cells as effector cells, the EC50 value [pM] of the DLL3xCD3 bispecific antibody construct was analyzed by a 51-chromium ( 51 Cr) release cytotoxicity assay TIFF2025106535000007.tif78128

[0320] Example 8.3 Ability to redirect stimulated human effector T cells to DLL3-positive human lung cancer cell line SHP-77 Using the DLL3-positive human lung cancer cell line SHP-77 as a target cell source and stimulated human CD8+ T cells as effector cells, the cytotoxic activity of the DLL3xCD3 bispecific antibody construct was analyzed by a 51-chromium ( 51 Cr) release cytotoxicity assay. This assay was performed as described in Example 8.1.

[0321] According to the results of the 51-chromium release assay using stimulated enriched human CD8+ T lymphocytes as effector cells and human DLL3-transfected CHO cells as target cells, the DLL3xCD3 bispecific antibody construct of the present invention also has strong cytotoxic activity against target cells of the natural expression form (see Table 6).

[0322] (Table 6) Using the DLL3-positive human lung cancer cell line SHP-77 as a target cell source and stimulated enriched human CD8 T cells as effector cells, the EC50 value [pM] of the DLL3xCD3 bispecific antibody construct was analyzed by an 18-hour 51-chromium ( 51 Cr) release cytotoxicity assay Rows 1-10: Antibody constructs according to the present invention that bind to DLL3 epitopes contained within the region shown in SEQ ID NO: 260. (Rows 1-7: Antibody constructs that bind to DLL3 epitopes contained within the EGF-3 region. Rows 8-10: Antibody constructs that bind to DLL3 epitopes contained within the EGF-4 region.) Rows 11-14: Antibody constructs that bind to DLL3 epitopes contained within the EGF-5 / [EGF-6] region. TIFF2025106535000008.tif101128

[0323] Example 8.4 Cytotoxicity assay by FACS using unstimulated human PBMC Isolation of effector cells Human peripheral blood mononuclear cells (PBMC) were prepared by Ficoll density gradient centrifugation from concentrated lymphocyte preparations (buffy coats), by-products of blood banks that collect blood for transfusion. Buffy coats were supplied by the local blood bank, and PBMC were prepared on the same day as blood collection. After Ficoll density centrifugation and thorough washing with Dulbecco's PBS (Gibco), residual red blood cells were removed from PBMC by incubation with red blood cell lysis buffer (155 mM NH4Cl, 10 mM KHCO3, 100 μM EDTA). PBMC were centrifuged at 100 x g, and platelets were removed from the supernatant. The remaining lymphocytes mainly include B lymphocytes, T lymphocytes, NK cells, and monocytes. PBMC were maintained at 37 °C / 5% CO2 in RPMI medium (Gibco) containing 10% FCS (Gibco) during culture.

[0324] CD14 + cells and CD56 + Removal of cells CD14 + For the removal of CD14 cells, human CD14 microbeads (Milteny Biotec, MACS, #130 - 050 - 201) were used, and for the removal of NK cells, human CD56 microbeads (MACS, #130 - 050 - 401) were used. PBMC were counted and centrifuged at 300 x g for 10 minutes at room temperature. The supernatant was discarded, and the cell pellet was resuspended in MACS separation buffer [80 μl / 10 7 cells; PBS (Invitrogen, #20012 - 043), 0.5% (v / v) FBS (Gibco, #10270 - 106), 2 mM EDTA (Sigma - Aldrich, #E - 6511)]. CD14 microbeads and CD56 microbeads (20 μl / 10 7Cells were added and incubated at 4 - 8 °C for 15 minutes. The cells were washed with MACS separation buffer (1 - 2 mL / 10 7 cells). After centrifugation (see above), the supernatant was discarded and the cells were resuspended in MACS separation buffer (500 μl / 10 8 cells). Next, CD14 / CD56 negative cells were isolated using an LS column (Miltenyi Biotec, #130 - 042 - 401). PBMCs without CD14+ / CD56+ cells were cultured at 37 °C in an incubator in RPMI complete medium, namely RPMI 1640 (Biochrom AG, #FG1215) supplemented with 10% FBS (Biochrom AG, #S0115), 1x non - essential amino acids (Biochrom AG, #K0293), 10 mM Hepes buffer (Biochrom AG, #L1613), 1 mM sodium pyruvate (Biochrom AG, #L0473), and 100 U / mL penicillin / streptomycin (Biochrom AG, #A2213) until needed.

[0325] Labeling of target cells To analyze cell lysis by flow cytometry assay, human DLL3 or cynomolgus DLL3 - transfected CHO cells were labeled as target cells and distinguished from effector cells using the fluorescent membrane dye DiOC 18 (DiO) (Molecular Probes, #V22886). Briefly, cells were harvested, washed once with PBS, and adjusted to 10 6 cells / mL in PBS containing 2% (v / v) FBS and the membrane dye DiO (5 μl / 10 6 cells). After incubation at 37 °C for 3 minutes, the cells were washed twice with complete RPMI medium and adjusted to 1.25x10 5 cells / mL. The viability of the cells was determined using 0.5% (v / v) isotonic eosin G solution (Roth, #45380).

[0326] Analysis by flow cytometry This assay aimed to quantify the lysis of cynomolgus monkey or human DLL3-transfected CHO cells in the presence of serial dilutions of a DLL3 bispecific antibody construct. DiO-labeled target cells and effector cells (i.e., PBMCs + without CD14 cells) were mixed in equal volumes to a 10:1 E:T cell ratio. 160 μl of this suspension was transferred to each well of a 96-well plate. Serial dilutions of a DLL3xCD3 bispecific antibody construct, and a negative control bispecific (a CD3-based bispecific antibody construct recognizing an irrelevant target antigen) or RPMI complete medium were added as additional negative controls. The bispecific antibody-mediated cytotoxic reaction was carried out for 48 hours in a 7% CO2 humidified incubator. The 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 dye that is normally excluded from live cells but is captured by dead cells and becomes distinguishable by fluorescence emission.

[0327] Samples were measured by flow cytometry on a FACSCanto II instrument and analyzed with FACSDiva software (both from Becton Dickinson). DiO-positive cells were determined to be target cells. PI-negative target cells were classified as live target cells. The percentage of cytotoxicity was calculated according to the following formula. TIFF2025106535000009.tif18128

[0328] The percentage of cytotoxicity was plotted against the concentration of the corresponding bispecific antibody construct using GraphPad Prism 5 software (Graph Pad Software, San Diego). The dose-response curve was analyzed using a four-parameter logistic regression model to derive a sigmoid dose-response curve with a fixed Hill slope and calculate the EC50 value.

[0329] Example 8.5 Ability of unstimulated human PBMCs to redirect to human DLL3-transfected CHO cells Using CHO cells transfected with human DLL3 as target cells and unstimulated human PBMCs as effector cells, the cytotoxic activity of the DLL3xCD3 bispecific antibody construct was analyzed by a cytotoxicity assay using FACS. This assay was performed as described in Example 8.4 above.

[0330] Table 7 shows the results of a cytotoxicity assay using FACS with unstimulated human PBMCs as effector cells and human DLL3-transfected CHO cells as targets.

