MUC1 and CD16a antibodies and methods of use

Abstract

The present disclosure provides multispecific antibodies and antigen binding fragments thereof that bind to human MUC1 and CD16A, pharmaceutical compositions comprising said antibodies and the use of said antibodies or said compositions for the treatment of diseases such as cancer.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of priority to PCT application No. PCT / CN2023 / 079507, filed on March 3, 2023, entitled “MUC1 and CD16A Antibodies and Methods of Use,” and PCT application No. PCT / CN2023 / 107724, filed on July 17, 2023, entitled “MUC1 and CD16A Antibodies and Methods of Use,” which are hereby incorporated by reference in their entirety.

[0003] Sequence Listing

[0004] This application is filed with a sequence listing in electronic format. The sequence listing is provided as a 179,126-byte file entitled "01368-0054-00PCT_SL" created on February 27, 2024. The information in the electronic sequence listing is incorporated herein by reference in its entirety. Technical Field

[0005] Disclosed herein are multispecific antibodies or antigen-binding fragments thereof that bind to human MUCl and human CD16A and compositions comprising the same. Background Art

[0006] Mucin 1 (MUC1; also known as CA 15-3, EMA, MCD, PEM, PUM, KL-6, MAM6, MCKD, PEMT, CD227, H23AG, MCKD1, ADMCKD, ADTKD2), which belongs to the mucin family, is a heavily glycosylated heterodimeric membrane-tethered protein normally expressed on the apical surface of glandular or luminal epithelial cells of various tissues, lubricating and nourishing the epithelial cell surface and protecting it from pathogens. However, in most human epithelial cancers, including pancreatic, breast, ovarian, lung, and colon cancers, MUC1 is hypoglycosylated and abnormally overexpressed. In addition, due to the loss of apical-basal polarity in cancer cells, MUC1 is expressed on the entire surface of tumor cells. Given these properties, MUC1 is considered a promising therapeutic target for human cancers.

[0007] MUC1 heterodimers consist of a longer N-terminal extracellular domain (ECD) (MUC1-N) and a shorter subunit (MUC1-C) containing a C-terminal cytoplasmic domain of 69 amino acids, a hydrophobic transmembrane domain of 28 amino acids, and a short ECD of 58 amino acids. The two subunits are non-covalently bound by hydrogen interactions. MUC1-N is usually shed from the surface of cells and released into the circulation. There is an increased level of shed MUC1 in the serum of patients with various cancers. Previously, various MUC1-N targeting therapeutics such as AS1402 (huHMFG-1) and BrevaRex (AR-20.5) failed in the clinic, mainly due to shed MUC1 sequestering these anti-MUC1 antibodies, thereby hindering their ability to bind to MUC1 on the surface of cancer cells. To overcome this problem, antibodies that bind to MUC1-C can be a more promising strategy to target MUC1-expressing cancer cells efficiently.

[0008] CD16A (also known as FcyRIIIa) is a low affinity receptor for IgGl and IgG3 expressed by natural killer (NK) cells, macrophages, and some circulating monocytes. CD16A can induce an activation signal alone and kill target cells opsonized by antibodies through antibody-dependent cell-mediated cytotoxicity (ADCC).

[0009] The ADCC mechanism contributes to the therapeutic efficacy of various widely used tumor-targeting monoclonal antibodies (mAbs) such as trastuzumab and rituximab. And this role of CD16A is supported by evidence of clinical responses to therapeutic antibodies that are affected by polymorphism of CD16A. In multiple indications, patients with homozygous high-affinity variants (CD16A-158V / V) show better clinical responses than those with heterozygous (CD16A-158V / F) or homozygous low-affinity variants (CD16A-158F / F).

[0010] To date, there are different strategies to enhance NK cell responses to tumor-targeting therapeutics. The most common strategy is genetic manipulation and glycoengineering of the antibody Fc region to enhance its interaction with CD16A. Another strategy is to handle the antibody format. Bispecific killer engagers or trispecific killer engagers are generated to efficiently redirect innate immune cells such as NK cells to tumor cells. These formats can better bridge NK cells and tumor cells and allow high-affinity engagement of CD16A, resulting in excellent killing activity of the antibodies.

[0011] The present disclosure provides targeting MUC1-C and CD16A via multispecific antibodies that recruit innate immune cells to MUC1 expressing cells and are useful for treating MUC1 expressing cancers. SUMMARY

[0012] The present disclosure relates to multispecific anti-MUC1xCD16A antibodies and antigen binding fragments thereof.

[0013] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment thereof comprising a first antigen binding domain that specifically binds to human MUC1 and a second antigen binding domain that specifically binds to human CD16A.

[0014] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the first antigen binding domain has high selectivity over human CD16B.

[0015] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the first antigen binding domain that specifically binds to human MUC1 comprises:

[0016] (i). a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 28, and (f) LCDR3 of SEQ ID NO: 29;

[0017] (ii). a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 65, and (f) LCDR3 of SEQ ID NO: 29;

[0018] (iii). a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 4, (b) HCDR2 of SEQ ID NO: 5, and (c) HCDR3 of SEQ ID NO: 6, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 7, (e) LCDR2 of SEQ ID NO: 8, and (f) LCDR3 of SEQ ID NO: 9; or

[0019] (iv). a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 14, (b) HCDR2 of SEQ ID NO: 15, and (c) HCDR3 of SEQ ID NO: 16, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 17, (e) LCDR2 of SEQ ID NO: 18, and (f) LCDR3 of SEQ ID NO: 19.

[0020] In embodiments, the disclosure relates to a multispecific antibody or antigen binding fragment, wherein the first antigen binding domain comprises:

[0021] (i). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 30, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 31;

[0022] (ii). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 61, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 62;

[0023] (iii). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 61, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 66;

[0024] (iv). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 61, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 68;

[0025] (v). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 10, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 11; or

[0026] (vi). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 20, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 21.

[0027] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein one, two, three, four, five, six, seven, eight, nine, or ten amino acids within one or more of SEQ ID NOs: 30, 31, 61, 62, 66, 68, 10, 11, 20, and 21 have been inserted, deleted, or substituted.

[0028] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the first antigen binding domain comprises:

[0029] (i). a heavy chain variable region (VH) comprising SEQ ID NO: 30, and a light chain variable region (VL) comprising SEQ ID NO: 31;

[0030] (ii). a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 62;

[0031] (iii). a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 66;

[0032] (iv). a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 68;

[0033] (v). a heavy chain variable region (VH) comprising SEQ ID NO: 10, and a light chain variable region (VL) comprising SEQ ID NO: 11; or

[0034] (vi). a heavy chain variable region (VH) comprising SEQ ID NO: 20, and a light chain variable region (VL) comprising SEQ ID NO: 21.

[0035] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the second antigen binding domain that specifically binds to human CD16A comprises:

[0036] (i). a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 109, (b) HCDR2 of SEQ ID NO: 110, and (c) HCDR3 of SEQ ID NO: 111; or

[0037] (ii). a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 109, (b) HCDR2 of SEQ ID NO: 114, and (c) HCDR3 of SEQ ID NO: 111.

[0038] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the second antigen binding domain comprises:

[0039] (i). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 112;

[0040] (ii). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 115;

[0041] (iii). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 117; or

[0042] (iv). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to SEQ ID NO: 119.

[0043] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein one, two, three, four, five, six, seven, eight, nine, or ten amino acids within one or more of SEQ ID NOs: 112, 115, 117, and 119 have been inserted, deleted, or substituted.

[0044] The multispecific antibody or antigen-binding fragment, wherein the second antigen-binding domain comprises:

[0045] (i). a heavy chain variable region (VH) comprising SEQ ID NO: 112;

[0046] (ii). a heavy chain variable region (VH) comprising SEQ ID NO: 115;

[0047] (iii). a heavy chain variable region (VH) comprising SEQ ID NO: 117; or

[0048] (iv). a heavy chain variable region (VH) comprising SEQ ID NO: 119.

[0049] In embodiments, the present disclosure relates to a multispecific antibody or antigen-binding fragment, wherein:

[0050] (i). the first antigen-binding domain that specifically binds to human MUC1 comprises a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 28, and (f) LCDR3 of SEQ ID NO: 29; and the second antigen-binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 110; and (c) HCDR3 of SEQ ID NO: 111;

[0051] (ii). the first antigen-binding domain that specifically binds to human MUC1 comprises a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 65, and (f) LCDR3 of SEQ ID NO: 29; and the second antigen-binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 110; and (c) HCDR3 of SEQ ID NO: 111;

[0052] (iii). The first antigen binding domain that specifically binds to human MUC1 comprises a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO:4, (b) HCDR2 of SEQ ID NO:5, and (c) HCDR3 of SEQ ID NO:6, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO:7, (e) LCDR2 of SEQ ID NO:8, and (f) LCDR3 of SEQ ID NO:9; and the second antigen binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 110; and (c) HCDR3 of SEQ ID NO: 111;

[0053] (iv). The first antigen binding domain that specifically binds to human MUC1 comprises a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 14, (b) HCDR2 of SEQ ID NO: 15, and (c) HCDR3 of SEQ ID NO: 16, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 17, (e) LCDR2 of SEQ ID NO: 18, and (f) LCDR3 of SEQ ID NO: 19; and the second antigen binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 110; and (c) HCDR3 of SEQ ID NO: 111;

[0054] (v). The first antigen binding domain that specifically binds to human MUC1 comprises a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 28, and (f) LCDR3 of SEQ ID NO: 29; and the second antigen binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 114; and (c) HCDR3 of SEQ ID NO: 111;

[0055] (vi). The first antigen binding domain that specifically binds to human MUC1 comprises a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 65, and (f) LCDR3 of SEQ ID NO: 29; and the second antigen binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 114; and (c) HCDR3 of SEQ ID NO: 111;

[0056] (vii). The first antigen binding domain that specifically binds to human MUC1 comprises a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 4, (b) HCDR2 of SEQ ID NO: 5, and (c) HCDR3 of SEQ ID NO: 6, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 7, (e) LCDR2 of SEQ ID NO: 8, and (f) LCDR3 of SEQ ID NO: 9; and the second antigen binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 114; and (c) HCDR3 of SEQ ID NO: 111; or

[0057] (viii). The first antigen binding domain that specifically binds to human MUC1 comprises a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 14, (b) HCDR2 of SEQ ID NO: 15, and (c) HCDR3 of SEQ ID NO: 16, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 17, (e) LCDR2 of SEQ ID NO: 18, and (f) LCDR3 of SEQ ID NO: 19; and the second antigen binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 114; and (c) HCDR3 of SEQ ID NO: 111.

[0058] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein:

[0059] (i). The first antigen binding domain that specifically binds to human MUC1 comprises:

[0060] a) a heavy chain variable region (VH) comprising SEQ ID NO: 30, and a light chain variable region (VL) comprising SEQ ID NO: 31 ;

[0061] b) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 62;

[0062] c) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 66;

[0063] d) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 68;

[0064] e) a heavy chain variable region (VH) comprising SEQ ID NO: 10, and a light chain variable region (VL) comprising SEQ ID NO: 11; or

[0065] f) a heavy chain variable region (VH) comprising SEQ ID NO: 20, and a light chain variable region (VL) comprising SEQ ID NO: 21;

[0066] (ii). and the second antigen binding domain that specifically binds to human CD16A comprises:

[0067] a) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 109, (b) HCDR2 of SEQ ID NO: 110, (c) HCDR3 of SEQ ID NO: 111; or

[0068] b) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 109, (b) HCDR2 of SEQ ID NO: 114, (c) HCDR3 of SEQ ID NO: 111.

[0069] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein:

[0070] (i). the first antigen binding domain that specifically binds to human MUC1 comprises:

[0071] a) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 28, and (f) LCDR3 of SEQ ID NO: 29;

[0072] b) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 65, and (f) LCDR3 of SEQ ID NO: 29;

[0073] c) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 4, (b) HCDR2 of SEQ ID NO: 5, (c) HCDR3 of SEQ ID NO: 6, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 7, (e) LCDR2 of SEQ ID NO: 8, and (f) LCDR3 of SEQ ID NO: 9;

[0074] d) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 14, (b) HCDR2 of SEQ ID NO: 15, (c) HCDR3 of SEQ ID NO: 16, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 17, (e) LCDR2 of SEQ ID NO: 18, and (f) LCDR3 of SEQ ID NO: 19;

[0075] (ii). and the second antigen binding domain that specifically binds to human CD16A comprises:

[0076] a) a heavy chain variable region (VH) comprising SEQ ID NO: 112;

[0077] b) a heavy chain variable region (VH) comprising SEQ ID NO: 115;

[0078] c) a heavy chain variable region (VH) comprising SEQ ID NO: 117; or

[0079] d) a heavy chain variable region (VH) comprising SEQ ID NO: 119.

[0080] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein:

[0081] (i). the first antigen binding domain that specifically binds to human MUC1 comprises:

[0082] a) a heavy chain variable region (VH) comprising SEQ ID NO: 30, and a light chain variable region (VL) comprising SEQ ID NO: 31 ;

[0083] b) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 62;

[0084] c) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 66;

[0085] d) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 68;

[0086] e) a heavy chain variable region (VH) comprising SEQ ID NO: 10, and a light chain variable region (VL) comprising SEQ ID NO: 11 ; or

[0087] f) a heavy chain variable region (VH) comprising SEQ ID NO: 20, and a light chain variable region (VL) comprising SEQ ID NO: 21 ;

[0088] (ii). and the second antigen binding domain that specifically binds to human CD16A comprises:

[0089] a) a heavy chain variable region (VH) comprising SEQ ID NO: 112;

[0090] b) a heavy chain variable region (VH) comprising SEQ ID NO: 115;

[0091] c) a heavy chain variable region (VH) comprising SEQ ID NO: 117; or

[0092] d) a heavy chain variable region (VH) comprising SEQ ID NO: 119.

