DLL3 / CD3 binding proteins for treatment of cancer

Abstract

The present invention relates to novel DLL3 / CD3 binding proteins. The invention also relates to the use of such proteins or compositions thereof, in particular for therapeutic purposes in the field of cancer diseases.

Description

[0001] This application is a divisional application of the Chinese patent application with application date of June 7, 2019 and application number 201980051335.7. Background of the Invention Technical Field

[0003] The present invention relates to a multispecific binding protein comprising a first antigen binding unit specific for DLL3 and a second antigen binding unit specific for CD3. The present invention also relates to nucleic acids encoding such binding proteins; methods for preparing such binding proteins; host cells expressing or capable of expressing such binding proteins; compositions comprising such binding proteins and uses of such binding proteins or such compositions, in particular in the field of cancer diseases for therapeutic purposes. Background Art

[0004] DLL3 belongs to the type I transmembrane protein of the Notch ligand family. DLL3 is mainly involved in somitogenesis, where it is mainly located at the high-gland bodies and acts as an inhibitor of Notch signaling compared to other DLL family members DLL1 and DLL4 located on the cell surface. Only in cancer cells that overexpress DLL3, some DLL3 molecules escape to the cell surface (Dylla, Molecular & Cellular Oncology 2016, Vol. 3, No. 2). DLL3 has been identified as a tumor-specific target by using several methods, such as LC / MS analysis of tumor tissue (WO 2014 / 125273), RNA sequencing (WO 2017 / 021349) and immunohistochemistry (Saunders et al., Sci Transl Med. 2015 Aug 26; 7(302):302ra136; Saunders et al., AACR 2017; WO 2013 / 126746) analysis.

[0005] DLL3 expression has been described in neuroendocrine tumors such as large cell neuroendocrine carcinoma (LCNEC), small cell lung cancer (SCLC), neuroendocrine prostate cancer, neuroendocrine pancreatic cancer, and glioblastoma (Saunders et al., Sci Transl Med. 2015 Aug 26;7(302):302ra136; Saunders et al., AACR 2017).

[0006] SCLC represents an indication with extremely high medical need. SCLC accounts for 13% of lung cancer diagnoses and has a poorer prognosis than NSCLC. Conventional treatments for these cancers mainly include chemotherapy, radiation therapy, surgery or a combination thereof, and no targeted therapies are available. Despite high initial response rates to chemotherapy, many patients rapidly relapse with chemotherapy-resistant disease, and therefore no treatment options are available. Although these cancer treatments have improved over the past few years, the overall survival rate for these tumor types has remained unchanged, and therefore, there is an unmet need for more targeted and potent therapies.

[0007] One targeted therapy approach is provided by antibody-drug conjugates (ADCs), however, for DLL3, this strategy is not preferred due to the low cell surface expression of DLL3 and the high resistance rate to chemotherapy. Because most patients relapse after chemotherapy treatment and due to the low expression of DLL3 on the cell surface, targeting DLL3 with ADCs may have limitations. In addition, ADC approaches often have off-target toxicity caused by free-state drugs due to linker instability or degradation.

[0008] Attempts have been made to target DLL3 in combination with targeting of other proteins. For example, WO2017 / 021349 describes a bispecific antibody construct that combines binding to human DLL3 on the surface of target cells with binding to human CD3 on the surface of T cells. However, such approaches have not proven successful and to date no targeted therapies are available for SCLC and glioblastomas and other tumors expressing DLL3.

[0009] In view of the poor outlook for such cancer patients, it is necessary to identify more effective therapies, especially effective therapies with improved tolerance. Therefore, one object of the present invention is to provide pharmacologically active agents, compositions and / or methods of treatment that provide certain advantages over currently used and / or known agents, compositions and / or methods in the art. These advantages include in vivo efficacy, improved therapeutic and pharmacological properties, increased specificity, improved safety profiles, fewer side effects, reduced immunogenicity, and other favorable properties, such as improved ease of preparation or reduced commodity cost, higher stability / longer half-life, less frequent demand for dosing regimens, especially compared to candidate drugs known in the art. Summary of the invention

[0010] The present invention is based on a bispecific T cell engagement approach that employs a multispecific binding protein having a binding arm that binds to CD3 on T cells and a binding arm that binds to DLL3 on the cell surface of tumor cells. Through simultaneous binding to T cells and tumor cells, the T cell engager forces the formation of a cytolytic synapse between the two cells and thereby selectively redirects T cell activity toward targeted tumor cells.

[0011] In one aspect, the present invention provides a multispecific binding protein comprising a first antigen binding unit that specifically binds to DLL3 and a second antigen binding unit that specifically binds to CD3, wherein the first antigen binding unit that specifically binds to DLL3 is selected from the group consisting of i) to xviii):

[0012] i) an antigen-binding unit comprising: a light chain CDR comprising the amino acid sequences of SEQ ID NO: 1 (CDR1), SEQ ID NO: 2 (CDR2) and SEQ ID NO: 3 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 4 (CDR1), SEQ ID NO: 5 (CDR2) and SEQ ID NO: 6 (CDR3);

[0013] ii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 7 (CDR1), SEQ ID NO: 8 (CDR2) and SEQ ID NO: 9 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 10 (CDR1), SEQ ID NO: 11 (CDR2) and SEQ ID NO: 12 (CDR3);

[0014] iii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 13 (CDR1), SEQ ID NO: 14 (CDR2) and SEQ ID NO: 15 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 16 (CDR1), SEQ ID NO: 17 (CDR2) and SEQ ID NO: 18 (CDR3);

[0015] iv) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 19 (CDR1), SEQ ID NO: 20 (CDR2) and SEQ ID NO: 21 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 22 (CDR1), SEQ ID NO: 23 (CDR2) and SEQ ID NO: 24 (CDR3);

[0016] v) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 25 (CDR1), SEQ ID NO: 26 (CDR2) and SEQ ID NO: 27 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 28 (CDR1), SEQ ID NO: 29 (CDR2) and SEQ ID NO: 30 (CDR3);

[0017] vi) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 31 (CDR1), SEQ ID NO: 32 (CDR2) and SEQ ID NO: 33 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 34 (CDR1), SEQ ID NO: 35 (CDR2) and SEQ ID NO: 36 (CDR3);

[0018] vii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 133 (CDR1), SEQ ID NO: 134 (CDR2) and SEQ ID NO: 135 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 136 (CDR1), SEQ ID NO: 137 (CDR2) and SEQ ID NO: 138 (CDR3);

[0019] viii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 139 (CDR1), SEQ ID NO: 140 (CDR2) and SEQ ID NO: 141 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 142 (CDR1), SEQ ID NO: 143 (CDR2) and SEQ ID NO: 144 (CDR3);

[0020] ix) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 145 (CDR1), SEQ ID NO: 146 (CDR2) and SEQ ID NO: 147 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 148 (CDR1), SEQ ID NO: 149 (CDR2) and SEQ ID NO: 150 (CDR3);

[0021] x) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 151 (CDR1), SEQ ID NO: 152 (CDR2) and SEQ ID NO: 153 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 154 (CDR1), SEQ ID NO: 155 (CDR2) and SEQ ID NO: 156 (CDR3);

[0022] xi) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 157 (CDR1), SEQ ID NO: 158 (CDR2) and SEQ ID NO: 159 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 160 (CDR1), SEQ ID NO: 161 (CDR2) and SEQ ID NO: 162 (CDR3);

[0023] xii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 163 (CDR1), SEQ ID NO: 164 (CDR2) and SEQ ID NO: 165 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 166 (CDR1), SEQ ID NO: 167 (CDR2) and SEQ ID NO: 168 (CDR3);

[0024] xiii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 169 (CDR1), SEQ ID NO: 170 (CDR2) and SEQ ID NO: 171 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 172 (CDR1), SEQ ID NO: 173 (CDR2) and SEQ ID NO: 174 (CDR3);

[0025] xiv) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 175 (CDR1), SEQ ID NO: 176 (CDR2) and SEQ ID NO: 177 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 178 (CDR1), SEQ ID NO: 179 (CDR2) and SEQ ID NO: 180 (CDR3);

[0026] xv) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 181 (CDR1), SEQ ID NO: 182 (CDR2) and SEQ ID NO: 183 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 184 (CDR1), SEQ ID NO: 185 (CDR2) and SEQ ID NO: 186 (CDR3);

[0027] xvi) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 187 (CDR1), SEQ ID NO: 188 (CDR2) and SEQ ID NO: 189 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 190 (CDR1), SEQ ID NO: 191 (CDR2) and SEQ ID NO: 192 (CDR3);

[0028] xvii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 193 (CDR1), SEQ ID NO: 194 (CDR2) and SEQ ID NO: 195 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 196 (CDR1), SEQ ID NO: 197 (CDR2) and SEQ ID NO: 198 (CDR3); and

[0029] xviii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 199 (CDR1), SEQ ID NO: 200 (CDR2) and SEQ ID NO: 201 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 202 (CDR1), SEQ ID NO: 203 (CDR2) and SEQ ID NO: 204 (CDR3).

[0030] In some embodiments of the binding protein of the present invention, the first antigen binding unit that specifically binds to DLL3 comprises a first light chain variable domain and a first heavy chain variable domain and is selected from the group consisting of i) to xviii):

[0031] i) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 37 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 38;

[0032] ii) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 39 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 40;

[0033] iii) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 41 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 42;

[0034] iv) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 43 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 44;

[0035] v) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 45 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 46;

[0036] vi) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 47 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 48;

[0037] vii) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 205 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 206;

[0038] viii) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 207 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 208;

[0039] ix) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 209 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 210;

[0040] x) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 211 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 212;

[0041] xi) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 213 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 214;

[0042] xii) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 215 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 216;

[0043] xiii) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 217 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 218;

[0044] xiv) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 219 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 220;

[0045] xv) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 221 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 222;

[0046] xvi) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 223 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 224;

[0047] xvii) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 225 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 226; and

[0048] xviii) an antigen binding unit comprising a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 227 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 228.

[0049] In some embodiments of the binding protein of the present invention, the second antigen binding unit that specifically binds to CD3 is selected from the group consisting of i) to iii):

[0050] i) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 55 (CDR1), SEQ ID NO: 56 (CDR2) and SEQ ID NO: 57 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 58 (CDR1), SEQ ID NO: 59 (CDR2) and SEQ ID NO: 60 (CDR3);

[0051] ii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 61 (CDR1), SEQ ID NO: 62 (CDR2) and SEQ ID NO: 63 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 64 (CDR1), SEQ ID NO: 65 (CDR2) and SEQ ID NO: 66 (CDR3); and

[0052] iii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 96 (CDR1), SEQ ID NO: 97 (CDR2) and SEQ ID NO: 98 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 99 (CDR1), SEQ ID NO: 100 (CDR2) and SEQ ID NO: 101 (CDR3).

[0053] In some embodiments of the binding protein of the present invention, the second antigen binding unit that specifically binds to CD3 comprises a second light chain variable domain and a second heavy chain variable domain and is selected from the group consisting of:

[0054] i) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 67 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 68;

[0055] ii) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 69 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 70; and

[0056] iii) an antigen binding unit comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 102 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 103.

[0057] In some embodiments of the present invention, the first antigen binding unit that specifically binds to DLL3 comprises, from its N-terminus to its C-terminus: a first light chain variable domain, a first light chain constant domain, a first peptide linker, a first heavy chain variable domain, and a first heavy chain constant CH1 domain; and the second antigen binding unit that specifically binds to CD3 comprises, from its N-terminus to its C-terminus: a second light chain variable domain, a second light chain constant domain, a second peptide linker, a second heavy chain variable domain, and a second heavy chain constant CH1 domain. In some embodiments of the present invention, the first and / or second peptide linker comprises 26 to 42 amino acids, preferably any one of 30 to 40 amino acids, 34 to 40 amino acids, or 36 to 39 amino acids, more preferably 38 amino acids. In some embodiments of the present invention, the first linker and / or the second linker is a Gly-Ser linker, which preferably comprises the amino acid sequence SEQ ID NO: 89, more preferably the first and second peptide linkers comprise the same sequence.

[0058] In some embodiments, the binding protein of the present invention further comprises first and second Fc domains, the first Fc domain is covalently linked to the first antigen binding unit, and the second Fc domain is covalently linked to the second antigen binding unit.

[0059] In some embodiments of the present invention,

[0060] i) the first Fc domain comprises a tyrosine (Y) at position 366 [T366Y], and the second Fc domain comprises a threonine (T) at position 407 [Y407T], or

[0061] ii) the first Fc domain comprises a tryptophan (W) at position 366 [T366W], and the second Fc domain comprises a serine (S) at position 366 [T366S], an alanine (A) at position 368 [L368A] and a valine (V) at position 407 [Y407V], or

[0062] iii) the second Fc domain comprises a tyrosine (Y) at position 366 [T366Y], and the first Fc domain comprises a threonine (T) at position 407 [Y407T], or

[0063] iv) the second Fc domain comprises a tryptophan (W) at position 366 [T366W], and the first Fc domain comprises a serine (S) at position 366 [T366S], an alanine (A) at position 368 [L368A] and a valine (V) at position 407 [Y407V],

[0064] Preferably the first or the second Fc domain further comprises an arginine at position 435 [H435R] and a phenylalanine at position 436 [Y436F]. In some embodiments, the first and / or second Fc domain further comprises an alanine at position 234 [L234A] and an alanine at position 235 [L235A]. In some embodiments, the first and second light chain constant domains comprise human kappa or lambda domains.

[0065] In a preferred embodiment, the binding protein of the invention comprises: a first polypeptide chain that specifically binds to DLL3, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO:73, SEQ ID NO:74, SEQ ID NO:75, SEQ ID NO:76, SEQ ID NO:77, SEQ ID NO:78, SEQ ID NO:241, 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, SEQ ID NO:251, and SEQ ID NO:252; and a second polypeptide chain that specifically binds to CD3, which comprises the amino acid sequence of SEQ ID NO:79, SEQ ID NO:80, or SEQ ID NO:105.

[0066] In another aspect, the present invention provides an isolated nucleic acid molecule, which i) encodes the first antigen binding unit and / or the second antigen binding unit of the protein of the present invention, optionally further encoding the first and / or the second Fc domain, or ii) encodes the first and / or the second polypeptide chain of the protein of the present invention. Other aspects provided herein are expression vectors comprising the nucleic acid molecules of the present invention, host cells transfected with such expression vectors, and methods for producing the proteins of the present invention.

[0067] In another aspect of the present invention, provided herein is a multispecific binding protein comprising a first polypeptide chain that specifically binds to DLL3 and a second polypeptide chain that specifically binds to CD3, wherein the first polypeptide chain comprises a first light chain, a first linker and a first heavy chain and the second polypeptide chain comprises a second light chain, a second linker and a second heavy chain, preferably the C-terminus of the first light chain is covalently bonded to the N-terminus of the first heavy chain via the first peptide linker and the C-terminus of the second light chain is covalently bonded to the N-terminus of the second heavy chain via the second peptide linker.

[0068] In some embodiments of the binding proteins of the invention, the first polypeptide chain that specifically binds to DLL3 comprises a light chain variable domain and a heavy chain variable domain, the light chain variable domain and the heavy chain variable domain comprising a CDR sequence and / or a VH / VL sequence as defined for the antigen binding unit of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, or DLL3#18 as described herein. In some embodiments, the second polypeptide chain that specifically binds to CD3 comprises a light chain variable domain and a heavy chain variable domain, the light chain variable domain and the heavy chain variable domain comprising a CDR sequence and / or a VH / VL sequence as defined for the antigen binding unit of CD3#1, CD3#2, or CD3#3 as described herein.

[0069] Other aspects, embodiments, uses and methods involving the binding proteins of the present invention will become apparent from the following detailed description of the invention and the appended claims.

[0070] The present invention provides novel binding proteins that allow for more efficient treatment of cancers expressing DLL3, such as SCLC, glioblastoma or neuroendocrine tumors expressing DLL3. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 : Schematic diagram of the bispecific binding protein of the present invention

[0072] Figure 2 : Schematic representation of DLL3 full-length protein and DLL3 domain constructs expressed on HEK293 cells for epitope domain mapping. For the DLL3 domain constructs, the transmembrane and intracellular domains are derived from EpCAM.

[0073] Figure 3A : Localization of DLL3 epitope domains of seven exemplary DLL3 / CD3 binding proteins. Exemplary DLL3 / CD3 binding proteins recognize the membrane-proximal peptides EGF4, EGF1, and DSL domains of DLL3. The y-axis depicts counts and the x-axis depicts PE-A (phycoerythrin signal region). The correspondence between DLL3 / CD3 binding proteins and epitope domains is as follows:

[0074]

[0075] Figure 3B: Localization of DLL3 epitope domains of six exemplary DLL3 / CD3 binding proteins. Exemplary DLL3 / CD3 binding proteins recognize EGF1, EGF3, EGF4 or EGF6 of DLL3. The y-axis depicts counts and the x-axis depicts PE-A (phycoerythrin signal). The correspondence between DLL3 / CD3 binding proteins and epitope domains is as follows:

[0076] DLL3 / CD3 binding protein Epitope domain DLL3#7 / CD3#1 EGF6 DLL3#8 / CD3#1 EGF4 DLL3#9 / CD3#1 EGF3 DLL3#10 / CD3#1 EGF1 DLL3#11 / CD3#1 EGF1 DLL3#12 / CD3#1 EGF1

[0077] Figure 3C : Localization of DLL3 epitope domains of six exemplary DLL3 / CD3 binding proteins. Exemplary DLL3 / CD3 binding proteins recognize the DSL domain, or recognize neither DSL, EGF, nor the membrane-proximal peptide of DLL3. The y-axis depicts counts, and the x-axis depicts PE-A (phycoerythrin signal). The correspondence between DLL3 / CD3 binding proteins and epitope domains is as follows:

[0078] DLL3 / CD3 binding protein Epitope domain DLL3#6 / CD3#1 DSL DLL3#13 / CD3#1 DSL DLL3#14 / CD3#1 DSL DLL3#15 / CD3#1 DSL DLL3#16 / CD3#1 Non-DSL-non-EGF1-6-non-membrane proximal peptide DLL3#17 / CD3#1 Non-DSL-non-EGF1-6-non-membrane proximal peptide DLL3#18 / CD3#1 Non-DSL-non-EGF1-6-non-membrane proximal peptide

[0079] Figure 4 : Binding of seven exemplary DLL3 / CD3 binding proteins to cell lines expressing human and cynomolgus monkey DLL3. The y-axis depicts counts and the x-axis depicts PE-A (phycoerythrin signal area).

[0080] Figure 5 : Binding of seven exemplary DLL3 / CD3 binding proteins to cell lines expressing human DLL1 and DLL4, no binding to human DLL1 and DLL4 was detected. The y-axis depicts counts and the x-axis depicts PE-A (phycoerythrin signal area).

[0081] Fig. 6A : Binding of seven exemplary DLL3 / CD3 binding proteins to SCLC cell lines and human T cells analyzed by flow cytometry.

[0082] Figure 6B : Binding of exemplary DLL3 / CD3 binding proteins (to peptides other than DSL, EGF1-6, membrane-proximal peptides) to SHP77 cells.

[0083] Figure 6C : Binding of exemplary DLL3 / CD3 binding proteins (directed to the DSL domain) to SHP77 cells.

[0084] Fig.6D : Binding of exemplary DLL3 / CD3 binding proteins (directed to the EGF1 domain) to SHP77 cells.

[0085] Fig. 6E : Binding of exemplary DLL3 / CD3 binding proteins (directed to the EGF3 domain) to SHP77 cells.

[0086] Fig. 6F : Binding of exemplary DLL3 / CD3 binding proteins (directed to the EGF4 domain) to SHP77 cells.

[0087] Figure 6G : Binding of exemplary DLL3 / CD3 binding proteins (directed to the DSL domain) to SHP77 cells.

[0088] Figure 6H : Binding of exemplary DLL3 / CD3 binding proteins (directed to the promembrane peptide domain) to SHP77 cells.

[0089] Fig.6I : Binding of exemplary DLL3 / CD3 binding proteins (to peptides other than DSL, EGF1-6, membrane-proximal peptides) to NCI-H82 cells.

[0090] Figure 6J : Binding of exemplary DLL3 / CD3 binding proteins (directed to the DSL domain) to NCI-H82 cells.

[0091] Figure 6K : Binding of exemplary DLL3 / CD3 binding proteins (directed to the EGF1 domain) to NCI-H82 cells.

[0092] Figure 6L : Binding of exemplary DLL3 / CD3 binding proteins (directed to the EGF3 domain) to NCI-H82 cells.

[0093] Figure 6M : Binding of exemplary DLL3 / CD3 binding proteins (directed to the EGF4 domain) to NCI-H82 cells.

[0094] Figure 6N : Binding of exemplary DLL3 / CD3 binding proteins (directed to the DSL domain) to NCI-H82 cells.

[0095] Fig.6O : Binding of exemplary DLL3 / CD3 binding proteins (directed against the membrane-proximal peptide domain) to NCI-H82 cells.

[0096] Figure 6P : Binding of exemplary DLL3 / CD3 binding proteins (to peptides other than DSL, EGF1-6, membrane-proximal peptides) to T cells.

[0097] Figure 6Q : Binding of exemplary DLL3 / CD3 binding proteins (directed to the DSL domain) to T cells.

[0098] Figure 6R: Binding of exemplary DLL3 / CD3 binding proteins (directed to the EGF1 domain) to T cells.

[0099] Figure 6S : Binding of exemplary DLL3 / CD3 binding proteins (directed to the EGF3 domain) to T cells.

[0100] Figure 6T : Binding of exemplary DLL3 / CD3 binding proteins (directed to the EGF4 domain) to T cells.

[0101] Figure 6U : Binding of exemplary DLL3 / CD3 binding proteins (directed to the EGF6 domain) to T cells.

[0102] Figure 6V : Binding of exemplary DLL3 / CD3 binding proteins (directed to the membrane-proximal peptide domain) to T cells.

[0103] Fig. 7A : Efficacy of seven DLL3 / CD3 binding proteins in lysing non-stimulatory PBMCs redirected to human SCLC cell lines.

[0104] Figure 7B : Efficacy of exemplary DLL3 / CD3 binding proteins (directed to peptides other than DSL, EGF1-6, membrane-proximal peptides) in redirecting non-stimulatory T cells to human SHP77 cells in lysing cells.

[0105] Figure 7C : Efficacy of exemplary DLL3 / CD3 binding proteins (directed to the DSL domain) in lysing non-stimulatory T cells redirected to human SHP77 cells.

[0106] Fig.7D : Efficacy of exemplary DLL3 / CD3 binding proteins (directed to the EGF1 domain) in lysing non-stimulatory T cells redirected to human SHP77 cells.

[0107] Fig. 7E : Efficacy of exemplary DLL3 / CD3 binding proteins (directed to the EGF3 domain) in lysing non-stimulatory T cells redirected to human SHP77 cells.

[0108] Figure 7F : Efficacy of exemplary DLL3 / CD3 binding proteins (directed to the EGF4 domain) in lysing non-stimulatory T cells redirected to human SHP77 cells.

[0109] Figure 7G : Efficacy of exemplary DLL3 / CD3 binding proteins (directed against the EGF6 domain) in lysing non-stimulatory T cells redirected to human SHP77 cells.

[0110] Figure 7H : Efficacy of exemplary DLL3 / CD3 binding proteins (directed against membrane-proximal peptides) in lysing non-stimulatory T cells redirected to human SHP77 cells.

[0111] Fig.7I : Efficacy of exemplary DLL3 / CD3 binding proteins (directed to peptides other than DSL, EGF1-6, membrane-proximal peptides) in lysing non-stimulatory T cells redirected to human NCI-H82 cells.

[0112] Figure 7J : Efficacy of exemplary DLL3 / CD3 binding proteins (directed to the DSL domain) in lysing non-stimulatory T cells redirected to human NCI-H82 cells.

[0113] Figure 7K : Efficacy of exemplary DLL3 / CD3 binding proteins (directed to the EGF1 domain) in lysing non-stimulatory T cells redirected to human NCI-H82 cells.

[0114] Figure 7L : Efficacy of exemplary DLL3 / CD3 binding proteins (directed against the EGF3 domain) in lysing non-stimulatory T cells redirected to human NCI-H82 cells.

[0115] Figure 7M : Efficacy of exemplary DLL3 / CD3 binding proteins (directed to the EGF4 domain) in lysing non-stimulatory T cells redirected to human NCI-H82 cells.

[0116] Figure 7N : Efficacy of exemplary DLL3 / CD3 binding proteins (directed against the EGF6 domain) in lysing non-stimulatory T cells redirected to human NCI-H82 cells.

[0117] Fig.7O : Efficacy of exemplary DLL3 / CD3 binding proteins (directed against membrane-proximal peptides) in lysing non-stimulatory T cells redirected to human NCI-H82 cells.

[0118] Figure 8 : In vitro internalization using two exemplary DLL3 / CD3 binding proteins. DLL3 / CD3 binding proteins were pre-incubated with DLL3-positive SHP77 cells for 2 and 4 hours, followed by co-incubation with human PBMCs for 48 hours.

[0119] Fig. 9: Pharmacokinetics of two exemplary DLL3 / CD3 binding proteins. Mean pharmacokinetic profiles of DLL3#3 / CD3#1 (open squares) and DLL3#3 / CD3#2 (closed squares) in male C57BL / 6 mice after a single 1 mg / kg intravenous administration.

[0120] Fig.10 : Anti-tumor activity of two exemplary DLL3 / CD3 binding proteins. The y-axis depicts median tumor volume and the x-axis depicts time.

[0121] Fig.11 : Efficacy of exemplary DLL3 / CD3 binding proteins in lysing non-stimulatory PBMCs redirected to human SHP77 and RKO-E6 cells.

[0122] Fig.12 : Efficacy of exemplary DLL3 / CD3 binding proteins on T cell activation in the presence of SHP77 cells

[0123] Fig.13 : Efficacy of exemplary DLL3 / CD3 binding proteins on T cell degranulation in the presence of SHP77 cells

[0124] Fig.14A : Efficacy of exemplary DLL3 / CD3 binding proteins on interferon secretion by PBMCs in the presence of SHP77 cells

[0125] Fig. 14B : Efficacy of exemplary DLL3 / CD3 binding proteins on MCP-1 secretion by PBMCs in the presence of SHP77 cells

[0126] Fig.15 : Efficacy of exemplary DLL3 / CD3 binding proteins on T cell proliferation in the presence of SHP77 cells

[0127] Fig.16 : Efficacy of exemplary DLL3 / CD3 binding proteins in redirecting non-stimulatory pan-T cells and naive T cells to human SHP77 in lysing cells.

[0128] Fig.17 :Non-stimulatory CD4 + and CD4 + Efficacy of exemplary DLL3 / CD3 binding proteins in lytic cells for redirecting central memory T cells to human SHP77 cells

[0129] Fig.18 :Non-stimulatory CD4 + and CD4 + Efficacy of exemplary DLL3 / CD3 binding proteins for redirecting effector memory T cells to human SHP77 in lytic cells

[0130] Fig.19 :Non-stimulatory CD8 + and CD8 + CD45RA + Efficacy of exemplary DLL3 / CD3 binding proteins for redirecting effector memory T cells to human SHP77 in lytic cells

[0131] Fig. 20 :Non-stimulatory CD8 + and CD8 + Efficacy of exemplary DLL3 / CD3 binding proteins for memory T cell redirection in human SHP77 cells

[0132] Fig.21 : T cell infiltration in SHP77 xenograft tumor tissues with exemplary DLL3 / CD3 binding proteins.

[0133] Fig. 22 : Binding of three exemplary anti-DLL3 antibodies to recombinant DLL3 protein in an ELISA assay. The y-axis depicts counts and the x-axis depicts the concentration of DLL3 protein.

[0134] Fig.23 : Binding of exemplary anti-DLL3 antibodies to DLL3-positive SCLC cell lines as determined by flow cytometry. The y-axis depicts counts and the x-axis depicts the concentration of the anti-DLL3 antibody.

[0135] Fig.24 : Representative staining of SCLC tissue samples with 5 exemplary anti-DLL3 antibodies. Anti-DLL3 antibodies DLL3#1 and DLL3#5 show that tumor cells exhibit punctate and / or diffuse cytoplasmic and / or membrane DLL3 staining. Images have been electronically generated by a LEICAS CN400 automated scanner at 10× magnification.

[0136] Fig.25 :Representative staining of cell aggregates from SCLC cell lines with different DLL3 expression levels. Anti-DLL3 antibody DLL3#5 DETAILED DESCRIPTION OF THE INVENTION

[0138] Terms and definitions used

[0139] The above and other aspects and embodiments of the present invention will become clear from the further description herein, in which:

[0140] Unless otherwise specified or defined, all terms used have their common meanings in the art, which will be clear to those skilled in the art. For example, reference is made to standard manuals, such as Sambrook et al., "Molecular Cloning: A Laboratory Manual" (2nd edition), Vol. 1-3, Cold Spring Harbor Laboratory Press (1989); Lewin, "Genes IV", Oxford University Press, New York, (1990) and Roitt et al., "Immunology" (2nd edition), Gower Medical Publishing, London, New York (1989), and the general background technology cited herein. In addition, as will be clear to those skilled in the art, unless otherwise specified, all methods, steps, techniques and operations not specifically described in detail can be performed in a manner known per se and have been performed in a manner known per se. In addition, reference is made to standard manuals, the general prior art mentioned above and other references cited therein for the embodiments.

[0141] When used herein, the term "comprising" and variations thereof, such as "comprises" and "comprise" may be replaced by the terms "containing" or "including" or "having".

[0142] Unless the context requires a more restricted interpretation, the term "sequence" as used herein (e.g., such as the following terms, "heavy chain / light chain sequence", "antibody sequence", "variable domain sequence", "constant domain sequence" or "protein sequence") should generally be understood to include the relevant amino acid sequence as well as the nucleic acid sequence or nucleotide sequence encoding the same.

[0143] As used herein, "antigen binding unit" refers to a polypeptide that is capable of binding to its specific target or antigen and comprises the minimum structural requirements derived from an antibody (usually present in an antibody) that allow target binding. Therefore, an antigen binding unit comprises at least three light chain and three heavy chain CDR sequences, preferably at least the presence of a light chain variable domain and a heavy chain variable domain.

[0144] The universal structure of antibodies or immunoglobulins is well known to those skilled in the art. These molecules are heterotetrameric glycoproteins, usually about 150,000 daltons, composed of two identical light (L) chains and two identical heavy (H) chains and are generally referred to as full-length antibodies. Each light chain is covalently linked to a heavy chain by a disulfide bond to form a heterodimer, and the heterotetrameric molecule is formed by covalent disulfide bonds between two identical heavy chains of the heterodimer. Although the light chain and the heavy chain are linked together by a disulfide bond, the number of disulfide bonds between the two heavy chains is different due to the immunoglobulin isotype. Each heavy chain and light chain also has regularly spaced intrachain disulfide bridges. Each heavy chain has a variable domain (VH) at the N-terminus, followed by three or four (for IgE) constant domains (CH1, CH2, CH3 and CH4), and a hinge region between CH1 and CH2. Each light chain has two domains: an N-terminal variable domain (VL) and a C-terminal constant domain (CL). The VL domain non-covalently associates with the VH domain, whereas the CL domain is usually covalently linked to the CH1 domain via a disulfide bond. Specific amino acid residues are known to form an interface between the light chain variable domain and the heavy chain variable domain (Chothia et al., 1985, J. Mol. Biol. 186: 651-663). The variable domain is also referred to herein as a variable region or Fv and represents the portion that confers specificity to an antibody for an antigen by carrying an antigen binding site.

[0145] As used herein, "light chain variable domains" (or "light chain variable regions") and "heavy chain variable domains" (or "heavy chain variable regions") have the same general structure and each domain is essentially composed of four framework (FR) regions, the sequences of which are widely conserved and are referred to in the art and hereinafter as "framework region 1" or "FR1"; "framework region 2" or "FR2"; "framework region 3" or "FR3"; and "framework region 4" or "FR4", respectively; and the framework regions are interspersed with three hypervariable regions HVR (or CDR), which are referred to in the art and hereinafter as "complementarity determining region 1" or "CDR1"; "complementarity determining region 2" or "CDR2"; and "complementarity determining region 3" or "CDR3", respectively. Thus, the general structure or sequence of an immunoglobulin variable domain can be represented as follows: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4. The framework regions adopt a β-sheet configuration and the CDRs can form loops connecting the β-sheet structures. The CDRs in each chain are held in their three-dimensional structure by the framework regions and together with the CDRs from the other chains form the antigen binding site.