[0331] (Table 7) EC50 values [pM] of the DLL3xCD3 bispecific antibody construct measured in a 48-hour cytotoxicity assay using FACS with unstimulated human PBMCs as effector cells and CHO cells transfected with human DLL3 as targets Rows 1-10: Antibody constructs according to the invention that bind to the DLL3 epitope contained within the region shown in SEQ ID NO: 260. (Rows 1-7: Antibody constructs that bind to the DLL3 epitope contained within the EGF-3 region. Rows 8-10: Antibody constructs that bind to the DLL3 epitope contained within the EGF-4 region.) Rows 11-14: Antibody constructs that bind to the DLL3 epitope contained within the EGF-5 / [EGF-6] region. TIFF2025106535000010.tif101128

[0332] As predicted, the EC50 values were generally higher in the cytotoxicity assay using unstimulated PBMCs as effector cells compared to the cytotoxicity assay using stimulated human CD8+ T cells (see Example 8.2).

[0333] Example 8.6 The ability of unstimulated human PBMCs to redirect to DLL3-positive lung cancer cell line SHP-77 Using the DLL3-positive human lung cancer cell line SHP-77 as the target cell source and unstimulated human PBMC as the effector cells, the cytotoxic activity of the DLL3xCD3 bispecific antibody construct was further analyzed by a cytotoxicity assay using FACS. This assay was performed as described in Example 8.4 above. The results are shown in Table 8.

[0334] (Table 8) EC50 values [pM] of the DLL3xCD3 bispecific antibody construct measured by a 48-hour cytotoxicity assay using FACS, with unstimulated human PBMC as the effector cells and the human cell line SHP-77 as the target cell source Rows 1-10: Antibody constructs according to the invention that bind to the DLL3 epitope contained within the region shown in SEQ ID NO: 260. (Rows 1-7: Antibody constructs that bind to the DLL3 epitope contained within the EGF-3 region. Rows 8-10: Antibody constructs that bind to the DLL3 epitope contained within the EGF-4 region.) Rows 11-14: Antibody constructs that bind to the DLL3 epitope contained within the EGF-5 / [EGF-6] region. TIFF2025106535000011.tif101128

[0335] Example 8.7 Ability to redirect cynomolgus T cells to CHO cells expressing cynomolgus DLL3 Finally, using CHO cells transfected with cynomolgus (cynomolgus macaque) DLL3 as the target cells and the cynomolgus T cell line 4119LnPx (Knappe et al. Blood 95: 3256-61 (2000)) as the effector cell source, the cytotoxic activity of the DLL3xCD3 bispecific antibody construct was analyzed by a cytotoxicity assay using FACS. Target cell labeling of cynomolgus DLL3-transfected CHO cells and analysis of cytotoxic activity by flow cytometry were performed as described above.

[0336] The results are shown in Table 9. The DLL3xCD3 bispecific antibody construct according to the present invention induced cynomolgus T cells derived from cell line 4119LnPx and efficiently killed cynomolgus DLL3-transfected CHO cells. In this assay, the antibody construct showed a potent EC50 value in the low picomolar range and was confirmed to be highly active in the cynomolgus system.

[0337] (Table 9) EC50 value [pM] of the DLL3xCD3 bispecific antibody construct measured by a 48-hour FACS cytotoxicity assay using cynomolgus T cell line 4119LnPx as effector cells and CHO cells transfected with cynomolgus DLL3 as target cells Rows 1 - 10: Antibody constructs according to the present invention that bind to the DLL3 epitope contained within the region shown in SEQ ID NO: 260. (Rows 1 - 7: Antibody constructs that bind to the DLL3 epitope contained within the EGF-3 region. Rows 8 - 10: Antibody constructs that bind to the DLL3 epitope contained within the EGF-4 region.) Rows 11 - 14: Antibody constructs that bind to the DLL3 epitope contained within the EGF-5 / [EGF-6] region. TIFF2025106535000012.tif101128

[0338] Example 9 Conversion from monomer to dimer after (i) 3 freeze / thaw cycles and (ii) 7-day incubation at 250 μg / ml After subjecting the monomeric construct of the bispecific DLL3xCD3 antibody to various stress conditions, it was subjected to high-speed SEC, and the percentage of the initial monomeric antibody construct converted to the dimeric antibody construct was measured.

[0339] (i) 25 μg of the monomeric antibody construct was adjusted to a concentration of 250 μg / ml in a general formulation buffer, then frozen at -80°C for 30 minutes and thawed at room temperature for 30 minutes. After 3 freeze / thaw cycles, the dimer content was measured by HP-SEC.

[0340] (ii) The monomeric antibody construct of 25 μg was adjusted to a concentration of 250 μg / ml with a general formulation buffer and then incubated at 37 °C for 7 days. The dimer content was measured by HP-SEC.

[0341] 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 composition of the column equilibration buffer and the running buffer was 100 mM KH2PO4 - 200 mM Na2SO4, adjusted to pH 6.6. The antibody solution (25 μg of protein) was applied to the equilibrated column, and elution was performed at a flow rate of 0.75 ml / min and a maximum pressure of 7 MPa. The entire run was monitored at absorbances of 280, 254, and 210 nm. The analysis was performed by peak integration of the 210 nm signal recorded on the run evaluation sheet of the Akta Unicorn software. The dimer content was calculated by dividing the peak area of the dimer by the total peak area of the monomer and dimer combined.

[0342] The results are shown in Table 10 below. The DLL3xCD3 bispecific antibody construct of the present invention showed a dimer ratio of 0.0% after 3 freeze / thaw cycles and a dimer ratio of ≤2% after incubation at 37 °C for 7 days.

[0343] (Table 10) Monomer - to - dimer percentage of the DLL3xCD3 bispecific antibody construct measured by high - performance size - exclusion chromatography (HP - SEC) TIFF2025106535000013.tif90161

[0344] Example 10 Thermal stability The aggregation temperature of the antibody was measured as follows. 40 μl of a 250 μg / ml antibody construct solution was transferred to a single-use cuvette and placed in a Wyatt dynamic light scattering device, DynaPro Nanostar (Wyatt). The sample was heated from 40 °C to 70 °C at a heating rate of 0.5 °C / min, and the measured radius was continuously acquired. The aggregation temperature of the antibody construct was calculated using the software package attached to the DLS device to utilize the increase in radius indicating protein melting and aggregation.

[0345] As shown in Table 11 below, all of the DLL3xCD3 bispecific antibody constructs of the present invention tested showed thermal stability with an aggregation temperature ≧ 45 °C. The group of antibody constructs that bind to the epitope of DLL3 contained within the EGF-4 region (shown in SEQ ID NO: 259) further had thermal stability of ≧ 50 °C, precisely ≧ 54 °C.

[0346] (Table 11) Thermal stability of bispecific antibody constructs measured by DLS (dynamic light scattering method) TIFF2025106535000014.tif95128

[0347] Example 11 Stability after 24-hour incubation in human plasma The purified bispecific antibody construct was incubated in a human plasma pool at a ratio of 1:5 at 37 °C at a final concentration of 2 - 20 μg / ml for 96 hours. After incubation with plasma, the antibody constructs were compared in a 51 chromium release assay using stimulated enriched human CD8+ T cells and human DLL3-transfected CHO cells at an effector cell to target cell (E:T) ratio of 10:1 at starting concentrations of 0.01 - 0.1 μg / ml (the assay described in Example 8.1). Freshly thawed bispecific antibody constructs that had not been incubated were included as controls.