[0093] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment that is a monoclonal antibody, a chimeric antibody, a humanized antibody, a human engineered antibody, a single chain antibody (scFv), a Fab fragment, a Fab’ fragment, or a F(ab’)2 fragment.

[0094] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the multispecific antibody is a bispecific antibody.

[0095] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the multispecific antibody is BG1222P (SEQ ID NO: 143, SEQ ID NO: 145, and SEQ ID NO: 147).

[0096] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the antibody or antigen binding fragment thereof has antibody dependent cellular cytotoxicity (ADCC) or complement dependent cellular cytotoxicity (CDC).

[0097] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the antibody or antigen binding fragment thereof has reduced glycosylation or no glycosylation or is hypofucosylated.

[0098] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the antibody or antigen binding fragment thereof comprises increased bisecting GlcNac structures.

[0099] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the Fc domain is an IgGl with reduced effector function.

[0100] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment, wherein the Fc domain is an IgG4.

[0101] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment thereof comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of a CDR, VH, VL, or complete chain as disclosed herein.

[0102] In embodiments, the present disclosure relates to a multispecific antibody or antigen binding fragment thereof, wherein one, two, three, four, five, six, seven, eight, nine, or ten amino acids within the amino acid sequence of one or more of a CDR, VH, VL, or complete chain as disclosed herein have been inserted, deleted, or substituted.

[0103] In embodiments, the present disclosure relates to a pharmaceutical composition comprising a multispecific antibody or antigen binding fragment thereof as disclosed herein and a pharmaceutically acceptable carrier. The pharmaceutical composition can comprise histidine / histidine HC1, trehalose dihydrate, and polysorbate 20.

[0104] In embodiments, the present disclosure relates to a method of treating cancer comprising administering to a patient in need thereof an effective amount of a multispecific antibody or antigen binding fragment as disclosed herein. In embodiments, the present disclosure relates to an isolated nucleic acid encoding a multispecific antibody or antigen binding fragment disclosed herein.

[0105] In embodiments, the present disclosure relates to a vector comprising a nucleic acid disclosed herein.

[0106] In embodiments, the present disclosure relates to a host cell comprising a nucleic acid disclosed herein or a vector disclosed herein.

[0107] In embodiments, the present disclosure relates to a method of producing a multispecific antibody or antigen binding fragment thereof comprising culturing a host cell disclosed herein and recovering the antibody or antigen binding fragment from the culture.

[0108] In embodiments, the multispecific antibody of the present disclosure is of IgGl, IgG2, IgG3, or IgG4 isotype. In one embodiment, the antibody of the present disclosure comprises the Fc domain of wild-type human IgGl (also referred to as human IgGl wt or huIgGl) or IgG2.

[0109] In one embodiment, the multispecific antibody of the present disclosure binds to MUC1 with a binding affinity (K -6 M to 1 x 10 -10 M. D In another embodiment, the antibody of the present disclosure binds to MUC1 with a binding affinity (K -6 M, about 1 x 10 -7 M, about 1 x 10 -8 M, about 1 x 10 -9 M, or about 1 x 10 -10 M. D In another embodiment, the antibody of the present disclosure binds to MUC1 with a binding affinity (K

[0110] In embodiments, the anti-human MUC1 multispecific antibody of the present disclosure exhibits cross-species binding activity for cynomolgus monkey MUC1.

[0111] In embodiments, the antibody of the present disclosure has strong Fc-mediated effector functions. In some embodiments, the antibody mediates antibody-dependent cellular cytotoxicity (ADCC) against MUC1 -expressing target cells. BRIEF DESCRIPTION OF DRAWINGS

[0112] Figure 1 Schematic of MUC1-SEA-mIgG2a (top) and MUC1-SEA-huIgGl (bottom), where "N" is the N-terminus and "C" is the C-terminus.

[0113] Figures 2A-2F Binding affinity of purified MUC1 antibodies to human and cynomolgus MUC1 overexpressing cells was shown by FACS assay, with human IgG1 as negative control. Figure 2A and 2B Binding affinity of chimeric anti-MUC1 monoclonal antibody BG138P to human and cynomolgus cells overexpressing MUC1 was shown. Figure 2C and 2D Binding affinity of chimeric BG346P to human and cynomolgus cells overexpressing MUC1 was shown. Figure 2E and 2F Binding affinity of chimeric BG219P to human and cynomolgus cells overexpressing MUC1 was shown.

[0114] Figures 3A-3C Epitope binning assay by competition SPR assay was shown, where purified human MUC1-mFc antigen was flown over the chip surface and captured by anti-mouse IgG antibody.

[0115] Figures 4A-4F Effect of soluble MUC1 on binding of MUC1 antibodies to MUC1 expressing cells was shown. Figure 4A -C shows binding profile of chBG138P and chBG219P at different concentrations (30, 3 and 0.3 pg / ml, respectively) in the presence of soluble MUC1, with HMFG1 as positive control and mIgG and hIgG1 as negative controls. Figure 4D -F shows binding profile of chBG138P and chBG346P at different concentrations (30, 3 and 0.3 pg / ml, respectively) in the presence of soluble MUC1, with HMFG1 as positive control and mIgG and hIgG1 as negative controls.

[0116] Figures 5A-5C Anti-MUC1 monoclonal antibody chBG138P targeting the MUC1 membrane proximal region was shown to bind to MUC1 positive cancer cell lines. Figures 5A-5C ChBG138P was shown to bind to MUC1 expressing tumor cell lines HCC827 Figure 5A ), H1975 Figure 5B ) and T-47D Figure 5C ) in a dose dependent manner (with human IgG1 as negative control).

[0117] Figure 6 Schematic representation of the FACS gating strategy for T cell binding assay was shown. Dotted boxes show the proportion of T cells binding the antibody.

[0118] Figures 7A-7HThe chimeric anti-MUC1 monoclonal antibody chBG138P ( Figure 7A 、 7B , 7E and 7F), chBG219P( Figure 7E and 7F ) or chBG346P( Figure 7E and 7F ) does not bind to activated T cells, whereas the MUC1 membrane distal portion targeting antibody HMFG1 ( Figure 7C and 7D ) and 16A( Figure 7G and 7H ) can bind to normal T cells.

[0119] Figures 8A-8D Depicts chBP138P and humanized MUC1 antibodies BG138P-hz2 and BG138P-hz4 ( Figure 8A and 8B ) does not bind to normal T cells, whereas the MUC1 membrane distal portion targeting antibody HMFG1 ( Figure 8C and 8D ) binds to normal activated T cells.

[0120] Figure 9 Humanized antibodies huBG219P-Bz0, -E39, and -E43 were shown to bind comparably to the MUCl overexpressing cell line ZR-75-1 compared to the chimeric antibody chBG219P.

[0121] Figure 10 Shown are the results of ELISA analysis of a representative leading clone, BG523P, compared to LS21.

[0122] Figures 11A-11C Shown are FACS analyses of a representative leading clone, BG523P, compared to LS21.

[0123] Figures 12A-12B : Figure 12A shows FACS binding of BG523P and BG524P to the CD16A 158F overexpressing cell line NK92mi / CD16A 158F; and Figure 12B Shown are FACS binding of BG523P, BG525P, and BG526P to NK92mi / CD16A F158 cells.

[0124] Figure 13 Shown are FACS binding signals of BG523P, BG525P, and BG526P at 300 nM to CD16B-overexpressing cell lines NK92mi / CD16B NA1 and NK92mi / CD16B NA2.

[0125] Figures 14A-14C FACS-based human IgG competition of the binding of NK92mi / CD16A 158F to BG523P and its humanized VHHs (BG525P and BG526P) in the presence or absence of 10 mg / mL recombinant CB6 human IgG1 is shown. Figure 14A IgG competition effect on the binding of BG523P to NK92mi / CD16A 158F is shown. Figure 14B IgG competition effect on the binding of BG525P to NK92mi / CD16A 158F is shown. Figure 14C IgG competition effect on the binding of BG526P to NK92mi / CD16A 158F is shown.

[0126] Figure 15 Schematic representation of the MUC1xCD16A multispecific antibody BG1222P format.

[0127] Figure 16 Binding affinity of the MUC1xCD16A multispecific antibody BG1222P to the MUC1 -expressing tumor cell line T47D is shown.

[0128] Figure 17 Comparison of the binding of the MUC1xCD16A multispecific antibody BG1222P and huBG219P-E39-AF to human CD16A F158 or V158 overexpressing NK92mi cells in the absence (A, C) or presence of human IgG competition (B, D) is shown. Figure 17 A, C) or with human IgG competition (B, D) is shown. Figure 17 Comparison of the binding of the MUC1xCD16A multispecific antibody BG1222P and huBG219P-E39-AF to human CD16A F158 or V158 overexpressing NK92mi cells in the absence (A, C) or presence of human IgG competition (B, D) is shown.

[0129] Figure 18 A-18H shows a comparison of the antibody-dependent cellular cytotoxicity activity of the MUC1xCD16A multispecific antibody BG1222P and huBG219P-E39-AF mediated by NK92mi / CD16A F158. In the absence (A, B, C, D) or presence (E, F, G, H) of human IgG competition. Figure 18 A, B, C, D) or with (E, F, G, H) human IgG competition is shown. Figure 18 A, B, C, D) or with (E, F, G, H) human IgG competition is shown.

[0130] Figure 19 A-19H shows a comparison of the antibody-dependent cellular cytotoxicity activity of the MUC1xCD16A multispecific antibody BG1222P and huBG219P-E39-AF mediated by NK92mi / CD16A V158. In the absence (A, B, C, D) or presence (E, F, G, H) of human IgG competition. Figure 19 A, B, C, D) or with (E, F, G, H) human IgG competition is shown. Figure 19E, F, G, H) Activity was characterized in T47D, HCC827, H358, and MDA-MB-453 cells in the presence of human IgG competition.

[0131] Figure 20 Shown is a comparison of the cytolytic activity of the MUClxCD16A multispecific antibody BG1222P and various anti-MUCl non-fucosylated antibodies in human whole blood against the cancer cell line T47D.

[0132] Figure 21 A-21D shows the effects of the MUC1xCD16A multispecific antibody BG1222P and huBG219P-E39-AF on T47D ( Figure 21 A), HCC827( Figure 21 B) H358( Figure 21 C) and MDA-MB-453( Figure 21 D) Comparison of cytolytic activity of cells.

[0133] Figure 22 A-22B shows the effect of human M2 macrophage-mediated MUC1xCD16A multispecific antibodies BG1222P and huBG219P-E39-AF on T47D ( Figure 22 A) and MDA-MB-453( Figure 22 B) Comparison of phagocytic activity of cells.

[0134] Figure 23 A and 23B show that the MUC1xCD16A multispecific antibody BG1222P and daratumumab induce NK fratricide in NK cells.

[0135] Figure 24 Shown are the pharmacokinetic profiles of the MUC1xCD16A multispecific antibody BG1222P in cynomolgus monkeys following intravenous infusion of 5 mg / kg and 25 mg / kg BG1222P.

[0136] Figures 25A-25C The effect of soluble MUC1 on the binding of the MUC1xCD16A multispecific antibody to MUC1-expressing cells is shown. Briefly, human MUC1-expressing cells were incubated with 30, 3, and 0.3 μg / ml BG1222P in the presence of serially diluted soluble MUC1 (Shanghai Linc-Bio Science Co. LTD), and then after washing and incubation with an anti-human IgG secondary antibody, fluorescence was measured by flow cytometry. Figure 25A- C shows the binding profile of BG1222P at different concentrations (30, 3 and 0.3 pg / ml respectively) in the presence of soluble MUC1, with HuVH-HMFG1 as positive control and mlgG and hlgG1 as negative controls. DETAILED DESCRIPTION

[0137] DEFINITIONS

[0138] Unless defined otherwise herein or in this document, all other technical and scientific terms used herein have the meaning commonly understood by one of ordinary skill in the art.

[0139] As used herein, including the appended claims, the singular forms "a," "an," and "the" include their corresponding plural references unless the context clearly dictates otherwise.

[0140] The term "or" as used herein, unless stated otherwise, is used to mean and / or, and is not meant to exclude the presence of claimable elements.

[0141] Unless specifically stated or otherwise clearly contradicted by context, as used herein the term "about" refers to a value or composition that is within a reasonable error range for the value or composition as determined by one of ordinary skill in the art, which will depend in part on how the value or composition is measured or determined, i.e., the limitations of the measurement system. For example, "about" can mean within one or more standard deviations, according to practice in the art. "About" can mean a range of up to 10% (i.e., ±10%). Thus, "about" can be understood to be within 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.05%, 0.01%, or 0.001% more or less than the stated value. For example, about 5 mg can include any amount between 4.5 mg and 5.5 mg. Also, especially with respect to biological systems or processes, the term can mean up to an order of magnitude or up to 5-fold of a value. When a particular value or composition is provided in the disclosure, the meaning of "about" should be assumed to be within an acceptable error range of the particular value or composition, unless otherwise stated.

[0142] The term "MUC1" or "mucin 1", also known as CA 15-3, EMA, MCD, PEM, PUM, KL-6, MAM6, MCKD, PEMT, CD227, H23AG, MCKD1, ADMCKD, or ADTKD2, is a member of the mucin family. The amino acid sequence of human MUC1 is set forth as SEQ ID NO: 1 and is also found at accession number P15941.

[0143] The term "CD16A" refers to a type I membrane protein with two Ig-like domains of low affinity for IgG and is also known as FCGRIIIA and FCGR3A. The amino acid sequence of human CD16A (P08637) can be found in Uniprot database at Uniprot P08637.

[0144] The term "administration" or "administering" as used herein, when applied to an animal, human, subject, cell, tissue, organ, or biological fluid, means the contact of an exogenous drug, therapeutic agent, diagnostic agent, or composition with the animal, human, subject, cell, tissue, organ, or biological fluid. Treatment of a cell encompasses contact of the agent with the cell, as well as contact of the agent with a fluid, where the fluid is in contact with the cell.