[0146] Within the context of the present invention, reference to the CDRs is to the CCG-based, also known as IMGT, definition (Lefranc MP, Pommie C, Ruiz M, Giudicelli V, Foulquier E, Truong L, Thouvenin-Contet V, Lefranc G. "IMGT unique numbering for immunoglobulin and T cell receptor variable domains and Ig superfamily V-like domains." Dev Comp Immunol. 2003 Jan; 27(1): 55-77; Giudicelli V, Brochet X, Lefranc MP. "IMGT / V-QUEST: IMGT standardized analysis of the immunoglobulin (IG) and T cell receptor (TR) nucleotide sequences". Cold Spring Harb Protoc. 2011; 2011(6): 695-715). Alternative definitions of CDRs known in the art are based on Chothia (Chothia and Lesk, J. Mol. Biol. 1987, 196:901-917) and Kabat (EA Kabat, TT Wu, H. Bilofsky, M. Reid-Miller and H. Perry, Sequence of Proteins of Immunological Interest, National Institutes of Health, Bethesda (1983)).

[0147] As used in this application, the term "constant domain" or "constant region" refers to the sum of the domains of an antibody excluding the variable region. Such constant domains and constant regions are well known in the art and are described, for example, by Kabat et al. ("Sequence of proteins of immunological interest", US Public Health Services, NIH Bethesda, MD, Publication No. 91).

[0148] The "Fc portion" or "Fc domain" of an antibody is not directly involved in the binding of the antibody to the antigen, but exhibits various effector functions. "Fc portion / domain of an antibody" is a term well known to those skilled in the art and is defined on the papain-cleaved matrix of the antibody. Depending on the amino acid sequence of the constant region of its heavy chain, antibodies or immunoglobulins are divided into the following categories: IgA, IgD, IgE, IgG and IgM. According to the heavy chain constant region, different classes of immunoglobulins are called α, δ, ε, γ and μ, respectively. Some of these can be further divided into subclasses (isotypes), such as IgG1, IgG2, IgG3 and IgG4, IgA1 and IgA2. The Fc portion of an antibody is directly involved in ADCC (antibody-dependent cell-mediated cytotoxicity) and CDC (complement-dependent cytotoxicity), based on complement activation, C1q binding and Fc substrate binding. Complement activation (CDC) is initiated by the binding of complement factor C1q to the Fc portion of most IgG antibody subclasses. Although the influence of antibodies on the complement system depends on certain conditions, C1q binding is caused by defined binding sites within the Fc portion. Such binding sites are, for example, L234, L235, D270, N297, E318, K320, K322, P331 and P329 (numbered according to EU numbering (Edelman et al., Proc Natl Acad Sci USA. 1969 May; 63(1):78-85). The most critical of these residues in mediating C1q and Fcγ substrate binding in IgG1 are L234 and L235 (Hezareh et al., J. Virology 75 (2001) 12161-12168, Shields et al., (2001) JBC, 276(9):6591-6604). Antibodies of subclasses IgG1 and IgG3 typically display complement activation and Clq and C3 binding, whereas IgG2 and IgG4 do not activate the complement system and do not bind Clq and C3.

[0149] The term "antibody" or "antibody molecule" (used synonymously herein) encompasses monoclonal antibodies, polyclonal antibodies, human antibodies, humanized antibodies, chimeric antibodies, multispecific antibodies (e.g., bispecific antibodies), antibody fragments (especially Fv, Fab, Fab' or F(ab')2 fragments), single-chain antibodies (especially single-chain variable fragments (scFv), single-chain Fab fragments (scFab)), small modular immunopharmaceuticals (SMIPs), domain antibodies, nanobodies, bifunctional antibodies. The antibody may have effector functions, such as ADCC or CDC, which are usually mediated by the Fc portion (antibody constant region) of the antibody, or it may not have effector functions, for example by lacking an Fc portion or having a blocking, shielded Fc portion, essentially an Fc portion that is not recognized or not fully recognized by immune cells or immune system components (such as the complement system).

[0150] Monoclonal antibodies (mAbs) are monospecific antibodies with identical amino acid sequences. They can be prepared by hybridoma technology from hybrid cell lines (called hybridomas), which represent pure lines of fusion of B cells and myeloma (B cell cancer) cells that produce specific antibodies (Kohler G, Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 1975; 256: 495-7). Alternatively, monoclonal antibodies can be prepared by recombinant expression in host cells (Norderhaug L, Olafsen T, Michaelsen TE, Sandlie I. (May 1997). "Versatile vectors for transient and stable expression of recombinant antibody molecules in mammalian cells." J Immunol Methods 204 (1): 77-87; see also below). "Recombinant antibodies" or "recombinant binding proteins" are antibodies or binding proteins produced by recombinantly engineered host cells. They are optionally separated or purified.

[0151] Full-length antibodies can be treated with enzymes such as papain or pepsin to produce useful antibody fragments. Papain digestion is used to produce two identical antigen-binding antibody fragments called "Fab" fragments, each with a single antigen-binding site, and a residual "Fc" fragment. The Fab fragment also contains the constant domain of the light chain and the CH1 domain of the heavy chain. Pepsin treatment produces a F(ab')2 fragment, which has two antigen-binding sites and is still able to cross-link antigens.

[0152] Fab' fragments differ from Fab fragments by the presence of additional residues at the C-terminus of the CH1 domain, including one or more cysteines from the antibody hinge region. F(ab')2 antibody fragments are pairs of Fab' fragments linked by cysteine ​​residues in the hinge region. Other chemical couplings of antibody fragments are also known.

[0153] The "Fv" fragment contains a complete antigen recognition and binding site consisting of a dimer of a heavy chain variable domain and a light chain variable domain in tight non-covalent association. In this configuration, the three CDRs of each variable domain interact to define an antigen binding site on the surface of the VH-VL dimer. The six CDRs collectively confer antigen binding specificity to the antibody.

[0154] "Single-chain Fv" or "scFv" antibody fragments are single-chain Fv variants comprising the VH and VL domains of an antibody, wherein the domains are present in a single polypeptide chain. Single-chain Fvs are capable of recognizing and binding antigens. ScFv polypeptides may also optionally contain a polypeptide linker between the VH and VL domains to facilitate the formation of the three-dimensional structure required for antigen binding by scFv (see, e.g., Pluckthun, 1994, The Pharmacology of monoclonal Antibodies, Vol. 113, Rosenburg and Moore, eds., Springer-Verlag, New York, pp. 269-315).

[0155] "Single-chain Fab" or "scFab" antibody fragments are single-chain Fab variants comprising the VL, CL, VH and CH1 domains of an antibody, wherein the domains are present in a single polypeptide chain. Single-chain Fab is capable of recognizing and binding antigen. scFab polypeptides may also optionally contain a polypeptide linker between the CL and VH domains (Hust et al., (2007) BMC Biotechnology).

[0156] For use in humans, it is often necessary to reduce the immunogenicity of therapeutic molecules (such as antibodies or binding proteins comprising antigen binding units as described herein) that were originally derived from other species, such as mice. This can be accomplished by constructing chimeric antibodies / binding proteins, or by a method known as "humanization". In this context, "chimeric antibodies"; or "chimeric antigen binding units" should be understood as antibodies or antigen binding units that contain sequence portions (e.g., variable domains) derived from one species (e.g., mice) fused to sequence portions (e.g., constant domains) derived from different species (e.g., humans). In this context, "humanized antibodies", "humanized binding proteins" or "humanized antigen binding units" are antibodies, proteins or antigen binding units that contain variable domains originally derived from non-human species, in which certain amino acids have been mutated to prepare variable domains that are more similar to the overall sequence of human variable domains. Methods for antibody humanization are well known in the art (Billetta R, Lobuglio AF. "Chimeric antibodies". Int Rev Immunol. 1993; 10(2-3): 165-76; Riechmann L, Clark M, Waldmann H, Winter G (1988). "Reshaping human antibodies for therapy". Nature: 332: 323).

[0157] "Optimized antibodies" or "optimized antigen binding units or proteins" are specific types of humanized antibodies or humanized antigen binding units / proteins, which include immunoglobulin amino acid sequence variants or fragments thereof, which are capable of binding to a predetermined antigen and include one or more FRs having substantially the amino acid sequence of a human immunoglobulin and one or more CDRs having substantially the amino acid sequence of a non-human immunoglobulin. This non-human amino acid sequence, often referred to as an "imported" sequence, is usually obtained from an "imported" antibody domain, especially a variable domain. In general, an optimized antibody includes at least a non-human antibody or a CDR (or HVL) derived from a non-human antibody inserted between the FRs of a human heavy or light chain variable domain. It should be understood that certain mouse FR residues are important for the function of the optimized antibody and therefore certain human reproductive sequence heavy and light chain variable domain residues are modified to be identical to those of the corresponding mouse sequence. During this process, undesired amino acids may also be removed or changed, for example, to avoid deamidation, undesired charge or lipophilicity or non-specific binding. "Optimized antibodies," "optimized antibody fragments," or "optimization" may sometimes be referred to as "humanized antibodies," "humanized antibody fragments," or "humanization," or "sequence optimized."

[0158] In addition, technologies have been developed for producing antibodies or VH / VL domains based on sequences derived from the human genome, for example by phage display or using transgenic animals (WWW.Ablexis.com / technology-alivamab.php; WO90 / 05144; D. Marks, HR Hoogenboom, TP Bonnert, J. McCafferty, AD Griffiths and G. Winter (1991) "By-passing immunisation. Human antibodies from V-gene libraries displayed on phage." J. Mol. Biol., 222, 581-597; Knappik et al., J. Mol. Biol. 296:57-86, 2000; S. Carmen and L. Jermutus, "Concepts in antibody phage display". Briefings in Functional Genomics and Proteomics 2002 1(2):189-203; Lonberg et al., J. Mol. Biol. 297:57-86, 2000; S. Carmen and L. Jermutus, "Concepts in antibody phage display". Briefings in Functional Genomics and Proteomics 2002 1(2):189-203; Lonberg et al., J. Mol. Biol. 298:57-86, 2000; S. Carmen and L. Jermutus, "Concepts in antibody phage display". Briefings in Functional Genomics and Proteomics 2002 1(2):189-203; Lonberg et al., J. Mol. Biol. 297:57-86, 2000; S. Carmen and L. Jermutus, "Concepts in antibody phage display". Briefings in Functional Genomics and Proteomics 2002 1(2):189-203; Lonberg et al., J. Mol. N, Huszar D. "Human antibodies from transgenic mice". Int Rev Immunol. 1995; 13(1): 65-93.; Brüggemann M, Taussig MJ. "Production of human antibody repertoires in transgenic mice". Curr Opin Biotechnol. 1997 August; 8(4): 455-8.). In the context of the present invention, such antibodies or antigen binding units or VH / VL domains are "human antibodies", "human antigen binding units" or "human VH / VL domains".

[0159] As used herein, the terms "human antibody", "human antigen binding unit" or "human VH / VL domain" are intended to include antibodies, antigen binding units or VH / VL domains having variable (and constant, if applicable) regions derived from human germline immunoglobulin sequences. The human antibodies, antigen binding units, proteins or VH / VL domains of the present technology may include amino acid residues that are not encoded by human germline immunoglobulin sequences (e.g., mutations induced by random or site-specific mutagenesis in vitro or introduced by somatic mutation in vivo). However, as used herein, the terms "human antibody", "human antigen binding unit" or "human VH / VL domain" are not intended to include antibodies in which CDR sequences derived from the germline of another (mammalian species) (such as mouse, rat or rabbit) have been grafted onto human framework sequences. Therefore, as used herein, the term "human antibody", "human antigen binding unit" or "human VH / VL domain" refers to the following antibodies, antigen binding units or VH / VL domains: wherein substantially every part of the protein (e.g., CDR, framework, CL, CH domain (e.g., CH1, CH2, CH3), hinge, VL, VH) is substantially non-immunogenic in humans, with only minor sequence changes or variations. Such changes or variations are optional and preferably retain or reduce immunogenicity in humans or other species relative to unmodified antibodies or antigen binding units.

[0160] Therefore, human antibodies, human antigen binding units or human VH / VL domains are different from, for example, chimeric or humanized antibodies. It should be noted that human antibodies, human antigen binding units or human VH / VL domains can be prepared by non-human animals or prokaryotic or eukaryotic cells that are capable of expressing functionally rearranged human immunoglobulin (e.g., heavy chain and / or light chain) genes.

[0161] The term "monomer" refers to a homogeneous form of an antibody or multispecific protein as described herein. For example, for a full-length antibody, a monomer means a monomeric antibody having two identical heavy chains and two identical light chains. In the context of the present invention, a monomer means a protein of the present invention having a single antigen binding unit with specificity for DLL3 as described herein, and a single antigen binding unit with specificity for CD3. For example, a monomer of a binding protein described herein may have two chains: a first chain comprising a single chain Fab with a first antigen binding unit and optionally a first Fc domain; and a second chain comprising a single chain Fab with a second antigen binding unit and optionally a second Fc domain.

[0162] An epitope is a region of an antigen that binds to an antibody or antigen binding portion (e.g., an antigen binding unit of a protein described herein). The term "epitope" includes any polypeptide determinant that is capable of specific binding to an antibody or antigen binding portion. In certain embodiments, epitope determinants include chemically active surface groups of molecules (such as amino acids, polysaccharide side chains, phosphoryl or sulfonyl groups), and in certain embodiments, may have specific three-dimensional structural characteristics, and / or specific charge characteristics. The difference between a conformational epitope and a non-conformational epitope is that in the presence of a denaturing solvent, binding to the former is lost, but not to the latter.

[0163] Antigen binding molecules / proteins (such as immunoglobulins, antibodies, antigen binding units, or fragments of such antigen binding molecules / proteins) that can "bind", "bind to", "specifically bind to", or "specifically bind to" an epitope, antigen, or protein (or at least a portion, fragment, or epitope thereof), "have affinity", "is specific for / has specificity for" an epitope, antigen, or protein are said to be "to" or "directed to" that epitope, antigen, or protein or to "bind" molecules / proteins relative to such epitope, antigen, or protein.

[0164] As used herein, the terms "binding" and "specific binding" refer to the binding of an antibody or antigen-binding portion (such as an immunoglobulin, an antibody, an antigen-binding unit, or a fragment of such an antigen-binding molecule / protein) to an epitope of an antigen in an in vitro assay, preferably in a plasma resonance assay (Malmqvist M., "Surface plasmon resonance for detection and measurement of antibody-antigen affinity and kinetics." Curr Opin Immunol. 1993 April; 5 (2): 282-6.) using purified wild-type antigen. Antibody affinity can also be measured using kinetic exclusion assay (KinExA) technology (Darling, RJ, and Brault PA., "Kinetic exclusion assay technology: Characterization of Molecular Interactions." ASSAY and Drug Development Technologies. 2004 December 2 (6): 647-657).

[0165] In general, the term "specificity" refers to the number of different types of antigens or epitopes to which a particular antigen binding molecule / protein (such as an immunoglobulin, an antibody, an antigen binding unit, or a fragment of such an antigen binding molecule / protein) can bind. The specificity of an antigen binding molecule / protein can be determined based on its affinity and / or avidity. The dissociation equilibrium constant (K) of the antigen and the antigen binding protein is D The affinity expressed as K is a measure of the binding strength between an epitope and the antigen binding site on an antigen binding molecule / protein: D The smaller the value, the stronger the binding strength between the epitope and the antigen-binding molecule (alternatively, affinity can also be expressed as an affinity constant (K A ), which is 1 / K D ). As will be clear to the skilled person (e.g. based on the other disclosures herein), affinity can be determined in a manner known per se, depending on the relevant specific antigen. Avidity is a measure of the strength of binding between an antigen binding molecule / protein (such as an immunoglobulin, an antibody, an antigen binding unit, or a fragment of such an antigen binding molecule / protein) and the relevant antigen. Avidity is related to both the affinity between an epitope and its antigen binding site on the antigen binding molecule / protein and the number of relevant binding sites present on the antigen binding molecule / protein.

[0166] As used herein, the term "isolated" refers to a substance being removed from its original or native environment (e.g., natural environment if it occurs naturally). For example, a naturally occurring polynucleotide or polypeptide present in a living animal is not isolated, but the same polynucleotide or polypeptide separated from some or all coexisting materials in the natural system by human intervention is isolated. The polynucleotide may be part of a vector and / or the polynucleotide or polypeptide may be part of a composition and still be isolated such that such vector or composition is not part of the environment in which it is found in nature. For example, a nucleic acid, protein / polypeptide molecule is considered "in substantially isolated (form)" when the nucleic acid, protein / polypeptide molecule has been separated from at least one other component, such as another nucleic acid, another protein / polypeptide, another biological component or macromolecule, or at least one adulterant, impurity, or trace component, when compared to the natural biological source and / or reaction medium or culture medium from which the nucleic acid, protein / polypeptide molecule was obtained, wherein the nucleic acid, protein / polypeptide molecule is normally associated with the at least one other component in the source or medium. In particular, a nucleic acid or protein / polypeptide molecule is considered to be "substantially isolated" when it is purified at least 2-fold, particularly at least 10-fold, more particularly at least 100-fold and up to 1000-fold or more than 1000-fold. Nucleic acids or protein / polypeptide molecules "in substantially isolated form" are preferably substantially homogeneous as determined using a suitable technique, such as a suitable chromatographic technique, such as polyacrylamide-gel electrophoresis.

[0167] As used herein, the term "identical" or "percentage of identity" in the context of two or more nucleic acid or polypeptide sequences refers to two or more sequences or subsequences that are identical or have a specified percentage of nucleotides or amino acid residues that are identical when compared and aligned for maximum correspondence. To determine the percentage of identity, sequence alignment is performed for the purpose of optimal comparison (e.g., for optimal alignment with a second amino acid or nucleic acid sequence, a gap may be introduced into the sequence of the first amino acid or nucleic acid sequence). The amino acid residues or nucleotides at corresponding amino acid positions or nucleotide positions are then compared. If the position in the first sequence is occupied by the same amino acid residue or nucleotide as the corresponding position in the second sequence, the molecule is consistent at that position. The percentage of identity between the two sequences is a function of the number of identical positions shared by the sequences (i.e., % identity = number of identical positions / total number of positions (i.e., overlapping positions) × 100). In some embodiments, when appropriate, after introducing a gap within the sequence, the two sequences being compared are of the same length (e.g., not including additional sequences extending beyond the comparison sequence). For example, when comparing variable region sequences, leader sequences and / or constant domain sequences are not considered. For sequence comparisons between two sequences, a "corresponding" CDR refers to a CDR at the same position in both sequences (eg, CDR-H1 in each sequence).

[0168] Determining the percent identity or percent similarity between two sequences can be accomplished using a mathematical algorithm. A preferred, non-limiting example of a mathematical algorithm for comparing two sequences is the algorithm of Karlin and Altschul, 1990, Proc. Natl. Acad. Sci. USA 87:2264-2268, as modified in Karlin and Altschul, 1993, Proc. Natl. Acad. Sci. USA 90:5873-5877. Such algorithms are incorporated into the NBLAST and XBLAST programs of Altschul et al., 1990, J. Mol. Biol. 215:403-410. BLAST nucleotide searches can be performed with the NBLAST program (score = 100, wordlength = 12) to obtain nucleotide sequences homologous to nucleic acids encoding a protein of interest. BLAST protein searches can be performed with the XBLAST program (score = 50, wordlength = 3) to obtain amino acid sequences homologous to a protein of interest. In order to achieve the purpose of comparison by gap comparison, gap BLAST can be used as described in Altschul et al., 1997, Nucleic Acids Res. 25: 3389-3402. Alternatively, PSI-Blast can be used to perform iterative searches, which detect long-distance relationships between molecules (ibid.). When using BLAST, gap BLAST and PSI-Blast programs, the preset parameters of the corresponding programs (e.g., XBLAST and NBLAST) can be used. Another preferred non-limiting embodiment of the mathematical algorithm for comparing sequences is the algorithm of Myers and Miller, CABIOS (1989). Such algorithms are incorporated into the ALIGN program (version 2.0), which is a part of the GCG sequence alignment software package. When using the ALIGN program to compare amino acid sequences, PAM120 weight residue tables, gap length penalties of 12 and gap penalties of 4 can be used. Additional algorithms for sequence analysis are known in the art and include ADVANCE and ADAM as described in Torellis and Robotti, 1994, Comput. Appl. Biosci. 10:3-5; and FASTA as described in Pearson and Lipman, 1988, Proc. Natl. Acad. Sci. USA 85:2444-8. Within FASTA, ktup is a control option that sets the sensitivity and speed of the search. If ktup=2, similar regions in the two sequences being compared are obtained by observing pairs of aligned residues; if ktup=1, single aligned amino acids are examined. For protein sequences, ktup can be set to 2 or 1, or to 6 for DNA sequences. If ktup is not specified, the default value is 2 for proteins and 6 for DNA.Alternatively, protein sequence alignment can be performed using the CLUSTAL W algorithm as described by Higgins et al., 1996, Methods Enzymol. 266:383-402.

[0169] As used herein, the term "covalently linked" means a direct covalent bond between residues, or an indirect association wherein two residues are not directly bonded but are both covalently bonded to an intermediate molecule or domain, such as an intermediate domain of an immunoglobulin.

[0170] The terms "compete" or "cross-compete" are used interchangeably herein and refer to the ability of an antibody molecule to interfere with the binding of an antibody molecule (e.g., an anti-DLL3 antibody molecule of the invention) to a target (e.g., human DLL3). Interference with binding may be direct or indirect (e.g., via allosteric modulation of the antibody molecule or target). The extent to which an antibody molecule is able to interfere with the binding of another antibody molecule to a target and therefore whether it can be said to compete can be determined using a competition binding assay (e.g., a FACS assay, an ELISA, or a BIACORE assay). In some embodiments, the competition binding assay is a quantitative competition assay. In some embodiments, a first anti-DLL3 antibody molecule is said to compete with a second anti-DLL3 antibody molecule for binding to a target when binding of the first antibody molecule to the target is reduced by 10% or more, e.g., 20% or more, 30% or more, 40% or more, 50% or more, 55% or more, 60% or more, 65% or more, 70% or more, 75% or more, 80% or more, 85% or more, 90% or more, 95% or more, 98% or more, 99% or more in a competition binding assay (e.g., a competition assay described herein).

[0171] In the absence of explicit recitation and unless otherwise intended, it is inferred that when the present technology relates to polypeptides, proteins, polynucleotides, or antibodies, equivalents or bioequivalents of such are intended to be within the scope of the present technology.

[0172] As used herein, the term "biological equivalent thereof" is intended to be synonymous with "equivalent thereof" when referring to a reference protein, antibody, polypeptide, polynucleotide, or nucleic acid, and means those having minimal homology while still maintaining the desired structure or functionality. Unless specifically stated herein, it is contemplated that any nucleic acid, polynucleotide, polypeptide, protein, or antibody referred to herein also includes equivalents thereof. For example, an equivalent means at least about 80% homology or identity and alternatively, at least about 85%, or alternatively at least about 90%, or alternatively at least about 95%, or alternatively 98% percent homology or identity and exhibits a biological activity substantially equivalent to that of a reference protein, polypeptide, antibody, or nucleic acid.

[0173] As used herein, the term "detectable label" refers to a molecule, compound or composition that can produce a detectable signal (i.e., a physical or chemical signal, including colorimetric, fluorescent, electrical, radioactive and chemiluminescent signals) that can be measured visually or instrumentally and which indicates the presence and / or amount / concentration of the label in a sample.

[0174] A "detectable signal" can be produced by a variety of mechanisms, including absorption, emission and / or scattering of photons (including radio frequency, microwave frequency, infrared frequency, visible frequency and ultraviolet frequency photons) and includes, but is not limited to, colorimetric, fluorescent, electrical, radioactive and chemiluminescent signals.

[0175] As used herein, "expression" refers to the process of transcribing a polynucleotide into mRNA and / or the process of transcribing mRNA and then translating it into a peptide, polypeptide or protein. If the polynucleotide is derived from genomic DNA, expression may include splicing of mRNA in eukaryotic cells. The amount of gene expression can be determined by measuring the amount of mRNA or protein in a cell or tissue sample.

[0176] As used herein, the term "biological sample" means a sample material derived from or in contact with living cells. The term is intended to include tissues, cells, and biological fluids isolated from an individual. As used herein, the term "tissue sample" shall refer to a cell sample that retains the cross-sectional spatial relationship between cells when leaving the individual from which the sample was obtained. Biological samples can also be obtained from sections of internal organs or cancers.

[0177] "Histochemistry" and "cytochemistry" are techniques used to identify molecules within the environment of intact cells by labeling the specimen with reagents that bind specifically to the molecules in a way that can be viewed under a microscope. "Immunohistochemistry" and "immunocytochemistry" are types of histochemistry and cytochemistry that use antibodies to label molecules.

[0178] Multispecific binding proteins of the invention

[0179] The present invention provides a multispecific binding protein comprising at least one antigen binding unit that specifically binds to DLL3 (first antigen binding unit) and at least one antigen binding unit that specifically binds to CD3 (second antigen binding unit). Such (multispecific) binding proteins are also referred to herein as (multispecific) binding molecules or DLL3 / CD3 binding proteins or DLL3 / CD3 binding molecules.

[0180] The inventors have surprisingly found that the multispecific binding proteins of the invention induce selective lysis of DLL3-positive SCLC cell lines in the presence of T cells and are already active at low effector:target cell ratios. Importantly, the binding proteins of the invention do not cause T cell activation, T cell proliferation and cytokine secretion, or lysis of DLL3-negative cells in the absence of DLL3-positive cells.

[0181] For the avoidance of doubt, DLL3 as used herein refers to human DLL3 of UniProt Q9NYJ7 and the nucleic acid sequence encoding the protein. CD3 as used herein refers to the human CD3ε (UniProt P07766) and CD3γ (Uniprot: P09693) complex (human CD3εγ complex).

[0182] Targeting DLL3 with a bispecific T cell engagement approach is expected to provide advantages over an ADC approach because redirected T cells are not affected by resistance to chemotherapy and low expression levels on the cell surface are less critical for this mode of action. T cell engagers are multispecific binding proteins that have a binding arm that binds to CD3 on T cells and a binding arm that binds to an antigen on the cell surface of a tumor cell. Through simultaneous binding to T cells and tumor cells, T cell engagers force the formation of a cytolytic synapse between the two cells and thereby selectively redirect T cell activity toward targeted tumor cells.

[0183] In one aspect, the multispecific binding protein of the invention comprises a first antigen binding unit that specifically binds to DLL3 and a second antigen binding unit that specifically binds to CD3, wherein the first binding unit is selected from the group consisting of i) to xviii):

[0184] i) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequences of SEQ ID NO: 1 (CDR1), SEQ ID NO: 2 (CDR2) and SEQ ID NO: 3 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 4 (CDR1), SEQ ID NO: 5 (CDR2) and SEQ ID NO: 6 (CDR3) (antigen binding unit DLL3#1);

[0185] ii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 7 (CDR1), SEQ ID NO: 8 (CDR2) and SEQ ID NO: 9 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 10 (CDR1), SEQ ID NO: 11 (CDR2) and SEQ ID NO: 12 (CDR3) (antigen binding unit DLL3#2);

[0186] iii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 13 (CDR1), SEQ ID NO: 14 (CDR2) and SEQ ID NO: 15 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 16 (CDR1), SEQ ID NO: 17 (CDR2) and SEQ ID NO: 18 (CDR3) (antigen binding unit DLL3#3);

[0187] iv) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 19 (CDR1), SEQ ID NO: 20 (CDR2) and SEQ ID NO: 21 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 22 (CDR1), SEQ ID NO: 23 (CDR2) and SEQ ID NO: 24 (CDR3) (antigen binding unit DLL3#4);

[0188] v) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 25 (CDR1), SEQ ID NO: 26 (CDR2) and SEQ ID NO: 27 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 28 (CDR1), SEQ ID NO: 29 (CDR2) and SEQ ID NO: 30 (CDR3) (antigen binding unit DLL3#5);

[0189] vi) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 31 (CDR1), SEQ ID NO: 32 (CDR2) and SEQ ID NO: 33 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 34 (CDR1), SEQ ID NO: 35 (CDR2) and SEQ ID NO: 36 (CDR3) (antigen binding unit DLL3#6);

[0190] vii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 133 (CDR1), SEQ ID NO: 134 (CDR2) and SEQ ID NO: 135 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 136 (CDR1), SEQ ID NO: 137 (CDR2) and SEQ ID NO: 138 (CDR3) (antigen binding unit DLL3#7);

[0191] viii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 139 (CDR1), SEQ ID NO: 140 (CDR2) and SEQ ID NO: 141 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 142 (CDR1), SEQ ID NO: 143 (CDR2) and SEQ ID NO: 144 (CDR3) (antigen binding unit DLL3#8);

[0192] ix) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 145 (CDR1), SEQ ID NO: 146 (CDR2) and SEQ ID NO: 147 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 148 (CDR1), SEQ ID NO: 149 (CDR2) and SEQ ID NO: 150 (CDR3) (antigen binding unit DLL3#9);

[0193] x) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 151 (CDR1), SEQ ID NO: 152 (CDR2) and SEQ ID NO: 153 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 154 (CDR1), SEQ ID NO: 155 (CDR2) and SEQ ID NO: 156 (CDR3) (antigen binding unit DLL3#10);

[0194] xi) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 157 (CDR1), SEQ ID NO: 158 (CDR2) and SEQ ID NO: 159 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 160 (CDR1), SEQ ID NO: 161 (CDR2) and SEQ ID NO: 162 (CDR3) (antigen binding unit DLL3#11);

[0195] xii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 163 (CDR1), SEQ ID NO: 164 (CDR2) and SEQ ID NO: 165 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 166 (CDR1), SEQ ID NO: 167 (CDR2) and SEQ ID NO: 168 (CDR3) (antigen binding unit DLL3#12);

[0196] xiii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 169 (CDR1), SEQ ID NO: 170 (CDR2) and SEQ ID NO: 171 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 172 (CDR1), SEQ ID NO: 173 (CDR2) and SEQ ID NO: 174 (CDR3) (antigen binding unit DLL3#13);

[0197] xiv) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 175 (CDR1), SEQ ID NO: 176 (CDR2) and SEQ ID NO: 177 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 178 (CDR1), SEQ ID NO: 179 (CDR2) and SEQ ID NO: 180 (CDR3) (antigen binding unit DLL3#14);

[0198] xv) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 181 (CDR1), SEQ ID NO: 182 (CDR2) and SEQ ID NO: 183 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 184 (CDR1), SEQ ID NO: 185 (CDR2) and SEQ ID NO: 186 (CDR3) (antigen binding unit DLL3#15);

[0199] xvi) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 187 (CDR1), SEQ ID NO: 188 (CDR2) and SEQ ID NO: 189 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 190 (CDR1), SEQ ID NO: 191 (CDR2) and SEQ ID NO: 192 (CDR3) (antigen binding unit DLL3#16);

[0200] xvii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 193 (CDR1), SEQ ID NO: 194 (CDR2) and SEQ ID NO: 195 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 196 (CDR1), SEQ ID NO: 197 (CDR2) and SEQ ID NO: 198 (CDR3) (antigen binding unit DLL3#17); and

[0201] xviii) An antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 199 (CDR1), SEQ ID NO: 200 (CDR2) and SEQ ID NO: 201 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 202 (CDR1), SEQ ID NO: 203 (CDR2) and SEQ ID NO: 204 (CDR3) (antigen binding unit DLL3#18).

[0202] Preferably, the first antigen binding unit of the binding protein of the present invention is any one of i) to iii) as defined in the above CDR sequences.

[0203] For the avoidance of doubt, each of the specific embodiments listed herein may also be considered a separate aspect of the present invention.