[0348] The results are shown in Table 12 below. Also, exemplary results for two antibody constructs, DLL-4 and DLL-14, are shown in Figure 5. All of the antibody constructs tested had extremely good plasma stability (EC 5O Plasma / EC 50had a control). The group of antibody constructs that bind to the epitope of DLL3 contained within the EGF-4 region (shown in SEQ ID NO: 259) further had plasma stability of ≤1.5, precisely ≤1.1.

[0349] (Table 12) EC50 values of antibody constructs with and without incubation with plasma, and calculated plasma / control values TIFF2025106535000015.tif89158

[0350] Example 12 Turbidity at an antibody concentration of 2500 μg / ml 1 ml of a purified antibody construct solution with a concentration of 250 μg / ml was concentrated to 2500 μg / ml using a rotary concentration unit. After storage at 5°C for 16 hours, the turbidity of the antibody solution was determined by measuring the light absorption at OD340 nm against a general formulation buffer.

[0351] The results are shown in Table 13 below. All of the tested antibody constructs had extremely good turbidity of ≤0.1, except that the turbidity of one construct slightly exceeded 0.1. The group of antibody constructs that bind to the epitope of DLL3 contained within the EGF-4 region (shown in SEQ ID NO: 259) further had a turbidity of ≤0.08.

[0352] (Table 13) Turbidity of antibody constructs after concentration to 2.5 mg / ml overnight TIFF2025106535000016.tif95128

[0353] Example 13 Protein homogeneity by high-resolution cation exchange chromatography The protein homogeneity of the antibody constructs of the present invention was analyzed by high-resolution cation exchange chromatography CIEX.

[0354] 50 μg of the antibody construct monomer was diluted with 50 ml of binding buffer A (20 mM sodium dihydrogen phosphate, 30 mM NaCl, 0.01% sodium octanoate, pH 5.5), and 40 ml of this solution was applied to a 1 ml BioPro SP-F column (YMC, Germany) connected to an Akta Micro FPLC device (GE Healthcare, Germany). After the sample was bound, a washing step was further performed with the binding buffer. To elute the protein, a salt gradient that linearly increased to 50% of buffer B (20 mM sodium dihydrogen phosphate, 1000 mM NaCl, 0.01% sodium octanoate, pH 5.5) was applied over 10 column volumes using buffer B. The entire run was monitored at light absorbances of 280, 254, and 210 nm. The analysis was performed by peak integration of the 280 nm signal recorded on the run evaluation sheet of the Akta Unicorn software.

[0355] The results are shown in Table 14 below. All of the antibody constructs tested had an extremely good homogeneity of ≥ 80% (area under the curve of the main peak (= AUC)). The group of antibody constructs that bind to the epitope of DLL3 contained within the EGF-3 region (shown in SEQ ID NO: 258) further had a homogeneity of ≥ 90%.

[0356] (Table 14) Protein homogeneity of the antibody construct (AUC% of the main peak) TIFF2025106535000017.tif95128

[0357] Example 14 Surface hydrophobicity measured by HIC butyl The surface hydrophobicity of the bispecific antibody construct of the present invention was tested by hydrophobic interaction chromatography HIC in flow-through mode.

[0358] The 50 μg antibody construct monomer was diluted to a final volume of 500 μl with a general formulation buffer (10 mM citric acid, 75 mM lysine HCl, 4% trehalose, pH 7.0) and applied to a 1 ml butyl sepharose FF column (GE Healthcare, Germany) connected to an Akta Purifier FPLC device (GE Healthcare, Germany). The entire run was monitored at light absorbances of 280, 254, and 210 nm. Analysis was performed by peak integration of the 280 nm signal recorded on the run evaluation sheet of the Akta Unicorn software. The elution behavior was evaluated by comparing the areas and velocities of the protein signals up and down, thereby indicating the interaction strength between the BiTE albumin and the matrix.

[0359] The antibody construct had good elution behavior, which was almost rapid and complete.

[0360] Example 15 Potency difference between monomeric and dimeric isoforms of bispecific antibody constructs To measure the difference in cytotoxic activity (referred to as potency difference) between the monomeric and dimeric isoforms of individual DLL3xCD3 bispecific antibody constructs, as described above herein (Example 8.1), a 51 chromium release cytotoxicity assay was performed for 18 hours using the purified monomeric and dimeric bispecific antibody constructs. The effector cells were stimulated enriched human CD8+ T cells. The target cells were human DLL3 transfected CHO cells. The effector cell to target cell (E:T) ratio was 10:1. The potency difference was calculated as the ratio of the EC50 values.

[0361] The results are shown in Table 15 below. Also, exemplary results of two antibody constructs, DLL-4 and DLL-14, are shown in Figure 6. The potency differences of the DLL3xCD3 bispecific antibody constructs tested were between 0.2 and 1.0. Therefore, there is substantially no dimer with higher activity compared to the corresponding monomer.

[0362] (Table 15) Potency difference between monomeric isoform and dimeric isoform TIFF2025106535000018.tif95166

[0363] Example 16 in vitro internalization assay The change in the potency of the DLL3xCD3 bispecific antibody construct was measured as a function of pre-incubation of the construct in target cells lacking T cells. If the antibody construct is internalized, it should undergo lysosomal degradation. The effective concentration should decrease over time, and thus the apparent potency should also decrease. Since this effect is a known phenomenon, it is observed with other targets, but no effect was observed with the DLL3xCD3 bispecific antibody construct. This assay was performed as follows.

[0364] T cells were counted and diluted to a concentration of 1x10 5 / ml in assay medium. SHP-77 cells were counted and seeded at 2500 cells / well (cpw). Starting at a concentration of 100 nM, the antibody construct was serially diluted 1:2 (using Bravo). The antibody construct was added to the culture assay plates, and T cells were added after 0, 1, or 2 hours of incubation. The T cells were then seeded at 25000 cpw, and the assay was incubated at 37 °C for 48 hours. The cell viability of SHP-77 was analyzed using the Steady-Glo® system (25 μl / well). The results are shown in Figure 7, which suggests that there is no significant internalization of the antibody construct DLL3-4xCD3 (I2C).

[0365] Example 17 Efflux assay To analyze whether the cytotoxic activity of the DLL3xCD3 bispecific antibody construct of the present invention is significantly impaired by the presence of the effluxed DLL, the following assay was performed. T cells were counted and diluted to a concentration of 1x10 5 / ml in assay medium. SHP-77 cells were counted and diluted to a concentration of 1.25x10 5Diluted to / ml and increased the concentration of soluble DLL3 to 0.3 nM to 12 nM. SHP-77 cells were seeded at 2500 cells / well (cpw), and T cells were added at 25000 cpw. The antibody construct was serially diluted 1:2 (with Bravo) and added to the culture assay (using Bravo). Incubation was carried out at 37 °C for 48 hours. The cell viability of SHP-77 was analyzed by the Steady-Glo® system (25 μl / well).