[0145] The term "subject" or "patient" herein includes any organism, preferably an animal, more preferably a mammal (e.g., rat, mouse, dog, cat, rabbit, primate) and most preferably a human (e.g., a patient including or at risk of having a disorder described herein).

[0146] "Treatment" of any disease or disorder refers in one aspect to ameliorating the disease or disorder (i.e., slowing or arresting or reducing the development of the disease or at least one of the clinical symptoms thereof). In another aspect, "treat," "treating," or "treatment" refers to the alleviation or amelioration of at least one physical parameter, including those that cannot be discerned by the patient. In another aspect, "treat," "treating," or "treatment" refers to the modulation (e.g., stabilization) of the disease or disorder, either physically, (e.g., stabilization of a discernible symptom), physiologically, (e.g., stabilization of a physical parameter), or both.

[0147] The term "affinity" as used herein refers to the strength of the interaction between an antibody and an antigen. Within an antigen, the variable region of an antibody interacts with the antigen at multiple sites through noncovalent forces. Generally, the greater the interaction, the stronger the affinity.

[0148] As used herein, the term "antibody" refers to a polypeptide of the immunoglobulin family that can bind to a corresponding antigen non-covalently, reversibly, and specifically. For example, naturally occurring IgG antibodies are tetramers comprising at least two heavy (H) chains and two light (L) chains interconnected by disulfide bonds. Each heavy chain consists of a heavy chain variable region (abbreviated herein as VH) and a heavy chain constant region. The heavy chain constant region consists of three domains, CH1, CH2, and CH3. Each light chain consists of a light chain variable region (abbreviated herein as VL or Vκ) and a light chain constant region. The light chain constant region consists of one domain, CL. The VH and VL regions can be further subdivided into hypervariable regions, called complementarity determining regions (CDRs), interspersed with more conserved regions called framework regions (FRs). Each VH and VL consists of three CDRs and four framework regions (FRs), arranged from amino terminus to carboxyl terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, and FR4. The variable regions of the heavy and light chains contain a binding domain that interacts with an antigen. The constant regions of antibodies mediate the binding of immunoglobulins to host tissues or factors, including various cells of the immune system (e.g., effector cells) and the first component (Clq) of the classical complement system.

[0149] The positions of the CDRs and framework regions can be defined using various definitions well known in the art, such as Kabat, Chothia, AbM, and IMGT (see, e.g., Johnson et al., Nucleic Acids Res., 29:205-206 (2001); Chothia and Lesk, J. Mol. Biol., 196:901-917 (1987); Chothia et al., Nature, 342:877-883 (1989); Chothia et al., J. Mol. Biol., 227:799-817 (1992); Al-Lazikani et al., J. Mol. Biol., 273:927-748 (1997); Lefranc, M.-P., The Immunologist, 7, 132-136 (1999); Lefranc, M.-P. et al., Dev. Comp. Immunol., 27, 55-77 (2003)).

[0150] The term "antibody" includes, but is not limited to, monoclonal antibodies, human antibodies, humanized antibodies, chimeric antibodies, and anti-idiotypic (anti-Id) antibodies. The antibody can be of any isotype / class (e.g., IgG, IgE, IgM, IgD, IgA, and IgY), or subclass (e.g., IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2).

[0151] In some embodiments, the anti-MUC1 antibody comprises at least one antigen binding site. In some embodiments, the anti-MUC1 antibody comprises an antigen binding fragment from a MUC1 antibody described herein. In some embodiments, the anti-MUC1 antibody is isolated or recombinant.

[0152] In some embodiments, the anti-CD16A antibody comprises at least one antigen binding site, at least one variable region. In some embodiments, the anti-CD16A antibody comprises an antigen binding fragment from a CD16A antibody described herein. In some embodiments, the anti-CD16A antibody is isolated or recombinant.

[0153] The term "monoclonal antibody" or "mAb" or "Mab" herein refers to a population of substantially homogeneous antibodies, i.e., the antibody molecules comprising the population are identical in amino acid sequence except for possible naturally occurring mutations that can be present in minor amounts. In contrast, conventional (polyclonal) antibody preparations typically include a multitude of different antibodies with different amino acid sequences in their variable domains, particularly their CDRs, that are often specific for different epitopes. The modifier "monoclonal" indicates the character of the antibody as being obtained from a substantially homogeneous population of antibodies, and should not be construed as requiring production of the antibody by any particular method. Monoclonal antibodies (mAbs) can be obtained by methods known to those skilled in the art. See, e.g., Kohler et al., Nature 1975 256:495-497; U.S. Pat. No. 4,376,110; Ausubel et al., CURRENT PROTOCOLS IN MOLECULAR BIOLOGY 1992; Harlow et al., ANTIBODIES: A LABORATORY MANUAL, Cold spring Harbor Laboratory 1988; and Colligan et al., CURRENT PROTOCOLS IN IMMUNOLOGY 1993. The antibodies disclosed herein can be of any immunoglobulin class including IgG, IgM, IgD, IgE, IgA, and any subclass thereof such as IgGl, IgG2, IgG3, IgG4. Hybridomas producing monoclonal antibodies can be cultured in vitro or in vivo. High titer monoclonal antibodies can be obtained in vivo production, in which cells from individual hybridomas are injected intraperitoneally into mice, such as naive primed Balb / c mice, to produce ascites containing high concentrations of the desired antibody. Monoclonal antibodies of isotype IgM or IgG can be purified from such ascites or from culture supernatants using column chromatography methods well known to those skilled in the art.

[0154] Generally, the basic antibody structural unit comprises a tetramer. Each tetramer includes two identical pairs of polypeptide chains, each pair having one "light" (about 25 kDa) and one "heavy" chain (about 50-70 kDa). The amino-terminal portion of each chain includes a variable region of about 100 to 110 or more amino acids primarily responsible for antigen recognition. The carboxy-terminal portion of each heavy chain can define a constant region primarily responsible for effector functions. Generally, human light chains are classified as kappa and lambda light chains. Furthermore, human heavy chains are traditionally classified as alpha, delta, epsilon, gamma, and mu, and define the class of an antibody as IgA, IgD, IgE, IgG, and IgM, respectively.

[0155] Within each light and heavy chain, the variable and constant regions are joined by a "J" region of about 12 or more amino acids, and, in the heavy chain, by a "D" region of about 10 or more amino acids.

[0156] The variable regions of each light / heavy chain pair contribute to the formation of the antibody binding site. Thus, generally, an intact antibody has two binding sites. Except in bifunctional or bispecific antibodies, the two binding sites are typically identical in their primary sequence.

[0157] Generally, the variable domains of heavy and light chains comprise three hypervariable regions, also called "complementarity determining regions (CDRs)", located between relatively conserved framework regions (FRs). The CDRs are typically aligned by the framework regions, enabling binding to a particular epitope. Generally, from N- to C-terminus, the light and heavy chain variable domains comprise FR-1 (or FR1), CDR-1 (or CDR1), FR-2 (FR2), CDR-2 (CDR2), FR-3 (FR3), CDR-3 (CDR3), and FR-4 (FR4). The location of CDRs and framework regions can be determined using various well-known definitions in the art, such as Kabat, Chothia, AbM, and IMGT (see, e.g., Johnson et al., Nucleic Acids Res., 29:205-206 (2001); Chothia and Lesk, J. Mol. Biol., 196:901-917 (1987); Chothia et al., Nature, 342:877-883 (1989); Chothia et al., J. Mol. Biol., 227:799-817 (1992); Al-Lazikani et al., J. Mol. Biol., 273:927-748 (1997) ImMunoGenTics (IMGT) numbering (Lefranc, M.-P., The Immunologist, 7, 132-136 (1999); Lefranc, M.-P. et al., Dev. Comp. Immunol., 27, 55-77 (2003) ("IMGT" numbering scheme). Definitions of antigen combining sites are also described in Ruiz et al., Nucleic Acids Res., 28:219-221 (2000); and Lefranc, M.P., Nucleic Acids Res., 29:207-209 (2001); MacCallum et al., J. Mol. Biol., 262:732-745 (1996); and Martin et al., Proc. Natl. Acad. Sci. USA, 86:9268-9272 (1989); Martin et al., Methods Enzymol., 203:121-153 (1991); and Rees et al., In Sternberg M.J.E. (ed.), Protein Structure Prediction, Oxford University Press, Oxford, 141-172 (1996).For example, under Kabat, CDR amino acid residues in the heavy chain variable domain (VH) are numbered 31-35 (HCDR1), 50-65 (HCDR2), and 95-102 (HCDR3); and in the light chain variable domain (VL) are numbered 24-34 (LCDR1), 50-56 (LCDR2), and 89-97 (LCDR3). Under Chothia, CDR amino acid residues in the VH are numbered 26-32 (HCDR1), 52-56 (HCDR2), and 95-102 (HCDR3); and in the VL are numbered 26-32 (LCDR1), 50-52 (LCDR2), and 91-96 (LCDR3). By combining the CDR definitions of Kabat and Chothia, the CDRs consist of amino acid residues 26-35 (HCDR1), 50-65 (HCDR2), and 95-102 (HCDR3) in a human VH, and 24-34 (LCDR1), 50-56 (LCDR2), and 89-97 (LCDR3) in a human VL. Under IMGT, CDR amino acid residues in the VH are numbered approximately 26-35 (HCDR1), 51-57 (HCDR2), and 93-102 (HCDR3), and in the VL are numbered approximately 27-32 (LCDR1), 50-52 (LCDR2), and 89-97 (LCDR3) (numbering according to Kabat). Under IMGT, CDR regions of an antibody can be determined using the program IMGT / DomainGapAlign.

[0158] The term "hypervariable region" refers to amino acid residues of an antibody that are responsible for antigen binding. The hypervariable region comprises amino acid residues from a "CDR" (e.g., LCDR1, LCDR2, and LCDR3 in a light chain variable domain and HCDR1, HCDR2, and HCDR3 in a heavy chain variable domain). See Kabat et al., (1991) Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md. (definition of CDR regions of an antibody by sequence); see also Chothia and Lesk (1987) J. Mol. Biol. 196:901-917 (definition of CDR regions of an antibody by structure). The term "framework" or "FR" residues refer to those variable domain residues other than the hypervariable region residues defined in this document as CDR residues.

[0159] Unless otherwise indicated, "antigen-binding fragment" means an antigen-binding fragment of an antibody, i.e., a fragment of an antibody that retains the ability to specifically bind to the antigen bound by the full-length antibody, e.g., a fragment that retains one or more CDR regions. Examples of antigen-binding fragments include, but are not limited to, Fab, Fab', F(ab')2, and Fv fragments; diabodies; linear antibodies; single-chain antibody molecules, e.g., single-chain Fv (scFv); nanobodies (or VHH antibodies); multi-specific antibodies formed from antibody fragments; and bicycling peptides (Hurov, K. et al., 2021. Journal for ImmunoTherapy of Cancer, 9(11)).

[0160] As used herein, an antibody or antigen-binding antibody fragment "specifically binds to" an antigen (e.g., a protein) means that the antibody exhibits preferential binding to the target as compared to other proteins, but this specificity does not require absolute binding specificity. The "specific" or "selective" binding reaction is determinative of the presence of an antigen in a heterogeneous population of proteins and other biological products, e.g., in a biological sample, blood, serum, plasma, or tissue sample. Thus, under designated immunoassay conditions, an antibody or antigen-binding fragment thereof specifically binds to a particular antigen at least two-fold greater than background level and does not significantly bind in a specific manner to other antigens present in the sample. In one aspect, under designated immunoassay conditions, an antibody or antigen-binding fragment thereof specifically binds to a particular antigen at least ten-fold greater than background level and does not significantly bind in a specific manner to other antigens present in the sample.

[0161] An "antigen-binding domain" as used herein comprises at least six CDRs (or three CDRs in the case of a single domain antibody) and specifically binds to an epitope. An "antigen-binding domain" of a multi-specific antibody (e.g., a bispecific antibody) comprises a first antigen-binding domain that specifically binds to a first epitope and a second antigen-binding domain that specifically binds to a second epitope and also comprises at least three CDRs. A multi-specific antibody can be bispecific, trispecific, tetraspecific, etc., and an antigen-binding domain is directed to each specific epitope. A multi-specific antibody can be multivalent (e.g., bispecific tetravalent antibody) comprising multiple antigen-binding domains, e.g., 2, 3, 4, or more antigen-binding domains that specifically bind to a first epitope and 2, 3, 4, or more antigen-binding domains that specifically bind to a second epitope.

[0162] The term "human antibody" herein refers to an antibody comprising only human immunoglobulin sequences. A human antibody can contain murine carbohydrate chains if produced in a mouse, in a mouse cell, or in a hybridoma derived from a mouse cell. Similarly, a "mouse antibody" or "rat antibody" refers to an antibody comprising only mouse or rat immunoglobulin sequences, respectively.

[0163] The term "humanized" or "humanized antibody" refers to forms of antibodies that contain sequences from non-human (e.g., murine) antibodies as well as human antibodies. Such antibodies contain minimal sequence derived from a non-human immunoglobulin. For the most part, humanized antibodies are human immunoglobulins (recipient antibody) in which hypervariable regions comprise the sequence of a non-human immunoglobulin (donor antibody). Generally, the humanized antibody will comprise substantially all of at least one, and typically two, variable domains, in which all or substantially all of the hypervariable loops correspond to those of a non-human immunoglobulin and all or substantially all of the FR regions are those of a human immunoglobulin sequence. The humanized antibody optionally also will comprise at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin. When necessary, the prefix "hum," "hu," "Hu," or "h" is added to the antibody clone designation to distinguish the humanized antibody from the parent rodent antibody. Humanized forms of rodent antibodies typically comprise the same CDR sequences as the parent rodent antibody, but can include certain amino acid substitutions to increase affinity, increase stability of the humanized antibody, remove post-translational modifications, or for other reasons.