[0204] In some embodiments of the binding protein of the present invention, the second antigen binding unit that specifically binds to CD3 is selected from the group consisting of i) to iii):

[0205] i) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 55 (CDR1), SEQ ID NO: 56 (CDR2) and SEQ ID NO: 57 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 58 (CDR1), SEQ ID NO: 59 (CDR2) and SEQ ID NO: 60 (CDR3) (antigen binding unit CD3#1);

[0206] ii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 61 (CDR1), SEQ ID NO: 62 (CDR2) and SEQ ID NO: 63 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 64 (CDR1), SEQ ID NO: 65 (CDR2) and SEQ ID NO: 66 (CDR3) (antigen binding unit CD3#2); and

[0207] iii) an antigen binding unit comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 96 (CDR1), SEQ ID NO: 97 (CDR2) and SEQ ID NO: 98 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 99 (CDR1), SEQ ID NO: 100 (CDR2) and SEQ ID NO: 101 (CDR3) (antigen binding unit CD3#3).

[0208] The first antigen binding units i) to xviii) as summarized above are referred to as DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 and DLL3#18, respectively, and the second antigen binding units i) to iii) as summarized above are referred to as CD3#1, CD3#2 and CD3#3, respectively. A sequence listing is provided herein that allows easy identification of individual amino acid sequences of specific antigen binding units and full-length binding proteins of the invention. An overview is provided in Table 1 in Example 2.

[0209] As used herein, the terms "first" and "second" with respect to antigen binding units are generally intended only to indicate that these units are two different units (because they bind to different target antigens). Therefore, these terms should not be understood to refer to the precise order or sequence of units within the binding proteins of the present invention.

[0210] In some embodiments, the binding protein of the present invention comprises: a first antigen binding unit selected from the group consisting of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 and DLL3#18 as defined by the respective CDR sequences as described above and a second antigen binding unit of CD3#1 as defined by the respective CDR sequences as described above. In a preferred embodiment, the binding protein of the present invention comprises a first antigen binding unit selected from the group consisting of DLL3#1, DLL3#2 and DLL3#3 as defined by the respective CDR sequences as described above and a second antigen binding unit of CD3#1 as defined by the respective CDR sequences as described above.

[0211] In some embodiments, the binding protein of the present invention comprises: a first antigen binding unit selected from the group consisting of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 and DLL3#18 as defined by the respective CDR sequences as described above and a second antigen binding unit of CD3#2 as defined by the respective CDR sequences as described above. In a preferred embodiment, the binding protein of the present invention comprises a first antigen binding unit selected from the group consisting of DLL3#1, DLL3#2 and DLL3#3 as defined by the respective CDR sequences as described above and a second antigen binding unit of CD3#2 as defined by the respective CDR sequences as described above.

[0212] In some embodiments, the binding protein of the present invention comprises: a first antigen binding unit selected from the group consisting of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 and DLL3#18 as defined by the respective CDR sequences as described above and a second antigen binding unit of CD3#3 as defined by the respective CDR sequences as described above. In a preferred embodiment, the binding protein of the present invention comprises a first antigen binding unit selected from the group consisting of DLL3#1, DLL3#2 and DLL3#3 as defined by the respective CDR sequences as described above and a second antigen binding unit of CD3#3 as defined by the respective CDR sequences as described above.

[0213] In a preferred embodiment, the binding protein of the present invention comprises: a first antigen binding unit that specifically binds to DLL3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 1 (CDR1), SEQ ID NO: 2 (CDR2) and SEQ ID NO: 3 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 4 (CDR1), SEQ ID NO: 5 (CDR2) and SEQ ID NO: 6 (CDR3); and a second antigen binding unit that specifically binds to CD3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 55 (CDR1), SEQ ID NO: 56 (CDR2) and SEQ ID NO: 57 (CDR3) and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 58 (CDR1), SEQ ID NO: 59 (CDR2) and SEQ ID NO: 60 (CDR3).

[0214] In a preferred embodiment, the binding protein of the present invention comprises: a first antigen binding unit that specifically binds to DLL3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 7 (CDR1), SEQ ID NO: 8 (CDR2) and SEQ ID NO: 9 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 10 (CDR1), SEQ ID NO: 11 (CDR2) and SEQ ID NO: 12 (CDR3); and a second antigen binding unit that specifically binds to CD3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 55 (CDR1), SEQ ID NO: 56 (CDR2) and SEQ ID NO: 57 (CDR3) and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 58 (CDR1), SEQ ID NO: 59 (CDR2) and SEQ ID NO: 60 (CDR3).

[0215] In a preferred embodiment, the binding protein of the present invention comprises: a first antigen binding unit that specifically binds to DLL3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 13 (CDR1), SEQ ID NO: 14 (CDR2) and SEQ ID NO: 15 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 16 (CDR1), SEQ ID NO: 17 (CDR2) and SEQ ID NO: 18 (CDR3); and a second antigen binding unit that specifically binds to CD3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 55 (CDR1), SEQ ID NO: 56 (CDR2) and SEQ ID NO: 57 (CDR3) and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 58 (CDR1), SEQ ID NO: 59 (CDR2) and SEQ ID NO: 60 (CDR3).

[0216] In a preferred embodiment, the binding protein of the present invention comprises: a first antigen binding unit that specifically binds to DLL3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 1 (CDR1), SEQ ID NO: 2 (CDR2) and SEQ ID NO: 3 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 4 (CDR1), SEQ ID NO: 5 (CDR2) and SEQ ID NO: 6 (CDR3); and a second antigen binding unit that specifically binds to CD3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 61 (CDR1), SEQ ID NO: 62 (CDR2) and SEQ ID NO: 63 (CDR3) and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 64 (CDR1), SEQ ID NO: 65 (CDR2) and SEQ ID NO: 66 (CDR3).

[0217] In a preferred embodiment, the binding protein of the present invention comprises: a first antigen binding unit that specifically binds to DLL3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 7 (CDR1), SEQ ID NO: 8 (CDR2) and SEQ ID NO: 9 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 10 (CDR1), SEQ ID NO: 11 (CDR2) and SEQ ID NO: 12 (CDR3); and a second antigen binding unit that specifically binds to CD3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 61 (CDR1), SEQ ID NO: 62 (CDR2) and SEQ ID NO: 63 (CDR3) and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 64 (CDR1), SEQ ID NO: 65 (CDR2) and SEQ ID NO: 66 (CDR3).

[0218] In a preferred embodiment, the binding protein of the present invention comprises: a first antigen binding unit that specifically binds to DLL3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 13 (CDR1), SEQ ID NO: 14 (CDR2) and SEQ ID NO: 15 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 16 (CDR1), SEQ ID NO: 17 (CDR2) and SEQ ID NO: 18 (CDR3); and a second antigen binding unit that specifically binds to CD3, comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 61 (CDR1), SEQ ID NO: 62 (CDR2) and SEQ ID NO: 63 (CDR3) and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 64 (CDR1), SEQ ID NO: 65 (CDR2) and SEQ ID NO: 66 (CDR3).

[0219] In addition to the CDR sequences as described herein, the antigen binding units of the binding proteins of the present invention include immunoglobulin framework region (FR) sequences. These sequences are preferably non-immunogenic in humans, and are therefore preferably human or humanized FR sequences. Suitable human or humanized FR sequences are known in the art. Specific preferred FR sequences can be taken from the embodiments shown below, which disclose all antigen binding units and thus disclose CDR sequences and FR sequences.

[0220] In a preferred embodiment of the binding protein of the present invention, the first and second binding units each comprise a light chain variable domain and a heavy chain variable domain derived from an antibody molecule, and the light chain / heavy chain variable domains are defined by the CDR sequence of any one of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 or DLL3#18 of the first antigen binding unit and the light / heavy chain variable domains are defined by the CDR sequence of any one of CD3#1, CD3#2 or CD3#3 of the second antigen binding unit. In some embodiments of the binding protein of the present invention, the VH and / or VL domains of the antigen binding unit of any one or more of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, DLL3#18, CD3#1, CD3#2 or CD3#3 are human or humanized VH and / or VL domains.

[0221] In a preferred embodiment of the present invention, the light / heavy chain variable domain of the first antigen binding unit is further defined as follows

[0222] i) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 37 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 38 (antigen binding unit DLL3#1); or

[0223] ii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 39 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 40 (antigen binding unit DLL3#2); or

[0224] iii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:41 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:42 (antigen binding unit DLL3#3); or

[0225] iv) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 43 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 44 (antigen binding unit DLL3#4); or

[0226] v) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 45 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 46 (antigen binding unit DLL3#5); or

[0227] vi) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 47 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 48 (antigen binding unit DLL3#6); ​​or

[0228] vii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 205 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 206 (antigen binding unit DLL3#7); or

[0229] viii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 207 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 208 (antigen binding unit DLL3#8); or

[0230] ix) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 209 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 210 (antigen binding unit DLL3#9); or

[0231] x) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 211 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 212 (antigen binding unit DLL3#10); or

[0232] xi) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 213 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 214 (antigen binding unit DLL3#11); or

[0233] xii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 215 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 216 (antigen binding unit DLL3#12); or

[0234] xiii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 217 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 218 (antigen binding unit DLL3#13); or

[0235] xiv) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 219 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 220 (antigen binding unit DLL3#14); or

[0236] xv) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 221 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 222 (antigen binding unit DLL3#15); or

[0237] xvi) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 223 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 224 (antigen binding unit DLL3#16); or

[0238] xvii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 225 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 226 (antigen binding unit DLL3#17); or

[0239] xviii) comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 227 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 228 (antigen binding unit DLL3#18).

[0240] Preferably, the first antigen binding unit of the binding protein of the present invention is any one of i) to iii) as defined in the above VL and VH sequences.

[0241] In a preferred embodiment of the present invention, the light / heavy chain variable domain of the second antigen binding unit is further defined as follows

[0242] i) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 67 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 68 (antigen binding unit CD3#1); or

[0243] ii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 69 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 70 (antigen binding unit CD3#2); or

[0244] iii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 102 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 103 (antigen binding unit CD3#3).

[0245] In some embodiments, the binding protein of the present invention comprises a combination of the first and second antigen binding units selected from the group consisting of: DLL3#1 / CD3#1, DLL3#2 / CD3#1, DLL3#3 / CD3#1, DLL3#4 / CD3#1, DLL3#5 / CD3#1, DLL3#6 / CD3#1, DLL3#7 / CD3#1, DLL3#8 / CD3#1, DLL3#9 / CD3#1, DLL3#10 / CD3#1, DLL3#11 / CD3#1, DLL3#12 / CD3#1, DLL3#13 / CD3#1, DLL3#14 / CD3#1, DLL3#15 / CD3#1, DLL3#16 / CD3#1, DLL3#17 / CD3#1, DLL3#18 / CD3#1, DLL3#1 / CD3#2, DLL3# 2 / CD3#2, DLL3#3 / CD3#2, DLL3#4 / CD3#2, DLL3#5 / CD3#2, DLL3#6 / CD3#2, DLL3#7 / CD3#2, DLL3#8 / CD3#2, DLL3#9 / CD3#2, DLL 3#10 / CD3#2, DLL3#11 / CD3#2, DLL3#12 / CD3#2, DLL3#13 / CD3#2, DLL3#14 / CD3#2, DLL3#15 / CD3#2, DLL3#16 / CD3#2, DLL3#1 7 / CD3#2, DLL3#18 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#3, DLL3#3 / CD3#3, DLL3#4 / CD3#3, DLL3#5 / CD3#3, DLL3#6 / CD3#3, DL L3#7 / CD3#3, DLL3#8 / CD3#3, DLL3#9 / CD3#3, DLL3#10 / CD3#3, DLL3#11 / CD3#3, DLL3#12 / CD3#3, DLL3#13 / CD3#3, DLL3#14 / CD3#3, DLL3#15 / CD3#3, DLL3#16 / CD3#3, DLL3#17 / CD3#3, DLL3#18 / CD3#3, the first and second antigen binding units are defined by the CDR and / or VH and VL sequences of the antigen binding units as described above.In a preferred embodiment, the binding protein of the present invention comprises a combination of a first and a second antigen binding unit selected from the group consisting of: DLL3#1 / CD3#1, DLL3#2 / CD3#1, DLL3#3 / CD3#1, DLL3#1 / CD3#2, DLL3#2 / CD3#2, DLL3#3 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#3 and DLL3#3 / CD3#3, the first and the second antigen binding unit being defined by the CDR and / or VH and VL sequences of the antigen binding unit as described above.

[0246] In a preferred embodiment, the binding protein of the present invention comprises (i) a first antigen binding unit that specifically binds to DLL3, which comprises: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 37 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 38 and (ii) a second antigen binding unit that specifically binds to CD3, which comprises: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 67 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 68.

[0247] In a preferred embodiment, the binding protein of the present invention comprises (i) a first antigen binding unit that specifically binds to DLL3, which comprises: a light chain variable domain comprising the amino acid sequence of SEQ ID NO:39 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:40 and (ii) a second antigen binding unit that specifically binds to CD3, which comprises: a light chain variable domain comprising the amino acid sequence of SEQ ID NO:67 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:68.

[0248] In a preferred embodiment, the binding protein of the present invention comprises (i) a first antigen binding unit that specifically binds to DLL3, which comprises: a light chain variable domain comprising the amino acid sequence of SEQ ID NO:41 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:42 and (ii) a second antigen binding unit that specifically binds to CD3, which comprises: a light chain variable domain comprising the amino acid sequence of SEQ ID NO:67 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:68.

[0249] In a preferred embodiment, the binding protein of the present invention comprises (i) a first antigen binding unit that specifically binds to DLL3, which comprises: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 37 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 38 and (ii) a second antigen binding unit that specifically binds to CD3, which comprises: a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 69 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 70.

[0250] In a preferred embodiment, the binding protein of the present invention comprises (i) a first antigen binding unit that specifically binds to DLL3, which comprises: a light chain variable domain comprising the amino acid sequence of SEQ ID NO:39 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:40 and (ii) a second antigen binding unit that specifically binds to CD3, which comprises: a light chain variable domain comprising the amino acid sequence of SEQ ID NO:69 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:70.

[0251] In a preferred embodiment, the binding protein of the present invention comprises (i) a first antigen binding unit that specifically binds to DLL3, comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO:41 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:42, and a second antigen binding unit that specifically binds to CD3, comprising: a light chain variable domain comprising the amino acid sequence of SEQ ID NO:69 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:70.

[0252] In some embodiments, the binding protein of the present invention comprises i) a first antigen binding unit that specifically binds to DLL3 (e.g., any one of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, or DLL3#18 as defined by the respective CDRs or VH / VL sequences above), which comprises a first light chain variable domain that is directly or indirectly covalently linked to a first heavy chain variable domain using a first peptide linker and / or ii) a second antigen binding unit that specifically binds to CD3 (e.g., any one of CD3#1, CD3#2, or CD3#3 as defined by the respective CDRs or VH / VL sequences above), which comprises a second light chain variable domain that is directly or indirectly covalently linked to a second heavy chain variable domain using a second peptide linker.

[0253] In some embodiments of the binding protein of the present invention, the first and / or second antigen binding unit further comprises a CL and a CH1 domain as in the light / heavy Fab domains of a conventional antibody molecule, so that the first binding unit comprises a) a VL domain covalently linked (directly or indirectly bonded) to the first CL domain (e.g., defined by the light chain CDR (LC CDR) or VL sequence of any one of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 or DLL3#18) and b) a VH domain covalently linked (directly or indirectly bonded) to the first CH1 domain (e.g., defined by the light chain CDR (LC CDR) or VL sequence of any one of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 or DLL3#18). The second antigen binding unit comprises a) a VL domain covalently linked (directly or indirectly bonded) to a second CL domain (e.g., defined by a LCCDR or VL sequence of any one of CD3#1, CD3#2 or CD3#3) and b) a VH domain covalently linked (directly or indirectly bonded) to a second CH1 domain (e.g., defined by a HCCDR or VH sequence of any one of CD3#1, CD3#2 or CD3#3).

[0254] In the context of the present invention, a CL domain is a constant domain of an antibody light chain, such as a kappa or lambda light chain. An example of a constant region of a kappa light chain is shown in SEQ ID NO: 87. An example of a constant region of a lambda light chain is shown in SEQ ID NO: 88. In some embodiments, the first and second CL domains are the same, such as the first and second CL domains are both kappa light chain constant domains or the first and second CL domains are both lambda light chain constant domains. In some embodiments, the first and second CL domains are different, such as the first CL domain is a constant kappa domain and the second CL domain is a constant lambda domain or vice versa.

[0255] In the context of the present invention, a CH1 domain is the first constant domain of an antibody heavy chain. An example of a constant CH1 domain is shown in SEQ ID NO: 253.

[0256] In a preferred embodiment of the binding protein of the present invention, the first antigen binding unit of the binding protein of the present invention (e.g., DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, DLL3#18, DLL3#19, DLL3#20, DLL3#21, DLL3#22, DLL3#23, DLL3#24, DLL3#25, DLL3#26, DLL3#27, DLL3#28, DLL3#29, DLL3#30, DLL3#31, DLL3#32, DLL3#33, The first and / or second binding units of the binding proteins of the present invention (e.g., any one of CD3#1, CD3#2 or CD3#3 defined by the CDR and / or VH / VL sequences as outlined above) comprise, from N-terminus to C-terminus: a first light chain variable domain, a first CL domain, a first linker peptide, a first VH domain and a first CH1 domain. In this embodiment, the first and / or second binding units have the structure of a single-chain Fab. For both, when the first and / or second antigen binding units form a single-chain Fab, the order may be reversed so that from N-terminus to C-terminus, the antigen binding units comprise: VH-CH1-[linker peptide]-VL-CL. In some embodiments of the protein of the present invention, when the first and / or second antigen binding unit comprises Fab or single-chain Fab, the constant domains may be of the same type (e.g., both CL domains are κ or λ light chain constant domains) or of different types (the first CL domain is κ and the second CL domain is λ light chain constant domain or vice versa).

[0257] In one aspect, the linker for the binding protein of the present invention comprises 26 to 42 amino acids, such as 30 to 40 amino acids. In another aspect, the linker for the protein of the present invention comprises 34 to 40 amino acids, such as 36 to 39 amino acids, such as 38 amino acids. In some embodiments, the linker comprises any one of the sequences SEQ ID NO: 89, 90, 91, 92, 93, 94 or 95, preferably SEQ ID NO: 89.

[0258] The linker sequence may be a naturally occurring sequence or a non-naturally occurring sequence. If used for therapeutic purposes, the linker is preferably not immunogenic in the individual to whom the binding protein of the invention is administered.

[0259] As described in WO1996 / 34103 and WO1994 / 04678, one useful group of linker sequences are linkers derived from the hinge region of a heavy chain antibody. Other examples are polyalanine linker sequences, such as Ala-Ala-Ala.

[0260] A more preferred embodiment of the linker sequence is a Gly / Ser linker with different lengths, such as a (glyxsery)z linker, including, for example, (gly4ser)3, (gly4ser)5, (gly4ser)7, (gly3ser)3, (gly3ser)5, (gly3ser)7, (gly3ser2)3, (gly3ser2)5 and (gly3ser2)7 or any one of SEQ ID NOs: 89 to 95.

[0261] In some embodiments of the binding protein of the present invention, the VL domain of the first antigen binding unit (e.g., defined by the light chain CDR (LC CDR) or VL sequence of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, or DLL3#18) is covalently linked (e.g., directly bonded) to the VH domain of the first antigen binding unit (e.g., defined by the light chain CDR (LC CDR) or VL sequence of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, or DLL3#18) via a first Gly / Ser linker (e.g., any one of 26 to 42 amino acids, 30 to 40 amino acids, 34 to 40 amino acids, or 36 to 39 amino acids, preferably a Gly / Ser linker of 38 amino acids). DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 or DLL3#18); and the VL domain of the second antigen binding unit (e.g., defined by the light chain CDR (HCCDR) or VH sequence of CD3#1, CD3#2 or CD3#3); The first and second linkers are covalently linked (e.g., directly bonded) to the VH domain of the second antigen binding unit (e.g., defined by the heavy chain CDR (HCCDR) or VH sequence of CD3#1, CD3#2 or CD3#3) via a second Gly / Ser linker (e.g., any one of 26 to 42 amino acids, 30 to 40 amino acids, 34 to 40 amino acids, or 36 to 39 amino acids, preferably a Gly / Ser linker of 38 amino acids). More preferably, the first and second linkers are identical. Even more preferably, the first and second linkers each comprise the amino acid sequence SEQ ID NO:89.

[0262] In some embodiments, the binding protein of the invention comprises: a first single-chain Fab, which forms a first antigen binding unit specific for DLL3 and comprises a sequence selected from the group consisting of SEQ ID NO:49, SEQ ID NO:50, SEQ ID NO:51, SEQ ID NO:52, SEQ ID NO:53, SEQ ID NO:54, 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:238, SEQ ID NO:239 and SEQ ID NO:240; and a second single-chain Fab of SEQ ID NO:71, which forms a second antigen binding unit specific for CD3. In a preferred embodiment, the binding protein of the present invention comprises: a first single-chain Fab, which forms a first antigen binding unit, comprising the sequence SEQ ID NO: 49; and a second single-chain Fab, which comprises the sequence SEQ ID NO: 71, forming a second antigen binding unit. In a preferred embodiment, the binding protein of the present invention comprises: a first single-chain Fab, which forms a first antigen binding unit, comprising the sequence SEQ ID NO: 50; and a second single-chain Fab, which comprises the sequence SEQ ID NO: 71, forming a second antigen binding unit. In a preferred embodiment, the binding protein of the present invention comprises: a first single-chain Fab, which forms a first antigen binding unit, comprising the sequence SEQ ID NO: 51; and a second single-chain Fab, which comprises the sequence SEQ ID NO: 71, forming a second antigen binding unit.

[0263] In some embodiments, the binding protein of the invention comprises: a first single-chain Fab, which forms a first antigen binding unit specific for DLL3 and comprises a sequence selected from the group consisting of SEQ ID NO:49, SEQ ID NO:50, SEQ ID NO:51, SEQ ID NO:52, SEQ ID NO:53, SEQ ID NO:54, 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:238, SEQ ID NO:239 and SEQ ID NO:240; and a second single-chain Fab of SEQ ID NO:72, which forms a second antigen binding unit specific for CD3. In a preferred embodiment, the binding protein of the present invention comprises: a first single-chain Fab, which forms a first antigen binding unit, comprising the sequence SEQ ID NO: 49; and a second single-chain Fab, which comprises the sequence SEQ ID NO: 72, forming a second antigen binding unit. In a preferred embodiment, the binding protein of the present invention comprises: a first single-chain Fab, which forms a first antigen binding unit, comprising the sequence SEQ ID NO: 50; and a second single-chain Fab, which comprises the sequence SEQ ID NO: 72, forming a second antigen binding unit. In a preferred embodiment, the binding protein of the present invention comprises: a first single-chain Fab, which forms a first antigen binding unit, comprising the sequence SEQ ID NO: 51; and a second single-chain Fab, which comprises the sequence SEQ ID NO: 72, forming a second antigen binding unit.

[0264] In some embodiments, the binding protein of the invention comprises: a first single-chain Fab, which forms a first antigen binding unit specific for DLL3 and comprises a sequence selected from the group consisting of SEQ ID NO:49, SEQ ID NO:50, SEQ ID NO:51, SEQ ID NO:52, SEQ ID NO:53, SEQ ID NO:54, 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:238, SEQ ID NO:239 and SEQ ID NO:240; and a second single-chain Fab of SEQ ID NO: 104, which forms a second antigen binding unit specific for CD3. In a preferred embodiment, the binding protein of the present invention comprises: a first single-chain Fab, which forms a first antigen binding unit, comprising the sequence SEQ ID NO: 49; and a second single-chain Fab, which comprises the sequence SEQ ID NO: 104, forming a second antigen binding unit. In a preferred embodiment, the binding protein of the present invention comprises: a first single-chain Fab, which forms a first antigen binding unit, comprising the sequence SEQ ID NO: 50; and a second single-chain Fab, which comprises the sequence SEQ ID NO: 104, forming a second antigen binding unit. In a preferred embodiment, the binding protein of the present invention comprises: a first single-chain Fab, which forms a first antigen binding unit, comprising the sequence SEQ ID NO: 51; and a second single-chain Fab, which comprises the sequence SEQ ID NO: 104, forming a second antigen binding unit.

[0265] In some embodiments, the first antigen binding unit (e.g., DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, or DLL3#18 as defined by the CDRs and / or VH / VL sequences outlined above) and / or the second antigen binding unit (e.g., CD3#1, CD3#2, or CD3#3) comprises a VL domain covalently linked (e.g., directly bonded) to a CL domain and a VH domain linked to a CH1 domain (i.e., together forming a Fab fragment derived from an antibody), and the CH1 domain is further covalently linked (e.g., directly bonded) to an Fc domain, thereby forming an arm of a conventional Y-shaped antibody molecule having one light chain and one heavy chain. In some embodiments, the first and second antigen binding units each form a Fab fragment covalently linked (eg, directly bonded) to the first and second Fc domains, respectively, ie, the first and second Fab fragments, thereby forming a conventional heterotetrameric bispecific and bivalent antibody molecule.

[0266] In a preferred embodiment, the first antigen binding unit (e.g., any of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 or DLL3#18 as defined by the CDR and / or VH / VL sequences outlined above) and / or the second antigen binding unit (e.g., any of CD3#1, CD3#2 or CD3#3) comprises a single-chain Fab, i.e., a Gly / Ser linker having 26 to 42 amino acids, 30 to 40 amino acids, 34 to 40 amino acids or 36 to 39 amino acids, preferably 38 amino acids, even more preferably SEQ ID NO: 1. NO:89) covalently linked to the VH-CH1 domain of the heavy chain. In a preferred embodiment, the binding protein of the present invention comprises: a first polypeptide chain comprising a polypeptide chain that is covalently linked to a DLL3 (e.g., DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 or DLL3#18 as defined by the respective CDRs or VH / VL sequences above). Any one of DLL3#18, preferably the first antigen binding unit is DLL3#1, DLL3#2, DLL3#3) and a first single-chain Fab and a first Fc domain (this polypeptide chain is also referred to herein as "DLL3 chain") that specifically bind to; and a second polypeptide chain, which comprises a second single-chain Fab and a second Fc domain (this polypeptide chain is also referred to herein as "CD3 chain") that specifically bind to CD3 (e.g., any one of CD3#1, CD3#2 or CD3#3 as defined by the respective CDR or VH / VL sequences above). In some embodiments, the first and second Fc domains are the same. In a preferred embodiment, the first and second Fc domains are different. The resulting binding protein of the present invention has a complete Fc and has two independent binding sites: a first binding unit for DLL3 and a second binding unit for CD3. In some embodiments, the first antigen binding unit consists of a single polypeptide chain (DLL3 chain). In some embodiments, the second antigen binding unit consists of a single polypeptide chain (CD3 chain). In some embodiments, both the first and second antigen binding units consist of a single polypeptide chain.

[0267] In some embodiments, the first antigen binding unit of the protein of the present invention is formed by a first polypeptide chain (DLL3 chain) and the second antigen binding unit is formed by a second polypeptide chain (CD3 chain). Therefore, in some embodiments, the binding protein of the present invention comprises two different polypeptide chains, each comprising an antigen binding unit with a different specificity, covalently linked to each other, or covalently linked via a disulfide bond or possibly via a peptide linker. In some embodiments, the binding protein of the present invention is a bispecific bivalent heterodimeric protein comprising two polypeptide chains, one polypeptide chain (the first polypeptide chain or DLL3 chain) comprises an antigen binding unit that specifically binds to DLL3 (e.g., any of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, or DLL3#18 defined by respective CDRs and / or VH / VL sequences) and the other polypeptide chain (the second polypeptide chain or CD3 chain) comprises an antigen binding unit that specifically binds to CD3 (e.g., any of CD3#1, CD3#2, or CD3#3).

[0268] In the context of the present invention, the Fc domain is derived, for example, from an IgG, such as a heavy chain of IgG1, IgG2 or IgG4. For example, the Fc domain of the present invention is an Fc domain of a heavy chain of IgG1 or IgG4 and comprises a hinge region and two constant domains (CH2 and CH3). Examples of Fc domains (including hinge regions) are shown in SEQ ID NOs: 81 and 84.

[0269] Unless otherwise specified, the numbering of amino acids in the amino acid chains of the proteins of the invention is herein according to the EU numbering system ((Edelman, Cunningham et al., 1969). Unless otherwise specified, this means that the amino acid numbers indicated herein correspond to the position in the heavy chain of the corresponding subtype (e.g. IgG1 or IgG4) according to the EU numbering system.

[0270] In some embodiments, the first Fc domain and the second Fc domain in the protein of the present invention each include one or more amino acid changes, and the amino acid changes reduce the formation of homodimers of the first or second polypeptide chains rather than heterodimers of the first and second polypeptide chains. Via these changes, a "protrusion" is generated in an Fc domain by replacing one or more small amino acid side chains from the interface of one of the heavy chains with a larger side chain (e.g., tyrosine or tryptophan). Complementary "cavities" of the same or similar size are produced on the interface of another Fc domain by replacing large amino acid side chains with smaller amino acid side chains (e.g., alanine or threonine). This provides a mechanism for increasing heterodimer yields compared to other undesired final products (such as homodimers, especially homodimers of Fc domains with "protrusions") (see, e.g., Ridgway et al., Protein Eng, 1996.9 (7): 617-21 pages; Atwell et al., JMB, 1997, 270, 26-35). In some embodiments, such amino acid changes are tyrosine (Y) [T366Y] at position 366 of the first Fc domain and threonine (T) [Y407T] at position 407 of the second Fc domain. In some embodiments, the first Fc domain comprises a serine (S) [T366S] at position 366 and the second Fc domain comprises a tryptophan (W) [T366W] at position 366, an alanine (A) [L368A] at position 368, and a valine (V) [Y407V] at position 407. In a preferred embodiment, the first Fc domain comprises a tryptophan (W) [T366W] at position 366, and the second Fc domain comprises a serine (S) [T366S] at position 366, an alanine (A) [L368A] at position 368, and a valine (V) [Y407V] at position 407. For example, position 366 of the Fc domain according to EU numbering corresponds to amino acid position 146 in the human IgG1 Fc sequence of SEQ ID NO: 81, which is changed from T at position 146 in SEQ ID NO: 81 to W at position 146 in SEQ ID NO: 82; and positions 366, 368, and 407 according to EU numbering correspond to amino acid positions 146, 148, and 187 in SEQ ID NO: 81, respectively, which are changed from T, L, and Y at these positions in SEQ ID NO: 81 to S, A, and V at these positions in SEQ ID NO: 83. In any of these embodiments, the amino acid changes described for the first Fc domain may be located in the second Fc domain and the respective amino acid changes of the second Fc domain may be located in the first Fc domain. In other words, the terms "first" and "second" are interchangeable in these embodiments. In some embodiments, such Fc domains are Fc domains derived from the heavy chain of IgG1 or IgG4.

[0271] In some embodiments, the first Fc domain comprises a cysteine ​​(C) at position 354 [S354C] in addition to a tryptophan (W) [T366W] at position 366 and the second Fc domain comprises a cysteine ​​(C) at position 349 [Y349C] in addition to a serine (S) at position 366 [T366S], an alanine (A) at position 368 [L368A], and a valine (V) at position 407 [Y407V]. In one aspect, such an Fc domain is an Fc domain derived from the heavy chain of IgG4.

[0272] In some embodiments, the first Fc domain or the second Fc domain in the binding protein of the present invention further comprises one or more amino acid changes that reduce the binding of the Fc domain to protein A. In some embodiments, such amino acid changes are arginine [H435R] at position 435 of one of the Fc domains and phenylalanine [Y436F] at position 436. Both changes are derived from the sequence of human IgG3 (IgG3 does not bind to protein A). These two mutations are located in the CH3 domain and incorporated into one of the Fc domains to reduce binding to protein A (see, e.g., Jendeberg et al., J Immunol Methods, 1997. 201 (1): pp. 25-34). These two changes facilitate the removal of homodimers of heavy chains containing these changes during protein purification.

[0273] In some embodiments, in the binding proteins of the invention, the Fc domain comprising a threonine (T) at position 407 [Y407T] further comprises an arginine at position 435 [H435R] and a phenylalanine at position 436 [Y436F]. In this case, the other heavy chain comprises a tyrosine (Y) at position 366 [T366Y], but does not include the two changes at positions 435 and 436. Alternatively, in some embodiments, in the proteins of the invention, the Fc domain comprising a serine (S) at position 366 [T366S], an alanine (A) at position 368 [L368A], and a valine (V) at position 407 [Y407V] further comprises [H435R] at position 435 and a phenylalanine at position 436 [Y436F]. In this case, the other Fc domain comprises a tryptophan (W) [T366W] at position 366, but does not include the two changes at positions 435 and 436. Therefore, the Fc domain comprising amino acid changes that produce a "cavity" as described above also comprises amino acid changes that reduce binding to protein A. Homodimers comprising this Fc domain are removed via reduced binding to protein A. The generation of homodimers of the other Fc domain comprising a "protrusion" is reduced by the presence of the "protrusion".