[0366] Example 18 Mouse xenograft efficacy test The antitumor activity of the HLE DLL3xCD3 bispecific antibody construct (SEQ ID NO: 517) was tested in female NOD / SCID mouse models subcutaneously injected with 5x10 6 human DLL3-positive SCLC (SHP-77 luc) cells or 5x10 6 human DLL3-positive melanoma (WM 266-4) cells. Effector cells (2x10 viable human CD3+ T cells proliferated and activated in vitro 7 ) were intraperitoneally injected on day 12. The start of treatment was on day 16 (WM 266-4) or day 18 (SHP-77 luc). The antibody construct was administered 4 times every 5 days (q5dx4) by intravenous bolus injection. The treatment groups were as follows: -SCLC model (SHP-77 luc) / 7 mice per group ·Vehicle-treated group with T cells ·DLL3xCD3 bispecific antibody construct: 10 mg / kg per administration -Melanoma model (WM266-4) / 9 mice per group ·Vehicle-treated group with T cells ·DLL3xCD3 bispecific antibody construct: 10 mg / kg per administration ·DLL3xCD3 bispecific antibody construct: 2 mg / kg per administration

[0367] During the experiment, tumors were measured with calipers, and progression was evaluated by comparing tumor volumes (TV) between groups. The daily tumor growth inhibition T / C [%] was measured by calculating the tumor volume as T / C% = 100 x (median TV of the analysis group) / (median TV of the control group), and the calculated values are shown in Table 16 below.

[0368] (Table 16) T / C values of mouse xenograft tests using SHP-77 luc cells and WM266-4 cells TIFF2025106535000019.tif90168

[0369] The results are further shown in Figures 8A and 8B. In both tumor models, significant tumor growth inhibition was seen at both tested dose levels of 2 and 10 mg / kg.

[0370] Example 19 Cynomolgus monkey exploratory toxicity test An exploratory toxicity test was conducted using a DLL3xCD3 bispecific antibody construct (SEQ ID NO: 554) that does not extend the half-life. It was administered to three female cynomolgus monkeys by continuous intravenous infusion for 16 days (5, 15, and 45 μg / kg / day for 3 days each, followed by 100 μg / kg / day for 7 days). No clinical observations related to the test article, or changes in body temperature, food consumption, or body weight were seen.

[0371] Consistent with the prediction regarding the pharmacology of the DLL3xCD3 bispecific antibody construct, the blood T lymphocyte population (total T lymphocytes, T helper and cytotoxic T lymphocytes, NK cells, B lymphocytes, and CD25+ activated T lymphocytes) decreased on the first day of administration and remained low throughout the test period in all animals. The activation markers (CD69 and CD25 of activated T cells) increased on the first day but did not increase at subsequent time points during the test.

[0372] In summary, the DLL3xCD3 bispecific antibody construct showed extremely excellent tolerance even at the maximum test dose (100 μg / kg / day, >300xEC 50 ).

[0373] Example 20 Cynomolgus monkey PK test The cynomolgus monkey PK test was performed using naive male cynomolgus monkeys. Three types of albumin-fused DLL3xCD3 (I2C) bispecific antibody constructs (DLL3-4, DLL3-6, and DLL3-14) were administered as a single intravenous bolus at a concentration of 12 μg / kg. Two animal groups were used for each molecule.

[0374] Each cynomolgus monkey PK test was performed under the same conditions, except for different Fc-fused DLL3-4.

[0375] Blood samples were collected before administration and at 0.05, 0.5, 1, 4, 8, 24, 48, 72, 120, 168, 240, and 336 hours after administration. Serum was prepared to measure the serum concentration of the molecule in the immunoassay. The assay was performed by capturing the antibody construct via its target moiety, and an antibody against the CD3-binding moiety of the construct was used for detection. The serum concentration-time profile was used to determine the PK parameters. Pharmacokinetic parameters were determined using the standard non-compartmental analysis (NCA) method. The following PK parameters were evaluated: AUC inf (area under the serum concentration-time curve), Vss (volume of distribution at steady state), CL (total body clearance), and terminal phase t 1 / 2 (terminal half-life). For all antibody constructs, serum levels could be quantified at all time points in all animals after administration. No clinical observations were made in any of the treated animals.

[0376] The pharmacokinetics of the tested antibody constructs are shown in Figure 9, and the PK parameters are summarized in Table 17 below as the mean values of n = 2.

[0377] The albumin-fused constructs showed a consistently good PK profile in a once-weekly or twice-weekly dosing schedule in human patients. For the Fc-fu...

Claims

1. A bispecific antibody construct comprising a first binding domain that specifically binds to human delta-like 3 (DLL3) on the surface of a target cell and a second binding domain that specifically binds to human CD3 on the surface of a T cell, wherein the first binding domain binds to an epitope of DLL3 contained within the DLL3 extracellular domain (ECD) sequence contained within the region shown in SEQ ID NO: 269, wherein the first binding domain comprises a VH region comprising CDR-H1, CDR-H2, and CDR-H3 having at least 90% sequence identity to CDR-H1, CDR-H2, CDR-H3, CDR-L1, CDR-L2, and CDR-L3, respectively, selected from the group consisting of: and a VL region comprising CDR-L1, CDR-L2, and CDR-L3, a bispecific antibody construct: a) CDR-H1 shown in SEQ ID NO: 171, CDR-H2 shown in SEQ ID NO: 172, CDR-H3 shown in SEQ ID NO: 173, CDR-L1 shown in SEQ ID NO: 174, CDR-L2 shown in SEQ ID NO: 175, and CDR-L3 shown in SEQ ID NO: 176; b) CDR-H1 shown in SEQ ID NO: 181, CDR-H2 shown in SEQ ID NO: 182, CDR-H3 shown in SEQ ID NO: 183, CDR-L1 shown in SEQ ID NO: 184, CDR-L2 shown in SEQ ID NO: 185, and CDR-L3 shown in SEQ ID NO: 186; c) CDR-H1 shown in SEQ ID NO: 191, CDR-H2 shown in SEQ ID NO: 192, CDR-H3 shown in SEQ ID NO: 193, CDR-L1 shown in SEQ ID NO: 194, CDR-L2 shown in SEQ ID NO: 195, and CDR-L3 shown in SEQ ID NO: 196; and d) CDR-H1 shown in SEQ ID NO: 201, CDR-H2 shown in SEQ ID NO: 202, CDR-H3 shown in SEQ ID NO: 203, CDR-L1 shown in SEQ ID NO: 204, CDR-L2 shown in SEQ ID NO: 205, and CDR-L3 shown in SEQ ID NO:

206.