[0164] The term "corresponding human germline sequence" refers to a nucleic acid sequence encoding a human variable region amino acid sequence or subsequence that shares the highest determined amino acid sequence identity to a reference variable region amino acid sequence or subsequence compared to all other known variable region amino acid sequences encoded by human germline immunoglobulin variable region sequences. A corresponding human germline sequence can also refer to a human variable region amino acid sequence or subsequence that has the highest amino acid sequence identity to a reference variable region amino acid sequence or subsequence compared to all other assessed variable region amino acid sequences. The corresponding human germline sequence can be a framework region only, a complementarity determining region only, a framework and complementarity determining region, a variable region, or other combination of sequences or subsequences. Sequence identity can be determined using the methods described herein, e.g., using BLAST, ALIGN, or another alignment algorithm known in the art, to align two sequences. A corresponding human germline nucleic acid or amino acid sequence can have at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to a reference variable region nucleic acid or amino acid sequence. In addition, if the antibody contains a constant region, the constant region is also derived from these human sequences, e.g., human germline sequences, or mutated versions of human germline sequences or antibodies containing consensus framework sequences derived from an analysis of human framework sequences as described in Knappik et al., J. Mol. Biol. 296:57-86, 2000.

[0165] The term "equilibrium dissociation constant" or "KD" or "M" refers to the dissociation rate constant (kd, time 1 ) divided by the association rate constant (ka, time -1 , M -l). Equilibrium dissociation constants can be measured using any known method in the art. Antibodies of the present disclosure typically have equilibrium dissociation constants of less than about 10 -7 M, for example, less than about 10 -8 M, for example, less than about 10 -9 M or 10 -10 M, in some aspects, less than about 10 -11 M, 10 - 12 M or 10 -13 M.

[0166] The term "cancer" or "tumor" as used herein has the broadest meaning understood in the art and refers to a physiological disorder in a mammal that is typically characterized by unregulated cell growth. In the context of the present disclosure, cancer is not limited to a certain type or location.

[0167] In the context of the present disclosure, the term "conservative substitution" when referring to an amino acid sequence means that the original amino acid is replaced with a new amino acid that does not substantially alter the chemical, physical, and / or functional properties of the antibody or fragment, e.g., its binding affinity to MUC1 or CD16A. Common conservative substitutions of amino acids are well known in the art.

[0168] The term "knob-into-hole" technology as used herein refers to the introduction of amino acids that direct two polypeptides to pair together in vitro or in vivo at the interface of a polypeptide interaction by introducing a steric bulge (knob) into one polypeptide and a pocket or cavity (hole) into the other polypeptide. For example, knob-into-hole has been introduced in the Fc:Fc binding interface, the C L :C H I interface, or the VH / VL interface of an antibody (see, e.g., US2011 / 0287009, US2007 / 0178552, WO 96 / 027011, WO 98 / 050431, and Zhu et al., 1997, Protein Science, 6:781-788). In some embodiments, knob-into-hole ensures that two different heavy chains are correctly paired together during the manufacture of a multispecific antibody. For example, a multispecific antibody having knob-into-hole amino acids in its Fc region can further comprise a single variable domain linked to each Fc region, or further comprise different heavy chain variable domains paired with similar or different light chain variable domains. The knob-into-hole technology can also be used in the VH or VL region to likewise ensure correct pairing.

[0169] An example of an algorithm that is suitable for determining percent sequence identity and sequence similarity is the BLAST algorithm, which is described in Altschul et al., Nuc. Acids Res. 25:3389-3402, 1977; and Altschul et al., J. Mol. Biol. 215:403-410, 1990. Software for performing BLAST analyses is publicly available through the National Center for Biotechnology Information. The algorithm involves first identifying high scoring sequence pairs (HSPs) by identifying short words in the query sequence that either match or satisfy some positive-valued threshold score T when aligned with a word of the same length in a database sequence. T is referred to as the neighborhood word score threshold. These initial neighborhood word hits act as seeds for initiating searches to find longer HSPs containing them. The word hits are extended in both directions along each sequence for as far as the cumulative alignment score can be increased. Cumulative scores are calculated using, for nucleotide sequences, the parameters M (reward score for a pair of matching residues; always > 0) and N (penalty score for mismatching residues; always < 0). For amino acid sequences, a matrix is used to calculate the cumulative score. Extension of the word hits in each direction are halted when: the cumulative alignment score falls off by the quantity X from its maximum achieved value; the cumulative score goes to zero or below, due to the accumulation of one or more negative-scoring residue alignments; or the end of either sequence is reached. The BLAST algorithm parameters W, T, and X determine the sensitivity and speed of the alignment. The BLASTN program (for nucleotide sequences) uses as defaults a word length (W) of 11, an expectation value (E) of 10, M=5, N=-4, and a comparison of both strands. For amino acid sequences, the BLAST program uses as defaults a word length of 3 and an expectation value (E) of 10, and the BLOSUM62 scoring matrix (see Henikoff and Henikoff, (1989) Proc. Natl. Acad. Sci. USA 89:10915) of 50, a comparison of both strands.

[0170] The BLAST algorithm also performs a statistical analysis of the similarity between two sequences (see, e.g., Karlin and Altschul, Proc. Natl. Acad. Sci. USA 90:5873-5787, 1993). One measure of similarity provided by the BLAST algorithm is the smallest sum probability (P(N)), which provides an indication of the probability by which an match between two nucleotide or amino acid sequences would occur by chance. For example, a nucleic acid is considered similar to a reference sequence if the smallest sum probability in a comparison of the test nucleic acid to the reference nucleic acid is less than about 0.2, more preferably less than about 0.01, and most preferably less than about 0.001.

[0171] The percent identity between two amino acid sequences can also be determined using the algorithm of E. Meyers and W. Miller, Comput. Appl. Biosci. 4: 11-17, (1988) incorporated into the ALIGN program (version 2.0), using a PAM120 weight residue table, a gap length penalty of 12, and a gap penalty of 4. In addition, the percent identity between two amino acid sequences can be determined using the Needleman and Wunsch, J. Mol. Biol. 48: 444-453, (1970) algorithm incorporated into the GAP program using either a BLOSUM62 matrix or a PAM250 matrix, and a gap weight of 16, 14, 12, 10, 8, 6, or 4 and a gap length penalty of 1, 2, 3, 4, 5, or 6.

[0172] The term "nucleic acid" is used interchangeably herein with the term "polynucleotide" and refers to deoxyribonucleotides or ribonucleotides and polymers thereof in either single- or double-stranded form. The term encompasses nucleic acids containing known nucleotide analogs or modified backbone residues or linkages, which are synthetic, naturally occurring, and non-naturally occurring, which have similar binding properties to reference nucleotides, and which are metabolized in a manner similar to reference nucleotides. Examples of such analogs include, without limitation, phosphorothioates, phosphoramidates, phosphoramidites, chiral-moieties phosphoramidites, 2-O-methyl ribonucleotides, peptide-nucleic acids (PNAs).

[0173] In the context of nucleic acids, the term "operably linked" refers to a functional relationship between two or more polynucleotide (e.g., DNA) segments. Typically, it refers to a functional relationship between a transcriptional regulatory sequence and a transcribed sequence. For example, a promoter or enhancer sequence is operably linked to a coding sequence if it stimulates or modulates the transcription of the coding sequence in a suitable host cell or other expression system. In general, a promoter transcriptional regulatory sequence that is operably linked to a transcribed sequence is physically contiguous with the transcribed sequence, i.e., it is cis-acting. However, some transcriptional regulatory sequences, such as enhancers, are not necessarily physically contiguous or positioned immediately adjacent to the coding sequence whose transcription they enhance.

[0174] In some aspects, the present disclosure provides compositions, e.g., pharmaceutically acceptable compositions, comprising an anti-MUC1xCD16A multispecific antibody as described herein formulated together with at least one pharmaceutically acceptable excipient. As used herein, the term "pharmaceutically acceptable excipient" includes any and all solvents, dispersion media, isotonic and absorption delaying agents, and the like that are physiologically compatible. The excipients can be suitable for intravenous, intramuscular, subcutaneous, gastrointestinal, rectal, spinal or epidermal administration (e.g., by injection or infusion).

[0175] The compositions disclosed herein can be in various forms. These include, for example, liquid, semi-solid and solid dosage forms, such as liquid solutions (e.g., injectable and infusible solutions), dispersions or suspensions, liposomes, and suppositories. The appropriate form depends on the intended mode of administration and therapeutic application. One suitable mode of administration is gastrointestinal (e.g., intravenous, subcutaneous, intraperitoneal, intramuscular). In some embodiments, the antibody is administered by intravenous infusion or injection. In certain embodiments, the antibody is administered by intramuscular or subcutaneous injection.

[0176] The term "therapeutically effective amount" as used herein refers to the amount of an antibody that, when administered to a subject for treating a disease, or at least one of the clinical symptoms of a disease or disorder, is sufficient to effect such treatment for the disease, disorder, or symptom. The "therapeutically effective amount" can vary with the antibody, the disease, disorder, and / or symptom of the disease or disorder, the severity of the disease, disorder, and / or symptom of the disease or disorder, the age of the subject to be treated, and / or the body weight of the subject to be treated. An appropriate amount in any given instance can be readily determined by one of ordinary skill in the art or can be determined by routine experimentation. In the case of combination therapy, "therapeutically effective amount" refers to the total amount of the combination components.

[0177] The term "combination therapy" refers to the administration of two or more therapeutic agents to treat a therapeutic condition or disorder. Such administration encompasses coadministration of these therapeutic agents in a substantially simultaneous manner, such as in a single dosage form. Such administration also encompasses use of each type of therapeutic agent in a sequential manner in the course of treatment. In either case, the treatment regimen provides beneficial effects of the drug combination in treating the conditions or disorders described herein.

[0178] As used herein, the phrase "in combination with" means that the anti-MUC1xCD16A multispecific antibody is administered to the subject at about the same time as, just prior to, or just after the additional therapeutic agent. In certain embodiments, the anti-MUC1xCD16A multispecific antibody is administered as a co-formulation with the additional therapeutic agent.

[0179] The present disclosure provides antibodies, antigen-binding fragments, and anti-MUC1xCD16A multispecific antibodies. In addition, the present disclosure provides antibodies that have desirable pharmacokinetic properties and other desirable attributes, and thus are useful for reducing the likelihood of cancer or treating cancer. The present disclosure further provides pharmaceutical compositions comprising the antibodies and methods of making and using such pharmaceutical compositions to prevent and treat cancer and related disorders.

[0180] Anti-MUC1 antibodies

[0181] The present disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to MUC1. The antibodies or antigen-binding fragments of the present disclosure include, but are not limited to, antibodies or antigen-binding fragments thereof produced as described below.

[0182] The present disclosure provides antibodies or antigen-binding fragments that specifically bind to MUC1, wherein the antibody or antibody fragment (e.g., antigen-binding fragment) comprises a VH domain having an amino acid sequence set forth in Table 2 and / or 8. The present disclosure also provides antibodies or antigen-binding fragments that specifically bind to MUC1, wherein the antibody or antigen-binding fragment comprises a HCDR having an amino acid sequence of any one of the HCDRs set forth in Table 2 and 8. In one aspect, the present disclosure provides antibodies or antigen-binding fragments that specifically bind to MUC1, wherein the antibody comprises (or alternatively consists of) one, two, three or more HCDRs having an amino acid sequence of any one of the HCDRs set forth in Table 2 and 8.

[0183] The present disclosure provides antibodies or antigen-binding fragments that specifically bind to MUC1, wherein the antibody or antigen-binding fragment comprises a VL domain having an amino acid sequence set forth in Table 2 and / or 8. The present disclosure also provides antibodies or antigen-binding fragments that specifically bind to MUC1, wherein the antibody or antigen-binding fragment comprises a LCDR having an amino acid sequence of any one of the LCDRs set forth in Table 2 and 8. In particular, the present disclosure provides antibodies or antigen-binding fragments that specifically bind to MUC1, comprising (or alternatively consisting of) one, two, three or more LCDRs having an amino acid sequence of any one of the LCDRs set forth in Table 2 and 8.

[0184] Other antibodies or antigen-binding fragments thereof of the present disclosure include those in which amino acids or nucleic acids encoding amino acids have been changed, but are at least 60%, 70%, 80%, 90%, 95%, or 99% identical to the CDR regions disclosed in Table 2 and 8. In some aspects, it includes changes in amino acid sequences in which no more than 1, 2, 3, 4, or 5 amino acids in the CDR regions have been changed when compared to the CDR regions described in the sequences in Table 2 and 8, while retaining substantially the same therapeutic activity.

[0185] Other antibodies of the present disclosure include those in which amino acids or nucleic acids encoding amino acids have been changed, but are at least 60%, 70%, 80%, 90%, 95%, or 99% identical to the sequences described in Table 2 and 8. In some aspects, it includes changes in amino acid sequences in which no more than 1, 2, 3, 4, or 5 amino acids in the variable regions have been changed when compared to the variable regions described in the sequences described in Table 2 and 8, while retaining substantially the same therapeutic activity.

[0186] The present disclosure also provides nucleic acid sequences encoding the VH, VL, full length heavy chain, and full length light chain of antibodies that specifically bind to MUC1. Such nucleic acid sequences can be optimized for expression in mammalian cells.

[0187] The present disclosure provides antibodies and antigen-binding fragments thereof that bind to an epitope of human MUC1. In certain aspects, the antibodies and antigen-binding fragments can bind to the same epitope of MUC1.

[0188] The present disclosure also provides antibodies and antigen-binding fragments thereof that bind to the same epitope as the anti-MUC1 antibodies described in Tables 2 and 8. Thus, additional antibodies and antigen-binding fragments thereof can be identified based on their ability to cross-compete (e.g., competitively inhibit the binding of) other antibodies in a binding assay. Demonstrating the ability of a test antibody to inhibit the binding of an antibody and antigen-binding fragment thereof of the present disclosure to MUC1 demonstrates that the test antibody can compete with the antibody or antigen-binding fragment thereof for binding to MUC1. Without being bound by any theory, such an antibody can bind to the same or related (e.g., structurally similar or spatially proximal) epitope on MUC1 as the antibody or antigen-binding fragment thereof with which it competes. In certain aspects, an antibody that binds to the same epitope on MUC1 as an antibody or antigen-binding fragment thereof of the present disclosure is a human or humanized monoclonal antibody. Such human or humanized monoclonal antibodies can be prepared and isolated as described herein.

