EA046123B1 - BISPECIFIC ANTIBODY CONSTRUCTION TARGETS MUC17 AND CD3 - Google Patents

Description

Technical field

The present invention relates to products and methods from the field of biotechnology, in particular to constructs based on bispecific antibodies, their preparation and their use.

Prerequisites for creating an invention

Some of the fastest and most promising emerging therapeutics are protein-based pharmaceuticals, which already play an important role in almost all areas of medicine and are among the most sought after therapeutics in (pre-)clinical development and as commercial products. (Leader, Nature Reviews Drug Discovery 2008 Jan 7, 21-39). Compared to small molecule drugs, protein pharmaceuticals have high specificity and activity at relatively low concentrations and are typically intended for the treatment of diseases that significantly reduce quality of life, such as various types of cancer, autoimmune diseases and metabolic disorders. (Roberts, Trends Biotechnol. 2014 Jul; 32(7):372-80, Wang, Int J Pharm. 1999 Aug 20;185(2):129-88).

Such new protein-based pharmaceuticals contain, for example, bispecific (monoclonal) antibodies, which typically can simultaneously bind to two different types of antigen. They are known in several structural formats, and to date, their applications for cancer immunotherapy and drug delivery have been explored (Fan, Gaowei; Wang, Zujian; Hao, Mingju; Li, Jinming (2015). Bispecific antibodies and their applications. Journal of Hematology & Oncology 8: 130).

Bispecific antibodies may refer to bispecific antibodies of the IgG type, i.e. full-length, or to bispecific antibodies other than the IgG type, which are constructs not based on full-length antibodies. Full-length bispecific antibodies typically retain the traditional monoclonal antibody (mAb) structure of two Fab arms and one F region, except that the two Fab sites bind to different antigens. Partial-length bispecific antibodies may completely lack the Fc region. These include chemically linked Fabs consisting of Fab regions only, and various types of divalent and trivalent single chain variable fragments (scFv). There are also fusion proteins that mimic the variable domains of two antibodies. An example of such a format is the Bispecific Tissue Recruiter (BiTE®) (Yang, Fa; Wen, Weihong; Qin, Weijun (2016). Bispecific Antibodies as a Development Platform for New Concepts and Treatment Strategies. International Journal of Molecular Sciences. 18 (1) :48).

Antibody-derived bispecific molecules, such as BiTE® antibody constructs, are recombinant protein constructs composed of two antibody-derived binding domains linked by a flexible linkage. One binding domain of BiTE® antibody constructs is specific for a selected tumor-associated surface antigen on target cells; the second binding domain is specific for CD3, a subunit of the T cell receptor complex on T cells. Due to their special structure, BiTE® antibody constructs are uniquely suited to transiently link T cells to target cells and, at the same time, effectively activate the cytolytic potential of T cells towards target cells. An important further development of the first generation BiTE® antibody constructs (see WO 99/54440 and WO 2005/040220), clinically developed as AMG 103 and AMG 110, was the provision of bispecific antibody constructs that bind to a context-independent epitope on N-terminus of the CD3ε chain (WO 2008/119567). BiTE® antibody constructs binding to this selected epitope demonstrate not only cross-species specificity for the human CD3ε chain and Macasa or Callithrix jacchus, Saguinus oedipus or Saimiri sciureus, but also by recognizing this specific epitope (instead of previously described CD3 binding epitopes molecules in bispecific T cell recruiting molecules), they do not exhibit nonspecific T cell activation to the same extent as was observed with the previous generation of T cell recruiting antibodies. This decrease in T cell activation was associated with less or reduced T cell redistribution in patients, the latter being associated with the risk of adverse events such as with pasotuximab.

The antibody-based constructs described in WO 2008/119567 are characterized by rapid clearance from the body; therefore, although they are capable of rapidly reaching most parts of the body, their in vivo applications may be limited by their short-term presence in vivo. On the other hand, their concentration in the body can be adapted and adjusted in a short time. Long-term administration by continuous intravenous infusion is used to achieve therapeutic effects due to the short in vivo half-life of this small single-chain molecule. However, bispecific antibody constructs are now available that have superior pharmacokinetic properties, including longer half-lives. The increased half-life is generally useful in in vivo applications of immunoglobulins, especially antibodies and most especially small-sized antibody fragments or constructs, for example, for patient compliance.

- 1 046123

Mucins have been identified as interesting markers of inflammatory and cancer diseases. Mucins are high molecular weight glycoproteins that are characterized by high levels of O-glycosylation at serine and threonine residues within tandem repeat domains (Johansson and Hansson, Nat. Rev. Immunology 2016). There are at least 20 members of the mucin family, including secreted proteins and transmembrane proteins, that are expressed by epithelial cells in various tissues (Corfield, Biochim. Biophys. Acta 2013). The primary function of mucins is to structure and regulate the mucus layer, which forms a protective barrier between epithelial cells and the environment (Hollingsworth and Swanson, Nat. Rev. Cancer 2004; Hattrup and Gendler, Annu. Rev. Physiol. 2008). Transmembrane mucins also play a role in cell signaling, including the regulation of proliferation and apoptosis, and in tumor formation (Hollingsworth and Swanson, Nat. Rev. Cancer 2004). Among the mucins, mucin 17 (MUC17) is a transmembrane mucin that was originally identified by its homology to MUC3 (Gum et al., Biochem. Biophys. Res. Comm. 2002).

Analysis of the complete coding sequence of MUC17 revealed that it has a large extracellular domain consisting of a central region of 61 tandem repeats, an epidermal growth factor (EGF) domain, sea urchin sperm protein, an enterokinase and agrin (SEA) domain, and a second EGF domain. The SEA domain contains a putative cleavage site that is conserved in other mucins (Moniaux et al., J. Biol. Chem. 2006). MUC17 is a bitopic transmembrane protein with an 80 amino acid cytoplasmic tail that is intracellular (Moniaux et al., J. Biol. Chem. 2006). MUC17 expression in healthy adults is restricted to the apical surface of enterocytes, or mature absorptive epithelial cells, that line the intestine (Moniaux et al., J. Biol. Chem. 2006; Johanasson and Hansson, Nat. Rev. Immunology 2016). MUC17 is also expressed in the stomach and pancreas (Moniaux et al., J. Biol. Chem. 2006; Moehle et al., J. Mol. Med. 2006). The biological function of MUC17 is thought to be to maintain the integrity of the mucosal barrier in the digestive tract, such as mucosal repair (Luu et al., Int. J. Biochem. Cell Biol. 2010; Resta-Lenert et al., Am. J. Physiology 2011; Johanasson and Hansson, Nat. Rev. Immunology 2016).

MUC17 is misexpressed in some cancers. MUC17 mRNA has been shown to be expressed in one pancreatic cancer cell line and three colon cancer cell lines (Gum et al. 2002). Immunohistochemical studies have confirmed the expression of MUC17 protein in pancreatic cancer (Moniauxet al. 2006). However, it has been shown that MUC17 protein expression was decreased in colon cancer (Senapati et al., J. Clin. Pathol. 2010). However, the expression pattern of MUC17 makes it a potential target for the treatment of various forms of malignancies.

Short description

Given the conflicting findings in the literature regarding the pathological condition for which MUC17 is a potential target, the purpose of the present invention is to clearly identify specific conditions associated with increased expression of MUC17 and provide constructs based on bispecific antibodies, such as T-recruiting molecules. cells that are specifically suited to bind MUC17 in a MUC17-associated condition are preferably used in the treatment of these particular conditions. Accordingly, the present invention provides an antibody construct characterized by comprising a first domain that binds to MUC17, a second domain that binds to an extracellular epitope of the human and/or non-human CD3ε chain, e.g., Masasa, and preferably a third domain , which is a specific form of Fc. Furthermore, the present invention provides a polynucleotide encoding an antibody construct, a vector containing the polynucleotide, and host cells expressing the construct, and a pharmaceutical composition containing the foregoing.

In a first aspect, the present invention provides for the provision of an antibody construct comprising:

the first domain that binds to MUC17; and a second domain that binds to an extracellular epitope of the human and Macasa CD3ε chain.

Within this aspect, the present invention further provides that the antibody construct comprises a third domain that contains two polypeptide monomers, each of which contains a hinge region, a CH2 domain and a CH3 domain, wherein said two polypeptide monomers are fused to each other by a peptide linker.

Within this aspect, the present invention further provides for the provision of an antibody construct that is a single chain antibody construct.

Within this aspect, it is also contemplated within the context of the present invention to provide an antibody construct wherein said third domain comprises, from the amino terminus to the carboxyl terminus, a hinge region-CH2-CH3-linker-hinge region-CH2-CH3.

Within this aspect, the present invention further provides for the provision of an antibody construct wherein each of said polypeptide monomers has an amino acid sequence that is at least 90% identical to a sequence selected from the group consisting of SEQ ID NOs: 17-24.

Within this aspect, it is also contemplated within the context of the present invention to provide an antibody construct wherein each of said polypeptide monomers has an amino acid sequence selected from SEQ ID NOs: 17-24.

Within this aspect, the present invention further provides for the provision of an antibody construct wherein the CH2 domain contains an intradomain cysteine disulfide bridge.

Within this aspect, the present invention also provides for the provision of an antibody-based construct, wherein:

(i) the first domain contains two antibody variable domains, and the second domain contains two antibody variable domains;

(ii) the first domain contains one antibody variable domain, and the second domain contains two antibody variable domains;

(iii) the first domain contains two antibody variable domains, and the second domain contains one antibody variable domain; or (iv) the first domain contains one antibody variable domain and the second domain contains one antibody variable domain.

Within this aspect, it is also contemplated within the context of the present invention to provide an antibody construct wherein the first and second domains are fused to a third domain via a peptide linker.

Within this aspect, it is also contemplated within the context of the present invention to provide an antibody construct, wherein the antibody construct comprises, from the amino terminus to the carboxyl terminus:

(a) first domain;

(b) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 1-3;

(c) second domain.

Within this aspect, the present invention also provides for the provision of an antibody construct, wherein the antibody construct, in addition to (a)-(c), comprises, in the amino-terminal to carboxyl-terminal direction:

(d) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 1, 2, 3, 9, 10, 11 and 12;

(e) the first polypeptide monomer of the third domain;

(f) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 5, 6, 7 and 8; and (g) a second polypeptide monomer of the third domain.

Within this aspect, the present invention further provides for the provision of an antibody construct, wherein the first domain of the antibody construct binds to an epitope within MUC17 that corresponds to SEQ ID NO: 528 (aa 4171-4296 according to Uniprot numbering Q685J3) .

Within this aspect, the present invention also provides for the provision of an antibody construct wherein the first domain of the antibody construct binds to an epitope within MUC17 that corresponds to SEQ ID NO: 529 (aa 4184-4291 according to Uniprot numbering Q685J3) .

Within this aspect, the present invention further provides for the provision of an antibody construct, wherein the first domain of the antibody construct binds to an epitope within MUC17 that corresponds to SEQ ID NO: 530 (aa 4131-4243 according to Uniprot numbering Q685J3) .

Within this aspect, the present invention also provides for the provision of an antibody construct, wherein the first domain of the antibody construct binds to an epitope within MUC17 that corresponds to SEQ ID NO: 531 (aa 4244-4389 according to Uniprot numbering Q685J3) .

Within this aspect, the present invention further provides for the provision of an antibody construct, wherein the first domain of the antibody construct binds to an epitope within MUC17 that corresponds to SEQ ID NO: 530 (aa 4131-4243 according to Uniprot numbering Q685J3) , but not with the epitope within MUC17, which corresponds to SEQ ID NO: 531 (aa 4244-4389 according to Uniprot numbering Q685J3).

Within this aspect, it is also contemplated within the context of the present invention to provide an antibody construct wherein the first domain of the antibody construct binds to an epitope within MUC17 that corresponds to SEQ ID NO: 532 (aa 4171-4390 according to the numbering Q685J3 in Uniprot) or SEQ ID NO: 533 (aa 4184-4390 according to Uniprot numbering Q685J3), but not with an epitope within MUC17 that corresponds to SEQ ID NO: 534 (aa 4291-4390 according to Q685J3 numbering Uniprot), or with an epitope within MUC17 that corresponds to SEQ ID NO: 535 (aa 4341-4390 according to Uniprot numbering Q685J3).

Within this aspect, the present invention further provides for the provision of an antibody construct wherein the VH VL moiety is given as 4 lambda 3. The nomenclature is known in the art.

Within this aspect, the present invention further provides for the provision of an antibody construct wherein the ratio of cytotoxicity to binding affinity (EC ₅₀ /K _d )x1000 is less than 250, wherein the cytotoxicity is indicated in pM and determined using NUGC-4 cells as in target cells and in huPBMC as effector cells, and the binding affinity is indicated in nM and determined using a surface plasmon resonance (SPR) assay, such as the Biacore assay. A factor of 1000 was introduced for better legibility, given the different values for typical EC ₅₀ and KD values.

Within this aspect, it is also contemplated within the context of the present invention to provide an antibody construct wherein the ratio of cytotoxicity to binding affinity (EC ₅₀ /K _d ) x 1000 is less than 125, wherein the cytotoxicity is indicated in pM and determined, for example, using NUGC- cells 4 as target cells and huPBMC as effector cells, wherein the binding affinity is indicated in nM and determined, for example, using a surface plasmon resonance (SPR) based assay.

Within this aspect, the present invention further provides for the provision of an antibody construct wherein the ratio of cytotoxicity to binding affinity (EC ₅₀ /KD)x1000 is less than 21, wherein the cytotoxicity is indicated in pM and determined, for example, using NUGC-4 cells as target cells and in huPBMC as effector cells, and the binding affinity is indicated in nM and determined using a surface plasmon resonance (SPR) based assay. Preferably, the cytotoxicity (EC ₅₀ ) is <100 pM and the binding affinity (KD) is <25 nM.

Within this aspect, it is also contemplated within the context of the present invention to provide an antibody construct wherein the first binding domain comprises a VH region comprising CDR-H1, CDR-H2 and CDR-H3 selected from:

(a) CDR-H1 shown under SEQ ID NO: 33, CDR-H2 shown under SEQ ID NO: 34, and CDRH3 shown under SEQ ID NO: 35;

(b) CDR-H1 shown under SEQ ID NO: 44, CDR-H2 shown under SEQ ID NO: 45, and CDRH3 shown under SEQ ID NO: 46;

(c) CDR-H1 shown under SEQ ID NO: 55, CDR-H2 shown under SEQ ID NO: 56, and CDRH3 shown under SEQ ID NO: 57;

(d) CDR-H1, shown under SEQ ID NO: 66, CDR-H2, shown under SEQ ID NO: 67, and CDRH3, shown under SEQ ID NO: 68;

(e) CDR-H1 shown under SEQ ID NO: 77, CDR-H2 shown under SEQ ID NO: 78, and CDRH3 shown under SEQ ID NO: 79;

(f) CDR-H1, shown under SEQ ID NO: 88, CDR-H2, shown under SEQ ID NO: 89, and CDRH3, shown under SEQ ID NO: 90;

(g) CDR-H1, shown under SEQ ID NO: 99, CDR-H2, shown under SEQ ID NO: 100, and CDRH3, shown under SEQ ID NO: 101;

(h) CDR-H1 shown under SEQ ID NO: 110, CDR-H2 shown under SEQ ID NO: 111, and CDR-H3 shown under SEQ ID NO: 112;

(i) CDR-H1 shown under SEQ ID NO: 121, CDR-H2 shown under SEQ ID NO: 122, and CDR-H3 shown under SEQ ID NO: 123;

(j) CDR-H1 shown under SEQ ID NO: 132, CDR-H2 shown under SEQ ID NO: 133, and CDR-H3 shown under SEQ ID NO: 134;

(k) CDR-H1 shown under SEQ ID NO: 143, CDR-H2 shown under SEQ ID NO: 144, and CDR-H3 shown under SEQ ID NO: 145;

(l) CDR-H1 shown under SEQ ID NO: 154, CDR-H2 shown under SEQ ID NO: 155, and CDR-H3 shown under SEQ ID NO: 156;

(m) CDR-H1 shown under SEQ ID NO: 165, CDR-H2 shown under SEQ ID NO: 166, and CDR-H3 shown under SEQ ID NO: 167;

(n) CDR-H1 shown under SEQ ID NO: 176, CDR-H2 shown under SEQ ID NO: 177, and CDR-H3 shown under SEQ ID NO: 178;

(o) CDR-H1 shown under SEQ ID NO: 187, CDR-H2 shown under SEQ ID NO: 188, and

- 4 046123

CDR-H3 shown under SEQ ID NO: 189;

(p) CDR-H1, shown under SEQ ID NO: 198, CDR-H2, shown under SEQ ID NO: 199, and

CDR-H3 given under SEQ ID NO: 200;

(q) CDR-H1 shown under SEQ ID NO: 209, CDR-H2 shown under SEQ ID NO: 210, and

CDR-H3, shown under SEQ ID NO: 211;

(r) CDR-H1 shown under SEQ ID NO: 220, CDR-H2 shown under SEQ ID NO: 221, and CDR-H3 shown under SEQ ID NO: 222;

(s) CDR-H1 shown under SEQ ID NO: 231, CDR-H2 shown under SEQ ID NO: 232, and CDR-H3 shown under SEQ ID NO: 233;

(t) CDR-H1 shown under SEQ ID NO: 242, CDR-H2 shown under SEQ ID NO: 243, and CDR-H3 shown under SEQ ID NO: 244;

(u) CDR-H1 shown under SEQ ID NO: 253, CDR-H2 shown under SEQ ID NO: 254, and CDR-H3 shown under SEQ ID NO: 255;

(v) CDR-H1 shown under SEQ ID NO: 264, CDR-H2 shown under SEQ ID NO: 265, and CDR-H3 shown under SEQ ID NO: 266;

(w) CDR-H1 shown under SEQ ID NO: 275, CDR-H2 shown under SEQ ID NO: 276, and CDR-H3 shown under SEQ ID NO: 276;

(x) CDR-H1 shown under SEQ ID NO: 286, CDR-H2 shown under SEQ ID NO: 287, and CDR-H3 shown under SEQ ID NO: 288;

(y) CDR-H1 shown under SEQ ID NO: 297, CDR-H2 shown under SEQ ID NO: 298, and CDR-H3 shown under SEQ ID NO: 299;

(z) CDR-H1 shown under SEQ ID NO: 308, CDR-H2 shown under SEQ ID NO: 309, and CDR-H3 shown under SEQ ID NO: 310;

(aa) CDR-H1 shown under SEQ ID NO: 319, CDR-H2 shown under SEQ ID NO: 320, and CDR-H3 shown under SEQ ID NO: 321;

(ab) CDR-H1 shown under SEQ ID NO: 330, CDR-H2 shown under SEQ ID NO: 331, and CDR-H3 shown under SEQ ID NO: 332;

(ac) CDR-H1 shown under SEQ ID NO: 341, CDR-H2 shown under SEQ ID NO: 342, and CDR-H3 shown under SEQ ID NO: 343;

(ad) CDR-H1 shown under SEQ ID NO: 352, CDR-H2 shown under SEQ ID NO: 353, and CDR-H3 shown under SEQ ID NO: 354;

(ae) CDR-H1 shown under SEQ ID NO: 363, CDR-H2 shown under SEQ ID NO: 364, and CDR-H3 shown under SEQ ID NO: 365;

(af) CDR-H1 shown under SEQ ID NO: 374, CDR-H2 shown under SEQ ID NO: 375, and CDR-H3 shown under SEQ ID NO: 376;

(ag) CDR-H1 shown under SEQ ID NO: 385, CDR-H2 shown under SEQ ID NO: 386, and CDR-H3 shown under SEQ ID NO: 386;

(ah) CDR-H1 shown under SEQ ID NO: 396, CDR-H2 shown under SEQ ID NO: 397, and CDR-H3 shown under SEQ ID NO: 398;

(ai) CDR-H1 shown under SEQ ID NO: 407, CDR-H2 shown under SEQ ID NO: 408, and CDR-H3 shown under SEQ ID NO: 409;

(aj) CDR-H1 shown under SEQ ID NO: 418, CDR-H2 shown under SEQ ID NO: 419, and CDR-H3 shown under SEQ ID NO: 420;

(ak) CDR-H1 shown under SEQ ID NO: 429, CDR-H2 shown under SEQ ID NO: 430, and CDR-H3 shown under SEQ ID NO: 431;

(al) CDR-H1 shown under SEQ ID NO: 440, CDR-H2 shown under SEQ ID NO: 441, and CDR-H3 shown under SEQ ID NO: 442;

(am) CDR-H1 shown under SEQ ID NO: 451, CDR-H2 shown under SEQ ID NO: 452, and CDR-H3 shown under SEQ ID NO: 453;

(an) CDR-H1 shown under SEQ ID NO: 462, CDR-H2 shown under SEQ ID NO: 463, and CDR-H3 shown under SEQ ID NO: 464;

(ao) CDR-H1 shown under SEQ ID NO: 473, CDR-H2 shown under SEQ ID NO: 474, and CDR-H3 shown under SEQ ID NO: 475;

(ap) CDR-H1 shown under SEQ ID NO: 484, CDR-H2 shown under SEQ ID NO: 485, and CDR-H3 shown under SEQ ID NO: 486;

(aq) CDR-H1 shown under SEQ ID NO: 495, CDR-H2 shown under SEQ ID NO: 496, and CDR-H3 shown under SEQ ID NO: 497;

(ar) CDR-H1 shown under SEQ ID NO: 506, CDR-H2 shown under SEQ ID NO: 507, and CDR-H3 shown under SEQ ID NO: 508; and (as) CDR-H1 shown under SEQ ID NO: 517, CDR-H2 shown under SEQ ID NO: 518, and CDR-H3 shown under SEQ ID NO: 519; where the preferred ones are:

(c) CDR-H1 given at SEQ ID NO: 55, CDR-H2 given at SEQ ID NO: 56, and CDR-5 046123

H3, shown under SEQ ID NO: 57;

(n) CDR-H1 shown under SEQ ID NO: 176, CDR-H2 shown under SEQ ID NO: 177, and

CDR-H3, shown under SEQ ID NO: 178;

(ac) CDR-H1, shown under SEQ ID NO: 341, CDR-H2, shown under SEQ ID NO: 342, and

CDR-H3, shown under SEQ ID NO: 343; and (aj) CDR-H1 set forth in SEQ ID NO: 418, CDR-H2 set forth in SEQ ID NO: 419, and CDR-H3 set forth in SEQ ID NO: 420.

Within this aspect, it is also contemplated within the context of the present invention to provide an antibody construct wherein the first binding domain comprises a VL region comprising CDR-H1, CDR-L2 and CDR-L3 selected from:

(a) CDR-L1 given at SEQ ID NO: 36, CDR-L2 given at SEQ ID NO: 37, and CDRL3 given at SEQ ID NO: 38;

(b) CDR-L1 given at SEQ ID NO: 47, CDR-L2 given at SEQ ID NO: 48, and CDRL3 given at SEQ ID NO: 49;

(c) CDR-L1 given at SEQ ID NO: 58, CDR-L2 given at SEQ ID NO: 59, and CDRL3 given at SEQ ID NO: 60;

(d) CDR-L1, shown under SEQ ID NO: 69, CDR-L2, shown under SEQ ID NO: 70, and CDRL3, shown under SEQ ID NO: 71;

(e) CDR-L1, shown under SEQ ID NO: 80, CDR-L2, shown under SEQ ID NO: 81, and CDRL3, shown under SEQ ID NO: 82;

(f) CDR-L1, shown under SEQ ID NO: 91, CDR-L2, shown under SEQ ID NO: 92, and CDRL3, shown under SEQ ID NO: 93;

(g) CDR-L1 shown under SEQ ID NO: 102, CDR-L2 shown under SEQ ID NO: 103, and CDR-L3 shown under SEQ ID NO: 104;

(h) CDR-L1 shown under SEQ ID NO: 113, CDR-L2 shown under SEQ ID NO: 114, and CDR-L3 shown under SEQ ID NO: 115;

(i) CDR-L1 given at SEQ ID NO: 124, CDR-L2 given at SEQ ID NO: 125, and CDR-L3 given at SEQ ID NO: 126;

(j) CDR-L1 shown under SEQ ID NO: 135, CDR-L2 shown under SEQ ID NO: 136, and CDR-L3 shown under SEQ ID NO: 137;

(k) CDR-L1 shown under SEQ ID NO: 146, CDR-L2 shown under SEQ ID NO: 147, and CDR-L3 shown under SEQ ID NO: 148;

(l) CDR-L1 shown under SEQ ID NO: 157, CDR-L2 shown under SEQ ID NO: 158, and CDR-L3 shown under SEQ ID NO: 159;

(m) CDR-L1 set forth at SEQ ID NO: 168, CDR-L2 set forth at SEQ ID NO: 169, and CDR-L3 set forth at SEQ ID NO: 170;

(n) CDR-L1 shown under SEQ ID NO: 179, CDR-L2 shown under SEQ ID NO: 180, and CDR-L3 shown under SEQ ID NO: 181;

(o) CDR-L1 set forth at SEQ ID NO: 190, CDR-L2 set forth at SEQ ID NO: 191, and CDR-L3 set forth under SEQ ID NO: 192;

(p) CDR-L1 given under SEQ ID NO: 201, CDR-L2 given under SEQ ID NO: 202, and CDR-L3 given under SEQ ID NO: 203;

(q) CDR-L1 shown under SEQ ID NO: 212, CDR-L2 shown under SEQ ID NO: 213, and CDR-L3 shown under SEQ ID NO: 214;

(r) CDR-L1 given at SEQ ID NO: 223, CDR-L2 given at SEQ ID NO: 224, and CDR-L3 given at SEQ ID NO: 225;

(s) CDR-L1 set forth at SEQ ID NO: 234, CDR-L2 set forth at SEQ ID NO: 235, and CDR-L3 set forth at SEQ ID NO: 236;

(t) CDR-L1 given at SEQ ID NO: 245, CDR-L2 given at SEQ ID NO: 246, and CDR-L3 given at SEQ ID NO: 247;

(u) CDR-L1 given at SEQ ID NO: 256, CDR-L2 given at SEQ ID NO: 257, and CDR-L3 given at SEQ ID NO: 258;

(v) CDR-L1 given at SEQ ID NO: 267, CDR-L2 given at SEQ ID NO: 268, and CDR-L3 given at SEQ ID NO: 269;

(w) CDR-L1 given at SEQ ID NO: 278, CDR-L2 given at SEQ ID NO: 279, and CDR-L3 given at SEQ ID NO: 280;

(x) CDR-L1 given at SEQ ID NO: 289, CDR-L2 given at SEQ ID NO: 290, and CDR-L3 given at SEQ ID NO: 291;

(y) CDR-L1 given under SEQ ID NO: 300, CDR-L2 given under SEQ ID NO: 301, and CDR-L3 given under SEQ ID NO: 302;

(z) CDR-L1 given under SEQ ID NO: 311, CDR-L2 given under SEQ ID NO: 312, and CDR-L3 given under SEQ ID NO: 313;

- 6 046123 (aa) CDR-L1 given under SEQ ID NO: 322, CDR-L2 given under SEQ ID NO: 323, and

CDR-L3 given under SEQ ID NO: 324;

(ab) CdR-LI, shown under SEQ ID NO: 333, CDR-L2, shown under SEQ ID NO: 334, and

CDR-L3 given under SEQ ID NO: 335;

(ac) CDR-L1 given at SEQ ID NO: 344, CDR-L2 given at SEQ ID NO: 345, and CDR-L3 given at SEQ ID NO: 346;

(ad) CDR-L1 set forth at SEQ ID NO: 355, CDR-L2 set forth at SEQ ID NO: 356, and CDR-L3 set forth under SEQ ID NO: 357;

(ae) CDR-L1 shown under SEQ ID NO: 366, CDR-L2 shown under SEQ ID NO: 367, and CDR-L3 shown under SEQ ID NO: 368;

(af) CDR-L1 shown under SEQ ID NO: 377, CDR-L2 shown under SEQ ID NO: 378, and CDR-L3 shown under SEQ ID NO: 379;

(ag) CDR-L1 shown under SEQ ID NO: 388, CDR-L2 shown under SEQ ID NO: 389, and CDR-L3 shown under SEQ ID NO: 390;

(ah) CdR-LI shown under SEQ ID NO: 399, CDR-L2 shown under SEQ ID NO: 400, and CDR-L3 shown under SEQ ID NO: 401;

(ai) CDR-L1 given under SEQ ID NO: 410, CDR-L2 given under SEQ ID NO: 411, and CDR-L3 given under SEQ ID NO: 412;

(aj) CDR-L1 given at SEQ ID NO: 421, CDR-L2 given at SEQ ID NO: 422, and CDR-L3 given at SEQ ID NO: 423;

(ak) CDR-L1 given at SEQ ID NO: 432, CDR-L2 given at SEQ ID NO: 433, and CDR-L3 given at SEQ ID NO: 434;

(al) CDR-L1 shown under SEQ ID NO: 443, CDR-L2 shown under SEQ ID NO: 444, and CDR-L3 shown under SEQ ID NO: 445;

(am) CDR-L1 set forth at SEQ ID NO: 454, CDR-L2 set forth at SEQ ID NO: 455, and CDR-L3 set forth under SEQ ID NO: 456;

(an) CDR-L1 shown under SEQ ID NO: 465, CDR-L2 shown under SEQ ID NO: 466, and CDR-L3 shown under SEQ ID NO: 467;

(ao) CDR-L1 given under SEQ ID NO: 476, CDR-L2 given under SEQ ID NO: 477, and CDR-L3 given under SEQ ID NO: 478;

(ap) CDR-L1 given under SEQ ID NO: 487, CDR-L2 given under SEQ ID NO: 488, and CDR-L3 given under SEQ ID NO: 489;

(aq) CDR-L1 shown under SEQ ID NO: 498, CDR-L2 shown under SEQ ID NO: 499, and CDR-L3 shown under SEQ ID NO: 500;

(ar) CDR-L1 set forth at SEQ ID NO: 509, CDR-L2 set forth at SEQ ID NO: 510, and CDR-L3 set forth under SEQ ID NO: 511; and (as) CDR-L1 given under SEQ ID NO: 520, CDR-L2 given under SEQ ID NO: 521, and CDR-L3 given under SEQ ID NO: 522; where the preferred ones are:

(c) CDR-L1 set forth at SEQ ID NO: 58, CDR-L2 set forth at SEQ ID NO: 59, and CDRL3 set forth at SEQ ID NO: 60;

(n) CDR-L1 set forth at SEQ ID NO: 179, CDR-L2 set forth at SEQ ID NO: 180, and CDRL3 set forth at SEQ ID NO: 181;

(ac) CDR-L1 set forth at SEQ ID NO: 344, CDR-L2 set forth at SEQ ID NO: 345, and CDR-L3 set forth at SEQ ID NO: 346; and (aj) CDR-L1 set forth in SEQ ID NO: 421, CDR-L2 set forth in SEQ ID NO: 422, and CDR-L3 set forth in SEQ ID NO: 423.

Within this aspect, it is also contemplated within the context of the present invention to provide an antibody construct wherein the first binding domain comprises a VL region and a VH region selected from the group consisting of:

(a) the VL region shown under SEQ ID NO: 40 and the VH region shown under SEQ ID NO: 39;

(b) the VL region shown under SEQ ID NO: 51 and the VH region shown under SEQ ID NO: 50;

(c) the VL region of SEQ ID NO: 62 and the VH region of SEQ ID NO: 61;

(d) the VL region of SEQ ID NO: 73 and the VH region of SEQ ID NO: 72;

(e) the VL region of SEQ ID NO: 84 and the VH region of SEQ ID NO: 83;

(f) the VL region of SEQ ID NO: 95 and the VH region of SEQ ID NO: 94;

(g) the VL region of SEQ ID NO: 106 and the VH region of SEQ ID NO: 105;

(h) the VL region of SEQ ID NO: 117 and the VH region of SEQ ID NO: 116;

(i) the VL region of SEQ ID NO: 128 and the VH region of SEQ ID NO: 127;

(j) the VL portion of SEQ ID NO: 139 and the VH portion of SEQ ID NO:

- 7 046123

138;

(k) the VL section shown under SEQ ID NO: 149;

(l) the VL region shown under SEQ ID NO: 160;

(m) the VL region shown under SEQ ID NO: 171;

(n) the VL region shown under SEQ ID NO: 182;

(o) the VL region shown under SEQ ID NO: 193;

(p) the VL region shown under SEQ ID NO: 204;

(q) the VL region shown under SEQ ID NO: 215;

(r) the VL region shown under SEQ ID NO: 226;

(s) the VL region shown under SEQ ID NO: 237;

(t) the VL region shown under SEQ ID NO: 248;

(u) the VL section shown under SEQ ID NO: 259;

(v) the VL region shown under SEQ ID NO: 270;

(w) the VL region shown under SEQ ID NO: 281;

(x) the VL region shown under SEQ ID NO: 292;

(y) the VL region shown under SEQ ID NO: 303;

(z) the VL region shown under SEQ ID NO: 314;

(aa) the VL region shown under SEQ ID NO: 325;

(ab) the VL region shown under SEQ ID NO: 336;

(ac) the VL region shown under SEQ ID NO: 347;

(ad) the VL region shown under SEQ ID NO: 358;

(ae) the VL region shown under SEQ ID NO: 369;

(af) the VL region shown under SEQ ID NO: 380;

(ag) the VL region shown under SEQ ID NO: 391;

(ah) the VL region shown under SEQ ID NO: 402;

(ai) the VL section shown under SEQ ID NO: 413;

(aj) the VL region shown under SEQ ID NO: 424;

(ak) the VL section shown under SEQ ID NO: 435;

(al) the VL region shown under SEQ ID NO: 446;

(am) the VL region shown under SEQ ID NO 457;

(an) the VL region shown under SEQ ID NO: 468;

(ao) of the VL section given under SEQ ID NO:

150, and the VH section given under SEQ ID NO:

161, and the VH section given under SEQ ID NO:

172, and the VH section given under SEQ ID NO:

183, and the VH section given under SEQ ID NO:

194, and the VH section given under SEQ ID NO:

205, and the VH section given under SEQ ID NO:

216, and the VH section given under SEQ ID NO:

227, and the VH section given under SEQ ID NO:

238, and the VH section given under SEQ ID NO:

249, and the VH section given under SEQ ID NO:

260, and the VH section given under SEQ ID NO:

271, and the VH section given under SEQ ID NO:

282, and the VH section given under SEQ ID NO:

293, and the VH section given under SEQ ID NO:

304, and the VH section given under SEQ ID NO:

315, and the VH section given under SEQ ID NO:

326, and the VH section given under SEQ ID NO:

337, and the VH region given under SEQ ID NO:

348, and the VH region given under SEQ ID NO:

359, and the VH region given under SEQ ID NO:

370, and the VH section given under SEQ ID NO:

381, and the VH region given under SEQ ID NO:

392, and the VH region given under SEQ ID NO:

403, and the VH section given under SEQ ID NO:

414, and the VH section given under SEQ ID NO:

425, and the VH section given under SEQ ID NO:

436, and the VH section given under SEQ ID NO:

447, and the VH region shown under SEQ ID NO: 458, and the VH region shown under SEQ ID NO:

469, and the VH region given under SEQ ID NO:

480, and the VH section given under SEQ ID NO:

- 8 046123

479;

(ap) the VL region of SEQ ID NO: 491 and the VH region of SEQ ID NO: 490;

(aq) the VL region of SEQ ID NO: 502 and the VH region of SEQ ID NO: 501;

(ar) the VL region of SEQ ID NO: 513 and the VH region of SEQ ID NO: 512; and (as) the VL portion of SEQ ID NO: 524 and the VH portion of SEQ ID NO: 523.

Within this aspect, the present invention further provides for the provision of an antibody construct, wherein the antibody construct comprises a sequence selected from the amino acid sequence set forth in any one of SEQ ID NO: 41, 52, 63, 74, 85, 96, 107 , 118, 129, 140, 151, 162, 173, 184, 195, 206, 217, 228, 239, 250, 261, 272, 283, 294, 305, 316, 327, 338, 349, 360, 371, 3 82 , 393, 404, 415, 426, 437, 448, 459, 470, 481, 492, 503, 514, and 525.

Within this aspect, it is also contemplated within the context of the present invention to provide an antibody construct, wherein the antibody construct comprises, from the amino terminus to the carboxyl terminus:

(a) a first domain having an amino acid sequence selected from the group consisting of SEQ ID NO: 41, 52, 63, 74, 85, 96, 107, 118, 129, 140, 151, 162, 173, 184, 195, 206, 217, 228, 239, 250, 261, 272, 283, 294, 305, 316, 327, 338, 349, 360, 371, 382, 393, 404, 415, 426, 437, 448, 459, 470 , 481, 492, 503, 514, and 525;

(b) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 1-3;

(c) a second domain having an amino acid sequence selected from the group consisting of SEQ ID NO: 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185 or 187 in WO 2008/119567, or given under SEQ ID NO: 15; and (d) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 1, 2, 3, 9, 10, 11 and 12.

Within this aspect, the present invention also provides for the provision of an antibody-based construct, wherein the antibody-based construct further comprises, in addition to (a)-(d), in the amino-terminal to carboxyl-terminal direction:

(e) a first third domain polypeptide monomer having a polypeptide sequence selected from the group consisting of SEQ ID NOs: 17-24;

(f) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 5, 6, 7 and 8; and (g) a second third domain polypeptide monomer having a polypeptide sequence selected from the group consisting of SEQ ID NOs: 17-24.

Within this aspect, the present invention also provides for the provision of an antibody construct having an amino acid sequence selected from the group consisting of SEQ ID NO: 42, 43, 53, 54, 64, 65, 75, 76, 86, 87, 97, 98, 108, 109, 119, 120, 130, 131,

141, 142, 152, 153, 163, 164, 174, 175, 185, 186, 196, 197, 207, 208, 218, 219, 229, 230, 240, 241, 251,252,

262, 263, 273, 274, 284, 285, 295, 296, 306, 307, 317, 318, 328, 329, 339, 340, 350, 351, 361, 362, 372,373,

383, 384, 394, 395,405, 406, 416, 417, 427, 428, 438, 439, 449, 450, 460, 461, 471, 472, 482, 483, 493,494,

504, 505, 515, 516, 526 and 527.

In a second aspect, the present invention further provides for the provision of a polynucleotide encoding an antibody construct of the present invention.

In a third aspect, the present invention also provides for the provision of a vector containing a polynucleotide of the present invention.

In a fourth aspect, the present invention further provides for providing a host cell transformed or transfected with a polynucleotide or vector of the present invention.

In a fifth aspect, the present invention also provides a method for producing an antibody construct of the present invention, the method comprising culturing a host cell of the present invention under conditions allowing expression of the antibody construct, and isolating the resulting antibody construct from culture.

In a sixth aspect, the present invention further provides for the provision of a pharmaceutical composition comprising an antibody construct of the present invention or prepared in accordance with a method of the present invention.

Within the scope of this aspect, it is also contemplated in the context of the present invention that the pharmaceutical composition is stable for at least four weeks at approximately -20°C.

- 9 046123

It is further contemplated within the context of the present invention to provide a construct based on an antibody of the present invention or produced in accordance with a method of the present invention for use in the prevention, treatment or amelioration of a disease selected from a proliferative disease, a neoplastic disease, cancer or an immunological disorder.

Within this aspect, it is also contemplated in the context of the present invention that the disease is gastrointestinal cancer (eg, gastric cancer, esophageal cancer, gastroesophageal cancer or colorectal cancer) or pancreatic cancer.

Within this aspect, it is also contemplated in the context of the present invention that the disease is gastric cancer.

In a seventh aspect, the present invention further provides a method of treating or ameliorating a proliferative disease, neoplastic disease, cancer or immunological disorder, comprising the step of administering to a subject in need thereof an antibody construct of the present invention or obtained in accordance with a method of the present invention. invention, wherein the disease is preferably gastrointestinal cancer or pancreatic cancer, most preferably gastric cancer.

In an eighth aspect, the present invention also provides for the provision of a kit comprising an antibody construct of the present invention or produced in accordance with a method of the present invention, a polynucleotide of the present invention, a vector of the present invention, and/or a host cell of the present invention.

In a ninth aspect, the present invention further provides a method of treating or ameliorating gastrointestinal cancer, comprising the step of administering to a subject in need thereof a bispecific antibody construct directed against MUC17 and CD3.

In a tenth aspect, the present invention further provides a bispecific antibody construct directed against MUC17 and CD3 for use in the treatment or amelioration of gastrointestinal cancer.

Description of graphic materials

In fig. Figure 1 shows binding to MUC17 epitope clusters. Epitopes E1, E2, E3, E4, E5A and 5B, as well as truncated variants of E2 (TR2, TR3, TR4 and TR5, respectively), are noted. Experiments with constructs where human MUC17 (brown/gray) was replaced by nonfunctional mouse MUC3 identified the corresponding epitopes. Forty-five anti-MUC17 scFc bispecific antibody constructs were identified that span the E2 epitope space containing the SEA domain.

Fig. 2. Mapping of the MUC17 epitope using cell surface binding of constructs based on bispecific antibodies with scFc to MUC17 in relation to cells expressing human/mouse chimeric constructs. Cell surface binding was assessed by fluorescence activated cell sorting (FACS), where loss of binding to the chimeric construct indicates that the corresponding (mutated) domain is required for binding of bispecific antibody constructs to the MUC17 scFc. For example, E2 exhibits loss of binding upon mutation. Therefore, E2 is required for binding of all four bispecific antibody constructs studied.

Fig. 3. MUC17 is expressed in gastric, pancreatic and colorectal cancer cell lines. Cell surface protein expression of MUC17 was determined by live cell flow cytometry and depicted as FACS readouts (A). MUC17 mRNA levels in cancer cell lines were determined using quantitative polymerase chain reaction (qPCR). Values are normalized to those of the constitutively expressed gene (B).

Fig. 4. Cytotoxicity assay using three different MUC17-carrying cell lines with different MUC17 expression (A: GSU, B: NUGC-4 and C: Ls174T). The designs tested are 1=32-G6; 2=1-B6; 3=2-C2 and 4=8-B7. Design 8-B7 is slightly more favorable in terms of cytotoxicity.

Fig. 5. Soluble MUC17 protein (sMUC17, aa 4131-4243 in Uniprot) was added to TDCC assays at 0-1000 ng/ml and the activity of constructs based on bispecific antibodies with scFc to MUC17 was assessed after incubation for 48 hours (target cells GSU (A) or NUGC-4 (B), human T cell to target cell ratio 10:1, Steady Glo readout). The addition of sMUC17 did not affect the cytotoxic activity of the bispecific antibody constructs.

Fig. 6. Design 8-B7 based on an antibody with scFc to MUC17 inhibits tumor growth in a xenograft model of colorectal cancer. Female NOD/SCID mice were implanted with ^2x106 Ls174T colorectal cancer cells. On day 15, activated T cells were administered in an increased amount of 2x10 ⁷ by intraperitoneal (IP) injection. Anti-MUC17 scFc antibody construct was dosed IP on day 16 and day 22. Tumor size was measured using calipers.

Fig. 7. List of preferred bispecific antibody constructs according to the present invention with group code (optimization library), molecule designation, epitope cluster to which the corresponding construct binds, affinity (K _D ) for SPR in [nM], cytotoxic activity (EC5 ₀ ) in NUGC-4 cells in [pM], their ratio (EC ₅₀ /K _d )x1000 and VH VL grouping.

Fig. 8. Anti-MUC17 scFc antibody construct 8-B7 also increased half-life in cynomolgus monkey (A). Exposure levels are consistent with expected exposures. Cynomolgus monkeys (B) (n=3 per group) were administered 100 mg/kg or 1000 mg/kg MUC17 HLE BiTE® at 0 h and 168 h. Serum was collected at the indicated time points and analyzed for the presence of a bispecific construct. anti-MUC17 scFc antibodies using either anti-CD3 or anti-MUC17 based ELISA. The data were fitted to a two-component model. The graph shows the individual data (dots) and the average (line).

Detailed Description of the Invention

The present invention provides a bispecific antibody construct specifically targeting MUC17 associated with cancer. In this regard, the first MUC17 was identified as a gene that is characterized by increased expression in gastric tumors relative to expression in normal tissue. In this regard, MUC17 protein has been shown to be expressed in 40-77% of gastric tumors according to immunohistochemical methods common in the art. Flow cytometry also demonstrated that MUC17 protein is expressed on the cell surface of gastric cancer cell lines and esophageal cancer cell lines, in addition to some pancreatic cancer cell lines and colorectal cancer cell lines. It has even been shown that such expression is particularly higher in gastric tumors from Chinese patients. Therefore, MUC17 has been identified as a useful target associated with gastrointestinal cancer, i.e. stomach cancer, cancer of the small intestine and large intestine (colon), cancer of the esophagus and cancer of the pancreas.

An unexpected finding in the context of the present invention is that the bispecific antibody constructs of the present invention preferentially target cancer cells, such as gastric and gastrointestinal cancer cells bearing MUC17, and are less likely to target cells other than cancerous MUC17 is normally expressed on the apical surface (i.e., the side away from the base of the corresponding cells) of intestinal epithelial cells other than cancer cells and forms part of the mucus layer. However, MUC17 is overexpressed in gastric and gastrointestinal cancers and is then not restricted to the apical surface, but is also expressed on non-apical surfaces. Without being limited by theory, MUC17 on the apical surface is considered less accessible to the bispecific antibody constructs of the present invention, while MUC17 expressed on a non-apical surface is more accessible in cancer cells. Therefore, the bispecific antibody constructs of the present invention preferably target MUC17-associated cancer cells and less preferably target non-cancerous cells. This was unexpectedly found when comparing good tolerability in healthy animals with high antitumor efficacy in an in vivo cancer model. In detail, although immunohistochemical study confirmed the expression of MUC17 on the apical surface of gastrointestinal tissue, such as small intestinal tissue collected from monkeys evaluated in a pilot toxicology study, predominantly no histopathological changes were detected in tissue expressing MUC17. The good tolerance of non-cancer cells to the bispecific antibody constructs of the present invention has likewise been demonstrated in vitro. In contrast, intravenous treatment of tumor-bearing mice with the bispecific antibody construct of the present invention results in statistically significant and dose-dependent inhibition of tumor growth compared to placebo-treated control mice. Accordingly, the bispecific antibody constructs of the present invention are preferably tolerated by the patient and exhibit a preferably well-controlled breadth of therapeutic action that has not previously been described for any MUC17-targeting agent.

The present invention provides bispecific antibody constructs that bind the EGF-SEA-EGF region of the MUC17 protein. Advantageously, targeting this region of the protein provides selectivity for nearby family members (MUC3A, MUC3B, MUC12; e.g., Hollingsworth and Swanson, Nat. Rev. Cancer 2004) and the ability to bind to membrane-bound MUC17. MUC17, like other transmembrane mucins, contains a potential cleavage site within the SEA domain.

- 11 046123

Accordingly, this document provides bispecific antibody constructs that target the EGF-SEA-EGF and CD3 region of MUC17 and have a single chain Fc format to specifically increase half-life. Advantageously, the bispecific antibody constructs of the present invention are preferably characterized by affinity for target cells bearing the target MUC17 (KD as a single digit number in nM) and potency (EC50<50 pM) to provide targeting to low or heterogeneous levels of MUC17 in cells tumors of interest.

The bispecific antibody constructs of the present invention are contemplated to be cross-reactive to, for example, cynomolgus MUC17 (in addition to human MUC17), to enable non-clinical toxicology studies. Significant sequence details of the cynomolgus macaque MUC17 EGF-SEA-EGF domain are presented here for the first time.

In the context of the present invention, it is contemplated that bispecific antibody constructs exhibit binding affinity, high cytotoxic activity, and the most consistent targeting of the SEA domain.

In the context of the present invention, it is contemplated that bispecific antibody constructs have a cysteine clamp, i.e. intramolecular disulfide bond, in the target binder to improve stability.

In the context of the present invention, it is provided that a bispecific antibody construct provided with a single chain fragment Fc(scFc) with an extended half-life (HLE) and directed against MUC17 is intended for use in the treatment of cancers of the gastrointestinal tract, including gastric cancer, gastrointestinal -esophageal cancer, esophageal cancer, pancreatic cancer and colorectal cancer.

In addition, in the context of the present invention, it is optionally, but advantageously, provided that the scFc antibody construct, i.e. HLE, provides intravenous dosing options that are administered only once a week, once every two weeks, once every three weeks, or even once every four weeks or less often.

In the context of the present invention, the preferred epitope to be targeted therapeutically is that identified by seed elimination of MUC17 tandem repeats, since they are highly glycosylated and repeat sequence. This leads, for example, to an undefined 376 aa region and a 177 aa region of the EGF-like/SEA domain. Advantageously, targeting the EGF-like/SEA domains provides selectivity for nearby family members such as MUC3, cross-reactivity with cynomolgus MUC17, and binding to cell membrane-bound MUC17. The present inventors then developed reagents and assays to assess T cell binding and redirection, activation, and cytokine release. Such assays were used to confirm that the preferred bispecific antibody constructs match the predefined product candidate profiles with respect to affinity, cytotoxic activity, and construct stability.