2. A bispecific antibody construct comprising a first binding domain that binds to DLL3 on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain binds to an epitope of DLL3 contained within the DLL3 extracellular domain (ECD) sequence contained within the region shown in SEQ ID NO: 269, Here, a bispecific antibody construct comprising a VH region comprising CDR-H1, CDR-H2, and CDR-H3 selected from the group consisting of the following, and a VL region comprising CDR-L1, CDR-L2, and CDR-L3: a) CDR-H1 shown in SEQ ID NO: 171, CDR-H2 shown in SEQ ID NO: 172, CDR-H3 shown in SEQ ID NO: 173, CDR-L1 shown in SEQ ID NO: 174, CDR-L2 shown in SEQ ID NO: 175, and CDR-L3 shown in SEQ ID NO: 176; wherein SEQ ID NO: 171 and SEQ ID NO: 175 each have 0 or 1 substitution, and further wherein SEQ ID NO: 172, SEQ ID NO: 173, SEQ ID NO: 174, and SEQ ID NO: 176 each have 0 to 2 substitutions; b) CDR-H1 shown in SEQ ID NO: 181, CDR-H2 shown in SEQ ID NO: 182, CDR-H3 shown in SEQ ID NO: 183, CDR-L1 shown in SEQ ID NO: 184, CDR-L2 shown in SEQ ID NO: 185, and CDR-L3 shown in SEQ ID NO: 186; wherein SEQ ID NO: 181 and SEQ ID NO: 185 each have 0 or 1 substitution, and further wherein SEQ ID NO: 182, SEQ ID NO: 183, SEQ ID NO: 184, and SEQ ID NO: 186 each have 0 to 2 substitutions; c) CDR-H1 shown in SEQ ID NO: 191, CDR-H2 shown in SEQ ID NO: 192, CDR-H3 shown in SEQ ID NO: 193, CDR-L1 shown in SEQ ID NO: 194, CDR-L2 shown in SEQ ID NO: 195, and CDR-L3 shown in SEQ ID NO: 196; wherein SEQ ID NO: 191 and SEQ ID NO: 195 each have 0 or 1 substitution, and further wherein SEQ ID NO: 192, SEQ ID NO: 193, SEQ ID NO: 194, and SEQ ID NO: 196 each have 0 to 2 substitutions; and d) CDR-H1 shown in SEQ ID NO: 201, CDR-H2 shown in SEQ ID NO: 202, CDR-H3 shown in SEQ ID NO: 203, CDR-L1 shown in SEQ ID NO: 204, CDR-L2 shown in SEQ ID NO: 205, and CDR-L3 shown in SEQ ID NO: 206; wherein SEQ ID NO: 201 and SEQ ID NO: 205 each have 0 or 1 substitution, and further wherein SEQ ID NO: 202, SEQ ID NO: 203, SEQ ID NO: 204, and SEQ ID NO: 206 each have 0 to 2 substitutions.

3. The antibody construct according to claim 1 or 2, wherein the first binding domain further binds to cynomolgus DLL3.

4. The antibody construct according to claim 3, wherein the cynomolgus DLL3 is Macaca fascicularis DLL3.

5. The antibody construct according to any one of claims 1 to 4, wherein the second binding domain further binds to cynomolgus CD3.

6. The antibody construct according to claim 5, wherein the second binding domain binds to human CD3 epsilon and cynomolgus CD3 epsilon.

7. The antibody construct according to any one of claims 1 to 6, wherein the second binding domain binds to human CD3 epsilon and CD3 epsilon of common marmoset (Callithrix jacchus), cottontop tamarin (Saguinus Oedipus), or common squirrel monkey (Saimiri sciureus).

8. The antibody construct according to any one of claims 1 to 7, wherein the first binding domain comprises a VH region having at least 90% sequence identity to a sequence selected from the group consisting of the sequences shown in SEQ ID NO: 177, SEQ ID NO: 187, SEQ ID NO: 197, and SEQ ID NO:

207.

9. The antibody construct according to any one of claims 1 to 8, wherein the first binding domain comprises a VL region having at least 90% sequence identity to a sequence selected from the group consisting of the sequences shown in SEQ ID NO: 178, SEQ ID NO: 188, SEQ ID NO: 198, and SEQ ID NO:

208.

10. The antibody construct according to any one of claims 1 to 9, wherein the first binding domain comprises a VH region and a VL region having at least 90% sequence identity to a pair of amino acid sequences respectively selected from the group consisting of the sequences shown in SEQ ID NO: 177 and 178, SEQ ID NO: 187 and 188, SEQ ID NO: 197 and 198, and SEQ ID NO: 207 and 208.

11. The antibody construct according to any one of claims 1 to 10, wherein the first binding domain comprises a polypeptide having at least 90% sequence identity to a sequence selected from the group consisting of the sequences shown in SEQ ID NO: 179, SEQ ID NO: 189, SEQ ID NO: 199, and SEQ ID NO:

209.

12. The antibody construct according to any one of claims 1 to 11, wherein the second binding domain binds to one or more residues included in the range of amino acid residues 1 to 27 of the human CD3 epsilon extracellular domain.

13. The antibody construct according to any one of claims 1 to 12, comprising a polypeptide having at least 90% sequence identity to a sequence selected from the group consisting of those set forth in SEQ ID NO: 180, SEQ ID NO: 190, SEQ ID NO: 200, and SEQ ID NO:

210.

14. (scFv) ２ The antibody construct according to any one of claims 1 to 13, which is in the form of

15. The antibody construct according to any one of claims 1 to 13, which is in the form of a scFv single domain mAb.

16. The antibody construct according to any one of claims 1 to 13, wherein the first binding domain, the second binding domain, or both are in Fab or scFab form.

17. The antibody construct according to any one of claims 1 to 16, further comprising a third domain that extends the serum half-life of the antibody construct.

18. The antibody construct according to claim 17, wherein the third domain comprises an Fc region.

19. A polynucleotide encoding the antibody construct according to any one of claims 1 to 18.

20. A vector comprising the polynucleotide according to claim 19.

21. A host cell transformed or transfected with the polynucleotide according to claim 19 or the vector according to claim 20.

22. A method for producing the antibody construct according to any one of claims 1 to 18, comprising culturing the host cell according to claim 21 under conditions that allow expression of the antibody construct, and recovering the antibody construct from the culture.

23. A pharmaceutical composition comprising the antibody construct according to any one of claims 1 to 18, or the antibody construct produced according to the method according to claim 22, and one or more excipients.

24. A kit comprising the pharmaceutical composition according to claim 23.

25. The antibody construct according to any one of claims 1 to 18, or the antibody construct produced according to the method according to claim 22, for use in the prevention, treatment, or amelioration of a tumor or cancer disease.

26. The antibody construct according to claim 25, wherein the tumor or cancer disease is a tumor or cancer disease that expresses DLL3.

27. The antibody construct according to claim 26, wherein the tumor or cancer disease is (a) a tumor or cancer of the lung, breast, cervix, colon, colorectal, endometrium, head and neck, liver, ovary, pancreas, prostate, skin, stomach, testis, thyroid, adrenal gland, kidney, bladder, uterus, esophagus, urothelium, or brain, (b) lymphoma, carcinoma, or sarcoma, or (c) a metastatic cancer disease derived from any of the foregoing.

28. The antibody construct according to any one of claims 25 to 27, wherein the tumor or cancer disease is small cell lung cancer (SCLC), non-small cell lung cancer (NSCLC), glioma, glioblastoma, melanoma, neuroendocrine prostate cancer, neuroendocrine pancreatic cancer, hepatoblastoma, hepatocellular carcinoma, or a metastatic cancer disease derived from any of the foregoing.

29. The kit according to claim 24, further comprising instructions for use for use in the prevention, treatment, or amelioration of a tumor or cancer disease expressing DLL3.