[0189] Anti-CD16A antibodies

[0190] The present disclosure provides antibodies or antigen-binding fragments thereof that specifically bind to CD16A. The antibodies or antigen-binding fragments thereof of the present disclosure include, but are not limited to, antibodies or antigen-binding fragments thereof produced as described below.

[0191] The present disclosure provides antibodies or antigen-binding fragments that specifically bind to CD16A, wherein the antibody or antibody fragment (e.g., antigen-binding fragment) comprises a VH domain having an amino acid sequence set forth in Table 23. The present disclosure also provides antibodies or antigen-binding fragments that specifically bind to CD16A, wherein the antibody or antigen-binding fragment comprises a HCDR having an amino acid sequence of any one of the HCDRs set forth in Table 23. In one aspect, the present disclosure provides antibodies or antigen-binding fragments that specifically bind to CD16A, wherein the antibody comprises (or alternatively consists of) one, two, three, or more HCDRs having an amino acid sequence of any one of the HCDRs set forth in Table 23.

[0192] Other antibodies or antigen-binding fragments thereof of the present disclosure include those in which amino acids or nucleic acids encoding amino acids have been changed, but which have at least 60%, 70%, 80%, 90%, 95%, or 99% identity to the sequences described in Table 23. In some aspects, these include changes in the amino acid sequence in which no more than 1, 2, 3, 4, or 5 amino acids in the variable region have been changed when compared to the variable region described in the sequences in Table 23, while retaining substantially the same therapeutic activity.

[0193] Other antibodies of the present disclosure include those in which amino acids or nucleic acids encoding amino acids have been changed, but which have at least 60%, 70%, 80%, 90%, 95%, or 99% identity to the sequences described in Table 23. In some aspects, these include changes in the amino acid sequence in which no more than 1, 2, 3, 4, or 5 amino acids in the variable region have been changed when compared to the variable region described in the sequences in Table 23, while retaining substantially the same therapeutic activity.

[0194] The present disclosure also provides nucleic acid sequences encoding the VH, VL, full length heavy chain, and full length light chain of antibodies that specifically bind to CD16A. These nucleic acid sequences can be optimized for expression in mammalian cells.

[0195] The present disclosure provides antibodies and antigen-binding fragments thereof that bind to an epitope of human CD16A. In certain aspects, the antibodies and antigen-binding fragments can bind to the same epitope of CD16A.

[0196] The present disclosure also provides antibodies and antigen-binding fragments thereof that bind to the same epitope as the anti-CD16A antibodies described in Table 23. Thus, additional antibodies and antigen-binding fragments thereof can be identified based on their ability to cross-compete (e.g., competitively inhibit their binding in a statistically significant manner) with other antibodies in a binding assay. Demonstrating the ability of a test antibody to inhibit the binding of an antibody and antigen-binding fragment thereof of the present disclosure to CD16A demonstrates that the test antibody can compete with the antibody or antigen-binding fragment thereof for binding to CD16A. Without being bound by any theory, such an antibody can bind to the same or related (e.g., structurally similar or spatially proximal) epitope on CD16A as the antibody or antigen-binding fragment thereof with which it competes. In certain aspects, an antibody that binds to the same epitope on CD16A as an antibody or antigen-binding fragment thereof of the present disclosure is a human or humanized monoclonal antibody. Such human or humanized monoclonal antibodies can be prepared and isolated as described herein.

[0197] Anti-MUC1xCD16A multispecific antibodies

[0198] In embodiments, the anti-MUC1 and anti-CD16A antibodies disclosed herein can be incorporated into an anti-MUC1xCD16A multispecific antibody. If, for example, the antibody comprises multiple antigen binding domains, it is a multispecific antibody, wherein at least one antigen binding domain sequence specifically binds MUC1 as a first epitope and a second antigen binding domain sequence specifically binds CD16A as a second epitope. In embodiments, the multispecific antibody comprises a third, fourth, or fifth antigen binding domain. In embodiments, the multispecific antibody is a bispecific antibody, a trispecific antibody, or a tetraspecific antibody. In each embodiment, the multispecific antibody comprises at least one anti-MUC1 antigen binding domain and at least one anti-CD16A antigen binding domain.

[0199] In one embodiment, the multispecific antibody is a bispecific antibody. As used herein, a bispecific antibody specifically binds only two antigens. The bispecific antibody comprises a first antigen binding domain that specifically binds MUC1 and a second antigen binding domain that specifically binds CD16A. Included are bispecific antibodies comprising a heavy chain variable domain and a light chain variable domain that specifically bind MUC1 as a first epitope and a heavy chain variable domain that specifically binds CD16A as a second epitope. In another embodiment, the bispecific antibody comprises an antigen binding fragment that specifically binds MUC1 and an antigen binding fragment that specifically binds CD16A. When the bispecific antibody comprises antigen binding fragments, the antigen binding fragments can be Fabs, F(ab')2s, Fvs, or single chain Fvs (scFvs).

[0200] Previous experiments (Coloma and Morrison, Nature Biotech. 15: 159-163 (1997)) describe tetravalent bispecific antibodies engineered by fusing DNA encoding a single chain anti-dansyl antibody Fv (scFv) either after the C-terminus of an IgG3 anti-dansyl antibody (CH3-scFv) or after the hinge (hinge-scFv). The present disclosure provides multivalent antibodies (e.g., tetravalent antibodies) having at least two antigen binding domains that can be readily produced by recombinant expression of nucleic acids encoding the polypeptide chains of the antibody. The multivalent antibodies herein comprise three to eight, but preferably four, antigen binding domains that specifically bind at least two antigens.

[0201] Linker

[0202] The domains and / or regions of the polypeptide chains of the bispecific tetravalent antibodies disclosed herein can be separated by linker regions of various lengths. In some embodiments, the antigen binding domains are separated from each other, CL, CH1, hinge, CH2, CH3, or the entire Fc region by linker regions. For example, the polypeptide chains can include the sequence VL1-CL-(linker)VH2-CH1, VH-linker-VL. Such linker regions can comprise randomly assorted amino acids or a restricted set of amino acids. Such linker regions can be flexible or rigid (see US 2009 / 0155275).

[0203] Multispecific antibodies are constructed by genetically fusing two single chain Fv (scFv) or Fab fragments with or without the use of flexible linkers (Mallender et al., J. Biol. Chem. 1994 269:199-206; Mack et al., Proc. Natl. Acad. Sci. USA. 1995 92:7021-5; Zapata et al., Protein Eng. 1995 8.1057-62); via dimerization devices such as leucine zippers (Kostelny et al., J. Immunol. 1992 148:1547-53; de Kruif et al J. Biol. Chem. 1996 271:7630-4); Ig C / CH1 domains (Muller et al., FEBS Lett. 422:259-64); by diabodies (Holliger et al., (1993) Proc. Nat. Acad. Sci. USA. 1998 90:6444-8; Zhu et al., Bio / Technology (NY) 1996 14:192-6); Fab-scFv fusions (Schoonjans et al., J. Immunol. 2000 165:7050-7); and minibody formats (Pack et al., Biochemistry 1992. 31:1579-84; Pack et al., Bio / Technology 1993 11:1271-7).

[0204] The bispecific tetravalent antibodies as disclosed herein comprise a linker region of at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75 or more amino acid residues between one or more of their antigen binding domains, CL domains, CHI domains, hinge regions, CH2 domains, CH3 domains, or Fc regions. In some embodiments, the linker region consists of the amino acids glycine and serine. Linkers may include the sequences GS, GG S, GSG, SGG, GGG, GGGS (SEQ ID NO: 149), SGGG (SEQ ID NO: 150), GGGGS (SEQ ID NO: 151), GGGGSGS (SEQ ID NO: 152), GGGGSGS (SEQ ID NO: 153), GGGGSGGS (SEQ ID NO: 154), GGGGSGGGGS (SEQ ID NO: 154), NO:155), GGGGSGGGGSGG GGS (SEQ ID NO:156), AKTTPKLEEGEFSEAR (SEQ ID NO:157), AKTTPKLEEGEFSEARV (SEQ ID NO:158), AKTTPKLGG (SEQ ID NO:159), SAKTTPKLGG (SEQ ID NO:160), AKTTPKLEEGEFSEARV (SEQ ID NO:161), SAKTTP (SEQ ID NO:162), SAKTTPKLGG (SEQ ID NO:163), RADAAP (SEQ ID NO:164), RADAAPTVS (SEQ ID NO:165), RADAAAAGGPGS (SEQ ID NO:166), RADAAAA(G4S)4 (SEQ ID NO:167), SAKTTP (SEQ ID NO:168), SAKTTPKLGG (SEQ ID NO:169), SAKTTPKLEEG EFSEARV(SEQ ID NO:170), ADAAP(SEQ ID NO:171), ADA APTVSIFPP(SEQ ID NO:172), TVAAP(SEQ ID NO:173), TV AAPSVFIFPP(SEQ ID NO:174), QPKAAP(SEQ ID NO:175), QPKAAPSVTLFPP(SEQ ID NO:176), AKTTPP(SEQ ID NO:177), AKTTPPSVTPLAP(SEQ IDAKTTAP (SEQ ID NO: 179), AKTTAPSVYPLAP (SEQ ID NO: 180), ASTKGP (SEQ ID NO: 181), ASTKGPSVFPLAP (SEQ ID NO: 182), GENKVEYAP ALMALS (SEQ ID NO: 183), GPAKELTPLKEAKVS (SEQ ID NO: 184), or GHEAAAVMQVQYPAS (SEQ ID NO: 185) or any combination thereof (see WO 2007 / 024715).

[0205] Dimerization-specific amino acids

[0206] In one embodiment, the multivalent antibody comprises at least one dimerization-specific amino acid change. The dimerization-specific amino acid change can create a “knob-into-hole” interaction and can increase the likelihood of correctly assembled multivalent antibodies. The dimerization-specific amino acids can be within the CH1 domain or the CL domain or a combination thereof. Suitable dimerization-specific amino acids for pairing CH1 domains with other CH1 domains (CH1-CH1) and CL domains with other CL domains (CL-CL) can be found in the disclosure of at least WO2014082179, WO2015181805, and WO2017059551. The dimerization-specific amino acids can also be within the Fc domain and can be combined with dimerization-specific amino acids within the CH1 or CL domains. In one embodiment, the present disclosure provides bispecific antibodies comprising at least one pair of dimerization-specific amino acids.

[0207] Further alterations of the framework of the Fc region

[0208] In an aspect, the Fc region is altered by replacing at least one amino acid residue with a different amino acid residue to alter the effector function of the antibody. For example, one or more amino acids can be replaced with a different amino acid residue such that the antibody has an altered affinity for an effector ligand but retains the antigen binding ability of the parent antibody. The effector ligand for which the affinity is altered can be, for example, an Fc receptor or the Cl component of complement. This approach is described in, for example, U.S. Patent Nos. 5,624,821 and 5,648,260 to Winter et al.

[0209] In another aspect, one or more amino acid residues can be replaced with one or more different amino acid residues such that the antibody has altered Clq binding and / or reduced or eliminated complement dependent cytotoxicity (CDC). This approach is described in, for example, U.S. Patent No. 6,194,551 to Idusogie et al.

[0210] In another aspect, one or more amino acid residues are changed to thereby alter the ability of the antibody to fix complement. This approach is described in, for example, the publication of Bodmer et al. WO 94 / 29351. In a particular aspect, one or more amino acids of an antibody or antigen-binding fragment thereof of the present disclosure are replaced by one or more allotypic amino acid residues of the IgGl subclass and kappa isotype. Allotypic amino acid residues also include, but are not limited to, the constant region of the heavy chain of the IgGl, IgG2, and IgG3 subclasses and the constant region of the light chain of the kappa isotype, as described by Jefferis et al., MAbs. 1 : 332-338 (2009).

[0211] In another aspect, the Fc region is modified by modifying one or more amino acids to increase the ability of the antibody to mediate antibody-dependent cellular cytotoxicity (ADCC) and / or increase the affinity of the antibody for Fcy receptors. This approach is described in, for example, the publication of Presta WO 00 / 42072. In addition, the binding sites on human IgGl for FcyRI, FcyRII, FcyRIII, and FcRn have been mapped and variants with improved binding have been described (see Shields et al., J. Biol. Chem. 276: 6591-6604, 2001).

[0212] In yet another aspect, the glycosylation of the multispecific antibody is modified. For example, an aglycosylated antibody can be made (i.e., an antibody lacking or having reduced glycosylation). Glycosylation can be altered, for example, to increase the affinity of the antibody for "antigen." Such carbohydrate modifications can be accomplished by, for example, altering one or more of the glycosylation sites in the antibody sequence. For example, one or more amino acid substitutions can be introduced into the variable region framework that result in elimination of one or more variable region framework glycosylation sites, thereby eliminating glycosylation at this site(s). This aglycosylation can increase the affinity of the antibody for antigen. This approach is described in, for example, U.S. Patent Nos. 5,714,350 and 6,350,861 of Co et al.

[0213] Additionally or alternatively, antibodies can be made that have altered types of glycosylation, such as hypofucosylated antibodies having reduced amounts of fucosyl residues or antibodies with increased bisecting GlcNac structures. Such altered glycosylation patterns have been demonstrated to increase the ADCC ability of antibodies. Such carbohydrate modifications can be accomplished by expressing the antibody in a host cell with altered glycosylation pathway. Cells with altered glycosylation pathway have been described in the art and can be used as host cells in which to express recombinant antibodies, thereby producing antibodies with altered glycosylation. For example, Hang et al. EP 1,176,195 describes a cell line with a functionally disrupted FUT8 gene, which encodes a fucosyl transferase, such that antibodies expressed in this cell line exhibit hypofucosylation. Presta's publication WO 03 / 035835 describes a variant CHO cell line, Lecl3 cells, which have a reduced ability to attach fucose to Asn(297)-linked carbohydrates, also resulting in hypofucosylation of antibodies expressed in the host cell (see also Shields et al., (2002) J. Biol. Chem. 277:26733-26740). Umana et al. WO 99 / 54342 describes cell lines engineered to express a glycoprotein-modifying glycosyl transferase, such as beta(l,4)-N-acetylglucosaminyltransferase III (GnTIII), such that antibodies expressed in the engineered cell line exhibit increased bisecting GlcNac structures, resulting in increased ADCC activity of the antibody (see also Umana et al., Nat. Biotech. 17:176-180, 1999).