[0274] In some embodiments, the Fc domain of the protein of the invention may or may not further comprise a YTE mutation (M252Y / S254T / T256E, EU numbering (Dall'Acqua, Kiener et al., 2006)). These mutations have been shown to improve the pharmacokinetic properties of the Fc domain by preferably increasing the binding affinity for the neonatal FcRn substrate at pH 6.0.

[0275] In some embodiments, the first and / or second Fc domain of the invention derived from IgG1 also includes a "KO" mutation (L234A, L235A). In another aspect, the first and / or second Fc domain of the invention derived from IgG4 also includes a Pro hinge mutation (S228P).

[0276] In a preferred embodiment of the present invention, the binding protein comprises i) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:73 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#1 / CD3#1), or ii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:74 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#2 / CD3#1), or iii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:75 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#3 / CD3#1), or iv) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:76 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#4 / CD3#1), or v) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:77 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#5 / CD3#1), or vi) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:78 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 NO:79 (DLL3#6 / CD3#1); or vii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:241 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#7 / CD3#1); or viii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:242 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#8 / CD3#1); or ix) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:243 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#9 / CD3#1); or x) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:244 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#10 / CD3#1); or xi) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:245 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#11 / CD3#1); or xii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:246 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 NO:246 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#12 / CD3#1); or xiii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:247 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#13 / CD3#1);Or xiv) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:248 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#14 / CD3#1); or xv) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:249 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#15 / CD3#1), or xvi) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:250 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#16 / CD3#1); or xvii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:251 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#17 / CD3#1); or xviii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:252 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:79 (DLL3#18 / CD3#1). Preferably, the first and second polypeptide chains are linked via one or more disulfide bonds and form an antibody-like structure similar to a conventional Y-shaped antibody molecule (; Figure 1 ).

[0277] In another preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO:73, SEQ ID NO:74 and SEQ ID NO:75; and a second polypeptide chain specific for CD3, which comprises the amino acid sequence of SEQ ID NO:79.

[0278] In a preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3 comprising the amino acid sequence of SEQ ID NO:73; and a second polypeptide chain specific for CD3 comprising the amino acid sequence of SEQ ID NO:79.

[0279] In a preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3 comprising the amino acid sequence of SEQ ID NO:74; and a second polypeptide chain specific for CD3 comprising the amino acid sequence of SEQ ID NO:79.

[0280] In a preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3 comprising the amino acid sequence of SEQ ID NO:75; and a second polypeptide chain specific for CD3 comprising the amino acid sequence of SEQ ID NO:79.

[0281] In a preferred embodiment of the present invention, the binding protein comprises i) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:73 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#1 / CD3#2), or ii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:74 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#2 / CD3#2), or iii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:75 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#3 / CD3#2), or iv) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:76 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#4 / CD3#2), or v) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:77 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#5 / CD3#2), or vi) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:78 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO: NO:80 (DLL3#6 / CD3#2); or vii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:241 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#7 / CD3#2); or viii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:242 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#8 / CD3#2); or ix) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:243 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#9 / CD3#2); or x) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:244 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#10 / CD3#2); or xi) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:245 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#11 / CD3#2); or xii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:246 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 NO:246 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#12 / CD3#2); or xiii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:247 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#13 / CD3#2);Or xiv) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:248 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#14 / CD3#2); or xv) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:249 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#15 / CD3#2), or xvi) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:250 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#16 / CD3#2); or xvii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:251 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#17 / CD3#2); or xviii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:252 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:80 (DLL3#18 / CD3#2). Preferably, the first and second polypeptide chains are linked via one or more disulfide bonds and form an antibody-like structure similar to a conventional Y-shaped antibody molecule (; Figure 1 ).

[0282] In another preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO:73, SEQ ID NO:74 and SEQ ID NO:75; and a second polypeptide chain specific for CD3, which comprises the amino acid sequence of SEQ ID NO:80.

[0283] In a preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3 comprising the amino acid sequence of SEQ ID NO:73; and a second polypeptide chain specific for CD3 comprising the amino acid sequence of SEQ ID NO:80.

[0284] In a preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3 comprising the amino acid sequence of SEQ ID NO:74; and a second polypeptide chain specific for CD3 comprising the amino acid sequence of SEQ ID NO:80.

[0285] In a preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3 comprising the amino acid sequence of SEQ ID NO:75; and a second polypeptide chain specific for CD3 comprising the amino acid sequence of SEQ ID NO:80.

[0286] In a preferred embodiment of the present invention, the binding protein comprises i) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:73 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#1 / CD3#3), or ii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:74 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#2 / CD3#3), or iii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:75 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#3 / CD3#3), or iv) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:76 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#4 / CD3#3), or v) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:77 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#5 / CD3#3), or vi) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:78 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO: NO: 105 (DLL3#6 / CD3#3); or vii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO: 241 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO: 105 (DLL3#7 / CD3#3); or viii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO: 242 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO: 105 (DLL3#8 / CD3#3); or ix) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO: 243 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO: 105 (DLL3#9 / CD3#3); or x) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO: 244 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO: 105 (DLL3#10 / CD3#3); or xi) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO: 245 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO: 105 (DLL3#11 / CD3#3); or xii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO: 245 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO: 105 NO:246 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#12 / CD3#3); or xiii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:247 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#13 / CD3#3);or xiv) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:248 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#14 / CD3#3); or xv) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:249 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#15 / CD3#3); or xvi) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:250 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#16 / CD3#3); or xvii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:251 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#17 / CD3#3); or xviii) a first polypeptide chain comprising the amino acid sequence of SEQ ID NO:252 and a second polypeptide chain comprising the amino acid sequence of SEQ ID NO:105 (DLL3#18 / CD3#3). Preferably, the first and second polypeptide chains are linked via one or more disulfide bonds and form an antibody-like structure similar to a conventional Y-shaped antibody molecule (; Figure 1 ).

[0287] In another preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO:73, SEQ ID NO:74 and SEQ ID NO:75; and a second polypeptide chain specific for CD3, which comprises the amino acid sequence of SEQ ID NO:105.

[0288] In a preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3 comprising the amino acid sequence of SEQ ID NO:73; and a second polypeptide chain specific for CD3 comprising the amino acid sequence of SEQ ID NO:105.

[0289] In a preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3 comprising the amino acid sequence of SEQ ID NO:74; and a second polypeptide chain specific for CD3 comprising the amino acid sequence of SEQ ID NO:105.

[0290] In a preferred embodiment, the binding protein comprises: a first polypeptide chain specific for DLL3 comprising the amino acid sequence of SEQ ID NO:75; and a second polypeptide chain specific for CD3 comprising the amino acid sequence of SEQ ID NO:105.

[0291] For all the above embodiments, it is to be understood that by using the term "comprising", it is intended to also include embodiments wherein the respective domain or molecule "consists of" the amino acid sequence as indicated.

[0292] In another aspect, the invention provides a protein comprising a first polypeptide chain that specifically binds to DLL3 and a second polypeptide chain that specifically binds to CD3, wherein the first chain comprises a first light chain covalently linked (e.g., directly bonded) to a first linker, which is itself covalently linked (e.g., directly bonded) to a first heavy chain, and wherein the second chain that specifically binds to CD3 comprises a second light chain covalently linked (e.g., directly bonded) to a second linker, which is itself covalently linked (e.g., directly bonded) to a second heavy chain.

[0293] In some embodiments, starting from its N-terminus, the first polypeptide chain comprises a first light chain variable region that specifically binds to DLL3, a first light chain constant region, a first linker, a first heavy chain variable region that is specific for DLL3, and a first heavy chain constant region. In some embodiments, starting from its N-terminus, the second polypeptide chain comprises a second light chain variable region that specifically binds to CD3, a second light chain constant region, a second linker, a second heavy chain variable region that is specific for CD3, and a second heavy chain constant region.

[0294] The resulting protein has an intact Fc that is slightly larger than that of IgG and has two independent binding sites (e.g., each binding site is monovalent for a respective antigen): a first binding site for DLL3 and a second binding site for CD3. Preferably, the first and second polypeptide chains are linked via one or more disulfide bonds. Thus, the protein of the invention is an antibody-like structure, having the Y-shaped structure of a conventional full-length antibody (see Figure 1 ). This bispecific format greatly reduces heterogeneity after expression and purification (e.g., by avoiding mispairing of light and heavy variable domains with different binding specificities), while maintaining the functional property that the binding portions within the structure are unlikely to generate undesirable immunogenic reactants. This also allows good expression of heterodimeric proteins (e.g., in mammalian cells).

[0295] In some embodiments of the protein of the invention, the first polypeptide chain that specifically binds to DLL3 comprises a first light chain variable domain and a first heavy chain variable domain, wherein the first light chain variable domain and the first heavy chain variable domain comprise a CDR sequence selected from the group consisting of i) to xviii):

[0296] i) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 1 (CDR1), SEQ ID NO: 2 (CDR2), and SEQ ID NO: 3 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 4 (CDR1), SEQ ID NO: 5 (CDR2), and SEQ ID NO: 6 (CDR3);

[0297] ii) a light chain CDR comprising the amino acid sequence of SEQ ID NO:7 (CDR1), SEQ ID NO:8 (CDR2) and SEQ ID NO:9 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO:10 (CDR1), SEQ ID NO:11 (CDR2) and SEQ ID NO:12 (CDR3);

[0298] iii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 13 (CDR1), SEQ ID NO: 14 (CDR2) and SEQ ID NO: 15 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 16 (CDR1), SEQ ID NO: 17 (CDR2) and SEQ ID NO: 18 (CDR3);

[0299] iv) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 19 (CDR1), SEQ ID NO: 20 (CDR2) and SEQ ID NO: 21 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 22 (CDR1), SEQ ID NO: 23 (CDR2) and SEQ ID NO: 24 (CDR3);

[0300] v) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 25 (CDR1), SEQ ID NO: 26 (CDR2), and SEQ ID NO: 27 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 28 (CDR1), SEQ ID NO: 29 (CDR2), and SEQ ID NO: 30 (CDR3); and

[0301] vi) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 31 (CDR1), SEQ ID NO: 32 (CDR2) and SEQ ID NO: 33 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 34 (CDR1), SEQ ID NO: 35 (CDR2) and SEQ ID NO: 36 (CDR3);

[0302] vii) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 133 (CDR1), SEQ ID NO: 134 (CDR2) and SEQ ID NO: 135 (CDR3) and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 136 (CDR1), SEQ ID NO: 137 (CDR2) and SEQ ID NO: 138 (CDR3);

[0303] viii) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 139 (CDR1), SEQ ID NO: 140 (CDR2), and SEQ ID NO: 141 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 142 (CDR1), SEQ ID NO: 143 (CDR2), and SEQ ID NO: 144 (CDR3);

[0304] ix) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 145 (CDR1), SEQ ID NO: 146 (CDR2) and SEQ ID NO: 147 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 148 (CDR1), SEQ ID NO: 149 (CDR2) and SEQ ID NO: 150 (CDR3);

[0305] x) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 151 (CDR1), SEQ ID NO: 152 (CDR2) and SEQ ID NO: 153 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 154 (CDR1), SEQ ID NO: 155 (CDR2) and SEQ ID NO: 156 (CDR3);

[0306] xi) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 157 (CDR1), SEQ ID NO: 158 (CDR2) and SEQ ID NO: 159 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 160 (CDR1), SEQ ID NO: 161 (CDR2) and SEQ ID NO: 162 (CDR3);

[0307] xii) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 163 (CDR1), SEQ ID NO: 164 (CDR2) and SEQ ID NO: 165 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 166 (CDR1), SEQ ID NO: 167 (CDR2) and SEQ ID NO: 168 (CDR3);

[0308] xiii) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 169 (CDR1), SEQ ID NO: 170 (CDR2) and SEQ ID NO: 171 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 172 (CDR1), SEQ ID NO: 173 (CDR2) and SEQ ID NO: 174 (CDR3);

[0309] xiv) comprising: a light chain CDR comprising the amino acid sequence of SEQ ID NO: 175 (CDR1), SEQ ID NO: 176 (CDR2) and SEQ ID NO: 177 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 178 (CDR1), SEQ ID NO: 179 (CDR2) and SEQ ID NO: 180 (CDR3);

[0310] xv) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 181 (CDR1), SEQ ID NO: 182 (CDR2) and SEQ ID NO: 183 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 184 (CDR1), SEQ ID NO: 185 (CDR2) and SEQ ID NO: 186 (CDR3);

[0311] xvi) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 187 (CDR1), SEQ ID NO: 188 (CDR2) and SEQ ID NO: 189 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 190 (CDR1), SEQ ID NO: 191 (CDR2) and SEQ ID NO: 192 (CDR3);

[0312] xvii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 193 (CDR1), SEQ ID NO: 194 (CDR2) and SEQ ID NO: 195 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 196 (CDR1), SEQ ID NO: 197 (CDR2) and SEQ ID NO: 198 (CDR3); and

[0313] xviii) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 199 (CDR1), SEQ ID NO: 200 (CDR2) and SEQ ID NO: 201 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 202 (CDR1), SEQ ID NO: 203 (CDR2) and SEQ ID NO: 204 (CDR3).

[0314] The individual light / heavy chain variable domains defined by these CDR sequences are respectively referred to as DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 and DLL3#18. Preferably, the CDR sequence is selected from the group consisting of i) to iii) (DLL3#1, DLL3#2, DLL3#3) as defined above.

[0315] In a preferred embodiment of the binding protein of the present invention, the second polypeptide chain that specifically binds to CD3 comprises a second light chain variable domain and a second heavy chain variable domain, and the second light chain variable domain and the second heavy chain variable domain comprise CDR sequences selected from the group consisting of:

[0316] i) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 55 (CDR1), SEQ ID NO: 56 (CDR2) and SEQ ID NO: 57 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 58 (CDR1), SEQ ID NO: 59 (CDR2) and SEQ ID NO: 60 (CDR3);

[0317] ii) a light chain CDR comprising the amino acid sequence of SEQ ID NO:61 (CDR1), SEQ ID NO:62 (CDR2) and SEQ ID NO:63 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO:64 (CDR1), SEQ ID NO:65 (CDR2) and SEQ ID NO:66 (CDR3); and

[0318] iii) a light chain CDR comprising the amino acid sequence of SEQ ID NO:96 (CDR1), SEQ ID NO:97 (CDR2) and SEQ ID NO:98 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO:99 (CDR1), SEQ ID NO:100 (CDR2) and SEQ ID NO:101 (CDR3).

[0319] The respective light / heavy chain variable domains defined by these CDR sequences are called CD3#1, CD3#2 and CD3#3, respectively.

[0320] In a preferred embodiment, the binding protein of the invention comprises (i) a first polypeptide chain that specifically binds to DLL3, comprising a first light chain variable domain, the first light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO: 1 (CDR1), SEQ ID NO: 2 (CDR2) and SEQ ID NO: 3 (CDR3) and a first heavy chain variable domain, the first heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO: 4 (CDR1), SEQ ID NO: 5 (CDR2) and SEQ ID NO: 6 (CDR3); and (ii) a second polypeptide chain that specifically binds to CD3, comprising a second light chain variable domain, the second light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO: 55 (CDR1), SEQ ID NO: 56 (CDR2) and SEQ ID NO: 57 (CDR3) and a second heavy chain variable domain, the second heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO: 58 (CDR1), SEQ ID NO: 59 (CDR2) and SEQ ID NO: Heavy chain CDR of IDNO: 60 (CDR3).

[0321] In a preferred embodiment, the binding protein of the invention comprises (i) a first polypeptide chain that specifically binds to DLL3, comprising a first light chain variable domain, the first light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO:7 (CDR1), SEQ ID NO:8 (CDR2), and SEQ ID NO:9 (CDR3), and a first heavy chain variable domain, the first heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO:10 (CDR1), SEQ ID NO:11 (CDR2), and SEQ ID NO:12 (CDR3); and (ii) a second polypeptide chain that specifically binds to CD3, comprising a second light chain variable domain, the second light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO:55 (CDR1), SEQ ID NO:56 (CDR2), and SEQ ID NO:57 (CDR3), and a second heavy chain variable domain, the second heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO:58 (CDR1), SEQ ID NO:59 (CDR2), and SEQ ID NO: Heavy chain CDR of NO:60 (CDR3).

[0322] In a preferred embodiment, the binding protein of the invention comprises (i) a first polypeptide chain that specifically binds to DLL3, comprising a first light chain variable domain, the first light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO: 13 (CDR1), SEQ ID NO: 14 (CDR2) and SEQ ID NO: 15 (CDR3) and a first heavy chain variable domain, the first heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO: 16 (CDR1), SEQ ID NO: 17 (CDR2) and SEQ ID NO: 18 (CDR3); and (ii) a second polypeptide chain that specifically binds to CD3, comprising a second light chain variable domain, the second light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO: 55 (CDR1), SEQ ID NO: 56 (CDR2) and SEQ ID NO: 57 (CDR3) and a second heavy chain variable domain, the second heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO: 58 (CDR1), SEQ ID NO: 59 (CDR2) and SEQ ID NO: Heavy chain CDR of NO:60 (CDR3).

[0323] In a preferred embodiment, the binding protein of the invention comprises (i) a first polypeptide chain that specifically binds to DLL3, comprising a first light chain variable domain, the first light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO: 1 (CDR1), SEQ ID NO: 2 (CDR2), and SEQ ID NO: 3 (CDR3) and a first heavy chain variable domain, the first heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO: 4 (CDR1), SEQ ID NO: 5 (CDR2), and SEQ ID NO: 6 (CDR3); and (ii) a second polypeptide chain that specifically binds to CD3, comprising a second light chain variable domain, the second light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO: 61 (CDR1), SEQ ID NO: 62 (CDR2), and SEQ ID NO: 63 (CDR3) and a second heavy chain variable domain, the second heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO: 64 (CDR1), SEQ ID NO: 65 (CDR2), and SEQ ID NO: Heavy chain CDR of NO:66 (CDR3).

[0324] In a preferred embodiment, the binding protein of the invention comprises (i) a first polypeptide chain that specifically binds to DLL3, comprising a first light chain variable domain, the first light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO:7 (CDR1), SEQ ID NO:8 (CDR2), and SEQ ID NO:9 (CDR3), and a first heavy chain variable domain, the first heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO:10 (CDR1), SEQ ID NO:11 (CDR2), and SEQ ID NO:12 (CDR3); and (ii) a second polypeptide chain that specifically binds to CD3, comprising a second light chain variable domain, the second light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO:61 (CDR1), SEQ ID NO:62 (CDR2), and SEQ ID NO:63 (CDR3), and a second heavy chain variable domain, the second heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO:64 (CDR1), SEQ ID NO:65 (CDR2), and SEQ ID NO:66 (CDR3). Heavy chain CDR of NO:66 (CDR3).

[0325] In a preferred embodiment, the binding protein of the invention comprises (i) a first polypeptide chain that specifically binds to DLL3, comprising a first light chain variable domain, the first light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO: 13 (CDR1), SEQ ID NO: 14 (CDR2), and SEQ ID NO: 15 (CDR3), and a first heavy chain variable domain, the first heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO: 16 (CDR1), SEQ ID NO: 17 (CDR2), and SEQ ID NO: 18 (CDR3); and (ii) a second polypeptide chain that specifically binds to CD3, comprising a second light chain variable domain, the second light chain variable domain having light chain CDRs comprising the amino acid sequence of SEQ ID NO: 61 (CDR1), SEQ ID NO: 62 (CDR2), and SEQ ID NO: 63 (CDR3), and a second heavy chain variable domain, the second heavy chain variable domain having heavy chain CDRs comprising the amino acid sequence of SEQ ID NO: 64 (CDR1), SEQ ID NO: 65 (CDR2), and SEQ ID NO: Heavy chain CDR of NO:66 (CDR3).

[0326] In a preferred embodiment of the protein of the present invention, the first polypeptide chain that specifically binds to DLL3 comprises a light chain variable domain (first light chain variable domain) and a heavy chain variable domain (first heavy chain variable domain) selected from the group consisting of i) to xviii):

[0327] i) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 37 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 38 (DLL3#1);

[0328] ii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:39 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:40 (DLL3#2);

[0329] iii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:41 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:42 (DLL3#3);

[0330] iv) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:43 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:44 (DLL3#4);

[0331] v) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:45 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:46 (DLL3#5);

[0332] vi) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:47 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:48 (DLL3#6);

[0333] vii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 205 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 206 (DLL3#7);

[0334] viii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 207 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 208 (DLL3#8);

[0335] ix) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 209 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 210 (DLL3#9);

[0336] x) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 211 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 212 (DLL3#10);

[0337] xi) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 213 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 214 (DLL3#11);

[0338] xii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 215 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 216 (DLL3#12);

[0339] xiii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 217 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 218 (DLL3#13);

[0340] xiv) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 219 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 220 (DLL3#14);

[0341] xv) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 221 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 222 (DLL3#15);

[0342] xvi) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 223 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 224 (DLL3#16);

[0343] xvii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 225 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 226 (DLL3#17); and

[0344] xviii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 227 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 228 (DLL3#18).

[0345] Preferably, the light chain variable domain and heavy chain variable domain sequences are selected from the group consisting of i) to iii) (DLL3#1, DLL3#2, DLL3#3) as defined above.

[0346] In a preferred embodiment of the protein of the present invention, the second polypeptide chain that specifically binds to CD3 comprises a light chain variable domain (second light chain variable domain) and a heavy chain variable domain (second heavy chain variable domain) selected from the group consisting of:

[0347] i) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 67 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 68 (CD3#1);

[0348] ii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 69 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 70 (CD3#2); and

[0349] iii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 102 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 103 (CD3#3).

[0350] In some embodiments, the binding protein of the present invention comprises a first and a second polypeptide chain, wherein the first and the second polypeptide chain comprise CDRs and / or VH and VL sequences of light / heavy chain variable domains selected from the list consisting of: DLL3#1 / CD3#1, DLL3#2 / CD3#1, DLL3#3 / CD3#1, DLL3#4 / CD3#1, DLL3#5 / CD3#1, DLL3#6 / CD3#1, DLL3#7 / CD3#1, DLL3#8 / CD3#1, DLL3#9 / CD3#1, DLL3#10 / CD3 #1, DLL3#11 / CD3#1, DLL3#12 / CD3#1, DLL3#13 / CD3#1, DLL3#14 / CD3#1, DLL3#15 / CD3#1, DLL3#16 / CD3#1, DLL3#17 / CD3#1 , DLL3#18 / CD3#1, DLL3#1 / CD3#2, DLL3#2 / CD3#2, DLL3#3 / CD3#2, DLL3#4 / CD3#2, DLL3#5 / CD3#2, DLL3#6 / CD3#2, DLL3#7 / CD3#2, DLL3#8 / CD3#2, DLL3#9 / CD3#2, DLL3#10 / CD3#2, DLL3#11 / CD3#2, DLL3#12 / CD3#2, DLL3#13 / CD3#2, DLL3#14 / CD3# 2. DLL3#15 / CD3#2, DLL3#16 / CD3#2, DLL3#17 / CD3#2, DLL3#18 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#3, DLL3#3 / CD3#3, DLL 3#4 / CD3#3, DLL3#5 / CD3#3 and DLL3#6 / CD3#3, DLL3#7 / CD3#3, DLL3#8 / CD3#3, DLL3#9 / CD3#3, DLL3#10 / CD3#3, DLL3#11 / CD3 #3, DLL3#12 / CD3#3, DLL3#13 / CD3#3, DLL3#14 / CD3#3, DLL3#15 / CD3#3, DLL3#16 / CD3#3, DLL3#17 / CD3#3, DLL3#18 / CD3#3. In a preferred embodiment, the binding protein of the present invention comprises a first and a second polypeptide chain, which comprises CDRs and / or VH and VL sequences of light / heavy chain variable domains selected from the list consisting of: DLL3#1 / CD3#1, DLL3#2 / CD3#1, DLL3#3 / CD3#1, DLL3#1 / CD3#2, DLL3#2 / CD3#2, DLL3#3 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#3 and DLL3#3 / CD3#3.

[0351] In a preferred embodiment, the binding protein of the invention comprises (i) a first polypeptide chain that specifically binds to DLL3, which comprises a light chain variable domain of SEQ ID NO:37 and a heavy chain variable domain of SEQ ID NO:38; and (ii) a second polypeptide chain that specifically binds to CD3, which comprises a light chain variable domain of SEQ ID NO:67 and a heavy chain variable domain of SEQ ID NO:68.

[0352] In a preferred embodiment, the binding protein of the invention comprises (i) a first polypeptide chain that specifically binds to DLL3, which comprises a light chain variable domain of SEQ ID NO:39 and a heavy chain variable domain of SEQ ID NO:40; and (ii) a second polypeptide chain that specifically binds to CD3, which comprises a light chain variable domain of SEQ ID NO:67 and a heavy chain variable domain of SEQ ID NO:68.

[0353] In a preferred embodiment, the binding protein of the present invention comprises (i) a first polypeptide chain that specifically binds to DLL3, which comprises a light chain variable domain of SEQ ID NO:41 and a heavy chain variable domain of SEQ ID NO:42; and (ii) a second polypeptide chain that specifically binds to CD3, which comprises a light chain variable domain of SEQ ID NO:67 and a heavy chain variable domain of SEQ ID NO:68.

[0354] In a preferred embodiment, the binding protein of the invention comprises (i) a first polypeptide chain that specifically binds to DLL3, which comprises a light chain variable domain of SEQ ID NO:37 and a heavy chain variable domain of SEQ ID NO:38; and (ii) a second polypeptide chain that specifically binds to CD3, which comprises a light chain variable domain of SEQ ID NO:69 and a heavy chain variable domain of SEQ ID NO:70.

[0355] In a preferred embodiment, the binding protein of the invention comprises (i) a first polypeptide chain that specifically binds to DLL3, which comprises a light chain variable domain of SEQ ID NO:39 and a heavy chain variable domain of SEQ ID NO:40; and (ii) a second polypeptide chain that specifically binds to CD3, which comprises a light chain variable domain of SEQ ID NO:69 and a heavy chain variable domain of SEQ ID NO:70.

[0356] In a preferred embodiment, the binding protein of the present invention comprises (i) a first polypeptide chain that specifically binds to DLL3, which comprises a light chain variable domain of SEQ ID NO:41 and a heavy chain variable domain of SEQ ID NO:42; and (ii) a second polypeptide chain that specifically binds to CD3, which comprises a light chain variable domain of SEQ ID NO:69 and a heavy chain variable domain of SEQ ID NO:70.

[0357] In some embodiments of the present invention, the first and / or second linker peptide comprises a linker as described above, such as a linker derived from a hinge region, a polyalanine linker or a Gly / Ser linker, wherein the linker comprises 26 to 42 amino acids, such as 30 to 40 amino acids, 34 to 40 amino acids, or any one of 36 to 39 amino acids, preferably 38 amino acids.

[0358] In preferred embodiments, the first polypeptide chain that is specific for DLL3 comprises the amino acid sequence of any one of SEQ ID NO:49, SEQ ID NO:50, SEQ ID NO:51, SEQ ID NO:52, SEQ ID NO:53, SEQ ID NO:54, 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:238, SEQ ID NO:239, or SEQ ID NO:240 and the second polypeptide chain that is specific for CD3 comprises the amino acid sequence of SEQ ID NO:71.

[0359] In some embodiments, the first polypeptide chain that is specific for DLL3 comprises the amino acid sequence of any one of SEQ ID NO:49, SEQ ID NO:50, SEQ ID NO:51, SEQ ID NO:52, SEQ ID NO:53, SEQ ID NO:54, 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:238, SEQ ID NO:239, or SEQ ID NO:240 and the second polypeptide chain that is specific for CD3 comprises the amino acid sequence of SEQ ID NO:72.

[0360] In some embodiments, the first polypeptide chain that is specific for DLL3 comprises the amino acid sequence of any one of SEQ ID NO:49, SEQ ID NO:50, SEQ ID NO:51, SEQ ID NO:52, SEQ ID NO:53, SEQ ID NO:54, 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:238, SEQ ID NO:239, or SEQ ID NO:240 and the second polypeptide chain that is specific for CD3 comprises the amino acid sequence of SEQ ID NO:104.

[0361] In a preferred embodiment, the first polypeptide chain specific for DLL3 comprises the amino acid sequence of SEQ ID NO:49 and the second polypeptide chain specific for CD3 comprises the amino acid sequence of SEQ ID NO:71.

[0362] In a preferred embodiment, the first polypeptide chain specific for DLL3 comprises the amino acid sequence of SEQ ID NO:50 and the second polypeptide chain specific for CD3 comprises the amino acid sequence of SEQ ID NO:71.

[0363] In a preferred embodiment, the first polypeptide chain specific for DLL3 comprises the amino acid sequence of SEQ ID NO:51 and the second polypeptide chain specific for CD3 comprises the amino acid sequence of SEQ ID NO:71.

[0364] In a preferred embodiment, the first polypeptide chain specific for DLL3 comprises the amino acid sequence of SEQ ID NO:49 and the second polypeptide chain specific for CD3 comprises the amino acid sequence of SEQ ID NO:72.

[0365] In a preferred embodiment, the first polypeptide chain specific for DLL3 comprises the amino acid sequence of SEQ ID NO:50 and the second polypeptide chain specific for CD3 comprises the amino acid sequence of SEQ ID NO:72.

[0366] In a preferred embodiment, the first polypeptide chain specific for DLL3 comprises the amino acid sequence of SEQ ID NO:51 and the second polypeptide chain specific for CD3 comprises the amino acid sequence of SEQ ID NO:72.

[0367] In some embodiments of the binding proteins of the present invention, the first and second polypeptide chains comprise an Fc domain derived from a heavy chain of an IgG (e.g., IgG1, IgG2, or IgG4). For example, the Fc domain of the present invention is an Fc domain of a heavy chain of an IgG1 or IgG4 and comprises a hinge region and two constant domains (CH2 and CH3). Examples of Fc domains of human IgG are shown in SEQ ID NOs: 81 and 84.

[0368] In some embodiments of the binding proteins of the present invention, the heavy chain comprises one or more amino acid changes. For example, such amino acid changes are tyrosine (Y) [T366Y] at position 366 of the first heavy chain and threonine (T) [Y407T] at position 407 of the second heavy chain. In some embodiments, the first heavy chain comprises a serine (S) [T366S] at position 366 and the second heavy chain comprises a tryptophan (W) [T366W] at position 366, an alanine (A) [L368A] at position 368, and a valine (V) [Y407V] at position 407. In a preferred embodiment, the first heavy chain comprises a tryptophan (W) [T366W] at position 366 and the second heavy chain comprises a serine (S) [T366S] at position 366, an alanine (A) [L368A] at position 368, and a valine (V) [Y407V] at position 407. For example, position 366 of the Fc domain according to EU numbering corresponds to amino acid position 146 in the human IgG1 Fc sequence of SEQ ID NO: 81, which is changed from T at position 146 in SEQ ID NO: 81 to W at position 146 in SEQ ID NO: 82; and positions 366, 368, and 407 according to EU numbering correspond to amino acid positions 146, 148, and 187 in SEQ ID NO: 81, respectively, which are changed from T, L, and Y at these positions in SEQ ID NO: 81 to S, A, and V at these positions in SEQ ID NO: 83. In any of these embodiments, the amino acid changes described for the first heavy chain may be located in the second heavy chain and the respective amino acid changes of the second heavy chain may be located in the first heavy chain. In other words, the terms "first" and "second" are interchangeable in these embodiments. In some embodiments, the heavy chain is derived from a heavy chain of IgG1 or IgG4.

[0369] In some embodiments, the first heavy chain or the second heavy chain in the protein of the invention further comprises one or more amino acid changes that reduce binding of the heavy chain to protein A. In some embodiments, such amino acid changes are arginine at position 435 [H435R] and phenylalanine at position 436 [Y436F] in one of the heavy chains.