30. A bispecific antibody construct comprising a first binding domain that specifically binds to human DLL3 on the surface of a target cell and a second binding domain that specifically binds to human CD3 on the surface of a T cell, wherein the first binding domain binds to an epitope of DLL3 contained within the region shown in SEQ ID NO: 259, wherein the first binding domain comprises a VH region comprising CDR-H1, CDR-H2, and CDR-H3 having at least 90% sequence identity to CDR-H1, CDR-H2, CDR-H3, CDR-L1, CDR-L2, and CDR-L3, respectively, selected from the group consisting of: and a VL region comprising CDR-L1, CDR-L2, and CDR-L3: a) CDR-H1 shown in SEQ ID NO: 121, CDR-H2 shown in SEQ ID NO: 122, CDR-H3 shown in SEQ ID NO: 123, CDR-L1 shown in SEQ ID NO: 124, CDR-L2 shown in SEQ ID NO: 125, and CDR-L3 shown in SEQ ID NO: 126; b) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; c) CDR-H1 shown in SEQ ID NO: 141, CDR-H2 shown in SEQ ID NO: 142, CDR-H3 shown in SEQ ID NO: 143, CDR-L1 shown in SEQ ID NO: 144, CDR-L2 shown in SEQ ID NO: 145, and CDR-L3 shown in SEQ ID NO: 146; d) CDR-H1 shown in SEQ ID NO: 151, CDR-H2 shown in SEQ ID NO: 152, CDR-H3 shown in SEQ ID NO: 153, CDR-L1 shown in SEQ ID NO: 154, CDR-L2 shown in SEQ ID NO: 155, and CDR-L3 shown in SEQ ID NO: 156; e) CDR-H1 shown in SEQ ID NO: 161, CDR-H2 shown in SEQ ID NO: 162, CDR-H3 shown in SEQ ID NO: 163, CDR-L1 shown in SEQ ID NO: 164, CDR-L2 shown in SEQ ID NO: 165, and CDR-L3 shown in SEQ ID NO: 166; f) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 439, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; g) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 440, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; h) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 441, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; i) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 442, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; j) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 443, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; k) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 444, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; l) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 439, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 441, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; and m) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 440, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 442, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO:

136. **Claim 31** A bispecific antibody construct comprising a first binding domain that specifically binds to human DLL3 on the target cell surface and a second binding domain that specifically binds to human CD3 on the T cell surface, wherein the first binding domain binds to an epitope of DLL3 contained within the region shown in SEQ ID NO: 259, wherein the first binding domain comprises a VH region comprising CDR-H1, CDR-H2, and CDR-H3 selected from the group consisting of: and a VL region comprising CDR-L1, CDR-L2, and CDR-L3: a) CDR-H1 shown in SEQ ID NO: 121, CDR-H2 shown in SEQ ID NO: 122, CDR-H3 shown in SEQ ID NO: 123, CDR-L1 shown in SEQ ID NO: 124, CDR-L2 shown in SEQ ID NO: 125, and CDR-L3 shown in SEQ ID NO: 126; wherein SEQ ID NO: 121 and SEQ ID NO: 125 each have 0 or 1 substitution, and further wherein SEQ ID NO: 122, SEQ ID NO: 123, SEQ ID NO: 124, and SEQ ID NO: 126 each have 0-2 substitutions; b) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; wherein SEQ ID NO: 131 and SEQ ID NO: 135 each have 0 or 1 substitution, and further wherein SEQ ID NO: 132, SEQ ID NO: 133, SEQ ID NO: 134, and SEQ ID NO: 136 each have 0 to 2 substitutions; c) CDR-H1 shown in SEQ ID NO: 141, CDR-H2 shown in SEQ ID NO: 142, CDR-H3 shown in SEQ ID NO: 143, CDR-L1 shown in SEQ ID NO: 144, CDR-L2 shown in SEQ ID NO: 145, and CDR-L3 shown in SEQ ID NO: 146; wherein SEQ ID NO: 141 and SEQ ID NO: 145 each have 0 or 1 substitution, and further wherein SEQ ID NO: 142, SEQ ID NO: 143, SEQ ID NO: 144, and SEQ ID NO: 146 each have 0 to 2 substitutions; d) CDR-H1 shown in SEQ ID NO: 151, CDR-H2 shown in SEQ ID NO: 152, CDR-H3 shown in SEQ ID NO: 153, CDR-L1 shown in SEQ ID NO: 154, CDR-L2 shown in SEQ ID NO: 155, and CDR-L3 shown in SEQ ID NO: 156; wherein SEQ ID NO: 151 and SEQ ID NO: 155 each have 0 or 1 substitution, and further wherein SEQ ID NO: 152, SEQ ID NO: 153, SEQ ID NO: 154, and SEQ ID NO: 156 each have 0 to 2 substitutions; e) CDR-H1 shown in SEQ ID NO: 161, CDR-H2 shown in SEQ ID NO: 162, CDR-H3 shown in SEQ ID NO: 163, CDR-L1 shown in SEQ ID NO: 164, CDR-L2 shown in SEQ ID NO: 165, and CDR-L3 shown in SEQ ID NO: 166; wherein SEQ ID NO: 161 and SEQ ID NO: 165 each have 0 or 1 substitution, and further wherein SEQ ID NO: 162, SEQ ID NO: 163, SEQ ID NO: 164, and SEQ ID NO: 166 each have 0 to 2 substitutions; f) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 439, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; wherein SEQ ID NO: 131 and SEQ ID NO: 135 each have 0 or 1 substitution, and further wherein SEQ ID NO: 439, SEQ ID NO: 133, SEQ ID NO: 134, and SEQ ID NO: 136 each have 0 to 2 substitutions; g) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 440, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 134, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; wherein SEQ ID NO: 131 and SEQ ID NO: 135 each have 0 or 1 substitution, and further wherein SEQ ID NO: 440, SEQ ID NO: 133, SEQ ID NO: 134, and SEQ ID NO: 136 each have 0 to 2 substitutions; h) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 441, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; wherein SEQ ID NO: 131 and SEQ ID NO: 135 each have 0 or 1 substitution, and further wherein SEQ ID NO: 441, SEQ ID NO: 132, SEQ ID NO: 133, and SEQ ID NO: 136 each have 0 to 2 substitutions; i) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 442, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; wherein SEQ ID NO: 131 and SEQ ID NO: 135 each have 0 or 1 substitution, and further wherein SEQ ID NO: 132, SEQ ID NO: 133, SEQ ID NO: 442, and SEQ ID NO: 136 each have 0 to 2 substitutions; j) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 443, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; wherein SEQ ID NO: 131 and SEQ ID NO: 135 each have 0 or 1 substitution, and further wherein SEQ ID NO: 132, SEQ ID NO: 133, SEQ ID NO: 443, and SEQ ID NO: 136 each have 0 to 2 substitutions; k) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 132, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 444, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; wherein SEQ ID NO: 131 and SEQ ID NO: 135 each have 0 or 1 substitution, and further wherein SEQ ID NO: 132, SEQ ID NO: 133, SEQ ID NO: 444, and SEQ ID NO: 136 each have 0 to 2 substitutions; l) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 439, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 441, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; wherein SEQ ID NO: 131 and SEQ ID NO: 135 each have 0 or 1 substitution, and further wherein SEQ ID NO: 439, SEQ ID NO: 133, SEQ ID NO: 441, and SEQ ID NO: 136 each have 0 to 2 substitutions; and m) CDR-H1 shown in SEQ ID NO: 131, CDR-H2 shown in SEQ ID NO: 440, CDR-H3 shown in SEQ ID NO: 133, CDR-L1 shown in SEQ ID NO: 442, CDR-L2 shown in SEQ ID NO: 135, and CDR-L3 shown in SEQ ID NO: 136; wherein SEQ ID NO: 131 and SEQ ID NO: 135 each have 0 or 1 substitution, and further wherein SEQ ID NO: 440, SEQ ID NO: 133, SEQ ID NO: 442, and SEQ ID NO: 136 each have 0 to 2 substitutions.