[0214] In another aspect, if reduction of ADCC is desired, human antibody subclass IgG4 has been shown in numerous previous reports to have only modest ADCC and little CDC effector function (Moore G L et al., 2010 MAbs, 2: 181-189). However, native IgG4 is found to be less stable under stress conditions such as in acidic buffer or at increased temperature (Angal, S. 1993 Mol Immunol, 30: 105-108; Dall'Acqua, W. et al., 1998 Biochemistry, 37: 9266-9273; Aalberse et al., 2002 Immunol, 105: 9-19). Reduction of ADCC can be achieved by operably linking the antibody to an IgG4 Fc engineered with varying combinations that reduce FcyR binding or Clq binding activity, thereby reducing or eliminating ADCC and CDC effector function. One of the less desirable intrinsic properties of IgG4, given the physicochemical properties of antibodies as biopharmaceuticals, is the dynamic separation of its two heavy chains in solution to form a half-antibody, leading to the production of bispecific antibodies in vivo via a process known as "Fab arm exchange" (Van der Neut Kolfschoten M, et al., 2007 Science, 317: 1554-157). A mutation of serine to proline at position 228 (EU numbering system) appears to inhibit IgG4 heavy chain separation (Angal, S. 1993 Mol Immunol, 30: 105-108; Aalberse et al., 2002 Immunol, 105: 9-19). Some amino acid residues in the hinge and gamma Fc region are reported to have an impact on antibody interaction with Fcy receptors (Chappel SM et al., 1991 Proc. Natl. Acad. Sci. USA, 88: 9036-9040; Mukherjee, J. et al., 1995 FASEB J, 9: 115-119; Armour, K. L. et al., 1999 Eur J Immunol, 29: 2613-2624; Clynes, R. A. et al., 2000 Nature Medicine, 6: 443-446; Arnold J. N., 2007 Annu Rev immunol, 25: 21-50). Furthermore, some rare occurrence of IgG4 isoforms in the human population can also elicit different physicochemical properties (Brusco, A. et al., 1998 Eur J Immunogenet, 25: 349-55; Aalberse et al., 2002 Immunol, 105: 9-19).To generate a multispecific antibody with low ADCC and CDC but with good stability, the hinge and Fc region of human IgG4 can be modified and a number of changes introduced. These modified IgG4 Fc molecules can be found in SEQ ID NOs: 83-88, U.S. Patent No. 8,735,553 to Li et al.

[0215] Antibody production

[0216] Antibodies and antigen-binding fragments thereof can be produced by any means known in the art, including but not limited to recombinant expression, chemical synthesis, and enzymatic digestion of antibody tetramers, while full-length monoclonal antibodies can be obtained by, e.g., hybridoma or recombinant production. Recombinant expression can be from any suitable host cell known in the art, e.g., mammalian host cells, bacterial host cells, yeast host cells, insect host cells, etc.

[0217] The present disclosure further provides polynucleotides encoding the antibodies described herein, e.g., polynucleotides encoding a heavy or light chain variable region or a segment comprising a complementarity determining region as described herein. In some aspects, a polynucleotide encoding a heavy chain variable region has at least 85%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% nucleic acid sequence identity to a polynucleotide selected from the group consisting of SEQ ID NO: 12, SEQ ID NO: 22, SEQ ID NO: 32, and SEQ ID NO: 63. In some aspects, a polynucleotide encoding a light chain variable region has at least 85%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% nucleic acid sequence identity to a polynucleotide selected from the group consisting of SEQ ID NO: 13, SEQ ID NO: 23, SEQ ID NO: 33, and SEQ ID NO: 64.

[0218] The polynucleotides of the present disclosure can encode variable region sequences of anti-MUC1xCD16A antibodies. They can also encode both the variable and constant regions of the antibodies. Some sequences encode polypeptides comprising the variable regions of the heavy and light chains of the exemplified anti-MUC1xCD16A antibodies.

[0219] The present disclosure also provides expression vectors and host cells that produce anti-MUC1xCD16A antibodies. The choice of expression vector depends on the intended host cell into which the vector is to be expressed. The expression vectors can contain a promoter and other regulatory sequences (e.g., enhancers) operably linked to the polynucleotide encoding the anti-MUC1xCD16A antibody chains or antigen-binding fragments. In some aspects, an inducible promoter is used to prevent expression of the inserted sequences except under the control of the inducing conditions. Inducible promoters include, for example, arabinose, lacZ, metallothionein promoters, or heat shock promoters. Cultures of transformed organisms can be expanded under non-inducing conditions without biasing the population toward coding sequences whose expression products are better tolerated by the host cell. In addition to the promoter, other control elements can also be included to effect efficient expression of the anti-MUC1xCD16A antibody or antigen-binding fragment thereof. These elements can include an ATG initiation codon and adjacent ribosome binding site or other sequences. In addition, expression efficiency can be enhanced by inclusion of enhancers appropriate to the cellular system in use (see, e.g., Scharf et al., Results Probl. Cell Differ. 20:125, 1994; and Bittner et al., Meth. Enzymol., 153:516, 1987). For example, an SV40 enhancer or a CMV enhancer can be used to increase expression in mammalian host cells.

[0220] Host cells for housing and expressing anti-MUC1xCD16A antibody vectors can be prokaryotic or eukaryotic. E. coli is one prokaryotic host suitable for cloning and expression of the polynucleotides of the present disclosure. Other microbial hosts suitable for use include bacilli, such as B. subtilis, and other enterobacteriaceae, such as Salmonella, Serratia, and various Pseudomonas species. In these prokaryotic hosts, expression vectors can also be made that typically contain expression control sequences compatible with the host cell (e.g., an origin of replication). In addition, any number of various well-known promoters can be present, such as the lactose promoter system, a tryptophan (trp) promoter system, a beta-lactamase promoter system, or a promoter system from phage lambda. The promoters are typically optionally associated with operator sequences to control expression and have ribosome binding site sequences and the like for initiation and termination of transcription and translation. Other microorganisms, such as yeast, can also be used as hosts. Insect cells in combination with baculovirus vectors can also be used.

[0221] In other aspects, mammalian host cells are used to express and produce the anti-MUC1xCD16A antibodies of the present disclosure. Examples include hybridoma cell lines that express endogenous immunoglobulin genes or mammalian cell lines that harbor exogenous expression vectors. These include any normal mortal or normal or abnormal immortal animal or human cell. For example, a variety of suitable host cell lines capable of secreting intact immunoglobulins have been developed, including CHO cell lines, various COS cell lines, HEK293 cells, myeloma cell lines, transformed B cells, and hybridomas. The use of mammalian tissue cell cultures to express polypeptides is generally discussed in, for example, Winnacker, From Genes to Clones, VCH Publishers, NY, N.Y., 1987. Expression vectors for mammalian host cells can include expression control sequences, such as origins of replication, promoters, and enhancers (see, e.g., Queen et al., Immunol. Rev. 89:49-68, 1986), and necessary processing information sites, such as ribosome binding sites, RNA splice sites, polyadenylation sites, and transcriptional terminator sequences. These expression vectors will typically contain a promoter derived from a mammalian gene or a mammalian virus. Suitable promoters can be constitutive, cell type-specific, stage-specific, and / or inducible or repressible. Promoters that can be used include, but are not limited to, the metallothionein promoter, the constitutive adenovirus major late promoter, the dexamethasone-inducible MMTV promoter, the SV40 promoter, the MRP pol III promoter, the constitutive MPSV promoter, the tetracycline-inducible CMV promoter (such as the human immediate early CMV promoter), the constitutive CMV promoter, and promoter-enhancer combinations known in the art.

[0222] Generating bispecific antibodies

[0223] The presently disclosed bispecific antibodies can be generated using a knobs-into-holes (KiH) design that introduces mutations at the core CH3 domain interface. The resulting heterodimers have a reduced CH3 melting temperature (69°C or less). In contrast, the ZW heterodimer Fc design has a thermal stability of 81.5°C comparable to wild-type CH3 domains. The specific methods for generating the presently disclosed antibodies are described in the Examples.

[0224] Detection and diagnostic methods

[0225] The antibodies or antigen-binding fragments of the present disclosure are suitable for use in a variety of applications including, but not limited to, methods of detecting MUC1. In one aspect, the antibodies or antigen-binding fragments can be used to detect the presence of MUC1 in a biological sample. The term "detecting" as used herein includes quantitative or qualitative detection. In certain aspects, the biological sample comprises a cell or tissue. In other aspects, such tissues include normal and / or cancerous tissues that express MUC1 at higher levels relative to other tissues.

[0226] In one aspect, the present disclosure provides methods of detecting the presence of MUC1 in a biological sample. In certain aspects, the methods comprise contacting a biological sample with an anti-MUC1xCD16A antibody under conditions that allow the antibody to bind to the antigen and detecting whether a complex is formed between the antibody and the antigen. The biological sample can include, but is not limited to, urine, tissue, saliva, or blood.

[0227] Also included are methods of diagnosing a disorder associated with expression of MUC1. In certain aspects, the methods comprise contacting a test cell with an anti-MUC1xCD16A antibody; determining the expression level (quantitative or qualitative) of MUC1 expressed by the test cell by detecting binding of the anti-MUC1xCD16A antibody to the MUC1 polypeptide; and comparing the expression level by the test cell to the expression level of MUC1 in a control cell (e.g., a normal cell of the same tissue origin as the test cell or a non-MUC1 expressing cell), wherein a higher level of MUC1 expression in the test cell compared to the control cell is indicative of the presence of a disorder associated with expression of MUC1.

[0228] Pharmaceutical compositions and formulations

[0229] Also provided are compositions, including pharmaceutical formulations, comprising an anti-MUC1xCD16A antibody or antigen-binding fragment thereof, or a polynucleotide comprising a sequence encoding an anti-MUC1xCD16A antibody or antigen-binding fragment. These compositions can further comprise suitable carriers well known in the art, such as pharmaceutically acceptable excipients including buffers.

[0230] Pharmaceutical formulations of the anti-MUC1xCD16A antibodies or antigen-binding fragments as described herein are prepared by mixing such antibody or antigen-binding fragment having the desired degree of purity with one or more optional pharmaceutically-acceptable carriers (Remington's Pharmaceutical Sciences 16th Edition, Osol, A. Ed. (1980)), in either freeze-dried or aqueous solution form, and in dosages and concentrations that are non-toxic to the recipient. Pharmaceutically-acceptable carriers are generally nontoxic to the recipient at the dosages and concentrations employed, and include, but are not limited to, buffers such as phosphate, citrate, and other organic acids; antioxidants including ascorbic acid and methionine; preservatives (such as octadecyldimethylbenzyl ammonium chloride; hexamethonium chloride; benzalkonium chloride; benzethonium chloride; phenol, butyl or benzyl alcohol; alkyl parabens such as methyl or propyl paraben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight (less than about 10 residues) polypeptides; proteins, such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers such as polyvinylpyrrolidone; amino acids such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates including glucose, mannose, or dextrins; chelating agents such as EDTA; sugars such as sucrose, mannitol, trehalose or sorbitol; salt-forming counter-ions such as sodium; metal complexes (e.g. Zn-protein complexes); and / or non-ionic surfactants such as polyethylene glycol (PEG). Exemplary pharmaceutically-acceptable carriers herein further include an interstitial drug dispersion agent such as a soluble neutral-active hyaluronidase glycoprotein (sHASEGP), for example, a human soluble PH-20 Hyaluronidase Glycoprotein, such as rHuPH20 (HYLENEX®, Ferring Pharmaceuticals Inc.) Certain exemplary sHASEGPs including rHuPH20 and methods of use are described in US Patent No. 7,871,607 and 2006 / 0104968. In one aspect, a sHASEGP is combined with one or more additional glycosaminoglycanases such as a chondroitinase. Baxter International, Inc.). Certain exemplary sHASEGPs including rHuPH20 and methods of use are described in US Patent No. 7,871,607 and 2006 / 0104968. In one aspect, a sHASEGP is combined with one or more additional glycosaminoglycanases such as a chondroitinase.

[0231] In embodiments, the formulation includes L-histidine / L-histidine hydrochloride monohydrate, trehalose, and polysorbate 20. In embodiments, the anti-MUC1xCD16A antibody formulation is an isotonic solution including 10 mg / mL anti-MUC1xCD16A antibody, 20 mM histidine / histidine HC1, 240 mM trehalose dihydrate, and 0.02% polysorbate 20 at a pH of about 5.5 after formulation with sterile water for injection.

[0232] Exemplary lyophilized antibody formulations are described in U.S. Patent No. 6,267,958. Aqueous antibody formulations include those described in U.S. Patent No. 6,171,586 and WO 2006 / 044908, the latter formulations including a histidine-acetate buffer.

[0233] Sustained-release preparations can be prepared. Suitable examples of sustained-release preparations include semipermeable matrices of solid hydrophobic polymers containing the antibody, which matrices are in the form of shaped articles, e.g., films, or microcapsules.

[0234] Formulations to be used for in vivo administration are generally sterile. Sterility is readily accomplished, e.g., by filtration through a sterile filtration membrane.

[0235] Equivalents

[0236] It is to be understood that while the anti-human 4Ig-B7H3 antibodies and antigen-binding fragments thereof have been described in conjunction with their detailed description, that the foregoing description is intended to illustrate and not limit the scope of the application, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the claims.