[0370] In some embodiments, in the protein of the invention, the heavy chain comprising a threonine (T) at position 407 [Y407T] further comprises an arginine at position 435 [H435R] and a phenylalanine at position 436 [Y436F]. In this case, the other heavy chain comprises a tyrosine (Y) at position 366 [T366Y], but does not include the two changes at positions 435 and 436. Alternatively, in some embodiments, in the protein of the invention, the heavy chain comprising a serine (S) at position 366 [T366S], an alanine (A) at position 368 [L368A], and a valine (V) at position 407 [Y407V] further comprises an arginine at position 435 [H435R] and a phenylalanine at position 436 [Y436F]. In this case, the other heavy chain comprises a tryptophan (W) at position 366 [T366W], but does not include the two changes at positions 435 and 436. Thus, the heavy chains comprising amino acid changes that produce a "cavity" as described above also comprise amino acid changes that reduce binding to protein A. Homodimers comprising these heavy chains are removed via reduced binding to protein A. The generation of homodimers of the other heavy chain comprising a "knob" is reduced by the presence of the "knob".

[0371] In some embodiments, the heavy chain of the protein of the invention may or may not further comprise a YTE mutation (M252Y / S254T / T256E, EU numbering (Dall'Acqua, Kiener et al., 2006)). These mutations have been shown to improve the pharmacokinetic properties of the heavy chain by preferably increasing the binding affinity for the neonatal FcRn substrate at pH 6.0.

[0372] In some embodiments, the first and / or second heavy chains of the invention derived from IgG1 also include a "KO" mutation (L234A, L235A). In another aspect, the first and / or second heavy chains of the invention derived from IgG4 also include a Pro hinge mutation (S228P).

[0373] In another aspect, the protein of the invention comprises a first antigen binding unit or polypeptide chain specific for DLL3 with an affinity for human and cynomolgus monkey DLL3 of preferably ≤10 nM, more preferably ≤1 nM, even more preferably ≤0.1 nM, even more preferably ≤0.01 nM. Affinity can be measured in an SPR (BIAcore) analysis using recombinant DLL3 proteins, as described, for example, in the examples or other methods known to the skilled person. The protein comprises a second antigen binding unit or polypeptide chain with an affinity for human and cynomolgus monkey CD3εγ complex of preferably ≤500 nM, more preferably ≤100 nM, even more preferably ≤10 nM.

[0374] In another aspect, the DLL3 / CD3 binding proteins of the invention do not bind to DLL3 negative cells and do not cross-react with human and DLL3 paralogs DLL1 and DLL4, as shown in Example 5.

[0375] In another aspect, the DLL3 / CD3 binding protein of the invention comprises a first antigen binding unit or a first polypeptide chain that specifically binds to a membrane-proximal peptide of a DLL3 protein. In another aspect, the protein of the invention exhibits weak binding to DLL3 protein-expressing cells, for example, saturation of protein binding to the cell surface is not achieved even at a concentration of 100 nM (see, e.g., FIG. 6 ).

[0376] In another aspect, the DLL3 / CD3 binding proteins of the invention are capable of mediating cytotoxicity against tumor cells by providing optimal spatial conditions for the formation of cytolytic synapses between DLL3-expressing tumor cells and T cells so as to selectively redirect T cell activity towards targeted tumor cells, thereby resulting in tumor cell lysis.

[0377] Various methods can be used to measure cytotoxicity mediated by the DLL3 / CD3 binding proteins of the invention. For example, cytotoxicity can be measured using the methods described in Example 10. Effector cells can be, for example, stimulated or unstimulated (human or cynomolgus monkey) T cells or subsets thereof (e.g., CD4, CD8) or unstimulated (human or cynomolgus monkey) peripheral blood mononuclear cells (PBMC). Target cells should express at least the extracellular domain of (human or cynomolgus monkey) DLL3 and can be cells with endogenous (natural) DLL3 expression, such as human small cell lung cancer cell lines SHP77, NCI-H82, or alternatively recombinant cells expressing full-length DLL3 or the extracellular domain of DLL3. The effector: target cell ratio (E:T) is typically about 10:1, but can vary. The cytotoxic activity of the DLL3 / CD3 binding molecules can be determined, for example, in an LDH-release assay after 48 or 72 hours of incubation. Modifications to the incubation time and readout for determining cytotoxicity are possible and known to the skilled person. Readout systems for cytotoxicity can include MTT / MTS assays, ATP-based assays, FACS-based assays, 51-chromium release assays, sulforhodamine B (SRB) assays, colorimetric (WST) assays, cell colony assays, ECIS technology, and bioluminescence assays.

[0378] The cytotoxic activity mediated by the DLL3 / CD3 binding proteins of the invention is preferably measured in a cell-based cytotoxicity assay. Cytotoxicity is measured by EC 90 The skilled person will appreciate that when using purified T cells as effector cells, ECs are expected to be higher than those of PBMCs. 90The skilled person will also realize that when using stimulated T cells, EC 90 It is further expected that when the target cells express a large amount of DLL3 on the cell surface, EC 90 The EC values ​​of DLL3 / CD3 binding proteins are lower. 90 Preferably ≤ 10 nM, more preferably ≤ 5 nM and even more preferably ≤ 1 nM.

[0379] Preferably, the multispecific binding protein of the invention does not induce / mediate lysis of DLL3-negative cells. The term "does not induce / mediate" "lysis" of DLL3-negative cells means that the DLL3 / CD3 binding molecule does not induce or mediate more than 30%, preferably no more than 20%, more preferably no more than 10% and especially no more than 5% of DLL3-negative cells, while lysis of DLL3-positive lung cancer cell lines is set to 100%. This generally applies to binding protein concentrations up to 1000 nM.

[0380] Preferably, the DLL3 / CD3 binding proteins of the invention are not internalized by the targeted cells. The internalization rate can be analyzed, for example, as described in Example 11. Preferably, after pre-incubation of the DLL3 / CD3 binding protein with the target cells for 4 hours, the internalization rate (e.g., measured as a reduction in cytotoxicity) is ≤50%, more preferably ≤40% and even more preferably ≤30%.

[0381] Furthermore, the DLL3 / CD3 binding proteins of the invention are shown to be stable, with a monomer content above 95% (e.g. at least 98%, see Example 14), have favorable pharmacokinetic properties and good downstream manufacturability and are further expected to have good biodistribution (see, e.g., Example 12). The proteins of the invention furthermore have favorable immunogenicity characteristics (see Example 16) and have good in vitro and in vivo stability (see, e.g., Examples 12 and 15). Furthermore, the DLL3 / CD3 binding proteins of the invention (e.g., DLL3#3 / CD3#1, DLL3#3 / CD3#2) exhibit favorable efficacy in a humanized in vivo xenograft mouse model. The DLL3 / CD3 binding proteins induce strong tumor regressions, which have already begun after the first dose of the DLL3 / CD3 binding proteins. Furthermore, the DLL3 / CD3 binding proteins of the invention induce tumor regressions at a very low dose of 0.25 mg / kg administered once a week (q7d), further supporting their therapeutic applicability. In detail, the DLL3 / CD3 binding protein of the present invention induces selective T cell proliferation, T cell activation, T cell degranulation and cytokine secretion only in the presence of DLL3-positive target cells and in the absence of DLL3-negative target cells (see Example 18), and further significantly increases T cell infiltration into tumor tissues (see Example 20). In addition, Example 19 shows that the DLL3 / CD3 binding protein mediates the lysis of CD4+ and CD8+ T cell redirection. In detail, naive T cells and CD4 + Effector memory cells, CD4 + Central memory cells, CD8 + CD45RA + Effector cells and CD8 + Memory cells contribute to T cell-redirected lysis of DLL3-expressing tumor cells.

[0382] Another aspect of the invention provides an isolated nucleic acid molecule encoding a first and / or second antigen binding unit of a multispecific binding protein of the invention. In some embodiments, the nucleic acid molecule further encodes a first and / or second Fc domain, respectively, as described herein, linked to the first and / or second Fc domain at the 3' end of the nucleic acid molecule encoding the first and / or second antigen binding unit. In some embodiments, the nucleic acid molecule encodes i) a first polypeptide chain comprising a first single-chain Fab specific for DLL3 (e.g., any one of DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, and DLL3#18), and an optional first Fc domain and / or ii) a second polypeptide chain comprising a second single-chain Fab specific for CD3 (e.g., any one of CD3#1, CD3#2, and CD3#3), and an optional second Fc domain.

[0383] Preferably, the nucleic acid molecule comprises a nucleotide sequence encoding a first single-chain Fab of any one of the following: SEQ ID NO:49, SEQ ID NO:50, SEQ ID NO:51, SEQ ID NO:52, SEQ ID NO:53, SEQ ID NO:54, 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:238, SEQ ID NO:239 or SEQ ID NO:240 and / or a second single-chain Fab of SEQ ID NO:71, SEQ ID NO:72 or SEQ ID NO:104. In some embodiments, the nucleic acid molecule comprises a first polypeptide chain encoding any one of SEQ ID NO:73, SEQ ID NO:74, SEQ ID NO:75, SEQ ID NO:76, SEQ ID NO:77, SEQ ID NO:78, SEQ ID NO:241, 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, SEQ ID NO:251, or SEQ ID NO:252 and / or a second polypeptide chain specific for CD3 comprising the amino acid sequence of SEQ ID NO:79, SEQ ID NO:80, or SEQ ID NO: 105.

[0384] Another aspect of the present invention provides an expression vector containing a DNA molecule comprising a nucleotide sequence encoding a first and / or second antigen binding domain (e.g., a first and / or second single-chain Fab of the present invention). Preferably, the expression vector further comprises a nucleic acid molecule encoding a first and / or second Fc domain connected to the nucleic acid molecule, preferably a DNA molecule, preferably a DNA molecule encoding a first and / or second antigen binding domain (e.g., a first and / or second single-chain Fab, respectively). Thus, the expression vector comprises: a nucleotide sequence encoding a polypeptide chain comprising a first single-chain Fab connected to a first Fc domain; and / or a nucleotide sequence encoding a polypeptide chain comprising a second single-chain Fab connected to a second Fc domain.

[0385] In a preferred embodiment, the expression vector contains a DNA molecule comprising a nucleotide sequence encoding the first and / or second polypeptide chain of the present invention. In a preferred embodiment, the expression vector comprises a nucleotide sequence encoding a first polypeptide chain of any of SEQ ID NO: 73, SEQ ID NO: 74, SEQ ID NO: 75, SEQ ID NO: 76, SEQ ID NO: 77, SEQ ID NO: 78, SEQ ID NO: 241, 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, SEQ ID NO: 251 or SEQ ID NO: 252 and / or a second polypeptide chain comprising SEQ ID NO: 79.

[0386] In another preferred embodiment, the expression vector comprises a nucleotide sequence encoding a first polypeptide chain of any one of the following: SEQ ID NO:73, SEQ ID NO:74, SEQ ID NO:75, SEQ ID NO:76, SEQ ID NO:77, SEQ ID NO:78, SEQ ID NO:241, 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, SEQ ID NO:251 or SEQ ID NO:252 and / or a second polypeptide chain comprising SEQ ID NO:80.

[0387] In another preferred embodiment, the expression vector comprises a nucleotide sequence encoding a first polypeptide chain of any one of the following: SEQ ID NO:73, SEQ ID NO:74, SEQ ID NO:75, SEQ ID NO:76, SEQ ID NO:77, SEQ ID NO:78, SEQ ID NO:241, 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, SEQ ID NO:251 or SEQ ID NO:252 and / or a second polypeptide chain comprising SEQ ID NO:105.

[0388] In a particularly preferred embodiment, two expression vectors can be used, one for expressing a first polypeptide chain specific for DLL3 and the other for expressing a second polypeptide chain specific for CD3, both of which can then be transfected into a host cell for recombinant protein expression.

[0389] Preferably, the expression vector will be a vector comprising the or said nucleic acid molecule operably linked to at least one regulatory sequence, wherein such regulatory sequence may be a promoter, enhancer or terminator sequence, and most preferably a heterologous promoter, enhancer or terminator sequence.

[0390] In another aspect, the present invention relates to a host cell having an expression vector encoding a first polypeptide chain of the present invention that is specific for DLL3 and an expression vector encoding a second polypeptide chain of the present invention that is specific for CD3.

[0391] According to a particularly preferred embodiment, the host cell is a eukaryotic cell, such as a mammalian cell. In another embodiment, such a host cell is a bacterial cell. Other useful cell lines are yeast cells or other fungal cells.

[0392] Suitable mammalian cells include, for example, CHO cells, BHK cells, HeLa cells, COS cells and the like. However, amphibian cells, insect cells, plant cells and any other cells used in the art to express heterologous proteins may also be used.

[0393] Anti-DLL3 Antibodies

[0394] DLL3 is normally expressed on intracellular membranes, including those of the high-glandular bodies in the fetal brain, and plays a key role in somitogenesis in the paraxial mesoderm (Geffers et al., J Cell Biol. 2007; 178: 465; Chapman et al., Hum. Mol. Genet. 2011; 20: 905-916). The pronounced induction of DLL3 expression in some tumor types, including SCLC, LCNEC, and glioblastoma, results in localization to the cell surface: this, together with the lack of detectable cell surface DLL3 in non-malignant cells, provides new opportunities for tumor cell-specific therapies and the use of anti-DLL3 antibodies as diagnostic and prognostic tools.

[0395] Several anti-DLL3 antibodies (such as LS-c167440, Lifespan; AP21739PU-N, Acris; LS-C148700, LSBio) are commercially available. However, these antibodies express poorly in immunohistochemistry (IHC), FACS and ELISA analysis and fail to specifically bind to DLL3 proteins expressed on cells, especially tumor cells. Anti-DLL3 antibodies have also been reported, for example, in WO2007111733 and suggested for diagnostic use in glioma patients, but no data for IHC analysis are presented. Therefore, the use of anti-DLL3 antibodies as a reliable diagnostic tool to accurately measure DLL3 expression in patients (e.g., tumor cells / tissues) and / or to evaluate the efficacy of DLL3 targeted therapies remains challenging.

[0396] Therefore, there is a need to identify alternative anti-DLL3 antibodies that can be used to accurately and specifically detect DLL3 protein expression in any kind of biological specimen in various analyses (such as FACS, ELISA, immunoprecipitation, Western blot, ELISA, radioimmunoassay, flow cytometry, IHC and immunoassay analysis) and can be used as reliable diagnostic reagents.

[0397] Therefore, another aspect of the invention provides an anti-DLL3 antibody molecule comprising

[0398] i) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 1 (CDR1), SEQ ID NO: 2 (CDR2) and SEQ ID NO: 3 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 4 (CDR1), SEQ ID NO: 5 (CDR2) and SEQ ID NO: 6 (CDR3); or

[0399] ii) a light chain CDR comprising the amino acid sequence of SEQ ID NO:7 (CDR1), SEQ ID NO:8 (CDR2) and SEQ ID NO:9 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO:10 (CDR1), SEQ ID NO:11 (CDR2) and SEQ ID NO:12 (CDR3); or

[0400] iii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 13 (CDR1), SEQ ID NO: 14 (CDR2) and SEQ ID NO: 15 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 16 (CDR1), SEQ ID NO: 17 (CDR2) and SEQ ID NO: 18 (CDR3); or

[0401] iv) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 19 (CDR1), SEQ ID NO: 20 (CDR2) and SEQ ID NO: 21 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 22 (CDR1), SEQ ID NO: 23 (CDR2) and SEQ ID NO: 24 (CDR3); or

[0402] v) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 25 (CDR1), SEQ ID NO: 26 (CDR2) and SEQ ID NO: 27 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 28 (CDR1), SEQ ID NO: 29 (CDR2) and SEQ ID NO: 30 (CDR3); or

[0403] vi) a light chain CDR comprising the amino acid sequence of SEQ ID NO:31 (CDR1), SEQ ID NO:32 (CDR2) and SEQ ID NO:33 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO:34 (CDR1), SEQ ID NO:35 (CDR2) and SEQ ID NO:36 (CDR3); or

[0404] vii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 133 (CDR1), SEQ ID NO: 134 (CDR2) and SEQ ID NO: 135 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 136 (CDR1), SEQ ID NO: 137 (CDR2) and SEQ ID NO: 138 (CDR3); or

[0405] viii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 139 (CDR1), SEQ ID NO: 140 (CDR2) and SEQ ID NO: 141 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 142 (CDR1), SEQ ID NO: 143 (CDR2) and SEQ ID NO: 144 (CDR3); or

[0406] ix) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 145 (CDR1), SEQ ID NO: 146 (CDR2) and SEQ ID NO: 147 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 148 (CDR1), SEQ ID NO: 149 (CDR2) and SEQ ID NO: 150 (CDR3); or

[0407] x) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 151 (CDR1), SEQ ID NO: 152 (CDR2) and SEQ ID NO: 153 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 154 (CDR1), SEQ ID NO: 155 (CDR2) and SEQ ID NO: 156 (CDR3); or

[0408] xi) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 157 (CDR1), SEQ ID NO: 158 (CDR2) and SEQ ID NO: 159 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 160 (CDR1), SEQ ID NO: 161 (CDR2) and SEQ ID NO: 162 (CDR3); or

[0409] xii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 163 (CDR1), SEQ ID NO: 164 (CDR2) and SEQ ID NO: 165 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 166 (CDR1), SEQ ID NO: 167 (CDR2) and SEQ ID NO: 168 (CDR3); or

[0410] xiii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 169 (CDR1), SEQ ID NO: 170 (CDR2) and SEQ ID NO: 171 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 172 (CDR1), SEQ ID NO: 173 (CDR2) and SEQ ID NO: 174 (CDR3); or

[0411] xiv) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 175 (CDR1), SEQ ID NO: 176 (CDR2) and SEQ ID NO: 177 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 178 (CDR1), SEQ ID NO: 179 (CDR2) and SEQ ID NO: 180 (CDR3); or

[0412] xv) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 181 (CDR1), SEQ ID NO: 182 (CDR2) and SEQ ID NO: 183 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 184 (CDR1), SEQ ID NO: 185 (CDR2) and SEQ ID NO: 186 (CDR3); or

[0413] xvi) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 187 (CDR1), SEQ ID NO: 188 (CDR2) and SEQ ID NO: 189 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 190 (CDR1), SEQ ID NO: 191 (CDR2) and SEQ ID NO: 192 (CDR3); or

[0414] xvii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 193 (CDR1), SEQ ID NO: 194 (CDR2) and SEQ ID NO: 195 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 196 (CDR1), SEQ ID NO: 197 (CDR2) and SEQ ID NO: 198 (CDR3); or

[0415] xviii) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 199 (CDR1), SEQ ID NO: 200 (CDR2) and SEQ ID NO: 201 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 202 (CDR1), SEQ ID NO: 203 (CDR2) and SEQ ID NO: 204 (CDR3).

[0416] Antibodies i) to xviii) as outlined above are respectively referred to as DLL3#1, DLL3#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17 or DLL3#18. A sequence listing is provided herein which allows easy identification of individual amino acid sequences of specific antibodies of the invention.

[0417] In a preferred embodiment of the invention, the antibody molecule comprises a CDR sequence as defined in i) or v) above corresponding to DLL3#1 or DLL3#5.

[0418] In some embodiments, the anti-DLL3 antibodies of the invention are chimeric, humanized or human antibody molecules. In some embodiments, the antibody molecule is a monoclonal antibody Fab, F(ab)2, Fv or scFv. In some embodiments, the anti-DLL3 antibody molecules of the invention comprise a heavy chain constant region selected from the group consisting of: IgG1, IgG2, IgG3, IgG4, IgM, IgA and IgE constant regions. In some embodiments, the light chain constant region of the anti-DLL3 antibody molecules of the invention is kappa or lambda.

[0419] In some embodiments, an anti-DLL3 antibody of the invention has a heavy chain variable domain comprising an amino acid sequence at least 85% identical to any one of SEQ ID NO:38, 40, 42, 44, 46, 48, 206, 208, 210, 212, 214, 216, 218, 220, 222, 224, 226, and 228. Preferably, the antibody molecule has a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:38, 40, 42, 44, 46, 48, 206, 208, 210, 212, 214, 216, 218, 220, 222, 224, 226, or 228.

[0420] In some embodiments, the anti-DLL3 antibody molecule has a light chain variable domain comprising an amino acid sequence at least 85% identical to any one of SEQ ID NO: 37, 39, 41, 43, 45, 47, 205, 207, 209, 211, 213, 215, 217, 219, 221, 223, 225, and 227. Preferably, the antibody molecule has a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 37, 39, 41, 43, 45, 47, 205, 207, 209, 211, 213, 215, 217, 219, 221, 223, 225, or 227.

[0421] Methods for calculating amino acid sequence identity are well known in the art and are further discussed in the definitions section of the specification herein.

[0422] In some embodiments, the anti-DLL3 antibody molecule has i) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:38 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO:37 (DLL3#1); or ii) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:40 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO:39 (DLL3#2); or iii) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:42 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO:41 (DLL3#3); or iv) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:44 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO:43 (DLL3#4); or v) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:46 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO:45 (DLL3#5); or vi) a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:48 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO: NO:47 (DLL3#6); ​​or vii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:205 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:206 (DLL3#7); or viii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:207 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:208 (DLL3#8); or ix) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:209 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:210 (DLL3#9); or x) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:211 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:212 (DLL3#10); or xi) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:213 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:214 (DLL3#11); or xii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:214 NO:215 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:216 (DLL3#12); or xiii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:217 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:218 (DLL3#13); or xiv) a light chain variable domain comprising the amino acid sequence of SEQ ID NO:219 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:220 (DLL3#14);or xv) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 221 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 222 (DLL3#15); or xvi) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 223 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 224 (DLL3#16); or xvii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 225 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 226 (DLL3#17); or xviii) a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 227 and a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 228 (DLL3#18). ;

[0423] In some embodiments, the anti-DLL3 antibodies of the present invention are mouse monoclonal antibodies. In the context of the present invention, mouse monoclonal antibodies include antibodies wherein the VH and VL are obtained by immunizing mice with human DLL3 protein, followed by selection of suitable VH and VL sequences that bind to human DLL3 with a certain affinity, and then further linking such VH and VL sequences to constant domains derived from mice (e.g., mouse IgG2a) by recombinant technology; and which are made by recombinant expression in host cells. The present invention further encompasses chimeric antibodies, e.g., comprising variable and constant regions from different species. In some embodiments, the antibody molecules of the present invention are chimeric antibodies comprising VH and VL domains derived from mice as described above and further comprising constant domains derived from another species, such as human, rabbit, rat, goat, donkey. In some embodiments, chimeric antibodies comprise VH and VL domains derived from mice and further humanized or sequence optimized as defined above and further comprising constant domains derived from another species. In some embodiments, a chimeric antibody comprises VH and VL domains derived from a transgenic animal (e.g., a mouse) comprising human IgG sequences, thus comprising human VH and VL sequences, and further comprising a constant domain derived from another species. In any of the embodiments of chimeric antibodies as outlined above, the heavy chain constant region is a mouse, human, rabbit, rat, goat, or donkey heavy chain region.

[0424] In some embodiments, the anti-DLL3 antibody molecules of the invention have a constant domain selected from the group consisting of IgG1, IgG2, IgG3, IgG4, IgM, IgA and IgE constant domains. In a preferred embodiment, the anti-DLL3 antibody has a constant domain of IgG2a, which preferably comprises the sequence of SEQ ID NO: 254. In some embodiments, the anti-DLL3 antibody molecules have a light chain constant domain that is kappa or lambda, preferably the light chain constant domain is a kappa light chain constant domain, which preferably comprises the sequence of SEQ ID NO: 255.

[0425] In some embodiments, the anti-DLL3 antibody molecule is capable of binding to human and cynomolgus monkey DLL3 with a dissociation constant (KD) preferably ≤ 10 nM, more preferably ≤ 1 nM, even more preferably ≤ 0.1 nM, even more preferably ≤ 0.01 nM. Affinity (KD value) can be measured using recombinant DLL3 protein in SPR (BIAcore) analysis, as described, for example, in the Examples or other methods known to the skilled person.

[0426] In some embodiments, the anti-DLL3 antibody molecule does not bind to mouse DLL3.

[0427] In some embodiments, the anti-DLL3 antibody molecule is capable of detecting DLL3 expression (e.g., cytoplasmic and surface protein expression) on a tissue sample (e.g., paraffin-embedded / formalin-fixed tissue sample), such as a tumor tissue sample or a cultured cell line. Optionally, the paraffin-embedded / formalin-fixed cell culture or tissue sample is further treated with an epitope retriever (e.g., proteinase K).

[0428] In some embodiments, the anti-DLL3 antibody molecules of the invention bind to the same epitope as any of the following: DLL3#1, DLL2#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, or DLL3#18 antibodies. In some embodiments, an anti-DLL3 antibody molecule of the invention cross-competes for binding to human DLL3 with any of the following: DLL3#1, DLL2#2, DLL3#3, DLL3#4, DLL3#5, DLL3#6, DLL3#7, DLL3#8, DLL3#9, DLL3#10, DLL3#11, DLL3#12, DLL3#13, DLL3#14, DLL3#15, DLL3#16, DLL3#17, or DLL3#18 antibody molecules.

[0429] Another aspect of the invention provides isolated nucleic acid molecules encoding the heavy chain variable domain and / or the light chain variable domain of the anti-DLL3 antibody molecules of the invention.

[0430] Preferably, the nucleic acid molecule comprises a nucleotide sequence encoding the heavy chain variable domain of any one of SEQ ID NO: 38, 40, 42, 44, 46, 48, 206, 208, 210, 212, 214, 216, 218, 220, 222, 224, 226 or 228. Preferably, the nucleic acid molecule comprises a nucleotide sequence encoding the light chain variable domain of any one of SEQ ID NO: 37, 39, 41, 43, 45, 47, 205, 207, 209, 211, 213, 215, 217, 219, 221, 223, 225 or 227.

[0431] Another aspect of the present invention provides an expression vector containing a DNA molecule comprising a nucleotide sequence encoding the heavy chain variable domain and / or light chain variable domain of the anti-DLL3 antibody molecule of the present invention.

[0432] Preferably, the expression vector further comprises a nucleic acid molecule, preferably a DNA molecule, encoding a heavy chain constant domain and / or a light chain constant domain, respectively, linked to a nucleic acid molecule, preferably a DNA molecule encoding a heavy chain variable domain and / or a light chain variable domain, respectively.

[0433] In a particularly preferred embodiment, two expression vectors can be used, one for expressing the heavy chain and the other for expressing the light chain, both of which can then be transfected into host cells for recombinant protein expression.

[0434] Preferably, the expression vector will be a vector comprising the or said nucleic acid molecule operably linked to at least one regulatory sequence, wherein such regulatory sequence may be a promoter, enhancer or terminator sequence, and most preferably a heterologous promoter, enhancer or terminator sequence.

[0435] In another aspect, the invention relates to a host cell having an expression vector encoding the heavy chain of an anti-DLL3 antibody molecule of the invention and an expression vector encoding the light chain of an anti-DLL3 antibody molecule of the invention.

[0436] According to a particularly preferred embodiment, the host cell is a eukaryotic cell, such as a mammalian cell. In another embodiment, such a host cell is a bacterial cell. Other useful cells are yeast cells or other fungal cells.

[0437] Suitable mammalian cells include, for example, CHO cells, BHK cells, HeLa cells, COS cells and the like. However, amphibian cells, insect cells, plant cells and any other cells used in the art to express heterologous proteins may also be used.

[0438] Manufacturing and purification methods

[0439] The invention further provides methods of making the multispecific binding proteins of the invention, the methods generally comprising the steps of:

[0440] - culturing a host cell under conditions that allow the formation of a binding protein of the invention, said host cell comprising an expression vector comprising a nucleic acid encoding a binding protein of the invention; and,

[0441] - recovering the binding protein expressed by the host cells from the culture; and

[0442] - Optionally further purifying and / or modifying and / or formulating the binding protein of the invention.

[0443] The present invention further provides methods of making an anti-DLL3 antibody of the present invention, the methods generally comprising the following steps:

[0444] - culturing a host cell under conditions that allow the formation of antibody molecules, said host cell comprising an expression vector comprising a nucleic acid encoding an antibody molecule of the invention; and,

[0445] - recovering the antibody molecules expressed by the host cells from the culture; and

[0446] - Optionally further purifying and / or modifying and / or formulating the antibody molecule of the invention.

[0447] The nucleic acid of the present invention may be, for example, a DNA molecule comprising a coding sequence and a regulatory sequence and optionally selected natural or artificial introns or may be a cDNA molecule. It may have its original codons or may have an optimized codon usage, which is particularly suitable for expression in a desired host cell or host organism. According to one embodiment of the present invention, as defined above, the nucleic acid of the present invention is in a substantially isolated form.

[0448] The nucleic acids of the invention can be prepared or obtained in a manner known per se (eg by automated DNA synthesis and / or recombinant DNA technology) based on the information given herein on the amino acid sequences of the proteins of the invention.

[0449] The nucleic acids of the invention will typically be incorporated into an expression vector, that is, a vector that can provide for protein expression when transfected into a suitable host cell or other expression system.

[0450] To produce the binding proteins or antibodies of the invention, the skilled artisan can choose from a large number of expression systems well known in the art, such as those reviewed by Kipriyanow and Le Gall, 2004.

[0451] Expression vectors include plasmids, retroviruses, musculosomes, EBV-derived episomes, and the like. Expression vectors and expression control sequences are selected to be compatible with host cells. The nucleotide sequence encoding the first antigen binding unit (e.g., DLL3-specific single-chain Fab or full-length DLL3 chain of the binding protein of the present invention) and the nucleotide sequence encoding the second antigen binding unit (e.g., CD3-specific single-chain Fab or full-length CD3 chain of the binding protein of the present invention) of the DLL3 / CD3 binding protein can be inserted into separate vectors. In certain embodiments, the two DNA sequences are inserted into the same expression vector. The nucleotide sequence encoding the light chain of the DLL3 antibody and the nucleotide sequence encoding the heavy chain of the DLL3 antibody can be inserted into separate vectors. In certain embodiments, the two DNA sequences are inserted into the same expression vector.

[0452] Suitable vectors are those encoding functional fully human CH (constant heavy) immunoglobulin sequences, with suitable restriction sites engineered to allow easy insertion and expression of any antigen binding unit as described above, such as a single chain Fab sequence or any heavy / light chain variable domain. For the antibody heavy chain, it may be, but is not limited to, any IgG isotype (IgG1, IgG2, IgG3, IgG4) or other immunoglobulins, including allele variants.

[0453] The recombinant expression vector may also encode a signal peptide that promotes secretion of full-length CD3 or DLL3 chains or secretion of light / heavy chains of anti-DLL3 antibodies by the host cell. The DNA encoding the protein chain can be cloned into the vector such that the signal peptide is linked in-frame to the amino terminus of the mature full-length chain DNA. The signal peptide may be an immunoglobulin signal peptide or a heterologous signal peptide from a non-immunoglobulin protein. Alternatively, the DNA sequence encoding the full-length chain of the protein of the invention may already contain a signal peptide sequence.

[0454] In addition to the DLL3 / CD3 chain encoding DNA sequence or the heavy / light chain encoding DNA sequence of the DLL3 antibody, the recombinant expression vector usually carries regulatory sequences, optionally heterologous regulatory sequences, including promoters, enhancers, termination and polyadenylation signals and other expression control components that control the expression of the protein chain in the host cell. Examples of promoter sequences (exemplified for expression in mammalian cells) are promoters and / or enhancers derived from CMV (such as the CMV Simian Virus 40 (SV40) promoter / enhancer), adenovirus (e.g., the adenovirus major late promoter (AdMLP)), polyoma virus, and strong mammalian promoters, such as the natural immunoglobulin and actin promoters. Examples of polyadenylation signals are BGH polyA, SV40 late or early polyA; alternatively, the 3′UTR of an immunoglobulin gene, etc., may be used.

[0455] The recombinant expression vector may also carry sequences regulating vector replication in the host cell (e.g., replication origin) and selectable marker genes. The following may be introduced into a host cell, such as a bacterial cell or a higher eukaryotic cell, such as a mammalian cell, according to transfection methods well known in the art (including liposome-mediated transfection, polycation-mediated transfection, protoplast fusion, microinjection, calcium phosphate precipitation, electroporation, or transfer by viral vectors): nucleic acid molecules encoding a full-length chain of a first antigen binding unit (single-chain Fab and Fc domains) or an antigen binding portion thereof and / or a full-length chain of a second antigen binding unit (single-chain Fab and Fc domains) or an antigen binding portion thereof, and vectors comprising these DNA molecules.