To determine the epitope(s) of preference for bispecific antibody constructs targeting MUC17, epitope mapping was performed as described herein. Preferred bispecific antibody constructs target the E2 epitope containing SEA domain. The E2 epitope contains the amino acid (aa) sequence characterized herein by SEQ ID NO: 528. This essentially corresponds to MUC17 aa 4171-4296 according to Uniprot numbering Q685J3. Generally, the aa numbering of MUC17 in the context of the present invention is always made or implied by reference to the numbering of MUC17 by Q685J3 in Uniprot. In contrast, bispecific antibody constructs targeting the E1 epitope of MUC17, e.g. epitope located N-terminal to the SEA domain (see Fig. 1) unexpectedly showed undesirable cross-reactivity with MUC3A and MUC3B, which will lead to off-target activity and ultimately an increased risk of side effects. Moreover, bispecific antibody constructs targeting the E3 and E4 epitopes located C-terminal to the SEA domain (see Fig. 1) unexpectedly did not cross-react with cynomolgus MUC17. Therefore, the bispecific antibody constructs of the present invention are envisioned to specifically and exclusively bind to the E2 epitope of MUC17.

Such preferred bispecific antibody constructs of the present invention may be further indicated based on their structure or their unique detailed epitope-binding characteristics. Preferred bispecific antibody constructs according to the present invention can be determined by calculating the new cytotoxicity-affinity ratio provided herein. For example, said ratio (EC ₅₀ /K _d )x1000 is preferably <(less) 250. Such a ratio is generally indicative of good binding to truncated variants of the E2 epitope, i.e. TR2 (truncated 2: SEQ ID NO: 532) and TR3 (truncated 3: SEQ ID NO:

- 12 046123

533), with a ratio > (greater than) 250 generally indicating good binding to TR2 but not TR3. In particular, more preferred constructs typically bind to the E2/E5A/part 5B and/or TR2/TR3 epitope cluster. They exhibit, for example, an (EC50: _Kd )x1000 ratio below about 21 and belong to related sequence families (eg, optimization library (ORT) nomenclature 4a, 4b, 5a and 10. Their VH/VL grouping is preferably described herein as 4 lambda 3 or 4I3.) Such structures are identified in the context of the present invention, for example, as 8-A7, 8-B7, 8-B8, 8-C7, 8-H8, 8-D7, 4-E7, 8-F9, 1A6, 8-Н9, 1-В6, 8-F11 and 5-Н1. Also preferred are constructs that bind to the E2/E5A/partial 5B epitope cluster and/or TR2/TR3 and that exhibit an EC ₅₀ :K _d ratio of less than approximately 125 and belong to sequence families (ORT library nomenclature) 1a, 1c, and 9 Their VH/VL grouping is described as 3 lambda 3 or 3I3. Such structures are identified in the context of the present invention, such as 2-D11, 8-E3, 32-G6, 2-C2, 9-C2, 1B10, 4-B1, 4-F6, 4-G4, 4-A8, 4 -B10, 4-H11 and 4-H2. Preferred, but less preferred than the above two sequence families, are binders that bind to the E2/part E5A/part 5B and/or TR2/part TR3 epitope cluster and exhibit an (EC ₅₀ /K _d )x1000 ratio below about 1500. typically between 250 and 1450, and belong to sequence families (ORT library nomenclature) 6, 7 and 8. Their VH/VL grouping is described as 2 kappa 3 or 3k3. Particularly preferred herein are designs 32-G6 (SEQ ID NO: 65), 1-B6 (SEQ ID NO: 483), 2-C2 (SEQ ID NO: 428) and 8-B7 (SEQ ID NO: 186). In the context of the present invention, affinity is generally measured using an SPC, such as the BiacoreB assay, and the results are typically reported in nM. Cytotoxic activity is typically determined using NUGC-4 cells as MUC17 target cells and unstimulated human PBMCs as CD3 effector cells.

In the context of the present invention, it is contemplated that preferred bispecific antibody constructs not only exhibit a favorable cytotoxicity to affinity ratio, but additionally exhibit sufficient stability characteristics to facilitate practical handling in the formulation, storage and administration of said constructs. Sufficient stability is, for example, characterized by a high monomer content (i.e., non-aggregated and/or unbound native molecule) after standard preparation, such as at least 65%, as determined by preparative size exclusion chromatography (SEC), more preferably at least 70% and even more preferably at least 75%. Turbidity was also measured, for example at 340 nm the optical absorbance at a concentration of 2.5 mg/ml should preferably be 0.025 or less, more preferably 0.020, for example, in order to infer the virtual absence of unwanted aggregates. Preferably high monomer content is maintained after incubation under stress conditions, such as freeze/thaw or incubation at 37 or 40°C.

Thus, the present invention provides an antibody construct comprising:

the first domain that binds to MUC17;

a second domain that binds to an extracellular epitope of the human and Macasa CD3ε chain; and optionally:

a third domain that contains two polypeptide monomers, each of which contains a hinge region, a CH2 domain and a CH3 domain, the two polypeptide monomers being fused to each other by a peptide linker.

In one embodiment, the present invention provides a bispecific antibody construct containing all three such domains.

The term antibody-based construct refers to a molecule whose structure and/or function is based on the structure and/or function of an antibody, for example, a full-length or whole immunoglobulin molecule. The antibody construct is therefore capable of binding to its particular target or antigen and/or it/they are/are derived from the heavy chain variable region (VH) and/or light chain variable region (VL) domains of the antibody or a fragment thereof. In addition, a domain that binds to its binding partner in accordance with the present invention is understood herein as the binding domain of an antibody construct in accordance with the present invention. Typically, the binding domain of the present invention provides the minimum structural requirements for the antibody to ensure target binding. This minimum requirement may, for example, be defined as the presence of at least three light chain CDRs (i.e., VL region CDR1, CDR2, and CDR3) and/or three heavy chain CDRs (i.e., VH-region CDR1, CDR2, and CDR3). ), preferably all six CDRs. An alternative approach to determining the minimum structural requirements of an antibody is to define the antibody epitope within the structure of the specific target, respectively, the protein domain of the target proteins forming the epitope region (epitope cluster), or by specifying the specific antibody,

- 13 046123 competing for the epitope of a specific antibody. Antibodies on which the constructs of the present invention are based include, for example, monoclonal, recombinant, chimeric, deimmunized, humanized and human antibodies.

The binding domain of an antibody construct according to the present invention may, for example, contain the above CDR groups. Preferably, the CDRs are located in the framework region of the antibody light chain variable region (VL) and the antibody heavy chain variable region (VH); however, it does not necessarily have to contain both sections. Fd fragments, for example, have two VH regions and often retain some of the antigen binding function of the intact antigen binding domain. Additional examples of the format of antibody fragments, antibody variants, or binding domains include (1) Fab fragment, a monovalent fragment with VL, VH, CL, and CH1 domains; (2) F(ab') ₂ fragment, a bivalent fragment with two Fab fragments connected by a disulfide bridge at the hinge region; (3) Fd fragment with two VH and CH1 domains; (4) an Fv fragment with VL and VH domains from one arm of the antibody, (5) a dAb fragment (Ward et al., (1989) Nature 341: 544-546), which has a VH domain; (6) a dedicated complementarity determining region (CDR); and (7) a single chain Fv (scFv), the latter being preferred (eg, derived from a scFv library). Examples of embodiments of antibody-based constructs in accordance with the present invention are described, for example, in WO 00/006605, WO 2005/040220, WO 2008/119567, WO 2010/037838, WO 2013/026837, WO 2013/026833, US 2014/ 0308285, US 2014/0302037, WO 2014/144722, WO 2014/151910 and WO 2015/048272.

Also within the definition of a binding domain or a domain that binds are full-length antibody fragments such as VH, VHH, VL, (s)dAb, Fv, Fd, Fab, Fab', F(ab') ₂ or r IgG (half-antibody) . The antibody constructs of the present invention may also contain modified antibody fragments, also called antibody variants, such as scFv, di-scFv or bi(s)-scFv, scFv-Fc, scFv-zip, scFab, Fab ₂ , Fab ₃ , diabodies, single chain diabodies, tandem diabodies (Tandab), tandem di-scFv, tandem tri-scFv, multibodies such as tribodies or tetrabodies, and single domain antibodies such as nanobodies or single variable domain antibodies containing only one variable domain that may be a VHH, VH or VL that specifically binds an antigen or epitope independent of other V regions or domains.

As used herein, the terms single chain Fv, single chain antibodies or scFv refer to single polypeptide chain antibody fragments that contain variable regions from both the heavy and light chains but lack constant regions. Typically, a single chain antibody further contains a polypeptide linker between the VH and VL domains, which allows the formation of the necessary structure that allows antigen binding. Single chain antibodies are discussed in detail by Pluckthun in Pharmacology of Monoclonal Antibodies, vol. 113, Rosenburg and Moore eds. Springer-Verlag, New York, pp. 269-315 (1994). Various methods for creating single-chain antibodies are known, including those described in US patents No. 4694778 and 5260203; in international publication of patent application No. WO 88/01649; Bird (1988) Science 242:423–442; Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883; Ward et al. (1989) Nature 334:54454; Skerra et al. (1988) Science 242:1038–1041. In specific embodiments, single chain antibodies may also be bispecific, multispecific, human and/or humanized and/or synthetic.

In addition, the definition of the term antibody construct includes monovalent, bivalent and polyvalent/multivalent constructs and therefore bispecific constructs that specifically bind only two antigenic structures, as well as polyspecific/multispecific constructs that specifically bind more than two antigenic structures, e.g. three, four or more, through specific binding domains. Moreover, the definition of the term antibody construct includes molecules consisting of only one polypeptide chain, as well as molecules consisting of more than one polypeptide chain, the chains of which may be either identical (homodimers, homotrimers or homo-oligomers) or different (heterodimers, heterotrimer or heterooligomer). Examples for the antibodies and variants or derivatives thereof defined above are described in, among others, Harlow and Lane, Antibodies a laboratory manual, CSHL Press (1988) and Using Antibodies: a laboratory manual, CSHL Press (1999), Kontermann and Dubel, Antibody Engineering , Springer, 2nd ed. 2010 and Little, Recombinant Antibodies for Immunotherapy, Cambridge University Press 2009.

As used herein, the term bispecific refers to an antibody construct that is at least bispecific, i.e. comprises at least a first binding domain and a second binding domain, wherein the first binding domain binds one antigen or target (in this case MUC17) and the second binding domain binds another antigen or target (in this case CD3). Accordingly, the antibody constructs of the present invention have specificity for at least two different antigens or targets. For example, the first domain preferably does not bind to an extracellular CD3ε epitope of one or more species, as described herein. The term target cell surface antigen refers to an antigenic structure that is expressed by a cell and that is present on the cell surface and is therefore accessible to the antibody construct described herein. It may be a protein, preferably an extracellular portion of a protein, or a carbohydrate structure, preferably a carbohydrate structure of a protein, such as a glycoprotein. Preferably it is a tumor antigen. The term bispecific antibody construct of the present invention also covers multispecific antibody constructs, such as trispecific antibody constructs where the latter include three binding domains, or constructs with more than three (e.g., four, five, etc.) types of specificity.

Given that the antibody constructs of the present invention are (at least) bispecific, they do not occur in nature and are very different from naturally occurring products. Therefore, a bispecific antibody construct or immunoglobulin is an artificial hybrid antibody or immunoglobulin with at least two different binding sites with different types of specificity. Constructs based on bispecific antibodies can be obtained using a variety of methods, including fusion of hybridomas or binding of Fab' fragments. See, for example, Songsivilai & Lachmann, Clin. Exp. Immunol. 79:315-321 (1990).

The at least two binding domains and variable domains (VH/VL) of the antibody construct of the present invention may or may not contain peptide linkers (spacer peptides). The term peptide linker according to the present invention provides an amino acid sequence by which the amino acid sequences of one (variable and/or binding) domain and another (variable and/or binding) domain of the antibody construct of the present invention are linked to each other. Peptide linkers can also be used to fuse the third domain with other domains of the antibody construct of the present invention. An important technical characteristic of such a peptide linker is that it has no polymerization activity. Suitable peptide linkers include those described in US patents 4751180 and 4935233 or WO 88/09344. Peptide linkers can also be used to attach other domains or modules or regions (such as half-life extending domains) to the antibody construct of the present invention.

The antibody constructs of the present invention are preferably antibody constructs produced in vitro. This term refers to an antibody-based construct as defined above, wherein all or part of the variable region (e.g., at least one CDR) is generated by selection on non-immune cells, e.g., by in vitro phage display, protein chip, or any other a method in which candidate sequences can be tested for their ability to bind an antigen. Preferably, the term excludes sequences created solely by genomic rearrangement in an animal immune cell. A recombinant antibody is an antibody produced using recombinant DNA technology or genetic engineering.

The term monoclonal antibody (mAb) or monoclonal antibody construct as used herein refers to an antibody derived from a substantially homogeneous population of antibodies, i.e. the individual antibodies making up the population are identical except for possible naturally occurring mutations and/or post-translational modifications (eg, isomerization, amidation), which may be present in minor quantities. Monoclonal antibodies are highly specific, being directed against one antigenic site or one determinant on an antigen, in contrast to traditional (polyclonal) antibody preparations, which typically contain different antibodies directed against different determinants (or epitopes). In addition to their specificity, monoclonal antibodies have the advantage that they are synthesized by the hybridoma in a culture uncontaminated with other immunoglobulins. The designation monoclonal refers to the nature of the antibody as being derived from a substantially homogeneous population of antibodies, and should not be construed as requiring the antibody to be produced by any particular method.

To obtain monoclonal antibodies, any technique that produces antibodies using continuous cultures of cell lines can be used. For example, monoclonal antibodies to be used can be produced using the hybridoma method first described in Koehler et al., Nature, 256: 495 (1975), or can be produced using recombinant DNA methods (see, for example, US Pat. No. 4,816,567 ). Examples of additional techniques for producing human monoclonal antibodies include the trioma technique, the human B-cell hybridoma technique (Kozbor, Immunology Today 4 (1983), 72) and the EBV hybridoma technique (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss , Inc. (1985), 77-96).

Hybridomas can then be screened using standard methods such as enzyme-linked immunosorbent assay (ELISA) and surface plasmonic assays.

- 15 046123 resonance, for example Biacore™, to identify one or more hybridomas that produce an antibody that specifically binds to a particular antigen. The immunogen may be any form of the relevant antigen, for example, a recombinant antigen, naturally occurring forms, any variants or fragments thereof, or an antigenic peptide thereof. Surface plasmon resonance, as used in the Biacore system, can be used to enhance the effectiveness of phage antibodies binding to an epitope of a target cell surface antigen (Schier, Human Antibodies Hybridomas 7 (1996), 97-105; Malmborg, J. Immunol. Methods 183 ( 1995), 7-13).

Another exemplary method for producing monoclonal antibodies involves screening libraries of expressed proteins, such as phage display or ribosome display libraries. Phage display is described, for example, in Ladner et al., US patent No. 5223409; Smith (1985) Science 228:1315-1317, Clackson et al., Nature, 352: 624-628 (1991) and Marks et al., J. Mol. Biol., 222: 581-597 (1991).

In addition to the use of display libraries, the relevant antigen can be used to immunize a non-human animal, such as a rodent (such as a mouse, hamster, rabbit or rat). In one embodiment, the non-human animal contains at least a portion of a human immunoglobulin gene. For example, it is possible to create lines of mice that are deficient in the production of mouse antibodies that have large fragments of human Ig (immunoglobulin) gene loci. Using hybridoma technology, it is possible to obtain and select antigen-specific monoclonal antibodies derived from genes with the required specificity. See, for example, XENOMOUSE™, Green et al. (1994) Nature Genetics 7:13-21, US 2003-0070185, WO 96/34096 and WO 96/33735.

The monoclonal antibody can be obtained from a non-human animal, and then a modified, eg, humanized, deimmunized, chimeric, antibody can be produced using recombinant DNA techniques known in the art. Examples of modified antibody constructs include humanized versions of non-human antibodies, affinity mature antibodies (see, for example, Hawkins et al. J. Mol. Biol. 254, 889-896 (1992) and Lowman et al., Biochemistry 30, 10832-10837 (1991)) and mutant antibodies with altered effector function(s) (see, for example, US patent 5648260, Kontermann and Dubel (2010), loc. cit. and Little (2009), loc. cit.).

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

A preferred type of amino acid replacement variant of antibody constructs involves replacing one or more hypervariable region residues of the original antibody (eg, humanized or human antibody). In general, the resulting variants selected for further development should have improved biological properties relative to the parent antibody from which they are derived. A convenient way to generate such substitution variants involves affinity maturation using phage display. Briefly, several hypervariable region sites (eg, 6-7 sites) are mutated to produce all possible amino acid substitutions at each site. The antibody variants produced in this manner are presented monovalently on filamentous phage particles as fusion molecules with the M13 gene III product packaged within each particle. The phage display variants are then screened for their biological activity (eg, binding affinity) as disclosed herein. To identify candidate hypervariable region regions for modification, alanine scanning mutagenesis can be performed to identify hypervariable region residues that contribute significantly to antigen binding. Alternatively or additionally, it may be useful to analyze the crystal structure of the antigen-antibody complex to identify points of contact between the binding domain and, for example, human MUC17. Such contacting residues and adjacent residues are candidates for replacement according to the methods developed herein. Once such variants are generated, the panel of variants is subjected to the screening described herein, and antibodies with superior properties in one or more relevant assays can be selected for further development.

Monoclonal antibodies and antibody constructs of the present invention particularly include chimeric antibodies (immunoglobulins) in which part of the heavy and/or light chain is identical or homologous to the corresponding antibody sequences derived from a particular species or belonging to a particular class or subclass of antibodies, while the rest is

- 16 046123 pi(chains) is identical or homologous to the corresponding sequences of antibodies obtained from another species or belonging to another class or subclass of antibodies, as well as fragments of such antibodies, provided that they exhibit the required biological activity (US patent No. 4816567; Morrison et al., Proc. Natl. Acad. Sci. USA, 81: 6851-6855 (1984)). Chimeric antibodies of interest herein include primatized antibodies containing antigen binding variable domain sequences derived from a non-human primate (eg, a monkey, ape, etc.) and human constant region sequences. A variety of approaches to the production of chimeric antibodies have been described. See, for example, Morrison et al., Proc. Natl. Acad. ScL U.S.A. 81:6851, 1985; Takeda et al., Nature 314:452, 1985, Cabilly et al., US Patent No. 4816567; Boss et al., US Patent No. 4816397; Tanaguchi et al., EP 0171496; EP 0173494 and GB 2177096.

An antibody, antibody construct, antibody fragment, or antibody variant can also be modified by specifically deleting human T cell epitopes (in a manner referred to as deimmunization) through methods disclosed, for example, in WO 98/52976 or WO 00/34317. Briefly, the variable domains of an antibody's heavy and light chains can be analyzed for the presence of peptides that bind to MHC class II; these peptides represent potential T cell epitopes (as defined in WO 98/52976 and WO 00/34317). A computer modeling approach called peptide pull-down can be used to identify potential T-cell epitopes and, in addition, a database of human MHC class Il binding peptides can be searched for motifs present in the VH and VL sequences as described in WO 98 /52976 and WO 00/34317. These motifs bind to any of the 18 major MHC class Il DR allotypes and thus constitute potential T cell epitopes. Identified potential T cell epitopes can be eliminated by substitution of a small number of amino acid residues in the variable domains or preferably by single amino acid substitutions. Conservative substitutions are usually made. Often, but not exclusively, an amino acid typically present at that position in human germline antibody sequences may be used. Human germline sequences are disclosed, for example, in Tomlinson, et al. (1992) J. Mol. Biol. 227:776-798; Cook, G.P. et al. (1995) Immunol. Today Vol. 16 (5): 237-242; and Tomlinson et al. (1995) EMBO J. 14: 14:4628-4638. The V BASE directory is a comprehensive directory of human immunoglobulin variable region sequences (compiled by Tomlinson, LA. et al. MRC Center for Protein Engineering, Cambridge, UK). These sequences can be used as a source of human sequences, for example for frameworks and CDRs. Consensus human frameworks, such as those described in US Pat. No. 6,300,064, can also be used.

Humanized antibodies, antibody constructs, variants or fragments thereof (such as Fv, Fab, Fab', F(ab') ₂ or other antigen-binding antibody sequences) are antibodies or immunoglobulins with predominantly human sequences that contain a minimal amount of sequence( sequences) derived from immunoglobulin other than human immunoglobulin. Humanized antibodies are generally human immunoglobulins (recipient antibody) in which residues from the hypervariable region (also CDR) of the recipient are replaced by residues from the hypervariable region of an antibody from a non-human (eg, rodent) species (donor antibody), such as mouse, rat, hamster or rabbit that has the required specificity, affinity and binding ability. In some cases, the Fv residues of the human immunoglobulin framework region (FR) are replaced by corresponding non-human residues. In addition, the humanized antibodies used herein may also contain residues not found in either the recipient antibody or the donor antibody. These modifications are made to further improve and optimize the effectiveness of the antibodies. The humanized antibody may also comprise at least a portion of an immunoglobulin constant region (Fc), typically from human immunoglobulin. For details, see Jones et al., Nature, 321: 522-525 (1986); Reichmann et al., Nature, 332: 323-329 (1988); and Presta, Curr. Op. Struct. Biol., 2: 593-596 (1992).

Humanized antibodies or fragments thereof can be generated by replacing Fv variable domain sequences not directly involved in antigen binding with equivalent human Fv variable domain sequences. Exemplary methods for creating humanized antibodies or fragments thereof are presented in Morrison (1985) Science 229:1202-1207; Oi et al. (1986) BioTechniques 4:214; and US 5585089; US 5693761; US 5693762; US 5,859,205 and US 6,407,213. These methods involve isolating, manipulating, and expressing nucleic acid sequences encoding entire or portions of immunoglobulin Fv variable domains from at least one heavy or light chain. Such nucleic acids can be obtained from a hybridoma producing an antibody to a predetermined target described above, as well as from other sources. The recombinant DNA encoding the humanized antibody molecule can then be cloned into a suitable expression vector.

- 17 046123

Humanized antibodies can also be produced using transgenic animals, such as mice, that express human heavy and light chain genes but are unable to express endogenous mouse immunoglobulin heavy and light chain genes. Winter describes an exemplary CDR grafting method that can be used to produce the humanized antibodies described herein (US Pat. No. 5,225,539). Each CDR of a particular human antibody may be at least partially replaced by a non-human CDR, or only some of the CDRs may be replaced by non-human CDRs. It is only necessary to replace the number of CDRs required to bind the humanized antibody to the predetermined antigen.

The humanized antibody can be optimized by introducing conservative substitutions, consensus sequence substitutions, substitutions with germline configuration residues, and/or back mutations. Such altered immunoglobulin molecules can be prepared according to any of several techniques known in the art (eg, Teng et al., Proc. Natl. Acad. Sci. U.S.A., 80: 7308-7312, 1983; Kozbor et al., Immunology Today , 4: 7279, 1983; Olsson et al., Meth. Enzymol., 92: 3-16, 1982, and EP 239400).

The term human antibody, human antibody construct, and human binding domain includes antibodies, antibody constructs, and binding domains with antibody regions, such as variable and constant regions, or domains that correspond substantially to human germline immunoglobulin sequences known in the art. , including, for example, those described in Kabat et al. (1991) (loc. cit.). The human antibodies, antibody constructs, or binding domains of the present invention may contain amino acid residues not encoded by human immunoglobulin gene sequences in a germline configuration (e.g., as a result of mutations introduced through random or site-specific mutagenesis in vitro or somatic mutations in vivo), for example in CDR, and in particular in CDR3. Human antibodies, antibody constructs, or binding domains may contain at least one, two, three, four, five or more positions replaced by an amino acid residue not encoded by a human germline immunoglobulin sequence. The definition of human antibodies, antibody constructs, and binding domains used herein also includes complete human antibodies, which include only non-artificial and/or genetically modified human human antibody sequences, such as those that can be produced using technologies or systems such as Xenomouse. Preferably, the fully human antibody does not include amino acid residues not encoded by human germline immunoglobulin sequences.

In some embodiments, the antibody constructs of the present invention are isolated or substantially pure antibody constructs. Isolated or substantially pure, as used to describe the antibody constructs disclosed herein, means an antibody construct that has been identified, separated, and/or recovered from a component of its production medium. Preferably, the antibody construct is unassociated or substantially unassociated with all other components of the environment in which it is produced. Contaminants in the production environment, such as those resulting from recombinant transfected cells, are materials that tend to interfere with the diagnostic or therapeutic use of the polypeptide, and may include enzymes, hormones, and other proteinaceous or nonproteinaceous solutes. The antibody constructs may, for example, represent at least about 5% or at least about 50% by weight of the total protein in a given sample. It is understood that the isolated protein may range from 5 to 99.9% by weight of the total protein content depending on the circumstances. The polypeptide can be produced at substantially higher concentrations by using an inducible promoter or a high-level expression promoter so that it is produced at increased concentration levels. This definition includes the production of an antibody construct in a wide range of organisms and/or host cells that are known in the art. In preferred embodiments of the present invention, the antibody construct is purified (1) to an extent sufficient to obtain at least 15 residues of N-terminal or internal amino acid sequence using a spinning beaker sequencer, or (2) to homogeneity by SDS-PAGE in non-reducing or reducing conditions using Coomassie blue or preferably silver staining. In this case, the antibody-based construct is usually obtained using at least one purification step.

The term binding domain, as used herein, describes a domain that (specifically) binds to/interacts with/recognizes a specified target epitope or a specified target site on target molecules (antigens), in this case MUC17 and CD3, respectively. The structure and function of the first binding domain (recognizing MUC17) and preferably also the structure and/or function of the second binding domain (recognizing CD3) is based on the structure and/or function of an antibody, for example, a full-length or whole immunoglobulin molecule

- 18 046123 and/or are derived from the heavy chain variable region (VH) and/or light chain variable region (VL) domains of an antibody or fragment thereof. Preferably, the first binding domain is characterized by the presence of three light chain CDRs (ie VL region CDR1, CDR2 and CDR3) and/or three heavy chain CDRs (ie VH region CDR1, CDR2 and CDR3). The second binding domain preferably also provides the minimum structural requirements of the antibody that provide target binding. More preferably, the second binding domain contains at least three light chain CDRs (ie, VL region CDR1, CDR2 and CDR3) and/or three heavy chain CDRs (ie, VH region CDR1, CDR2 and CDR3). It is intended that the first and/or second binding domain is or can be obtained by phage display or library screening methods rather than by grafting CDR sequences from a pre-existing (monoclonal) antibody into a scaffold.

In accordance with the present invention, the binding domains are in the form of one or more polypeptides. Such polypeptides may contain protein portions and non-protein portions (eg, chemical linkers or chemical cross-linkers such as glutaraldehyde). Proteins (including fragments thereof, preferably biologically active fragments, and peptides typically having less than 30 amino acids) contain two or more amino acids linked together by a covalent peptide bond (resulting in a chain of amino acids).

The term polypeptide as used herein describes a group of molecules typically containing more than 30 amino acids. Polypeptides can further form multimers such as dimers, trimers and higher oligomers, i.e. containing more than one polypeptide molecule. The polypeptide molecules forming such dimers, trimers, etc. may be identical or non-identical. The corresponding higher order structures of such multimers are therefore called homo- or heterodimers, homo- or heterotrimers, etc. An example of a heteromultimer is an antibody molecule, which in its naturally occurring form contains two identical light polypeptide chains and two identical heavy polypeptide chains. The terms peptide, polypeptide and protein also refer to naturally modified peptides/polypeptides/proteins, where the modification is through, for example, post-translational modifications such as glycosylation, acetylation, phosphorylation and the like. The peptide, polypeptide or protein referred to herein may also be modified chemically, such as by PEGylation. Such modifications are well known in the art and are described below herein.

Preferably, the binding domain that binds to MUC17 and/or the binding domain that binds to CD3ε are human binding domains. Antibodies and antibody constructs containing at least one human binding domain avoid a number of problems associated with antibodies or antibody constructs that have non-human variable and/or constant regions, such as those from rodents ( e.g. mouse, rat, hamster or rabbit). The presence of such rodent-derived proteins may result in rapid clearance of antibodies or antibody constructs or may result in an immune response to the antibody or antibody construct in the patient. To avoid the use of rodent-derived antibodies or antibody constructs, human or fully human antibodies/antibody constructs can be generated by introducing functional portions of a human antibody into a rodent so that fully human antibodies are produced in the rodent.

The ability to clone and reconstitute human loci on the order of millions of base pairs in yeast YAC artificial chromosomes and introduce them into the mouse germline provides an effective approach for studying the functional components of very large or coarsely mapped loci and for generating useful models of human disease. In addition, the use of such technology to replace mouse loci with their human equivalents may provide unique information about the expression and regulation of human gene products during development, their interactions with other systems, and their role in disease induction and progression.

An important practical application of this strategy is the humanization of the mouse humoral immune system. Introduction of human immunoglobulin (Ig) loci into mice in which endogenous Ig genes have been inactivated provides an opportunity to study the mechanisms underlying programmed expression and assembly of antibodies, as well as their role in B cell development. In addition, such a strategy could provide an ideal source for the production of fully human monoclonal antibodies (mAbs), a key point on the path to developing antibody therapeutics to treat human diseases. Fully human antibodies or antibody-based constructs are expected to minimize the immunogenic and allergic responses inherent in murine or mouse-derived mAbs and therefore enhance the efficacy and safety of the administered antibody/antibody-based constructs. The use of fully human antibodies or antibody constructs can be expected to provide significant benefits in the treatment of chronic and recurrent human diseases such as inflammation, autoimmune reactions and cancer, in

- 19 046123 which require repeated administration of compounds.

One approach toward this goal has been to construct mouse lines deficient in mouse antibody production with large fragments of human Ig loci, with the expectation that such mice will produce a large repertoire of human antibodies in the absence of mouse antibodies. Large human Ig fragments must maintain a wide variety of variable genes, as well as proper regulation of antibody production and expression. By using a mouse apparatus to enhance the diversity and selection of antibodies and the absence of immunological tolerance to human proteins, the reconstituted repertoire of human antibodies in these mouse strains should result in high-affinity antibodies to any antigen of interest, including human antigens. Using hybridoma technology, antigen-specific human mAbs with the required specificity can be easily prepared and screened. This general strategy was demonstrated in connection with the creation of the first lines of XenoMouse mice (see Green et al. Nature Genetics 7:13-21 (1994)). XenoMouse lines with yeast YACs containing 245 kb fragments were constructed. and 190 t.o. germline configurations of the human heavy chain locus and the kappa light chain locus, respectively, which contained internal variable and constant region sequences. Human Ig-containing YACs were found to be compatible with the mouse system in terms of both rearrangement and antibody expression, and were able to replace inactivated mouse Ig genes. This has been demonstrated by their ability to induce B cell development, produce a human adult repertoire of fully human antibodies, and produce antigen-specific human mAbs. These results also suggest that introduction of larger portions of human Ig loci, containing more V genes, additional regulatory elements, and human Ig constant regions, can essentially recapitulate the full repertoire that characterizes the human humoral response to infection and immunization. The work of Green et al. was recently expanded to introduce more than approximately 80% of the human antibody repertoire by introducing YAC fragments on the order of a million base pairs of the germline configuration of the human heavy chain locus and the kappa light chain locus, respectively. See Mendez et al. Nature Genetics 15:146-156 (1997) and US Patent Application Serial No. 08/759620.

Production of XenoMouse animals is further discussed and described in US Patent Applications Serial No. 07/466008, Serial No. 07/610515, Serial No. 07/919297, Serial No. 07/922649, Serial No. 08/031801, Serial No. 08/112848, Serial No. 08/234145, serial No. 08/376279, serial No. 08/430938, serial No. 08/464584, serial No. 08/464582, serial No. 08/463191, serial No. 08/462837, serial No. 08/486853, serial number 08 /486857, serial No. 08/486859, serial No. 08/462513, serial No. 08/724752 and serial No. 08/759620; and US patents No. 6162963; 6150584; 6114598; 6075181 and 5939598, and Japanese patents No. 3068180 B2, 3068506 B2 and 3068507 B2. See also Mendez et al. Nature Genetics 15:146–156 (1997) and Green and Jakobovits J. Exp. Med. 188:483-495 (1998), EP 0463151 B1, WO 94/02602, WO 96/34096, WO 98/24893, WO 00/76310 and WO 03/47336.

In an alternative approach, others, including GenPharm International, Inc., have used a minilocus approach. In the minilocus approach, the exogenous Ig locus is mimicked by incorporating parts (individual genes) from the Ig locus. Thus, one or more VH genes, one or more DH genes, one or more JH genes, a mu chain constant region, and a second constant region (preferably a gamma chain constant region) are constituted into a construct for insertion into an animal. This approach is described in US Patent No. 5545807 by Surani et al. and US patents No. 5545806; 5625825; 5625126; 5633425; 5661016; 5770429; 5789650; 5814318; 5877397; 5,874,299 and 6,255,458 each to Lonberg and Kay, US Patent Nos. 5,591,669 and 6,023,010 to Krimpenfort and Berns, US Patent No. 5,612,205; 5721367 and 5789215 by Berns et al. and U.S. Patent No. 5643763 to Choi and Dunn, and U.S. Patent Application Serial No. 07/574748, Serial No. 07/575962, Serial No. 07/810279, Serial No. 07/853408, Serial No. 07/904068, Serial No. No. 07/990860, serial No. 08/053131, serial No. 08/096762, serial No. 08/155301, serial No. 08/161739, serial No. 08/165699, serial No. 08/209741. Also see EP 0546073 B1, WO 92/03918, WO 92/22645, WO 92/22647, WO 92/22670, WO 93/12227, WO 94/00569, WO 94/25585, WO 96/14436, WO 97/ 13852 and WO 98/24884 and US Patent No. 5981175. See further Taylor et al. (1992), Chen et al. (1993), Tuaillon et al. (1993), Choi et al. (1993), Lonberg et al. (1994), Taylor et al. (1994) and Tuaillon et al. (1995), Fishwild et al. (1996).

Kirin also demonstrated the creation of human antibodies using mice injected with large sections of chromosomes or entire chromosomes via microcell-mediated fusion. See European Patent Applications Nos. 773288 and 843961. Xenerex Biosciences is developing technology to potentially create human antibodies. In this technology, SCID mice are regenerated using human lymphatic cells, such as B and/or T cells. Mice are then immunized with the antigen and can mount an immune response to the antigen. See US Pat. No. 5,476,996; 5698767 and 5958765.

- 20 046123

Human anti-mouse antibody (HAMA) responses have led the industry to the development of chimeric or otherwise humanized antibodies. It is expected that some human antibody responses to chimeras (NACA) will be observed, particularly with chronic or multi-dose applications of the antibody. Thus, it would be desirable to generate antibody constructs containing a human anti-MUC17 binding domain and a human anti-CD3ε binding domain to mitigate concerns and/or effects regarding NAMA or NASA response.

The terms (specifically) binds to, (specifically) recognizes, (specifically) targets and (specifically) reacts with in accordance with the present invention mean that the binding domain interacts or specifically interacts with a specific epitope or specific site on target molecules ( antigens), in this case MUC17 and CD3ε, respectively.

The term epitope refers to the region of an antigen to which a binding domain, such as an antibody or immunoglobulin or a derivative, fragment or variant of an antibody or immunoglobulin, specifically binds. An epitope is antigenic, and therefore the term epitope is sometimes also referred to herein as an antigenic structure or antigenic determinant. Thus, the binding domain represents the site of interaction with the antigen. Said binding/interaction is also understood to determine specific recognition.

Epitopes can be formed either by contiguous amino acids or by non-contiguous amino acids placed side by side when the protein is folded into the tertiary structure. A linear epitope is an epitope where the primary amino acid sequence forms a recognizable epitope. A linear epitope typically contains at least 3 or at least 4, and more often at least 5 or at least 6 or at least 7, such as about 8 to about 10 amino acids in a unique sequence.

A conformational epitope, as opposed to a linear epitope, is an epitope in which the primary amino acid sequence forming the epitope is not the sole defining component of the epitope being recognized (e.g., an epitope in which the primary amino acid sequence is not necessarily recognized by the binding domain). Typically, a conformational epitope contains an increased number of amino acids compared to a linear epitope. With respect to conformational epitope recognition, the binding domain recognizes the three-dimensional structure of the antigen, preferably a peptide or protein or fragment thereof (in the context of the present invention, the antigenic structure for one of the binding domains is contained within the target cell surface antigen protein). For example, when a protein molecule is folded into a three-dimensional structure, the specific amino acids and/or polypeptide backbone that form the conformational epitope are placed adjacent to each other, allowing recognition of the epitope by an antibody. Methods for determining the conformation of epitopes include, but are not limited to, x-ray crystallography, two-dimensional nuclear magnetic resonance (2DNMR) spectroscopy, and site-directed spin labeling electron paramagnetic resonance (EPR) spectroscopy.

The epitope mapping method is described as follows: when a region (contiguous amino acid segment) in human MUC17 is replaced or replaced with a corresponding region of non-human and non-primate MUC17 (e.g. mouse MUC17, but others such as chicken, rat, hamster, rabbit, etc.) etc. are also allowed), reduced binding domain binding is expected unless the binding domain is cross-reactive for use with non-human and non-primate MUC17. Said reduction is preferably at least 10, 20, 30, 40 or 50%; more preferably at least 60, 70 or 80% and most preferably 90, 95 or even 100% compared to binding to the corresponding region of the human MUC17 protein, where binding to the corresponding region of the human MUC17 protein is taken to be 100%. The above-mentioned human MUC17/non-human MUC17 is envisioned to be expressed in CHO cells. It is also contemplated that human MUC17/non-human MUC17 is fused to the transmembrane domain and/or cytoplasmic domain of another membrane-bound protein, such as EpCAM.

In an alternative or complementary epitope mapping method, multiple truncated variants of the extracellular domain of human MUC17 can be generated to determine the specific region that is recognized by the binding domain. In these truncated variants, various domains/subdomains or regions of extracellular MUC17 are removed stepwise, starting from the N-terminus. It is envisaged that truncated variants of MUC17 may be expressed in CHO cells. It is also contemplated that truncated variants of MUC17 may be fused to the transmembrane domain and/or cytoplasmic domain of another membrane-bound protein such as EpCAM. It is also contemplated that truncated variants of MUC17 may contain a signal peptide domain at their N-terminus, for example a signal peptide that is derived from a mouse IgG heavy chain signal peptide. Moreover, it is envisaged that truncated variants of MUC17 may contain a v5 domain at their N-terminus

- 21 046123 (after the signal peptide), which makes it possible to confirm their correct expression on the cell surface. Reduction or loss of binding is expected to occur in those truncated MUC17 variants that no longer contain a region of MUC17 that is recognized by the binding domain. The reduction in binding is preferably at least 10, 20, 30, 40, 50%; more preferably at least 60, 70, 80% and most preferably 90, 95 or even 100%, with binding to the entire human MUC17 protein (or extracellular region or domain thereof) being taken as 100.

An additional method for determining the contribution of a particular MUC17 residue to recognition by an antibody construct or binding domain is alanine scanning (see, e.g., Morrison KL & Weiss GA. Cur Opin Chem Biol. 2001 Jun;5(3):302-7). wherein each residue analyzed is replaced by alanine, for example, by site-directed mutagenesis. Alanine's use is due to its small, chemically inert methyl functional group, which nevertheless mimics the secondary structure preferences of many other amino acids. In cases where the size of the mutated residues needs to be maintained, large amino acids such as valine or leucine can sometimes be used. Alanine scanning is a proven technology that has been used for a long time.

An interaction between a binding domain and an epitope or epitope-containing region means that the binding domain exhibits acceptable affinity for the epitope/epitope-containing region in a particular protein or antigen (in this case MUC17 and CD3, respectively) and generally does not exhibit significant reactivity with proteins or antigens other than MUC17 or CD3. Significant affinity includes binding with an affinity of approximately 10 ^-6 M (KD) or higher. Preferably, binding is considered specific if the binding affinity is about 10 ^-12 -10 ^-8 M, 10 ^-12 -10 ^-9 M, 10 ^-12 -10 ^-10 M, 10 ^-11 -10 ^-8 M, preferably about 10 ^{- 11} -10 ^-9 M. The specificity of the reaction or binding of the binding domain to the target can be easily verified by, among others, comparing the reaction of the specified binding domain with the target protein or antigen and the reaction of the specified binding domain with proteins or antigens other than MUC17 or CD3. Preferably, the binding domain of the present invention has substantially or substantially no binding to proteins or antigens other than MUC17 or CD3 (i.e., the first binding domain is incapable of binding to proteins other than MUC17 and the second binding domain is incapable of binding to proteins other than CD3). An implied characteristic of the antibody constructs of the present invention is that they have superior affinity characteristics compared to other HLE formats. Therefore, this superior affinity implies an extended half-life in vivo. The longer half-life of the antibody constructs of the present invention may reduce the duration and frequency of administration, which generally results in improved patient compliance. This is extremely important because the antibody constructs of the present invention are extremely beneficial for severely frail or even multimorbid cancer patients.

The term essentially/substantially does not bind or is incapable of binding means that the binding domain of the present invention does not bind to a protein or antigen other than MUC17 or CD3, i.e. exhibits no more than 30%, preferably no more than 20%, more preferably no more than 10%, especially preferably no more than 9, 8, 7, 6 or 5% reactivity with proteins or antigens other than MUC17 or CD3, where binding to MUC17 or CD3, respectively, is taken as 100%.

It is believed that specific binding occurs due to specific motifs in the amino acid sequence of the binding domain and the antigen. Thus, binding is achieved through their primary, secondary and/or tertiary structure, as well as through secondary modifications of these structures. The specific interaction of an antigen interacting site and its specific antigen may result in the site simply binding to the antigen. Moreover, the specific interaction of the antigen interacting portion and its specific antigen may alternatively or additionally result in a signal, for example, due to the induction of a change in antigen conformation, antigen oligomerization, or the like.

The term variable refers to portions of antibody or immunoglobulin domains that exhibit sequence variability and are involved in determining the specificity and binding affinity of a particular antibody (ie, variable domain(s)). By pairing a heavy chain variable region (VH) and a light chain variable region (VL), a single antigen-binding region is formed.

Variability is not evenly distributed among the variable domains of antibodies; it is concentrated in the subdomains of each of the variable regions of the heavy and light chains. These subdomains are called hypervariable regions or complementarity determining regions (CDRs). The more conserved (i.e., non-hypervariable) regions of the variable domains are called framework regions (FRM or FR) and provide the framework for the six CDRs in three dimensions

- 22 046123 nom space for the formation of an antigen-binding surface. Each of the naturally occurring heavy and light chain variable domains contains four FRM regions (FR1, FR2, FR3 and FR4) adopting a largely β-sheet configuration, connected by three hypervariable regions that form loops connecting, and in some cases forming part of, β-folded structure. The hypervariable regions on each chain are held together in close proximity by FRM and, together with the hypervariable regions of the other chain, contribute to the formation of the antigen binding site (see Kabat et al., loc. cit.).

The term CDR and its plural refer to the complementarity determining region, three of which impart binding character to the light chain variable region (CDR-L1, CDR-L2, and CDR-L3) and three that confer binding character to the heavy chain variable region (CDR- H1, CDR-H2 and CDR-H3). CDRs contain the majority of residues responsible for the specific interaction of an antibody with an antigen and, therefore, contribute to the functional activity of the antibody molecule: they are the main determinants of antigen specificity.

The precise definition of the boundaries and dimensions of a CDR is subject to various classifications and numbering systems. Accordingly, CDRs may be indicated according to Kabat, Chothia, contact or any other delineation system including the numbering system described herein. Despite their differing boundaries, each of these systems has a certain degree of overlap, representing so-called hypervariable regions in variable sequences. Consequently, CDR definitions under these systems may differ in length and boundary portions relative to the adjacent frame portion. See for example Kabat (an approach based on interspecies sequence variability), Chothia (an approach based on crystallographic studies of antigen-antibody complexes) and/or MacCallum (Kabat et al., loc. cit.; Chothia et al., J. Mol Biol, 1987, 196: 901-917; and MacCallum et al., J Mol Biol, 1996, 262: 732). Another standard for characterizing the antigen-binding site is the AbM definition used in Oxford Molecular's AbM antibody modeling software. See, for example, Protein Sequence and Structure Analysis of Antibody Variable Domains. In Antibody Engineering Lab Manual (Ed.: Duebel, S. and Kontermann, R., Springer-Verlag, Heidelberg). When two residue identification techniques identify residues from overlapping but not identical regions, they can be combined to identify a hybrid CDR. However, numbering according to the so-called Kabat system is preferred.

Typically, CDRs form a loop structure, which can be classified as a canonical structure. The term canonical structure refers to the backbone conformation adopted by the antigen-binding loops (CDRs). From comparative structural studies, five of the six antigen-binding loops were found to have only a limited set of available conformations. Each canonical structure can be characterized by the helical angles of the polypeptide backbone. Consequently, the corresponding loops among antibodies can be characterized by a high degree of similarity in three-dimensional structures, despite high amino acid variability in most parts of the loops (Chothia and Lesk, J. Mol. Biol., 1987, 196: 901; Chothia et al., Nature, 1989, 342: 877; Martin and Thornton, J Mol Biol, 1996, 263: 800). In addition, there is a relationship between the structure adopted by the loop and the amino acid sequences surrounding it. The conformation of a particular canonical class is determined by the length of the loop and the amino acid residues located at key positions in the loop, as well as within the conserved framework region (i.e., outside the loop). Thus, assignment to a particular canonical class can be made based on the presence of these key amino acid residues.

The term canonical structure may also include aspects related to the linear sequence of the antibody, such as those listed according to Kabat (Kabat et al., loc. cit.). The Kabat numbering scheme is a widely accepted standard for numbering the amino acid residues of an antibody variable domain in a sequential manner and is the preferred scheme used in the present invention, which is also mentioned elsewhere herein. Additional structure factors can also be used to determine the canonical structure of an antibody. For example, differences not fully captured by the Kabat numbering system can be described using the Chothia et al. numbering system. and/or identified using other techniques, such as crystallography or two- or three-dimensional computer modeling. Accordingly, the antibody sequence can be placed in a canonical class, allowing, among other things, the identification of relevant framework sequences (eg, based on the need to include a variety of canonical structures in the library). Kabat's antibody amino acid sequence numbering system and structural features described by Chothia et al., loc. cit, as well as their significance for interpreting canonical aspects of antibody structure, are described in the literature. The structural subunits and three-dimensional configurations of various classes of immunoglobulins are well known in the art. For a review of antibody structure, see Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, eds. Harlow et al., 1988.

- 23 046123

Light chain CDR3 and particularly heavy chain CDR3 may constitute the most important determinants in antigen binding within the light and heavy chain variable regions. In some antibody constructs, the heavy chain CDR3 constitutes the major contact area between the antigen and the antibody. In vitro selection schemes in which only CDR3 is varied can be used to alter the binding properties of an antibody or determine which residues promote antigen binding. Therefore, CDR3 is typically the source of the greatest molecular diversity at the antibody binding site. For example, H3 can be as short as two amino acid residues or over 26 amino acids long.

In classical full-length immunoglobulin, each light (L) chain is linked to the heavy (H) chain by a single covalent disulfide bond, while the two H chains are linked to each other by one or more disulfide bonds, depending on the H chain isotype. The CH domain closest to VH is usually designated CH1. Constant domains (C) are not directly involved in antigen binding, but exhibit various effector functions, such as antibody-dependent cell-mediated cytotoxicity and activation of the complement system. The Fc region of the antibody is located within the constant domains of the heavy chain and, for example, is capable of interacting with Fc receptors located on the cell surface.

The sequences of antibody genes after assembly and somatic mutation are highly variable, and these variable genes are estimated to encode 10 different antibody molecules (Immunoglobulin Genes, ^2nd ed., eds. Jonio et al., Academic Press, San Diego, CA, 1995). Accordingly, a repertoire of immunoglobulins is formed in the immune system. The term repertoire refers to at least one nucleotide sequence derived in whole or in part from at least one sequence encoding at least one immunoglobulin.

The sequence(s) can be obtained by in vivo rearrangement of the V, D and J segments of the heavy chain genes and the V and J segments of the light chain genes. Alternatively, the sequence(s) can be obtained from the cell through, for example, in vitro stimulation, in response to which rearrangement occurs. Alternatively, part or all of the sequence(s) can be obtained through DNA splicing, nucleotide synthesis, mutagenesis and other methods, see, for example, US patent 5565332. The repertoire may include only one sequence or may include a collection of sequences, including those found in genetically diverse collection.