[0237] It is to be understood that one, some, any or all of the features of the various embodiments disclosed herein can be combined to form further embodiments of the present disclosure. These and other aspects of the present disclosure are readily apparent to one skilled in the art.

[0238] Examples

[0239] Example 1. Generation of Anti-MUC1 Monoclonal Antibodies Targeting the MUC1 Membrane Proximal Region

[0240] MUC1 recombinant proteins for immunization and binding assays

[0241] The cDNA encoding full-length human MUC1 (SEQ ID NO: 1) was synthesized based on its Uniprot sequence (UniProtKB: P15941) by Genewiz (Suzhou, China) and purchased from them. The coding region of the SEA domain (sea urchin sperm protein, enterokinase and agrin) of full-length human MUC1 consisting of amino acid (AA) 1036-1155 of full-length MUC1 (SEQ ID NO: 2) was PCR-amplified and cloned into an expression vector based on pcDNA3.4 (Invitrogen, Carlsbad, CA, USA) with a C-terminal fusion to the Fc domain of mouse IgG2a or the Fc domain of human IgG1 heavy chain, resulting in two recombinant fusion protein expression plasmids, MUC1-SEA-mIgG2a and MUC1-SEA-huIgG1, respectively. A schematic depiction of the MUC1 fusion proteins is shown in Figure 1For recombinant fusion protein production, MUC1-SEA-mIgG2a and MUC1-SEA-huIgG1 plasmids were transiently transfected into Expi293 cells (Thermo Fisher, Waltham, MA, USA) and cultured for 6 days in a CO2 incubator equipped with a rotary shaker. The supernatant containing the recombinant protein was collected and clarified by centrifugation. MUC1-SEA-mIgG2a and MUC1-SEA-huIgG1 were purified using a protein A column (Cat. No. 17549852, Cytiva Life Sciences) followed by a HiLoad 16 / 600 Superdex 200pg size exclusion column (Cat. No. 28989335, Cytiva Life Sciences). MUC1-SEA-mIgG2a and MUC1-SEA-huIgG1 proteins were dialyzed against phosphate-buffered saline (PBS) and stored in a -80°C freezer in smaller aliquots.

[0242] Table 1. Sequence of MUC1

[0243]

[0244]

[0245] Generation of cell lines stably expressing human or cynomolgus macaque MUC1 for antibody production, screening, and validation

[0246] Cell lines stably expressing human MUCl were generated and validated, including the PT67 / human MUCl cell line (cell line generated in-house), the HEK293 / human MUCl cell line (HEK293 was obtained from ATCC, CRL-1573), and the HCT116 / human MUCl cell line (ATCCCL-247).

[0247] Cell lines stably expressing cynomolgus monkey MUC1 (SEQ ID NO: 3) were generated and validated, including HEK293 / cynomolgus monkey MUC1, L929 / cynomolgus monkey MUC1 cell lines (L929 was obtained from ATCC, CCL-1), HCT116 / cynomolgus monkey MUC1 cell lines, and Daudi / cynomolgus monkey MUC1 cell lines (Daudi was obtained from ATCC, CCL-213).

[0248] To generate stable cell lines expressing human or cynomolgus MUC1, episomal vectors were constructed using retroviral construct PFBneo (STRATAGENE, Cat# 217561-51). Retroviral construct transfection into PLAT-E cells (Cyagen, Cat# IPMPC-01001) was performed using Lipofectamine 2000 (Invitrogen, REF. #52758) according to the manufacturer's instructions. Virus supernatant was collected at 24, 48 and 72 hours post transfection and filtered (0.45 μιη) before use. The episomal virus generated above was used to transduce the bi-directional tropic packaging cell line PT67 in the presence of polybrene (final concentration: 8 μg / ml). After 3 rounds of transduction, PT67 cells were selected in G418 (final concentration: 1 mg / ml) for 7 days. To collect bi-directional virus produced from PT67 cells, when the cells became 100% confluent, the medium was changed to fresh DMEM complete medium without G418. Virus was collected once a day for 3 days. Cell lines were infected with virus containing human or cynomolgus MUC1. After 3 rounds of transduction, infected cells were selected in G418 (final concentration: 1 mg / ml) for 7 days.

[0249] Immunization

[0250] To generate antibodies against MUC1, cohorts of 30 inbred strain BALB / C, MRL strains of mice were immunized with different MUC1 antigens, and each cohort underwent an immunization strategy comprising a unique combination of MUC1 antigens (including proteins and cell lines from Example 1), dose, route of injection, adjuvant, and time of immunization. A total of 5 animals in 6 cohorts were immunized. Animals received immunizations over varying periods of time between 0 and 90 days. To monitor the immune response, serum was titrated after 2-6 immunizations between 30-90 days and screened by ELISA and FACS. Serum screening was selected for binding of antibodies to MUC1 antigens. MUC1 specific antibody responses were measured in each animal, and animals with sufficient titers of anti-MUC1 Ig were selected for a final boost 4 days.

[0251] Hybridoma fusion and screening

[0252] Lymphoid organs including spleen and lymph nodes were isolated from mice immunized as described above. Hybridomas were generated by fusion with immortalized mouse myeloma cells derived from SP2 / 0 via PEG-based fusion. The resulting cells were plated in 96-well cell culture plates using conventional 1640 culture medium supplemented with HAT for hybridoma selection. After 10-13 days of culture and replacement of growth medium, hybridoma culture supernatants were collected from individual wells and screened to identify wells with secreted MUC1-specific antibodies. All supernatants were initially screened against recombinant protein huMUC1-SEA-huIgG1 (from Example 1). Antibody binding on recombinant protein huMUC1-SEA-huIgG1 was measured by ELISA. Supernatants from culture wells of 3 hybridoma fusions were screened for MUC1 antibodies. Briefly, 2 μg / mL huMUC1-SEA-huIgG1 was coated in a 96-well ELISA plate and 50 μl of hybridoma culture supernatant was incubated for 30-60 min, washed, and incubated with an anti-mouse IgG Fc secondary antibody conjugated to HRP. After incubation and washing, the plate was developed with HRP substrate and the absorbance was measured.

[0253] Hybridomas from positive wells were transferred to 24-well plates with fresh medium to grow for 2-3 days and then screened again by flow cytometry to confirm antibody binding to human MUCl and cynomolgus monkey MUCl overexpressing cell lines.

[0254] The binding of antibodies (Abs) to human MUC1 and cynomolgus monkey MUC1 overexpressing cell lines was measured by FACS. Briefly, 100 μl of hybridoma culture supernatant was incubated with human MUC1 overexpressing cells or cynomolgus monkey MUC1 overexpressing cells for 30-60 min, washed, and incubated with an anti-mouse IgG Fc secondary antibody conjugated to APC. After incubation and washing, fluorescence was measured by flow cytometry.

[0255] Subcloning and sequence analysis

[0256] Selected anti-MUC1 antibody-secreting hybridomas were subcloned once or twice to ensure monoclonality. Briefly, positive hybridoma clones were subcloned by limiting dilution. After 7-10 days, culture supernatants were screened by ELISA and flow cytometry as described above to confirm binding to human and cynomolgus monkey MUC1 antibodies. Stable hybridoma subclones were cultured in vitro for cell cryopreservation and antibody VH and VL gene cloning and sequencing.

[0257] Following subcloning, anti-MUC1 antibody secreting hybridomas were lysed by lysis buffer. The lysate containing mRNA was then transferred to a 96-deep well plate for mRNA isolation, cDNA synthesis and DNA sequencing by standard sequencing technique (Sanger sequencing). Typically, total RNA of cell lysate was prepared and cDNA was generated by reverse transcription of mRNA using Super Script III First-Strand Synthesis SuperMix (Invitrogen) according to the manufacturer’s instructions. Sequences of BG138P antibody are listed in Table 2.

[0258] Single B screening

[0259] Immunized mice were sacrificed and spleen was harvested. Enriched plasma cells were loaded onto 14K chip. hMUC1 beads, cynoMUC1 beads and HEK293-cynoMUC1 cells were used for screening on chip. Hits were selected and exported into lysis buffer. Single cell RNA was purified and Ig sequences were recovered using BLI cDNA synthesis kit (BERKELEY LIGHTS) according to the manufacturer’s instructions. Sequences of BG219P and BG346P are listed in Table 2.

[0260] Table 2. Sequences of murine antibodies against human MUC1

[0261]

[0262]

[0263]

[0264]

[0265] Large scale expression and purification of chimeric BG219P, BG138P and BG346P

[0266] Chimeric antibodies chBG219P, chBG138P and chBG346P were produced by transient transfection of in-house produced heavy and light chain containing plasmids in ExpiCHO-s cells. Conditioned media were harvested and antibodies were purified using MabSelect SuRe columns (Cytiva) followed by POROS TM 50HS columns (Thermofisher Scientific) and G-25 desalting columns (Cytiva). All purified antibodies were stored in -80°C freezer in small aliquots.

[0267] Example 2. Binding kinetics and affinity determination of anti-MUC1 antibodies by SPR

[0268] Chimeric anti-MUC1 antibodies were assayed for binding kinetics by SPR using Biacore TM T-200 (GE Life Sciences). Briefly, mouse anti-human IgG Fc antibody was immobilized on an activated CM5 biosensor chip (catalog no. BR100530, GE Life Sciences). Purified chimeric anti-MUC1 antibodies were flowed over the chip surface and captured by the anti-mouse IgG antibody. Serial dilutions of human or cynomolgus MUC1-SEA protein were then flowed over the chip surface and changes in surface plasmon resonance signals were analyzed using a one-to-one Langmuir binding model (BIA Evaluation Software, GE Life Sciences) to calculate the association rate (k on ) and dissociation rate (k off ). The equilibrium dissociation constant (K D ) was calculated as the ratio k off / k on . The binding affinity profiles of chimeric anti-MUC1 antibodies chBG138P, chBG346P, and chBG219P are shown in Table 3 below. ChBG138P, chBG346P, and chBG219P exhibited high affinity for human and cynomolgus MUC1-SEA.

[0269] Table 3. Antigen binding affinity of chBG138P, chBG346P, and chBG219P

[0270]

[0271] Example 3. Determining binding affinity of anti-MUC1 antibodies for MUC1 expressed in stable expression cell lines

[0272] The binding affinity of chimeric anti-MUC1 antibodies for human and cynomolgus MUC1 overexpressing cell lines (HEK293 / human MUC1 and HEK293 / cynomolgus MUC1) was determined by FACS. Briefly, human or cynomolgus MUC1 overexpressing cells were incubated with serial dilutions of purified antibodies, washed, and incubated with anti-human IgG secondary antibody conjugated to APC. After incubation and washing, fluorescence was measured by flow cytometry. The binding affinity profiles of chimeric anti-MUC1 antibodies are shown in Table 4 and Figures 2A-2F below (with hIgGl as a negative control). The results demonstrate that all three chimeric anti-MUC1 antibodies have favorable binding affinity for human and cynomolgus MUC1 expressed in stable expression cell lines.

[0273] Table 4. Cell binding affinity of chBG138P, chBG346P, and chBG219P

[0274]

[0275] Example 4. Epitope binning of anti-MUC1 antibodies

[0276] Epitope binning of chimeric anti-MUC1 antibodies was determined by a competitive SPR assay. Briefly, anti-mouse IgG Fc antibody was immobilized on an activated CM5 biosensor chip. Purified huMUC1-SEA-mIgG2a (human MUC1 linked to mouse IgG2a Fc) antigen was flowed over the chip surface and captured by the anti-mouse IgG antibody. Reference MUC1-SEA antibody 5F3 (Cancer Immunol Immunother. 2020 Jul;69(7): 1337-1352) was first injected under saturating antigen binding conditions, followed by injection of chBG138P (SEQ ID NO: 1), chBG219P (SEQ ID NO: 2) or chBG346P (SEQ ID NO: 3). Sensograms of epitope binning are shown in Figure 3A Figure 3B Figure 3C Figures 3A-3C As shown in Table 5 below, the three chimeric MUC1 antibodies were grouped into two epitope bins in the MUC1-SEA domain. ChBG138P and chBG219P bound to the same epitope on MUC1-SEA as 5F3, which is a different MUC1 epitope from chBG346P. More specifically, chBG138P and chBG219P bound to epitope bin A, while chBG346P bound to bin B.

[0277] Table 5. Epitope bins of chBG138P, chBG346P and chBG219P

[0278] Antibody Number Box chBG138P A chBG346P B chBG219P A

[0279] Example 5. Chimeric anti-MUC1 antibodies chBG138P, chBG219P and chBG346P show reduced interference by soluble MUC1

[0280] The presence of soluble MUC1 in specific binding of MUC1 antibodies to MUC1 expressing cells was determined by a competitive FACS assay. Briefly, human MUC1 expressing cells were incubated with 30, 3 and 0.3 pg / ml MUC1 antibodies in the presence of serially diluted soluble MUC1 (Shanghai Linc-Bio Science Co. LTD). After washing and incubation with anti-human IgG secondary antibody, fluorescence was measured by flow cytometry. IC 50 values of soluble MUC1 blocking the binding of MUC1 antibodies to MUC1 expressing cells are shown in Table 6, and blocking curves are shown in Figures 4A-4F ​​​, where HMFG1 is a positive control and mIgG and hIgG1 are negative controls. The profiles indicate that HMFG binding to MUC1-expressing cells can be easily perturbed at high, medium, and low antibody concentrations (i.e., 30, 3, and 0.3 μg / ml), but chBG138P, chBG219P, and chBG346P binding is only slightly perturbed at low antibody concentrations (i.e., 0.3 μg / ml). Figure 4A -F). Overall, the profiles in Figure 4 indicate that binding of the MUCl antibodies to MUCl expressing cells showed significantly reduced interference by soluble MUCl compared to HMFG1 (Abeam), which targets the N-terminus of MUCl.

[0281] Table 6. IC of soluble MUC1 blocking activity 50

[0282]

[0283] (NA: Not available. Figure 4A 、 4B As shown in the binding curves in 4E and 4F, IC 50 The data cannot be well fitted and extracted due to extremely low noise.)