[0456] Preferably, the DNA molecules encoding the DLL3 and CD3 chains of the protein of the present invention are present on two expression vectors, and the expression vectors are co-transfected into a host cell, preferably a mammalian cell.

[0457] Mammalian cell lines that can be used as hosts for expression are well known in the art and include Chinese hamster ovary (CHO) cells, NS0, SP2 / 0 cells, HeLa cells, baby hamster kidney (BHK) cells, monkey kidney cells (COS), human cancer cells (e.g., Hep G2 and A-549 cells), 3T3 cells, or derivatives / progeny of any such cell lines. Other mammalian cells, including but not limited to human, mouse, rat, monkey, and rodent cell lines, or other eukaryotic cells, including but not limited to yeast, insect, and plant cells, or prokaryotic cells, such as bacteria, can be used.

[0458] The protein of the present invention is prepared by culturing a host cell for a period of time sufficient to allow expression of the protein in the host cell. The protein molecule is preferably recovered from the culture medium in the form of secreted polypeptides or, if, for example, expressed without a secretion signal, it can be recovered from the host cell lysate. It is necessary to use the standard protein purification method for recombinant proteins and host cell proteins to obtain a substantially homogeneous preparation mode purification of protein molecules. By means of an embodiment, the most advanced purification method suitable for obtaining the protein molecule of the present invention includes removing cells and / or granular cell fragments from the culture medium or lysate as a first step. Subsequently, for example, by fractionation on an immunoaffinity or ion exchange column, ethanol precipitation, reversed phase HPLC, dextran gel chromatography, silica gel chromatography, or on a cation exchange resin, purifying the protein from pollutant soluble proteins, polypeptides, and nucleic acids. As the final step in the process for obtaining a protein molecule preparation, it can be dried, for example, lyophilized purified protein molecules, as described in the following therapeutic applications.

[0459] The present invention relates to binding proteins having binding specificities for at least two different targets. With respect to the present invention, the binding molecules are derived from antibodies. Techniques for making binding molecules include, but are not limited to, recombinant co-expression of two immunoglobulin chains with different specificities (see Milstein and Cuello, Nature 305:537 (1983)), WO 93 / 08829, and Traunecker et al., EMBO J. 10:3655 (1991)), and "knob-in-hole" engineering (see, e.g., U.S. Pat. No. 5,731,168; Atwell et al., JMB, 1997, 270, 26-35). The binding proteins of the present invention can also be prepared as follows: Engineered electrostatic director effects for preparing antibody Fc-heterodimers (WO 93 / 08829, and Traunecker et al., EMBO J. 10:3655 (1991)). 2009 / 089004A1); cross-linking two or more antibodies or fragments (see, e.g., U.S. Pat. No. 4,676,980 and Brennan et al., Science, 229:81 (1985)); using leucine zippers to produce bispecific proteins (see, e.g., Kostelny et al., Immunol., 148(5):1547-1553 (1992)); using "bifunctional" antibodies for making bispecific antibody fragments and preparation of trispecific antibodies as described in, e.g., Tutt et al., J. Immunol. 147:60 (1991).

[0460] The compositions (e.g., multispecific binding proteins and anti-DLL3 antibodies) and methods disclosed herein encompass polypeptides and nucleic acids having a specified sequence, or a sequence that is substantially identical or similar to a specified sequence, such as a sequence that is at least 85%, 90%, 95% identical or higher to a specified sequence. In the context of amino acid sequences, the term "substantially identical" is used herein to refer to a first amino acid sequence that contains a sufficient or minimum number of amino acid residues that are i) identical to the aligned amino acid residues in the second amino acid sequence or ii) conservative substitutions for the aligned amino acid residues in the second amino acid sequence such that the first and second amino acid sequences may have a common domain and / or a common functional activity. For example, an amino acid sequence containing a common domain has at least about 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identity to a reference sequence, such as a sequence provided herein. In the context of nucleotide sequences, the term "substantially identical" is used herein to refer to a first nucleic acid sequence that contains a sufficient or minimum number of nucleotides that are identical to the aligned nucleotides in a second nucleic acid sequence such that the first and second nucleotide sequences encode polypeptides having a common functional activity, or encode a common structural polypeptide domain or a common functional polypeptide activity, such as a nucleotide sequence having at least about 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identity to a reference sequence.

[0461] Nucleic acid molecules of the present invention include, but are not limited to, DNA molecules encoding the polypeptide sequences shown in the sequence listing. In addition, the present invention also relates to nucleic acid molecules that hybridize with DNA molecules encoding the polypeptide sequences shown in the sequence listing under high stringency binding and washing conditions, as defined in WO 2007 / 042309. Preferred molecules (from the perspective of mRNA) are those having at least 75% or 80% (preferably at least 85%, more preferably at least 90% and optimally at least 95%) homology or sequence identity with one of the DNA molecules described herein. By way of example, in view of expressing antibodies in eukaryotic cells, the DNA sequences shown in the sequence listing have been designed to match the codon usage in eukaryotic cells. If it is desired to express antibodies in E. coli, these sequences can be altered to match the E. coli codon usage. Variants of the DNA molecules of the present invention can be constructed in several different ways, as described, for example, in WO 2007 / 042309.

[0462] The proteins of the invention may have a modified N-terminal sequence, such as deletion of one or more N-terminal amino acids, or exchange of, for example, the first N-terminal amino acid (e.g., glutamate for alanine), to optimize the molecule for expression using certain expression systems (such as specific vectors or host cells), or for expression in inclusion bodies or soluble form, or for secretion into the culture medium or periplasmic space or contained within the cell, or for obtaining a more homogeneous product. The polypeptides of the invention may have a modified C-terminal sequence, such as additional alanine, and / or other amino acid exchanges at other defined positions in the C-terminal portion or in any of the framework regions, as explained, for example, in WO2012 / 175741, WO2011 / 075861 or WO2013 / 024059, in order, for example, to further enhance the stability of such polypeptides or to reduce the immunogenicity of such polypeptides.

[0463] For the avoidance of doubt, all embodiments relating to pharmaceutical compositions, kits, methods of treatment, medical uses, combinations, methods of administration, and dosages as described herein are contemplated for use with any of the multispecific binding proteins described herein, alone or in combination with other therapeutic agents (as specified in more detail below).

[0464] Pharmaceutical composition; Method of administration; Dosage

[0465] The present invention further relates to pharmaceutical compositions for treating a disease (as specified in more detail below), wherein such compositions comprise at least one multispecific binding protein of the present invention. The present invention further encompasses methods of treating a disease (as specified in more detail below) using at least one multispecific protein of the present invention or a pharmaceutical composition as set forth below, and further encompasses the preparation of a medicament for treating such a disease using such a binding protein or pharmaceutical composition of the present invention.

[0466] Depending on the specific pharmaceutical formulation or composition to be used, the binding proteins of the invention (e.g., DLL3#1 / CD3#1, DLL3#2 / CD3#1, DLL3#3 / CD3#1, DLL3#4 / CD3#1, DLL3#5 / CD3#1, DLL3#6 / CD3#1, DLL3#7 / CD3#1, DLL3#8 / CD3#1, DLL3#9 / CD3#1, DLL3#10 / CD3#1, DLL3#11 / CD3#1, DLL3#12 / CD3#1, DLL3#13 / CD3#1, DLL3#14 / CD3#1) can be administered to a patient in need thereof in any suitable manner. D3#1, DLL3#15 / CD3#1, DLL3#16 / CD3#1, DLL3#17 / CD3#1, DLL3#18 / CD3#1, DLL3#1 / CD3#2, DLL3#2 / CD3#2, DLL3#3 / CD3#2, DLL3#4 / CD3#2, D LL3#5 / CD3#2, DLL3#6 / CD3#2, DLL3#7 / CD3#2, DLL3#8 / CD3#2, DLL3#9 / CD3#2, DLL3#10 / CD3#2, DLL3#11 / CD3#2, DLL3#12 / CD3#2, DLL3#13 / C D3#2, DLL3#14 / CD3#2, DLL3#15 / CD3#2, DLL3#16 / CD3#2, DLL3#17 / CD3#2, DLL3#18 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#3, DLL3#3 / CD3#3, DLL3#4 / CD3#3, DLL3#5 / CD3#3 and DLL3#6 / CD3#3, DLL3#7 / CD3#3, DLL3#8 / CD3#3, DLL3#9 / CD3#3, DLL3#10 / CD3#3, DLL3#11 / CD3#3, DLL3#12 / C Any one of DLL3#1 / CD3#3, DLL3#13 / CD3#3, DLL3#14 / CD3#3, DLL3#15 / CD3#3, DLL3#16 / CD3#3, DLL3#17 / CD3#3, and DLL3#18 / CD3#3, preferably any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, and DLL3#3 / CD3#3) and / or a composition comprising the same.Therefore, the binding protein of the present invention and / or the composition comprising the same can be administered in an effective amount or dosage, for example, in the following manner: intravenous (iv), subcutaneous (sc), intramuscular (im), intraperitoneal (ip), transdermal, oral, sublingual (e.g., in the form of a sublingual lozenge, spray or drops placed under the tongue and absorbed through the mucosa into the capillary network under the tongue), nasal (intranasal) (e.g., in the form of a nasal spray and / or in the form of an aerosol), topically, by means of a suppository, by inhalation or any other suitable means. The binding protein can be administered by infusion, bolus or injection. In a preferred embodiment, administration is by intravenous infusion or subcutaneous injection.

[0467] According to the treatment course suitable for treating and / or alleviating the disease to be treated or alleviated, illness or condition, the binding protein of the present invention and / or the composition comprising it are given. Clinicians will generally be able to determine a suitable treatment regimen depending on factors such as the disease to be treated or alleviated, illness or condition, the severity of the disease, the severity of its symptoms, the specific binding protein of the present invention to be used, the specific route of administration and the pharmaceutical formulation to be used or the composition, the patient's age, sex, weight, diet, general conditions, and similar factors known to clinicians. In general, the treatment regimen will include administering one or more binding proteins of the present invention with a therapeutically effective amount or dosage, or one or more compositions comprising it.

[0468] In general, for the treatment and / or alleviation of the diseases, disorders and conditions mentioned herein and depending on the specific disease, disorder or condition to be treated, the efficacy of the specific binding protein of the present invention to be used, the specific route of administration and the specific pharmaceutical formulation or composition used, the binding protein of the present invention will generally be in an amount and dosage between 0.005 and 20.0 mg per kilogram of body weight, preferably between 0.05 and 10.0 mg / kg / dose, continuously (e.g., by infusion) or more preferably in a single dose (such as twice a week, once a week or once a month; see below), but may vary significantly, especially depending on the parameters mentioned above. Therefore, in some cases, it may be sufficient to use less than the minimum dose given above, while in other cases the upper limit may be exceeded. When a larger amount is given, it may be desirable to divide it into multiple smaller doses in one day.

[0469] Depending on the specific binding protein of the present invention and its specific pharmacokinetic and other properties, it can be administered daily, every two days, every three days, every four days, every five days or every six days, weekly, monthly and in a similar manner. The administration course may include long-term treatment once a week. "Long-term" means a duration of at least two weeks and preferably several months or years.

[0470] The efficacy of the multispecific proteins of the invention and compositions comprising the same may be tested using any suitable in vitro assay, cell-based assay, in vivo assay, and / or animal model known per se, or any combination thereof, depending on the particular disease involved. Suitable assays and animal models will be clear to the skilled artisan, and include, for example, those used in the Examples below.

[0471] Preparation

[0472] For pharmaceutical uses, the binding protein of the present invention can be formulated into a pharmaceutical preparation, which comprises (i) at least one binding protein of the present invention (e.g., DLL3#1 / CD3#1, DLL3#2 / CD3#1, DLL3#3 / CD3#1, DLL3#4 / CD3#1, DLL3#5 / CD3#1, DLL3#6 / CD3#1, DLL3#7 / CD3#1, DLL3#8 / CD3#1, DLL3#9 / CD3#1, DLL3#10 / CD3#1, DLL3#11 / CD3#1, DLL3#12 / CD3#1, DLL3#13 / CD3#1, DLL3#14 / CD3#1, DLL3#15 / CD 3#1, DLL3#16 / CD3#1, DLL3#17 / CD3#1, DLL3#18 / CD3#1, DLL3#1 / CD3#2, DLL3#2 / CD3#2, DLL3#3 / CD3#2, DLL3#4 / CD3#2, DLL3#5 / CD3#2, DLL3#6 / CD3 #2, DLL3#7 / CD3#2, DLL3#8 / CD3#2, DLL3#9 / CD3#2, DLL3#10 / CD3#2, DLL3#11 / CD3#2, DLL3#12 / CD3#2, DLL3#13 / CD3#2, DLL3#14 / CD3#2, DLL3#15 / CD 3#2, DLL3#16 / CD3#2, DLL3#17 / CD3#2, DLL3#18 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#3, DLL3#3 / CD3#3, DLL3#4 / CD3#3, DLL3#5 / CD3#3 and DLL3#6 / CD3 #3, DLL3#7 / CD3#3, DLL3#8 / CD3#3, DLL3#9 / CD3#3, DLL3#10 / CD3#3, DLL3#11 / CD3#3, DLL3#12 / CD3#3, DLL3#13 / CD3#3, DLL3#14 / CD3#3, DLL3#15 / CD DLL3#3 / CD3#3, DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) and (ii) at least one pharmaceutically acceptable carrier, diluent, excipient, adjuvant and / or stabilizer, and (iii) optionally one or more other pharmacologically active polypeptides and / or compounds."Pharmaceutically acceptable" means that when administered to an individual, the corresponding substance does not exhibit any biological or undesirable effects and does not interact in a harmful manner with any of the other components (such as pharmaceutically active ingredients) contained in the pharmaceutical composition. Specific embodiments can be found in standard manuals, such as Remington's Pharmaceutical Sciences, 18th edition, Mack Publishing Company, USA (1990). For example, the binding protein of the present invention can be formulated and administered in any manner known per se for conventional antibodies and antibody fragments and other pharmaceutically active proteins. Therefore, according to another embodiment, the present invention relates to a pharmaceutical composition or formulation containing at least one binding protein of the present invention and at least one pharmaceutically acceptable carrier, diluent, excipient, adjuvant and / or stabilizer, and optionally one or more other pharmacologically active substances, in the form of a lyophilized or otherwise dried formulation or an aqueous or non-aqueous solution or suspension.

[0473] Pharmaceutical preparations for parenteral administration, such as intravenous, intramuscular, subcutaneous injection or intravenous infusion, may, for example, be sterile solutions, suspensions, dispersions, emulsions or powders containing the active ingredient and, optionally after another dissolution or dilution step, suitable for infusion or injection. Suitable carriers or diluents for the preparations include, but are not limited to, sterile water and pharmaceutically acceptable buffered aqueous solutions and solutions such as phosphate-buffered saline, Ringer's solution, dextrose solution and Hank's solution; water oil; glycerol; ethanol; glycols such as propylene glycol and mineral oils, animal oils and vegetable oils, such as peanut oil, soybean oil and suitable mixtures thereof.

[0474] The solution of the binding protein of the present invention may also contain preservatives, such as antibacterial and antifungal agents to prevent microbial growth, for example, p-hydroxybenzoate / paraben, chlorobutanol, phenol, sorbic acid, thimerosal, ethylenediaminetetraacetic acid (alkali metal salts) and the like. In many cases, it will be preferred to include isotonic agents, for example, sugars, buffers or sodium chloride. Optionally, emulsifiers and / or dispersants may be used. Appropriate fluidity can be maintained, for example, by forming liposomes, by maintaining the desired particle size in the case of dispersions, or by using surfactants. Other agents that delay absorption, such as aluminum monostearate and gelatin, may also be added. The solution can be filled into injection vials, ampoules, infusion bottles and similar bottles thereof.

[0475] In all cases, the final dosage form must be sterile, fluid and stable under the conditions of manufacture and storage. Sterile injectable solutions are prepared by incorporating the required amount of the active compound in an appropriate solvent, optionally along with the various other ingredients listed above, followed by filtered sterilization. In the case of sterile powders for the preparation of sterile injectable solutions, the preferred methods of preparation are vacuum drying and freeze drying techniques, which produce a powder of the active ingredient plus any additional desired ingredients present in a previously sterile-filtered solution.

[0476] Typically, aqueous solutions or suspensions will be preferred. In general, suitable formulations for therapeutic proteins (such as the binding proteins of the present invention) are buffered protein solutions, such as solutions comprising: proteins at a suitable concentration (such as 0.001 to 400 mg / mL, preferably 0.005 to 200 mg / mL, more preferably 0.01 to 200 mg / mL, more preferably 1.0-100 mg / mL, such as 1.0 mg / ml (intravenous administration) or 100 mg / ml (subcutaneous administration)) and an aqueous buffer such as:

[0477] - Phosphate buffered saline, pH 7.4,

[0478] -Other phosphate buffers, pH 6.2 to pH 8.2,

[0479] - Acetate buffer, pH 3.2 to pH 7.5, preferably pH 4.8 to pH 5.5

[0480] -Histidine buffer, pH 5.5 to pH 7.0,

[0481] - Succinate buffer, pH 3.2 to pH 6.6, and

[0482] - Citrate buffer, pH 2.1 to pH 6.2,

[0483] and optionally salts (e.g. NaCl) and / or sugars (e.g. sucrose and trehalose) and / or other polyols (e.g. mannitol and glycerol) selected to provide isotonicity of the solution.

[0484] In addition, other agents, such as detergents, for example 0.02% Tween-20 or Tween-80, may be included in the solution. Subcutaneously administered formulations may include significantly higher concentrations, such as up to 100 mg / ml or even greater than 100 mg / ml of the antibodies of the invention. However, it will be clear to those skilled in the art that the ingredients and amounts thereof given above represent only a preferred choice. Alternatives and variants thereof will be immediately apparent to those skilled in the art, or may be easily conceived starting from the above disclosure. The above formulations may optionally be provided as lyophilized formulations to be reconstituted in a solution, such as water for injection (WFI).

[0485] According to another aspect of the invention, the binding proteins of the invention can be used in combination with a device suitable for administering the protein, such as a syringe, injection pen, micropump or other device.

[0486] Treatment

[0487] Another aspect of the present invention provides a method for treating cancer, comprising administering a therapeutically effective amount of a binding protein of the present invention to a patient in need thereof.

[0488] Another aspect of the invention provides a binding protein of the invention for use in a method of treating cancer.

[0489] Another aspect of the present invention is the use of the binding protein of the present invention for preparing a pharmaceutical composition for treating cancer.

[0490] For the avoidance of doubt, the medical use aspects of the present invention may comprise any one of the specific binding proteins of the present invention as described above (e.g. DLL3#1 / CD3#1, DLL3#2 / CD3#1, DLL3#3 / CD3#1, DLL3#4 / CD3#1, DLL3#5 / CD3#1, DLL3#6 / CD3#1, DLL3#7 / CD3#1, DLL3#8 / CD3#1, DLL3#9 / CD3#1, DLL3#10 / CD3#1, DLL3#11 / CD3#1, DLL3#12 / CD3#1, DLL3#13 / CD3#1, DLL3#14 / CD3#1, DLL3#15 / CD3#1, DLL3#16 / CD3#1, DLL3#17 / CD3#1, DLL3#18 / CD3#1, DLL3#19 / CD3#1, DLL3#20 / CD3#1, DLL3#21 / CD3#1, DLL3#22 / CD3#1, DLL3#23 / CD3#1, DLL3#24 / CD3#1, DLL3#25 / CD3#1, DLL3#26 / CD3#1, DLL3#27 / CD3#1, DLL3#28 / CD3#1, DLL3#29 / CD3#1, DLL3#30 / CD3#1, DLL3#31 / CD3#1, DLL3#32 / CD3#1, DLL3#33 / CD3#1, DLL3#34 / CD3#1, D3#1, DLL3#15 / CD3#1, DLL3#16 / CD3#1, DLL3#17 / CD3#1, DLL3#18 / CD3#1, DLL3#1 / CD3#2, DLL3#2 / CD3#2, DLL3#3 / CD3#2, DLL3#4 / CD3#2, DLL3#5 / CD3#2, DLL3#6 / CD3#2, DLL3#7 / CD3#2, DLL3#8 / CD3#2, DLL3#9 / CD3#2, DLL3#10 / CD3#2, DLL3#11 / CD3#2, DLL3#12 / CD3#2, DLL3#1 3 / CD3#2, DLL3#14 / CD3#2, DLL3#15 / CD3#2, DLL3#16 / CD3#2, DLL3#17 / CD3#2, DLL3#18 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#3, DLL3#3 / CD 3#3, DLL3#4 / CD3#3, DLL3#5 / CD3#3 and DLL3#6 / CD3#3, DLL3#7 / CD3#3, DLL3#8 / CD3#3, DLL3#9 / CD3#3, DLL3#10 / CD3#3, DLL3#11 / CD3#3, DLL Any one of DLL3#12 / CD3#3, DLL3#13 / CD3#3, DLL3#14 / CD3#3, DLL3#15 / CD3#3, DLL3#16 / CD3#3, DLL3#17 / CD3#3, DLL3#18 / CD3#3, preferably any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3).

[0491] As used herein, the term "cancer" is intended to include all types of neoplastic growth or oncogenic processes, metastatic tissues or malignantly transformed cells, tissues or organs, regardless of histopathological type or stage of invasiveness.

[0492] Exemplary cancers whose growth can be inhibited using the multispecific binding proteins described herein are any tumor expressing DLL3, preferably SCLC, LCNEC, glioma, glioblastoma, melanoma or other neuroendocrine tumors expressing DLL3 (e.g., neuroendocrine prostate cancer or neuroendocrine pancreatic cancer, small cell bladder cancer).

[0493] Neuroendocrine tumors (NETs) are caused by the dispersed endocrine system and can occur in many different body regions. Traditionally, NETs have been classified according to their original anatomical units and are usually highly invasive. They are most commonly located in the gastrointestinal tract, pancreas or lungs (small cell lung carcinomas and large cell neuroendocrine carcinomas) and kidneys or urogenital tracts (air bladder, prostate, ovaries, cervix and endometrium). NETs include certain tumors of the gastrointestinal tract and pancreatic islet cells, certain thymus and lung tumors, and medullary carcinoma of parafollicular cells of the thyroid gland.

[0494] Other cancers whose growth may be inhibited using the multispecific binding proteins described herein are pseudo neuroendocrine tumors (pNETs), which share certain genotypic, expression, or biochemical features with traditionally defined neuroendocrine tumors.

[0495] In some embodiments, the following cancers, tumors, and other proliferative diseases can be treated with the multispecific binding proteins of the invention: lung cancer; preferably SCLC, NSCLC, or LCNEC; breast cancer; cervical cancer; colon cancer; colorectal cancer; endometrial cancer; head and neck cancer; liver cancer (hepatoblastoma or hepatocellular carcinoma); ovarian cancer; pancreatic cancer; prostate cancer; skin cancer; gastric cancer; testicular cancer; thyroid cancer; adrenal cancer; kidney cancer; air bladder cancer; uterine cancer; esophageal cancer; urothelial carcinoma; brain tumors; lymphomas; Ewing's sarcoma; and other neuroendocrine and small blue round cell tumors.

[0496] In a preferred embodiment of the invention, the cancer is small cell lung cancer (SCLC) or glioblastoma.

[0497] All of the above cancers, tumors, neoplasms, etc. characterized by their specific location / origin in the body are meant to include the primary tumor and metastatic tumors derived therefrom.

[0498] If a patient has a cancer characterized by high DLL3 expression, it is possible that the patient is more likely to respond to treatment with a binding protein of the invention (as described herein). Thus, in some embodiments, the cancer to be treated with a binding protein of the invention is a cancer with high DLL3 expression, e.g., DLL3 expression is higher than the average expression in cancer cells of a population of patients suffering from the same type of DLL3 expressing cancer.

[0499] The binding proteins of the invention can be used in a course of treatment in the context of first line, second line or any other line of treatment.

[0500] The binding proteins of the invention can be used for the prevention, short-term or long-term treatment of the above-mentioned diseases, optionally also in combination with radiation therapy, one or more additional therapeutic agents and / or surgery.

[0501] In a preferred embodiment, the protein of the invention is used in combination with a PD-1 antagonist (such as an anti-PD-1 antibody or an anti-PDL-1 antibody) for the treatment of cancer. Preferably, the anti-PD-1 antibody is selected from the group consisting of pembrolizumab, nivolumab, pidilizumab, PD1-1, PD1-2, PD1-3, PD1-4 and PD1-5, as described herein (as defined by the sequences in Table A below) and WO2017 / 198741 (incorporated herein by reference). Preferably, the anti-PDL-1 antibody is selected from the group consisting of atezolizumab, avelumab and durvalumab. In a particularly preferred embodiment, the binding protein of the present invention (preferably any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, or DLL3#3 / CD3#3) is used in combination with PD1-1 for the treatment of cancer. In a particularly preferred embodiment, the binding protein of the present invention (preferably any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, or DLL3#3 / CD3#3) is used in combination with PD1-2 for the treatment of cancer. In a particularly preferred embodiment, the binding protein of the present invention (preferably, any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) is used in combination with PD1-3 to treat cancer. In a particularly preferred embodiment, the binding protein of the present invention (preferably, any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) is used in combination with PD1-4 to treat cancer.In a particularly preferred embodiment, the binding protein of the present invention (preferably, any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) is used in combination with PD1-5 to treat cancer.

[0502] Table A: Amino acid sequences and SEQ ID NOs of the heavy and light chain sequences of anti-PD1 antibodies PD1-1, PD1-2, PD1-3, PD1-4, and PD1-5.

[0503]

[0504]

[0505]

[0506]

[0507]

[0508] According to these preferred embodiments and any other aspects of the invention, antibodies PD1-1, PD1-2, PD1-3, PD1-4 and PD1-5 are antibody molecules as disclosed in WO2017 / 198741 and are defined by the sequences as shown in Table A above.

[0509] Thus, PD1-1 has a heavy chain comprising the amino acid sequence of SEQ ID NO: 256 and a light chain comprising the amino acid sequence of SEQ ID NO: 257;

[0510] PD1-2 has a heavy chain comprising the amino acid sequence of SEQ ID NO: 258 and a light chain comprising the amino acid sequence of SEQ ID NO: 259;

[0511] PD1-3 has a heavy chain comprising the amino acid sequence of SEQ ID NO: 260 and a light chain comprising the amino acid sequence of SEQ ID NO: 261;

[0512] PD1-4 has a heavy chain comprising the amino acid sequence of SEQ ID NO: 262 and a light chain comprising the amino acid sequence of SEQ ID NO: 263; and

[0513] PD1-5 has a heavy chain comprising the amino acid sequence of SEQ ID NO:264 and a light chain comprising the amino acid sequence of SEQ ID NO:265.

[0514] The above also includes the use of the binding proteins of the present invention in various methods for treating the above diseases by administering a therapeutically effective dose to a patient in need thereof, as well as the use of these binding proteins for the manufacture of medicaments for treating such diseases, as well as pharmaceutical compositions comprising such binding proteins of the present invention, and the preparation and / or manufacture of medicaments comprising such binding proteins of the present invention, and the like.

[0515] Combination with other active substances or treatments

[0516] The binding proteins of the invention can be used by themselves or in combination with one or more additional therapeutic agents, particularly in combination with chemotherapeutic DNA damaging agents, therapeutically active compounds that inhibit angiogenesis, signal transduction pathway inhibitors, EGFR inhibitors, immunomodulators, immune checkpoint inhibitors, mitotic checkpoint inhibitors, or hormone therapy agents.

[0517] The additional therapeutic agent can optionally be administered simultaneously with the components of the same pharmaceutical formulation, or before or after administration of the DLL3 / CD3 binding protein.

[0518] Cytostatic and / or cytotoxic active substances that can be administered in combination with the binding molecules of the invention include, but are not limited to, hormones, hormone analogs and anti-hormones, aromatase inhibitors, LHRH agonists and antagonists, inhibitors of growth factors (growth factors such as platelet-derived growth factor (PDGF), fibroblast growth factor (FGF), vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), insulin-like growth factor (IGF), human epidermal growth factor (HER, such as HER2, HER3, HER4) and hepatocyte growth factor (HGF)), inhibitors such as (anti) growth factor antibodies, (anti) growth factor substrate antibodies and tyrosine kinase inhibitors such as cetuximab, gefitinib, afatinib, nivolumab, selenomethionine, tadalafil, tadalafil, selenomethionine ... nintedanib, imatinib, lapatinib, bosutinib, and trastuzumab; antimetabolites (e.g., antifolates such as methotrexate, raltitrexed, pyrimidine analogs such as 5-fluorouracil (5-FU), gemcitabine, irinotecan, cranberry, TAS-102, capecitabine, and gemcitabine; purine and adenosine analogs such as mercaptopurine, thioguanine, cladribine, and pentostatin; cytarabine (arabinoside) C), fludarabine); antitumor antibiotics (e.g., anthracyclines); platinum derivatives (e.g., cisplatin, oxaliplatin, carboplatin); alkylating agents (e.g., estramustine, nitrogen mustard, melphalan, chlorambucil, busulfan, dacarbazin, cyclophosphamide, ifosfamide, temozolomide, nitrosoureas such as carmustin and lomustin, thiotepa), antimitotic agents (e.g., vinca alkaloids such as vinblastine, vindesin, vinorelbin, and vincristine; and taxanes such as paclitaxel, docetaxel); angiogenesis inhibitors, including bevacizumab, ramucirumab, and aflibercept, tubulin inhibitors;DNA synthesis inhibitors, PARP inhibitors, topoisomerase inhibitors (e.g., epipodophyllotoxins such as etoposide and etopophos, teniposide, amsacrin, topotecan, irinotecan, mitoxantrone), serine / threonine kinase inhibitors (e.g., PDK1 inhibitors, Raf inhibitors, A-Raf inhibitors, B-Raf inhibitors, C-Raf inhibitors, mTOR inhibitors, mTORC1 / 2 inhibitors, PI3K inhibitors agents, PI3Kα inhibitors, dual mTOR / PI3K inhibitors, STK33 inhibitors, AKT inhibitors, PLK1 inhibitors (such as volasertib, inhibitors of CDK, including CDK9 inhibitors, Aurora kinase inhibitors), tyrosine kinase inhibitors (e.g., PTK2 / FAK inhibitors), protein-protein interaction inhibitors, MEK inhibitors, ERK inhibitors, FLT3 inhibitors, BRD4 inhibitors, IGF-1R inhibitors, Bcl-xL inhibitors, Bcl-2 inhibitors, Bcl-2 / Bcl-xL inhibitors, ErbB substrate inhibitors, BCR-ABL inhibitors , ABL inhibitors, Src inhibitors, rapamycin analogs (e.g., everolimus, temsirolimus, ridaforolimus, sirolimus), androgen synthesis inhibitors, androgen substrate inhibitors, DNMT inhibitors, HDAC inhibitors, ANG1 / 2 inhibitors, CYP17 inhibitors, radiopharmaceuticals, immunotherapeutics, such as immune checkpoint inhibitors (e.g., CTLA4, PD1, PD-L1, LAG3 and TIM3 binding molecules / immunoglobulins, such as ipilimumab, nivolumab, pembrolizumab) and various chemotherapeutic agents such as amifostin, anagrelid, clodronat, filgrastin, interferon, interferon alpha, leucovorin, rituximab, procarbazine, levamisole, mesna, mitotane, pamidronate, and porfimer; proteasome inhibitors such as bortezomib; Smac and BH3 mimetics; agents that restore p53 functionality, including mdm2-p53 antagonists; inhibitors of the Wnt / β-catenin signaling pathway; and / or cyclin-dependent kinase 9 inhibitors. ;

[0519] Particularly preferred is treatment with a binding molecule of the invention in combination with one or more immunotherapeutic agents (including anti-PD-1 and anti-PD-L1 agents and anti-LAG3 agents): Exemplary anti-PD1 agents include, but are not limited to, anti-PD-1 antibody PDR-001, pembrolizumab, nivolumab, pilizumab, and PD1-1, PD1-2, PD1-3, PD1-4, and PD1-5 as disclosed herein (Table A) and WO2017 / 198741. Exemplary anti-PDL-1 agents include, but are not limited to, atezolizumab, avelumab, and durvalumab. In a preferred embodiment, the binding molecules of the present invention (preferably any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) are combined with PD1-1. In a preferred embodiment, the binding molecules of the present invention (preferably any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) are combined with PD1-2. In a preferred embodiment, the binding molecules of the present invention (preferably any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) are combined with PD1-3. In a preferred embodiment, the binding molecules of the present invention (preferably any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) are combined with PD1-4. In a preferred embodiment, the binding molecules of the present invention (preferably any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) are combined with PD1-5.