In connection with the present invention, the term Fc portion or Fc monomer means a polypeptide containing at least one domain having the function of a CH2 domain and at least one domain having the function of a CH3 domain of an immunoglobulin molecule. From the term Fc monomer it is clear that the polypeptide containing these CH domains is a polypeptide monomer. The Fc monomer may be a polypeptide containing at least a fragment of an immunoglobulin constant region excluding the first domain of an immunoglobulin heavy chain constant region (CH1), but retaining at least a functional portion of one CH2 domain and a functional portion of one CH3 domain, wherein the CH2 domain is amino terminal to the CH3 domain. In a preferred aspect of this definition, the Fc monomer may be a polypeptide constant region comprising a portion of an Ig-Fc hinge region, a CH2 region, and a CH3 region, the hinge region being amino terminal to the CH2 domain. It is understood that the hinge region of the present invention stimulates dimerization. Such Fc polypeptide molecules can be prepared, for example, without limitation, by digestion of the immunoglobulin portion with papain (of course, resulting in a dimer of two Fc polypeptides). In another aspect of this definition, an Fc monomer may be a polypeptide region comprising part of a CH2 region and a CH3 region. Such Fc polypeptide molecules can be produced, for example, without limitation, by digestion of an immunoglobulin molecule with pepsin. In one embodiment, the polypeptide sequence of the Fc monomer is substantially similar to the sequence of the Fc polypeptide: IgGi Fc region, IgG ₂ Fc region, IgG ₃ Fc region, IgG ₄ Fc region, IgM Fc region, IgA Fc region, Fc- IgD region and IgE Fc region. (See, for example, Padlan, Molecular Immunology, 31(3), 169217 (1993)). Because there is some variability between immunoglobulins, and for clarity only, Fc monomer means the last two domains of the immunoglobulin heavy chain constant region of IgA, IgD and IgG and the last three domains of the immunoglobulin heavy chain constant region of IgE and IgM. As mentioned, the Fc monomer may also contain a flexible hinge region located N-terminal to these domains. In the case of IgA and IgM, the Fc monomer may include a J chain. In the case of IgG, the Fc portion contains the CH2 and CH3 domains of the immunoglobulin and the hinge region between the first two domains and CH2. Although the boundaries of the Fc portion may vary, for example, in the case of a human IgG heavy chain Fc portion containing a functional hinge region, the CH2 and CH3 domains may be specified, for example, to contain residues from D231 (the hinge domain - in accordance with D234 in Table 1 below) to P476, respectively L476 (for IgG ₄ ) carboxy terminus of the CH3 domain, where numbering is used according to Kabat. Two Fc moieties or two Fc monomers fused to each other via a peptide linker define a third domain of the antibody construct of the present invention, which may also be defined as a scFc domain.

In one embodiment of the present invention, it is contemplated that the scFc domain described herein, or the Fc monomers fused to each other, are present only in the third domain of the antibody construct.

In accordance with the present invention, the hinge region of an IgG can be determined by analogy using Kabat numbering, as shown in table. 1. In accordance with the above, it is assumed that for the hinge domain/region of the present invention the minimum requirement contains amino acid residues corresponding to the region of the IgG1 sequence from D231, D234 to P243 according to Kabat numbering. Likewise, the hinge domain/hinge region of the present invention is intended to contain or consist of the IgG1 hinge sequence DKTHTCPPCP (SEQ ID NO: 477) (corresponding to the region D234 to P243 as shown in Table 1 below - variants of this sequence are also contemplated in provided that the hinge region still promotes dimerization). In a preferred embodiment of the present invention, the glycosylation site at Kabat position 314 of the CH2 domains in the third domain of the antibody construct is removed by substitution N314X, where X is any amino acid excluding Q. This substitution is preferably the substitution N314G. In a more preferred embodiment, said CH2 domain further contains the following substitutions (positions according to Kabat): V321C and R309C (these substitutions introduce an intradomain cysteine disulfide bridge at positions 309 and 321 according to Kabat).

It is also contemplated that the third domain of the antibody construct of the present invention comprises, from the amino terminus to the carboxyl terminus of DKTHTCPPCP (SEQ ID NO: 477) (i.e., the hinge region), a -CH2-CH3 linker-DKTHTCPPCP (SEQ ID NO: 477) (i.e. hinged section) CH2-CH3. The peptide linker from the above antibody construct is in a preferred embodiment characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e. Gly ₄ Ser (SEQ ID NO: 1), or polymers thereof, i.e. (Gly ₄ Ser)x, where x is an integer of 5 or greater (eg, 5, 6, 7, 8, etc. or greater), with 6 ((Gly ₄ Ser) 6) being preferred. This construct may further contain the above substituents: N314X, preferably N314G and/or additional substituents V321C and R309C. In a preferred embodiment of the antibody constructs of the present invention, as previously defined herein, the second domain is provided to bind to an extracellular epitope of the human CD3ε chain and/or Masaca.

Table 1

Numbering of amino acid residues of the hinge region according to Kabat

IMGT numbering for hinge section Amino acid translation for IgGi Numbering according to Kabat 1 (E) 226 2 R 227 3 TO 228 4 S 232 5 With 233 6 D 234 7 TO 235 8 T 236 9 N 237 10 T 238 eleven WITH 239 12 R 240 13 R 241 14 With 242 15 R 243

In further embodiments, the hinge domain/region comprises or consists of an IgG2 subtype hinge sequence ERKCCVECPPCP (SEQ ID NO: 478), an IgG3 subtype hinge sequence ELKTPLDTTHTCPRCP (SEQ ID NO: 479) or ELKTPLGDTTHTCPRCP (SEQ ID NO: 486) , and/or the IgG4 subtype hinge region sequence, ESKYGPPCPSCP (SEQ ID NO: 480). The IgG1 subtype hinge region sequence may be: EPKSCDKTHTCPPCP (as shown in Table 1 and SEQ ID NO: 487). Thus, these main hinge portions are also intended in the context of the present invention.

The position and sequence of the CH2 domain of IgG and the CD3 domain of IgG can be determined by analogy using Kabat numbering, as shown in Table. 2.

table 2

Numbering of amino acid residues of the CH2 and CH3 region of IgG according to Kabat

IgG subtype Conversion to aa CH2 Numbering CH2 no Kabat Conversion to aa CH3 Numbering CH3 no Kabat IgGi A.P.E. .....HOW 244... ...360 GQP. .....PGK 361... ...478 IgG ₂ APP ......KTK 244... ...360 GQP. .....PGK 361... ...478 IgG ₃ A.P.E. ......KTK 244... ...360 GQP. .....PGK 361... ...478 IgG ₄ A.P.E. ......HOW 244... ...360 GQP ......LGK 361... ...478

In one embodiment of the present invention, the bolded amino acid residues in the CH3 domain of the first or both Fc monomers are deleted.

The peptide linker by which the polypeptide monomers (Fc portion or Fc monomer) of the third domain are fused to each other preferably contains at least 25 amino acid residues (25, 26, 27, 28, 29, 30, etc.). More preferably, the peptide linker contains at least 30 amino acid residues (30, 31, 32, 33, 34, 35, etc.). It is also preferable that the linker contains up to 40 amino acid residues, more preferably up to 35 amino acid residues, most preferably 30 amino acid residues. A preferred embodiment of such a peptide linker is one having the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e. Gly ₄ Ser (SEQ ID NO: 1), or polymers thereof, i.e. (Gly ₄ Ser)x, where x is an integer 5 or greater (for example, 6, 7 or 8). Preferably the integer is 6 or 7, more preferably the integer is 6.

In the event that a linker is used to fuse a first domain to a second domain, or a first or second domain to a third domain, the linker is preferably of length and sequence sufficient to ensure that each of the first and second domains can independently maintain its differential specificity binding. For peptide linkers that link at least two binding domains (or two variable domains) in an antibody construct of the present invention, preferred peptide linkers are those that contain only a few amino acid residues, for example, 12 amino acid residues or less. Thus, peptide linkers of 12, 11, 10, 9, 8, 7, 6 or 5 amino acid residues are preferred. A peptide linker with less than 5 amino acids is intended to contain 4, 3, 2 or one amino acid, with Gly-rich linkers being preferred. A preferred embodiment of the peptide linker for fusion of the first and second domains is set forth in SEQ ID NO: 1. The preferred linker in the embodiment of the peptide linker for fusion of the second and third domains is a G^A) ₄ -linker, also referred to as G^uH^p.

A particularly preferred single amino acid in the context of the one peptide linker described above is Gly. Accordingly, said peptide linker may consist of a single Gly amino acid. In a preferred embodiment of the present invention, the peptide linker is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e. Gly ₄ Ser (SEQ ID NO: 1), or polymers thereof, i.e. (Gly ₄ Ser)x, where x is an integer 1 or greater (for example, 2 or 3). Preferred linkers are given in SEQ ID NO: 1-12.

The characteristics of said peptide linker, which include the absence of stimulation of the formation of secondary structures, are known in the art and are described, for example, in Dall'Acqua et al. (Biochem. (1998) 37, 9266-9273), Cheadle et al. (Mol Immunol (1992) 29, 21-30) and Raag and Whitlow (FASEB (1995) 9(1), 73-80). Preferred are peptide linkers that, among other things, do not promote the formation of any secondary structures. The connection of these domains with each other can be achieved, for example, using genetic engineering, as described in the examples. Methods for preparing fused and operably linked bispecific single-strand constructs and expressing them in mammalian cells or bacteria are well known in the art (eg, WO 99/54440 or Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2001).

In a preferred embodiment of the antibody construct of the present invention, the first and second domains form an antibody construct in a format selected from the group consisting of (scFv) ₂ , scFv single domain mAb, diabody, and oligomers of any of these formats.

In a particularly preferred embodiment, and as documented in the accompanying examples, the first and second domains of the antibody construct of the present invention are a bispecific single chain antibody construct, more preferably a bispecific single chain Fv (scFv). Although the two domains of the Fv fragment, VL and VH, are encoded by separate genes, they can be joined recombinantly using the synthetic linker described above, allowing them to form a single protein chain in which the VL and VH regions are paired to form a monovalent

- 26 046123 molecules; see for example Huston et al. (1988) Proc. Natl. Acad. Sci USA 85:5879–5883). These antibody fragments are prepared using conventional techniques known to those skilled in the art, and the fragments are evaluated for function in the same manner as full-length antibodies. Therefore, a single chain variable fragment (scFv) is a fusion protein of the variable heavy chain (VH) and light chain (VL) of immunoglobulins, typically linked by a short linker peptide of about ten to about 25 amino acids in length, preferably about 15 to 20 amino acids. The linker is typically rich in glycine for flexibility and serine or threonine for solubility and can either connect the N-terminus of VH to the C-terminus of VL or vice versa. This protein retains the specificity of the original immunoglobulin, despite the removal of constant regions and the introduction of a linker.

Constructs based on bispecific single chain antibodies are known in the art and are described in WO 99/54440, Mack, J. Immunol. (1997), 158, 3965-3970, Mack, PNAS, (1995), 92, 7021-7025, Kufer, Cancer Immunol. Immunother., (1997), 45, 193-197, Loffler, Blood, (2000), 95, 6, 20982103, Bruhl, Immunol., (2001), 166, 2420-2426, Kipriyanov, J. Mol. Biol., (1999), 293, 41-56. The techniques described for the production of single chain antibodies (see, but not limited to, US Pat. No. 4,946,778, Kontermann and Dubel (2010), loc. cit. and Little (2009), loc. cit.) can be adapted to produce single chain antibody constructs specifically recognizing the selected target(s).

Bivalent (also called divalent) or bispecific single chain variable fragments (bi-scFv or di-scFv in (scFv) ₂ format can be constructed by joining two scFv molecules (for example, with the linkers described herein above). If the two molecules scFvs have the same binding specificity, the resulting (scFv)2 molecule would preferably be called bivalent (i.e., it has two antigenic determinants for the same target epitope).If two scFv molecules have different binding specificities, then the resulting As a result, the (scFv)2 molecule will preferably be called bispecific. Linking can be accomplished by producing a single peptide chain with two VH regions and two VL regions, resulting in the formation of tandem scFvs (see, for example, Kufer P. et al., (2004 ) Trends in Biotechnology 22(5):238-244) Another possibility is to create scFv molecules with linker peptides that are too short for two variable regions to fold together (eg, approximately five amino acids), forcing the scFv to dimerize. This type is known as a diabody (see, for example, Hollinger, Philipp et al., (July 1993) Proceedings of the National Academy of Sciences of the United States of America 90 (14): 6444-8).

In accordance with the present invention, any of the first, second, or first and second domains may constitute a single domain antibody, respectively a variable domain, or at least a CDR portion of a single domain antibody. Single domain antibodies contain only one (monomeric) antibody variable domain that is capable of selectively binding to a specific antigen, independent of other V regions or domains. The first single-domain antibodies were constructed from the heavy chain of antibodies identified in camels and are called VHH fragments. Cartilaginous fish also have heavy chain antibodies (IgNARs) from which single domain antibodies called VNAR fragments can be produced. An alternative approach is to separate the dimeric variable domains of common immunoglobulins, such as human or rodent immunoglobulins, into monomers, thereby producing VH or VL in as a single domain Ab. Although most single-domain antibody research is currently based on heavy chain variable domains, light chain-derived nanobodies have also been shown to specifically bind to target epitopes. Examples of single domain antibodies are called sdAbs, nanobodies, or single variable domain antibodies.

Therefore, (single domain mAb)2 is a monoclonal antibody construct consisting of (at least) two single domain monoclonal antibodies that are individually selected from the group consisting of VH, VL, VHH and VNAR. The linker preferably takes the form of a peptide linker. Similarly, an scAv single-domain mAb is a monoclonal antibody construct consisting of at least one single-domain antibody described above and one scFv molecule described above. Again, the linker is preferably in the form of a peptide linker.

Whether an antibody construct competes for binding with another specified antibody construct can be determined in a competition assay, such as a competitive ELISA or a cell competition assay. Avidin-coated microparticles (granules) can also be used. Similarly, with an avidin-coated ELISA plate when reacting with a biotinylated protein, each of these beads can be used as a substrate for the assay. The antigen is applied to the bead and then the first antibody is pre-coated. A second antibody is added and the presence of any additional binding is determined. Possible means for reading the data include flow cytometry.

T cells or T lymphocytes are a type of lymphocyte (which themselves are a type

- 27 046123 white blood cells), which plays a major role in cell-mediated immunity. There are several subsets of T cells, each with a distinct function. T cells can be distinguished from other lymphocytes such as B cells and NK cells by the presence of the T cell receptor (TCR) on the cell surface. TCR is responsible for recognizing antigens bound to major histocompatibility complex (MHC) molecules and consists of two different protein chains. In 95% of T cells, TCR consists of alpha (α) and beta (b) chains. When the TCR comes into contact with an antigenic peptide and the MHC (peptide/MHC complex), the T cell is activated through a sequence of biochemical events mediated by associated enzymes, coreceptors, specialized adapter molecules, and activated or released transcription factors.

The CD3 receptor complex is a protein complex and consists of four chains. In mammals, the complex contains a CD3y (gamma) chain, a CD3δ (delta) chain, and two CD3ε (epsilon) chains. These chains bind to the T-cell receptor (TRC) and the so-called ζ (zeta) chain to form the T-cell receptor-CD3 complex and generate an activation signal in T lymphocytes. The CD3Y (gamma), CD3δ (delta), and CD3ε (epsilon) chains are highly related cell surface proteins of the immunoglobulin superfamily, containing a single extracellular immunoglobulin domain. The intracellular tails of CD3 molecules contain one conserved motif known as the immunoreceptor tyrosine activating motif, or ITAM for short, which is key to the function of the TKA signaling pathway. The SEZ-epsilon molecule is a polypeptide that in humans is encoded by the CD3E gene, which is located on chromosome 11. The most preferred epitope of SEZ-epsilon is located within amino acid residues 1-27 of the extracellular domain of human CD3-epsilon. It is understood that the antibody constructs of the present invention are generally and advantageously less likely to exhibit nonspecific T cell activation, which is undesirable in specific immunotherapy. This leads to a reduced risk of side effects.

Redirected lysis of target cells through T cell recruitment by a multispecific, at least bispecific antibody construct involves formation of a cytolytic synapse and delivery of perforin and granzymes. Recruited T cells are capable of sequential lysis of target cells and are not affected by immunosurveillance evasion mechanisms that interfere with the processing and presentation of peptide antigens or clonal differentiation of T cells; see for example WO 2007/042261.

Cytotoxicity mediated by the antibody constructs of the present invention can be determined in various ways. The effector cells may be, for example, enriched (human) CD8-positive T cells or unstimulated (human) peripheral blood mononuclear cells (PBMC). If the target cells are derived from macaque or express or are transfected with macaque MUC17, which is bound by the first domain, the effector cells must also be derived from macaque, such as in the case of a macaque T cell line such as 4119LnPx. The target cells must express (at least the extracellular domain) MUC17, for example, human or macaque MUC17. The target cells may be a cell line (such as CHO cells) stably or transiently transfected with MUC17, for example, human or macaque MUC17. EC ₅₀ values are generally expected to be lower for target cell lines expressing MUC17 on the cell surface at higher levels. The ratio of effector cells to target cells (E:T) is usually approximately 10:1, but can also vary. The cytotoxic activity of MUC17 bispecific antibody constructs can be measured in a ⁵¹ Cr release assay (incubation time approximately 18 hours) or a FACS cytotoxicity assay (incubation time approximately 48 hours). Modifications to the incubation time (cytotoxic reaction) in the assay are also possible. Other methods for determining cytotoxicity are well known to one skilled in the art and include the MTT or MTS assay, ATP-based assays including bioluminescence assay, sulforhodamine B (SRB) assay, WST assay, clonogenic assay and ECIS technique.

The cytotoxic activity mediated by the MUC17xCD3 bispecific antibody constructs of the present invention is preferably measured in a cell-based cytotoxicity assay. It can also be measured in a ⁵¹ Cr release assay. It is represented by the EC50 value, which corresponds to the half-maximal effective concentration (the concentration of the antibody construct that induces half the cytotoxic response between the minimum and maximum). Preferably, the EC ₅₀ for MUC17xCD3 bispecific antibody constructs is <5000 pM or <4000 pM, more preferably <3000 pM or <2000 pM, even more preferably <1000 pM or <500 pM, even more preferably <400 pM or < 300 pM, even more preferably <200 pM, even more preferably <100 pM, even more preferably <50 pM, even more preferably <20 pM or <10 pM, and most preferably <5 pM.

The above EC ₅₀ values can be obtained from various assays. A specialist in this

- 28 046123 it is known in the art that the EC50 value can be expected to be lower when stimulated/enriched CD8+ T cells are used as effector cells compared to unstimulated PBMCs. In addition, EC50 values can be expected to be lower when target cells express high amounts of MUC17 compared to low target expression in rats. For example, when stimulated/enriched human CD8+ T cells are used as effector cells (and either MUC17-transfected cells, such as CHO cells, or MUC17-positive human cell lines are used as target cells), the EC50 value for the The anti-MUC17xCD3 bispecific antibody is preferably <1000 pM, more preferably <500 pM, even more preferably <250 pM, even more preferably <100 pM, even more preferably <50 pM, even more preferably <10 pM, and most preferably <5 pM. When human PBMCs are used as effector cells, the EC50 value for the MUC17xCD3 bispecific antibody construct is preferably <5000 pM or <4000 pM (particularly when the target cells are MUC17 positive human cell lines), more preferably <2000 pM (particularly when the target cells are cells transfected with MUC17, such as CHO cells), more preferably <1000 pM or <500 pM, even more preferably <200 pM, even more preferably <150 pM, even more preferably <100 pM and most preferably <50 pM or lower. If a macaque T cell line, such as LnPx4119, is used as the effector cells, and macaque MUC17-transfected cells, such as CHO cells, are used as the target cell line, the EC ₅₀ value for the MUC17xCD3 bispecific antibody construct is preferably <2000 pM or <1500 pM, more preferably <1000 pM or <500 pM, even more preferably <300 pM or <250 pM, even more preferably <100 pM and most preferably <50 pM.

Preferably, the MUC17xCD3 bispecific antibody constructs of the present invention do not induce/mediate lysis or substantially induce/mediate lysis of MUC17 negative cells such as CHO cells. The term does not induce lysis, does not induce lysis, does not mediate lysis, or essentially does not mediate lysis means that the antibody construct of the present invention induces or mediates lysis of no more than 30%, preferably no more than 20%, more preferably no more than 10 %, particularly preferably no more than 9, 8, 7, 6 or 5% of MUC17 negative cells, where lysis of the MUC17 positive human cell line is taken to be 100%. This typically corresponds to antibody construct concentrations of no more than 500 nM. One skilled in the art will know how to detect cell lysis without undue effort. In addition, specific instructions for determining cell lysis are provided herein.

The difference in cytotoxic activity between the monomeric and dimeric isoforms of individual MUC17xCD3 bispecific antibody constructs is referred to as potency discrepancy. This difference in efficiency can, for example, be calculated as the ratio of EC ₅₀ values for monomeric and dimeric forms of molecules. The difference in potency of the MUC17xCD3 bispecific antibody constructs of the present invention is preferably <5, more preferably <4, even more preferably <3, even more preferably <2, and most preferably <1.

The first and/or second (or any additional) binding domain(s) of the antibody construct of the present invention are preferably cross-species specific for mammalian primates. Cross-species specific CD3 binding domains are described, for example, in WO 2008/119567. In one embodiment, the first and/or second binding domain, in addition to binding to human MUC17 and human CD3, respectively, will also bind to MUC17/CD3, including but not limited to new world primates (such as Callithrix jacchus, Saguinus Oedipus, or Saimiri sciureus) , old world primates (such as baboons and macaques), gibbons, and non-human homininae.

In one embodiment of the antibody construct of the present invention, the first domain binds to human MUC17 and further binds to macaque MUC17, such as Macaca fascicularis MUC17, and more preferably macaque MUC17 expressed on the surface of cells, such as CHO or 293 cells The affinity of the first domain for MUC17, preferably for human MUC17, is preferably <100 nM or <50 nM, more preferably <25 nM or <20 nM, more preferably <15 nM or <10 nM, even more preferably <5 nM , even more preferably <2.5 nM or <2 nM, even more preferably <1 nM, even more preferably <0.6 nM, even more preferably <0.5 nM and most preferably <0.4 nM. Affinity can be determined, for example, in the BIAcore assay or in the Scatchard assay. Other methods for determining affinity are also well known to those skilled in the art. The affinity of the first domain with MUC17 is preferably <15 nM, more preferably <10 nM, even more preferably <5 nM, even more preferably <1 nM, even more preferably

- 29 046123 <0.5 nM, even more preferably <0.1 nM, and most preferably <0.05 nM or even <0.01 nM.

Preferably, the difference in affinity of the antibody constructs of the present invention for binding to macaque MUC17 versus human MUC17 [ma MUC17:hu MUC17] (as determined, for example, by BiaCore or Scatchard assay) is <100, preferably <20, more preferably <15, even more preferably <10, even more preferably <8, more preferably <6 and most preferably <2. Preferred ranges for the difference in affinity of the antibody constructs of the present invention for binding to macaque MUC17 compared to human MUC17 are from 0.1 to 20, more preferably from 0.2 to 10, even more preferably from 0.3 to 6. even more preferably from 0.5 to 3 or from 0.5 to 2.5 and most preferably from 0.5 to 2 or from 0.6 to 2.

The second domain of the antibody construct of the present invention binds to human CD3epsilon and/or Masasa CD3epsilon. In a preferred embodiment, the second domain further binds to the epsilon SEZ of Callithrix jacchus, Saguinus Oedipus or Saimiri sciureus. Both Callithrix jacchus and Saguinus oedipus are new world primates belonging to the family Callitrichidae, while Saimiri sciureus is a new world primate belonging to the family Cebidae.

For the antibody construct of the present invention, it is preferred that the second domain, which binds to an extracellular epitope of the human and/or Macasa SEZ epsilon chain, comprises a VL region containing CDR-L1, CDR-L2 and CDR-L3 selected from:

(a) CDR-L1 given at SEQ ID NO: 27 in WO 2008/119567, CDR-L2 given at SEQ ID NO: 28 in WO 2008/119567, and CDR-L3 given at SEQ ID NO: 29 in WO 2008/119567;

(b) CDR-L1 given under SEQ ID NO: 117 in WO 2008/119567, CDR-L2 given under SEQ ID NO: 118 in WO 2008/119567, and CDR-L3 given under SEQ ID NO: 119 in WO 2008/119567; and (c) CDR-L1 given under SEQ ID NO: 153 in WO 2008/119567, CDR-L2 given under SEQ ID NO: 154 in WO 2008/119567, and CDR-L3 given under SEQ ID NO: 155 in WO 2008/119567.

In an even more preferred embodiment of the antibody construct of the present invention, the second domain that binds to an extracellular epitope of the human CD3 epsilon chain and/or Masasa chain comprises a VH region comprising CDR-H1, CDR-H2 and CDR-H3 selected from :

(a) CDR-H1 given under SEQ ID NO: 12 in WO 2008/119567, CDR-H2 given under SEQ ID NO: 13 in WO 2008/119567, and CDR-H3 given under SEQ ID NO: 14 in WO 2008/119567;

(b) CDR-H1 given at SEQ ID NO: 30 in WO 2008/119567, CDR-H2 given at SEQ ID NO: 31 in WO 2008/119567, and CDR-H3 given at SEQ ID NO: 32 in WO 2008/119567;

(c) CDR-H1 given at SEQ ID NO: 48 in WO 2008/119567, CDR-H2 given at SEQ ID NO: 49 in WO 2008/119567, and CDR-H3 given at SEQ ID NO: 50 in WO 2008/119567;

(d) CDR-H1 given under SEQ ID NO: 66 in WO 2008/119567, CDR-H2 given under SEQ ID NO: 67 in WO 2008/119567, and CDR-H3 given under SEQ ID NO: 68 in WO 2008/119567;

(e) CDR-H1 given at SEQ ID NO: 84 in WO 2008/119567, CDR-H2 given at SEQ ID NO: 85 in WO 2008/119567, and CDR-H3 given at SEQ ID NO: 86 in WO 2008/119567;

(f) CDR-H1 given under SEQ ID NO: 102 in WO 2008/119567, CDR-H2 given under SEQ ID NO: 103 in WO 2008/119567, and CDR-H3 given under SEQ ID NO: 104 in WO 2008/119567;

(g) CDR-H1 given under SEQ ID NO: 120 in WO 2008/119567, CDR-H2 given under SEQ ID NO: 121 in WO 2008/119567, and CDR-H3 given under SEQ ID NO: 122 in WO 2008/119567;

(h) CDR-H1 given under SEQ ID NO: 138 in WO 2008/119567, CDR-H2 given under SEQ ID NO: 139 in WO 2008/119567, and CDR-H3 given under SEQ ID NO: 140 in WO 2008/119567;

(i) CDR-H1 given under SEQ ID NO: 156 in WO 2008/119567, CDR-H2 given under SEQ ID NO: 157 in WO 2008/119567, and CDR-H3 given under SEQ ID NO: 158 in WO 2008/119567; and (j) CDR-H1 given under SEQ ID NO: 174 in WO 2008/119567, CDR-H2 given under SEQ ID NO: 175 in WO 2008/119567, and CDR-H3 given under SEQ ID NO: 176 in WO 2008/119567.

In a preferred embodiment of the antibody construct of the present invention, the above-described three VL CDR groups are combined with the above-described ten VH CDR groups within the second binding domain to produce (30) groups, each containing CDR-L 1-3 and CDR-H 1 -3.

For the antibody construct of the present invention, it is preferred that the second domain that binds CD3 comprises a VL region selected from the group consisting of those set forth in SEQ ID NOs: 17, 21, 35, 39, 53, 57. 71, 75, 89, 93, 107, 111, 125, 129, 143, 147, 161, 165, 179 or 183 of WO 2008/119567 or given under SEQ ID NO: 13 in accordance with the present invention.

It is also preferred that the second domain that binds CD3 comprises a VH region selected from the group consisting of those set forth under SEQ ID NO: 15, 19, 33, 37, 51, 55, 69, 73, 87, 91 , 105, 109, 123, 127, 141, 145, 159, 163, 177 or 181 in WO 2008/119567 or given under SEQ ID NO: 14.

- 30 046123

More preferably, the antibody construct of the present invention is characterized by a second domain that binds to CD3, comprising a VL region and a VH region selected from the group consisting of:

(a) the VL portion given by SEQ ID NO: 17 or 21 in WO 2008/119567 and the VH portion given by SEQ ID NO: 15 or 19 in WO 2008/119567;

(b) the VL portion given by SEQ ID NO: 35 or 39 in WO 2008/119567 and the VH portion given by SEQ ID NO: 33 or 37 in WO 2008/119567;

(c) the VL region given by SEQ ID NO: 53 or 57 in WO 2008/119567, and the VH region given by SEQ ID NO: 51 or 55 in WO 2008/119567;

(d) the VL portion given by SEQ ID NO: 71 or 75 in WO 2008/119567 and the VH portion given by SEQ ID NO: 69 or 73 in WO 2008/119567;

(e) the VL portion given by SEQ ID NO: 89 or 93 in WO 2008/119567 and the VH portion given by SEQ ID NO: 87 or 91 in WO 2008/119567;

(f) the VL portion given by SEQ ID NO: 107 or 111 in WO 2008/119567 and the VH portion given by SEQ ID NO: 105 or 109 in WO 2008/119567;

(g) the VL portion of SEQ ID NO: 125 or 129 in WO 2008/119567 and the VH portion of SEQ ID NO: 123 or 127 in WO 2008/119567;

(h) the VL portion given by SEQ ID NO: 143 or 147 in WO 2008/119567 and the VH portion given by SEQ ID NO: 141 or 145 in WO 2008/119567;

(i) the VL portion given by SEQ ID NO: 161 or 165 in WO 2008/119567 and the VH portion given by SEQ ID NO: 159 or 163 in WO 2008/119567; and (j) the VL portion given by SEQ ID NO: 179 or 183 in WO 2008/119567, and the VH portion given by SEQ ID NO: 177 or 181 in WO 2008/119567.

Also in connection with the antibody design of the present invention, it is preferred that the second domain that binds to CD3 comprises a VL region set forth in SEQ ID NO: 13 and a VH region set forth in SEQ ID NO: 14.

In accordance with a preferred embodiment of the antibody construct of the present invention, the first and/or second domain have the following format: pairs of VH regions and VL regions are in single chain antibody (scFv) format. Sections VH and VL are arranged in the order VH-VL or VL-VH. Preferably, the VH region is located N-terminal to the linker sequence and the VL region is located C-terminal to the linker sequence.

A preferred embodiment of the above-described antibody construct of the present invention is characterized by a second domain that binds to CD3 containing an amino acid sequence selected from the group consisting of SEQ ID NO: 23, 25, 41, 43, 59, 61, 77, 79 , 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185 or 187 in WO 2008/119567, or given under SEQ ID NO: 15.

It is also intended that the first binding domain of the antibody construct of the present invention comprises a VL region comprising CDR-L1, CDR-L2 and CDR-L3 and a VH region containing CDR-H1, CDR-H2 and CDR-3 selected from the group consisting of:

(a) CDR-L1 given under SEQ ID NO: 36, CDR-L2 given under SEQ ID NO: 37, and CDRL3 given under SEQ ID NO: 38, and CDR-H1 given under SEQ ID NO: 33 , CDR-H2, shown under SEQ ID NO: 34, and CDR-H3, shown under SEQ ID NO: 35;

(b) CDR-L1 given at SEQ ID NO: 47, CDR-L2 given at SEQ ID NO: 48, and CDRL3 given at SEQ ID NO: 49, and CDR-H1 given at SEQ ID NO: 44 , CDR-H2, shown under SEQ ID NO: 45, and CDR-H3, shown under SEQ ID NO: 46;

(c) CDR-L1, shown under SEQ ID NO: 58, CDR-L2, shown under SEQ ID NO: 59, and CDRL3, shown under SEQ ID NO: 60, and CDR-H1, shown under SEQ ID NO: 55 , CDR-H2, shown under SEQ ID NO: 56, and CDR-H3, shown under SEQ ID NO: 57;

(d) CDR-L1, shown under SEQ ID NO: 69, CDR-L2, shown under SEQ ID NO: 70, and CDRL3, shown under SEQ ID NO: 71, and CDR-H1, shown under SEQ ID NO: 66 , CDR-H2, shown under SEQ ID NO: 67, and CDR-H3, shown under SEQ ID NO: 68;

(e) CDR-L1 given under SEQ ID NO: 80, CDR-L2 given under SEQ ID NO: 81, and CDRL3 given under SEQ ID NO: 82, and CDR-H1 given under SEQ ID NO: 77 , CDR-H2, shown under SEQ ID NO: 78, and CDR-H3, shown under SEQ ID NO: 79;

(f) CDR-L1 given at SEQ ID NO: 91, CDR-L2 given at SEQ ID NO: 92, and CDRL3 given at SEQ ID NO: 93, and CDR-H1 given at SEQ ID NO: 88 , CDR-H2, shown under SEQ ID NO: 89, and CDR-H3, shown under SEQ ID NO: 90;

(g) CDR-L1 given under SEQ ID NO: 102, CDR-L2 given under SEQ ID NO: 103, and CDR-L3 given under SEQ ID NO: 104, and CDR-H1 given under SEQ ID NO: : 99, CDR-H2 given under SEQ ID NO: 100, and CDR-H3 given under SEQ ID NO: 101;

(h) CDR-L1 set forth at SEQ ID NO: 113, CDR-L2 set forth at SEQ ID NO: 114, and

- 31 046123

CDR-L3, shown under SEQ ID NO: 115, and CDR-H1, shown under SEQ ID NO: software, CDR-H2, shown under SEQ ID NO: 111, and CDR-H3, shown under SEQ ID NO: 112;

(i) CDR-L1 given under SEQ ID NO: 124, CDR-L2 given under SEQ ID NO: 125, and

CDR-L3, shown under SEQ ID NO: 126, and CDR-H1, shown under SEQ ID NO: 121, CDR-H2, shown under SEQ ID NO: 122, and CDR-H3, shown under SEQ ID NO: 123;

(j) CDR-L1 given under SEQ ID NO: 135, CDR-L2 given under SEQ ID NO: 136, and CDR-L3 given under SEQ ID NO: 137, and CDR-H1 given under SEQ ID NO: 137 : 132, CDR-H2 given under SEQ ID NO: 133, and CDR-H3 given under SEQ ID NO: 134;

(k) CDR-L1 given under SEQ ID NO: 146, CDR-L2 given under SEQ ID NO: 147, and CDR-L3 given under SEQ ID NO: 148, and CDR-H1 given under SEQ ID NO: : 143, CDR-H2 given under SEQ ID NO: 144, and CDR-H3 given under SEQ ID NO: 145;

(l) CDR-L1 given under SEQ ID NO: 157, CDR-L2 given under SEQ ID NO: 158, and CDR-L3 given under SEQ ID NO: 159, and CDR-H1 given under SEQ ID NO: : 154, CDR-H2 given under SEQ ID NO: 155, and CDR-H3 given under SEQ ID NO: 156;

(m) CDR-L1 given under SEQ ID NO: 168, CDR-L2 given under SEQ ID NO: 169, and CDR-L3 given under SEQ ID NO: 170, and CDR-H1 given under SEQ ID NO: : 165, CDR-H2 given under SEQ ID NO: 166, and CDR-H3 given under SEQ ID NO: 167;

(n) CDR-L1 given under SEQ ID NO: 179, CDR-L2 given under SEQ ID NO: 180, and CDR-L3 given under SEQ ID NO: 181, and CDR-H1 given under SEQ ID NO: : 176, CDR-H2 given under SEQ ID NO: 177, and CDR-H3 given under SEQ ID NO: 178;

(o) CDR-L1 given under SEQ ID NO: 190, CDR-L2 given under SEQ ID NO: 191, and CDR-L3 given under SEQ ID NO: 192, and CDR-H1 given under SEQ ID NO: : 187, CDR-H2 given under SEQ ID NO: 188, and CDR-H3 given under SEQ ID NO: 189;

(p) CDR-L1 given under SEQ ID NO: 201, CDR-L2 given under SEQ ID NO: 202, and CDR-L3 given under SEQ ID NO: 203, and CDR-H1 given under SEQ ID NO: : 198, CDR-H2 given under SEQ ID NO: 199, and CDR-H3 given under SEQ ID NO: 200;

(q) CDR-L1 given under SEQ ID NO: 212, CDR-L2 given under SEQ ID NO: 213, and CDR-L3 given under SEQ ID NO: 214, and CDR-H1 given under SEQ ID NO: : 209, CDR-H2 given under SEQ ID NO: 210, and CDR-H3 given under SEQ ID NO: 211;

(r) CDR-L1 given under SEQ ID NO: 223, CDR-L2 given under SEQ ID NO: 224, and CDR-L3 given under SEQ ID NO: 225, and CDR-H1 given under SEQ ID NO: : 220, CDR-H2 given under SEQ ID NO: 221, and CDR-H3 given under SEQ ID NO: 222;

(s) CDR-L1 given under SEQ ID NO: 234, CDR-L2 given under SEQ ID NO: 235, and CDR-L3 given under SEQ ID NO: 236, and CDR-H1 given under SEQ ID NO: : 231, CDR-H2 given under SEQ ID NO: 232, and CDR-H3 given under SEQ ID NO: 233;

(t) CDR-L1 given under SEQ ID NO: 245, CDR-L2 given under SEQ ID NO: 246, and CDR-L3 given under SEQ ID NO: 247, and CDR-H1 given under SEQ ID NO: : 242, CDR-H2 given under SEQ ID NO: 243, and CDR-H3 given under SEQ ID NO: 244;

(u) CDR-L1 given under SEQ ID NO: 256, CDR-L2 given under SEQ ID NO: 257, and CDR-L3 given under SEQ ID NO: 258, and CDR-H1 given under SEQ ID NO: : 253, CDR-H2 given under SEQ ID NO: 254, and CDR-H3 given under SEQ ID NO: 255;

(v) CDR-L1 given under SEQ ID NO: 267, CDR-L2 given under SEQ ID NO: 268, and CDR-L3 given under SEQ ID NO: 269, and CDR-H1 given under SEQ ID NO: : 264, CDR-H2 given under SEQ ID NO: 265, and CDR-H3 given under SEQ ID NO: 266;

(w) CDR-L1 given under SEQ ID NO: 278, CDR-L2 given under SEQ ID NO: 279, and CDR-L3 given under SEQ ID NO: 280, and CDR-H1 given under SEQ ID NO: : 275, CDR-H2 given under SEQ ID NO: 276, and CDR-H3 given under SEQ ID NO: 276;

(x) CDR-L1 given under SEQ ID NO: 289, CDR-L2 given under SEQ ID NO: 290, and CDR-L3 given under SEQ ID NO: 291, and CDR-H1 given under SEQ ID NO: : 286, CDR-H2 given under SEQ ID NO: 287, and CDR-H3 given under SEQ ID NO: 288;

(y) CDR-L1 given under SEQ ID NO: 300, CDR-L2 given under SEQ ID NO: 301, and CDR-L3 given under SEQ ID NO: 302, and CDR-H1 given under SEQ ID NO: : 297, CDR-H2 given under SEQ ID NO: 298, and CDR-H3 given under SEQ ID NO: 299;

(z) CDR-L1 given under SEQ ID NO: 311, CDR-L2 given under SEQ ID NO: 312, and CDR-L3 given under SEQ ID NO: 313, and CDR-H1 given under SEQ ID NO: : 308, CDR-H2 given under SEQ ID NO: 309, and CDR-H3 given under SEQ ID NO: 310;

(aa) CDR-L1 given under SEQ ID NO: 322, CDR-L2 given under SEQ ID NO: 323, and CDR-L3 given under SEQ ID NO: 324, and CDR-H1 given under SEQ ID NO: : 319, CDR-H2 given under SEQ ID NO: 320, and CDR-H3 given under SEQ ID NO: 321;

(ab) CDR-L1 given under SEQ ID NO: 333, CDR-L2 given under SEQ ID NO: 334, and CDR-L3 given under SEQ ID NO: 335, and CDR-H1 given under SEQ ID NO: : 330, CDR-H2 given under SEQ ID NO: 331, and CDR-H3 given under SEQ ID NO: 332;

- 32 046123 (ac) CDR-L1 given under SEQ ID NO: 344, CDR-L2 given under SEQ ID NO: 345, and

CDR-L3, shown under SEQ ID NO: 346, and CDR-H1, shown under SEQ ID NO: 341, CDR-H2, shown under SEQ ID NO: 342, and CDR-H3, shown under SEQ ID NO: 343;

(ad) CDR-L1 given under SEQ ID NO: 355, CDR-L2 given under SEQ ID NO: 356, and CDR-L3 given under SEQ ID NO: 357, and CDR-H1 given under SEQ ID NO: : 352, CDR-H2 given under SEQ ID NO: 353, and CDR-H3 given under SEQ ID NO: 354;

(ae) CDR-L1 given under SEQ ID NO: 366, CDR-L2 given under SEQ ID NO: 367, and CDR-L3 given under SEQ ID NO: 368, and CDR-H1 given under SEQ ID NO: : 363, CDR-H2 given under SEQ ID NO: 364, and CDR-H3 given under SEQ ID NO: 365;

(af) CDR-L1 given under SEQ ID NO: 377, CDR-L2 given under SEQ ID NO: 378, and CDR-L3 given under SEQ ID NO: 379, and CDR-H1 given under SEQ ID NO: : 374, CDR-H2 given under SEQ ID NO: 375, and CDR-H3 given under SEQ ID NO: 376;

(ag) CDR-L1 given under SEQ ID NO: 388, CDR-L2 given under SEQ ID NO: 389, and CDR-L3 given under SEQ ID NO: 390, and CDR-H1 given under SEQ ID NO: : 385, CDR-H2 given under SEQ ID NO: 386, and CDR-H3 given under SEQ ID NO: 386;

(ah) CDR-L1 given under SEQ ID NO: 399, CDR-L2 given under SEQ ID NO: 400, and CDR-L3 given under SEQ ID NO: 401, and CDR-H1 given under SEQ ID NO: : 396, CDR-H2 given under SEQ ID NO: 397, and CDR-H3 given under SEQ ID NO: 398;

(ai) CDR-L1 given under SEQ ID NO: 410, CDR-L2 given under SEQ ID NO: 411, and CDR-L3 given under SEQ ID NO: 412, and CDR-H1 given under SEQ ID NO: : 407, CDR-H2 given under SEQ ID NO: 408, and CDR-H3 given under SEQ ID NO: 409;

(aj) CDR-L1 given under SEQ ID NO: 421, CDR-L2 given under SEQ ID NO: 422, and CDR-L3 given under SEQ ID NO: 423, and CDR-H1 given under SEQ ID NO: : 418, CDR-H2 given under SEQ ID NO: 419, and CDR-H3 given under SEQ ID NO: 420;

(ak) CDR-L1 given under SEQ ID NO: 432, CDR-L2 given under SEQ ID NO: 433, and CDR-L3 given under SEQ ID NO: 434, and CDR-H1 given under SEQ ID NO: : 429, CDR-H2 given under SEQ ID NO: 430, and CDR-H3 given under SEQ ID NO: 431;

(al) CDR-L1 given under SEQ ID NO: 443, CDR-L2 given under SEQ ID NO: 444, and CDR-L3 given under SEQ ID NO: 445, and CDR-H1 given under SEQ ID NO: : 440, CDR-H2 given under SEQ ID NO: 441, and CDR-H3 given under SEQ ID NO: 442;

(am) CDR-L1 given under SEQ ID NO: 454, CDR-L2 given under SEQ ID NO: 455, and CDR-L3 given under SEQ ID NO: 456, and CDR-H1 given under SEQ ID NO: : 451, CDR-H2 given under SEQ ID NO: 452, and CDR-H3 given under SEQ ID NO: 453;

(an) CDR-L1 given under SEQ ID NO: 465, CDR-L2 given under SEQ ID NO: 466, and CDR-L3 given under SEQ ID NO: 467, and CDR-H1 given under SEQ ID NO: 467 : 462, CDR-H2 given under SEQ ID NO: 463, and CDR-H3 given under SEQ ID NO: 464;

(ao) CDR-L1 given under SEQ ID NO: 476, CDR-L2 given under SEQ ID NO: 477, and CDR-L3 given under SEQ ID NO: 478, and CDR-H1 given under SEQ ID NO: : 473, CDR-H2 given under SEQ ID NO: 474, and CDR-H3 given under SEQ ID NO: 475;

(ap) CDR-L1 given under SEQ ID NO: 487, CDR-L2 given under SEQ ID NO: 488, and CDR-L3 given under SEQ ID NO: 489, and CDR-H1 given under SEQ ID NO: : 484, CDR-H2 given under SEQ ID NO: 485, and CDR-H3 given under SEQ ID NO: 486;

(aq) CDR-L1 given under SEQ ID NO: 498, CDR-L2 given under SEQ ID NO: 499, and CDR-L3 given under SEQ ID NO: 500, and CDR-H1 given under SEQ ID NO: : 495, CDR-H2 given under SEQ ID NO: 496, and CDR-H3 given under SEQ ID NO: 497;

(ar) CDR-L1 given under SEQ ID NO: 509, CDR-L2 given under SEQ ID NO: 510, and CDR-L3 given under SEQ ID NO: 511, and CDR-H1 given under SEQ ID NO: : 506, CDR-H2 given under SEQ ID NO: 507, and CDR-H3 given under SEQ ID NO: 508; and (as) CDR-L1 given under SEQ ID NO: 520, CDR-L2 given under SEQ ID NO: 521, and CDR-L3 given under SEQ ID NO: 522, and CDR-H1 given under SEQ ID NO: 522 NO: 517, CDR-H2 given under SEQ ID NO: 518, and CDR-H3 given under SEQ ID NO: 519; where preferred are, for example:

(c) CDR-L1 given at SEQ ID NO: 58, CDR-L2 given at SEQ ID NO: 59, and CDRL3 given at SEQ ID NO: 60, and CDR-H1 given at SEQ ID NO: 55 , CDR-H2, shown under SEQ ID NO: 56, and CDR-H3, shown under SEQ ID NO: 57;

(n) CDR-L1 given at SEQ ID NO: 179, CDR-L2 given at SEQ ID NO: 180, and CDRL3 given at SEQ ID NO: 181, and CDR-H1 given at SEQ ID NO: 176 , CDR-H2, shown under SEQ ID NO: 177, and CDR-H3, shown under SEQ ID NO: 178;

(ac) CDR-L1 given under SEQ ID NO: 344, CDR-L2 given under SEQ ID NO: 345, and CDR-L3 given under SEQ ID NO: 346, and CDR-H1 given under SEQ ID NO: : 341, CDR-H2 given under SEQ ID NO: 342, and CDR-H3 given under SEQ ID NO: 343; and (aj) CDR-L1 set forth at SEQ ID NO: 421, CDR-L2 set forth at SEQ ID NO: 422, and

- 33 046123

CDR-L3, shown at SEQ ID NO: 423, and CDR-H1, shown at SEQ ID NO: 418, CDR-H2, shown at SEQ ID NO: 419, and CDR-H3, shown at SEQ ID NO: 420.