32. The antibody construct according to claim 30 or 31, wherein the first binding domain further binds to cynomolgus DLL3.

33. The antibody construct according to claim 32, wherein the cynomolgus DLL3 is Macaca fascicularis DLL3.

34. The antibody construct according to any one of claims 30 to 33, wherein the second binding domain further binds to cynomolgus CD3.

35. The antibody construct according to claim 34, wherein the second binding domain binds to human CD3 epsilon and cynomolgus CD3 epsilon.

36. The antibody construct according to any one of claims 30 to 35, wherein the second binding domain binds to human CD3 epsilon and CD3 epsilon of common marmoset (Callithrix jacchus), cotton-top tamarin (Saguinus Oedipus), or common squirrel monkey (Saimiri sciureus).

37. (scFv) ２ The antibody construct according to any one of claims 30 to 36, which is in a form selected from the group consisting of (scFv), scFv single-domain mAb, diabody, and oligomers thereof.

38. (scFv) ２ The antibody construct according to claim 37, which is in the form of

39. The antibody construct according to claim 37, which is in the form of a scFv single-domain mAb.

40. The antibody construct according to any one of claims 30 to 39, further comprising a third domain that extends the serum half-life of the antibody construct.

41. The antibody construct according to any one of claims 30 to 40, wherein the first binding domain comprises a VH region having at least 90% sequence identity to a sequence selected from the group consisting of the sequences shown in SEQ ID NO: 127, SEQ ID NO: 137, SEQ ID NO: 147, SEQ ID NO: 157, SEQ ID NO: 167, SEQ ID NO: 445, SEQ ID NO: 446, SEQ ID NO: 447, SEQ ID NO: 448, SEQ ID NO: 449, SEQ ID NO: 450, SEQ ID NO: 451, SEQ ID NO: 452, SEQ ID NO: 453, SEQ ID NO: 454, and SEQ ID NO:

455.

42. The antibody construct according to any one of claims 30 to 41, wherein the first binding domain comprises a VL region having at least 90% sequence identity to a sequence selected from the group consisting of the sequences shown in SEQ ID NO: 128, SEQ ID NO: 138, SEQ ID NO: 148, SEQ ID NO: 158, SEQ ID NO: 168, SEQ ID NO: 456, SEQ ID NO: 457, SEQ ID NO: 458, SEQ ID NO: 459, SEQ ID NO: 460, SEQ ID NO: 461, SEQ ID NO: 462, SEQ ID NO: 463, SEQ ID NO: 464, SEQ ID NO: 465, SEQ ID NO: 466, SEQ ID NO: 467, SEQ ID NO: 468, SEQ ID NO: 469, and SEQ ID NO:

470.

43. The antibody construct according to any one of claims 30 to 42, wherein the first binding domain comprises a VH region and a VL region having at least 90% sequence identity to a pair of VH region and VL region selected from the group consisting of pairs of VH region and VL region shown in SEQ ID NO: 127 + 128, SEQ ID NO: 137 + 138, SEQ ID NO: 147 + 148, SEQ ID NO: 157 + 158, SEQ ID NO: 167 + 168, SEQ ID NO: 137 + 456, SEQ ID NO: 137 + 457, SEQ ID NO: 137 + 458, SEQ ID NO: 137 + 459, SEQ ID NO: 137 + 460, SEQ ID NO: 445 + 138, SEQ ID NO: 446 + 138, SEQ ID NO: 447 + 138, SEQ ID NO: 445 + 460, SEQ ID NO: 448 + 461, SEQ ID NO: 449 + 462, SEQ ID NO: 450 + 463, SEQ ID NO: 450 + 464, SEQ ID NO: 450 + 465, SEQ ID NO: 450 + 466, SEQ ID NO: 450 + 467, SEQ ID NO: 450 + 468, SEQ ID NO: 451 + 463, SEQ ID NO: 452 + 463, SEQ ID NO: 453 + 463, SEQ ID NO: 451 + 468, SEQ ID NO: 454 + 469, and SEQ ID NO: 455 + 470.

44. The antibody construct according to any one of claims 30 to 43, wherein the first binding domain comprises a polypeptide having at least 90% sequence identity to a sequence selected from the group consisting of those shown in SEQ ID NO: 129, SEQ ID NO: 139, SEQ ID NO: 149, SEQ ID NO: 159, SEQ ID NO: 169, SEQ ID NO: 471, SEQ ID NO: 472, SEQ ID NO: 473, SEQ ID NO: 474, SEQ ID NO: 475, SEQ ID NO: 476, SEQ ID NO: 477, SEQ ID NO: 478, SEQ ID NO: 479, SEQ ID NO: 480, SEQ ID NO: 481, SEQ ID NO: 482, SEQ ID NO: 483, SEQ ID NO: 484, SEQ ID NO: 485, SEQ ID NO: 486, SEQ ID NO: 487, SEQ ID NO: 488, SEQ ID NO: 489, SEQ ID NO: 490, SEQ ID NO: 491, SEQ ID NO: 492, and SEQ ID NO:

493.

45. An antibody construct according to any one of claims 30 to 44, comprising a polypeptide having at least 90% sequence identity to a sequence selected from the group consisting of SEQ ID NO: 130, SEQ ID NO: 140, SEQ ID NO: 150, SEQ ID NO: 160, SEQ ID NO: 170; SEQ ID NO: 218, SEQ ID NO: 219, SEQ ID NO: 220, SEQ ID NO: 494, SEQ ID NO: 495, SEQ ID NO: 496, SEQ ID NO: 497, SEQ ID NO: 498, SEQ ID NO: 499, SEQ ID NO: 500, SEQ ID NO: 501, SEQ ID NO: 502, SEQ ID NO: 503, SEQ ID NO: 504, SEQ ID NO: 505, SEQ ID NO: 506, SEQ ID NO: 507, SEQ ID NO: 508, SEQ ID NO: 509, SEQ ID NO: 510, SEQ ID NO: 511, SEQ ID NO: 512, SEQ ID NO: 513, SEQ ID NO: 514, SEQ ID NO: 515, and SEQ ID NO:

516.

46. The antibody construct according to any one of claims 30 to 45, wherein the second binding domain binds to one or more residues included in the range of amino acid residues 1 to 27 of the human CD3 epsilon extracellular domain.