[0520] In certain embodiments, the additional therapeutic agent may be other immunotherapeutic agents, such as modulators of TIM-1, TIM-3, TIM-4, PD-L2, LAG3, CTLA-4, Galectin-9, Galectin-1, CD69, CD113, GPR56, CD48, GARP, CAECAM-1, BTLA, TIGIT, CD160, LAIR1, 2B4, CEACAM, CD39, TGFβ, IL-10, Fas Ligand, ICOS, B7 family (B7-1, B7-2, B7-H1 (PDL-1), B7-DC (PD-L2), B7-H2 (ICOS-L), B7-H3, B7-H4, B7-H5 (VISTA) or B7-H6).

[0521] In some embodiments, the additional immunotherapeutic agent is a member of the TNF family of molecules that bind to a cognate TNF substrate family member, including CD40 and CD40L, OX-40, OX-40L, CD70, CD27L, CD30, CD30L, 4-1BBL, CD137, GITR, TRAIL / Apo2-L, TRAILR1 / DR4, TRAILR2 / DR5, TRAILR3, TRAILR4, OPG, RANK, RANKL , TWEAKR / Fn14, TWEAK, BAFFR, EDAR, XEDAR, TACI, APRIL, BCMA, LIGHT, DcR3, HVEM, VEGI / TLlA, TRAMP / DR3, E DAR, EDAl,

[0522] In some embodiments, the additional therapeutic agent is a SMAC mimetic. A SMAC mimetic is a compound that binds to an inhibitor of apoptosis protein (IAP) and has an immunomodulatory function and mediates systemic cytokines (e.g., IL-6, TNFα, etc.) and chemokines (e.g., MCP-1) when administered to an animal or human. In some embodiments, the SMAC mimetic is (i) LCL161, i.e., Compound A in Example 1 of WO 2008 / 016893 (pages 28 / 29;

[122] ), or a pharmaceutically acceptable salt thereof; (ii) a SMAC mimetic called Debio-1143, or a pharmaceutically acceptable salt thereof; (iii) a SMAC mimetic called birinapant, or a pharmaceutically acceptable salt thereof; (iv) a SMAC mimetic called ASTX-660, or a pharmaceutically acceptable salt thereof; or (v) a SMAC mimetic called CUDC-427, or a pharmaceutically acceptable salt thereof.

[0523] In some embodiments, the additional immunotherapeutic agent is selected from (i) antagonists of cytokines that inhibit T cell activation (e.g., IL-6, IL-10, TGF-B, VEGF; "immunosuppressive cytokines") and / or (ii) agonists of cytokines that stimulate T cell activation and / or are used to stimulate an immune response, e.g., for the treatment of proliferative diseases (such as cancer) (such as IL2).

[0524] In some embodiments, the additional immunotherapeutic agent is an agonist of a protein that stimulates T cell activation, such as CD28, GITRL, OX40L, CD27, and CD28H.

[0525] In some embodiments, the additional therapeutic agent is an oncolytic virus, including but not limited to an oncolytic virus derived from pox virus, adenovirus (AdV), herpes simplex virus (HSV1 or HSV2), reovirus, myxoma virus (MYXV), poliovirus, vesicular stomatitis virus (VSV), Maraba virus, varicella virus, measles virus (MV), or Newcastle disease virus (NDV).

[0526] Diagnostic Uses

[0527] The anti-DLL3 antibody molecules of the invention are useful in diagnostic and prognostic methods and can be used to label, localize or identify cells or tissues expressing DLL3 (e.g., in an ELISA assay, FACS assay, immunohistology or similar assay) by attaching a dye, drug or another molecule that has binding specificity for a different antigen. For example, a detectable label can be bound to an anti-DLL3 antibody molecule of the invention or a second reagent (e.g., a secondary antibody) that specifically binds to an antibody molecule of the invention and that is bound to a detectable label can be used in such diagnostic methods. In some embodiments, DLL3-specific antibodies specifically bind to DLL3 expressed in cells (in the cytoplasm and / or at the cell surface) and are used to localize and / or identify such cells. In some embodiments, the DLL3-specific antibodies provided herein are used to identify cells or tissues (e.g., tumor cells) that express DLL3.

[0528] The antibodies of the invention are selected such that they have a high level of epitope binding specificity and a high binding affinity to the DLL3 polypeptide. Generally speaking, the higher the binding affinity of the antibody, the more stringent washing conditions can be performed in the immunoassay to remove non-specifically bound material without removing the target polypeptide.

[0529] Therefore, the present invention further provides a method for detecting DLL3 in a biological sample. Specifically, in one aspect, the present invention provides a method for detecting DLL3 in a sample (eg, a tissue sample, a tumor tissue sample), comprising:

[0530] (a) contacting a sample with an antibody of the present invention; and

[0531] (b) Detection of anti-DLL3 antibodies binding to DLL3.

[0532] The biological sample can be obtained / isolated from any tissue (including biological specimens), cells or body fluids of an individual. In a preferred embodiment, the sample is a tissue sample, preferably a tumor tissue sample. In a specific embodiment, the sample is a fixed (tumor) tissue sample, preferably a formalin-fixed, paraffin-embedded (FFPE) tissue sample, more preferably a formalin-fixed, paraffin-embedded (FFPE) tumor tissue sample.

[0533] The sample is usually divided into several parts and attached to the culture medium for microscopic analysis, such as a microscope slide. In the case where the sample is a tissue sample, several parts may be tissue sections. In some embodiments, serial sections are taken from FFPE tissue samples. In some embodiments, serial sections are taken from a plurality of different FFPE block sites, which can be completed to obtain heterogeneity within the section and heterogeneity within the block. In some embodiments, serial sections are taken from a plurality of different biopsy samples obtained from different locations of the same tumor, which can be performed to obtain heterogeneity within the section and heterogeneity within the tumor. Typically, tissue samples are first deparaffinized and antigen extracted (e.g., with proteinase K) before antigen detection. Tissue samples can be obtained from / separated from a variety of tissues (e.g., biopsy), specifically from tissues of tumors, including but not limited to SCLC, LCNEC, glioma, glioblastoma, melanoma or other neuroendocrine tumors (e.g., neuroendocrine prostate cancer or neuroendocrine pancreatic cancer, small cell bladder cancer).

[0534] Other exemplary tumors from which (tissue) samples may be obtained / isolated include lung cancer; preferably SCLC, NSCLC or LCNEC; breast cancer; cervical cancer; colon cancer; colorectal cancer; endometrial cancer; head and neck cancer; liver cancer (hepatoblastoma or hepatocellular carcinoma); ovarian cancer; pancreatic cancer; prostate cancer; skin cancer; stomach cancer; testicular cancer; thyroid cancer; adrenal cancer; kidney cancer; air bladder cancer; uterine cancer; esophageal cancer; urothelial cancer; brain tumors; lymphomas; Ewing sarcoma; and other neuroendocrine (including pseudoendocrine) and small blue round cell tumors.

[0535] Exemplary methods for detection of DLL3 in a sample are immunocytochemistry (ICC), immunohistochemistry (IHC), Western blot, flow cytometry and / or ELISA.

[0536] In a preferred embodiment, DLL3 is detected using IHC.

[0537] In one specific aspect of the invention, a method for detecting DLL3 in a tissue sample comprises

[0538] a) contacting a tissue sample (e.g., tumor tissue, such as SCLC, glioblastoma, or neuroendocrine tumor) with an antibody of the present invention (e.g., DLL3#1 or DLL3#5), preferably the tissue sample is a fixed tissue sample (e.g., formalin-fixed and paraffin-embedded)

[0539] b) allowing antibody-antigen complexes to form in the sample; and

[0540] c) Detection of anti-DLL3 antibodies binding to DLL3.

[0541] In some embodiments of the methods of the invention, the anti-DLL3 antibody is detected by a detectable signal, preferably, the detectable signal is produced by a detectable label. In one preferred embodiment of the methods of the invention, the detectable signal is detected in an IHC analysis. In some embodiments, the detectable signal is detected in an ELISA analysis, by flow cytometry or by Western blot. In some embodiments, the methods described herein can be used to detect DLL3 present on the cell surface of a cell (e.g., a tumor cell), for example, using flow cytometry. In some embodiments, the methods described herein can be used to detect the presence of DLL3 in a tissue sample (e.g., a tumor tissue sample) using ELISA, Western blot, or IHC.

[0542] In some embodiments, the detectable label is directly bound to the anti-DLL3 antibody and is thus deposited on the sample after the anti-DLL3 antibody binds to DLL3, which is generally referred to as direct expression. Direct expression is often more directly quantitative, but often lacks sensitivity. In other embodiments, the deposition of the detectable label is achieved by using a secondary detection reagent that specifically binds to the antibody molecule of the invention and to the detectable label (e.g., a secondary antibody), which is generally referred to as indirect expression. In some embodiments, the specific secondary detection reagent can be a species-specific secondary antibody, an anti-hapten antibody that binds to a hapten-bound anti-DLL3 antibody, or a biotin-binding protein that binds to a biotin-labeled anti-DLL3 antibody.

[0543] Detectable labels that can be conjugated to an anti-DLL3 antibody of the invention or to a secondary detection agent include chromogenic, fluorescent, phosphorescent, luminescent, and radioactive molecules and materials, catalysts that convert one substance into another to provide a detectable difference (such as by converting a colorless substance into a colored substance or vice versa, or by producing a precipitate or increasing sample turbidity), haptens that can be detected via antibody-hapten binding interactions using an additional detectably labeled antibody conjugate, and paramagnetic and magnetic molecules or materials.

[0544] For example, detectable labels can be enzymes such as horseradish peroxidase (HRP), alkaline phosphatase (AP), acid phosphatase, glucose oxidase, β-galactose, β-glucuronidase, and β-lactamase; fluorophores such as fluorescein, luminophores, coumarins, BODIPY dyes, resorufin, and rhodamine; nanoparticles such as quantum dots (U.S. Pat. Nos. 6,815,064, 6,682,596, and 6,649,138); metal chelates such as DOTA and DPTA chelates of radioactive or paramagnetic metal ions (Gd<3+>); and liposomes, for example, liposomes containing captured fluorescent molecules.

[0545] Where the detectable label comprises an enzyme, a detectable substrate (such as a chromogen, a fluorescent compound, or a luminescent compound) may be used in combination with the enzyme to produce a detectable signal. Specific examples of chromogenic compounds include diaminobenzidine (DAB), 4-nitrophenyl phosphate (pNPP), fast red ( red), bromochloroindolyl phosphate (BCIP), nitro blue tetrazolium (NBT), BCIP / NBT, fast red, AP orange, AP blue, tetramethylbenzidine (TMB), 2,2'-azobis-[3-ethylbenzothiazoline sulfonate] (ABTS), o-dimethoxyaniline, 4-chloronaphthol (4-CN), nitrophenyl--D-galactopyranoside (ONPG), o-phenylenediamine (OPD), 5-bromo-4-chloro-3-indolyl--galactopyranoside (X-Gal), methylumbelliferyl--D-galactopyranoside (MU-Gal), p-nitrophenyl-aD-galactopyranoside (PNP), 5-bromo-4-chloro-3-indolyl--D-glucuronide (X-Gluc), 3-amino-9-ethylcarbazole (AEC), fuchsin, iodo-nitrotetrazolium (INT), tetrazolium blue and tetrazolium violet.

[0546] In some embodiments, the detectable moiety is a fluorophore, which belongs to several common chemical classes, including coumarins, fluorescein (or fluorescein derivatives and analogs), rhodamines, resorufins, luminophores, and cyanines.

[0547] In other embodiments, the detectable moiety is a molecule detectable via bright field microscopy, such as a dye including diaminobenzidine (DAB), 4-(dimethylamino)azobenzene-4'-sulfonamide (DABSYL), tetramethylrhodamine (DISCOVERY Purple), N,N'-dicarboxypentyl-5,5'-disulfonato-indole-dicarbocyanine (Cy5), and rhodamine 110 (rhodamine).

[0548] A hapten is a small molecule that binds specifically to an antibody, but by itself will not elicit an immune response in an animal and must first be attached to a larger carrier molecule, such as a protein, to produce an immune response. Examples of haptens include di-nitrophenyl, biotin, digoxigenin, and fluorescein.

[0549] Further provided herein in one aspect are methods of diagnosing or identifying a tumor as a DLL3-expressing tumor (expressing DLL3 in the cytoplasm and / or on the cell surface), comprising detecting DLL3 in a sample of an individual (e.g., using IHC on a tumor tissue sample) using an anti-DLL3 antibody of the invention (e.g., a tumor tissue sample) (e.g., DLL3#1 or DLL3#5). In another aspect, provided herein are methods of selecting a DLL3-targeted therapy for an individual, comprising detecting DLL3 in a sample of an individual (e.g., a tumor tissue sample) using an anti-DLL3 antibody of the invention (e.g., DLL3#1 or DLL3#5) (e.g., using IHC on a tumor tissue sample). In another aspect, provided herein are methods of monitoring the efficacy of a treatment in an individual (e.g., the efficacy of a DLL3-targeted therapy), comprising detecting DLL3 in a sample of an individual (e.g., a tumor tissue sample) using an anti-DLL3 antibody of the invention (e.g., DLL3#1 or DLL3#5) (e.g., using IHC on a tumor tissue sample).

[0550] The anti-DLL3 antibody molecules of the present invention can also be used to target cells. In some embodiments, the DLL3-specific antibodies provided herein are used to deliver drugs or cytotoxic agents to target cells (e.g., tumor cells expressing DLL3) by linking such drugs or cytotoxic agents to the DLL3 antibodies, thereby, for example, killing the target cells.

[0551] Reagent test kit

[0552] The present invention also encompasses a kit comprising at least a multispecific binding protein of the present invention (e.g., DLL3#1 / CD3#1, DLL3#2 / CD3#1, DLL3#3 / CD3#1, DLL3#4 / CD3#1, DLL3#5 / CD3#1, DLL3#6 / CD3#1, DLL3#7 / CD3#1, DLL3#8 / CD3#1, DLL3#9 / CD3#1, DLL3#10 / CD3#1, DLL3#11 / CD3#1, DLL3#12 / CD3#1, DLL3#13 / CD3#1, DLL3#14 / CD3#1, DLL3#15 / CD3#1, DLL3#16 / CD3#1, DLL3#17 / CD3#1, DLL3#18 / CD3#1, DLL3#19 / CD3#1, DLL3#20 / CD3#1, DLL3#21 / CD3#1, DLL3#22 / CD3#1, DLL3#23 / CD3#1, DLL3#24 / CD3#1, DLL3#25 / CD3#1, DLL3#26 / CD3#1, DLL3#27 / CD3#1, DLL3#28 / CD3#1, DLL3#29 / CD3#1, DLL3#30 / CD3#1, DLL3#31 / CD3#1, DLL3#32 / CD3#1, DLL3#33 / CD3#1, DLL3#34 / CD3#1, DLL3#35 / CD3#1, DLL3#36 / CD3#1 L3#16 / CD3#1, DLL3#17 / CD3#1, DLL3#18 / CD3#1, DLL3#1 / CD3#2, DLL3#2 / CD3#2, DLL3#3 / CD3#2, DLL3#4 / CD3#2, DLL3#5 / CD3#2, DLL3#6 / CD3# 2. DLL3#7 / CD3#2, DLL3#8 / CD3#2, DLL3#9 / CD3#2, DLL3#10 / CD3#2, DLL3#11 / CD3#2, DLL3#12 / CD3#2, DLL3#13 / CD3#2, DLL3#14 / CD3#2, DLL3#1 5 / CD3#2, DLL3#16 / CD3#2, DLL3#17 / CD3#2, DLL3#18 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#3, DLL3#3 / CD3#3, DLL3#4 / CD3#3, DLL3#5 / CD3#3 and D LL3#6 / CD3#3, DLL3#7 / CD3#3, DLL3#8 / CD3#3, DLL3#9 / CD3#3, DLL3#10 / CD3#3, DLL3#11 / CD3#3, DLL3#12 / CD3#3, DLL3#13 / CD3#3, DLL3#14 / CD DLL3#3 / CD3#3, any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3), and optionally one or more other components selected from the group consisting of other drugs for treating the diseases and conditions mentioned above.

[0553] In one embodiment, the kit includes a composition containing an effective amount of a binding protein of the invention in unit dosage form.

[0554] The invention also encompasses kits comprising at least a multispecific binding protein of the invention, and one or more additional components selected from the group consisting of other drugs for treating diseases and disorders as described above.

[0555] In one embodiment, the kit includes a composition containing an effective amount of a multispecific binding protein of the present invention (preferably any one of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) in a unit dosage form. In another embodiment, the kit comprises both: a composition containing an effective amount of a multispecific binding protein of the invention (preferably any of DLL3#1 / CD3#1, DLL3#1 / CD3#2, DLL3#1 / CD3#3, DLL3#2 / CD3#1, DLL3#2 / CD3#2, DLL3#2 / CD3#3, DLL3#3 / CD3#1, DLL3#3 / CD3#2, DLL3#3 / CD3#3) in a unit dosage form; and a composition containing an effective amount of a PD-1 antagonist (such as an anti-PD-1 antibody as described herein (e.g., Table A) and WO2017 / 198741, most preferably PD1-1, PD1-2, PD1-3, PD1-4, and PD1-5) in a unit dosage form.

[0556] In some embodiments, the kit comprises a sterile container containing such a composition; these containers may be in the form of boxes, ampoules, bottles, vials, tubes, bags, pouches, blister packs, or other suitable containers known in the art. Such containers may be made of plastic, glass, laminated paper, metal foil, or other materials suitable for preserving medicaments. In addition, the kit may include a pharmaceutical composition in a lyophilized form in a first container with the binding protein of the present invention and a second container with a pharmaceutically acceptable diluent (e.g., sterile water) for injection. Pharmaceutically acceptable diluents can be used to reconstitute or dilute the binding protein.

[0557] Optionally, the multispecific binding proteins of the invention are provided together with instructions for administering the multispecific binding proteins to an individual suffering from cancer. The instructions will generally include information about the use of the composition for treating or preventing cancer. In other embodiments, the instructions include at least one of the following: instructions for the therapeutic agent, dosage schedule and administration for treating or preventing cancer or its symptoms, precautions, warnings, indications, contraindications, overdose information, adverse reactions, animal pharmacology, clinical studies, and / or references. The instructions may be printed directly on the container (if present) or as a label applied to the container, or as a separate sheet of paper, brochure, card, or folder supplied in or with the container.

[0558] As described herein, the invention further provides diagnostic methods for determining DLL3 expression in a biological sample. In one aspect, the invention provides kits for performing these methods and instructions for practicing the methods of the invention (such as collecting tissues and / or performing detection assays, and / or analyzing results).

[0559] The kit comprises, or alternatively consists essentially of, or further consists of, a DLL3 antibody composition of the invention (e.g., a monoclonal antibody) and instructions for use. The kit is suitable for detecting the presence of a DLL3 polypeptide in a biological sample (e.g., a biopsy of any body fluid or body tissue). In some embodiments, the kit may comprise: one or more DLL3 antibodies (e.g., anti-DLL3 antibodies DLL3#1 or DLL3#5) that are capable of binding to a DLL3 polypeptide in a sample; means for detecting a DLL3 polypeptide in a sample; and, optionally, means for comparing the signal / amount of a DLL3 polypeptide in a sample to a control. One or more anti-DLL3 antibodies of the invention may be labeled. Kit components (e.g., reagents) may be packaged in suitable containers. The kit may further comprise instructions for using the kit to detect a DLL3 polypeptide. In certain embodiments, the kit comprises an anti-DLL3 antibody of the invention (a first antibody) and a different second antibody that binds to a DLL3 polypeptide or the first antibody and is conjugated to a detectable label.

[0560] The kit may also include, for example, a buffer, a preservative, or a protein stabilizer. The kit may further include components necessary for detecting a detectable label, such as an enzyme or a substrate. The kit may also contain a control sample or a series of control samples, which may be analyzed and compared with the test sample. The components of the kit may be sealed in separate containers, and all of the multiple containers may be in a single package together with instructions for explaining the results of the analysis performed using the kit. The kit of the present invention may contain a written product on or in the kit container. The written product describes how to use the reagents contained in the kit.

[0561] For convenience, these kit components can be packaged in a manner commonly used by those skilled in the art. For example, these kit components can be provided in the form of a solution or a liquid dispersion or the like. Example

[0562] The following examples illustrate the present invention. These examples should not be construed as limiting the scope of the present invention.

[0563] Example 1: Design and construction of CD3 / DLL3 binding protein

[0564] The present inventors have developed a multispecific binding protein that binds to DLL3 and CD3 and induces T cell activation, leading to lysis of DLL3 expressing tumor cells. The molecular design used has an IgG antibody backbone and an IgG-like structure. It is characterized by knob-hole technology in the Fc for the heterodimeric knob (anti-DLL3) and hole (anti-CD3) arms. In addition, the binding protein has a flexible peptide sequence between the light chain and the corresponding heavy chain in each arm. Thus, the binding protein comprises two arms, one binding to CD3 and the other to DLL3, each arm comprising a single chain Fab and an Fc region.

[0565] Preferably, the binding molecules are bispecific and bivalent (monovalent for each of the two targets).

[0566] Preparation of binding domains recognizing DLL3 and CD3 using high-throughput V genes recovered from hybridomas and cultured single B cells

[0567] To obtain anti-DLL3 binders, hybridomas or single B cells derived from wild-type and AlivaMab humanized mice (Ablexis, San Francisco, CA, USA: AlivaMab transgenic mouse platform with human immunoglobulin loci) immunized with DLL3 were cultured in vitro. Reactivity to recombinant human DLL3 was determined by AlphaLISA (PerkinElmer, Waltham, MA, USA) and to SHP77 cells expressing human DLL3 by flow cytometry. CRL-2195TM) and screen the supernatant.

[0568] Immunoglobulin (Ig) VH and VL genes were subsequently amplified from the identified positive clones. 6Cells from a single pure line were pelleted and used as source material. For a single B cell, 100 to 500 cells amplified from a single isolated B cell were used as source material. RNA was isolated using RNeasy Plus (Qiagen, Hilden, Germany). cDNA was subsequently synthesized using a Smarter cDNA synthesis kit (Clontech, Mountain View, CA) according to the manufacturer's instructions. To promote cDNA synthesis, oligodT was used to initiate reverse transcription of all messenger RNAs, followed by "5' capping" with Smarter IIA oligonucleotides. 2-step PCR amplification was used to perform subsequent amplification of VH and VL fragments using a 5' primer targeting the Smarter IIA end cap and a 3' primer targeting the common region in CH1. In brief, each 50 μl PCR reaction consisted of 20 μM forward and reverse primer mixtures, 25 μl PrimeStar Max DNA polymerase premix (Clontech), 2 μl unpurified cDNA, and 21 μl double distilled H2O. The cycling program started at 94°C for 3 min, followed by 35 cycles (94°C for 30 sec, 50°C for 1 min, 68°C for 1 min), and ended at 72°C for 7 min. A second round of PCR was performed with VL and VH second round primers containing 15 bp complementary extensions that "overlapped" their respective regions (VH and VL) in their respective pTT5 parent vectors. The second round of PCR was performed with the following program: 94°C for 3 min; 35 cycles (94°C for 30 sec, 50°C for 1 min, 68°C for 1 min), and ended at 72°C for 7 min.

[0569] use The VL gene was directionally cloned into the pTT5huIgK vector using the HD cloning kit (Clontech, USA) and the VH gene was directionally cloned into the pTT5 huIgG1KO vector. HD cloning, PCR products were purified and treated with cloning enhancers before In-Fusion HD cloning. Cloning and transformation were performed according to the manufacturer's protocol (Clontech, USA). Mini-prep DNA was subjected to Sanger sequencing to confirm that the complete V gene segment was obtained.

[0570] Using this method, Ig VH and VL gene pairs were generated that encode binding domains specific for DLL3. Recombinant antibodies were produced by transiently transfecting CHO-E37 cells with the corresponding heavy and light chain encoding plasmids.

[0571] Validation and screening of recombinant antibodies

[0572] Supernatants containing expressed recombinant antibodies were analyzed by flow cytometry for binding to cell lines expressing human or cynomolgus macaque DLL3. Briefly, cells were incubated with recombinant supernatants, washed, and bound mAbs from supernatants were detected with anti-human-IgG-APC (Jackson ImmunoResearch 109-136-098). Signal / background ratios (S / B) were calculated by dividing the median fluorescence intensity (MFI) of the sample by that of the isotype control (variable region vs. irrelevant protein and different constant region backbones).

[0573] Surface plasmon resonance (SPR) was performed on the recombinant supernatant on a Biacore 400. Briefly, non-optimized IgG in HTP supernatant was captured on the sensor surface via protein A / G at 10 ul / min for 60 sec. Binding of 100 nM human DLL3 to the captured IgG was monitored, association for 180 sec at 30 ul / min followed by dissociation for 120 sec in HBS-EP buffer. Regeneration of the protein A / G surface was performed between each binding cycle with glycine pH 2.1. The following materials were used for this analysis: Protein reagent: recombinantly expressed human DLL3. System operating buffer: HBS-EP (10 mM HEPES pH 7.4, 150 mM NaCl, 3 mM EDTA and 0.005% v / v polysorbate P20). Capture reagent: protein A / GG (ThermoFisher Scientific, Waltham, MA, USA), specific for all human IgG isotypes.

[0574] Relevant clones (Kd < 300 pM) were selected for multispecific formatting.Multispecific binding proteins were generated and further evaluated in mechanistic and functional screens (such as cell binding, cytotoxicity, and T cell activation as described below).

[0575] Humanization of DLL3 and CD3 binders

[0576] The sequences of DLL3 binders as described above and CD3 binders described in the literature (Pessano et al., EMBO J. 1985 Feb; 4(2): 337-44; Salmerón A et al., J Immunol. 1991 Nov 1; 147(9): 3047-52) are humanized and / or optimized. Sequence optimization / humanization of antibodies is a method for engineering antibodies produced in non-human species (for a specific antigen / epitope) for use as therapeutic agents that are similar to antibodies produced in humans and thereby eliminate possible adverse effects such as immunogenicity while retaining specificity. The sequence optimization / humanization method utilized herein is described by Singh et al., 2015 (Singh S et al., mAbs 2015: 7(4): 778-91). Briefly, close matches to human germlines are identified by in silico hybridization, and optimized / humanized variants are evaluated using phage screening methods. Final lead candidate sequences were selected based on binding, percent human score and Epivax (an in silico hybridization prediction tool for potential immunogenicity) score.

[0577] Construction of bispecific proteins that bind DLL3 and CD3

[0578] Using common molecular biology techniques, the variable regions of DLL3 and CD3 binders were cloned into the expression vector pTT5 (National Research Council, Canada) to form a bispecific binding protein having: a DLL3-specific binding unit comprising a single-chain Fab and Fc region that binds to DLL3 (such binding units are also referred to herein as "DLL3 arms" or "DLL3 chains"); and a CD3-specific binding unit comprising a single-chain Fab and Fc region that binds to CD3 (such binding units are also referred to herein as "CD3 arms" or "CD3 chains"). The Fc regions of the DLL3 and CD3 arms include "knob" or "hole" mutations (Atwell et al., JMB, 1997, 270, 26-35) and the respective chains are referred to as knob or hole chains. For multi-fragment DNA assembly, Gibson assembly and NEBuilder HiFi DNA assembly methods (New England Biolabs, Ipswich, MA, USA) were used according to the manufacturer's protocol. Miniprep DNA was sequenced.

[0579] Each expression vector contains a chain encoding gene (DLL3 or CD3 arm / chain), a eukaryotic promoter component encoding the signal sequence and the genes for the light and heavy chains, an expression cassette for a prokaryotic selectable marker gene (such as ampicillin), and an origin of replication. These DNA plasmids were propagated in ampicillin-resistant E. coli colonies and cultures and purified.

[0580] Example 2: Expression and purification of bispecific bivalent binding proteins that bind DLL3 and CD3

[0581] Bispecific molecules binding to DLL3 and CD3 were prepared by transiently transfecting CHO-E cells with the pTT5 vector carrying the genes encoding the DLL3 / CD3-chain. Briefly, transfected CHO-E cells grown in suspension in serum-free medium were cultured in shake flasks under agitation at 140 rpm, 37°C and 5% CO2 and maintained under exponential growth conditions. On the day of transfection, cells were chemically transfected with a 1:3 mass ratio of a pestle-chain plasmid and a hole-chain plasmid. Cells were then seeded in 1 L of 100 μL culture medium at 1 to 2 × 10^6 cells / mL. FreeStyle TM CHO expression medium (Life Technologies, NY, US) was added. The cells were then cultured for 10 days under ambient shaking and fed once with 200 ml of commercial feed solution for protein expression. The antibody titer in the cell culture supernatant was determined using the ELISA instrument (Pall ForteBio, CA, US) and portA biosensor probe tip according to the manufacturer's instructions.

[0582] Recombinant DLL3 / CD3 binding protein was purified from culture supernatant in two steps: first by protein A affinity chromatography using MabSelect TM Column (GE Healthcare); second, by cation exchange chromatography using a 50HS column (Applied Biosystems, Carlsbad, CA, USA). The two-step purified material was stored in a final buffer of 50 mM sodium acetate and 100 mM NaCl, pH 5.0 purity and the degree of heterogeneity of the samples was assessed by analytical size exclusion chromatography, mass spectrometry, and analytical ultracentrifugation. The samples subjected to functional testing contained the two-step purified material with approximately 99% monomer content.

[0583] Table 1: Amino acid sequences and SEQ ID NOs of CDRs, VH, VL, scFab, DLL3 arms and CD3 arms of the proteins / antibody constructs described herein:

[0584]

[0585]

[0586]

[0587]

[0588]

[0589]

[0590]

[0591]

[0592]

[0593]

[0594]

[0595]

[0596]

[0597]

[0598]

[0599]

[0600]

[0601]

[0602]

[0603]

[0604]

[0605]

[0606]

[0607]

[0608]

[0609] *The underlined sequence indicates the hinge region

[0610] Example 3: Production of recombinant protein

[0611] Human DLL3-His

[0612] The cell line for producing human DLL3-His was generated using HEK-293 cells (Thermo Fisher), Lenti-X lentiviral system (Clontech), and plasmid encoding human DLL3-His (human DLL3 accession number Q9NYJ7 huDLL3-His: SEQ ID NO: 106). For expression, cells were cultured and expanded at 37°C, 5% CO2 and shaking at 140 rpm. On day 0 of expression, cells were pelleted and resuspended in Expi293 medium. On day 3 of expression, conditioned culture supernatant was collected by pelleting cells at 4700 rpm for 40 minutes. Protease inhibitors were added to the biomass prior to purification. Expression was confirmed by Western blotting. Conditioned culture supernatant was conditioned with 0.5 mM TCEP, 0.02% CHAPS, 10 mM imidazole. Purification was performed on a HisTrap Ni excel column and buffer A: 50 mM MES, 50 mM NaCl, 0.5 mM TCEP, 0.02% CHAPS, pH 6.5. The protein of interest was eluted using an elution gradient from 20 mM imidazole to 500 mM imidazole in buffer A, pH 8.5 supplemented with 0.5 M imidazole. The pooled fractions were dialyzed in the following buffer: 50 mM MES, 50 mM NaCl, 1 mM TCEP, 0.02% CHAPS, 0.2 M arginine, 3% glycerol, pH 6.5. The purified material was checked by mass spectrometry and analytical ultracentrifugation.