It is further contemplated that the first binding domain of the antibody construct of the present invention comprises a VH region and a VL region selected from the group consisting of:

(a) the VL region shown under SEQ ID NO: 40 and the VH region shown under SEQ ID NO: 39;

(b) the VL region shown under SEQ ID NO: 51 and the VH region shown under SEQ ID NO: 50;

(c) the VL region of SEQ ID NO: 62 and the VH region of SEQ ID NO: 61;

(d) the VL region of SEQ ID NO: 73 and the VH region of SEQ ID NO: 72;

(e) the VL region of SEQ ID NO: 84 and the VH region of SEQ ID NO: 83;

(f) the VL region of SEQ ID NO: 95 and the VH region of SEQ ID NO: 94;

(g) the VL region shown under SEQ ID NO: 105;

(h) the VL region shown under SEQ ID NO: 116;

(i) the VL section shown under SEQ ID NO: 127;

(j) the VL region shown under SEQ ID NO: 138;

(k) the VL section shown under SEQ ID NO: 149;

(l) the VL region shown under SEQ ID NO: 160;

(m) the VL region shown under SEQ ID NO: 171;

(n) the VL region shown under SEQ ID NO: 182;

(o) the VL region shown under SEQ ID NO: 193;

(p) the VL region shown under SEQ ID NO: 204;

(q) the VL region shown under SEQ ID NO: 215;

(r) the VL region shown under SEQ ID NO: 226;

(s) the VL region shown under SEQ ID NO: 237;

(t) the VL region shown under SEQ ID NO: 248;

(u) the VL section shown under SEQ ID NO: 259;

(v) the VL region shown under SEQ ID NO: 270;

(w) the VL region shown under SEQ ID NO: 281;

(x) the VL region shown under SEQ ID NO: 292;

(y) the VL region shown under SEQ ID NO: 303;

(z) the VL region shown under SEQ ID NO: 314;

(aa) the VL region shown under SEQ ID NO: 325;

(ab) the VL region shown under SEQ ID NO: 336;

(ac) the VL region shown under SEQ ID NO: 347;

(ad) the VL region shown under SEQ ID NO: 358;

(ae) the VL region shown under SEQ ID NO: 369;

(af) the VL region shown under SEQ ID NO: 380;

106, and the VH section given under SEQ ID NO:

117, and the VH section given under SEQ ID NO:

128, and the VH section given under SEQ ID NO:

139, and the VH section given under SEQ ID NO:

150, and the VH section given under SEQ ID NO:

161, and the VH section given under SEQ ID NO:

172, and the VH section given under SEQ ID NO:

183, and the VH section given under SEQ ID NO:

194, and the VH section given under SEQ ID NO:

205, and the VH section given under SEQ ID NO:

216, and the VH section given under SEQ ID NO:

227, and the VH section given under SEQ ID NO:

238, and the VH section given under SEQ ID NO:

249, and the VH section given under SEQ ID NO:

260, and the VH section given under SEQ ID NO:

271, and the VH section given under SEQ ID NO:

282, and the VH section given under SEQ ID NO:

293, and the VH section given under SEQ ID NO:

304, and the VH section given under SEQ ID NO:

315, and the VH section given under SEQ ID NO:

326, and the VH section given under SEQ ID NO:

337, and the VH region given under SEQ ID NO:

348, and the VH region given under SEQ ID NO:

359, and the VH region given under SEQ ID NO:

370, and the VH section given under SEQ ID NO:

381, and the VH region given under SEQ ID NO:

- 34 046123 (ag) section VL, given under SEQ ID NO: 392, and section VH, given under SEQ ID NO:

391;

(ah) the VL region of SEQ ID NO: 403 and the VH region of SEQ ID NO: 402;

(ai) the VL portion of SEQ ID NO: 414 and the VH portion of SEQ ID NO: 413;

(aj) the VL region of SEQ ID NO: 425 and the VH region of SEQ ID NO: 424;

(ak) the VL region of SEQ ID NO: 436 and the VH region of SEQ ID NO: 435;

(al) the VL region of SEQ ID NO: 447 and the VH region of SEQ ID NO: 446;

(am) the VL region of SEQ ID NO: 458 and the VH region of SEQ ID NO: 457;

(an) the VL region of SEQ ID NO: 469 and the VH region of SEQ ID NO: 468;

(ao) the VL portion of SEQ ID NO: 480 and the VH portion of SEQ ID NO: 479;

(ap) the VL region of SEQ ID NO: 491 and the VH region of SEQ ID NO: 490;

(aq) the VL region of SEQ ID NO: 502 and the VH region of SEQ ID NO: 501;

(ar) the VL region of SEQ ID NO: 513 and the VH region of SEQ ID NO: 512; and (as) the VL portion of SEQ ID NO: 524 and the VH portion of SEQ ID NO: 523.

It is further contemplated that the first binding domain of the antibody construct of the present invention comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 41, 52, 63, 74, 85, 96, 107, 118, 129, 140 , 151, 162, 173, 184, 195, 206, 217, 228, 239, 250, 261, 272, 283, 294, 305, 316, 327, 338, 349, 360, 371, 382, 393, 404, 4 15 , 426, 437, 448, 459, 470, 481, 492, 503, 514, and 525 or having an amino acid sequence characterized by at least 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identity with the specified sequences.

The present invention further provides an antibody construct comprising or having an amino acid sequence (complete bispecific antibody constructs) selected from the group consisting of SEQ ID NO: 42, 43, 53, 54, 64, 65, 75, 76, 86, 87, 97, 98, 108, 109, 119, 120, 130, 131, 141, 142, 152, 153, 163, 164, 174, 175, 185, 186, 196, 197, 207, 208, 218, 219, 229, 230, 240, 241, 251, 252, 262, 263, 273, 274, 284, 285, 295, 296, 306, 307, 317, 318, 328, 329, 339, 340, 350, 351 , 361, 362, 372, 373, 383, 384, 394, 395, 405, 406, 416, 417, 427, 428, 438, 439, 449, 450, 460, 461, 471, 472, 482, 483, 493 , 494, 504, 505, 515, 516, 526 and 527, or having an amino acid sequence characterized by at least 90, 91, 92, 93, 94, 95, 96, 97, 98 or 99% identity with these sequences.

Covalent modifications of antibody-based constructs are also included within the scope of the present invention and are generally, but not in every case, carried out post-translationally. For example, several types of covalent modifications of the antibody construct are introduced into the molecule by reacting specific amino acid residues of the antibody construct with an organic derivatizing agent that is capable of reacting with selected side chains of the N- or C-terminal residues.

Cysteinyl residues most often react with α-haloacetates (and corresponding amines) such as chloroacetic acid or chloroacetamide to form carboxymethyl or carboxyamidomethyl derivatives. Cysteinyl residues are also derivatized by reaction with bromotrifluoroacetone, a-bromo-b-(5-imidozoyl)propionic acid, chloroacetyl phosphate, Nalkylmaleimides, 3-nitro-2-pyridyl disulfide, methyl 2-pyridyl disulfide, chloromercuric benzoate, 2-chloromercuric-4- nitrophenol or chloro-7-nitrobenzo-2-oxa-1,3-diazole.

Histidyl residues are derivatized by reaction with diethyl pyrocarbonate at pH 5.5-7.0, since this agent is relatively specific for the histidyl side chain. Para-bromophenacyl bromide is also used; the reaction is preferably carried out in 0.1 M sodium cacodylate at pH 6.0. Lysinyl and amino-terminal residues are reacted with succinic anhydrides or other carboxylic acids. Derivatization by these means has the effect of reversing the charge of the lysinyl residues. Other suitable reagents for derivatization of alpha-amino group-containing residues include imidoesters such as methyl picolinimidate; pyridoxal phosphate; pyridoxal, chloroborohydride; trinitrobenzenesulfonic acid; O-methylisourea;

- 35 046123

2,4-pentanedione and glyoxylate in a transaminase-catalyzed reaction.

Arginyl residues can be modified by reaction with one or more standard reagents, including phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin. Derivatization of arginine residues requires the reaction to be carried out under alkaline conditions due to the high pKa of the guanidine functional group. In addition, these reagents can react with lysine groups as well as the epsilon amino group of arginine.

Specific modification of tyrosyl residues can be carried out, with particular interest being the introduction of spectral labels into tyrosyl residues by reaction with aromatic diazonium compounds or with tetranitromethane. For the formation of O-acetyltyrosyl compounds and 3-nitro derivatives, N-acetylimidizole and tetranitromethane are most often used, respectively. Tyrosyl residues are iodinated using ¹²⁵ I or ¹³¹ I to obtain labeled proteins for use in radioimmunoassays, the chloramine T method described above being suitable.

Carboxyl side groups (aspartyl or glutamyl) are selectively modified by reaction with carbodiimides (R'-N=C=N--R'), where R and R' optionally represent different alkyl groups such as 1-cyclohexyl-3- (2-morpholinyl-4-ethyl)carbodiimide or 1-ethyl-3-(4azonia-4,4-dimethylphenyl)carbodiimide. In addition, aspartyl and glutamyl residues are converted to asparaginyl and glutaminyl residues by reaction with ammonium ions.

Derivatization with bifunctional agents is used to cross-link the antibody constructs of the present invention to a water-insoluble immobilization matrix or surface for use in various methods. Commonly used cross-linking agents include, for example, 1,1-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, Nhydroxysuccinimide esters, for example, 4-azidosalicylic acid esters, homobifunctional imido esters, including disuccinimidyl esters such as 3,3 'dithiobis(succinimidylpropionate) and bifunctional maleimides such as bis-N-maleimido-1,8octane. Derivatizers such as methyl 3-[(p-azidophenyl)dithio]propioimidate result in photoactivatable intermediates that are capable of cross-linking in the presence of light. Alternatively, reactive water-insoluble matrices, such as cyanogen bromide-activated carbohydrates, and reactive substrates described in US Pat. No. 3,969,287 are used to immobilize proteins; 3691016; 4195128; 4247642; 4229537 and 4330440, used for protein immobilization.

Glutaminyl and aspartyl residues are often deamidated to the corresponding glutamyl and aspartyl residues. Alternatively, these residues are deamidated under mildly acidic conditions. Any form of these residues is within the scope of the present invention.

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

Another type of covalent modification of antibody constructs included within the scope of the present invention involves altering the glycosylation profile of a protein. As is known in the art, glycosylation profiles can depend on both the sequence of the protein (eg, the presence or absence of specific amino acid glycosylation residues, discussed below) and the host cell or organism in which the protein is produced. Specific expression systems are discussed below.

Glycosylation of polypeptides is typically either N-linked or O-linked. Nlinked refers to the attachment of a carbohydrate moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine, where X is any amino acid except proline, are recognition sequences for the enzymatic attachment of a carbohydrate moiety to the asparagine side chain. Thus, the presence of any of these tripeptide sequences in a polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the addition of one of the sugars N-acetylgalactosamine, galactose or xylose, to a hydroxyamino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine can also be used.

The addition of glycosylation sites to an antibody construct is conveniently accomplished by altering the amino acid sequence so that it contains one or more of the tripeptide sequences described above (for N-linked glycosylation sites). The change can also be made by adding or replacing one or more serine or threonine residues in the original sequence (for O-linked glycosylation sites). Conveniently, the amino acid sequence of the antibody construct is preferably altered through changes at the DNA level, in particular by mutating the DNA encoding the polypeptide at preselected bases so as to create codons that

- 36 046123 will be translated into the necessary amino acids.

Another means of increasing the amount of carbohydrate components in an antibody construct is to chemically or enzymatically couple the glycosides to the protein. These procedures are preferred because they do not require protein production in a host cell that has glycosylation capabilities for N- and O-linked glycosylation. Depending on the type of conjugation used, the sugar(s) may be attached to (a) arginine and histidine, (b) free carboxyl groups, (c) free sulfhydryl groups such as cysteine groups, (d) free hydroxyl groups such as for example, serine, threonine or hydroxyproline groups, (e) aromatic residues, such as phenylalanine, tyrosine or tryptophan residues, or (f) a glutamine amide group. These methods are described in WO 87/05330 and in Aplin and Wriston, 1981, CRC Crit. Rev. Biochem., pp. 259-306.

Removal of carbohydrate components present in the original antibody construct can be accomplished chemically or enzymatically. Chemical deglycosylation requires exposing the protein to a trifluoromethanesulfonic acid compound or an equivalent compound. This treatment results in the cleavage of most or all of the sugars except the binding sugar (N-acetylglucosamine or N-acetylgalactosamine), while the polypeptide remains intact. Chemical deglycosylation is described in Hakimuddin et al., 1987, Arch. Biochem. Biophys. 259:52 and in Edge et al., 1981, Anal. Biochem. 118:131. Enzymatic cleavage of carbohydrate moieties into polypeptides can be achieved by using a number of endo- and exo-glycosidases, as described in Thotakura et al., 1987, Meth. Enzymol. 138:350. Glycosylation at potential glycosylation sites can be prevented by use of the compound tunicamycin, as described in Duskin et al., 1982, J. Biol. Chem. 257:3105. Tunicamycin blocks the formation of protein-N-glycosidic bonds.

Other modifications to the antibody-based design are also provided herein. For example, another type of covalent modification of an antibody construct involves linking the antibody construct to various non-protein polymers, including, without limitation, various polyols such as polyethylene glycol, polypropylene glycol, polyoxyalkylenes, or copolymers of polyethylene glycol and polypropylene glycol, as described in US Pat. No. 4,640,835; 4496689; 4301144; 4670417; 4791192 or 4179337. In addition, as is known in the art, amino acid substitutions can be made at various positions in the antibody construct, for example, to facilitate the addition of polymers such as PEG.

In some embodiments of the present invention, covalent modification of the antibody constructs of the present invention includes the addition of one or more tags. The labeling group can be attached to the antibody construct using spacer groups of varying lengths to reduce possible steric inconsistency. Various protein labeling methods are known in the art that can be used in the practice of the present invention. The term label or labeling group refers to any detectable label. In general, marks are divided into many classes depending on the analysis in which they are supposed to be identified; wherein the following examples include, without limitation, isotopic tracers, which may be radioactive or heavy isotopes, such as radioisotopes or radionuclides (for example, ³ H, ¹⁴ C, ¹⁵ N, ³⁵ S, ⁸⁹ Zr, ⁹⁰ Y, ⁹⁹ Tc, ^m In, ¹²⁵ 1, ¹³¹ I);

magnetic tags (eg magnetic particles);

redox-active fragments;

optical dyes (including without limitation chromophores, phosphors and fluorophores) such as fluorescent groups (eg, FITC, rhodamine, lanthanide phosphors), chemiluminescent groups and fluorophores, which can be either low molecular weight fluorescent agents or protein fluorescent agents;

enzymatic groups (eg horseradish peroxidase, β-galactosidase, luciferase, alkaline phosphatase);

biotinylated groups;

predefined polypeptide epitopes recognized by the secondary reporter (eg, paired leucine zipper sequences, secondary antibody binding sites, metal binding domains, epitope tags, etc.).

A fluorescent label refers to any molecule that can be detected by its inherent fluorescent properties. Suitable fluorescent labels include, but are not limited to, fluorescein, rhodamine, tetramethylrhodamine, eosin, erythrosine, coumarin, methylcoumarins, pyrene, malachite green, stilbene, Lucifer Yellow, Cascade Blue J, Texas Red, IAEDANS, EDANS, BODIPY FL, LC Red 640, Cy 5 , Cy 5.5, LC Red 705, Oregon green, Alexa-Fluor dyes (Alexa Fluor 350, Alexa Fluor 430, Alexa Fluor 488, Alexa Fluor 546, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor 633, Alexa Fluor 660, Alexa Fluor 680), Cascade Blue, Cascade Yellow and R-phycoerythrin (PE) (Molecular Probes, Eugene, OR), FITC, Rhodamine and Texas Red (Pierce, Rockford, IL), Cy5, Cy5.5, Cy7 (Amersham Life Science, Pittsburgh, Pennsylvania). Suitable optical dyes, including fluorophores, are described in

- 37 046123

Molecular Probes Handbook by Richard P. Haugland.

Suitable protein fluorescent tags also include, but are not limited to, green fluorescent protein, including GFP from Renilla, Ptilosarcus or Aequorea species (Chalfie et al., 1994, Science 263:802-805), EGFP (Clontech Laboratories, Inc., GenBank accession number U55762 ), blue fluorescent protein (BFP, Quantum Biotechnologies, Inc. 1801 de Maisonneuve Blvd. West, 8th Floor, Montreal, Quebec, Canada H3H 1J9; Stauber, 1998, Biotechniques 24:462-471; Heim et al., 1996, Curr . Biol. 6:178-182), enhanced yellow fluorescent protein (EYFP, Clontech Laboratories, Inc.), luciferase (Ichiki et al., 1993, J. Immunol. 150:5408-5417), p-galactosidase (Nolan et al. al., 1988, Proc. Natl. Acad. Sci. U.S.A 85:26032607) and Renilla (WO 92/15673, WO 95/07463, WO 98/14605, WO 98/26277, WO 99/49019, US patents No. 5292658 ; 5418155; 5683888; 5741668; 5777079; 5804387; 5874304; 5876995; 5925558).

The antibody construct of the present invention may also contain additional domains that, for example, facilitate release of the molecule or are associated with a tailored pharmacokinetic profile of the molecule. The domains that facilitate the release of the antibody construct can be selected from peptide motifs or secondarily introduced components that can be captured in the recovery process, for example, in a recovery column. Non-limiting embodiments of such additional domains include peptide motifs known as Myc tag, HAT tag, HA tag, TAP tag, GST tag, chitin binding domain (CBD tag), maltose binding domain (MBP tag) ), Flag tag, Strep tag and its variants (for example, StrepII tag) and His tag. All antibody constructs disclosed herein may contain a His tag domain, which is generally known as a repeat of consecutive His residues in the amino acid sequence of a molecule, preferably five or more, preferably six His (hexa-histidine) residues. The His tag may be located, for example, at the N- or C-terminus of the antibody construct, preferably it is located at the C-terminus. Most preferably, a hexa-histidine tag (HHHHHH) (SEQ ID NO: 16) is linked via a peptide bond to the C terminus of the antibody construct of the present invention. Additionally, a PLGA-PEG-PLGA conjugate system can be combined with a polyhistidine tag for sustained release applications and to improve the pharmacokinetic profile.

Modifications to the amino acid sequences of the antibody constructs described herein are also contemplated. For example, it may be necessary to improve the binding affinity and/or other biological properties of the antibody construct. Amino acid sequence variants of antibody constructs are obtained by introducing appropriate nucleotide substitutions into the nucleic acid of the antibody constructs or by peptide synthesis. All of the amino acid sequence modifications described below should result in an antibody construct that still retains the desired biological activity (MUC17 and CD3 binding) of the unmodified parent molecule.

The terms amino acid or amino acid residue generally refer to an amino acid as defined in the art, such as an amino acid selected from the group consisting of alanine (Ala or A); arginine (Arg or R); asparagine (Asn or N); aspartic acid (Asp or D); cysteine (Cys or C); glutamine (Gln or Q); glutamic acid (Glu or E); glycine (Gly or G); histidine (His or H); isoleucine (Ile or I); leucine (Leu or L); lysine (Lys or K); methionine (Met or M); phenylalanine (Phe or F); proline (Pro or P); serine (Ser or S); threonine (Thr or T); tryptophan (Trp or W); tyrosine (Tyr or Y) and valine (Val or V), although modified, synthetic or rare amino acids can be used if necessary. Generally, amino acids can be divided into groups having a non-polar side chain (eg, Ala, Cys, Ile, Leu, Met, Phe, Pro, Val); negatively charged side chain (eg Asp, Glu); a positively charged side chain (eg Arg, His, Lys) or an uncharged polar side chain (eg Asn, Cys, Gln, Gly, His, Met, Phe, Ser, Thr, Trp and Tyr).

Amino acid modifications include, for example, deletions and/or insertions and/or substitutions of residues within the amino acid sequences of antibody constructs. Any combination of deletion, insertion and substitution is performed to obtain the final construct, provided that the final construct has the required characteristics. Changes in amino acids can also alter the post-translational processes that antibody constructs undergo, such as changing the number or position of glycosylation sites.

For example, 1, 2, 3, 4, 5 or 6 amino acids can be inserted, replaced or deleted in each CDR (depending of course on their length), while 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 25 amino acids may be inserted, replaced, or deleted in each FR. Preferably, amino acid sequence insertions into an antibody construct include amino- and/or carboxyl-terminated fusions ranging in length from 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 residues to polypeptides containing one hundred or more residues, as well as insertions within the sequence of one or more amino acid residues. Appropriate modifications can also be made within the third domain of the antibody construct of the present invention. An insertional embodiment of the antibody construct of the present invention comprises the fusion

- 38 046123 enzyme with N-terminus or C-terminus of an antibody-based construct or fusion with a polypeptide.

Sites of greatest interest for substitution mutagenesis include, but are not limited to, the heavy and/or light chain CDRs, particularly the hypervariable regions, but FR changes in the heavy and/or light chains are also contemplated. Substitutions are preferably conservative substitutions described herein. Preferably, 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 amino acids may be substituted in the CDR, whereas 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 12, 13, 14, 15, 16, 17, 18, 19, 20, or 25 amino acids may be substituted in framework regions (FRs) depending on the length of the CDR or FR. For example, if the CDR sequence spans 6 amino acids, it is contemplated that one, two, or three of those amino acids may be replaced. Likewise, if the CDR sequence spans 15 amino acids, it is contemplated that one, two, three, four, five or six of those amino acids may be substituted.

A useful method for identifying specific residues or sites in antibody constructs that are preferred locations for mutagenesis is called alanine scanning mutagenesis, described by Cunningham and Wells in Science, 244: 1081-1085 (1989). Here, the target residue or group of residues within the antibody construct is identified (e.g., charged residues such as Arg, Asp, His, Lys and Glu) and replaced with neutral or negatively charged amino acids (most preferably alanine or polyalanine), influencing the interaction of amino acids with the epitope.

Those amino acid locations that exhibit functional sensitivity to substitution are then adjusted by introducing additional or different variants at or for the substitution sites. Thus, although the site for introducing the amino acid sequence variant is predetermined, the actual nature of the mutation does not need to be predetermined. For example, to analyze or optimize the characteristics of a mutation at a given site, alanine scanning or random mutagenesis can be performed at the target codon or site and the expressed variants of the antibody construct can be screened for the optimal combination of desired activity. Methods for performing mutations with substitutions in predetermined regions in DNA with a known sequence are well known, for example, mutagenesis with primer M13 and PCR mutagenesis. Screening for mutants is carried out using antigen binding activity assays such as MUC17 or CD3 binding.

In general, if amino acids are replaced in one or more or all of the CDRs of the heavy and/or light chains, it is preferred that the resulting sequence with substitutions be at least 60 or 65%, more preferably 70 or 75%, still more preferably 80 or 85% and especially preferably 90 or 95% identical to the original CDR sequence. This means that the degree of its identity with respect to the sequence with substitutions depends on the length of the CDR. For example, a CDR containing 5 amino acids is preferably 80% identical to its substituted sequence such that at least one amino acid has been replaced. Accordingly, antibody-based CDR constructs may have varying degrees of identity to their substituted sequences, for example CDRL1 may be 80% while CDRL3 may be 90%.

Preferred substitution(s) are conservative substitutions. However, any substitution (including a non-conservative substitution or one or more of the exemplary substitutions listed in Table 3 below) is contemplated so long as the antibody construct retains its ability to bind to MUC17 via the first domain and to SEZ-epsilon via a second domain and/or its CDRs having at least 60 or 65% identity with such substitution sequence, more preferably 70 or 75%, even more preferably 80 or 85%, and especially preferably 90 or 95% identical to the original CDR sequence).

Conservative substitutions are given in table. 3 under the heading preferred substitutions. If such substitutions lead to a change in biological activity, then more significant changes can be made, indicated in the table. 3 as illustrative substitutions or as further described below with respect to amino acid classes, and the products can be screened for the desired characteristic.

- 39 046123

Table 3

Amino acid substitutions

Parent amino acid Illustrative Substitutions Preferred Substitutions Ala (A) val, leu, ile Val Arg(R) lys, gin, asn Lys Asn(N) gin, his, asp, lys, arg Gin Asp (D) glu, asn Glu Cys(C) ser, ala ser Gin (Q) asn,glu asn Glu(E) asp, gin asp Gly (G) Ala ala His(H) asn, gin, lys, arg arg He(I) leu, val, met, ala, phe leu Leu (L) Norleucine, ile, val, met, ala ile Lys (K) arg, gin, asn arg Met(M) leu, phe, ile leu Phe (F) leu, val, ile, ala, tyr tyr Pro (P) Ala ala Ser (S) Thr thr Thr (T) Ser ser Trp(W) tyr, phe tyr Tyr (Y) trp, phe, thr, ser phe Val(V) ile, leu, met, phe, ala leu

Significant modifications to the biological properties of the antibody construct of the present invention are accomplished by selecting substitutions that differ significantly in their effect with respect to maintaining (a) the structure of the polypeptide backbone in the region of the substitution, for example, a sheet or helical conformation, (b) the charge or hydrophobicity of the molecule at the target site or (c) the volume of the side chain. Naturally occurring residues are classified into groups based on general side chain properties: (1) hydrophobic: norleucine, met, ala, val, leu, ile; (2) neutral hydrophilic: cys, ser, thr, asn, gln; (3) acidic: asp, glu; (4) basic: his, lys, arg; (5) residues that affect chain orientation: gly, pro; and (6) aromatic: trp, tyr, phe.

Non-conservative substitutions involve replacing a member of one of these classes with a member of another class. Any cysteine residue not involved in maintaining proper conformation of the antibody construct can be replaced, typically with serine, to improve the oxidative stability of the molecule and prevent abnormal cross-linking. In contrast, cysteine linkage(s) can be added to the antibody to improve its stability (especially if the antibody is an antibody fragment such as an Fv fragment).

For amino acid sequences, sequence identity and/or similarity is determined using standard techniques known in the art, including, without limitation, the local sequence identity algorithm of Smith and Waterman, 1981, Adv. Appl. Math. 2:482, sequence identity alignment algorithm after Needleman and Wunsch, 1970, J. Mol. Biol. 48:443, similarity search method according to Pearson and Lipman, 1988, Proc. Nat. Acad. Sci. U.S.A 85:2444, computerized implementations of these algorithms (GAP, BESTFIT, FASTA and TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group, 575 Science Drive, Madison, Wis.), the Best Fit sequence program described in Devereux et al., 1984, Nucl. Acid Res. 12:387-395, preferably using standard settings or by brute force. Preferably, the percent identity is calculated using FastDB based on the following parameters: mismatch penalty 1; penalty for introducing gap 1; penalty for gap length 0.33; and binding penalty 30, Current Methods in Sequence Comparison and Analysis, Macromolecule Sequencing and Synthesis, Selected Methods and Applications, pp 127-149 (1988), Alan R. Liss, Inc.

An example of an algorithm used is PILEUP. PILEUP generates a multiple sequence alignment from a group of related sequences using sequential pairwise alignments. It can also construct a tree showing the clustering relationships used to create the alignment. PILEUP uses a simplification of the sequential alignment method Feng & Doolittle, 1987, J. Mol. Evol. 35:351-360; the method is similar to that described in Higgins and Sharp, 1989, CABIOS 5:151-153. PILEUP parameters used include standard

- 40 046123 penalty for introducing a gap of 3.00, standard penalty for extending a gap of 0.10 and assessed end gaps.

Another example of an algorithm used is the BLAST algorithm described in Altschul et al., 1990, J. Mol. Biol. 215:403-410; Altschul et al., 1997, Nucleic Acids Res. 25:3389-3402; and Karin et al., 1993, Proc. Natl. Acad. Sci. U.S.A 90:5873-5787. A particularly useful BLAST program is the WU-BLAST-2 program created by Altschul et al., 1996, Methods in Enzymology 266:460-480. WU-BLAST-2 uses several search parameters, most of which are set to default values. The adjusted parameters are set with the following values: overlap length =1, overlap fraction =0.125, threshold word length (T)=II. The HSP S and HSP S2 parameters are dynamic values and are set by the program itself, depending on the composition of the specific sequence and the composition of the specific database in which the sequence of interest is searched; however, these values can be adjusted to increase sensitivity.

An additional algorithm used is BLAST with gap injection according to Altschul et al., 1993, Nucl. Acids Res. 25:3389–3402. BLAST with gap injection uses BLOSUM-62 replacement scores; threshold parameter T is set to 9; the two-hit method for initiating extensions without introducing gaps gives the gap extension k a value of 10+k; Xu is set to 16 and Xg is set to 40 for the database search stage and 67 for the final stage of the algorithms. Gap-introduced alignments are triggered using a score corresponding to approximately 22 bits.

Typically, the amino acid homology, similarity or identity between individual variant CDRs or VH/VL sequences is at least 60% with the sequences provided herein, and, most often, with preferably increasing homology or identity of at least 65 or 70%, more preferably at least 75 or 80%, even more preferably at least 85, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99 and almost 100%. Similarly, the percentage (%) of nucleic acid sequence identity relative to the nucleic acid sequence of the binding proteins identified herein is defined as the percentage of nucleotide residues in the candidate sequence that are identical to nucleotide residues in the coding sequence of the antibody construct. The specific method uses the BLASTN module from WU-BLAST-2 with the default parameters set, with overlap length and overlap fraction set to 1 and 0.125, respectively.

Typically, the homology, similarity or identity of nucleotide sequences between the nucleotide sequences encoding individual variant CDRs or VH/VL sequences and the nucleotide sequences provided herein is at least 60% and, more often, with preferably increasing homology or identity of up to at least 65, 70, 75, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99, and almost 100%. Thus, a CDR variant or VH/VL region variant is one that has certain homology, similarity or identity with the parent CDR/VH/VL of the present invention, and has a common biological function including, but not limited to, at least 60, 65 , 70, 75, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% specificity and/or parental CDR or VH/VL activity.

In one embodiment of the present invention, the percentage of human germline identity of the antibody constructs of the present invention is >70 or >75%, more preferably >80 or >85%, even more preferably >90%, and most preferably >91. >92, >93, >94, >95 or even >96%. Identity with human germline antibody gene products is believed to be an important factor in reducing the risk that therapeutic proteins will induce an immune response to the drug in the patient during treatment. Hwang and Foote (Immunogenicity of engineered antibodies; Methods 36 (2005) 3-10) demonstrate that reducing the non-human portions of drug antibody constructs results in a reduced risk of inducing anti-drug antibodies in patients during treatment. When comparing a sufficient number of clinically evaluated antibody drugs and corresponding immunogenicity data, there was a trend that humanization of the V regions of antibodies rendered the proteins less immunogenic (on average in 5.1% of patients) than antibodies bearing intact non-human V regions. areas (on average for 23.59% of patients). Therefore, a high degree of identity to human sequences is required for V region protein therapeutics in the form of antibody constructs. To determine germline identity, VL V regions can be aligned to human germline V and J segment amino acid sequences (http://vbase.mrc-cpe.cam.ac.uk/) using Vector NTI software and calculate a percentage for the amino acid sequence by dividing identical amino acid residues by the total number of VL amino acid residues. The same can be done for VH segments (http://vbase.mrc-cpe.cam.ac.uk/) except that CDR3 VH

- 41 046123 may be excluded due to its high degree of diversity and the current lack of alignment partners among the human germline VH CDR3s. Recombinant techniques can then be used to increase sequence identity to human germline antibody genes.

In a further embodiment, the bispecific antibody constructs of the present invention exhibit high monomer yields under standard research scale conditions, for example, in a standard two-step purification method. Preferably, the monomer yield of the antibody constructs of the present invention is >0.25 mg/L supernatant, more preferably >0.5 mg/L, even more preferably >1 mg/L, and most preferably >3 mg/L supernatant.

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

In one embodiment, the antibody constructs preferably have a stability in blood plasma (ratio of EC ₅₀ with blood plasma to EC ₅₀ without blood plasma) of <5 or <4, more preferably <3.5 or <3, even more preferably < 2.5 or <2 and most preferably <1.5 or <1. Plasma stability of the antibody construct can be tested by incubating the construct in human plasma at 37°C for 24 hours followed by determination of EC ₅₀ in a ^51Cr chromium release cytotoxicity assay. The effector cells in the cytotoxicity assay may be stimulated enriched human CD8-positive T cells. The target cells can be, for example, CHO cells transfected with human MUC17. The ratio of effector cells to target cells (E:T) can be selected as 10:1 or 5:1. The pool of human blood plasma used for these purposes is obtained from the blood of healthy donors collected in EDTA-coated syringes. Cellular components are removed by centrifugation, and the upper phase in the form of blood plasma is collected and then pooled. As a control, antibody constructs were diluted immediately prior to cytotoxicity assay in RPMI-1640 medium. Stability in blood plasma is calculated as the ratio of _EC50 (after incubation of blood plasma) and EC50 (control).

It is also preferred that the conversion of monomers to dimers of antibody-based constructs be low. The degree of conversion can be determined under various conditions and analyzed using high performance exclusion chromatography. For example, incubation of monomeric isoforms of antibody-based constructs can be carried out for 7 days at 37°C and concentrations of, for example, 100 μg/ml or 250 μg/ml in an incubator. Under such conditions, it is preferred that the antibody constructs of the present invention exhibit a dimer percentage of <5%, more preferably <4%, even more preferably <3%, even more preferably <2.5%, even more preferably <2%, even more preferably <1.5% and most preferably <1% or <0.5% or even 0%.

It is also preferred that the bispecific antibody constructs of the present invention exhibit very low levels of conversion to dimers after a number of freeze/thaw cycles. For example, the antibody construct monomer is adjusted to a concentration of 250 μg/mL, for example, in generic formulation buffer and subjected to three freeze/thaw cycles (freezing at -80°C for 30 min followed by thawing for 30 min at room temperature ) followed by high-throughput SEC to determine the percentage of the original monomeric antibody construct that was converted to a dimeric antibody construct. Preferably, the percentage of dimers of bispecific antibody constructs is <5%, more preferably <4%, even more preferably <3%, even more preferably <2.5%, even more preferably <2%, even more preferably <1.5 % and most preferably <1% or even <0.5%, for example after three freeze/thaw cycles.

The bispecific antibody constructs of the present invention preferably exhibit good thermal stability with an aggregation temperature of >45 or >50°C, more preferably >52 or >54°C, even more preferably >56 or >57°C, and most preferably >58 or > 59°C. The thermal stability parameter can be defined in the context of the antibody aggregation temperature as follows: A 250 μg/mL antibody solution is transferred to a disposable cuvette and placed in a dynamic light scattering (DLS) device. The sample is heated from 40 to 70°C at a heating rate of 0.5°C/min with constant measurement of the determined radius. The increase in radius, indicating melting and aggregation of the protein, is used to calculate the aggregation temperature of the antibody.

Alternatively, temperature melting curves can be determined by differential scanning calorimetry (DSC) to determine the characteristic biophysical protein stability of antibody constructs. These experiments are performed using a MicroCal LLC (Northampton, MA, USA) VP-DSC device. The energy absorption of the sample containing the antibody construct is recorded over the range from 20 to 90°C and compared with the sample containing only the formulation buffer. Antibody constructs were adjusted to a final concentration of 250 μg/ml, for example, in SEC running buffer. To obtain the appropriate melting curve, the overall temperature of the sample is increased step by step. At each temperature T, the energy absorption of the sample and the reference in the form of a composition buffer is recorded. The difference in energy absorption Cp (kkyal/mol/°C) of the sample minus the standard is plotted as a function of the corresponding temperature. The melting point is defined as the temperature at the first maximum of energy absorption.

The MUC17xCD3 bispecific antibody constructs of the present invention are also expected to have a defined turbidity (measured at OD340 after concentrating the purified monomeric antibody construct to 2.5 mg/ml and incubating overnight) of <0.2, preferably <0.15, more preferably <0.12, even more preferably <0.1, and most preferably <0.08.

In a further embodiment, the antibody construct of the present invention is stable at physiological or slightly lower pH, i.e. pH from about 7.4 to 6.0. The more robust the antibody construct is at a non-physiological pH, such as pH 6.0, the greater the recovery of the antibody construct eluting from the ion exchange column relative to the total amount of protein loaded. Recovery of the antibody construct from the ion exchange (eg, cationic) column at approximately pH 6.0 is preferably >30%, more preferably >40%, more preferably >50%, even more preferably >60%, even more preferably >70 %, even more preferably >80%, even more preferably >90%, even more preferably >95% and most preferably >99%.

It is further intended that the bispecific antibody constructs of the present invention exhibit therapeutic efficacy or antitumor activity. This can be assessed, for example, in a study as described in the following generalized example of a late-stage human tumor xenograft model.

On day 1 of the study, 5x106 cells from a cancer cell line positive for the human target cellular antigen (in this case MUC17) were injected subcutaneously into the right posterior flank of female NOD/SCID mice. When the average tumor volume reached approximately 100 mm ³ , mice were transplanted with in vitro expanded CD3-positive human T cells by injecting approximately 2x107 cells into the peritoneal cavity of the animals. Mice from vehicle control group 1 did not receive effector cells and were used as a non-transplanted control for comparison with vehicle control group 2 (receiving effector cells) to monitor the effect of T cells individually on tumor growth . Antibody treatment began when the average tumor volume reached approximately 200 mm ³ . The mean tumor size in each treatment group on the day treatment began should not be statistically different from the volume in any other group (analysis of variance). Mice are treated with 0.5 mg/kg/day of the MUC17xCD3 bispecific antibody construct by intravenous bolus injection for approximately 15 to 20 days. Tumors are measured with calipers throughout the study, and progression is assessed by intergroup comparison of tumor volumes (TV). Tumor growth inhibition T/C [%] is determined by calculating TV as T/C%=100x (median TV in the analyzed group)/(median TV in control group 2).

One skilled in the art will know how to modify or adapt certain parameters of this assay, such as the number of tumor cells injected, the injection site, the number of human T cells transplanted, the number of bispecific antibody constructs to be administered, and the schedule of administration, while obtaining a significant and reproducible results. Preferably, the tumor growth inhibition T/C [%] is <70 or <60, more preferably <50 or <40, even more preferably <30 or <20, and most preferably <10 or <5 or even <2.5. Tumor growth inhibition is preferably about 100%.

In a preferred embodiment of the antibody construct of the present invention, the antibody construct is a single chain antibody construct.

Also in a preferred embodiment of the antibody construct of the present invention, said third domain comprises an amino-terminal to carboxyl-terminal direction

- 43 046123 end hinge section-CH2-CH3-linker-hinge section-CH2-CH3.

In one embodiment of the present invention, each of these third domain polypeptide monomers has an amino acid sequence that is at least 90% identical to a sequence selected from the group consisting of SEQ ID NOs: 17-24. In a preferred embodiment of the present invention, each of said polypeptide monomers has an amino acid sequence selected from SEQ ID NOs: 17-24.

Also in one embodiment of the present invention, the CH2 domain of one or preferably each (both) third domain polypeptide monomers contains an intradomain cysteine disulfide bridge. As is known in the art, the term cysteine disulfide bridge refers to a functional group with the general structure R-S-S-R. This bond is also called an SS bond or a disulfide bridge and is formed by joining two thiol groups of cysteine residues. For the antibody construct of the present invention, it is particularly preferred that the cysteines forming the cysteine disulfide bridge in the mature antibody construct are introduced into the amino acid sequence of the CH2 domain corresponding to positions 309 and 321 (Kabat numbering).

In one embodiment of the present invention, the glycosylation site at position 314 of the Kabat CH2 domain is removed. Preferably, this removal of the glycosylation site is achieved by the substitution N314X, where X is any amino acid excluding Q. The substitution is preferably N314G. In a more preferred embodiment, said CH2 domain further contains the following substitutions (positions according to Kabat): V321C and R309C (these substitutions introduce an intradomain cysteine disulfide bridge at positions 309 and 321 according to Kabat).

It is believed that the advantageous characteristics of the antibody construct of the present invention compared to, for example, a prior art hetero-Fc bispecific antibody construct (Figure 1b) may be due to, among other things, the introduction of the above-described modifications to the CH2 domain. Thus, for the construct of the present invention, it is preferred that the CH2 domains in the third domain of the antibody construct of the present invention contain an intradomain cysteine disulfide bridge at Kabat positions 309 and 321, and/or the glycosylation site at Kabat position 314 is deleted, preferably by replacing N314G.

In a further preferred embodiment of the present invention, the CH2 domains in the third domain of the antibody construct of the present invention contain an intradomain cysteine disulfide bridge at Kabat positions 309 and 321, and the glycosylation site at Kabat position 314 is removed by substitution N314G. Most preferably, the polypeptide monomer in the third domain of the antibody construct of the present invention has an amino acid sequence selected from the group consisting of SEQ ID NOs: 17 and 18.

In one embodiment of the present invention, an antibody construct is provided wherein:

(i) the first domain contains two antibody variable domains, and the second domain contains two antibody variable domains;

(ii) the first domain contains one antibody variable domain, and the second domain contains two antibody variable domains;

(iii) the first domain contains two antibody variable domains, and the second domain contains one antibody variable domain; or (iv) the first domain contains one antibody variable domain and the second domain contains one antibody variable domain.

Accordingly, the first and second domains may be binding domains, each of which contains two antibody variable domains, such as VH and VL domains. Examples of such binding domains containing two antibody variable domains have been described above herein and include, for example, the Fv fragments, scFv fragments or Fab fragments described above herein. Alternatively, one or both of these binding domains may contain only one variable domain. Examples of such single domain binding domains are those described above and contain, for example, single variable domain nanobodies or antibodies containing only one variable domain, which may be VHH, VH or VL, which specifically bind an antigen or epitope independently of others V regions or domains.

In a preferred embodiment, the antibody constructs of the present invention are fused to the third domain via a peptide linker. A preferred peptide linker is described herein above and is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e. Gly ₄ Ser (SEQ ID NO: 1), or polymers thereof, i.e. (Gly=Ser)x, where x is an integer 1 or greater (for example, 2 or 3). A particularly preferred linker for fusing the first and second domains to the third domain is given in SEQ

- 44 046123

ID NO: 1.

In a preferred embodiment, the antibody construct of the present invention is characterized in that it contains, from the amino terminus to the carboxyl terminus:

(a) first domain;

(b) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 1-3;

(c) second domain;

(d) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 1, 2, 3, 9, 10, 11 and 12;

(e) the first polypeptide monomer of the third domain;

(f) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 5, 6, 7 and 8; and (g) a second polypeptide monomer of the third domain.

The antibody construct of the present invention contains a first domain that binds to MUC17, preferably the extracellular domain (ECD) of MUC17. It is known that the term binding to the extracellular domain of MUC17 in the context of the present invention means that the binding domain binds to MUC17 expressed on the surface of the target cell. Therefore, the first domain of the present invention preferentially binds to MUC17 when expressed by naturally expressing cells or cell lines and/or cells or cell lines transformed or (stably/transiently) transfected with MUC17. In a preferred embodiment, the first binding domain also binds to MUC17 when MUC17 is used as a target molecule or ligand in an in vitro binding assay such as BIAcore or Scatchard assay. The target cell can be any prokaryotic or eukaryotic cell expressing MUC17 on its surface; preferably the target cell is a cell that is part of a human or animal body, such as specific MUC17-expressing cancer or tumor cells.

Preferably, the first binding domain binds to human MUC17/ECD MUC17. In a further preferred embodiment, it binds to macaque MUC17/ECD MUC17. In a most preferred embodiment, it binds to both human and macaque MUC17/ECD MUC17. The MUC17 extracellular domain or MUC17 ECD refers to a region or sequence of MUC17 that is substantially free of the transmembrane and cytoplasmic domains of MUC17. One skilled in the art will appreciate that the transmembrane domain defined for the MUC17 polypeptide of the present invention is determined based on criteria commonly used in the art to define a hydrophobic domain of this type. The exact boundaries of the transmembrane domain may vary, but are likely to be within no more than about 5 amino acids at each end of the domain specifically referred to herein.

Preferred binding domains that bind to MUC17 are described in WO 2010/037836 and WO 2011/121110. Any binding domain for MUC17 described in these applications can be used in the context of the present invention.

In one aspect of the present invention, the antibody construct comprises, from the amino terminus to the carboxyl terminus:

(a) a first domain having an amino acid sequence selected from the group consisting of SEQ ID NO: 50, 56, 68, 74, 86, 92, 104, 110, 122, 128, 140, 146, 158, 164, 176, 182, 194, 200, 212, 218, 230, 236, 248, 254, 266, 272, 284, 290, 302, 308, 320, 335, 350, 365, 380, 395, 410, 425, 440, 455 , 470;

(b) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 1-3;

(c) a second domain having an amino acid sequence selected from the group consisting of SEQ ID NO: 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185 or 187 in W0 2008/119567, or set forth under SEQ ID NO: 15;

(d) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 1, 2, 3, 9, 10, 11 and 12;

(e) a first third domain polypeptide monomer having a polypeptide sequence selected from the group consisting of SEQ ID NOs: 17-24;

(f) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 5, 6, 7 and 8; and (g) a second third domain polypeptide monomer having a polypeptide sequence selected from the group consisting of SEQ ID NOs: 17-24.

According to this preferred embodiment, the first and second domains, which are fused via a peptide linker to a single chain polypeptide, comprise a sequence selected from the group consisting of SEQ ID NO: 51, 57, 69, 75, 87, 93, 105, 111, 123, 129, 141, 147, 159, 165, 177, 183, 195, 201, 213, 219, 231, 237, 249, 255, 267, 273, 285, 291, 303, 309, 321, 324, 336 ,

- 45 046123

339, 351, 354, 366, 369, 381, 384, 396, 399, 411, 414, 426, 429, 441, 444, 456, 459, 471 and 474.

In one aspect, the antibody construct of the present invention is characterized in that it has an amino acid sequence selected from the group consisting of SEQ ID NO: 52, 53,

58, 59, 70, 71, 76, 77, 88, 89, 94, 95, 106, 107, 112, 113, 124, 125, 130, 131, 142, 143, 148, 149, 160, 161, 166,

167, 178, 179, 184, 185, 196, 197, 202, 203, 214, 215, 220, 221, 232, 233, 238, 239, 250, 251, 256, 257, 268,

269, 274, 275, 286, 287, 292, 293, 304 305, 310, 311, 322, 323, 325, 326, 337, 338, 340, 341, 352, 353, 355,

356, 367, 368, 370, 371, 382, 383, 385, 386, 397, 398, 400, 401, 412, 413, 415, 416, 427, 428, 430, 431, 442,

443, 445, 446, 457, 458, 460, 461, 472, 473, 475 and 476.

The present invention further provides a polynucleotide/nucleic acid molecule encoding an antibody construct of the present invention. A polynucleotide is a biopolymer consisting of 13 or more nucleotide monomers covalently linked in a chain. DNA (such as cDNA) and RNA (such as mRNA) are examples of polynucleotides with a specific biological function. Nucleotides are organic molecules that serve as monomers or subunits of nucleic acid molecules such as DNA or RNA. A nucleic acid molecule or polynucleotide can be double-stranded or single-stranded, linear or circular. Preferably it is contained in a vector, which is preferably located in a host cell. Said host cell, for example, after transformation or transfection with a vector or polynucleotide of the present invention, is capable of expressing an antibody construct. For this purpose, the polynucleotide or nucleic acid molecule is operably linked to regulatory sequences.

The genetic code is a set of rules by which information encoded in genetic material (nucleic acids) is translated into protein. Biological transcription in living cells is carried out by the ribosome, which binds amino acids in the order specified by mRNA, using tRNA molecules to carry amino acids and to read trinucleotides of mRNA at a time. The code specifies how the sequences of these nucleotide triplets, called codons, determine which amino acid should be added next during protein synthesis. With some exceptions, a trinucleotide codon in a nucleic acid sequence specifies one amino acid. Since the vast majority of genes are encoded by exactly the same code, this particular code is often called the canonical or standard genetic code. Although the genetic code determines the protein sequence for a given coding region, other genomic regions can influence when and where these proteins are produced.

In addition, the present invention provides a vector containing a polynucleotide/nucleic acid molecule of the present invention. A vector is a nucleic acid molecule used as a carrier medium to transfer (foreign) genetic material into a cell. The term vector includes, without limitation, plasmids, viruses, cosmids and artificial chromosomes. Typically, constructed vectors contain an origin of replication, a multicloning site, and a selectable marker. The vector itself is typically a nucleotide sequence, usually a DNA sequence, containing an insert (transgene) and a larger sequence that serves as the backbone of the vector. Modern vectors may include additional elements beyond the transgene insert and backbone: promoter, genetic marker, antibiotic resistance, reporter gene, targeting sequence, protein purification tag. Vectors, called expression vectors (expression constructs), are specifically designed to express a transgene in a target cell and usually contain regulatory sequences.

The term regulatory sequences refers to DNA sequences necessary for the expression of an operably linked coding sequence in a particular host organism. Regulatory sequences suitable for prokaryotes, for example, include a promoter, optionally an operator sequence and a ribosome binding site. Promoters, polyadenylation signals, and enhancers are known to be used in eukaryotic cells.

A nucleic acid is operably linked if it is placed in a functional relationship with another nucleic acid sequence. For example, DNA encoding a presequence or leader sequence for secretion is operably linked to DNA encoding a polypeptide if it is expressed as a precursor protein involved in secretion of the polypeptide; a promoter or enhancer is operably linked to a coding sequence if it affects transcription of the sequence; or the ribosome binding site is operably linked to the coding sequence if it is located in such a way as to facilitate translation. In general, operably linked means that the linked DNA sequences are contiguous, and in the case of a secretion leader sequence, contiguous and in reading phase. In this case, enhancers should not be adjacent. Binding is accomplished by ligation at convenient restriction sites. If such regions do not exist, synthetic adapters or linkers are used in accordance with traditional practice.

Transfection is the process of deliberate introduction of nucleic acid molecules or polynucleotides (including vectors) into target cells. This term is used primarily for non-viral techniques in eukaryotic cells. Transduction is often used to describe the virus-mediated transfer of nucleic acid molecules or polynucleotides. Transfection of animal cells typically involves opening temporary pores or holes in the cell membrane to allow uptake of material. Transfection can be done using calcium phosphate, by electroporation, by squeezing cells, or by mixing a cationic lipid with the material to produce liposomes, which fuse with the cell membrane and deposit their cargo inside.