47. The antibody construct according to any one of claims 30 to 46, wherein the second binding domain comprises a VL region comprising CDR-L1, CDR-L2, and CDR-L3 selected from the following, and a VH region comprising CDR-H1, CDR-H2, and CDR-H3: a) CDR-L1 shown in SEQ ID NO: 342, CDR-L2 shown in SEQ ID NO: 343, CDR-L3 shown in SEQ ID NO: 344, CDR-H1 shown in SEQ ID NO: 345, CDR-H2 shown in SEQ ID NO: 346, and CDR-H3 shown in SEQ ID NO: 347; b) CDR-L1 shown in SEQ ID NO: 351, CDR-L2 shown in SEQ ID NO: 352, CDR-L3 shown in SEQ ID NO: 353, CDR-H1 shown in SEQ ID NO: 354, CDR-H2 shown in SEQ ID NO: 355, and CDR-H3 shown in SEQ ID NO: 356; c) CDR-L1 shown in SEQ ID NO: 360, CDR-L2 shown in SEQ ID NO: 361, CDR-L3 shown in SEQ ID NO: 362, CDR-H1 shown in SEQ ID NO: 363, CDR-H2 shown in SEQ ID NO: 364, and CDR-H3 shown in SEQ ID NO: 365; d) CDR-L1 shown in SEQ ID NO: 369, CDR-L2 shown in SEQ ID NO: 370, CDR-L3 shown in SEQ ID NO: 371, CDR-H1 shown in SEQ ID NO: 372, CDR-H2 shown in SEQ ID NO: 373, and CDR-H3 shown in SEQ ID NO: 374; e) CDR-L1 shown in SEQ ID NO: 378, CDR-L2 shown in SEQ ID NO: 379, CDR-L3 shown in SEQ ID NO: 380, CDR-H1 shown in SEQ ID NO: 381, CDR-H2 shown in SEQ ID NO: 382, and CDR-H3 shown in SEQ ID NO: 383; f) CDR-L1 shown in SEQ ID NO: 387, CDR-L2 shown in SEQ ID NO: 388, CDR-L3 shown in SEQ ID NO: 389, CDR-H1 shown in SEQ ID NO: 390, CDR-H2 shown in SEQ ID NO: 391, and CDR-H3 shown in SEQ ID NO: 392; g) CDR-L1 shown in SEQ ID NO: 396, CDR-L2 shown in SEQ ID NO: 397, CDR-L3 shown in SEQ ID NO: 398, CDR-H1 shown in SEQ ID NO: 399, CDR-H2 shown in SEQ ID NO: 400, and CDR-H3 shown in SEQ ID NO: 401; h) CDR-L1 shown in SEQ ID NO: 405, CDR-L2 shown in SEQ ID NO: 406, CDR-L3 shown in SEQ ID NO: 407, CDR-H1 shown in SEQ ID NO: 408, CDR-H2 shown in SEQ ID NO: 409, and CDR-H3 shown in SEQ ID NO: 410; i) CDR-L1 shown in SEQ ID NO: 414, CDR-L2 shown in SEQ ID NO: 415, CDR-L3 shown in SEQ ID NO: 416, CDR-H1 shown in SEQ ID NO: 417, CDR-H2 shown in SEQ ID NO: 418, and CDR-H3 shown in SEQ ID NO: 419; or j) CDR-L1 shown in SEQ ID NO: 423, CDR-L2 shown in SEQ ID NO: 424, CDR-L3 shown in SEQ ID NO: 425, CDR-H1 shown in SEQ ID NO: 426, CDR-H2 shown in SEQ ID NO: 427, and CDR-H3 shown in SEQ ID NO:

428.

48. The antibody construct according to claim 47, comprising a VH region comprising an amino acid sequence having at least 90% sequence identity to a sequence selected from the group consisting of those set forth in SEQ ID NO: 348, SEQ ID NO: 357, SEQ ID NO: 366, SEQ ID NO: 375, SEQ ID NO: 384, SEQ ID NO: 393, SEQ ID NO: 402, SEQ ID NO: 411, SEQ ID NO: 420, and SEQ ID NO:

429.

49. The antibody construct according to claim 47 or 48, comprising a VL region comprising an amino acid sequence having at least 90% sequence identity to a sequence selected from the group consisting of those set forth in SEQ ID NO: 349, SEQ ID NO: 358, SEQ ID NO: 367, SEQ ID NO: 376, SEQ ID NO: 385, SEQ ID NO: 394, SEQ ID NO: 403, SEQ ID NO: 412, SEQ ID NO: 421, and SEQ ID NO:

430.

50. The antibody construct according to any one of claims 47 to 49, comprising a VH region and a VL region having at least 90% sequence identity to a pair of VH regions and VL regions each comprising a pair of amino acid sequences selected from the group consisting of those set forth in SEQ ID NO: 348 and 349, SEQ ID NO: 357 and 358, SEQ ID NO: 366 and 367, SEQ ID NO: 375 and 376, SEQ ID NO: 384 and 385, SEQ ID NO: 393tp394, SEQ ID NO: 402 and 403, SEQ ID NO: 411 and 412, SEQ ID NO: 420 and 421, and SEQ ID NO: 429 and 430.

51. The antibody construct according to any one of claims 47 to 50, comprising an amino acid sequence having at least 90% sequence identity to a sequence selected from the group consisting of those set forth in SEQ ID NO: 350, SEQ ID NO: 359, SEQ ID NO: 368, SEQ ID NO: 377, SEQ ID NO: 386, SEQ ID NO: 395, SEQ ID NO: 404, SEQ ID NO: 413, SEQ ID NO: 422, and SEQ ID NO:

431.

52. The antibody construct according to any one of claims 47 to 51, comprising a polypeptide having at least 90% sequence identity to a sequence selected from the group consisting of those set forth in SEQ ID NO: 525, SEQ ID NO: 526, SEQ ID NO: 527, and SEQ ID NO:

528.

53. A polynucleotide encoding the antibody construct according to any one of claims 30 to 52.

54. A vector comprising the polynucleotide according to claim 53.

55. A host cell transformed or transfected with the polynucleotide according to claim 53 or the vector according to claim 54.

56. A method for producing an antibody construct according to any one of claims 30 to 52, the method comprising culturing the host cell according to claim 55 under conditions enabling the expression of the antibody construct, and recovering the antibody construct from the culture.

57. A pharmaceutical composition comprising an antibody construct according to any one of claims 30 to 52, or an antibody construct produced according to the method of claim 56, and a carrier, stabilizer, excipient, diluent, solubilizer, surfactant, emulsifier, preservative, or adjuvant.

58. An antibody construct according to any one of claims 30 to 52 or an antibody construct produced according to the method of claim 56 for use in the prevention, treatment, or amelioration of a tumor or cancer disease.

59. A kit comprising an antibody construct according to any one of claims 30 to 52, an antibody construct produced according to the method of claim 56, the polynucleotide according to claim 53, the vector according to claim 54, and / or the host cell according to claim 55, and instructions for use for use in the prevention, treatment, or amelioration of a tumor or cancer disease.

60. The antibody construct according to claim 58 or the kit according to claim 59, wherein the tumor or cancer disease is (a) a tumor or cancer of the lung, breast, cervix, colon, colorectal, endometrium, head and neck, liver, ovary, pancreas, prostate, skin, stomach, testis, thyroid, adrenal gland, kidney, bladder, uterus, esophagus, urothelium, or brain, (b) lymphoma, carcinoma, or sarcoma, or (c) a metastatic cancer disease derived from any of the foregoing.

61. The antibody construct according to claim 58 or 60, or the kit according to claim 59 or 60, wherein the tumor or cancer disease is small cell lung cancer (SCLC), non-small cell lung cancer (NSCLC), glioma, glioblastoma, melanoma, neuroendocrine prostate cancer, neuroendocrine pancreatic cancer, hepatoblastoma, hepatocellular carcinoma, or a metastatic cancer disease derived from any of the foregoing.

62. The antibody construct according to claim 58, 60 or 61, or the kit according to claim 59, 60 or 61, wherein the tumor or cancer disease is a tumor or cancer disease that expresses DLL3.

Images