[0613] · Macaca fascicularis DLL3-His

[0614] The cell line used to produce cynomolgus macaque DLL3-His was generated using HEK-293 cells (Thermo Fisher), Lenti-X lentiviral system (Clontech) and plasmids encoding cynomolgus macaque DLL3-His (cynomolgus macaque DLL3 accession number: XM_005589196.1 (RefSeq); cynomolgus macaque DLL3-His: SEQ ID NO: 107). For expression, cells were cultured and expanded at 37°C, 5% CO2 and shaking at 140 rpm. On day 0 of expression, cells were pelleted and resuspended in Expi 293 medium. On day 3 of expression, conditioned culture supernatants were collected by pelleting cells at 4700 rpm for 40 minutes. Protease inhibitors were added to the biomass before purification. Expression was confirmed by Western blotting. Conditioned culture supernatants were conditioned with 0.5 mM TCEP, 0.02% CHAPS, 10 mM imidazole. Purification was performed on a HisTrap Ni excel column and buffer A: 50 mM MES, 50 mM NaCl, 0.5 mM TCEP, 0.02% CHAPS, pH 6.5. The protein of interest was eluted using an elution gradient from 20 mM imidazole to 500 mM imidazole in buffer A, pH 8.5 supplemented with 0.5 M imidazole. The pooled fractions were dialyzed in the following buffer: 50 mM MES, 50 mM NaCl, 1 mM TCEP, 0.02% CHAPS, 0.2 M arginine, 3% glycerol, pH 6.5. The purified material was checked by mass spectrometry and analytical ultracentrifugation.

[0615] Human CD3 E+G HuFc-6xHis (E+G indicates εγ subunit)

[0616] The cell line for the production of human CD3 E+G HuFc-6xHis was generated using HEK-293 cells (Thermo Fisher), Lenti-X lentiviral system (Clontech) and plasmids encoding human CD3 E+G HuFc-6xHis (human CD3E accession number: P07766; human CD3E+G-HuFc-His: SEQ ID NO: 108). For expression, cells were cultured and expanded in Freestyle 293 medium at 37°C, humidified 8% CO2 environment and shaking at 135 rpm. Conditioned culture supernatants were collected on day 6 by centrifugation at 9300xg for 30 minutes. Expression was monitored by SDS-PAGE and Western blotting. Conditioned culture supernatants were adjusted with 0.2 M sucrose, 5% glycerol, 0.01% CHAPS and 10 mM imidazole. The pH was then adjusted to 7.2. Purification was performed in a two-step process: affinity purification using Ni / NTA resin (overnight incubation at 4°C and elution with 250 mM imidazole); followed by size exclusion chromatography on a Superdex 200 column in target buffer PBS containing 0.2 M sucrose, 5% glycerol, 0.01% CHAPS, 1 mM TCEP, pH 7.2. Pooled material was concentrated using 10K MWCO PES membrane viva cell 100 centrifugal devices prior to final analysis and storage. Purified material was checked by mass spectrometry and analytical ultracentrifugation.

[0617] · Cynomolgus macaque CD3 E+G HuFc-6xHis (E+G indicates εγ subunit)

[0618] The cell line used to produce cynomolgus macaque CD3 E+G HuFc-6xHis was generated using HEK-293 cells (Thermo Fisher), Lenti-X lentiviral system (Clontech) and plasmids encoding cynomolgus macaque CD3 E+G HuFc-6xHis (cynomolgus CD3E accession number: Q95LI5<; cynomolgus macaque CD3E+G-huFc-His: SEQ ID NO: 109). For expression, cells were cultured and expanded in Freestyle 293 medium at 37°C, humidified 8% CO2 environment and shaking at 135 rpm. The conditioned culture supernatant was collected on day 6 by centrifugation at 9300xg for 30 minutes. Expression was monitored by SDS-PAGE and Western blotting. The conditioned culture supernatant was conditioned with 0.2M sucrose, 5% glycerol, 0.01% CHAPS and 10mM imidazole. The pH was then adjusted to 7.2. Purification was performed in a two-step process: affinity purification using Ni / NTA resin (overnight incubation at 4°C and elution with 250 mM imidazole); followed by size exclusion chromatography on a Superdex 200 column in target buffer PBS containing 0.2 M sucrose, 5% glycerol, 0.01% CHAPS, 1 mM TCEP, pH 7.2. Pooled material was concentrated using a 10K MWCO PES membrane viva cell 100 centrifugal device prior to final analysis and storage. Purified material was checked by mass spectrometry and analytical ultracentrifugation.

[0619] Example 4A: SPR-based determination of affinity and cross-species reactivity with recombinant DLL3 and CD3εγ subunits

[0620] The binding affinity of the purified DLL3 / CD3 bispecific constructs to recombinant human and cynomolgus monkey DLL3-ECD and Fc fusion proteins of human and cynomolgus monkey CD3 epsilon gamma subunits was determined by surface plasmon resonance (SPR) using a ProteOn XPR36 instrument (Bio-Rad). The operating buffer and all dilutions (unless otherwise stated) were obtained in PBS-T-EDTA. GLM sensor chips (Bio-Rad) were standardized and preconditioned according to the manufacturer's recommendations. The sensor chips were activated with an equal mixture of EDC / s-NHS in the horizontal direction at a flow rate of 30 ul / min for 300 seconds and fixed with protein A / G (20 ug / ml in 10 mM acetate pH 4.5) in the horizontal direction at a flow rate of 30 ul / min for 300 seconds to obtain approximately 5000 RU of protein A / G on the surface. The sensor chips were deactivated with 1 M ethanolamine hydrochloride in the horizontal direction at a flow rate of 30 μl / min for 300 seconds. The sensor chip was stabilized with 0.85% phosphoric acid at a flow rate of 100 μl / min for 18 seconds, three times in the horizontal direction and three times in the vertical direction.

[0621] For the binding kinetics assay for DLL3, bispecific molecules were captured vertically for 250 seconds on a Protein A / G surface at a flow rate of 25 μl / min, respectively. The baseline was stabilized by injecting PBS-T-EDTA horizontally for 60 seconds at a flow rate of 40 μl / min. The analyte (hu DLL3 or cy DLL3) was injected horizontally over the captured antibody for 600 seconds at a flow rate of 40 μl / min and dissociated for 2700 seconds. The concentrations of the analytes were 20 nM, 10 nM, 5 nM, 2.5 nM and 1.25 nM. The surface was regenerated by injecting 0.85% phosphoric acid at a flow rate of 100 μl / min for 18 seconds, once horizontally and once vertically.

[0622] For binding kinetics assays to CD3, hu CD3εγ-Fc and cy CD3εγ-Fc (dissolved in pH 4.5 acetate) were directly immobilized on the sensor surface via vertical amine coupling at a flow rate of 30 μl / min for 140 seconds. The baseline was stabilized by horizontal injection of PBS-T-EDTA at a flow rate of 40 μl / min for 60 seconds. The analyte (bispecific molecule) was horizontally injected over the CD3εγ-Fc construct for 600 seconds at a flow rate of 40 μl / min and dissociated for 600 seconds. The concentrations of the analytes were 250 nM, 125 nM, 62.5 nM, 31.3 nM and 15.6 nM or 2 μM, 1 μM, 0.5 μM, 0.25 μM and 0.125 μM. The surface was regenerated by injecting 0.85% phosphoric acid at a flow rate of 100 μl / min for 18 seconds, once horizontally and once vertically.

[0623] Intervals (interactions with the sensor surface) and blanks (PBS-T-EDTA) were subtracted from the raw data. The sensor spectra were then fitted to a 1:1 kinetic model to provide rate constants (ka, kd) and affinity (KD) values ​​for hu DLL3 and cy DLL3, and globally fitted to equilibrium to provide affinity (KD) values ​​for hu CD3εγ and cy CD3εγ.

[0624] The DLL3 / CD3 binding proteins described herein show affinities for DLL3 in the pM range. The affinity for the CD3 epsilon gamma subunit is in the low nM range. Interspecies differences between humans and cynomolgus macaques are balanced. The affinities of three exemplary DLL3 / CD3 binding proteins (DLL3 / CD3 binding proteins comprising a DLL3 chain of SEQ ID NO: 75 and a CD3 chain of SEQ ID NO: 79, SEQ ID NO: 80 or SEQ ID NO: 105, respectively) are shown in Table 2A.

[0625] Table 2A: Affinity of DLL3 / CD3 binding proteins for human and cynomolgus monkey DLL3 and CD3εγ subunit as determined by SPR analysis (KD, and interspecies differences

[0626]

[0627] Example 4B: SPR-based determination of affinity for recombinant DLL3 and CD3εγ

[0628] The binding affinity of the purified DLL3 / CD3 bispecific constructs to recombinant human DLL3-ECD and Fc fusion proteins of the human CD3 epsilon gamma subunit was determined by surface plasmon resonance (SPR) using a ProteOn XPR36 instrument (Bio-Rad). The method was similar to Example 4A, with minor differences as outlined below.

[0629] For the binding kinetics assay for DLL3, anti-DLL3 / CD3 molecules were captured vertically for 200 seconds on a Protein A / G surface at a flow rate of 30 μl / min, respectively. The baseline was stabilized by injecting PBS-T-EDTA horizontally for 60 seconds at a flow rate of 30 μl / min. The analyte (hu DLL3) was injected horizontally over the captured antibody for 300 seconds at a flow rate of 30 μl / min and dissociated for 1800 seconds. The concentrations of the analyte were 20 nM, 10 nM, 5 nM, 2.5 nM, and 1.25 nM. The surface was regenerated by injecting 0.85% phosphoric acid at a flow rate of 100 μl / min for 18 seconds, once horizontally and once vertically.

[0630] For binding kinetics assays to CD3, HuCD3E_HuCD3G-Fc (dissolved in pH 4.5 acetate) was directly immobilized on the sensor surface via amine coupling vertically for 360 seconds at a flow rate of 30 μl / min. The baseline was stabilized by injecting PBS-T-EDTA horizontally for 60 seconds at a flow rate of 30 μl / min. Analytes (all anti-DLL3 / CD3 molecules) were injected horizontally over the captured antibodies for 600 seconds at a flow rate of 30 μl / min and dissociated for 1800 seconds. The concentrations of analytes were 250 nM, 83.3 nM, 27.8 nM, 9.3 nM and 3.1 nM. The surface was regenerated by injecting 0.85% phosphoric acid at a flow rate of 100 μl / min for 18 seconds, once horizontally and once vertically.

[0631] Affinities of exemplary DLL3 / CD3 binding proteins (comprising a DLL3 / CD3 binding protein of SEQ ID NO:73, SEQ ID NO:74, SEQ ID NO:75, SEQ ID NO:76, SEQ ID NO:77, SEQ ID NO:78, SEQ ID NO:241, 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, SEQ ID NO:251, or SEQ ID NO:252 and a CD3 chain of SEQ ID NO:79; a DLL3 binding protein comprising a DLL3 chain of SEQ ID NO:75 and a CD3 chain of SEQ ID NO:80; and a DLL3 binding protein comprising a DLL3 chain of SEQ ID NO:75 and a CD3 chain of SEQ ID NO:105) are shown in 2B.

[0632] Table 2B: Affinity (KD) of DLL3 / CD3 binding proteins for human DLL3 and CD3εγ subunit as determined by SPR analysis

[0633]

[0634]

[0635] Example 5: Generation of HEK293 cells expressing the extracellular domain and its subdomains on the cell surface

[0636] To generate stable HEK293 cells expressing full-length human DLL3 (Uniprot: Q9NYJ7), macaque DLL3 (UPI0003AB95BD), human DLL1 (Q00548) and human DLL4 (Q9NR61), respectively, the respective coding sequences were cloned in pcDNA3.1 () and the FLAG tag was inserted between the signal sequence and the extracellular domain. Expression on the cell surface was examined by using an anti-FLAG antibody (clone M2, Sigma) followed by a monoclonal anti-mouse IgG1 (Acris).

[0637] The extracellular domain of DLL3 is composed of different subdomains:

[0638] Hu DLL3 signal peptide: aa 1-26

[0639] Hu DLL3 N-terminal ECD domain: aa 27-175

[0640] Hu DLL3 DSL: aa 176-215

[0641] Hu DLL3 EGF1: aa 216-249

[0642] Hu DLL3 EGF2: aa 274-310

[0643] Hu DLL3 EGF3: aa 312-351

[0644] Hu DLL3 EGF4: aa 353-389

[0645] Hu DLL3 EGF5: aa 391-427

[0646] Hu DLL3 EGF6: aa 429-465

[0647] · Proximal membrane peptide: aa 466-492

[0648] The following subdomains of DLL3 are expressed on the cell surface of HEK293 cells:

[0649] ·Hu DLL3 EGF1+2:Uniprot: Q9NYJ7 aa 216-310

[0650] ·Hu DLL3 EGF 2+3:Uniprot: Q9NYJ7 aa 274-351

[0651] ·Hu DLL3 EGF3+4:Uniprot: Q9NYJ7 aa 312-389

[0652] ·Hu DLL3 EGF4+5:Uniprot: Q9NYJ7 aa 353-427

[0653] ·Hu DLL3 EGF5+6:Uniprot: Q9NYJ7 aa 391-465

[0654] Hu DLL3 EGF6+ membrane-proximal peptide: Uniprot: Q9NYJ7 aa 429-492

[0655] To generate HEK293 cells expressing 6-His-myc tagged subdomains of human DLL3, the respective coding sequences were cloned in pcDNA 3.1 (Thermo Fisher Scientific). The constructs contained an N-terminal mouse IgG Vk leader sequence followed by a 6-Histidin-myc tag and the respective human DLL3 domain. To ensure cell surface localization of the human DLL3 domain, all constructs were followed by a Ser / Gly-linker, the transmembrane domain (aa266-288) and the intracellular domain (aa289-314) of human EpCAM (P16422). Subdomain expression was verified using mouse monoclonal IgG2a antibodies to the various domains as described in WO 2013126746. Bound monoclonal antibodies were detected with goat anti-mouse IgG (F(ab')2)-RPE (Jackson Immuno Research). Samples were measured by flow cytometry. The expression of human DLL1 and DLL4 was confirmed by FACS analysis using anti-DLL1 (R&D Systems, MAB1818) and anti-DLL4 (R&D Systems MAB1506) antibodies followed by PE-labeled anti-mouse or anti-rat secondary antibodies (see Figure 5 , right).

[0656] Table 3: Sequences of various DLL3 subdomains expressed on HEK293 cells

[0657]

[0658]

[0659]

[0660] Example 6: Epitope Domain Mapping of DLL3 / CD3 Binding Protein

[0661] Determination of epitope domains by binding DLL3 / CD3 proteins to recombinant cell lines expressing DLL3 subdomains (see Figure 2 ). The generation of these cell lines is described under Example 5.

[0662] The DLL3 / CD3 binding protein was prepared by transiently transfecting CHO-E cells with the pTT5 vector carrying the strand encoding gene as described in Example 2.

[0663] Cells expressing the DLL3 subdomains were stained with two-step purified DLL3 / CD3 binding protein at 1.6 nM in FACS buffer (PBS / 0.5% BSA / 0.05% sodium azide). Bound molecules were detected with PE-conjugated anti-human secondary antibodies (Sigma-Aldrich, #P8047). As a negative control, cells were incubated with secondary antibodies only. Samples were measured by flow cytometry. Figure 3A -C shows exemplary binders that bind to different regions of DLL3 (comprising the following DLL3 / CD3 binding proteins: SEQ ID NO: 73, SEQ ID NO: 74, SEQ ID NO: 75, SEQ ID NO: 76, SEQ ID NO: 77, SEQ ID NO: 78, SEQ ID NO: 241, 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, SEQ ID NO: 251 or SEQ ID NO: 252 DLL3 chain and SEQ ID NO: 79 CD3 chain, and DLL3 binding proteins comprising SEQ ID NO: 75 DLL3 chain and SEQ ID NO: 80 CD3 chain). The sequences of the epitope domains are listed in Example 5, Table 3.

[0664] Example 7: Interspecies cross-reaction

[0665] Cross-species reactivity is demonstrated by the interaction of DLL3 / CD3 binding protein with NCI-H82 (HTB-175 TM ) and SHP77( CRL-2195 TM ) cells, two SCLC cell lines, and binding assays with a recombinant cell line expressing macaque DLL3 (cell line generation described in Example 5).

[0666] DLL3 / CD3 binding protein was prepared by transiently transfecting CHO-E cells with pTT5 vector carrying the strand encoding gene, as described in Example 2. Cells expressing human or macaque DLL3 were stained with 1.6 nM of two-step purified DLL3 / CD3 binding protein in FACS buffer (PBS / 0.5% BSA / 0.05% sodium azide). Bound molecules were detected with PE-conjugated anti-human secondary antibody (Sigma-Aldrich, #P8047). As a negative control, cells were incubated with secondary antibody alone. Samples were measured by flow cytometry. Figure 4 Binding to human and cynomolgus monkey DLL3 expressing cells with seven exemplary DLL3 / CD3 binding proteins (including the following DLL3 / CD3 binding proteins: the DLL3 chain of SEQ ID NO:73 and the CD3 chain of SEQ ID NO:79, the DLL3 chain of SEQ ID NO:74 and the CD3 chain of SEQ ID NO:79, the DLL3 chain of SEQ ID NO:75 and the CD3 chain of SEQ ID NO:79, the DLL3 chain of SEQ ID NO:75 and the CD3 chain of SEQ ID NO:80, the DLL3 chain of SEQ ID NO:76 and the CD3 chain of SEQ ID NO:79, the DLL3 chain of SEQ ID NO:77 and the CD3 chain of SEQ ID NO:79, or the DLL3 chain of SEQ ID NO:78 and the CD3 chain of SEQ ID NO:79) is shown.

[0667] Example 8: Confirmation of the absence of binding to human DLL1 and DLL4

[0668] The absence of cross-reactivity with human DLL1 and DLL4 was determined by binding of the DLL3 / CD3 binding protein to recombinant cells expressing human DLL1 or DLL4 (produced in the cell lines described in Example 5).

[0669] The DLL3 / CD3 binding protein was prepared by transiently transfecting CHO-E cells with the pTT5 vector carrying the strand encoding gene as described in Example 2.

[0670] Cells expressing DLL1 and DLL4 were stained with 1.6 nM of two-step purified DLL3 / CD3 binding protein at increasing concentrations in FACS buffer (PBS / 0.5% BSA / 0.05% sodium azide). Bound molecules were detected with PE-conjugated anti-human secondary antibodies (Sigma-Aldrich, #P8047). As a negative control, cells were incubated with secondary antibodies alone. Samples were measured by flow cytometry. Expression of human DLL1 and DLL4 was confirmed by FACS analysis using anti-DLL1 (R&D Systems, MAB1818) and anti-DLL4 (R&D Systems, MAB1506) antibodies followed by PE-labeled anti-mouse or anti-rat secondary antibodies. Figure 5 Binding of seven exemplary DLL3 / CD3 binding proteins (including the DLL3 chain of SEQ ID NO:73 and the CD3 chain of SEQ ID NO:79, the DLL3 chain of SEQ ID NO:74 and the CD3 chain of SEQ ID NO:79, the DLL3 chain of SEQ ID NO:75 and the CD3 chain of SEQ ID NO:79, the DLL3 chain of SEQ ID NO:75 and the CD3 chain of SEQ ID NO:80, the DLL3 chain of SEQ ID NO:76 and the CD3 chain of SEQ ID NO:79, the DLL3 chain of SEQ ID NO:77 and the CD3 chain of SEQ ID NO:79, or the DLL3 chain of SEQ ID NO:78 and the CD3 chain of SEQ ID NO:79) to human DLL1 and DLL4 expressing cells is shown.

[0671] Example 9: Binding of DLL3 / CD3 binding protein to SCLC cell lines and T cells

[0672] Through NCI-H82 (HTB-175TM) and SHP77 ( The binding of DLL3 / CD3 binding protein to human SCLC cell lines was tested by flow cytometry of 100 μg / mL CRL-2195™). DLL3 / CD3 binding protein was prepared as described in Example 2.

[0673] By Ficoll density gradient centrifugation, human peripheral blood mononuclear cells (PBMC) are prepared from the preparation (leukocyte layer) (collecting the byproduct of the blood bank of the blood for transfusion) of enriched lymphocyte.After obtaining informed consent according to the Declaration of Helsinki and under the approval of the cantonal ethics committee of Austria, all leukocyte layers are obtained and PBMC is prepared on the same day of collection.Therefore, by Ficoll density gradient centrifugation (35min, without deceleration at 1400rpm), mononuclear cells are separated and washed in large quantities with PBS. Remaining red blood cells were removed by incubation in ACK lysis buffer (Thermo Fisher Scientific, A1049201) for 3 minutes, followed by washing in PBS and then resuspended in assay medium containing RPMI 1640 GlutaMAX (Gibco #61870-010), 5% human AB serum AB (Gemini, GemCell cat #100-512LOT #H56500I) + 1% MEM-NEAA (Gibco #11140-035), 10 mM HEPES (Affymetrix #7365-49-9), 10 μM β-mercaptoethanol (Gibco #21985-023) and sodium pyruvate (Gibco #11360-039).

[0674] T cells were isolated by negative selection using the Disk T Cell Isolation Kit II (Miltenyi Biotec #130-091-156). Briefly, cells were resuspended in 40 μl buffer PBS / 0.5% BSA (Gibco ref #041-94553M) / 2mM EDTA (Invitrogen ref #15575-038) / 10 Mio cells and incubated with 10 μl biotin-antibody mixture per 10 Mio cells for 5 min at 4°C. Next, 30 μl buffer and 20 μl anti-biotin MicroBeds / 10 million cells were added and incubated for 10 min at 4°C. Next, the mixture was placed in a pre-rinsed 25LS column (Miltenyi Biotec #130-042-401) in a magnetic field suitable for a MACS separator (Miltenyi Biotec). The flow-through was collected and washed in assay medium.

[0675] Cells (T cells or human SCLC cells) were stained with increasing concentrations of two-step purified DLL3 / CD3 binding proteins in FACS buffer (PBS / 0.5% BSA / 0.05% sodium azide). Bound molecules were detected with PE-conjugated anti-human secondary antibodies (Sigma-Aldrich, #P8047). DLL3 / CD3 binding proteins targeting DSL (DLL3#6), EGF1 (DLL3#5), or EGF4 (DLL#4) domains showed stronger binding to SCLC cell lines expressing DLL3 than DLL3 / CD3 binding proteins targeting proximal membrane peptides (DLL3#1, DLL3#2, DLL3#3). Binding of exemplary DLL3 / CD3 binding proteins for different epitope domains (comprising DLL3 / CD3 binding proteins of SEQ ID NO:73, SEQ ID NO:74, SEQ ID NO:75, SEQ ID NO:76, SEQ ID NO:77, SEQ ID NO:78, SEQ ID NO:241, 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, SEQ ID NO:251, or SEQ ID NO:252 and a CD3 chain of SEQ ID NO:79; a DLL3 binding protein comprising a DLL3 chain of SEQ ID NO:75 and a CD3 chain of SEQ ID NO:80) to SHP77, NCI-H82 cells, and T cells is shown in Fig. 6A -V.

[0676] Example 10A: Efficacy of redirecting non-stimulated PBMCs to human SCLC cell lines

[0677] The efficacy of non-stimulated PBMCs against SCLC cell lines was determined using lactate dehydrogenase (LDH) release as a readout for cell lysis. In this assay, DLL3-positive SCLC cell lines, SHP77 and NCI-H82, were co-cultured with human PBMCs as effector cells and increasing concentrations of DLL3 / CD3 binding protein at an effector:target cell ratio of 10:1 for 72 hours. DLL3 / CD3 binding protein (comprising the following DLL3 / CD3 binding proteins: SEQ ID NO:73, SEQ ID NO:74, SEQ ID NO:75, SEQ ID NO:76, SEQ ID NO:77, SEQ ID NO:78 DLL3 chain and SEQ ID NO:79 CD3 chain, and DLL3 binding protein comprising SEQ ID NO:75 DLL3 chain and SEQ ID NO:80 CD3 chain) was prepared by transiently transfecting CHO-E cells with pTT5 vectors carrying the chain encoding genes as described in Example 2.

[0678] By Ficoll density gradient centrifugation, human peripheral blood mononuclear cells (PBMC) are prepared from the preparation (leukocyte layer) (collecting the byproduct of the blood bank of the blood for transfusion) of enriched lymphocyte.After obtaining informed consent according to the Declaration of Helsinki and under the approval of the cantonal ethics committee of Austria, all leukocyte layers are obtained and PBMC is prepared on the same day of collection.Therefore, by Ficoll density gradient centrifugation (35min, without deceleration at 1400rpm), mononuclear cells are separated and washed in large quantities with PBS. Remaining red blood cells were removed by incubation in ACK lysis buffer (Thermo Fisher Scientific, A1049201) for 3 minutes, followed by washing in PBS and then resuspended in assay medium containing RPMI 1640 GlutaMAX (Gibco #61870-010), 5% human AB serum AB (Gemini, GemCell cat #100-512LOT #H56500I) + 1% MEM-NEAA (Gibco #11140-035), 10 mM HEPES (Affymetrix #7365-49-9), 10 μM β-mercaptoethanol (Gibco #21985-023) and sodium pyruvate (Gibco #11360-039).

[0679] Next, target cells, SHP77 and NCI-H82, and PBMC were incubated with DLL3 / CD3 binding protein at a ratio of 1:10 at concentrations ranging from 0.0001 nM to 100 nM for 72 hours.

[0680] Using Cytotoxicity Assay Kit PLUS(Roche), according to the manufacturer's instructions to determine cytotoxic activity. In brief, this method is based on the use of LDH released from dead or plasma membrane damaged cells. Cell culture supernatant was incubated with the reaction mixture from the kit for 30 minutes and formazan formation caused by LDH activity as a substitute for cell lysis was measured in a spectrophotometer at 500nm. The percentage of cytotoxicity relative to the maximum lysis control was calculated according to the following formula:

[0681]

[0682] Background: target cells + effector cells

[0683] Maximum lysis: target cells + 5% Triton X-100

[0684] Minimal lysis: target cells

[0685] The percentage of cytotoxicity relative to the maximum lysis control was plotted against the corresponding DLL3 / CD3 binding protein concentration using GraphPad Prism 5 software. Dose-response curves were analyzed using a four-parameter logistic equation model to estimate sigmoidal dose-response curves and calculate EC 90 Value (EC 90 Values ​​are shown in Table 4A).

[0686] Fig. 7A Examples of the efficacy of cell lysis of seven exemplary DLL3 / CD3 binding proteins (comprising the DLL3 chain of SEQ ID NO:73, SEQ ID NO:74, SEQ ID NO:75, SEQ ID NO:76, SEQ ID NO:77 or SEQ ID NO:78, and the CD3 chain of SEQ ID NO:79, and a DLL3 binding protein comprising the DLL3 chain of SEQ ID NO:75 and the CD3 chain of SEQ ID NO:80) that bind to different epitope domains are shown.

[0687] Table 4A: EC of DLL3 / CD3 binding proteins as measured in a 72-hour cytotoxicity assay with non-stimulated PBMCs as effector cells and SCLC cell lines, NCI-H82 and SHP77 as target cells 90 Value [nM].

[0688] DLL3 / CD3 binding protein NCI-H82 SHP77 DLL3#1 / CD3#1 1.7 1.4 DLL3#2 / CD3#1 2.9 3.4 DLL3#3 / CD3#1 0.9 1.5 DLL3#3 / CD3#2 3.9 16.5 DLL3#4 / CD3#1 No saturation at 100 nM No saturation at 100 nM DLL3#5 / CD3#1 No saturation at 100 nM No saturation at 100nM DLL3#6 / CD3#1 No saturation at 100 nM No saturation at 100 nM

[0689] Example 10B: Efficacy of redirected ...

Claims

1. An anti-DLL3 antibody molecule comprising i) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 1 (CDR1), SEQ ID NO: 2 (CDR2) and SEQ ID NO: 3 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 4 (CDR1), SEQ ID NO: 5 (CDR2) and SEQ ID NO: 6 (CDR3); or ii) a light chain CDR comprising the amino acid sequence of SEQ ID NO:7 (CDR1), SEQ ID NO:8 (CDR2) and SEQ ID NO:9 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO:10 (CDR1), SEQ ID NO:11 (CDR2) and SEQ ID NO:12 (CDR3); or iii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 13 (CDR1), SEQ ID NO: 14 (CDR2) and SEQ ID NO: 15 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 16 (CDR1), SEQ ID NO: 17 (CDR2) and SEQ ID NO: 18 (CDR3); or iv) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 19 (CDR1), SEQ ID NO: 20 (CDR2) and SEQ ID NO: 21 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 22 (CDR1), SEQ ID NO: 23 (CDR2) and SEQ ID NO: 24 (CDR3); or v) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 25 (CDR1), SEQ ID NO: 26 (CDR2) and SEQ ID NO: 27 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 28 (CDR1), SEQ ID NO: 29 (CDR2) and SEQ ID NO: 30 (CDR3); or vi) a light chain CDR comprising the amino acid sequence of SEQ ID NO:31 (CDR1), SEQ ID NO:32 (CDR2) and SEQ ID NO:33 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO:34 (CDR1), SEQ ID NO:35 (CDR2) and SEQ ID NO:36 (CDR3); or vii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 133 (CDR1), SEQ ID NO: 134 (CDR2) and SEQ ID NO: 135 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 136 (CDR1), SEQ ID NO: 137 (CDR2) and SEQ ID NO: 138 (CDR3); or viii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 139 (CDR1), SEQ ID NO: 140 (CDR2) and SEQ ID NO: 141 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 142 (CDR1), SEQ ID NO: 143 (CDR2) and SEQ ID NO: 144 (CDR3); or ix) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 145 (CDR1), SEQ ID NO: 146 (CDR2) and SEQ ID NO: 147 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 148 (CDR1), SEQ ID NO: 149 (CDR2) and SEQ ID NO: 150 (CDR3); or x) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 151 (CDR1), SEQ ID NO: 152 (CDR2) and SEQ ID NO: 153 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 154 (CDR1), SEQ ID NO: 155 (CDR2) and SEQ ID NO: 156 (CDR3); or xi) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 157 (CDR1), SEQ ID NO: 158 (CDR2) and SEQ ID NO: 159 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 160 (CDR1), SEQ ID NO: 161 (CDR2) and SEQ ID NO: 162 (CDR3); or xii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 163 (CDR1), SEQ ID NO: 164 (CDR2) and SEQ ID NO: 165 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 166 (CDR1), SEQ ID NO: 167 (CDR2) and SEQ ID NO: 168 (CDR3); or xiii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 169 (CDR1), SEQ ID NO: 170 (CDR2) and SEQ ID NO: 171 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 172 (CDR1), SEQ ID NO: 173 (CDR2) and SEQ ID NO: 174 (CDR3); or xiv) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 175 (CDR1), SEQ ID NO: 176 (CDR2) and SEQ ID NO: 177 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 178 (CDR1), SEQ ID NO: 179 (CDR2) and SEQ ID NO: 180 (CDR3); or xv) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 181 (CDR1), SEQ ID NO: 182 (CDR2) and SEQ ID NO: 183 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 184 (CDR1), SEQ ID NO: 185 (CDR2) and SEQ ID NO: 186 (CDR3); or xvi) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 187 (CDR1), SEQ ID NO: 188 (CDR2) and SEQ ID NO: 189 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 190 (CDR1), SEQ ID NO: 191 (CDR2) and SEQ ID NO: 192 (CDR3); or xvii) a light chain CDR comprising the amino acid sequence of SEQ ID NO: 193 (CDR1), SEQ ID NO: 194 (CDR2) and SEQ ID NO: 195 (CDR3); and a heavy chain CDR comprising the amino acid sequence of SEQ ID NO: 196 (CDR1), SEQ ID NO: 197 (CDR2) and SEQ ID NO: 198 (CDR3); or xviii) a light chain CDR comprising the amino acid sequences of SEQ ID NO: 199 (CDR1), SEQ ID NO: 200 (CDR2) and SEQ ID NO: 201 (CDR3); and a heavy chain CDR comprising the amino acid sequences of SEQ ID NO: 202 (CDR1), SEQ ID NO: 203 (CDR2) and SEQ ID NO: 204 (CDR3).

2. The anti-DLL3 antibody molecule of claim 1, wherein the antibody molecule is a monoclonal, chimeric, humanized or human antibody molecule.

3. The anti-DLL3 antibody molecule of claim 1 or 2, wherein the antibody molecule is Fab, F(ab')2, Fab', Fv or scFv.

Images