The term transformation is used to describe the non-viral transfer of nucleic acid molecules or polynucleotides (including vectors) into bacteria and also into non-animal eukaryotic cells, including plant cells. Transformation, therefore, is a genetic change in a bacterial or non-animal eukaryotic cell resulting from direct absorption through the cellular membrane(s) from the environment and the subsequent incorporation of exogenous genetic material (nucleic acid molecules). Transformation can be carried out using artificial means. For transformation to occur, the cells or bacteria must be in a state of competence, which can occur as a time-limited response to external conditions such as starvation and cell density.

In addition, the present invention provides a host cell transformed or transfected with a polynucleotide/nucleic acid molecule or vector of the present invention. As used herein, the terms host cell or recipient cell are intended to include any single cell or culture of cells that can be or has been the recipient of vectors, exogenous nucleic acid molecules and polynucleotides encoding the antibody construct of the present invention; and/or recipients of the antibody construct itself. The introduction of the corresponding material into the cell is carried out by transformation, transfection, etc. The term host cell is also intended to include the progeny or potential progeny of a single cell. Since certain modifications may occur in subsequent generations due to natural, accidental or intentional mutation or due to environmental influences, such offspring may not actually be completely identical (in morphology or in genomic or total DNA) to the parent cell, but continue to be included within the scope of the term used in this document. Suitable host cells include prokaryotic or eukaryotic cells and also include, but are not limited to, bacteria, yeast cells, fungal cells, plant cells and animal cells, such as insect cells and mammalian cells, such as mouse, rat, macaque or human.

The antibody construct of the present invention can be produced in bacteria. Once expressed, the antibody construct of the present invention is isolated from the E. coli cell mass in a soluble fraction and can be purified by, for example, affinity and/or size exclusion chromatography. The final purification can be carried out in a manner similar to that for purifying an antibody expressed, for example, in CHO cells.

In addition to prokaryotes, eukaryotic microorganisms such as filamentous fungi or yeast are suitable hosts for cloning or expression of the antibody construct of the present invention. Saccharomyces cerevisiae, or common baker's yeast, is the most commonly used lower eukaryotic host microorganism. However, many other genera, species and strains are widely available and applicable herein, such as Schizosaccharomyces pombe, Kluyveromyces hosts such as K. lactis, K. fragilis (ATCC 12424), K. bulgaricus (ATcC 16045), K. wickeramii (ATCC 24178), K. waltii (ATCC 56500), K. drosophilarum (ATCC 36906), K. thermotolerans and K. marxianus; yarrowia (EP 402226); Pichia pastoris (EP 183070); Candida; Trichoderma reesia (EP 244234); Neurospora crassa; Schwanniomyces such as Schwanniomyces occidentalis; and filamentous fungi such as Neurospora, Penicillium, Tolypocladium, and Aspergillus hosts such as A. nidulans and A. niger.

Suitable host cells for expression of the glycosylated antibody construct of the present invention are derived from multicellular organisms. Examples of invertebrate cells include plant and insect cells. A variety of baculovirus strains and variants and corresponding permissive host cells have been identified from insect hosts such as Spodoptera frugiperda (caterpillar), Aedes aegypti (mosquito), Aedes albopictus (mosquito), Drosophila melanogaster (fruit fly) and Bombyx mori. Various strains of viruses are publicly available for transfection, such as Autographa californica NPV variant L-1 and Bombyx mori NPV strain Bm-5, and such viruses can be used herein as the virus of the present invention, particularly for transfecting Spodoptera frugiperda cells.

Plant cell cultures from cotton, corn, potato, soybean, petunia, tomato, Arabidopsis and tobacco can also be used as hosts. Cloning and expression vectors useful for producing proteins in plant cell culture are known to those skilled in the art.

- 47 046123 equipment. See, for example, Hiatt et al., Nature (1989) 342: 76-78, Owen et al. (1992) Bio/Technology 10: 790794, Artsaenko et al. (1995) The Plant J 8: 745-750, and Fecker et al. (1996) Plant Mol Biol 32: 979–986.

However, vertebrate cells were of greatest interest, and propagation of vertebrate cells in culture (tissue culture) became a routine procedure. Examples of mammalian host cell lines used include the monkey kidney cell line CV1 transformed with SV40 (C0S-7, ATCC CRL 1651); human embryonic kidney cell line (293 or 293 cells subcloned for growth in suspension culture, Graham et al., J. Gen Virol. 36: 59 (1977)); newborn hamster kidney cells (BHK, ATCC CCL 10); Chinese hamster ovary cells/-DHFR (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA 77: 4216 (1980)); mouse Sertoli cells (TM4, Mather, Biol. Reprod. 23: 243-251 (1980)); monkey kidney cells (CVI, ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL1587); human cervical carcinoma cells (HELA, ATCC CCL 2); dog kidney cells (MDCK, ATCC CCL 34); Buffalo rat liver cells (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human liver cells (Hep G2.1413 8065); mouse mammary tumor cells (MMT 060562, ATCC CCL5 1); TRI cells (Mather et al., Annals N. Y Acad. Sci. (1982) 383: 44-68); MRC 5 cells; FS4 cells; and the human hepatoma line (Hep G2).

In a further embodiment, the present invention provides a method for producing an antibody construct of the present invention, the method comprising culturing a host cell of the present invention under conditions allowing expression of the antibody construct of the present invention, and isolating the resulting antibody construct from culture.

As used herein, the term culture refers to the maintenance, differentiation, growth, proliferation and/or propagation of cells in vitro under suitable conditions in a culture medium. The term expression includes any step involved in the production of an antibody construct of the present invention, including, without limitation, transcription, post-transcriptional modification, translation, post-translational modification and secretion.

When using recombinant technologies, the antibody-based construct can be produced inside the cell, in the periplasmic space, or it can be secreted directly into the medium. If the antibody construct is produced intracellularly, in a first step, particulate debris, either host cells or lysed fragments, is removed, for example by centrifugation or ultrafiltration. Carter et al., Bio/Technology 10: 163-167 (1992) describes a procedure for isolating antibodies secreted into the periplasmic space of E. coli. Briefly, the cell mass is thawed in the presence of sodium acetate (pH 3.5), EDTA and phenylmethylsulfonyl fluoride (PMSF) for approximately 30 min. Cellular debris can be removed by centrifugation. If the antibody is secreted into the medium, supernatant samples from such expression systems are typically first concentrated using a commercially available protein concentration filter, such as an Amicon or Millipore Pellicon ultrafiltration unit. At any of the previous steps, a protease inhibitor such as PMSF may be included to inhibit proteolysis, and antibiotics may be included to prevent the growth of inadvertent contaminants.

The antibody construct of the present invention obtained from host cells can be isolated or purified using, for example, hydroxylapatite chromatography, gel electrophoresis, dialysis and affinity chromatography. Other protein purification techniques are also available, such as ion exchange column fractionation, ethanol precipitation, reverse phase HPLC, silica gel chromatography, heparin-Sepharose chromatography (SEPHAROSE™), anion exchange or cation exchange resin chromatography (such as a polyaspartic acid column). acid), chromatofocusing, SDS-PAGE and ammonium sulfate precipitation depending on the antibody to be recovered. If the antibody construct of the present invention contains a CH3 domain, Bakerbond ABX resin (J.T. Baker, Phillipsburg, NJ) is used for purification.

Affinity chromatography is the preferred purification technique. The matrix to which the affinity ligand is attached is most often agarose, but other matrices are also available. Mechanically stable matrices such as controlled pore glass or poly(styrene divinyl)benzene provide higher flow rates and shorter processing times compared to agarose.

In addition, the present invention provides a pharmaceutical composition containing an antibody construct of the present invention or an antibody construct prepared in accordance with the method of the present invention. In the case of the pharmaceutical composition of the present invention, it is preferable that the homogeneity of the antibody construct is >80%, more preferably >81, >82, >83, >84 or >85%, even more preferably >86, >87, >88, > 89 or >90%, even more preferably >91, >92, >93, >94 or >95% and most preferably >96, >97, >98 or >99%.

- 48 046123

As used herein, the term pharmaceutical composition refers to a composition for administration to a patient, preferably a human patient. In particular, a preferred pharmaceutical composition of the present invention contains one or a plurality of antibody constructs of the present invention, preferably in a therapeutically effective amount. Preferably, the pharmaceutical composition further comprises suitable formulations of one or more (pharmaceutically effective) carriers, stabilizers, excipients, diluents, solubilizers, surfactants, emulsifiers, preservatives and/or adjuvants. Suitable composition constituents are preferably non-toxic to recipients at the dosages and concentrations used. The pharmaceutical compositions of the present invention include, but are not limited to, liquid, frozen and lyophilized compositions.

The compositions of the present invention may contain a pharmaceutically acceptable carrier. In general, as used herein, a pharmaceutically acceptable carrier means any and all aqueous and non-aqueous solutions, sterile solutions, solvents, buffers, such as phosphate buffered saline (PBS), water, suspensions, emulsions, such as oil/oil emulsions. water, various types of wetting agents, liposomes, dispersion media and coatings that are compatible with pharmaceutical administration, in particular parenteral administration. The use of such media and agents in pharmaceutical compositions is well known in the art, and compositions containing such carriers can be formulated using well known conventional methods.

In certain embodiments, pharmaceutical compositions are provided comprising an antibody construct of the present invention and additionally one or more excipients, such as those described for illustrative purposes herein and elsewhere herein. In this regard, the present invention may employ excipients for a wide range of purposes, such as adjusting the physical, chemical or biological properties of the formulations, such as viscosity control, and/or the methods of the present invention to enhance the effectiveness and/or stabilization of such formulations, and methods against degradation and deterioration, for example, due to stresses encountered during manufacturing, shipping, storage, pre-use, administration and post-use preparation.

In some embodiments, the pharmaceutical composition may contain formulation materials designed to alter, maintain, or maintain, for example, pH, osmolarity, viscosity, clarity, color, isotonicity, odor, sterility, stability, dissolution or release rate, adsorption, or permeation of the composition ( see, REMINGTON'S PHARMACEUTICAL SCIENCES, 18 Edition, (A.R. Genrmo, ed.), 1990, Mack Publishing Company). In such embodiments, suitable formulation materials may include, but are not limited to, amino acids such as glycine, alanine, glutamine, asparagine, threonine, proline, 2-phenylalanine, including charged amino acids, preferably lysine, lysine acetate, arginine, glutamate and/or histidine;

antimicrobial agents such as antibacterial and antifungal agents;

antioxidants such as ascorbic acid, methionine, sodium sulfite or sodium hydrogensulfite;

buffers, buffer systems and buffer agents that are used to maintain the composition at a physiological pH value or a slightly lower pH value, preferably a pH value less than 4.0-6.5; examples of buffers are borate, bicarbonate, Tris-HCl, citrates, phosphates or other organic acids, succinate, phosphate and histidine; for example, Tris buffer at a pH of approximately 7.0-8.5;

non-aqueous solvents such as propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate;

aqueous vehicles, including water, alcohol/aqueous vehicles, emulsions or suspensions, including saline and buffered media;

biodegradable polymers such as polyesters;

bulking agents such as mannitol or glycine;

chelating agents such as ethylenediaminetetraacetic acid (EDTA);

isotonic agents and agents that slow down absorption;

complexing agents such as caffeine, polyvinylpyrrolidone, beta-cyclodextrin or hydroxypropyl-beta-cyclodextrin;

fillers;

monosaccharides; disaccharides; and other carbohydrates (such as glucose, mannose or dextrins); wherein the carbohydrates may be non-reducing sugars, preferably trehalose, sucrose, octosulfate, sorbitol or xylitol;

(low molecular weight) proteins, polypeptides or protein carriers such as human or bovine serum albumin, gelatin or immunoglobulins, preferably of human origin;

dyes and flavors;

sulfur-containing reducing agents such as glutathione, thioctic acid, sodium thioglycolate, thioglycerol, [alpha]-monothioglycerol and sodium thiosulfate;

thinners;

emulsifiers;

hydrophilic polymers such as polyvinylpyrrolidone;

salt-forming counterions such as sodium;

preservatives such as antimicrobial agents, antioxidants, chelating agents, inert gases and the like; examples are benzalkonium chloride, benzoic acid, salicylic acid, thimerosal, phenethyl alcohol, methylparaben, propylparaben, chlorhexidine, sorbic acid or hydrogen peroxide);

metal complexes such as Zn-protein complexes;

solvents and co-solvents (such as glycerin, propylene glycol or polyethylene glycol);

sugars and sugar alcohols such as trehalose, sucrose, octosulfate, mannitol, sorbitol or xylitol, stachyose, mannose, sorbose, xylose, ribose, myoinositose, galactose, lactitol, ribitol, myoinositol, galactitol, glycerol, cyclitols (e.g. inositol), polyethylene glycol ; and polyhydric sugar alcohols;

suspending agents;

surfactants or wetting agents such as pluronics, PEG, sorbitol esters, polysorbates such as polysorbate 20, polysorbate, Triton, tromethamine, lecithin, cholesterol, tyloxapal; wherein the surfactants may be detergents with a preferred molecular weight of >1.2 kDa and/or a polyether with a preferred molecular weight of >3 kDa; non-limiting examples of preferred detergents are Tween 20, Tween 40, Tween 60, Tween 80 and Tween 85; non-limiting examples for preferred polyethers are PEG 3000, PEG 3350, PEG 4000 and PEG 5000;

stability enhancers such as sucrose or sorbitol;

tonicity agents such as alkali metal halides, preferably sodium or potassium chloride, mannitol, sorbitol;

carrier media for parenteral administration, including sodium chloride solution, Ringer's solution with dextrose, sodium chloride with dextrose, Ringer's solution with lactate or fixed oils;

IV carrier media, including fluid and nutrient supplements, electrolyte supplements (eg, those underlying Ringer's dextrose solution).

In the context of the present invention, the pharmaceutical composition, which is preferably a liquid composition, or which may be a solid composition obtained by lyophilization, or may be a reconstituted liquid composition, contains:

(a) an antibody construct comprising at least three domains, wherein the first domain binds to a surface antigen of a target cell and has an isoelectric point (pI) ranging from 4 to 9.5;

the second domain binds to the second antigen and has a pI in the range from 8 to 10, preferably from 8.5 to 9.0; and optionally the third domain comprises two polypeptide monomers, each of which contains a hinge region, a CH2 domain and a CH3 domain, wherein said two polypeptide monomers are fused to each other by a peptide linker;

(b) at least one buffering agent;

(c) at least one saccharide; and (d) at least one surfactant; and wherein the pH value of the pharmaceutical composition is in the range from 3.5 to 6.

In the context of the present invention, it is further provided that the at least one buffer agent is present at a concentration in the range of 5-200 mM, more preferably at a concentration in the range of 10-50 mM. In the context of the present invention, it is provided that at least one saccharide is selected from the group consisting of a monosaccharide, a disaccharide, a cyclic polysaccharide, a sugar alcohol, a linear branched dextran or a linear linear dextran. It is also contemplated within the context of the present invention that the disaccharide is selected from the group consisting of sucrose, trehalose and mannitol, sorbitol and combinations thereof. In the context of the present invention, it is further contemplated that the sugar alcohol is sorbitol. In the context of the present invention, it is provided that the at least one saccharide is present at a concentration in the range of 1 to 15% (w/v), preferably at a concentration in the range of 9 to 12% (w/v).

It is also contemplated within the context of the present invention that at least one surfactant is selected from the group consisting of polysorbate 20, polysorbate 40, polysorbate 60, polysorbate 80, poloxamer 188, Pluronic F68, Triton X-100, polyoxyethylene, PEG 3350, PEG 4000 and their combinations. In the context of the present invention it is further provided that

- 50 046123 at least one surfactant is present at a concentration in the range of 0.004 to 0.5% (w/v), preferably in the range of 0.001 to 0.01% (w/v). In the context of the present invention, it is envisaged that the pH value of the composition is in the range from 4.0 to 5.0, preferably 4.2. It is also contemplated within the context of the present invention that the pharmaceutical composition has an osmolarity in the range of 150 to 500 mOsm. In the context of the present invention, it is further provided that the pharmaceutical composition further comprises an excipient selected from the group consisting of one or more polyols and one or more amino acids. In the context of the present invention, it is contemplated that said one or more adjuvants are present at a concentration in the range of 0.1-15% (w/v).

It is also contemplated within the context of the present invention that the pharmaceutical composition contains:

(a) the antibody-based construct discussed above;

(b) 10 mM glutamate or acetate;

(c) 9% (w/v) sucrose or 6% (w/v) sucrose and 6% (w/v) hydroxypropyl-β-cyclodextrin;

(d) 0.01% (w/v) polysorbate 80, and wherein the pH of the liquid pharmaceutical composition is 4.2.

In the context of the present invention, it is further contemplated that the antibody construct is present at a concentration in the range of 0.1-8 mg/ml, preferably 0.2-2.5 mg/ml, more preferably 0.25-1.0 mg/ml .

Those skilled in the art will appreciate that different constituents of a pharmaceutical composition (eg, those listed above) may have different effects, for example, an amino acid may act as a buffer, stabilizer and/or antioxidant; mannitol may act as a bulking agent and/or a tonicity agent; sodium chloride may act as a delivery vehicle and/or tonicity agent, etc.

It is understood that the composition of the present invention may contain, in addition to the polypeptide of the present invention as defined herein, additional biologically active agents depending on the intended use of the composition. Such agents may be drugs that act on the gastrointestinal tract, drugs that act as cytotoxic agents, drugs that prevent hyperuricemia, drugs that suppress immune responses (eg, corticosteroids), drugs that modulate the inflammatory response, drugs agents acting on the circulatory system and/or agents such as cytokines known in the art. It is also contemplated that the antibody construct of the present invention be used as a combination therapy, i.e. in combination with another anticancer drug.

In some embodiments, the optimal pharmaceutical composition will be determined by one skilled in the art depending, for example, on the route of administration, delivery format, and dosage required. See, for example, REMINGTON'S PHARMACEUTICAL SCIENCES, supra. In certain embodiments, such compositions may affect the physical state, stability, in vivo release rate, and in vivo clearance rate of the antibody construct of the present invention. In some embodiments, the primary carrier medium or primary carrier in the pharmaceutical composition may be either aqueous or non-aqueous in nature. For example, a suitable vehicle or carrier may be water for injection, physiological saline or artificial cerebrospinal fluid, possibly supplemented with other materials conventionally used in compositions for parenteral administration. Neutral buffered saline or saline mixed with serum albumin are additional examples of carrier media. In some embodiments, the antibody construct of the compositions of the present invention may be prepared for storage by mixing the selected composition of the desired degree of purity with optional formulation agents (REMINGTON'S PHARMACEUTICAL SCIENCES, supra) as a lyophilized mass or an aqueous solution. Additionally, in certain embodiments, the antibody construct of the present invention can be formulated as a lyophilisate using appropriate excipients, such as sucrose.

If parenteral administration is contemplated, therapeutic compositions for use in the present invention can be prepared as a pyrogen-free, parenterally acceptable aqueous solution containing the desired antibody construct of the present invention in a pharmaceutically acceptable carrier vehicle. In particular, a suitable carrier medium for parenteral administration is sterile distilled water, in which the antibody construct of the present invention is prepared as a sterile, isotonic solution

- 51 046123 ra, properly maintained. In some embodiments, the preparation may involve formulation of the desired molecule with a vehicle such as injectable microspheres, biodegradable particles, polymeric compounds (such as polylactic acid or polyglycolic acid), beads or liposomes, which can provide a controlled or sustained release of the product that can be delivered by injection of a substance with delayed absorption. In some embodiments, hyaluronic acid may also be used, which has the effect of providing a sustained action in the systemic circulation. In specific embodiments, implantable drug delivery devices can be used to administer the desired antibody construct.

Additional pharmaceutical compositions will be apparent to those skilled in the art, including formulations containing the antibody construct of the present invention in sustained release or controlled delivery/release formulations. Also known to those skilled in the art are methods for formulating a number of other delayed or controlled delivery agents, such as liposomal carriers, biodegradable microparticles or porous granules, and delayed absorption injectables. See, for example, International Patent Application No. PCT/US93/00829, which describes the controlled release of porous polymer microparticles for the delivery of pharmaceutical compositions. Sustained release formulations may include semi-permeable polymer matrices in the form of molded articles, such as films or microcapsules. Sustained release matrices may include polyesters, hydrogels, polylactides (as described in US Pat. No. 3,773,919 and European Patent Application Publication EP 058481), copolymers of L-glutamic acid and gamma-ethyl-Lglutamate (Sidman et al., 1983, Biopolymers 2:547-556), poly(2-hydroxyethyl methacrylate) (Langer et al., 1981, J. Biomed. Mater. Res. 15:167-277 and Langer, 1982, Chem. Tech. 12:98-105) , ethylene vinyl acetate (Langer et al., 1981, supra) or poly-I(-)-3-hydroxybutyric acid (European Patent Application Publication No. EP 133988). Sustained release compositions may also include liposomes, which can be prepared by any of several methods known in the art. See, for example, Eppstein et al., 1985, Proc. Natl. Acad. Sci. U.S.A. 82:3688-3692; publication of applications for European patent No. EP 036676; EP 088046 and EP 143949.

The antibody construct can also be encapsulated into microcapsules produced, for example, by coacervation or interfacial polymerization techniques (e.g., hydroxymethylcellulose or gelatin microcapsules and polymethyl methacrylate microcapsules, respectively), into colloidal drug delivery systems (e.g. , liposomes, albumin microspheres, microemulsions, nanoparticles and nanocapsules) or in macroemulsions. Such techniques are disclosed in Remington's Pharmaceutical Sciences, 16th edition, Oslo, A., Ed., (1980).

Pharmaceutical compositions used for in vivo administration are generally presented in the form of sterile preparations. Sterilization can be achieved by filtration using sterile filtration membranes. If the composition is lyophilized, then sterilization using this method can be performed either before or after lyophilization and reconstitution. Compositions for parenteral administration can be stored in lyophilized form or in solution. Typically, compositions for parenteral administration are placed in a container having a sterile entry port, such as an intravenous solution bag or vial with a stopper pierced by a hypodermic needle.

Another aspect of the present invention includes self-buffering compositions based on the antibody constructs of the present invention, which can be used as pharmaceutical compositions, as described in international patent application WO 06138181A2 (PCT/US2006/022599). Various explanations are available regarding protein stabilization and formulation materials and methods useful in this regard, such as Arakawa et al., Solvent interactions in pharmaceutical formulations, Pharm Res. 8(3): 285-91 (1991); Kendrick et al., Physical stabilization of proteins in aqueous solution in RATIONAL DESIGN OF STABLE PROTEIN FORMULATIONS: THEORY AND PRACTICE, Carpenter and Manning, eds. Pharmaceutical Biotechnology. 13: 61-84 (2002), and Randolph et al., Surfactantprotein interactions, Pharm Biotechnol. 13: 159-75 (2002), see in particular the parts relating to excipients and associated processes for self-buffering protein-based formulations in accordance with the present invention, in particular protein-based pharmaceutical products and processes for veterinary applications and/or human medicine.

Salts can be used in accordance with some embodiments of the present invention for the purpose of, for example, adjusting the ionic strength and/or isotonicity of the composition and/or improving the solubility and/or physical stability of the protein or other ingredient of the composition in accordance with the present invention. It is well known that ions can stabilize the native state of proteins by binding to charged residues on the protein surface and by shielding charged and polar groups in the protein and reducing the strength of their electrostatic, attractive and repulsive interactions. Ions can also stabilize the denatured state of a protein by binding, in particular, denatured peptide bonds (-CONH) of the protein. In addition, ionic interaction with charged and polar groups in a protein can also reduce intermolecular electrostatic interactions and thereby prevent or reduce protein aggregation and insolubility.

Ionic compounds vary significantly in their effects on proteins. A number of categorical gradations of ions and their effects on proteins have been developed that can be used in the formulation of pharmaceutical compositions in accordance with the present invention. One example is the Hofmeister series, which ranks ionic and polar nonionic solutes according to their effect on the conformational stability of proteins in solution. Stabilizing solutes are called kosmotropic. Destabilizing solutes are called chaotropic. Cosmotropes are typically used in high concentrations (eg, >1 mol/L ammonium sulfate) to precipitate proteins from solution (salting out). Chaotropes are commonly used for dental prosthetics and/or for protein solubilization (salting-in). The relative efficiency of ions in salting-in and salting-out determines their position in the Hofmeister series.

Free amino acids can be used in the compositions of the antibody construct of the present invention in accordance with various embodiments of the present invention as bulking agents, stabilizers and antioxidants, as well as for other standard uses. Lysine, proline, serine and alanine can be used to stabilize proteins in the formulation. Glycine is used in lyophilization to ensure the correct structure and properties of the mass. Arginine may be useful for inhibiting protein aggregation in both liquid and lyophilized formulations. Methionine is used as an antioxidant.

Polyols include sugars, such as mannitol, sucrose and sorbitol, and polyhydric alcohols, such as, for example, glycerol and propylene glycol, and for purposes of discussion herein, polyethylene glycol (PEG) and related substances. Polyols are kosmotropic. They are useful stabilizing agents in both liquid and lyophilized formulations to protect proteins from physical and chemical degradation processes. Polyols are also useful for adjusting the tonicity of formulations. Polyols useful in some embodiments of the present invention include mannitol, which is commonly used to provide structural stability to the mass in lyophilized formulations. This ensures the structural stability of the mass. It is usually used with a lyoprotectant such as sucrose. Sorbitol and sucrose are among the preferred agents for adjusting tonicity and as stabilizers to protect against freeze-thaw stresses during transportation or receiving bulk materials in the manufacturing process. Reducing sugars (which contain free aldehyde or ketone groups) such as glucose and lactose can glycate surface lysine and arginine residues. Therefore, they are generally not among the preferred polyols for use in accordance with the present invention. In addition, sugars that form such reactive compounds, such as sucrose, which hydrolyzes to fructose and glucose under acidic conditions, and therefore causes glycation, are also not among the preferred polyols of the present invention in this regard. PEG is useful for stabilizing proteins and as a cryoprotectant, and can be used in this regard in the present invention.

Embodiments of the antibody construct compositions of the present invention further provide surfactants. Protein molecules can be subject to adsorption to surfaces and denaturation and subsequent aggregation at air-liquid, solid-liquid, and liquid-liquid interfaces. These effects are usually inversely proportional to protein concentration. In general, these deleterious interactions are inversely proportional to protein concentration and are usually exacerbated by physical agitation, such as that encountered during transport and handling of the product. Surfactants are generally used to prevent, minimize or reduce surface adsorption. Useful surfactants in the present invention in this regard include polysorbate 20, polysorbate 80, other fatty acid esters of sorbitan polyethoxylates and poloxamer 188. Surfactants are also commonly used to control the conformational stability of a protein. The use of surfactants in this regard is protein specific, since any given surfactant typically stabilizes some proteins and destabilizes others.

Polysorbates are sensitive to oxidative degradation and often contain sufficient peroxides when supplied to cause oxidation of protein side chains, especially methionine. Therefore, polysorbates should be used with caution and when used, they should be used at the lowest effective concentration. In this regard, polysorbates exemplify the general rule that excipients should be used at their lowest effective concentrations.

Embodiments of the antibody construct compositions of the present invention further provide antioxidants. To some extent, harmful protein oxidation can be prevented in pharmaceutical formulations by maintaining proper oxygen levels and

- 53 046123 ambient temperature and avoidance of exposure to light. Antioxidant excipients can also be used to prevent oxidative degradation of proteins. Useful antioxidants in this regard include reducing agents, oxygen/free radical scavengers, and chelating agents. Antioxidants for use in therapeutic protein compositions in accordance with the present invention are preferably water soluble and remain active throughout the shelf life of the product. EDTA is the preferred antioxidant of the present invention in this regard. Antioxidants can damage proteins. For example, reducing agents, such as glutathione in particular, can disrupt intramolecular disulfide bonds. Thus, antioxidants for use in the present invention are selected among others to eliminate or sufficiently reduce the potential for damage to proteins in the formulation.

The compositions of the present invention may include metal ions that are protein cofactors and that are necessary for the formation of protein coordination complexes, such as zinc, necessary for the formation of certain insulin suspensions. Metal ions can also inhibit certain processes that break down proteins. However, metal ions also catalyze physical and chemical processes that break down proteins. Magnesium ions (10120 mM) can be used to inhibit the isomerization of aspartic acid to isoaspartic acid. Ca ⁺² ions (up to 100 mM) can increase the stability of human deoxyribonuclease. However, Mg ⁺² , Mn ⁺² and Zn ⁺² can destabilize rhDNase. Similarly, Ca ⁺² and Sr ⁺² can stabilize factor VIII, while it can be destabilized by Mg ⁺² , Mn ⁺² and Zn ⁺² , Cu ⁺² and Fe ⁺² ions, and its aggregation can be increased by Al ⁺² ions.

Embodiments of the antibody construct compositions of the present invention further include the use of one or more preservatives. Preservatives are essential in the development of multi-dose parenteral formulations that include more than one withdrawal from a single container. Their main function is to inhibit the growth of microorganisms and ensure sterility of the product throughout the shelf life or period of use of the medicinal product. Traditionally used preservatives include benzyl alcohol, phenol and m-cresol. Although preservatives have long been used with small molecule drugs for parenteral administration, developing protein-based formulations that include preservatives can be challenging. Preservatives almost always have a destabilizing effect (aggregation) on proteins, and this has become a major factor limiting their use in multidose protein-based formulations. To date, most protein-based drugs have been formulated for single use only. However, if multi-dose formulations can be prepared, they have the added benefit of providing patient convenience and improved marketability. A good example is human growth hormone (hGH), where the development of preservative formulations has led to the introduction of more convenient dosage forms in the form of refillable pens. At least four such pen devices containing hGH-based formulations with added preservatives are currently available on the market. Norditropin (liquid, Novo Nordisk), Nutropin AQ (liquid, Genentech) and Genotropin (lyophilized-dual-chamber cartridge, Pharmacia & Upjohn) contain phenol, while Somatrope (Eli Lilly) contains m-cresol. Several aspects must be taken into account when formulating and developing dosage forms containing added preservatives. The effective concentration of the preservative in the drug product must be optimized. To do this, it is necessary to test this preservative in dosage form in concentration ranges that provide antimicrobial effectiveness without compromising protein stability.

As expected, developing liquid formulations containing preservatives is more challenging than developing lyophilized formulations. Freeze-dried products can be lyophilized without preservative and reconstituted with a preservative-containing diluent during use. This reduces the time the preservative is in contact with the protein, significantly minimizing the associated stability risks. When using liquid formulations, the effectiveness and stability of the preservative must be maintained throughout the shelf life of the product (18 to 24 months). It is important to note that the effectiveness of the preservative must be demonstrated in the final formulation containing the active drug and all excipients.

The antibody constructs described herein can also be formulated as immunoliposomes. A liposome is a small vesicle composed of various types of lipids, phospholipids and/or surfactant that is useful for delivering a drug to a mammal. Liposome components typically have a bilayer structure similar to the lipid structure of biological membranes. Liposomes containing the antibody construct are prepared using methods known in the art, such as those described in Epstein et al., Proc. Natl. Acad. Sci. USA, 82: 3688 (1985); Hwang et al., Proc. Natl Acad. Sci. USA, 77: 4030 (1980); in US patents No. 4485045 and 4544545 and WO 97/38731. Liposomes with increased circulation time in the bloodstream are disclosed in US Pat. No. 5,013,556. Using the reverse phase method

- 54 046123 evaporation can produce particularly useful liposomes whose lipid composition contains phosphatidylcholine, cholesterol and PEG-derivatized phosphatidylethanolamine (PEG-PE). Liposomes are extruded through filters with a specific pore size to obtain liposomes of the desired diameter. Fab' fragments of the antibody construct of the present invention can be conjugated to liposomes as described in Martin et al. J Biol. Chem. 257: 286-288 (1982), via a disulfide exchange reaction. The liposome does not necessarily contain a chemotherapeutic agent. See Gabizon et al. J. National Cancer Inst. 81 (19) 1484 (1989).

Once the pharmaceutical composition has been formulated, it can be stored in sterile vials as a solution, suspension, gel, emulsion, solid, crystal, or as a dehydrated or lyophilized powder. Such formulations can be stored either in a ready-to-use form or in a form (eg, lyophilized) that is reconstituted before administration.

The biological activity of the pharmaceutical composition defined herein can be determined, for example, using cytotoxicity assays described in the following examples, in WO 99/54440 or in Schlereth et al. (Cancer Immunol. Immunother. 20 (2005), 1-12). The term efficacy or in vivo efficacy as used herein refers to response to therapy with a pharmaceutical composition of the present invention using, for example, NCI standardized response criteria. The success or effectiveness of in vivo therapy using the pharmaceutical composition of the present invention relates to the effectiveness of the composition in relation to its intended purpose, i.e. the ability of the composition to achieve the desired effect, i.e. depletion of pathological cells, such as tumor cells. Efficacy in vivo can be monitored using generally accepted standard methods for the relevant disease entities, including, but not limited to, white blood cell count, leukocyte count, fluorescence activated cell sorting, and bone marrow sampling. In addition, clinical and chemical parameters specific to a particular disease can be used, as well as other generally accepted standard methods. In addition, computed tomography, x-ray, nuclear magnetic resonance imaging (eg, to assess response based on National Cancer Institute criteria) can be used [Cheson BD, Horning SJ, Coiffier B, Shipp MA, Fisher RI, Connors JM, Lister TA, Vose J, Grillo-Lopez A, Hagenbeek A, Cabanillas F, Klippensten D, Hiddemann W, Castellino R, Harris NL, Armitage JO, Carter W, Hoppe R, Canellos GP. Report of an international workshop to standardize response criteria for non- Hodgkin's lymphomas. NCI Sponsored International Working Group. J Clin Oncol. 1999 Apr; 17(4): 1244]), positron emission tomography scanning, white blood cell count, leukocyte count determination, fluorescence activated cell sorting, bone marrow aspiration, research biopsies/histosections of lymph nodes and various clinical and chemical indicators specific for lymphoma (for example, lactate dehydrogenase level), as well as other generally accepted standard methods.

Another significant problem in the development of drugs, such as the pharmaceutical composition of the present invention, is the predictable modulation of pharmacokinetic properties. In this regard, it is possible to establish the pharmacokinetic profile of the drug candidate, i.e. a profile of pharmacokinetic parameters that influence the ability of a particular drug to treat a specified condition. Pharmacokinetic parameters of a drug that influence the drug's ability to treat a particular disease entity include, but are not limited to, half-life, volume of distribution, first-pass metabolism, and degree of serum protein binding. The effectiveness of this drug may be affected by any of the indicators mentioned above. An intended characteristic of the antibody constructs of the present invention provided with a specific form of Fc is, for example, that they differ in pharmacokinetic behavior. The extended half-life antibody targeting construct of the present invention preferably exhibits a remarkably increased in vivo residence time compared to canonical non-HLE variants of the same antibody construct.

Half-life means the time when 50% of the administered drug is eliminated through biological processes, such as metabolism, excretion, etc. First-pass metabolism refers to the susceptibility of a drug to metabolism after first contact with the liver, i.e. during its first passage through the liver. The volume of distribution means the extent to which a drug is retained in various compartments of the body, such as, for example, intracellular and extracellular space, tissues and organs, etc., and the distribution of the drug in these compartments. Serum protein binding refers to the ability of a drug to interact and bind to serum proteins such as albumin, resulting in a decrease or loss of the drug's biological activity.

Pharmacokinetic parameters also include bioavailability, lag period (Tlag), Tmax, absorption rate, onset of action and/or Cmax for a given amount of drug administered

- 55 046123 funds. Bioavailability refers to the amount of drug in the blood compartment. Lag period refers to the time delay between the administration of a drug and its detection and measurability in the blood or plasma. Tmax is the time at which the maximum blood concentration of a drug is reached, and Cmax is the maximum blood concentration that the drug reaches. The time required to achieve the drug concentration in the blood or tissue required for biological action is influenced by all parameters. The pharmacokinetics of bispecific antibody constructs exhibiting cross-species specificity that can be determined in preclinical animal testing involving non-chimpanzee primates as noted above are also outlined, for example, in Schlereth et al. (Cancer Immunol. Immunother. 20 (2005), 1-12).

In a preferred aspect of the present invention, the pharmaceutical composition is stable for at least four weeks at approximately -20°C. As is evident from the accompanying examples, the quality of the antibody constructs of the present invention compared to the quality of prior art antibody constructs can be examined using a variety of systems. These tests are intended to comply with the ICH Harmonized Tripartite Guideline: Stability Testing of Biotechnological/Biological Products Q5C and Specifications: Test procedures and Acceptance Criteria for Biotech Biotechnological/Biological Products Q6B, and are thus selected to provide a profile indicating stability that provides confidence is that changes in the identity, purity and potency of the product are detected. It is generally accepted that the term purity is a relative term. Due to the influence of glycosylation, deamidation, or other types of heterogeneity, the absolute purity of a biotechnological/biological product typically must be assessed using more than one method, and the resulting purity value will vary by method. For stability testing purposes, purity studies should focus on methods for determining degradation products.

To assess the quality of a pharmaceutical composition containing the antibody construct of the present invention, analysis can be performed, for example, analysis of the content of soluble aggregates in solution (HMWS, assessed using size exclusion chromatography). Preferably, stability for at least four weeks at about -20°C is characterized by less than 1.5% HMWS, preferably less than 1% HMWS.

A preferred formulation for an antibody construct as a pharmaceutical composition may, for example, contain the formulation components described below.

Ingredients: potassium phosphate, L-arginine hydrochloride, trehalose dihydrate, polysorbate 80 at pH 6.0.

Other examples of evaluating the stability of the antibody construct of the present invention in the form of a pharmaceutical composition are provided in the accompanying Examples 4-12. In these examples, embodiments of the antibody constructs of the present invention are tested against various stress conditions in various pharmaceutical formulations, and the results are compared with other half-life enhancing (HLE) formats of bispecific T cell recruiting antibody constructs that are known in the art. technology. In general, it is contemplated that antibody constructs provided with a specific form of the Fc domain of the present invention are generally more stable under a wide range of stress conditions, such as temperature and light stress, compared to antibody constructs provided with and without various HLE formats. them (for example, canonical antibody-based constructs). The specified temperature stability may refer to both low (below room temperature, including freezing) and elevated (above room temperature, including temperatures up to or above body temperature). One skilled in the art will appreciate that this improved stability against stress, which is difficult to avoid when practiced in a clinical setting, makes the antibody design safer since fewer degradation products will occur when practiced in a clinical setting. Accordingly, said increased stability means increased safety.

In one embodiment, an antibody construct of the present invention or an antibody construct prepared in accordance with a method of the present invention is provided for use in the prevention, treatment, or amelioration of a cancer correlated with MUC17 expression or overexpression of MUC17, such as prostate cancer glands.

The compositions described herein are useful as pharmaceutical compositions in the treatment, reduction of severity and/or prevention of the pathological condition described herein in a patient in need thereof. The term treatment refers to both therapeutic treatment and prophylactic or preventive measures. Treatment involves the use or administration of a composition to the body, isolated tissue or cell of a patient who has a disease/disorder, a symptom of a disease/disorder, or a predisposition to a disease/disorder, for the purpose of treating, curing, mitigating, alleviating, altering, correcting, reducing severity, improvements or changes

- 56 046123 effects on a disease, a symptom of a disease or a predisposition to a disease.

The term amelioration as used herein refers to improving the disease state of a patient having a disease, as defined herein below, by administering an antibody construct of the present invention to a subject in need thereof. Such improvement may also be seen as slowing or stopping the progression of the patient's disease. The term provides, as used herein, means preventing the occurrence or reoccurrence in a patient having a tumor or cancer or metastatic cancer, as defined herein below, by administering an antibody construct of the present invention to a subject in need thereof.

The term disease refers to any condition that would benefit from treatment with an antibody construct or pharmaceutical composition described herein. It includes chronic and acute disorders or diseases, including those pathological conditions that predispose the mammal to the disease in question.

A neoplasm is an abnormal growth of tissue, usually, but not always, forming a tumor mass. When a mass also forms, it is usually called a tumor. The growths or tumors may be benign, potentially malignant (precancerous), or malignant. Malignant neoplasms are usually called cancer. They usually invade and destroy surrounding tissue and can form metastases, i.e. they spread to other parts, tissues or organs of the body. Therefore, the term metastatic cancer covers metastases to other tissues or organs other than the origin of the tumor. Lymphomas and leukemias are lymphoid neoplasms. For the purposes of the present invention, they are also covered by the terms tumor or cancer.

The term viral disease describes diseases that result from a viral infection in a subject.

The term immunological disorder, as used herein, describes immunological disorders, such as autoimmune diseases, hypersensitivity, immunodeficiencies, in accordance with the general definition of the term.

In one embodiment, the present invention provides a method of treating or reducing cancer correlated with MUC17 expression or MUC17 overexpression, comprising the step of administering to a subject in need thereof an antibody construct of the present invention or an antibody construct prepared in accordance with the method according to the present invention. The bispecific single chain antibody to MUC17xCD3 is particularly advantageous for the therapy of cancer, preferably solid tumors, more preferably carcinomas and prostate cancer.

The terms subject in need of treatment or in need of treatment include those who already have a disorder as well as those in whom the disorder is to be prevented. The subject or patient in need includes human or other mammalian subjects who are receiving either prophylactic or therapeutic treatment.

The antibody construct of the present invention is typically designed for specific routes and modes of administration, for specific dosages and frequencies of administration, for specific treatments for specific diseases, with specific ranges of bioavailability and persistence in the body, among other things. The materials for the composition are preferably formulated in concentrations that are acceptable for the given site of administration.

Thus, the formulations and compositions can be prepared in accordance with the present invention for delivery by any suitable route of administration. In the context of the present invention, routes of administration include, but are not limited to:

local routes (such as cutaneous, inhalation, nasal, ocular, auricular/ear, vaginal, mucosal);

enteral routes (such as oral, gastrointestinal, sublingual, sublabial, buccal, rectal);

parenteral routes (such as intravenous, intraarterial, intraosseous, intramuscular, intracerebral, intracerebroventricular, epidural, intrathecal, subcutaneous, intraperitoneal, extra-amniotic, intraarticular, intracardiac, intradermal, intralesional, intrauterine, intravesical, intravitreal, transdermal, intranasal, transmucosal, intrasynovial, in traluminal ).

The pharmaceutical compositions and antibody construct of the present invention are particularly useful for parenteral administration, for example, subcutaneous or intravenous delivery, for example, by injection, such as bolus injection, or by infusion, such as continuous infusion. Pharmaceutical compositions can be administered using medical devices. Examples of medical devices for administering pharmaceutical compositions are described in US Pat. No. 4,475,196; 4439196; 4447224; 4447233; 4486194; 4487603; 4596556;

- 57 046123

4790824;4941880;5064413;5312335;5312335; 5383851 and 5399163.

In particular, the present invention provides for continuous administration of a suitable composition. As a non-limiting example, continuous or essentially continuous, i.e. Long-term administration can be accomplished using a small pump system worn by the patient to meter the delivery of the therapeutic agent into the patient's body. A pharmaceutical composition containing the antibody construct of the present invention can be administered using the above pump systems. In general, such pumping systems are known in the art and their operation is typically based on periodic replacement of cartridges containing the therapeutic agent to be infused. Replacing the cartridge in such a pump system entails a temporary interruption or some other interruption in the flow of therapeutic agent into the patient's body. In such a case, the administration phase before changing the cassette and the administration phase after changing the cassette would still fall within the definition of the pharmaceutical agents and methods of the present invention, together constituting one continuous administration of such therapeutic agent.

Long-term or continuous administration of the antibody constructs of the present invention may be intravenous or subcutaneous and accomplished by means of a fluid delivery device or a small pump system including a fluid delivery mechanism for dispensing fluid from a reservoir and a drive mechanism for driving the delivery mechanism. Subcutaneous pump systems may include a needle or cannula to penetrate the patient's skin and deliver a suitable composition into the patient's body. These pumping systems can be directly attached or attached to the patient's skin regardless of the location of veins, arteries and blood vessels, thereby ensuring direct contact between the pumping system and the patient's skin. The pump system can be attached to the patient's skin for 24 hours to several days. The pumping system may be small in size with a reservoir for small volumes. As a non-limiting example, the volume of the reservoir for administering a suitable pharmaceutical composition may range from 0.1 to 50 ml.

Long-term administration can also be accomplished by the transdermal route via a patch worn on the skin and replaced at regular intervals. Suitable patch-based drug delivery systems for this purpose are known to those skilled in the art. It should be noted that transdermal administration is particularly suitable for continuous administration, since replacement of the first used patch can advantageously be carried out simultaneously with the placement of a new second patch, for example, on the surface of the skin immediately adjacent to the first used patch and immediately before removal of the first used patch. There are no problems with interruption of flow or termination of the feed cell.

If the pharmaceutical composition has been lyophilized, the lyophilized material is first reconstituted in an appropriate liquid before administration. The lyophilized material can be reconstituted, for example, in bacteriostatic water for injection (BWFI), physiological saline, phosphate buffered saline (PSB), or in the same formulation as the protein before lyophilization.

The compositions of the present invention can be administered to a subject at an appropriate dose, which can be determined, for example, through dose escalation studies by administering increasing doses of an antibody construct of the present invention described herein demonstrating cross-species specificity to non-chimpanzee primates, e.g. macaques. As stated above, the antibody construct of the present invention described herein, demonstrating cross-species specificity, can advantageously be used in identical form in preclinical studies in non-chimpanzee primates and as a drug in humans. The dosage regimen is determined by the attending physician and clinical factors. As is well known in the medical field, dosages for any patient depend on many factors, including body size, body surface area, age of the patient, the specific compound to be administered, gender, time and route of administration, general health and other medications. , introduced together.

The term effective dose or effective dosage is defined as an amount sufficient to achieve or at least partially achieve the desired effect. The term therapeutically effective dose is defined as an amount sufficient to cure or at least partially abort a disease and its complications in a patient who is already suffering from the disease. Amounts or dosages effective for this use will depend on the condition being treated (indication), the antibody construct being delivered, the therapeutic context and objectives, the severity of the disease, prior therapy, the patient's medical history and response to the therapeutic agent, route of administration, the size (body weight, body surface or organ size) and/or condition (age and general health) of the patient and the general condition of the patient's own immune system. The appropriate dose can be adjusted according to the judgment of the attending physician so that it can be administered to the patient at a time or through a series of administrations to obtain optimal therapeutic effect.

A typical dosage may range from about 0.1 μg/kg to about 30 mg/kg or more depending on the above factors. In specific embodiments, the dosage may range from 1.0 μg/kg to about 20 mg/kg, optimally from 10 μg/kg to about 10 mg/kg, or from 100 μg/kg to about 5 mg/kg.

A therapeutically effective amount of the antibody construct of the present invention preferably results in a reduction in the severity of symptoms of a disease, an increase in the frequency or duration of asymptomatic periods, or the prevention of impairment or disability due to disease involvement. For the treatment of diseases correlated with MUC17 expression, as described above herein, a therapeutically effective amount of an antibody construct of the present invention, in this case an anti-MUC17/CD3 antibody construct, preferably inhibits cell growth or tumor growth by at least about 20%, at least about 40%, at least about 50%, at least about 60%, at least about 70%, at least about 80%, or at least about 90% relative to patients not exposed treatment. The ability of a compound to inhibit tumor growth can be assessed in an animal model predictive of efficacy.

The pharmaceutical composition can be administered as a single therapeutic agent or in combination with additional therapeutic agents, such as anticancer therapies, as appropriate, for example, other protein and non-protein drugs. These drugs may be administered concomitantly with a composition containing an antibody construct of the present invention as defined herein, or separately, before or after administration of said antibody construct at specific times and dosages.

The term effective and non-toxic dose as used herein refers to a tolerable dose of the antibody construct of the present invention high enough to cause pathological cell depletion, tumor elimination, tumor shrinkage or disease stabilization with or substantially no significant toxicity. effects. Such effective and non-toxic doses can be determined, for example, using dose escalation studies described in the art, and should be below the dose that induces serious adverse side effects (dose limiting toxicity, DLT).

The term toxicity as used herein refers to the toxic effects of a drug, manifested as adverse events or serious adverse events. These side effects may relate to the lack of tolerability of the drug in general and/or the lack of local tolerability after administration. Toxicity may also include teratogenic or carcinogenic effects caused by the drug.

The term safety, in vivo safety, or tolerability as used herein defines the administration of a drug without inducing serious adverse events immediately after administration (local tolerability) and over a longer period of drug use. Safety, in vivo safety or tolerability can be assessed, for example, at regular intervals during the treatment period and follow-up. Measurements include clinical assessment, such as visceral findings, and examination of abnormal laboratory values. Clinical evaluation may be performed and abnormal findings recorded/coded according to NCI-CTC and/or MedDRA standards. Organ manifestations may include criteria such as allergy/immunology, blood/bone marrow, cardiac arrhythmia, coagulation, etc., as described, for example, in the Common Terminology Criteria for Adverse Events v3.0 (CTCAE). Laboratory parameters that can be examined include, for example, hematology, clinical chemistry, coagulation profile and urinalysis, as well as examination of other body fluids such as serum, plasma, lymphatic or cerebrospinal fluid, cerebrospinal fluid, etc.

Thus, safety can be assessed, for example, through physical examination, imaging techniques (i.e. ultrasound, x-ray, CT scan, magnetic resonance imaging (MRI), other measurements using technical devices (i.e. electrocardiogram ), vital signs, by measuring laboratory parameters and recording adverse events.

For example, adverse events in non-chimpanzee primates in the uses and methods of the present invention can be examined using histopathological and/or histochemical methods.

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

Finally, the present invention provides a kit containing an antibody construct of the present invention or prepared in accordance with a method of the present invention, a pharmaceutical composition of the present invention, a polynucleotide of the present invention, a vector of the present invention, and/or a cell - the owner of the present invention.

In the context of the present invention, the term kit means two or more components, one of which corresponds to an antibody construct, pharmaceutical composition, vector or host cell of the present invention, packaged together in a container, recipient or otherwise. Therefore, a kit can be described as a collection of products and/or tools sufficient to achieve a specific purpose that can be sold as a single unit.

The kit may contain one or more containers (such as vials, ampoules, containers, syringes, bottles, bags) of any suitable shape, size and material (preferably, for example, waterproof plastic or glass) containing the antibody construct or pharmaceutical composition according to the present invention in a suitable dosage for administration (see above). The kit may further contain instructions for use (for example, in the form of a leaflet or brochure), means for administering the antibody construct of the present invention, such as a syringe, pump, infusion system and the like, means for reconstituting the antibody construct of the present invention. the present invention and/or a means for diluting the antibody construct of the present invention.

The present invention also provides kits with a single-dose administration device. The kit of the present invention may also comprise a first container containing a dried/lyophilized antibody construct and a second container containing an aqueous formulation. In some embodiments, the present invention provides kits containing single and multi-chamber prefilled syringes (eg, liquid or lyophilized syringes).

It is noted that singular forms used herein include references to the plural unless the context clearly indicates otherwise. Thus, for example, reference to a reagent includes one or more of these different reagents, and reference to a method includes reference to equivalent steps and methods known to those skilled in the art that may be modified or used in place of the methods described herein.

Unless otherwise specified, the term at least preceding sequences of elements should be understood to refer to each element in the series. Those skilled in the art will appreciate, or will be able to determine through routine experimentation alone, many equivalents to the specific embodiments of the present invention described herein. Such equivalents are intended to be included in the present invention.

The expression and/or, as used in any part herein, includes the meanings of and, or and all elements or any other combination thereof, combined by the specified expression.

As used herein, the term approximately or approximately means within 20%, preferably within 10%, and more preferably within 5% of a given value or range. Moreover, it also includes a specific number, for example, approximately 20 includes 20.

The term less than or more than includes a specific number. For example, less than 20 means less than or equal to. Similarly, more than or greater than means more than or equal to or greater than or equal to, respectively.

Throughout this specification and the claims that follow, unless the context otherwise requires, the word contain and its variants, such as contains and containing, will be understood to include the specified integer or step, or group of integers or steps, but not the exclusion of some other whole number or stage, or group of whole numbers or stages. The term including, when used herein, may be replaced by the terms containing or including, or sometimes when used herein by the term having.

As used herein, consisting of excludes any element, step or ingredient not defined in the element of the claim.

In the context of this document, the expression consisting essentially of does not exclude materials or stages that do not significantly affect the basic and novel characteristics of the claimed object.

In each case, any of the terms containing, consisting essentially of, and consisting of can be replaced by either of the remaining two terms.

It should be understood that the present invention is not limited to the specific methodologies, protocols, materials, reagents and substances, etc. described herein and, therefore, may vary. The terminology used herein is used only for the purpose of describing specific embodiments, and is not intended to limit the present invention, which is defined solely by the claims.

All publications and patents referred to throughout this specification (including all patents, patent applications, scientific publications, specifications and instructions from the manufacturer, etc.) above or below are hereby incorporated by reference in their entirety. completeness. Nothing set forth herein should be construed as an admission that the present invention is not entitled to be opposed to such description based on an earlier invention. To the extent that material incorporated by reference is inconsistent with or inconsistent with this description, this description takes precedence over any such material.

A better understanding of the present invention and its advantages will be provided by the following examples, which are provided for illustrative purposes only. The examples are not intended to limit the scope of the present invention in any way.

Examples

Example 1: Evaluation of cell surface expression of MUC17.

Cell surface MUC17 levels were determined by flow cytometry using QIFIkit (Dako). Adherent cells were separated using nonenzymatic cell dissociation buffer (Cellstripper, Coining) and then stained with anti-MUC17 antibody 4C11. Antibody 4C11 is a monoclonal antibody generated by immunization of B6 mice with a DNA construct spanning the EGF-SEA-EGF region of MUC17 (aa 4131-4493). MUC17 was detected by incubation with a FITC-conjugated secondary antibody and analyzed by flow cytometry. Relative antibody binding capacity was determined using QIFIkit (Dako) using the bead samples provided in the kit as a standard. The results are shown in Fig. 3 (A) MUC17 gene expression levels in cancer cell lines were determined by quantitative polymerase chain reaction (qPCR) using methods and probes from Applied Biosystems/Thermo Fisher. RNA was isolated from cancer cell lines and then transcribed into cDNA. MUC17 cDNA was amplified with MUC17-specific probes using qPCR. MUC17 gene expression levels were normalized to those of a constitutively expressed gene, such as those depicted in FIG. 3 (B).

Example 2. Evaluation of the in vitro effectiveness of constructs based on bispecific antibodies to MUC17.

The cellular activity of MUC17 HLE antibody constructs was assessed in T cell-dependent cytotoxicity (TDCC) assays. Effector cells were obtained from commercial sources such as AllCells or Cepheus Biosciences, Inc. Human pan-T cells, human PBMCs, or cynomolgus PBMCs were incubated at a 10:1 or 5:1 ratio with target cells expressing human or cynomolgus MUC17 in the presence of a range of doses of antibody constructs. After 48 h of incubation, cellular cytotoxicity was assessed using a luminescent assay (T-glo cell or Steady-glo (Promega)) or high-resolution imaging (Cellomics ArrayScan) as a readout parameter of cellular cytotoxicity. The results are shown in Fig. 4 and 5.

Example 3. Xenograft study to evaluate the in vivo efficacy of a bispecific antibody to MUC17 construct.

The purpose of the xenograft study was to evaluate the antitumor activity of a bispecific anti-MUC17/CD3 antibody construct with extended half-life after intravenous administration in a subcutaneous GSU-luc xenograft model of human gastric cancer in female NOD/SCID mice.

Obtaining target and effector cells for inoculation.

Target cells.

GSU human gastric carcinoma cells transfected with the lentiviral vector LV417Luc to stably express firefly luciferase (GSU-luc) were harvested, centrifuged, washed with cold DPBS, counted and adjusted to a concentration of 5 x 10 ⁷ cells/ml. A total of 5 x 10 ⁶ cells/mouse (SC) were injected subcutaneously into the right posterior flank of female NOD/SCID mice (supplier: Envigo) in a final volume of 100 μl.

Effector cells.

Human T cells were isolated from fresh blood of healthy donors (#0801), enriched for CD3+ T cells using a pan T cell isolation kit (#130-096-535) and activated and expanded in vitro using a T activation/expansion kit -human cells (No. 130-091-441, both from Miltenyi Biotec) according to the manufacturer's instructions.

On the day of injection, T cells were counted, isolated from beads, and washed with 2x cold DPBS. The cell count was adjusted to 1 x 108 cells/ml and kept on ice until injection. A total of 2 x 10 ⁷ cells/mouse were injected into the peritoneal cavity (IP) in a final volume of 200 μl. Cells were kept on ice until injection.

Experimental design.

Animals received the MUC17 bispecific antibody construct or control compound via intravenous (IV) bolus injection (tail vein). Mice were treated in accordance with table. 4.

- 61 046123

Table 4

Study Design for Efficacy Studies

Group Mouse/ Group Target cells/ Effector cells/ Treatment RoA Level doses Volume doses Days of education mouse (SC) mouse (IP) (mg/kg) (ml) boots 1 5 5x10 ^b GSU-luc - Control connection IV 0 0.1 12.19 2 10 5x10 ⁶ GSU-luc 2χ10 ^ζ CD3 ⁺ Control connection IV 0 o,1 12.19 3 10 5x10 ⁶ GSU-luc 2χ10 ⁷ CD3 ⁺ Bispecific construct KMUC17 IV 2.5 0.1 12.19, 26 4 10 5x10 ⁶ GSU-luc 2χ10 ⁷ CD3 ⁺ Bispecific construct KMUC17 IV 0.25 0.1 12.19, 26 5 10 5x10 ⁶ GSU-luc 2χ10 ⁷ CD3 ⁺ Bispecific construct KMUC17 IV 0.025 0.1 12.19, 26 55 Additional animals (residual) to ensure equal tumor volume at start of treatment Σ 100 Animals at the beginning of the study

Sequence of research.

Day 1: Subcutaneous injection of tumor cells (GSU-luc) into the right posterior flank of female NOD/SCID mice (see above). Animals were 6 weeks old at the start of the study.

Day 7. Anti-Asialo antibody treatment.

To deplete remaining NK cells/NK cell activity, mice were treated with a single dose of polyclonal (rabbit anti-mouse Ig) anti-asialo-GM1 antibody · Anti-asialo-GM1 antibody was reconstituted according to the manufacturer's instructions and a 50 μl diluted 1:2.5 dose of H2O was injected IV into the lateral tail vein.

Day 8: Injection of CD3+ T cells into the peritoneal cavity of mice (see above).

Days 11, 18 and 25: FcR blocking.

The Fc region of the test compound was mutated to prevent binding to Fcy receptors. However, since NOD/Scid mice lack B cells resulting in low immunoglobulin levels, FcR blocking was performed to avoid the potential reduction of CD3+ effector cells through antibody-dependent cell-mediated cytotoxicity. On days 11, 18, and 25, mice received a mixture of anti-FcyR antibody 2.4G2 (8 mg/kg) and Kiovig (400 mg/kg) via intraperitoneal bolus injection in a final volume of 200 μl per mouse per injection.

Days 12, 19 and 26: treatment with test or control compound (see Table 4).

Animals received test compound (anti-MUC17/CD3 bispecific antibody construct) or control compound (vehicle) by intravenous (IV) bolus injection into the lateral tail vein on days 12, 19, and 26 according to Table 1. 4. The dose volume was kept constant at a total of 100 μL per mouse per injection.

The test compound was formulated in 10 mM L-glutamic acid, 9% (w/v) sucrose, 0.01% (w/v) PS80; pH 4.2 at a stock solution concentration of 1.04 mg/ml and diluted in carrier medium (25 mM L-lysine monohydrochloride, 0.002% (w/v) polysorbate 80 in 0.9% (w/v) chloride sodium pH 7.0) according to the last determined mean body weight (BW) of the group. Dose concentration (s) was calculated using the formula:

- 62 046123 _G µg, C -] = ml dose1|x average BW [kg] dose volume [µl]

Day 33. End of study.

Experimental studies and calculations.

All animals were observed daily during the course of the study to assess general appearance, activity, behavior, and survival. All results and comments were noted on the appropriate sheet in the study documentation. Body weight values were determined 3 times per week during the study. Tumor growth progression was determined by measuring tumor height and width using an external caliper. Tumor growth was determined 3 times per week and tumor volumes (TV) were calculated using the formula:

height x width^ where width is defined as the smaller value and height is defined as the larger value of the two dimensions.

All measured raw data were downloaded to a computer and automatically imported into VIVO Manager software for further data processing. Values not calculated by VIVO Calculations were calculated using a calculation table document from MS Excel or GraphPad Prism for Windows.

Graphical results are presented in Fig. 6 as group means ± standard error of the mean. Data were analyzed using one-way analysis of variance (ANOVA) and differences in experimental results for tumor growth were assessed using Dunnett's post hoc test for comparison with control group 2.

Relative tumor volume (RTV) was calculated by dividing the group mean tumor volume on day p by the group mean tumor volume on the day before treatment (day 11).

Tumor growth inhibition was quantified on day 20, the last day when all animals in the control group were alive, according to the formula:

- _gp ,, . ί average tumor volume in the treatment group [mm3],

Tumor growth inhibition [%] = 100 - „ „-----------—----4----4—Ax 100 anterior tumor volume in the control group [mm3] 7

Results.

Intravenous treatment of GSU-luc tumor-bearing mice with a bispecific antibody construct to MUC17/CD3 (test compound, SEQ ID NO: 186) resulted in a statistically significant and dose-dependent inhibition of tumor growth compared to vehicle-treated control group 2. carrier. After starting treatment on day 12, p < 0.01 (at 0.25 mg/kg) or p < 0.001 (at 2.5 mg/kg) values were achieved on days 18 and 20. Since most animals (6/10) control group were sacrificed, no statistical analysis was performed after day 20. Tumor growth inhibition observed at day 20 was 24% (0.025 mg/kg), 58% (0.25 mg/kg), and 77% (2.5 mg/kg). Comparison of relative tumor volumes (RTV) at day 20 showed that while tumors growing in vehicle-treated mice had an average of 4.2 times greater volume relative to the day before treatment, RTV in the vehicle-treated groups compound was 3.4 (0.025 mg/kg), 2.4 (0.25 mg/kg) and 1.0 (2.5 mg/kg) times greater. Following treatment at 2.5 mg/kg, the RTV was <2.0 until day 29.

Comparison of two control groups treated with vehicle showed that T cells did not affect the growth of GSU-luc cells in the absence of the test compound. The test compound was well tolerated and drug-related adverse events were not expected or observed as the mouse is not a suitable species.

Ultimately, intravenous administration of the bispecific antibody constructs of the present invention (test compound SEQ ID NO: 186) at 2.5 or 0.25 mg/kg resulted in statistically significant and dose-dependent inhibition of the growth of subcutaneous GSU-luc tumors in female NOD/Scid mice.

Example 4: Experimental toxicological study on cynomolgus monkeys.

The anti-MUC17 HLE BiTE antibody construct (SEQ ID NO: 186, construct 8-B7) was evaluated in a pilot toxicology study in cynomolgus monkeys. Three monkeys were administered either 100 μg/kg or 1000 μg/kg of the MUC17 scFc bispecific antibody construct via intravenous injection on day 1 and day 8 of the study. The anti-MUC17 scFc bispecific antibody construct (SEQ ID NO: 186) was well tolerated at both doses with no associated clinical signs or changes in body weight. Temporary increases in body temperature were recorded at 100 μg/kg and 1000 μg/kg. We observed some signs of activity of the bispecific scFc anti-MUC17 construct (lymphocyte redistribution, increased levels of neutrophils and monocytes, elevated C-reactive protein, slight increase in cytokine levels) in blood samples from monkeys treated with the bispecific construct. antibodies with scFc to MUC17. Although immunohistochemical study confirmed the expression

MUC17 on the apical surface of small intestinal specimens from monkeys assessed in this experimental toxicology study did not reveal histopathological changes in tissues expressing MUC17.

Toxicokinetic parameters of an anti-MUC17 scFc bispecific antibody construct in cynomolgus monkeys.

The toxicokinetic parameters of an anti-MUC17 scFc bispecific antibody construct (SEQ ID NO: 186) were assessed in blood samples collected from monkeys evaluated in a pilot toxicology study. Blood samples were collected before dosing and at 0.083, 4, 8, 24, 48, 96, and 168 hours after each dosing. The concentration of the anti-MUC17 scFc bispecific antibody construct in serum was determined by an immunoassay using a rutenylated mouse anti-human Fc IgG 1.35.1 mAb directed against MUC17 to capture the antibody construct and an Fc fragment-directed antibody for detection. designs. Serum levels of the anti-MUC17 scFc bispecific antibody construct were determined at all time points analyzed after the first dose. The data were fit to a two-component model. In fig. Figure 8 (B) shows the individual data (dots) and the average (line). Several pharmacokinetic parameters were assessed, including systemic clearance (CL), intracompartmental clearance (Q), serum volume/central compartment volume (Vp), tissue volume/compartment tissue volume (Vt), terminal half-life (t1/2), and the second dose of 1000 mcg/kg - the average maximum dose concentration (Cmax) and the area under the serum concentration versus time curve (AUCinf).

Example 5 T-cell dependent cytotoxicity assays in normal intestinal cells.

To further test the idea that MUC17 localization to the apical surface of normal human and cynomolgus intestinal cells is inaccessible to the cytotoxic activity of the MUC17 scFc bispecific antibody construct (SEQ ID NO: 186), MUC17 expression and activity of the bispecific antibody construct with scFc to MUC17 was assessed in normal cells in vitro. Cell surface expression of MUC17 was assessed by fluorescence-activated cell sorting. The cytotoxic activity of the MUC17 scFc bispecific antibody construct was assessed in T cell-dependent cytotoxicity (TDCC) assays where the anti-MUC17 scFc bispecific antibody construct was incubated with MUC17-positive human or monkey target and effector cells ( i.e. T cells or peripheral blood mononuclear cells), and then cell viability was assessed. Such experiments are initially tested using standard two-dimensional cell culture. However, in order to better observe the localization of MUC17 at the apical surface, normal cells are cultured in a manner that maintains epithelial cell polarity, such as growth on an extracellular matrix or in organ-like culture. A construct based on a bispecific antibody with scFc to MUC17 did not show a significant increase in TDCC relative to normal, i.e. different from cancerous intestinal cells.

Table 5

Sequence table

SEQID NO: Designation Source Subsequence 1. Linker (G4S) Artificial ahh GGGGS 2. Linker (G4S)2 Artificial ahh GGGGSGGGGS 3. Linker (G4S)3 Artificial ahh GGGGSGGGGSGGGGS 4. Linker (G4S)4 Artificial ahh GGGGSGGGGSGGGGSGGGGS 5. Linker (G4S)5 Artificial ahh GGGGSGGGGSGGGGSGGGGSGGGGS 6. Linker (G4S)6 Artificial ahh GGGGSGGGGSGGGGSGGGGSGGGGSG GGGS 7. Linker (G4S)7 Artificial ahh GGGGSGGGGSGGGGSGGGGSGGGGSG GGGSGGGGS 8. Linker (G4S)8 Artificial ahh GGGGSGGGGSGGGGSGGGGSGGGGSG GGGSGGGGSGGGGS 9. Peptide linker Artificial ahh PGGGGS 10. Peptide linker Artificial ahh PGGDGS eleven. Peptide Artificial ahh SGGGGS

- 64 046123

linker 12. Peptide linker Artificial ahh GGGG 13. Contact CD3sVL Artificial ahh QTVVTQEPSLTVSPGGTVTLTCGSSTGA VTSGNYPNWVQQKPGQAPRGLIGGTKF LAPGTPARFSGSLLGGKAALTLSGVQPE DEAEYYCVLWYSNRWVFGGGTKLTVL 14. Contact CD3sVH Artificial ahh EVQLVESGGGLVQPGGSLRLSCAASGFT FNSYAMNWVRQAPGKGLEWVARIRSK YNNYATYYADSVKGRFTISRDDSKNTA YLQMNSLKTEDTAVYYCVRHGNFGNS YVSWWAYWGQGTLVTVSS 15. Contact CD3s scFv Artificial ahh EVQLVESGGGLVQPGGSLRLSCAASGFT FNSYAMNWVRQAPGKGLEWVARIRSK YNNYATYYADSVKGRFTISRDDSKNTA YLQMNSLKTEDTAVYYCVRHGNFGNS YVSWWAYWGQGTLVTVSSGGGGSGG GGSGGGGSQTVVTQEPSLTVSPGGTVTL TCGSSTGAVTSGNYPNWVQQKPGQAPR GLIGGTK FLAPGTPARFSGSLLGGKAAL TLSGVQPEDEAEYYCVLWYSNRWVFG GGTKLTVL 16. Hexahistidine tag Artificial ahh HHHHHH 17. Fc-monomer-1 +c/-g Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPCEEQYGSTYRC VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLY SKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGK

- 65 046123

18. PC-monomer-2 +c/-g/delGK Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPCEEQYGSTYRC VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLY SKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSP 19. Fc-monomer-3 -c/+g Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPREEQYNSTYRV VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFF LYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGK 20. Fc-MOHOMep-4 -c/+g/delGK Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPREEQYNSTYRV VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFF LYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSP 21. Fc-MOHOMep-5 -c/-g Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPREEQYGSTYRV VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFF LYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ

- 66 046123

KSLSLSPGK 22. Fc-monomer-b -c/-g/delGK Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPREEQYGSTYRV VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFF LYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSP 23. FC-monomer-? +c/+g Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPCEEQYNSTYRC VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLY SKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGK 24. Fc-MOHOMep-8 +c/+g/delGK Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPCEEQYNSTYRC VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLY SKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSP 25. scFc-1 Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPCEEQYGSTYRC VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLY SKLTV

- 67 046123

DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGKGGGGSGGGGSGGGGSGGG GSGGGGSGGGGSDKTHTCPPCPAPELLG GPSVFLFPPKPKDTLMISRTPEVTCVVVD VSHEDPEVKFNWYVDGVEVHNAKTKP CEEQYGSTYRCVSVLTVLHQDWLNGKE YKCKVSNKALPAPIEKTISKAKGQ PREP QVYTLPPSREEMTKNQVSLTCLVKGFYP SDIAVEWESNGQPENNYKTTPPVLDSDG SFFLYSKLTVDKSRWQQGNVFSCSVMH EALHNHYTQKSLSLSPGK 26. scFc-2 Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPCEEQYGSTYRC VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLY SKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGGGGSGGGGSGGGGSGGGGS GGGGSGGGGSDKTHTCPPCPAPELLGGP SVFLFPPKPKDTLMISRTPEVTCVVVDVS HEDPEVKFNWYVDGVEVHNAKTKPCE EQYGSTYRCVSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAKGQ PREPQ VYTLPPSREEMTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGS FFLYSKLTVDKSRWQQGNVFSCSVMHE ALHNHYTQKSLSLSP 27. scFc-3 Artificial ahh DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPREEQYNSTYRV VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN

- 68 046123

GQPENNYKTTPPVLDSDGSFFLYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGKGGGGSGGGGSGGGGSGGG GSGGGGSGGGGSDKTHTCPPCPAPELLG GPSVFLFPPKPKDTLMISRTPEVTCVVVD VSHEDPEVKFNWYVDGVEVHNAKTKP REEQYNSTYRVVSVLTVLHQDWL NGKE YKCKVSNKALPAPIEKTISKAKGQPREP QVYTLPPSREEMTKNQVSLTCLVKGFYP SDIAVEWESNGQPENNYKTTPPVLDSDG SFFLYSKLTVDKSRWQQGNVFSCSVMH EALHNHYTQKSLSLSPGK 28. scFc-4 Artificial aa DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPREEQYNSTYRV VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFF LYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGGGGSGGGGSGGGGSGGGGS GGGGSGGGGSDKTHTCPPCPAPELLGGP SVFLFPPKPKDTLMISRTPEVTCVVVDVS HEDPEVKFNWYVDGVEVHNAKTKPRE EQYNSTYRVVSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAK GQPREPQ VYTLPPSREEMTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGS FFLYSKLTVDKSRWQQGNVFSCSVMHE ALHNHYTQKSLSLSP 29. scFc-5 Artificial aa DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPREEQYGSTYRV VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE

- 69 046123

MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGKGGGGSGGGGSGGGGSGGG GSGGGGSGGGGSDKTHTCPPCPAPELLG GPSVFLFPPKPKDTLMISRTPEVTCVVVD VSHEDPEVKFNWYVDGVEVHNAKTKP REEQYGSTYRVVSVLTVLHQDWLNGKE YKCKVSNKALPAPIEKTISKAKGQPREP QVYTLPPSREEMTKNQVSLTCLVKGFYP SDIAVEWESNGQPENNYKTTPPVLDSDG SFFLYSKLTVDKSRWQQGNVFSCSVMH EALHNHYTQKSLSLSPGK thirty. scFc-6 Artificial aa DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPREEQYGSTYRV VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFF LYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGGGGSGGGGSGGGGSGGGGS GGGGSGGGGSDKTHTCPPCPAPELLGGP SVFLFPPKPKDTLMISRTPEVTCVVVDVS HEDPEVKFNWYVDGVEVHNAKTKPRE EQYGSTYRVVSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAK GQPREPQ VYTLPPSREEMTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGS FFLYSKLTVDKSRWQQGNVFSCSVMHE ALHNHYTQKSLSLSP 31. scFc-7 Artificial aa DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPCEEQYNSTYRC VSVLTVLHQDWLNGKEYKCKVSNKAL

- 70 046123

PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGKGGGGSGGGGSGGGGSGGG GSGGGGSGGGGSDKTHTCPPCPAPELLG GPSVFLFPPKPKDTLMISRTPEVTCVVVD VSHEDPEVKFNWYVDGVEVHNAKTKP CEEQYNSTYRCVSVLTVLHQDWLNGKE YKCKVSNKALPAPIEKTISKAKGQPREP QVYTLPPSREEMTKNQVSLTCLVKGFYP SDIAVEWESNGQPENNYKTTPPVLDSDG SFFLYSKLTVDKSRWQQGNVFSCSVMH EALHNHYTQKSLSLSPGK 32. scFc-8 Artificial aa DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPCEEQYNSTYRC VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLY SKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGGGGSGGGGSGGGGSGGGGS GGGGSGGGGSDKTHTCPPCPAPELLGGP SVFLFPPKPKDTLMISRTPEVTCVVVDVS HEDPEVKFNWYVDGVEVHNAKTKPCE EQYNSTYRCVSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAKGQ PREPQ VYTLPPSREEMTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGS FFLYSKLTVDKSRWQQGNVFSCSVMHE ALHNHYTQKSLSLSP 33. CDR1 VH MU 92- G6 CC x I2C0-scFc Artificial aa SYGMH 34. CDR2 VH MU 92- Artificial aa VISFEGSNKYYASSVKG

- 71 046123

G6 CC x I2C0-scFc 35. CDR3 VH MU 92- G6 CC x I2C0-scFc Artificial ahh GAYTYGFDY 36. CDR1 VL MU 92- G6 CC x I2C0-scFc Artificial ahh RASQSVNRYLA 37. CDR2 VL MU 92- G6 CC x I2C0-scFc Artificial ahh GASNRAT 38. CDR3 VL MU 92- G6 CC x I2C0-scFc Artificial ahh HHYGSSIFA 39. VH MU 92-G6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISFE GSNKYYASSVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSS 40. VL MU 92-G6 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSV NRYLAWYQQKPGQAPRLLIYGASNRAT GIPDRFTGSGSGTDFTLTISRLEPEDFAV YFCHHYGSSIFAFGCGTKVEIK 41. scFv MU 92-G6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISFE GSNKYYASSVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTG SSGSGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIK 42. Bispecific molecule MU 92G6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISFE GSNKYYASSVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG

- 72 046123

IPDRFTGSGSGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIKSSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGS QTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL 43. Bispecific molecule with HLE MU 92-G6 SS x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISFE GSNKYYASSVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTG SGSGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIKSGGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEP SLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTYRCVSVLTVLHQD WLNGKEYK CKVSNKALPAPIEKTISKAKGQPREPQV

- 73 046123

YTLPPSREEMTKNQVSLTCLVKGFYPSD IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELLGGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 44. CDR1 VH MU 92- C12 CC x I2C0scFc Artificial ahh SFGMH 45. CDR2 VH MU 92- C12 CC x I2C0scFc Artificial ahh VIWFSGSNKYYAEAVKG 46. CDR3 VH MU 92- C12 CC x I2C0scFc Artificial ahh GGYTYGFDY 47. CDR1 VL MU 92- C12 CC x I2C0scFc Artificial ahh RANQAINRYLA 48. CDR2 VL MU 92- C12 CC x I2C0scFc Artificial ahh GASSRAT 49. CDR3 VL MU 92C12 CC x I2C0scFc Artificial ahh HHYGSSIFT 50. VH MU 92-C12 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWFS GSNKYYAEAVKGRFTISRDNSKNTLYL

- 74 046123

QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSS 51. VL MU 92-C12 CC x I2C0-scFc Artificial ahh EIVLTQSPATLSLSPGERATLSCRANQAI NRYLAWYQQKPGQAPRLLIYGASSRAT GIPDRFSGSGSGTDFTLTISRLEPEDFAVY YCHHYGSSIFTFGCGTKVEIK 52. scFv MU 92-C12 CC X I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWFS GSNKYYAEAVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPATLSLSPGERATLSCRANQAIN RYLAWYQQKPGQAPRLLIYGASSRATGI PDRFSGSG SGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 53. Bispecific molecule MU 92C12 CC x I2C0scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWFS GSNKYYAEAVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPATLSLSPGERATLSCRANQAIN RYLAWYQQKPGQAPRLLIYGASSRATGI PDRFSGSG SGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIKSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSL TVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL 54. Bispecific Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF

- 75 046123

molecule with HLE MU 92-C12 CC x I2C0-scFc TFSSFGMHWVRQAPGKCLEWVAVIWFS GSNKYYAEAVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPATLSLSPGERATLSCRANQAIN RYLAWYQQKPGQAPRLLIYGASSRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSS IFTFGCGTKVEIKSGGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQ KPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTYRCVSVLTVLHQDWLNGKEYK CKVSNKALPAPIEKTISKA KGQPREPQV YTLPPSREEMTKNQVSLTCLVKGFYPSD IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTTVDKSRWQQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELLGGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFN WYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK

- 76 046123

55. CDR1 VH MU 32- G6 CC x I2C0-scFc Artificial ahh NHAMH 56. CDR2 VH MU 32- G6 CC x I2C0-scFc Artificial ahh GIWSEGNSKYYAESVKG 57. CDR3 VH MU 32- G6 CC x I2C0-scFc Artificial ahh ATYTTGWSYFDY 58. CDR1 VL MU 32- G6 CC x I2C0-scFc Artificial ahh SGDKLGDKYAS 59. CDR2 VL MU 32- G6 CC x I2C0-scFc Artificial ahh QDRKRPS 60. CDR3 VL MU 32- G6 CC x I2C0-scFc Artificial ahh QAYDASTWV 61. VH MU 32-G6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNHAMHWVRQAPGKCLEWVAGIWS EGSNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARATYTTGWSY FDYWGQGTLVTVSS 62. VL MU 32-G6 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASFTCSGDKLGD KYASWYQQKSGQSPVLVIYQDRKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAYDASTWVFGCGTQLTVL 63. scFv MU 32-G6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNHAMHWVRQAPGKCLEWVAGIWS EGSNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARATYTTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYASWYQQKSGQSPVLVIYQDRKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAYDASTWVFGCGTQLTVL 64. Bispecific molecule MU 32G6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNHAMHWVRQAPGKCLEWVAGIWS EGSNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARATYTTGWSY

- 77 046123

FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASrrCSGDK LGDKYASWYQQKSGQSPVLVIYQDRKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAYDASTWVFGCGTQLTVLSG GGGSEVQLVESGGGLVQPGGSLKLSCA ASGFTFNKYAMNWVRQAPGKGL EWVA RIRSKYNNYATYYADSVKDRFTISRDDS KNTAYLQMNNLKTEDTAVYYCVRHGN FGNSYISYWAYWGQGTLVTVSSGGGGS GGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW V FGGGTKLTVL 65. Bispecific molecule with HLE MU 32-G6 SS x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNHAMHWVRQAPGKCLEWVAGIWS EGSNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARATYTTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASrrCSGDK LGDKYASWYQQKSGQSPVLVIYQDRK R PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAYDASTWVFGCGTQLTVLSG GGGSEVQLVESGGGLVQPGGSLKLSCA ASGFTFNKYAMNWVRQAPGKGLEWVA RIRSKYNNYATYYADSVKDRFTISRDDS KNTAYLQMNNLKTEDTAVYYCVRHGN FGNSYISYWAYWGQGTLVTVSSGGGGS GGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP ELLGGPSVFLFPPKPKDTLMISRTPEVTC

- 78 046123

WVDVSHEDPEVKFNWYVDGVEVHNA KTKPCEEQYGSTYRCVSVLTVLHQDWL NGKEYKCKVSNKALPAPIEKTISKAKGQ PREPQVYTLPPSREEMTKNQVSLTCLVK GFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQGNVFSC SVMHEALHNHYTQKS LSLSPGKGGGGS GGGGSGGGGSGGGGSGGGGSGGGGSD KTHTCPPCPAPELLGGPSVFLFPPKPKDT LMISRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS VLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLPPSREEMT KNQVSLTCLVKGFYPSDI AVEWESNGQP ENNYKTTPPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNHYTQKSL SLSPGK 66. CDR1 VH MU 9- C2 CC x I2C0-scFc Artificial ahh NFGMH 67. CDR2 VH MU 9- C2 CC x I2C0-scFc Artificial ahh VIWFDASKTYYASSVKG 68. CDR3 VH MU 9- C2 CC x I2C0-scFc Artificial ahh ATYSTGWSYFDY 69. CDR1 VL MU 9- C2 CC x I2C0-scFc Artificial ahh SGDKLGDKYTS 70. CDR2 VL MU 9- C2 CC x I2C0-scFc Artificial ahh HDAKRPS 71. CDR3 VL MU 9- C2 CC x I2C0-scFc Artificial ahh QAWDASTAWV 72. VH MU 9-C2 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSS

- 79 046123

73. VL MU 9-C2 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASFTCSGDKLGD KYTSWYQQKPGQSPVLVIYHDAKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAWDASTAWVFGCGTKLTVL 74. scFv MU 9-C2 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWDASTAWVFGCGTKLTVL 75. Bispecific molecule MU 9C2 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWDASTAWVFGCGTKLTVLS GGGGSEVQLVESGGGLVQPGGSLKLSC AASGFTFNKYAMNWVRQAPGKGLEWV ARIRSKYNNYATYYADSVKDRFTISRDD SKNTAYLQMNNLKTEDTAVYYCVRHG NFGNSYISYWAYWGQGTLVTVSSGGGG SGG GGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVL 76. Bispecific molecule c HLE MU 9-C2 CC x Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL

- 80 046123

I2C0-scFc QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASETCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWDASTAWVFGCGTKLTVLS GGGGSEVQLVESGGGLVQPGGSLKLSC AASGFTFNKYAMNWVRQAPGKGLEWV ARIRSKYNNYATYYADSVKDRFTISRDD SKNTAYLQMNNLKTEDTAVYYCVRHG NFGNSYISYWAYWGQGTLVTVSSGGGG SGGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSG VQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP ELLGGPSVFLFPPKPKDTLMISRTPEVTC WVDVSHEDPEVKFNWYVDGVEVHNA KTKPCEEQYGSTYRCVSVLTVLHQDWL NGKEYKCKVSNKALPAPIEKTISKAKGQ PREPQVYTLPPSREEMTKNQVSLTCLV K GFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQGNVFSC SVMHEALHNHYTQKSLSLSPGKGGGGS GGGGSGGGGSGGGGSGGGGSGGGGSD KTHTCPPCPAPELLGGPSVFLFPPKPKDT LMISRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS V LTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLPPSREEMT KNQVSLTCLVKGFYPSDIAVEWESNGQP ENNYKTTPPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNHYTQKSL SLSPGK 77. CDR1 VH MU 8- Artificial aa GYYWS

- 81 046123

H9 CC x I2C0-scFc 78. CDR2 VH MU 8- H9 CC x I2C0-scFc Artificial ahh DIEHSGSTKYNPSLKS 79. CDR3 VH MU 8- H9 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDY 80. CDR1 VL MU 8- H9 CC x I2C0-scFc Artificial ahh SGDKLGDKYAS 81. CDR2 VL MU 8- H9 CC x I2C0-scFc Artificial ahh HDNKRPS 82. CDR3 VL MU 8- H9 CC x I2C0-scFc Artificial ahh QAYGSSSAV 83. VH MU 8-H9 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIEHSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSS 84. VL MU 8-H9 CC x I2C0-scFc Artificial ahh SYELTQSPSASVSPGQTASrrCSGDKLGD KYASWYQQKPGQSPVLVIYHDNKRPSG IPERFSGSNSGNTATLTISGTQAMDEAD YYCQAYGSSSAVFGCGTKLTVL 85. scFv MU 8-H9 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIEHSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQSPSASVSPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIY HDNKRPSGIP ERFSGSNSGNTATLTISGTQAMDEADYY CQAYGSSSAVFGCGTKLTVL 86. Bispecific molecule MU 8H9 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIEHSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY

- 82 046123

ELTQSPSASVSPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIYHDNKRPSGIP ERFSGSNSGNTATLTISGTQAMDEADYY CQAYGSSSAVFGCGTKLTVLSGGGGSE VQLVESGGGLVQPGGSLKLSCAASGFTF NKYAMNWVRQAPGKGLEWVARIRSKY NNYATYYADSVKDRFTISRDDSKNTAY L QMNNLKTEDTAVYYCVRHGNFGNSYI SYWAYWGQGTLVTVSSGGGGSGGGGS GGGGSQTVVTQEPSLTVSPGGTVTLTCG SSTGAVTSGNYPNWVQQKPGQAPRGLI GGTKFLAPGTPARFSGSLLGGKAALTLS GVQPEDEAEYYCVLWYSNRWVFGGGT KLTVL 87. Bispecific molecule with HLE MU 8-H9 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIEHSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQSPSASVSPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIYHDNKRPSGIP ERFSGSNSGNTATLTISGTQAMDEADYY CQAYGSSSAVFGCGTKLTVLSGGGGSE VQLVESGGGLVQPGGSLKLSCAASGFTF NKYAMNWVRQAPGKGLEWVARIRSKY NNYATYYADSVKDRFTISRDDSKNTAY LQMNNLKTEDTAVYYCVRHGNFGNSYI SYWAYWGQGTLVTVSSGGGGSGGGGS GGGGSQTVVTQEPSLTVSPGGTVTLTCG SSTGAVTSGNYPNWVQQKPGQAPRGLI GGTKFLAPGTPARFSGSLLGGKAALTLS GVQPEDEAEYYCVLWYSNRWVFGGGT KLTVLGGGGDKTHTCPPCPAPELLGGPS VFLFPPKPKDTLMISRTPEVTCVVVDVS HEDPEVKFNWYVDGVEVHNAKTKPCE

- 83 046123

P CPAPELLGGPSVFLFPPKPKDTLMISRT PEVTCVWDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSR WQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 88. CDR1 VH MU 8- H8 CC x I2C0-scFc Artificial ahh GYYWS 89. CDR2 VH MU 8- H8 CC x I2C0-scFc Artificial ahh DIDASGSTKYNPSLKS 90. CDR3 VH MU 8- H8 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDY 91. CDR1 VL MU 8- H8 CC x I2C0-scFc Artificial ahh SGDKLGDKYAS 92. CDR2 VL MU 8- H8 CC x I2C0-scFc Artificial ahh QDRKRPS 93. CDR3 VL MU 8- H8 CC x I2C0-scFc Artificial ahh QAWGSSTAV 94. VH MU 8-H8 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSS 95. VL MU 8-H8 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVPPGQTASHCSGDKLGD KYASWYQQKPGQSPVLHYQDRKRPSGV

- 84 046123

PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSTAVFGCGTKLTVL 96. scFv MU 8-H8 CC x I2C0-scFc Artificial aa QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPPSVSVPPGQTASrrCSGDKLGDK YASWYQQKPGQSPVLIIY QDRKRPSGVP ERFSGSNSGNTATLTISGTQAMDEADYY CQAWGSSTAVFGCGTKLTVL 97. Bispecific molecule MU 8H8 CC x I2C0-scFc Artificial aa QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPPSVSVPPGQTASrrCSGDKLGDK YASWYQQKPGQSPVLIIY QDRKRPSGVP ERFSGSNSGNTATLTISGTQAMDEADYY CQAWGSSTAVFGCGTKLTVLSGGGGSE VQLVESGGGLVQPGGSLKLSCAASGFTF NKYAMNWVRQAPGKGLEWVARIRSKY NNYATYYADSVKDRFTISRDDSKNTAY LQMNNLKTEDTAVYYCVRHGNFGNSYI SYWAYWGQGTLVTVSSGGGGSG GGGS GGGGSQTVVTQEPSLTVSPGGTVTLTCG SSTGAVTSGNYPNWVQQKPGQAPRGLI GGTKFLAPGTPARFSGSLLGGKAALTLS GVQPEDEAEYYCVLWYSNRWVFGGGT KLTVL 98. Bispecific molecule with HLE MU 8-H8 CC x I2C0-scFc Artificial aa QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY

- 85 046123

ELTQPPSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLIIYQDRKRPSGVP ERFSGSNSGNTATLTISGTQAMDEADYY CQAWGSSTAVFGCGTKLTVLSGGGGSE VQLVESGGGLVQPGGSLKLSCAASGFTF NKYAMNWVRQAPGKGLEWVARIRSKY NNYATYYADSVKDRFTISRDDSKNTAY LQMNNLKTEDTAVYYCVRHGNFGNSYI SYWAYWGQGTLVTVSSGGGGSGGGGS GGGGSQTVVTQEPSLTVSPGGTVTLTCG SSTGAVTSGNYPNWVQQKPGQAPRGLI GGTKFLAPGTPARFSGSLLGGKAALTLS GVQPEDEAEYYCVLWYSNRWVFGGGT KLTVLGGGGDKTHTCPPCPAPELLGGPS VFLF PPKPKDTLMISRTPEVTCVVVDVS HEDPEVKFNWYVDGVEVHNAKTKPCE EQYGSTYRCVSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAKGQPREPQ VYTLPPSREEMTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGS FFLYSKLTVDKSRWQQGNVFSCSVM HE ALHNHYTQKSLSLSPGKGGGGSGGGGS GGGGSGGGGSGGGGSGGGGSDKTHTCP PCPAPELLGGPSVFLFPPKPKDTLMISRT PEVTCVWDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQV SLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 99. CDR1 VH MU 8- H5 CC x I2C0-scFc Artificial aa SFGMH 100. CDR2 VH MU 8- H5 CC x I2C0-scFc Artificial aa VIWFDASNKYYAESVKG

- 86 046123

101. CDR3 VH MU 8- H5 CC x I2C0-scFc Artificial ahh GGYTYGFDY 102. CDR1 VL MU 8- H5 CC x I2C0-scFc Artificial ahh RASQAVNRYLA 103. CDR2 VL MU 8- H5 CC x I2C0-scFc Artificial ahh GASSRAT 104. CDR3 VL MU 8- H5 CC x I2C0-scFc Artificial ahh QQYGSSIFT 105. VH MU 8-H5 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSS 106. VL MU 8-H5 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQAV NRYLAWYQQKPGQAPRLLIYGASSRAT GIPDRFSGSGSGTDFTLTISRLEPEDFAVY YCQQYGSSIFTFGCGTKVEIK 107. scFv MU 8-H5 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSLPGERATLSCRASQAVN RYLAWYQQKPGQAPRLLIYGASSRATGI PDRF SGSSGSGTDFTLTISRLEPEDFAVYY CQQYGSSIFTFGCGTKVEIK 108. Bispecific molecule MU 8H5 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSLPGERATLSCRASQAVN RYLAWYQQKPGQAPRLLIYGASSRATGI PDRF SSGSGSGTDFTLTISRLEPEDFAVYY

- 87 046123

CQQYGSSIFTFGCGTKVEIKSSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGAVT SGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL 109. Bispecific molecule with HLE MU 8-H5 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSLPGERATLSCRASQAVN RYLAWYQQKPGQAPRLLIYGASSRATGI PDRF SSGSGSGTDFTLTISRLEPEDFAVYY CQQYGSSIFTFGCGTKVEIKSGGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVT QEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTYRCVSVLTVLH QDWLNGKEYK CKVSNKALPAPIEKTISKAKGQPREPQV YTLPPSREEMTKNQVSLTCLVKGFYPSD

- 88 046123

IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELLGGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVL H QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK BY. CDR1 VH MU 8- Fll CC x I2C0scFc Artificial ahh SHYWS 111. CDR2 VH MU 8- Fll CC x I2C0scFc Artificial ahh RIDVSGSANYNPALKS 112. CDR3 VH MU 8- Fll CC x I2C0scFc Artificial ahh APYSSGWGYFDY 113. CDR1 VL MU 8- Fll CC x I2C0scFc Artificial ahh SGDKLGDKYAS 114. CDR2 VL MU 8- Fll CC x I2C0scFc Artificial ahh HDNKRPS 115. CDR3 VL MU 8Fll CC x I2C0scFc Artificial ahh QAWDITTAV 116. VH MU 8-F11 CC x I2C0-scFc Artificial ahh QVQLQESGPGLVKPSETLSLTCTVSGGSI SSHYWSWIRQSAGKCLEWIGRIDVSGSA NYNPALKSRATMSADTSKNQFSLRLSSV TAADTAVYYCARAPYSSGWGYFDYWG

- 89 046123

QGTLVTVSS 117. VL MU 8-F11 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASITCSGDKLGD KYASWYQQQPGQSPVLVIYHDNKRPSG IPERFSGSNSGNTATLTISGTQAMDEAD YFCQAWDnTAVFGCGTKLTVL 118. scFv MU 8-F11 CC x I2C0-scFc Artificial ahh QVQLQESGPGLVKPSETLSLTCTVSGGSI SSHYWSWIRQSAGKCLEWIGRIDVSGSA NYNPALKSRATMSADTSKNQFSLRLSSV TAADTAVYYCARAPYSSGWGYFDYWG QGTLVTVSSGGGGSGGGGSGGGGSSYE LTQPPSVSVSPGQTASITCSGDKLGDKY ASWYQQQPGQSPVLVIYHDNKRPS GIPE RFSGSNSGNTATLTISGTQAMDEADYFC QAWDITTAVFGCGTKLTVL 119. Bispecific molecule MU 8Fll CC x I2C0scFc Artificial ahh QVQLQESGPGLVKPSETLSLTCTVSGGSI SSHYWSWIRQSAGKCLEWIGRIDVSGSA NYNPALKSRATMSADTSKNQFSLRLSSV TAADTAVYYCARAPYSSGWGYFDYWG QGTLVTVSSGGGGSGGGGSGGGGSSYE LTQPPSVSVSPGQTASITCSGDKLGDKY ASWYQQQPGQSPVLVIYHDNKRPS GIPE RFSGSNSGNTATLTISGTQAMDEADYFC QAWDHTAVFGCGTKLTVLSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GS QTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL 120. Bispecific molecule with HLE Artificial ahh QVQLQESGPGLVKPSETLSLTCTVSGGSI SSHYWSWIRQSAGKCLEWIGRIDVSGSA

- 90 046123

MU 8-F11 CC X I2C0-scFc NYNPALKSRATMSADTSKNQFSLRLSSV TAADTAVYYCARAPYSSGWGYFDYWG QGTLVTVSSGGGGSGGGGSGGGGSSYE LTQPPSVSVSPGQTASITCSGDKLGDKY ASWYQQQPGQSPVLVIYHDNKRPSGIPE RFSGSNSGNTATLTISGTQAMDEADYFC QAWDITTAVFGCGTKLTVLSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGT PARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTYRCVSVLTVLHQDWLNGKEYK CKVSNKALPAPIEKTISKAKGQPREPQV YTLPPSREEMT KNQVSLTCLVKGFYPSD IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELLGGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPC EEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 121. CDR1 VH MU 8- Artificial aa GYYWS

- 91 046123

F9 CC x I2C0-scFc 122. CDR2 VH MU 8- F9 CC x I2C0-scFc Artificial ahh DIDASGSTKYNPSLKS 123. CDR3 VH MU 8- F9 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDY 124. CDR1 VL MU 8- F9 CC x I2C0-scFc Artificial ahh SGDKLGDKYAS 125. CDR2 VL MU 8- F9 CC x I2C0-scFc Artificial ahh QDRKRPS 126. CDR3 VL MU 8- F9 CC x I2C0-scFc Artificial ahh QAWGSSAAV 127. VH MU 8-F9 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSS 128. VL MU 8-F9 CC x I2C0-scFc Artificial ahh SYELTQPSSVSVSPGQTASITCSGDKLGD KYASWYQQKPGQSPVLVIYQDRKRPSG VPERFSGSNSGNTATLTISGTQAMDEAD YYCQAWGSSAAVFGCGTKLTVL 129. scFv MU 8-F9 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVSPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIYQ DRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSAAVFGCGTKLTVL 130. Bispecific molecule MU 8-F9 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY

- 92 046123

ELTQPSSVSVSPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIYQDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSAAVFGCGTKLTVLSGGGGS EVQLVESGGGLVQPGGSLKLSCAASGFT FNKYAMNWVRQAPGKGLEWVARIRSK YNNYATYYADSVKDRFTISRDDSKN TA YLQMNNLKTEDTAVYYCVRHGNFGNS YISYWAYWGQGTLVTVSSGGGGSGGG GSGGGGSQTVVTQEPSLTVSPGGTVTLT CGSSTGAVTSGNYPNWVQQKPGQAPRG LIGGTKFLAPGTPARFSGSLLGGKAALT LSGVQPEDEAEYYCVLWYSNRWVFGG GTKLTVL 131. Bispecific molecule with HLE MU 8-F9 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVSPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIYQ DRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSAAVFGCGTKLTVLSGGGGS EVQLVESGGGLVQPGGSLKLSCAASGFT FNKYAMNWVRQAPGKGLEWVARIRSK YNNYATYYADSVKDRFTISRDDSKNTA YLQMNNLKTEDTAVYYCVRHGNFGNS YISYWAYWGQGTLVTVSSGGGGG GG GSGGGGSQTVVTQEPSLTVSPGGTVTLT CGSSTGAVTSGNYPNWVQQKPGQAPRG LIGGTKFLAPGTPARFSGSLLGGKAALT LSGVQPEDEAEYYCVLWYSNRWVFGG GTKLTVLGGGGDKTHTCPPCPAPELLGG PSVFLFPPKPKDTLMISRTPEVTCWVDV SHEDPEVKFNWYVDGVEVHNAKTKPCE

- 93 046123

P CPAPELLGGPSVFLFPPKPKDTLMISRT PEVTCVWDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSR WQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 132. CDR1 VH MU 8E3 CC X I2C0scFc Artificial ahh NHGMH 133. CDR2 VH MU 8E3 CC X I2C0scFc Artificial ahh GIWSDASNKYYADAVKG 134. CDR3 VH MU 8E3 CC X I2C0scFc Artificial ahh ATYTTGWSYFDY 135. CDR1 VL MU 8E3 CC X I2C0scFc Artificial ahh SGDKLGDKYAS 136. CDR2 VL MU 8E3 CC X I2C0scFc Artificial ahh QDNKRPS 137. CDR3 VL MU 8E3 CC X I2C0scFc Artificial ahh QAYDASTWV 138. VH MU 8-E3 CC Artificial ahh QVQLVESGGGVVQPGRSLRLACAASGF

- 94 046123

X I2C0-scFc TFSNHGMHWVRQAPGKCLEWVAGIWS DASNKYYADAVKGRFTISRDNSKNTLY LQMNSLRAEDTAVYYCARATYTTGWS YFDYWGQGTLVTVSS 139. VL MU 8-E3 CC X I2C0-scFc Artificial ahh SYELTQPASVSVSPGQTASITCSGDKLG DKYASWYQQKSGQSPVLVIYQDNKRPS GIPERFSGSNSGNTATLTISGTQAMDEA DYYCQAYDASTWVFGCGTQLTVL 140. scFv MU 8-E3 CC X I2C0-scFc Artificial ahh R PSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAYDASTWVFGCGTQLTVL 141. Bispecific molecule MU 8E3 CC X I2C0scFc Artificial ahh R PSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAYDASTWVFGCGTQLTVLSG GGGSEVQLVESGGGLVQPGGSLKLSCA ASGFTFNKYAMNWVRQAPGKGLEWVA RIRSKYNNYATYYADSVKDRFTISRDDS KNTAYLQMNNLKTEDTAVYYCVRHGN FGNSYISYWAYWGQGTLVTVSSGGGGS GGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW

- 95 046123

VFGGGTKLTVL 142. Bispecific molecule with HLE MU 8-E3 CC X I2C0-scFc Artificial ahh R PSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAYDASTWVFGCGTQLTVLSG GGGSEVQLVESGGGLVQPGGSLKLSCA ASGFTFNKYAMNWVRQAPGKGLEWVA RIRSKYNNYATYYADSVKDRFTISRDDS KNTAYLQMNNLKTEDTAVYYCVRHGN FGNSYISYWAYWGQGTLVTVSSGGGGS GGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP ELLGGPSVFLFPPKPKDTLMISRTPEVTC WVDVSHEDPEVKFNWYVDGVEVHNA KTKPC KTHTC PPCPAPELLGGPSVFLFPPKPKDT LMISRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS VLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLPPSREEMT KNQVSLTCLVKGFYPSDIAVEWESNGQP

- 96 046123

ENNYKTPPPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNHYTQKSL SLSPGK 143. CDR1 VH MU 8- D7 CC x I2C0-scFc Artificial ahh GYYWS 144. CDR2 VH MU 8- D7 CC x I2C0-scFc Artificial ahh DTDASGSTKYNPSLKS 145. CDR3 VH MU 8- D7 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDY 146. CDR1 VL MU 8- D7 CC x I2C0-scFc Artificial ahh SGDKLGEKYAS 147. CDR2 VL MU 8- D7 CC x I2C0-scFc Artificial ahh QDRKRPS 148. CDR3 VL MU 8- D7 CC x I2C0-scFc Artificial ahh QAWGSSAAV 149. VH MU 8-D7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW RQGTLVTVSS 150. VL MU 8-D7 CC x I2C0-scFc Artificial ahh SYELTQPSSVSVPPGQTASHCSGDKLGE KYASWYQQKPGQSPVLIIYQDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSAAVFGCGTKLTVL 151. scFv MU 8-D7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW RQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASITCSGDKLGEKY ASWYQQKPGQSPVLIIYQDR KRPSGVPE RFSGSNSGNTATLTISGTQAMDEADYYC QAWGSSAAVFGCGTKLTVLS 152. Bispecific Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG

- 97 046123

molecule MU 8D7 CC x I2C0-scFc SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW RQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASITCSGDKLGEKY ASWYQQKPGQSPVLIIYQDRKRPSGVPE RFSGSNSGNTATLTISGTQAMDE ADYYC QAWGSSAAVFGCGTKLTVLSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGA VTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL 153. Bispecific molecule with HLE MU 8-D7 CC x I2C0-scFc Artificial aa QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW RQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASITCSGDKLGEKY ASWYQQKPGQSPVLIIYQDR KRPSGVPE RFSGSNSGNTATLTISGTQAMDEADYYC QAWGSSAAVFGCGTKLTVLSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGG SG GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG

- 98 046123

VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTYRCVSVLTVLHQDWLNGKEYK CKVSNKALPAPIEKTISKAKGQPREPQV YTLPPSREEMTKNQVSLTCLVK GFYPSD IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELLGGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPCSVEEQYGSTYRCV LTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 154. CDR1 VH MU 8- C7 CC x I2C0-scFc Artificial ahh GYYWS 155. CDR2 VH MU 8- C7 CC x I2C0-scFc Artificial ahh DIDQSGSTKYNPSLKS 156. CDR3 VH MU 8- C7 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDY 157. CDR1 VL MU 8- C7 CC x I2C0-scFc Artificial ahh SGDKLGDKYAS 158. CDR2 VL MU 8- C7 CC x I2C0-scFc Artificial ahh QDRKRPS 159. CDR3 VL MU 8- C7 CC x I2C0-scFc Artificial ahh QAWGSSAAV 160. VH MU 8-C7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS

- 99 046123

VTAADTAVYFCARKKYSTVWSYFDYW RQGTLVTVSS 161. VL MU 8-C7 CC x I2C0-scFc Artificial ahh SYELTQPSSVSVSPGQTASTTCSGDKLGD KYASWYQQKPGQSPVLIIYQDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSAAVFGCGTKLTVL 162. scFv MU 8-C7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW RQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVSPGQTASITCSGDKLGDK YASWYQQKPGQSPVLIIY QDRKRPSGVP ERFSGSNSGNTATLTISGTQAMDEADYY CQAWGSSAAVFGCGTKLTVL 163. Bispecific molecule MU 8C7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW RQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVSPGQTASITCSGDKLGDK YASWYQQKPGQSPVLIIY QDRKRPSGVP ERFSGSNSGNTATLTISGTQAMDEADYY CQAWGSSAAVFGCGTKLTVLSGGGGSE VQLVESGGGLVQPGGSLKLSCAASGFTF NKYAMNWVRQAPGKGLEWVARIRSKY NNYATYYADSVKDRFTISRDDSKNTAY LQMNNLKTEDTAVYYCVRHGNFGNSYI SYWAYWGQGTLVTVSSGGGGSG GGGS GGGGSQTVVTQEPSLTVSPGGTVTLTCG SSTGAVTSGNYPNWVQQKPGQAPRGLI GGTKFLAPGTPARFSGSLLGGKAALTLS GVQPEDEAEYYCVLWYSNRWVFGGGT KLTVL 164. Bispecific Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG

- 100 046123

molecule with HLE MU 8-C7 SS x I2C0-scFc SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW RQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVSPGQTASITCSGDKLGDK YASWYQQKPGQSPVLIIYQDRKRPSGVP ERFSGSNSGNTATLTISGTQAM DEADYY CQAWGSSAAVFGCGTKLTVLSGGGGSE VQLVESGGGLVQPGGSLKLSCAASGFTF NKYAMNWVRQAPGKGLEWVARIRSKY NNYATYYADSVKDRFTISRDDSKNTAY LQMNNLKTEDTAVYYCVRHGNFGNSYI SYWAYWGQGTLVTVSSGGGGSGGGGS GGGGSQTVVTQEPSLTVSPGGTVTLTCG S STGAVTSGNYPNWVQQKPGQAPRGLI GGTKFLAPGTPARFSGSLLGGKAALTS GVQPEDEAEYYCVLWYSNRWVFGGGT KLTVLGGGGDKTHTCPPCPAPELLGGPS VFLFPPKPKDTLMISRTPEVTCVVVDVS HEDPEVKFNWYVDGVEVHNAKTKPCE EQYGSTYRCVSVLTVLHQDWLNGK EY KCKVSNKALPAPIEKTISKAKGQPREPQ VYTLPPSREEMTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGS FFLYSKLTVDKSRWQQGNVFSCSVMHE ALHNHYTQKSLSLSPGKGGGGSGGGGS GGGGSGGGGSGGGGSGGGGSDKTHTCP PCPAPELLGGPSVFLFPPKPKDTLMISRT PEVT CVWDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLS LSPGK

- 101 046123

165. CDR1 VH MU 8- B8 CC x I2C0-scFc Artificial ahh GYYWS 166. CDR2 VH MU 8- B8 CC x I2C0-scFc Artificial ahh DIDQSGSTKYNPSLKS 167. CDR3 VH MU 8- B8 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDY 168. CDR1 VL MU 8- B8 CC x I2C0-scFc Artificial ahh SGDKLGDKYAS 169. CDR2 VL MU 8- B8 CC x I2C0-scFc Artificial ahh QDRKRPS 170. CDR3 VL MU 8- B8 CC x I2C0-scFc Artificial ahh QAWGSSAAV 171. VH MU 8-B8 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSS 172. VL MU 8-B8 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVPPGQTASHCSGDKLGD KYASWYQQKPGQSPVLHYQDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSAAVFGCGTKLTVL 173. scFv MU 8-B8 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPPSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLIIY QDRKRPSGVP ERFSGSNSGNTATLTISGTQAMDEADYY CQAWGSSAAVFGCGTKLTVL 174. Bispecific molecule MU 8B8 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW

- 102 046123

GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPPSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLIIYQDRKRPSGVP ERFSGSNSGNTATLTISGTQAMDEADYY CQAWGSSAAVFGCGTKLTVLSGGGGSE VQLVESGGGLVQPGGSLKLSCAASGFTF NKYAMNWVRQAPGKGLEWVARIRSKY NNYATYYADSVKDRFTISRDDSKNTAY LQMNNLKTEDTAVYYCVRHGNFGNSYI SYWAYWGQGTLVTVSSGGGGSGGGGS GGGGSQTVVTQEPSLTVSPGGTVTLTCG SSTGAVTSGNYPNWVQQKPGQAPRGLI GGTKFLAPGTPARFSGSLLGGKAALTLS GVQPEDEAEYYCVLWYSNRWVFGGGT KLTV L 175. Bispecific molecule with HLE MU 8-B8 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPPSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLIIY QDRKRPSGVP ERFSGSNSGNTATLTISGTQAMDEADYY CQAWGSSAAVFGCGTKLTVLSGGGGSE VQLVESGGGLVQPGGSLKLSCAASGFTF NKYAMNWVRQAPGKGLEWVARIRSKY NNYATYYADSVKDRFTISRDDSKNTAY LQMNNLKTEDTAVYYCVRHGNFGNSYI SYWAYWGQGTLVTVSSGGGGSG GGGS GGGGSQTVVTQEPSLTVSPGGTVTLTCG SSTGAVTSGNYPNWVQKPGQAPRGLI GGTKFLAPGTPARFSGSLLGGKAALTLS GVQPEDEAEYYCVLWYSNRWVFGGGT KLTVLGGGGDKTHTCPPCPAPELLGGPS VFLFPPKPKDTLMISRTPEVTCVVVDVS

- 103 046123

HEDPEVKFNWYVDGVEVHNAKTKPCE EQYGSTYRCVSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAKGQPREPQ VYTLPPSREEMTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGS FFLYSKLTVDKSRWQQGNVFSCSVMHE ALHNHYTQKSLSLSPGKGG GGSGGGGS GGGGSGGGGSGGGGSGGGGSDKTHTCP PCPAPELLGGPSVFLFPPKPKDTLMISRT PEVTCVWDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQ PENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 176. CDR1 VH MU 8- B7 CC x I2C0-scFc Artificial ahh GYYWS 177. CDR2 VH MU 8- B7 CC x I2C0-scFc Artificial ahh DIDASGSTKYNPSLKS 178. CDR3 VH MU 8- B7 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDN 179. CDR1 VL MU 8- B7 CC x I2C0-scFc Artificial ahh SGDKLGDKYAS 180. CDR2 VL MU 8- B7 CC x I2C0-scFc Artificial ahh QDRKRPS 181. CDR3 VL MU 8- B7 CC x I2C0-scFc Artificial ahh QAWGSSTAV 182. VH MU 8-B7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDNW GQGTLVTVSS 183. VL MU 8-B7 CC x Artificial ahh SYELTQPSSVSVPPGQTASITCSGDKLGD

- 104 046123

I2C0-scFc KYASWYQQKPGQSPVLVIYQDRKRPSG VPERFSGSNSGNTATLTISGTQAMDEAD YYCQAWGSSTAVFGCGTKLTVL 184. scFv MU 8-B7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDNW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIYQ DRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSTAVFGCGTKLTVL 185. Bispecific molecule MU 8- B7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDNW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIYQ DRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSTAVFGCGTKLTVLSGGGGS EVQLVESGGGLVQPGGSLKLSCAASGFT FNKYAMNWVRQAPGKGLEWVARIRSK YNNYATYYADSVKDRFTISRDDSKNTA YLQMNNLKTEDTAVYYCVRHGNFGNS YISYWAYWGQGTLVTVSSGGGGG GG GSGGGGSQTVVTQEPSLTVSPGGTVTLT CGSSTGAVTSGNYPNWVQQKPGQAPRG LIGGTKFLAPGTPARFSGSLLGGKAALT LSGVQPEDEAEYYCVLWYSNRWVFGG GTKLTVL 186. Bispecific molecule with HLE MU 8-B7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDASG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDNW

- 105 046123

GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASFTCSGDKLGDK YASWYQQKPGQSPVLVIYQDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSTAVFGCGTKLTVLSGGGGS EVQLVESGGGLVQPGGSLKLSCAASGFT FNKYAMNWVRQAPGKGLEWVARIRSK YNNYATYYADSVKDRFTISRDDSKNTA YLQMNNLKTEDTAVYYCVRHGNFGNS YISYWAYWGQGTLVTVSSGGGGSGGG GSGGGGSQTVVTQEPSLTVSPGGTVTLT CGSSTGAVTSGNYPNWVQQKPGQAPRG LIGGTKFLAPGTPARFSGSLLGGKAALT LSGVQPEDEAEYYCVLWYSNRWVFGG GTKLTVL GGGGDKTHTCPPCPAPELLGG PSVFLFPPKPKDTLMISRTPEVTCWVDV SHEDPEVKFNWYVDGVEVHNAKTKPCE EQYGSTYRCVSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAKGQPREPQ VYTLPPSREEMTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGS FFLY SKLTVDKSRWQQGNVFSCSVMHE ALHNHYTQKSLSLSPGKGGGGSGGGGS GGGGSGGGGSGGGGSGGGGSDKTHTCP PCPAPELLGGPSVFLFPPKPKDTLMISRT PEVTCVWDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKG QPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 187. CDR1 VH MU 8- A7 CC x I2C0-scFc Artificial ahh GYYWS 188. CDR2 VH MU 8- Artificial ahh DIDQSGSTKYNPSLKS

- 106 046123

A7 CC x I2C0-scFc 189. CDR3 VH MU 8- A7 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDY 190. CDR1 VL MU 8- A7 CC x I2C0-scFc Artificial ahh SGDKLGDKYAS 191. CDR2 VL MU 8- A7 CC x I2C0-scFc Artificial ahh QDRKRPS 192. CDR3 VL MU 8- A7 CC x I2C0-scFc Artificial ahh QAWGSSTAV 193. VH MU 8-A7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSS 194. VL MU 8-A7 CC x I2C0-scFc Artificial ahh SYELTQPSSVSVPPGQTASHCSGDKLGD KYASWYQQKPGQSPVLVIYQDRKRPSG VPERFSGSNSGNTATLTISGTQAMDEAD YYCQAWGSSTAVFGCGTKLTVL 195. scFv MU 8-A7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIY QDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSTAVFGCGTKLTVL 196. Bispecific molecule MU 8A7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIY QDRKRPSGV

- 107 046123

PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSTAVFGCGTKLTVLSGGGGS EVQLVESGGGLVQPGGSLKLSCAASGFT FNKYAMNWVRQAPGKGLEWVARIRSK YNNYATYYADSVKDRFTISRDDSKNTA YLQMNNLKTEDTAVYYCVRHGNFGNS YISYWAYWGQGTLVTVSSGGGGSGGG GSGGG GSQTVVTQEPSLTVSPGGTVTLT CGSSTGAVTSGNYPNWVQQKPGQAPRG LIGGTKFLAPGTPARFSGSLLGGKAALT LSGVQPEDEAEYYCVLWYSNRWVFGG GTKLTVL 197. Bispecific molecule with HLE MU 8-A7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIY QDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSTAVFGCGTKLTVLSGGGGS EVQLVESGGGLVQPGGSLKLSCAASGFT FNKYAMNWVRQAPGKGLEWVARIRSK YNNYATYYADSVKDRFTISRDDSKNTA YLQMNNLKTEDTAVYYCVRHGNFGNS YISYWAYWGQGTLVTVSSGG SGGG GSGGGGSQTVVTQEPSLTVSPGGTVTLT CGSSTGAVTSGNYPNWVQQKPGQAPRG LIGGTKFLAPGTPARFSGSLLGGKAALT LSGVQPEDEAEYYCVLWYSNRWVFGG GTKLTVLGGGGDKTHTCPPCPAPELLGG PSVFLFPPKPKDTLMISRTPEVTCWVDV SHEDPEVKFNWYVDGVEVHNAKTKPCE EQYGSTYRCVSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAKGQPREPQ

- 108 046123

VYTLPPSREEMTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGS FFLYSKLTVDKSRWQGNVFSCSVMHE ALHNHYTQKSLSLSPGKGGGGSGGGGS GGGGSGGGGSGGGGSGGGGSDKTHTCP PCPAPELLGGPSVFLFPPKPKDTLMISRT PEVTCVWDVSHEDPEVKFNWYVDGVE VH NAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 198. CDR1 VH MU 7- G6 CC x I2C0-scFc Artificial ahh SYGMH 199. CDR2 VH MU 7- G6 CC x I2C0-scFc Artificial ahh VIWYSGSNKYYATSVKG 200. CDR3 VH MU 7- G6 CC x I2C0-scFc Artificial ahh GAYTYGFDY 201. CDR1 VL MU 7- G6 CC x I2C0-scFc Artificial ahh RASQSINRYLA 202. CDR2 VL MU 7- G6 CC x I2C0-scFc Artificial ahh TASNRAT 203. CDR3 VL MU 7- G6 CC x I2C0-scFc Artificial ahh HHYGSSIFT 204. VH MU 7-G6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGGSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY SGSNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSS 205. VL MU 7-G6 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK

- 109 046123

206. scFv MU 7-G6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGGSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY SGSNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSGTD FTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIK 207. Bispecific molecule MU 7G6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGGSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY SGSNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSGTD FTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTVSP GGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVL 208. Bispecific molecule with HLE MU 7-G6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGGSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY SGSNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP

- 110 046123

DRFSGSGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSSGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQ TVVTQEPSLTVSPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTCPPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRC VSVLTVLHQDWLNGKEYKC KVSNKALPAPIEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTCLVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGGGGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSV FLFPPKPKDTLMISRTPE VTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPCEEQYGSTYRCVSVLTVLHQ DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTC LVKGFYPSDIAVEWESNGQPENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGK 209. CDR1 VH MU 6- B12 CC x I2C0scFc Artificial ahh SYGMH 210. CDR2 VH MU 6- B12 CC x I2C0scFc Artificial ahh VIWFDASNKYYAESVKG

- 111 046123

211. CDR3 VH MU 6- B12 CC x I2C0scFc Artificial ahh GAYTYGFDY 212. CDR1 VL MU 6- B12 CC x I2C0scFc Artificial ahh RASQSINRYLA 213. CDR2 VL MU 6- B12 CC x I2C0scFc Artificial ahh TASNRAT 214. CDR3 VL MU 6B12 CC x I2C0scFc Artificial ahh HHYGSSIFT 215. VH MU 6-B12 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSS 216. VL MU 6-B12 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 217. scFv MU 6-B12 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSG TDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIK 218. Bispecific molecule MU 6B12 CC x I2C0scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY

- 112 046123

WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYY ADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTVSPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVL 219. Bispecific molecule with HLE MU 6-B12 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSG TDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTV SPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTCPPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE

- 113 046123

DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRCVSVLTVLHQDWLNGKEYKC KVSNKALPAPIEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTCLVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGG GGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSVFLFPPKPKDTLMISRTPE VTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPCEEQYGSTYRCVSVLTVLHQ DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTC LVKGFYPSDIAVEWESNGQ PENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGK 220. CDR1 VH MU 5- H4 CC x I2C0-scFc Artificial ahh SYGMH 221. CDR2 VH MU 5- H4 CC x I2C0-scFc Artificial ahh VIWFQGSNKYYADAVKG 222. CDR3 VH MU 5- H4 CC x I2C0-scFc Artificial ahh GGYTYGFDY 223. CDR1 VL MU 5- H4 CC x I2C0-scFc Artificial ahh RASQSINRYLA 224. CDR2 VL MU 5- H4 CC x I2C0-scFc Artificial ahh TASNRAT 225. CDR3 VL MU 5- H4 CC x I2C0-scFc Artificial ahh HHYGSSIFT 226. VH MU 5-H4 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVQPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFQ GSNKYYADAVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSS 227. VL MU 5-H4 CC x Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN

- 114 046123

I2C0-scFc RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 228. scFv MU 5-H4 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVQPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFQ GSNKYYADAVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSLPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSG SGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIK 229. Bispecific molecule MU 5H4 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVQPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFQ GSNKYYADAVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSLPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSG SGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSL TVSPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVL 230. Bispecific molecule with HLE MU 5-H4 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVQPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFQ GSNKYYADAVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY

- 115 046123

WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYY ADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTVSPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTC PPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRCVSVLTVLHQDWLNGKEYKC KVSNKALPAPIEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTCLVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTV DKSRWQQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGGGGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSVFLFPPKPKDTLMISRTPE VTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPCEEQYGSTYRCVSVLTVLHQ DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQ VYTLPPSREEMTKNQVSLTC LVKGFYPSDIAVEWESNGQPENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGK 231. CDR1 VH MU 5- H1 CC x I2C0-scFc Artificial aa SGGYNWA 232. CDR2 VH MU 5- Artificial aa YIYYSGSTYYNPSLKS

- 116 046123

Hl CC X I2C0-scFc 233. CDR3 VH MU 5- H1 CC x I2C0-scFc Artificial ahh EKYSSRWTFFDY 234. CDR1 VL MU 5- H1 CC x I2C0-scFc Artificial ahh SGDKLGDNYAS 235. CDR2 VL MU 5- H1 CC x I2C0-scFc Artificial ahh HDNKRPS 236. CDR3 VL MU 5- H1 CC x I2C0-scFc Artificial ahh QAFQSSTVV 237. VH MU 5-H1 CC x I2C0-scFc Artificial ahh QVQLQESGPGLVKPSETLSLTCTVSGDSI SSGGYNWAWIRQHPGKCLEWIGYIYYS GSTYYNPSLKSRVTISVDTSKNQFSLKLS SVTAADTAVYYCAREKYSSRWTFFDY WGQGTLVTVSS 238. VL MU 5-H1 CC x I2C0-scFc Artificial ahh S YELTQPPSVSVSPGQTASITCSGDKLGD NYASWYQQKPGQSPVLVIYHDNKRPSG IPERFSGSNSGNTATLTISGTQAMDEAD YYCQAFQSSTVVFGCGTKLTVL 239. scFv MU 5-H1 CC x I2C0-scFc Artificial ahh QVQLQESGPGLVKPSETLSLTCTVSGDSI SSGGYNWAWIRQHPGKCLEWIGYIYYS GSTYYNPSLKSRVTISVDTSKNQFSLKLS SVTAADTAVYYCAREKYSSRWTFFDY WGQGTLVTVSSGGGGSGGGGSGGGGSS YELTQPPSVSVSPGQTASITCSGDKLGD NYASWYQQKPGQSPVLVI YHDNKRPSG IPERFSGSNSGNTATLTISGTQAMDEAD YYCQAFQSSTVVFGCGTKLTVL 240. Bispecific molecule MU 5H1 CC X I2C0-scFc Artificial ahh QVQLQESGPGLVKPSETLSLTCTVSGDSI SSGGYNWAWIRQHPGKCLEWIGYIYYS GSTYYNPSLKSRVTISVDTSKNQFSLKLS SVTAADTAVYYCAREKYSSRWTFFDY WGQGTLVTVSSGGGGSGGGGSGGGGSS YELTQPPSVSVSPGQTASITCSGDKLGD NYASWYQQKPGQSPVLVI YHDNKRPSG

- 117 046123

G SGGGGSQTVVTQEPSLTVSPGGTVTLT CGSSTGAVTSGNYPNWVQQKPGQAPRG LIGGTKFLAPGTPARFSGSLLGGKAALT LSGVQPEDEAEYYCVLWYSNRWVFGG GTKLTVL 241. Bispecific molecule with HLE MU 5-H1 CC x I2C0-scFc Artificial ahh QVQLQESGPGLVKPSETLSLTCTVSGDSI SSGGYNWAWIRQHPGKCLEWIGYIYYS GSTYYNPSLKSRVTISVDTSKNQFSLKLS SVTAADTAVYYCAREKYSSRWTFFDY WGQGTLVTVSSGGGGSGGGGSGGGGSS YELTQPPSVSVSPGQTASITCSGDKLGD NYASWYQQKPGQSPVLVI YHDNKRPSG IPERFSGSNSGNTATLTISGTQAMDEAD YYCQAFQSSTVVFGCGTKLTVLSGGGG SEVQLVESGGGLVQPGGSLKLSCAASGF TFNKYAMNWVRQAPGKGLEWVARIRS KYNNYATYYADSVKDRFTISRDDSKNT AYLQMNNLKTEDTAVYYCVRHGNFGN SYISYWAYWGQGTLVTVSSGG GGSGGG GSGGGGSQTVVTQEPSLTVSPGGTVTLT CGSSTGAVTSGNYPNWVQQKPGQAPRG LIGGTKFLAPGTPARFSGSLLGGKAALT LSGVQPEDEAEYYCVLWYSNRWVFGG GTKLTVLGGGGDKTHTCPPCPAPELLGG PSVFLFPPKPKDTLMISRTPEVTCWVDV SHEDPEVKFNWYVDGVEVHNAKTKP CE EQYGSTYRCVSVLTVLHQDWLNGKEY KCKVSNKALPAPIEKTISKAKGQPREPQ

- 118 046123

VYTLPPSREEMTKNQVSLTCLVKGFYPS DIAVEWESNGQPENNYKTTPPVLDSDGS FFLYSKLTVDKSRWQGNVFSCSVMHE ALHNHYTQKSLSLSPGKGGGGSGGGGS GGGGSGGGGSGGGGSGGGGSDKTHTCP PCPAPELLGGPSVFLFPPKPKDTLMISRT PEVTCVWDVSHEDPEVKFNWYVDGVE VH NAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 242. CDR1 VH MU 4- Hll CC x I2C0scFc Artificial ahh NFGMH 243. CDR2 VH MU 4- Hll CC x I2C0scFc Artificial ahh VIWFDASKTYYAEAVKG 244. CDR3 VH MU 4Hll CC x I2C0scFc Artificial ahh ATYSTGWSYFDY 245. CDR1 VL MU 4Hll CC x I2C0scFc Artificial ahh SGDKLGDKYTS 246. CDR2 VL MU 4Hll CC x I2C0scFc Artificial ahh HDAKRPS 247. CDR3 VL MU 4Hll CC x I2C0scFc Artificial ahh QAYEASTAWV 248. VHMU4-H11 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAEAVKGRFTISRDTSMNTLYL

- 119 046123

QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSS 249. VL MU 4-H11 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASITCSGDKLGD KYTSWYQQKPGQSPVLVIYHDAKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAYEASTAWVFGCGTKLTVL 250. scFv MU 4-H11 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAEAVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAK R PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAYEASTAWVFGCGTKLTVL 251. Bispecific molecule MU 4Hll CC X I2C0scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAEAVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAK R PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAYEASTAWVFGCGTKLTVLSG GGGSEVQLVESGGGLVQPGGSLKLSCA ASGFTFNKYAMNWVRQAPGKGLEWVA RIRSKYNNYATYYADSVKDRFTISRDDS KNTAYLQMNNLKTEDTAVYYCVRHGN FGNSYISYWAYWGQGTLVTVSSGGGGS GGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVL 252. Bispecific Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF

- 120 046123

molecule with HLE MU 4-NP SS x I2C0-scFc TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAEAVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASrrCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAM DE ADYYCQAYEASTAWVFGCGTKLTVLSG GGGSEVQLVESGGGLVQPGGSLKLSCA ASGFTFNKYAMNWVRQAPGKGLEWVA RIRSKYNNYATYYADSVKDRFTISRDDS KNTAYLQMNNLKTEDTAVYYCVRHGN FGNSYISYWAYWGQGTLVTVSSGGGGS GGGGSGGGGSQTVVTQEPSLTVSPGGT V TLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP ELLGGPSVFLFPPKPKDTLMISRTPEVTC WVDVSHEDPEVKFNWYVDGVEVHNA KTKPCEEQYGSTYRCVSVLTVLHQDWL NGKEYKCKVSNKALPAPIEKTISKAKGQ PREPQVYTLPPSREEMTKNQVSLTCLVK GFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQGNVFSC SVMHEALHNHYTQKSLSLSPGKGGGGS GGGGSGGGGSGGGGSGGGGSGGGGSD KTHTCPPCPAPELLGGPSVFLFPPKPKDT LMI SRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS VLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLPPSREEMT KNQVSLTCLVKGFYPSDIAVEWESNGQP ENNYKTTPPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNHYT QKSL

- 121 046123

SLSPGK 253. CDR1 VH MU 4- H2 CC x I2C0-scFc Artificial ahh NFGMH 254. CDR2 VH MU 4- H2 CC x I2C0-scFc Artificial ahh VIWFDASKTYYAESVKG 255. CDR3 VH MU 4- H2 CC x I2C0-scFc Artificial ahh ATYSTGWSYFDY 256. CDR1 VL MU 4- H2 CC x I2C0-scFc Artificial ahh SGDKLGDKYTS 257. CDR2 VL MU 4- H2 CC x I2C0-scFc Artificial ahh HDAKRPS 258. CDR3 VL MU 4- H2 CC x I2C0-scFc Artificial ahh QAWEASTAWV 259. VH MU 4-H2 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAESVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSS 260. VL MU 4-H2 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASHCSGDKLGD KYTSWYQQKPGQSPVLVIYHDAKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAWEASTAWVFGCGTKLTVL 261. scFv MU 4-H2 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAESVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAK R PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWEASTAWVFGCGTKLTVL 262. Bispecific molecule MU 4H2 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAESVKGRFTISRDTSMNTLYL

- 122 046123

QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASrrCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWEASTAWVFGCGTKLTVLS GGGGSEVQLVESGGGLVQPGGSLKLS C AASGFTFNKYAMNWVRQAPGKGLEWV ARIRSKYNNYATYYADSVKDRFTISRDD SKNTAYLQMNNLKTEDTAVYYCVRHG NFGNSYISYWAYWGQGTLVTVSSGGGG SGGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTL SGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVL 263. Bispecific molecule with HLE MU 4-H2 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAESVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASrrCSGDK LGDKYTSWYQQKPGQSPVLVIYHDA KR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWEASTAWVFGCGTKLTVLS GGGGSEVQLVESGGGLVQPGGSLKLSC AASGFTFNKYAMNWVRQAPGKGLEWV ARIRSKYNNYATYYADSVKDRFTISRDD SKNTAYLQMNNLKTEDTAVYYCVRHG NFGNSYISYWAYWGQGTLVTVSSGG SGGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP

- 123 046123

ELLGGPSVFLFPPKPKDTLMISRTPEVTC WVDVSHEDPEVKFNWYVDGVEVHNA KTKPCEEQYGSTYRCVSVLTVLHQDWL NGKEYKCKVSNKALPAPIEKTISKAKGQ PREPQVYTLPPSREEMTKNQVSLTCLVK GFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSR WQQGNVFSC SVMHEALHNHYTQKSLSLSPGKGGGGS GGGGSGGGGSGGGGSGGGGSGGGGSD KTHTCPPCPAPELLGGPSVFLFPPKPKDT LMISRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS VLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQV YTLPPSREEMT KNQVSLTCLVKGFYPSDIAVEWESNGQP ENNYKTTPPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNHYTQKSL SLSPGK 264. CDR1 VH MU 4- G4 CC x I2C0-scFc Artificial ahh NFGMH 265. CDR2 VH MU 4- G4 CC x I2C0-scFc Artificial ahh VIWFDASKTYYADAVKG 266. CDR3 VH MU 4- G4 CC x I2C0-scFc Artificial ahh ATYSTGWSYFDY 267. CDR1 VL MU 4- G4 CC x I2C0-scFc Artificial ahh SGDKLGDKYTS 268. CDR2 VL MU 4- G4 CC x I2C0-scFc Artificial ahh HDAKRPS 269. CDR3 VL MU 4- G4 CC x I2C0-scFc Artificial ahh QAWDASTAWV 270. VH MU 4-G4 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYADAVKGRFTISRDTSMNTLY LQMNSLRAEDTAVYYCARATYSTGWS

- 124 046123

YFDYWGQGTLVTVSS 271. VL MU 4-G4 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASITCSGDKLGD KYTSWYQQKPGQSPVLVIYHDAKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAWDASTAWVFGCGTKLTVL 272. scFv MU 4-G4 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYADAVKGRFTISRDTSMNTLY LQMNSLRAEDTAVYYCARATYSTGWS YFDYWGQGTLVTVSSGGGGSGGGGSG GGGSSYELTQPPSVSVSPGQTASITCSGD KLGDKYTSWYQQKPGQSPVLVIYHDAK R PSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAWDASTAWVFGCGTKLTVL 273. Bispecific molecule MU 4G4 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYADAVKGRFTISRDTSMNTLY LQMNSLRAEDTAVYYCARATYSTGWS YFDYWGQGTLVTVSSGGGGSGGGGSG GGGSSYELTQPPSVSVSPGQTASITCSGD KLGDKYTSWYQQKPGQSPVLVIYHDAK R PSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAWDASTAWVFGCGTKLTVL SGGGGSEVQLVESGGGLVQPGGSLKLS CAASGFTFNKYAMNWVRQAPGKGLEW VARIRSKYNNYATYYADSVKDRFTISRD DSKNTAYLQMNNLKTEDTAVYYCVRH GNFGNSYISYWAYWGQGTLVTVSSGGG GS GGGGSGGGGSQTVVTQEPSLTVSPG GTVTLTCGSSTGAVTSGNYPNWVQQKP GQAPRGLIGGTKFLAPGTPARFSGSLLG GKAALTLSGVQPEDEAEYYCVLWYSNR WVFGGGTKLTVL 274. Bispecific molecule with HLE Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF

- 125 046123

MU 4-G4 CC X I2C0-scFc DASKTYYADAVKGRFTISRDTSMNTLY LQMNSLRAEDTAVYYCARATYSTGWS YFDYWGQGTLVTVSSGGGGSGGGGSG GGGSSYELTQPPSVSVSPGQTASITCSGD KLGDKYTSWYQQKPGQSPVLVIYHDAK RPSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAWDASTAWVFGCGTKLTV L SGGGGSEVQLVESGGGLVQPGGSLKLS CAASGFTFNKYAMNWVRQAPGKGLEW VARIRSKYNNYATYYADSVKDRFTISRD DSKNTAYLQMNNLKTEDTAVYYCVRH GNFGNSYISYWAYWGQGTLVTVSSGGG GSGGGGSGGGGSQTVVTQEPSLTVSPG GTVTLTCGSSTGAVTSGNYPNWVQQKP GQAP RGLIGGTKFLAPGTPARFSGSLLG GKAALTLSGVQPEDEAEYYCVLWYSNR WVFGGGTKLTVLGGGGDKTHTCPPCPA PELLGGPSVFLFPPKPKDTLMISRTPEVT CVWDVSHEDPEVKFNWYVDGVEVHN AKTKPCEEQYGSTYRCVSVLTVLHQDW LNGKEYKCKVSNKALPAPIEKTISKAK G QPREPQVYTLPPSREEMTKNQVSLTCLV KGFYPSDIAVEWESNGQPENNYKTTPPV LDSDGSFFLYSKLTVDKSRWQQGNVFS CSVMHEALHNHYTQKSLSLSPGKGGGG SGGGGSGGGGSGGGGSGGGGSGGGGS DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYV DGVEVHNAKTKPCEEQYGSTYRC VSVLTVLHQDWLNGKEYKCKVSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGK

- 126 046123

275. CDR1 VH MU 4- F6 CC x I2C0-scFc Artificial ahh NFGMH 276. CDR2 VH MU 4- F6 CC x I2C0-scFc Artificial ahh VIWFDASKTYYASSVKG 277. CDR3 VH MU 4- F6 CC x I2C0-scFc Artificial ahh ATYSTGWSYFDY 278. CDR1 VL MU 4- F6 CC x I2C0-scFc Artificial ahh SGDKLGDKYTS 279. CDR2 VL MU 4- F6 CC x I2C0-scFc Artificial ahh HDAKRPS 280. CDR3 VL MU 4- F6 CC x I2C0-scFc Artificial ahh QAYSASTAWV 281. VH MU 4-F6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSS 282. VL MU 4-F6 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASrrCSGDKLGD KYTSWYQQKPGQSPVLVIYHDAKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAYSASTAWVFGCGTKLTVL 283. scFv MU 4-F6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAYSASTAWVFGCGTKLTVL 284. Bispecific molecule MU 4-F6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY

- 127 046123

FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASETCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAYSASTAWVFGCGTKLTVLSG GGGSEVQLVESGGGLVQPGGSLKLSCA ASGFTFNKYAMNWVRQAPGKGLEW VF GGGTKLTVL 285. Bispecific molecule with HLE MU 4-F6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAYSASTAWVFGCGTKLTVLSG GGGSEVQLVESGGGLVQPGGSLKLSCA ASGFTFNKYAMNWVRQAPGKGLEWVA RIRSKYNNYATYYADSVKDRFTISRDDS KNTAYLQMNNLKTEDTAVYYCVRHGN FGNSYISYWAYWGQGTGGTVSSGGGGS GGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP ELLGGPSVFLFPPKPKDTLMISRTPEVTC

- 128 046123

WVDVSHEDPEVKFNWYVDGVEVHNA KTKPCEEQYGSTYRCVSVLTVLHQDWL NGKEYKCKVSNKALPAPIEKTISKAKGQ PREPQVYTLPPSREEMTKNQVSLTCLVK GFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQGNVFSC SVMHEALHNHYTQKS LSLSPGKGGGGS GGGGSGGGGSGGGGSGGGGSGGGGSD KTHTCPPCPAPELLGGPSVFLFPPKPKDT LMISRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS VLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLPPSREEMT KNQVSLTCLVKGFYPSDI AVEWESNGQP ENNYKTTPPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNHYTQKSL SLSPGK 286. CDR1 VH MU 4- E7 CC x I2C0-scFc Artificial ahh GYYWS 287. CDR2 VH MU 4- E7 CC x I2C0-scFc Artificial ahh DIDYSGSTKYNPSLKS 288. CDR3 VH MU 4- E7 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDY 289. CDR1 VL MU 4- E7 CC x I2C0-scFc Artificial ahh SGDKLGEKYAS 290. CDR2 VL MU 4- E7 CC x I2C0-scFc Artificial ahh QDRKRPS 291. CDR3 VL MU 4- E7 CC x I2C0-scFc Artificial ahh QAWGSSAAV 292. VH MU 4-E7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDYSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSS

- 129 046123

293. VL MU 4-E7 CC x I2C0-scFc Artificial ahh SYELTQPSSVSVSPGQTASTTCSGDKLGE KYASWYQQKPGQSPVLIIYQDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSAAVFGCGTKLTVL 294. scFv MU 4-E7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDYSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVSPGQTASITCSGDKLGEKY ASWYQQKPGQSPVLIIYQ DRKRPSGVPE RFSGSNSGNTATLTISGTQAMDEADYYC QAWGSSAAVFGCGTKLTVL 295. Bispecific molecule MU 4E7 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDYSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVSPGQTASITCSGDKLGEKY ASWYQQKPGQSPVLIIYQ DRKRPSGVPE RFSGSNSGNTATLTISGTQAMDEADYYC QAWGSSAAVFGCGTKLTVLSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGG GGSG GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL 296. Bispecific molecule c HLE MU 4-E7 CC x Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDYSG STKYNPSLKSRVTISLDTSKNQFSLKLNS

- 130 046123

I2C0-scFc VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVSPGQTASITCSGDKLGEKY ASWYQQKPGQSPVLIIYQDRKRPSGVPE RFSGSNSGNTATLTISGTQAMDEADYYC QAWGSSAAVFGCGTKLTVLSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTF N KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYY IAVEWES NGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELLGGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWL NGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 297. CDR1 VH MU 4- Cll CC x I2C0- Artificial aa SYGMH

- 131 046123

scFc 298. CDR2 VH MU 4Cll CC x I2C0scFc Artificial ahh VISYDASNKYYASAVKG 299. CDR3 VH MU 4- Cll CC x I2C0scFc Artificial ahh GAYTYGFDY 300. CDR1 VL MU 4- Cll CC x I2C0scFc Artificial ahh RASQSVNRYLA 301. CDR2 VL MU 4Cll CC x I2C0scFc Artificial ahh GASNRAT 302. CDR3 VL MU 4Cll CC x I2C0scFc Artificial ahh HHYGSSIFA 303. VHMU4-C11 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYD ASNKYYASAVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSS 304. VL MU 4-C11 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSV NRYLAWYQQKPGQAPRLLIYGASNRAT GIPDRFTGSGSGTDFTLTISRLEPEDFAV YFCHHYGSSIFAFGCGTKVEIK 305. scFv MU 4-C11 CC X I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYD ASNKYYASAVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTGSG SGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIK

- 132 046123

306. Bispecific molecule MU 4SP CC X I2C0scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYD ASNKYYASAVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTGSG SGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIKSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSL TVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL 307. Bispecific molecule with HLE MU 4-SP CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYD ASNKYYASAVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTGSG SGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIKSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSL TVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG

- 133 046123

GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTYRCVSVLTVLHQDWLNGKEYK CKVSNKALPAPIEKTISKAKGQPREPQ V YTLPPSREEMTKNQVSLTCLVKGFYPSD IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELLGGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 308. CDR1 VH MU 4- C4 CC x I2C0-scFc Artificial ahh SYGMH 309. CDR2 VH MU 4- C4 CC x I2C0-scFc Artificial ahh VIWFDASNKYYAESVKG 310. CDR3 VH MU 4- C4 CC x I2C0-scFc Artificial ahh GAYTYGFDY 311. CDR1 VL MU 4- C4 CC x I2C0-scFc Artificial ahh RASQSVNRYLA 312. CDR2 VL MU 4- C4 CC x I2C0-scFc Artificial ahh GASNRAT 313. CDR3 VL MU 4- C4 CC x I2C0-scFc Artificial ahh HHYGSSIFA 314. VH MU 4-C4 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF

- 134 046123

DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSS 315. VL MU 4-C4 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSV NRYLAWYQQKPGQAPRLLIYGASNRAT GIPDRFTGSGSGTDFTLTISRLEPEDFAV YFCHHYGSSIFAFGCGTKVEIK 316. scFv MU 4-C4 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTGSG SGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIK 317. Bispecific molecule MU 4- C4 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTGSG SGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIKSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSL TVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL

- 135 046123

318. Bispecific molecule with HLE MU 4-C4 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTGSG SGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIKSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSL TVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTYRCVSVLTVLHQDW LNGKEYK CKVSNKALPAPIEKTISKAKGQPREPQV YTLPPSREEMTKNQVSLTCLVKGFYPSD IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELLGGPSVFLFPPKPKD TLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG

- 136 046123

NVFSCSVMHEALHNHYTQKSLSLSPGK 319. CDR1 VH MU 4- C3 CC x I2C0-scFc Artificial ahh SYGMH 320. CDR2 VH MU 4- C3 CC x I2C0-scFc Artificial ahh VISYEGSNKYYAESVKG 321. CDR3 VH MU 4- C3 CC x I2C0-scFc Artificial ahh GAYTYGFDY 322. CDR1 VL MU 4- C3 CC x I2C0-scFc Artificial ahh RASQSVNRYLA 323. CDR2 VL MU 4- C3 CC x I2C0-scFc Artificial ahh GASNRAT 324. CDR3 VL MU 4- C3 CC x I2C0-scFc Artificial ahh HHYGSSIFA 325. VH MU 4-C3 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYE GSNKYYAESVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSS 326. VL MU 4-C3 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSV NRYLAWYQQKPGQAPRLLIYGASNRAT GIPDRFTGSGSGTDFTLTISRLEPEDFAV YFCHHYGSSIFAFGCGTKVEIK 327. MU 4-C3 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYE GSNKYYAESVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRF TGSGSGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIK 328. Bispecific molecule MU 4C3 CC X I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYE GSNKYYAESVKGRFTISRDNSKNTLYLQ

- 137 046123

MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTGSGSGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIKSSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMN WVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYS NRWVFGGGTK LTVL 329. Bispecific molecule with HLE MU 4-СЗ СС x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYE GSNKYYAESVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRF TGSGSGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIKSGGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQ EPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV

- 138 046123

FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTYRCVSVLTVLHQDWLNGKEYK CKVSNKALPAPIEKTISKAKGQPREPQV YTLPPSREEMTKNQVSLTCLVKGFYPSD IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQQGNVFS CSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELLGGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREE MTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 330. CDR1 VH MU 4- B10 CC x I2C0scFc Artificial ahh NFGMH 331. CDR2 VH MU 4- B10 CC x I2C0scFc Artificial ahh VIWFDASKTYYASSVKG 332. CDR3 VH MU 4- B10 CC x I2C0scFc Artificial ahh ATYSTGWSYFDY 333. CDR1 VL MU 4B10 CC x I2C0scFc Artificial ahh SGDKLGDKYTS 334. CDR2 VL MU 4- B10 CC x I2C0scFc Artificial ahh HDAKRPS 335. CDR3 VL MU 4B10 CC x I2C0scFc Artificial ahh QAWSASTAWV

- 139 046123

336. VHMU4-B10 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSS 337. VL MU 4-B10 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASITCSGDKLGD KYTSWYQQKPGQSPVLVIYHDAKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAWSASTAWVFGCGTKLTVL 338. scFv MU 4-B10 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWSASTAWVFGCGTKLTVL 339. Bispecific molecule MU 4B10 CC X I2C0scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWSASTAWVFGCGTKLTVLS GGGGSEVQLVESGGGLVQPGGSLKLSC AASGFTFNKYAMNWVRQAPGKGLEWV ARIRSKYNNYATYYADSVKDRFTISRDD SKNTAYLQMNNLKTEDTAVYYCVRHG NFGNSYISYWAYWGQGTLVTVSSGGGG SGG GGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK

- 140 046123

AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVL 340. Bispecific molecule with HLE MU 4-B10 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYASSVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWSASTAWVFGCGTKLTVLS GGGGSEVQLVESGGGLVQPGGSLKLSC AASGFTFNKYAMNWVRQAPGKGLEWV ARIRSKYNNYATYYADSVKDRFTISRDD SKNTAYLQMNNLKTEDTAVYYCVRHG NFGNSYISYWAYWGQGTLVTVSSGGGG SGG GGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP ELLGGPSVFLFPPKPKDTLMISRTPEVTC WVDVSHEDPEVKFNWYVDGVEVHNA KTKPCEE QYGSTYRCVSVLTVLHQDWL NGKEYKCKVSNKALPAPIEKTISKAKGQ PREPQVYTLPPSREEMTKNQVSLTCLVK GFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQGNVFSC SVMHEALHNHYTQKSLSLSPGKGGGGS GGGGSGGGGSGGGGSGGGGSGGGGSD KTHTCPP CPAPELLGGPSVFLFPPKPKDT LMISRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS VLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLPPSREEMT

- 141 046123

KNQVSLTCLVKGFYPSDIAVEWESNGQP ENNYKTTPPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNHYTQKSL SLSPGK 341. CDR1 VH MU 4- B6 CC x I2C0-scFc Artificial ahh SYGMH 342. CDR2 VH MU 4- B6 CC x I2C0-scFc Artificial ahh VISYDASNKYYASSVKG 343. CDR3 VH MU 4- B6 CC x I2C0-scFc Artificial ahh GAYTYGFDY 344. CDR1 VL MU 4- B6 CC x I2C0-scFc Artificial ahh RASQSVNRYLA 345. CDR2 VL MU 4- B6 CC x I2C0-scFc Artificial ahh GASNRAT 346. CDR3 VL MU 4- B6 CC x I2C0-scFc Artificial ahh HHYGSSIFA 347. VH MU 4-B6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYD ASNKYYASSVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSS 348. VL MU 4-B6 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSV NRYLAWYQQKPGQAPRLLIYGASNRAT GIPDRFTGSGSGTDFTLTISRLEPEDFAV YFCHHYGSSIFAFGCGTKVEIK 349. scFv MU 4-B6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYD ASNKYYASSVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTGSG SGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIK

- 142 046123

350. Bispecific molecule MU 4- B6 CC X I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYD ASNKYYASSVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTGSG SGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIKSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSL TVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL 351. Bispecific molecule with HLE MU 4-B6 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVISYD ASNKYYASSVKGRFTISRDNSKNTLYLQ MNSLRAEDTAVYYCARGAYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSPGERATLSCRASQSVN RYLAWYQQKPGQAPRLLIYGASNRATG IPDRFTGSG SGTDFTLTISRLEPEDFAVYF CHHYGSSIFAFGCGTKVEIKSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQEPSL TVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG

- 143 046123

GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTYRCVSVLTVLHQDWLNGKEYK CKVSNKALPAPIEKTISKAKGQPREPQ V YTLPPSREEMTKNQVSLTCLVKGFYPSD IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELLGGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 352. CDR1 VH MU 4- B1 CC x I2C0-scFc Artificial ahh NFGMH 353. CDR2 VH MU 4- B1 CC x I2C0-scFc Artificial ahh VIWFDASKTYYAESVKG 354. CDR3 VH MU 4- B1 CC x I2C0-scFc Artificial ahh ATYSTGWSYFDY 355. CDR1 VL MU 4- B1 CC x I2C0-scFc Artificial ahh SGDKLGDKYTS 356. CDR2 VL MU 4- B1 CC x I2C0-scFc Artificial ahh HDAKRPS 357. CDR3 VL MU 4- B1 CC x I2C0-scFc Artificial ahh QAWSASTAWV 358. VHMU 4-BlCCx I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF

- 144 046123

DASKTYYAESVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSS 359. VLMU 4-B1 CCx I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASnCSGDKLGD KYTSWYQQKPGQSPVLVIYHDAKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAWSASTAWVFGCGTKLTVL 360. scFv MU 4-B1 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAESVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAK R PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWSASTAWVFGCGTKLTVL 361. Bispecific molecule MU 4- B1 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAESVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAK S GGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVL

- 145 046123

362. Bispecific molecule with HLE MU 4-B1 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAESVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWSASTAWVFGCGTKLTVLS GGGGSEVQLVESGGGLVQPGGSLKLSC AASGFTFNKYAMNWVRQAPGKGLEWV ARIRSKYNNYATYYADSVKDRFTISRDD SKNTAYLQMNNLKTEDTAVYYCVRHG NFGNSYISYWAYWGQGTLVTVSSGGGG SGGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP ELLGGPSVFLFPPKPKDTLMISRTPEVTC WVDVSHEDPEVKFNWYVDGVEVHNA KTKPCEEQYGSTYRCVSVLTVLHQDWL NGKEYKCKVSNKALPAPIEKTISKAKGQ PREPQVYTLPPSREEMTKNQVSLTCLVK GFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQGNVFSC SVMHEALHNHYTQKSLSLSPGKGGGGS GGGGSGGGGSGGGGSGGGGSGGGGSD KTHTCPPCPAPELLGGPSVFLFPPKPKDT LMISRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS VLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLPPSREEMT KNQVSLTCLVKGFYPSDIAVEWESNGQP ENNYKTTPPVLDSDGSFFLYSKLTVDKS

- 146 046123

RWQQGNVFSCSVMHEALHNHYTQKSL SLSPGK 363. CDR1 VH MU 4- A8 CC x I2C0-scFc Artificial ahh NFGMH 364. CDR2 VH MU 4- A8 CC x I2C0-scFc Artificial ahh VIWFDASKTYYADAVKG 365. CDR3 VH MU 4- A8 CC x I2C0-scFc Artificial ahh ATYSTGWSYFDY 366. CDR1 VL MU 4- A8 CC x I2C0-scFc Artificial ahh SGDKLGDKYTS 367. CDR2 VL MU 4- A8 CC x I2C0-scFc Artificial ahh HDAKRPS 368. CDR3 VL MU 4- A8 CC x I2C0-scFc Artificial ahh QAWSASTAWV 369. VH MU 4-A8 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYADAVKGRFTISRDTSMNTLY LQMNSLRAEDTAVYYCARATYSTGWS YFDYWGQGTLVTVSS 370. VL MU 4-A8 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASHCSGDKLGD KYTSWYQQKPGQSPVLVIYHDAKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAWSASTAWVFGCGTKLTVL 371. scFv MU 4-A8 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYADAVKGRFTISRDTSMNTLY LQMNSLRAEDTAVYYCARATYSTGWS YFDYWGQGTLVTVSSGGGGSGGGGSG GGGSSYELTQPPSVSVSPGQTASITCSGD KLGDKYTSWYQQKPGQSPVLVIYHDAK R PSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAWSASTAWVFGCGTKLTVL 372. Bispecific molecule MU 4- Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF

- 147 046123

A8 CC x I2C0-scFc DASKTYYADAVKGRFTISRDTSMNTLY LQMNSLRAEDTAVYYCARATYSTGWS YFDYWGQGTLVTVSSGGGGSGGGGSG GGGSSYELTQPPSVSVSPGQTASITCSGD KLGDKYTSWYQQKPGQSPVLVIYHDAK RPSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAWSASTAWVFGCGTKLTV L SGGGGSEVQLVESGGGLVQPGGSLKLS CAASGFTFNKYAMNWVRQAPGKGLEW VARIRSKYNNYATYYADSVKDRFTISRD DSKNTAYLQMNNLKTEDTAVYYCVRH GNFGNSYISYWAYWGQGTLVTVSSGGG GSGGGGSGGGGSQTVVTQEPSLTVSPG GTVTLTCGSSTGAVTSGNYPNWVQQKP GQAP RGLIGGTKFLAPGTPARFSGSLLG GKAALTLSGVQPEDEAEYYCVLWYSNR WVFGGGTKLTVL 373. Bispecific molecule with HLE MU 4-A8 CC x I2C0-scFc Artificial aa QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYADAVKGRFTISRDTSMNTLY LQMNSLRAEDTAVYYCARATYSTGWS YFDYWGQGTLVTVSSGGGGSGGGGSG GGGSSYELTQPPSVSVSPGQTASITCSGD KLGDKYTSWYQQKPGQSPVLVIYHDAK R PSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAWSASTAWVFGCGTKLTVL SGGGGSEVQLVESGGGLVQPGGSLKLS CAASGFTFNKYAMNWVRQAPGKGLEW VARIRSKYNNYATYYADSVKDRFTISRD DSKNTAYLQMNNLKTEDTAVYYCVRH GNFGNSYISYWAYWGQGTLVTVSSGGG GS GGGGSGGGGSQTVVTQEPSLTVSPG GTVTLTCGSSTGAVTSGNYPNWVQQKP GQAPRGLIGGTKFLAPGTPARFSGSLLG GKAALTLSGVQPEDEAEYYCVLWYSNR

- 148 046123

WVFGGGTKLTVLGGGGDKTHTCPPCPA PELLGGPSVFLFPPKPKDTLMISRTPEVT CVWDVSHEDPEVKFNWYVDGVEVHN AKTKPCEEQYGSTYRCVSVLTVLHQDW LNGKEYKCKVSNKALPAPIEKTISKAKG QPREPQVYTLPPSREEMTKNQVSLTCLV KGFYPSDIAVEWESNGQPEN NYKTTPPV LDSDGSFFLYSKLTVDKSRWQQGNVFS CSVMHEALHNHYTQKSLSLSPGKGGGG SGGGGSGGGGSGGGGSGGGGSGGGGS DKTHTCPPCPAPELLGGPSVFLFPPKPKD TLMISRTPEVTCVVVDVSHEDPEVKFN WYVDGVEVHNAKTKPCEEQYGSTYRC VSVLTVLHQDWLNGKEYKCK VSNKAL PAPIEKTISKAKGQPREPQVYTLPPSREE MTKNQVSLTCLVKGFYPSDIAVEWESN GQPENNYKTTPPVLDSDGSFFLYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQ KSLSLSPGK 374. CDR1 VH MU 3C10 CC x I2C0scFc Artificial ahh SYGMH 375. CDR2 VH MU 3C10 CC x Г2С0scFc Artificial ahh VIWYSGSNKYYATSVKG 376. CDR3 VH MU 3C10 CC x I2C0scFc Artificial ahh GGYTYGFDY 377. CDR1 VL MU 3C10 CC x I2C0scFc Artificial ahh RASQSINRYLA 378. CDR2 VL MU 3C10 CC x I2C0scFc Artificial ahh TASNRAT 379. CDR3 VL MU 3- Artificial ahh HHYGSSIFT

- 149 046123

CIO CC X I2C0scFc 380. VH MU 3-C10 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY SGSNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSS 381. VL MU 3-C10 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 382. scFv MU 3-C10 CC X I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY SGSNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSG TDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIK 383. Bispecific molecule MU 3C10 CC X I2C0scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY SGSNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSG TDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTV SPGGTVTLTCGSS

- 150 046123

TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVL 384. Bispecific molecule with HLE MU 3-C10 SS x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY SGSNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSG TDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTV SPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTCPPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRCVSVLTVLH QDWLNGKEYKC KVSNKALPAPIEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTCLVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGGGGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSVFLFPPKPK DTLMISRTPE VTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPCEEQYGSTYRCVSVLTVLHQ

- 151 046123

DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTC LVKGFYPSDIAVEWESNGQPENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGK 385. CDR1 VH MU 2- F7 CC x I2C0-scFc Artificial ahh SYGMH 386. CDR2 VH MU 2- F7 CC x I2C0-scFc Artificial ahh VIWFDASNKYYAESVKG 387. CDR3 VH MU 2- F7 CC x I2C0-scFc Artificial ahh GGYTYGFDY 388. CDR1 VL MU 2- F7 CC x I2C0-scFc Artificial ahh RASQSINRYLA 389. CDR2 VL MU 2- F7 CC x I2C0-scFc Artificial ahh TASNRAT 390. CDR3 VL MU 2- F7 CC x I2C0-scFc Artificial ahh HHYGSSIFT 391. VH MU 2-F7 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFD ASNKYYAESVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSS 392. VL MU 2-F7 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 393. scFv MU 2-F7 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFD ASNKYYAESVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSG SGSGTDFTLTISRLEPEDFAVYYC

- 152 046123

HHYGSSIFTFGCGTKVEIK 394. Bispecific molecule MU 2-F7 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFD ASNKYYAESVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSG SGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEP SLTVSPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVL 395. Bispecific molecule with HLE MU 2-F7 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFD ASNKYYAESVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSG SGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEP SLTVSPGGTVTLTCGSS

- 153 046123

TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTCPPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRCVSVLTVLHQDWLNGK EYKC KVSNKALPAPIEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTCLVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGGGGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSVFLFPPKPKDTLMISRTPE VTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPCEEQYGSTYRCVSVLTVLHQ DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTC LVKGFYPSDIAVEWESNGQPENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQ LSLSPGK 396. CDR1 VH MU 02- E7 CC x I2C0-scFc Artificial ahh SYGMH 397. CDR2 VH MU 02- E7 CC x I2C0-scFc Artificial ahh VIWYTGSNKYYAHSVKG 398. CDR3 VH MU 02- E7 CC x I2C0-scFc Artificial ahh GAYTYGFDY 399. CDR1 VL MU 02- E7 CC x I2C0-scFc Artificial ahh RASQSINRYLA 400. CDR2 VL MU 02- E7 CC x I2C0-scFc Artificial ahh TASNRAT 401. CDR3 VL MU 02- E7 CC x I2C0-scFc Artificial ahh HHYGSSIFT 402. VH MU 02-E7 CC Artificial ahh QVQLVESGGGVVQPGGSLRLSCAASGF

- 154 046123

x I2C0-scFc TFSSYGMHWVRQAPGKCLEWVAVIWY TGSNKYYAHSVKGRFAISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSS 403. VL MU 02-E7 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 404. scFv MU 02-E7 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGGSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY TGSNKYYAHSVKGRFAISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSGTD FTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIK 405. Bispecific molecule MU 02E7 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGGSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY TGSNKYYAHSVKGRFAISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSGTD FTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTVSP GGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL

- 155 046123

TVL 406. Bispecific molecule with HLE MU 02-E7 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGGSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY TGSNKYYAHSVKGRFAISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSGTD FTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTVSP GGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTCPPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRCVSVLTVLHQ DWLNGKEYKC KVSNKALPAPIEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTCLVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGGGGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSVFLFPPKPKDT LMISRTPE VTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPCEEQYGSTYRCVSVLTVLHQ DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTC LVKGFYPSDIAVEWESNGQPENNYKTTP

- 156 046123

PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGK 407. CDR1 VH MU 2Dll CC x I2C0scFc Artificial ahh NHGMH 408. CDR2 VH MU 2Dll CC x I2C0scFc Artificial ahh GIWSDASNKYYAEAVKG 409. CDR3 VH MU 2Dll CC x I2C0scFc Artificial ahh ATYTTGWSYFDY 410. CDR1 VL MU 2Dll CC x I2C0scFc Artificial ahh SGDKLGDKYTS 411. CDR2 VL MU 2Dll CC x I2C0scFc Artificial ahh HDRKRPS 412. CDR3 VL MU 2Dll CC x I2C0scFc Artificial ahh QAYDRSTAWV 413. VHMU2-D11 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNHGMHWVRQAPGKCLEWVAGIWS DASNKYYAEAVKGRFTISRDTSKNTLYL QMNSLRAEDTAVYYCARATYTTGWSY FDYWGQGTLVTVSS 414. VL MU 2-D11 CC x I2C0-scFc Artificial ahh SYELTQSPSVSVSPGQTASITCSGDKLGD KYTSWYQQKPGQSPVLVIYHDRKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAYDRSTAWVFGCGTKLTVL 415. scFv MU 2-D11 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNHGMHWVRQAPGKCLEWVAGIWS DASNKYYAEAVKGRFTISRDTSKNTLYL QMNSLRAEDTAVYYCARATYTTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG

- 157 046123

GGSSYELTQSPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDRKR PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAYDRSTAWVFGCGTKLTVL 416. Bispecific molecule MU 2Dll CC x I2C0scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNHGMHWVRQAPGKCLEWVAGIWS DASNKYYAEAVKGRFTISRDTSKNTLYL QMNSLRAEDTAVYYCARATYTTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQSPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDRK S GGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVL 417. Bispecific molecule with HLE MU 2-D11 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNHGMHWVRQAPGKCLEWVAGIWS DASNKYYAEAVKGRFTISRDTSKNTLYL QMNSLRAEDTAVYYCARATYTTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQSPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDRK R PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAYDRSTAWVFGCGTKLTVLS GGGGSEVQLVESGGGLVQPGGSLKLSC AASGFTFNKYAMNWVRQAPGKGLEWV ARIRSKYNNYATYYADSVKDRFTISRDD

- 158 046123

SKNTAYLQMNNLKTEDTAVYYCVRHG NFGNSYISYWAYWGQGTLVTVSSGGGG SGGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP ELL GGPSVFLFPPKPKDTLMISRTPEVTC WVDVSHEDPEVKFNWYVDGVEVHNA KTKPCEEQYGSTYRCVSVLTVLHQDWL NGKEYKCKVSNKALPAPIEKTISKAKGQ PREPQVYTLPPSREEMTKNQVSLTCLVK GFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQ GNVFSC SVMHEALHNHYTQKSLSLSPGKGGGGS GGGGSGGGGSGGGGSGGGGSGGGGSD KTHTCPPCPAPELLGGPSVFLFPPKPKDT LMISRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS VLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLP PSREEMT KNQVSLTCLVKGFYPSDIAVEWESNGQP ENNYKTTPPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNHYTQKSL SLSPGK 418. CDR1 VH MU 2- C2 CC x I2C0-scFc Artificial ahh NHGMH 419. CDR2 VH MU 2- C2 CC x I2C0-scFc Artificial ahh GIWSEGNSKYYADAVKG 420. CDR3 VH MU 2- C2 CC x I2C0-scFc Artificial ahh ATYTTGWSYFDY 421. CDR1 VL MU 2- C2 CC x I2C0-scFc Artificial ahh SGDKLGDKYAS 422. CDR2 VL MU 2- C2 CC x I2C0-scFc Artificial ahh QDAKRPS

- 159 046123

423. CDR3 VL MU 2- C2 CC x I2C0-scFc Artificial ahh QAFHQSTWV 424. VH MU 2-C2 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNHGMHWVRQAPGKCLEWVAGIWS EGSNKYYADAVKGRFTISRDNSKNTLY LQMNSLRAEDTAVYYCARATYTTGWS YFDYWGQGTLVTVSS 425. VL MU 2-C2 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASrrCSGDKLGD KYASWYQQKSGQSPVLVIYQDAKRPSG IPERFSGSNSGNTATLTISGTQAMDEAD YYCQAFHQSTWVFGCGTQLTVL 426. scFv MU 2-C2 CC x I2C0-scFc Artificial ahh R PSGIPERFSGSNSGNTATLTISGTQAMD EAD YYCQAFHQSTWVFGCGTQLTVL 427. Bispecific molecule MU 2C2 CC x I2C0-scFc Artificial ahh R PSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAFHQSTWVFGCGTQLTVLSG GGGSEVQLVESGGGLVQPGGSLKLSCA ASGFTFNKYAMNWVRQAPGKGLEWVA RIRSKYNNYATYYADSVKDRFTISRDDS KNTAYLQMNNLKTEDTAVYYCVRHGN FGNSYISYWAYWGQGTLVTVSSGGGGS GGGGSGGGGSQTVVTQEPSLTVSPGGT

- 160 046123

VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVL 428. Bispecific molecule with HLE MU 2-C2 CC x I2C0-scFc Artificial ahh R PSGIPERFSGSNSGNTATLTISGTQAMD EADYYCQAFHQSTWVFGCGTQLTVLSG GGGSEVQLVESGGGLVQPGGSLKLSCA ASGFTFNKYAMNWVRQAPGKGLEWVA RIRSKYNNYATYYADSVKDRFTISRDDS KNTAYLQMNNLKTEDTAVYYCVRHGN FGNSYISYWAYWGQGTLVTVSSGGGGS GGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP ELLGGPSVFLFPPKPKDTLMISRTPEVTC WVDVSHEDPEVKFNWYVDGVEVHNA KTK PCEEQYGSTYRCVSVLTVLHQDWL NGKEYKCKVSNKALPAPIEKTISKAKGQ PREPQVYTLPPSREEMTKNQVSLTCLVK GFYPSDIAVEWESNGQPENNYKTTPPVL DSDGSFFLYSKLTVDKSRWQQGNVFSC SVMHEALHNHYTQKSLSLSPGKGGGGS GGGGSGGGGSGGGGSGGGGSGGGGSD KTH TCPPCPAPELLGGPSVFLFPPKPKDT LMISRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS

- 161 046123

VLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLPPSREEMT KNQVSLTCLVKGFYPSDIAVEWESNGQP ENNYKTTPPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNHYTQKSL SLSPGK 429. CDR1 VH MU 2- A3 CC x I2C0-scFc Artificial ahh SFGMH 430. CDR2 VH MU 2- A3 CC x I2C0-scFc Artificial ahh VIWFDASNKYYAESVKG 431. CDR3 VH MU 2- A3 CC x I2C0-scFc Artificial ahh GGYTYGFDY 432. CDR1 VL MU 2- A3 CC x I2C0-scFc Artificial ahh RASQAINRYLA 433. CDR2 VL MU 2- A3 CC x I2C0-scFc Artificial ahh GASSRAT 434. CDR3 VL MU 2- A3 CC x I2C0-scFc Artificial ahh QHYGSSIFT 435. VH MU 2-A3 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSS 436. MU 2-A3 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLSVSPGERATLLSCRASQAI NRYLAWYQQKPGQAPRLLIYGASSRAT GIPDRFSGSGSGTDFTLTISRLEPEDFAVY YCQHYGSSIFTFGCGTKVEIK 437. scFv VL MU 2-A3 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLSVSPGERAATLSCRASQAIN RYLAWYQQKPGQAPRLLIYGASSRATGI

- 162 046123

PDRFSGSGSGTDFTLTISRLEPEDFAVYY CQHYGSSIFTFGCGTKVEIK 438. Bispecific molecule MU 2AZ CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLSVSPGERAATLSCRASQAIN RYLAWYQQKPGQAPRLLIYGASSRATGI PDRFSG SGSGTDFTLTISRLEPEDFAVYY CQHYGSSIFTFGCGTKVEIKSGGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG GGGSQTVVTQ EPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL 439. Bispecific molecule with HLE MU 2-AZ CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSSFGMHWVRQAPGKCLEWVAVIWF DASNKYYAESVKGRFTISRDNSKNTLYL QMNSLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLSVSPGERAATLSCRASQAIN RYLAWYQQKPGQAPRLLIYGASSRATGI PDRFSG SGSGTDFTLTISRLEPEDFAVYY CQHYGSSIFTFGCGTKVEIKSGGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG

- 163 046123

GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTY RCVSVLTVLHQDWLNGKEYK CKVSNKALPAPIEKTISKAKGQPREPQV YTLPPSREEMTKNQVSLTCLVKGFYPSD IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELL GGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQ QG NVFSCSVMHEALHNHYTQKSLSLSPGK 440. CDR1 VH MU 1- H2 CC x I2C0-scFc Artificial ahh SYGMH 441. CDR2 VH MU 1- H2 CC x I2C0-scFc Artificial ahh VIWYDASNKYYATSVKG 442. CDR3 VH MU 1- H2 CC x I2C0-scFc Artificial ahh GGYTYGFDY 443. CDR1 VL MU 1- H2 CC x I2C0-scFc Artificial ahh RASQSINRYLA 444. CDR2 VL MU 1- H2 CC x I2C0-scFc Artificial ahh TASNRAT 445. CDR3 VL MU 1- H2 CC x I2C0-scFc Artificial ahh HHYGSSIFT

- 164 046123

446. VH MU 1-H2 CC X I2C0-scFc Artificial ahh QVQLVESGGGLVKPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWYD ASNKYYATSVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSS 447. VL MU 1-H2 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 448. scFv MU 1-H2 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWYD ASNKYYATSVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSG SGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIK 449. Bispecific molecule MU 1H2 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWYD ASNKYYATSVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSG SGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEP SLTVSPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV

- 165 046123

QPEDEAEYYCVLWYSNRWVFGGGTKL TVL 450. Bispecific molecule with HLE MU 1-H2 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGGSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWYD ASNKYYATSVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSG SGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEP SLTVSPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTCPPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRCVSVLTV LHQDWLNGKEYKC KVSNKALPAPIEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTCLVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGGGGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSVFLFPPK PKDTLMISRTPE VTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPCEEQYGSTYRCVSVLTVLHQ DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTC

- 166 046123

LVKGFYPSDIAVEWESNGQPENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGK 451. CDR1 VH MU 1- E9 CC x I2C0-scFc Artificial ahh SYGMH 452. CDR2 VH MU 1- E9 CC x I2C0-scFc Artificial ahh VIWFHGSNKYYAESVKG 453. CDR3 VH MU 1- E9 CC x I2C0-scFc Artificial ahh GAYTYGFDY 454. CDR1 VL MU 1- E9 CC x I2C0-scFc Artificial ahh RASQSINRYLA 455. CDR2 VL MU 1- E9 CC x I2C0-scFc Artificial ahh TASNRAT 456. CDR3 VL MU 1- E9 CC x I2C0-scFc Artificial ahh HHYGSSIFT 457. VH MU 1-E9 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF HGSNKYYAESVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSS 458. VL MU 1-E9 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 459. scFv MU 1-E9 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF HGSNKYYAESVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSG TDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIK 460. Bispecific Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF

- 167 046123

molecule MU 1 - E9 CC x I2C0-scFc TFSSYGMHWVRQAPGKCLEWVAVIWF HGSNKYYAESVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSLPGERATLLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSGTDFTLTISRLEPEDFAVYYC HHYGSS IFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTVSPGGTVTLTCGSS TGAVTSGNYPNWVQQ KPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVL 461. Bispecific molecule with HLE MU 1-E9 CC x I2C0-scFc Artificial aa QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWF HGSNKYYAESVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGAYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSG TDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTV SPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV

- 168 046123

QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTCPPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRCVSVLTVLHQDWLNGKEYKC KVSNKALPAPIEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTC LVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGGGGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSVFLFPPKPKDTLMISRTPE VTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPCSVEEQYGSTYRCV LTVLHQ DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTC LVKGFYPSDIAVEWESNGQPENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGK 462. CDR1 VH MU 1- B10 CC x I2C0scFc Artificial ahh NFGMH 463. CDR2 VH MU 1- B10 CC x I2C0scFc Artificial ahh VIWFDASKTYYAEAVKG 464. CDR3 VH MU 1- B10 CC x I2C0scFc Artificial ahh ATYSTGWSYFDY 465. CDR1 VL MU 1- B10 CC x I2C0scFc Artificial ahh SGDKLGDKYTS 466. CDR2 VL MU 1B10 CC x I2C0scFc Artificial ahh HDAKRPS 467. CDR3 VL MU 1- Artificial ahh QAWSASTAWV

- 169 046123

BIO CC X I2C0scFc 468. VH MU 1-B10 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAEAVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSS 469. VL MU 1-B10 CC x I2C0-scFc Artificial ahh SYELTQPPSVSVSPGQTASTTCSGDKLGD KYTSWYQQKPGQSPVLVIYHDAKRPSGI PERFSGSNSGNTATLTISGTQAMDEADY YCQAWSASTAWVFGCGTKLTVL 470. scFv MU 1-B10 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAEAVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAK R PSGIPERFSGSNSGNTATLTISGTQAMDE ADYYCQAWSASTAWVFGCGTKLTVL 471. Bispecific molecule MU 1 BIO CC X I2C0scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAEAVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAK S GGGGSGGGGSQTVVTQEPSLTVSPGGT

- 170 046123

VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVL 472. Bispecific molecule with HLE MU 1-B10 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVQPGRSLRLSCAASGF TFSNFGMHWVRQAPGKCLEWVAVIWF DASKTYYAEAVKGRFTISRDTSMNTLYL QMNSLRAEDTAVYYCARATYSTGWSY FDYWGQGTLVTVSSGGGGSGGGGSGG GGSSYELTQPPSVSVSPGQTASITCSGDK LGDKYTSWYQQKPGQSPVLVIYHDAK S GGGGSGGGGSQTVVTQEPSLTVSPGGT VTLTCGSSTGAVTSGNYPNWVQQKPGQ APRGLIGGTKFLAPGTPARFSGSLLGGK AALTLSGVQPEDEAEYYCVLWYSNRW VFGGGTKLTVLGGGGDKTHTCPPCPAP ELLGGPSVFLFPPKPKDTLMISRTPEVTC WVDVSHEDPEVKFNWYVDGVEVHNA KTKPC KTHTC PPCPAPELLGGPSVFLFPPKPKDT LMISRTPEVTCWVDVSHEDPEVKFNW YVDGVEVHNAKTKPCEEQYGSTYRCVS

- 171 046123

VLTVLHQDWLNGKEYKCKVSNKALPA PIEKTISKAKGQPREPQVYTLPPSREEMT KNQVSLTCLVKGFYPSDIAVEWESNGQP ENNYKTTPPVLDSDGSFFLYSKLTVDKS RWQQGNVFSCSVMHEALHNHYTQKSL SLSPGK 473. CDR1 VH MU 1- B6 CC x I2C0-scFc Artificial ahh GYYWS 474. CDR2 VH MU 1- B6 CC x I2C0-scFc Artificial ahh DIDYSGSTKYNPSLKS 475. CDR3 VH MU 1- B6 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDY 476. CDR1 VL MU 1- B6 CC x I2C0-scFc Artificial ahh SGDKLGDKYAN 477. CDR2 VL MU 1- B6 CC x I2C0-scFc Artificial ahh HDNKRPS 478. CDR3 VL MU 1- B6 CC x I2C0-scFc Artificial ahh QAYGISSAV 479. VH MU 1-B6 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDYSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSS 480. VL MU 1-B6 CC x I2C0-scFc Artificial ahh SYELTQPASASVSPGQTASITCSGDKLG DKYANWYQQKPGQSPILVIYHDNKRPS GIPERFSGSNSGNTATLTISGTQAMDEA DYYCQAYGISSAVFGCGTKLTVL 481. scFv MU 1-B6 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDYSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPASASVSPGQTASHCSGDKLGDK YANWYQQKPGQSPILVIYHD NKRPSGIP

- 172 046123

ERFSGSNSGNTATLTISGTQAMDEADYY CQAYGISSAVFGCGTKLTVL 482. Bispecific molecule MU 1 - Wb CC X I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDYSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPASASVSPGQTASITCSGDKLGDK YANWYQQKPGQSPILVIYHD NKRPSGIP ERFSGSNSGNTATLTISGTQAMDEADYY CQAYGISSAVFGCGTKLTVLSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGG GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVL 483. Bispecific molecule with HLE MU 1-B6 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDYSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPASASVSPGQTASITCSGDKLGDK YANWYQQKPGQSPILVIYHDNKRPSGIP ERFSGSNSGNTATLTISGTQAMDEADYY CQAYGISSAVFGCGTKLTVLSGGGGSEV QLVESGGGLVQPGGSLKLSCAASGFTFN KYAMNWVRQAPGKGLEWVARIRSKYN NYATYYADSVKDRFTISRDDSKNTAYL QMNNLKTEDTAVYYCVRHGNFGNSYIS YWAYWGQGTLVTVSSGGGGSGGGGSG

- 173 046123

GGGSQTVVTQEPSLTVSPGGTVTLTCGS STGAVTSGNYPNWVQQKPGQAPRGLIG GTKFLAPGTPARFSGSLLGGKAALTLSG VQPEDEAEYYCVLWYSNRWVFGGGTK LTVLGGGGDKTHTCPPCPAPELLGGPSV FLFPPKPKDTLMISRTPEVTCVWDVSH EDPEVKFNWYVDGVEVHNAKTKPCEE QYGSTY RCVSVLTVLHQDWLNGKEYK CKVSNKALPAPIEKTISKAKGQPREPQV YTLPPSREEMTKNQVSLTCLVKGFYPSD IAVEWESNGQPENNYKTTPPVLDSDGSF FLYSKLTVDKSRWQQGNVFSCSVMHEA LHNHYTQKSLSLSPGKGGGGSGGGGSG GGGSGGGGSGGGGSGGGGSDKTHTCPP CPAPELL GGPSVFLFPPKPKDTLMISRTP EVTCVVVDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQ QG NVFSCSVMHEALHNHYTQKSLSLSPGK 484. CDR1 VH MU 1- A6 CC x I2C0-scFc Artificial ahh GYYWS 485. CDR2 VH MU 1- A6 CC x I2C0-scFc Artificial ahh DIDQSGSTKYNPSLKS 486. CDR3 VH MU 1- A6 CC x I2C0-scFc Artificial ahh KKYSTVWSYFDY 487. CDR1 VL MU 1- A6 CC x I2C0-scFc Artificial ahh SGDKLGDKYAS 488. CDR2 VL MU 1- A6 CC x I2C0-scFc Artificial ahh QDRKRPS 489. CDR3 VL MU 1- A6 CC x I2C0-scFc Artificial ahh QAWGSSAAV

- 174 046123

490. VH MU 1-A6 CC X I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSS 491. VL MU 1-A6 CC x I2C0-scFc Artificial ahh SYELTQPSSVSVPPGQTASITCSGDKLGD KYASWYQQKPGQSPVLVIYQDRKRPSG VPERFSGSNSGNTATLTISGTQAMDEAD YYCQAWGSSAAVFGCGTKLTVL 492. scFv MU 1-A6 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIY QDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSAAVFGCGTKLTVL 493. Bispecific molecule MU 1A6 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASITCSGDKLGDK YASWYQQKPGQSPVLVIY QDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSAAVFGCGTKLTVLSGGGGS EVQLVESGGGLVQPGGSLKLSCAASGFT FNKYAMNWVRQAPGKGLEWVARIRSK YNNYATYYADSVKDRFTISRDDSKNTA YLQMNNLKTEDTAVYYCVRHGNFGNS YISYWAYWGQGTLVTVSSGG SGGG GSGGGGSQTVVTQEPSLTVSPGGTVTLT CGSSTGAVTSGNYPNWVQQKPGQAPRG LIGGTKFLAPGTPARFSGSLLGGKAALT

- 175 046123

LSGVQPEDEAEYYCVLWYSNRWVFGG GTKLTVL 494. Bispecific molecule with HLE MU 1-A6 CC x I2C0-scFc Artificial ahh QVQLQQWGAGLLKPSETLSLTCAVYGG SFSGYYWSWIRQPPGKCLEWIGDIDQSG STKYNPSLKSRVTISLDTSKNQFSLKLNS VTAADTAVYFCARKKYSTVWSYFDYW GQGTLVTVSSGGGGSGGGGSGGGGSSY ELTQPSSVSVPPGQTASETCSGDKLGDK YASWYQQKPGQSPVLVIY QDRKRPSGV PERFSGSNSGNTATLTISGTQAMDEADY YCQAWGSSAAVFGCGTKLTVLSGGGGS EVQLVESGGGLVQPGGSLKLSCAASGFT FNKYAMNWVRQAPGKGLEWVARIRSK YNNYATYYADSVKDRFTISRDDSKNTA YLQMNNLKTEDTAVYYCVRHGNFGNS YISYWAYWGQGTLVTVSSGG SGGG GSGGGGSQTVVTQEPSLTVSPGGTVTLT CGSSTGAVTSGNYPNWVQQKPGQAPRG LIGGTKFLAPGTPARFSGSLLGGKAALT LSGVQPEDEAEYYCVLWYSNRWVFGG GTKLTVLGGGGDKTHTCPPCPAPELLGG PSVFLFPPKPKDTLMISRTPEVTCWVDV SHEDPEVKFNWYVDGVEVHNAKTKPCE PCP APELLGGPSVFLFPPKPKDTLMISRT PEVTCVWDVSHEDPEVKFNWYVDGVE VHNAKTKPCEEQYGSTYRCVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISK AKGQPREPQVYTLPPSREEMTKNQVSLT

- 176 046123

CLVKGFYPSDIAVEWESNGQPENNYKT TPPVLDSDGSFFLYSKLTVDKSRWQQG NVFSCSVMHEALHNHYTQKSLSLSPGK 495. CDR1 VH MU Ο- ΡΟ CC x I2C0-scFc Artificial ahh SFGMH 496. CDR2 VH MU Ο- ΡΟ CC x I2C0-scFc Artificial ahh VIWYTGSNKYYASSVKG 497. CDR3 VH MU Ο- ΡΟ CC x I2C0-scFc Artificial ahh GGYTYGFDY 498. CDR1 VL MU Ο- ΡΟ CC x I2C0-scFc Artificial ahh RASQSINRYLA 499. CDR2 VL MU Ο- ΡΟ CC x I2C0-scFc Artificial ahh TASNRAT 500. CDR3 VL MU Ο- ΡΟ CC x I2C0-scFc Artificial ahh HHYGSSIFT 501. VH MU 0-F0 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGGSLRLSCAASGFT FSSFGMHWVRQAPGKCLEWVAVIWYT GSNKYYASSVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSS 502. VL MU 0-F0 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 503. scFv MU 0-F0 CC x I2C0-scFc Artificial ahh DRF SSGSGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIK 504. Bispecific Artificial ahh QVQLVESGGGLVKPGGSLRLSCAASGFT

- 177 046123

molecule MU 0-F9 CC x I2C0-scFc HH YGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTVSPGGTVTLTCGSS TGAVTSGNYPNW VQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVL 505. Bispecific molecule with HLE MU 0-F9 CC x I2C0-scFc Artificial aa DRF SGSGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQ EPSLTVSPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV

- 178 046123

QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTCPPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRCVSVLTVLHQDWLNGKEYKC KVSNKALPAPIEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTC LVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGGGGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSVFLFPPKPKDTLMISRTPE VTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPCSVEEQYGSTYRCV LTVLHQ DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTC LVKGFYPSDIAVEWESNGQPENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGK 506. CDR1 VH MU 0- F6 CC x I2C0-scFc Artificial ahh SYGMH 507. CDR2 VH MU 0- F6 CC x I2C0-scFc Artificial ahh VIWFDASNKYYAESVKG 508. CDR3 VH MU 0- F6 CC x I2C0-scFc Artificial ahh GGYTYGFDY 509. CDR1 VL MU 0- F6 CC x I2C0-scFc Artificial ahh RASQSINRYLA 510. CDR2 VL MU 0- F6 CC x I2C0-scFc Artificial ahh TASNRAT 511. CDR3 VL MU 0- F6 CC x I2C0-scFc Artificial ahh HHYGSSIFT 512. VH MU 0-F6 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGRSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFD ASNKYYAESVKGRFTISRDNSKNTLYLQ

- 179 046123

MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSS 513. VL MU 0-F6 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 514. scFv MU 0-F6 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGRSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFD ASNKYYAESVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSG SGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIK 515. Bispecific molecule MU 0-F6 CC x I2C0-scFc Artificial ahh QVQLVESGGGLVKPGRSLRLSCAASGFT FSSYGMHWVRQAPGKCLEWVAVIWFD ASNKYYAESVKGRFTISRDNSKNTLYLQ MNNLRAEDTAVYYCARGGYTYGFDYW GQGTLVTVSSGGGGSGGGGSGGGGSEI VLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSG SGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEP SLTVSPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVL 516. Bispecific Artificial ahh QVQLVESGGGLVKPGRSLRLSCAASGFT

- 180 046123

molecule with HLE MU 0-F6 SS x I2C0-scFc HHY GSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTVSPGGTVTLTCGSS TGAVTSGNYPNWV QQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTCPPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRCVSVLTVLHQDWLNGKEYKC KVSNKALPAP IEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTCLVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGGGGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSVFLFPPKPKDTLMISRTPE VTCVVVDVSHEDPEV KFNWYVDGVEV HNAKTKPCEEQYGSTYRCVSVLTVLHQ DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTC LVKGFYPSDIAVEWESNGQPENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGK

- 181 046123

517. CDR1 VH MU 0- E5 CC x I2C0-scFc Artificial ahh SYGMH 518. CDR2 VH MU 0- E5 CC x I2C0-scFc Artificial ahh VIWYDASNKYYATSVKG 519. CDR3 VH MU 0- E5 CC x I2C0-scFc Artificial ahh GGYTYGFDY 520. CDR1 VL MU 0- E5 CC x I2C0-scFc Artificial ahh RASQSINRYLA 521. CDR2 VL MU 0- E5 CC x I2C0-scFc Artificial ahh TASNRAT 522. CDR3 VL MU 0- E5 CC x I2C0-scFc Artificial ahh HHYGSSIFT 523. VH MU 0-E5 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY DASNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSS 524. VL MU 0-E5 CC x I2C0-scFc Artificial ahh EIVLTQSPGTLLSLSPGERATLSCRASQSIN RYLAWYQQKPGQAPRLLIYTASNRATGI PDRFSGSGSGTDFTLTISRLEPEDFAVYY CHHYGSSIFTFGCGTKVEIK 525. scFv MU 0-E5 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY DASNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSG TDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIK 526. Bispecific molecule MU 0E5 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY DASNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY

- 182 046123

WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSGTDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYY ADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTVSPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVL 527. Bispecific molecule with HLE MU 0-E5 CC x I2C0-scFc Artificial ahh QVQLVESGGGVVKPGRSLRLSCAASGF TFSSYGMHWVRQAPGKCLEWVAVIWY DASNKYYATSVKGRFTISRDNSKNTLYL QMNNLRAEDTAVYYCARGGYTYGFDY WGQGTLVTVSSGGGGSGGGGSGGGGSE IVLTQSPGTLLSPGERATLSCRASQSINR YLAWYQQKPGQAPRLLIYTASNRATGIP DRFSGSGSG TDFTLTISRLEPEDFAVYYC HHYGSSIFTFGCGTKVEIKSGGGGSEVQ LVESGGGLVQPGGSLKLSCAASGFTFNK YAMNWVRQAPGKGLEWVARIRSKYNN YATYYADSVKDRFTISRDDSKNTAYLQ MNNLKTEDTAVYYCVRHGNFGNSYISY WAYWGQGTLVTVSSGGGGSGGGGSGG GGSQTVVTQEPSLTV SPGGTVTLTCGSS TGAVTSGNYPNWVQQKPGQAPRGLIGG TKFLAPGTPARFSGSLLGGKAALTLSGV QPEDEAEYYCVLWYSNRWVFGGGTKL TVLGGGGDKTHTCPPCPAPELLGGPSVF LFPPKPKDTLMISRTPEVTCVVVDVSHE

- 183 046123

DPEVKFNWYVDGVEVHNAKTKPCEEQ YGSTYRCVSVLTVLHQDWLNGKEYKC KVSNKALPAPIEKTISKAKGQPREPQVY TLPPSREEMTKNQVSLTCLVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFF LYSKLTVDKSRWQQGNVFSCSVMHEAL HNHYTQKSLSLSPGKGG GGSGGGGSGG GGSGGGGSGGGGSGGGGSDKTHTCPPC PAPELLGGPSVFLFPPKPKDTLMISRTPE VTCVVVDVSHEDPEVKFNWYVDGVEV HNAKTKPCEEQYGSTYRCVSVLTVLHQ DWLNGKEYKCKVSNKALPAPIEKTISKA KGQPREPQVYTLPPSREEMTKNQVSLTC LVKGFYPSDIAVEWESNGQ PENNYKTTP PVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGK 528. Epitope E2 MUC17 Artificial ahh EVVSSIDIGPPET1SAQMELTVTVTSVKFT EELKNHSSQEFQEFKQTFTEQMNIVYSGI PEYVGVNITKLRLGSVVVEHDVLLRTK YTPEYKTVLDNATEVVKEKITKVTTQQI MINDICSDMMCF 529. Epitope E2 MUC17 (shortened N-terminus) Artificial ahh SAQMELTVTVTSVKFTEELKNHSSQEFQ EFKQTFTEQMNIVYSGIPEYVGVNITKLR LGSVVVEHDVLLRTKYTPEYKTVLDNA TEVVKEKTTKVTTQQIMINDICS 530. Epitope 5A MUC17 (contains part of E2) Artificial ahh RTTTCFGDGCQNTASRCKNGGTWDGLK CQCPNLYYGELCEEVVSSIDIGPPETISA QMELTVTVTSVKFTEELKNHSSQEFQEF KQTFTEQMNIVYSGIPEYVGVNITKLRL G 531. Epitope 5B MUC17 (contains part of E2) Artificial ahh SVVVEHDVLLRTKYTPEYKTVLDNATE WKEKITKVTTQQIMINDICSDMMCFNT TGTQVQNETVTQYDPEEDCRKMAKEYG DYFVVEYRDQKPYCISPCEPGFSVSKNC NLGKCQMSLSGPQCLCVTTETHWYSGE

- 184 -

Claims (4)

TCNQGTQKS 532. Truncate 2 epitope E2 MUC17 Artificial ahh EVVSSmiGPPETISAQMELTVTVTSVKFT EELKNHSSQEFQEFKQTFTEQMNIVYSGI PEYVGVNITKLRLGSVVVEHDVLLRTK YTPEYKTVLDNATEVVKEKITKVTTQQI MINDICSDMMCFNTTGTQVQNITVTQY DPEEDCRKMAKEYGDYFWEYRDQKP YCISPCEPGFSVSKNCNLGKC QMSLSGP QCLCVTTETHWYSGETCNQGTQKSL 533. Truncated 3 epitope E2 MUC17 Artificial ahh ISAQMELTVTVTSVKFTEELKNHSSQEF QEFKQTFTEQMNIVYSGIPEYVGVNITKL RLGSVVVEHDVLLRTKYTPEYKTVLDN ATEVVKEKTTKVTTQQIMINDICSDMMC FNTTGTQVQNITVTQYDPEEDCRKMAK EYGDYFVVEYRDQKPYCISPCEPGFSVS KNCNLGKCQMSLSGPQ CLCVTTETHWY SGETCNQGTQKSL 534. Truncate 4 epitope E2 MUC17 Artificial ahh DMMCFNTTGTQVQNrrVTQYDPEEDCR KMAKEYGDYFVVEYRDQKPYCISPCEP GFSVSKNCNLGKCQMSLSGPQCLCVTT ETHWYSGETCNQGTQKSL 535. Truncate 5 epitope E2 MUC17 Artificial ahh SPCEPGFSVSKNCNLGKCQMSLSGPQCL CVTTETHWYSGETCNQGTQKSL CLAIM 1. A design based on a bispecific antibody, characterized in that it contains: the first domain that binds to MUC17; and a second domain that binds to an extracellular epitope of the human and Macasa CD3ε chain; wherein the first binding domain comprises a VH region containing CDR-H1, CDR-H2 and CDR-H3 selected from: (a) CDR-H1 shown under SEQ ID NO: 55, CDR-H2 shown under SEQ ID NO: 56, and CDRH3 shown under SEQ ID NO: 57; (b) CDR-H1 shown under SEQ ID NO: 176, CDR-H2 shown under SEQ ID NO: 177, and CDR-H3 shown under SEQ ID NO: 178; (c) CDR-H1 shown under SEQ ID NO: 418, CDR-H2 shown under SEQ ID NO: 419, and CDR-H3 shown under SEQ ID NO: 420; (d) CDR-H1 shown under SEQ ID NO: 473, CDR-H2 shown under SEQ ID NO: 474, and CDR-H3 shown under SEQ ID NO: 475; and the first binding domain comprises a VL region containing CDR-H1, CDR-L2 and CDR-L3 selected from: (a) CDR-L1 given at SEQ ID NO: 58, CDR-L2 given at SEQ ID NO: 59, and CDRL3 given at SEQ ID NO: 60; (b) CDR-L1 shown under SEQ ID NO: 179, CDR-L2 shown under SEQ ID NO: 180, and CDR-L3 shown under SEQ ID NO: 181; (c) CDR-L1 shown under SEQ ID NO: 421, CDR-L2 shown under SEQ ID NO: 422, and CDR-L3 shown under SEQ ID NO: 423; and (d) CDR-L1 set forth at SEQ ID NO: 476, CDR-L2 set forth at SEQ ID NO: 477, and CDR-L3 set forth at SEQ ID NO: 478. 2. A construct based on a bispecific antibody according to claim 1, characterized in that it additionally contains a third domain, which contains two polypeptide monomers, each of which contains a hinge region, a CH2 domain and a CH3 domain, where these two polypeptide monomers are fused to each other by peptide linker. 3. The bispecific antibody construct according to claim 1 or 2, wherein the antibody construct is a single chain antibody construct. 4. A bispecific antibody construct according to any one of claims 1 to 3, characterized in that the first binding domain contains a VL region and a VH region selected from the group consisting of: - 185 -