[0284] Example 6. Anti-MUCl monoclonal antibodies targeting the membrane-proximal region of MUCl bind to cancer cell lines but not to normal T cells, whereas antibodies targeting the MUCl N-terminus, HMFG1 or 16A, bind to normal T cells.

[0285] To evaluate whether anti-MUC1 monoclonal antibodies targeting the membrane-proximal region of MUC1 can differentially bind to MUC1-expressing tumor cells compared to MUC1-expressing normal cells, such as activated T cells, FACS binding assays were performed. For tumor cell line binding experiments, cells were harvested and stained for 1 hour with the anti-human MUC1 antibody chBG138P or a control (human IgG1). The cells were then washed twice and subsequently incubated with a secondary antibody (Alexa Fluor 500). 647 anti-human IgG Fc) for 30 minutes. The cells were washed and fixed with 1% paraformaldehyde (PFA) in DPBS before FACS analysis. All flow cytometry data were obtained using a NovoCyte flow cytometer (ACEA Biosciences, Inc.) and the data were analyzed using NovoExpress software. Figures 5A-5C The results showed that the chimeric antibody chBG138P targeting the membrane proximal region of human MUC1 bound to the MUC1-expressing tumor cell line HCC827 ( Figure 5A )、H1975( Figure 5B ) and T-47D(Figure 5C ) (human IgGl as negative control) indicates that MUC1 membrane-proximal region-targeting antibodies can be used to target cancer cells and are thus suitable for treating MUC1 -expressing cancers.

[0286] To assess whether MUC1 membrane-proximal region-targeting monoclonal antibodies can avoid binding to normal cells expressing MUC1, an activated T cell binding assay was performed. Briefly, human peripheral blood mononuclear cells (PBMCs) from six healthy donors purchased from Allcells or Stemcell were stimulated with 1 pg / ml PHA-L for 3 days. FACS staining was then performed using the stimulated PBMCs. Cell suspensions were pre-incubated with LIVE / DEAD® Fixable Dead Cell Stain Kit (Invitrogen, REF. #L34964) and Fc receptor blocking solution (100 pg / mL human IgG in FACS buffer) and then stained with anti-human antibodies. Cells were washed twice and incubated with 10 pg / mL of anti-MUC1 monoclonal antibodies targeting the MUC1 membrane-proximal region or MUC1 N-terminal targeting antibodies HMFG1 (as positive control, Abeam, Cat# ab215670) or 16A (as positive control, Biolegend, Cat# 355608) for 1 hour. Cells were then washed and stained with PE-CY7 anti-human aP TCR (eBioscience, Cat# 25-9986-42), AF647 anti-human IgG Fc (Biolegend, REF. #409320) for 30 minutes. Cells were washed and fixed with 1% PFA in DPBS before FACS analysis. All flow cytometry data were acquired using a NovoCyte flow cytometer (ACEA Biosciences, Inc.) and data were analyzed using NovoExpress software. As shown in Figure 6, none of the chimeric antibodies chBG138P (7A), chBG219P (7B), chBG346P (7C), chBG138H (7D), chBG219H (7E), and chBG346H (7F) bound to activated T cells expressing MUC1. However, MUC1 -N-terminal targeting antibodies HMFG1 (7G) and 16A (7H) bound to a significant fraction of activated T cells. TM Fixable Dead Cell Stain Kit (Invitrogen, REF. #L34964) and Fc receptor blocking solution (100 pg / mL human IgG in FACS buffer) before staining with anti-human antibodies. Cells were washed twice and incubated with 10 pg / mL of anti-MUC1 monoclonal antibodies targeting the MUC1 membrane-proximal region or MUC1 N-terminal targeting antibodies HMFG1 (as positive control, Abeam, Cat# ab215670) or 16A (as positive control, Biolegend, Cat# 355608) for 1 hour. Cells were then washed and stained with PE-CY7 anti-human aP TCR (eBioscience, Cat# 25-9986-42), AF647 anti-human IgG Fc (Biolegend, REF. #409320) for 30 minutes. Cells were washed and fixed with 1% PFA in DPBS before FACS analysis. All flow cytometry data were acquired using a NovoCyte flow cytometer (ACEA Biosciences, Inc.) and data were analyzed using NovoExpress software. As shown in Figure 6, none of the chimeric antibodies chBG138P (7A), chBG219P (7B), chBG346P (7C), chBG138H (7D), chBG219H (7E), and chBG346H (7F) bound to activated T cells expressing MUC1. However, MUC1 -N-terminal targeting antibodies HMFG1 (7G) and 16A (7H) bound to a significant fraction of activated T cells. Figure 6 and 7A -7H show that none of the chimeric antibodies chBG138P (7A), chBG219P (7B), chBG346P (7C), chBG138H (7D), chBG219H (7E), and chBG346H (7F) bound to activated T cells expressing MUC1. However, MUC1 -N-terminal targeting antibodies HMFG1 (7G) and 16A (7H) bound to a significant fraction of activated T cells. Figure 7A , 7B , 7E, and 7F), chBG219P (7B), chBG346P (7C), chBG138H (7D), chBG219H (7E), and chBG346H (7F) did not bind to activated T cells expressing MUC1. However, MUC1 -N-terminal targeting antibodies HMFG1 (7G) and 16A (7H) bound to a significant fraction of activated T cells. Figure 7E and 7F ) or chBG346P (7C, 7D, 7E, and 7F) did not bind to activated T cells expressing MUC1. However, MUC1 -N-terminal targeting antibodies HMFG1 (7G) and 16A (7H) bound to a significant fraction of activated T cells. Figure 7E and 7F ) or chBG346P (7C, 7D, 7E, and 7F) did not bind to activated T cells expressing MUC1. However, MUC1 -N-terminal targeting antibodies HMFG1 (7G) and 16A (7H) bound to a significant fraction of activated T cells. Figure 7C and 7D ) or chBG346P (7C, 7D, 7E, and 7F) did not bind to activated T cells expressing MUC1. However, MUC1 -N-terminal targeting antibodies HMFG1 (7G) and 16A (7H) bound to a significant fraction of activated T cells. Figure 7G and 7H ) or chBG346P (7C, 7D, 7E, and 7F) did not bind to activated T cells expressing MUC1. However, MUC1 -N-terminal targeting antibodies HMFG1 (7G) and 16A (7H) bound to a significant fraction of activated T cells. Figures 8A-8DThe antibody BG138P in chimeric (chBG138P) and humanized (huBG138P-Hz2 and huBG138P-Hz4) formats showed similar binding properties to BG138P, with no binding to normal activated human T cells expressing MUC1 Figure 8A and 8B ), whereas the MUC1-N-terminal targeting antibody HMFG1 Figure 8C and 8D bound to normal activated T cells.

[0287] The results indicate that antibodies targeting the MUC1 membrane proximal region specifically target cancer cells while sparing normal T cells compared to MUC1-N-terminal targeting antibodies that bind to both cancer cells and normal T cells, and thus can confer an optimized safety profile when used as an anti-tumor therapy in humans.

[0288] Example 7. Humanization of murine anti-human MUC1 antibody BG219P

[0289] For humanization of BG219P, human germline IgG genes were searched for sequences sharing high homology to the protein sequences of the variable regions of BG219P by sequence comparison against the human immunoglobulin gene database in IMGT. Human IGHV and IGKV genes that are present at high frequency in the human antibody repertoire and that share high homology to murine BG219P were selected as templates for humanization.

[0290] Humanization was performed by CDR grafting followed by incorporation of key back mutations. The humanized antibodies were engineered into human IgGl wild type format by using in-house developed expression vectors. In the initial round of humanization, mutations of murine to human amino acid residues in the framework regions were guided by 3D structure analysis and murine framework residues that are structurally important for maintaining the canonical structure of the CDRs were preserved in the first round of humanization design. Among all 19 variants generated, BG219P-Bz0 is the CDR grafted version of the antibody variant with all back mutation sites that is the variant with binding capacity close to the theoretical binding capacity of the parental murine antibody BG219P.

[0291] Specifically, BG219P-Bz0 was generated as described herein. Human germline variable genes IGKV1-39*01 and IGKJ2*01 and human germline variable genes IGHV3-23*01 and IGHJ6*01 were selected as the recipient frameworks for the BG219P VL and VH sequences. The LCDRs of murine BG219P were grafted into the frameworks of human germline variable genes IGKV1-39*01 and IGKJ2*01 with D17E, A43S, I48V, T69P, and F71Y murine framework residues. The amino acid sequence and DNA sequence of the resulting BG219P-Bz0 VL are shown in Table 8. The HCDRs of murine BG219P were grafted into the frameworks of human germline variable genes IGHV3-23*01 and IGHJ6*01 with S30N, S49A, A93T, and K94R murine framework residues retained. The amino acid sequence and DNA sequence of the resulting BG219P-Bz0 VH are shown in Table 8.

[0292] Beginning with the humanized BG219P antibody huBG219P-Bz0, several additional amino acid changes in the CDR regions of the VH and VL were made to further improve biophysical properties for therapeutic use in humans. Considerations included removal of post-translational modifications and improvement of thermal stability (Tm) while maintaining binding activity.

[0293] Over thirty humanized BG219P (also referred to as HuBG219P) variants were constructed using an internal IgG1 / Ck eukaryotic expression vector with readily adaptable subcloning sites containing the constant region of human wild-type IgG1 and kappa chains, respectively. Variants were produced by transient transfection of plasmids in ExpiCHO-s cells (Thermofisher Scientific). Conditioned media were harvested and variants purified using MabSelect TM SuRe columns (Cytiva) followed by UF / DF to exchange buffer. All purified antibodies were stored in -80 °C freezer in small aliquots.

[0294] For affinity determination, antibodies were captured by anti-human Fc surface and used in an affinity assay based on surface plasmon resonance (SPR) technology. Results of SPR determination of binding profiles for anti-MUC1 antibodies are summarized in Table 7. huBG219P-E39 and huBG219P-E43 had similar binding affinities with dissociation constants of 35.2 pM and 30.3 pM, respectively, similar to the dissociation constant of chimeric BG219P (39.8 pM). Sequences of huBG219P-E39 and huBG219P-E43 are provided in Table 8.

[0295] Table 7. Comparison of binding affinity of huBG219P to MUC1 SEA-Fc by SPR

[0296] Antibody k on (M-1s-1)] k off (s-1)]]> K D (nM) chBG219P 1.44E+06 5.74E-05 3.98E-011 huBG219P-E39 1.40E+06 4.92E-05 3.52E-011 huBG219P-E43 1.61E+06 4.86E-05 3.03E-011

[0297] Table 8. Sequence Listing

[0298]

[0299]

[0300]

[0301]

[0302]

[0303] To assess the binding activity of anti-MUC1 antibodies to native MUC1 on live cells, ZR-75 cells were used for FACS-based binding assays. Live ZR-75 cells were seeded in 96-well plates and incubated with serial dilutions of chimeric or humanized BG219Ps. Goat anti-human IgG was used as a secondary antibody to detect the binding of antibodies to the cell surface. EC 50 Values were determined by fitting the dose response data to a four-parameter logistic model in GraphPad Prism. As Figure 9 Table 9 and Figure 9 show that humanized BG219P antibodies huBG219P-Bz0, E39 and E43 maintained comparable binding affinity to native MUC1 compared to chimeric BG219P.

[0304] Table 9. Comparison of binding affinity of chBG219P and huBG219P to native MUC1 by FACS

[0305] Antibody ZR-75 binding EC 50 (nM)]]> chBG219P 12.3 huBG219P-bz0 20.9 huBG219P-E39 14.0 huBG219P-E43 15.7

[0306] Example 8. Generation of anti-CD16A VHH

[0307] Human CD16A recombinant proteins and cell lines for immunization and assays

[0308] A recombinant hexahistidine-tagged extracellular domain (ECD) fragment of human CD16A protein (V158) (SEQ ID NO: 101), referred to as human CD16A-His6 (V158), was purchased from a commercial source (Sino Biologies) and used as the antigen for immunizing llamas. Recombinant hexahistidine-tagged ECD fragments of human CD16A (F158) (SEQ ID NO: 102), human CD16B (NA1) (SEQ ID NO: 103), human CD16B (NA2) (SEQ ID NO: 104), human CD16B (SH) (SEQ ID NO: 105), cynomolgus CD16 (SEQ ID NO: 106), referred to as human CD16A-His6 (F158), human CD16B-His6 (NA1), human CD16B-His6 (NA2), human CD16B-His6 (SH), and cynomolgus CD16-His6, respectively, were purchased from a commercial source (Sino Biologies) and used in various in vitro assays.

[0309] To facilitate screening and testing, a DNA fragment of human CD16A (V158) ECD (AA 1-208 of SEQ ID NO: 101) was fused to a C-terminal human IgGl mf Fc tag (SEQ ID NO: 107), mouse IgG2a Fc tag, or llama IgG2b Fc tag and subjected to transient expression in Expi293 cells (Thermofisher Scientific). Culture supernatants were harvested and clarified, and subjected to affinity purification with a protein A column (Cytiva). The final product was buffer exchanged to DPBS by ultrafiltration / diafiltration (UF / DF) and stored at -80 °C.

[0310] To assess the binding activity of antibodies to CD16A expressed on live cells, NK92mi (ATCC, CRL-2407) cells were engineered to overexpress human CD16A by co-transducing expression plasmids containing CD16A (F158 or V158) and FcRy cDNAs (NK92mi / CD16A F158 and NK92mi / CD16A V158). NK92mi / CD16B (NA1) and NK92mi / CD16B (NA2) expression cell lines were prepared in a similar manner from CD16B (NA1) or CD16B (NA2) expression plasmids.

[0311] Immunization and screening

[0312] An alpaca was immunized with recombinant protein human CD16A-His6 (V158) as antigen by an external contract research organization and after the third immunization, an immune VHH phage library was constructed from isolated alpaca peripheral blood mononuclear cells (PBMCs) (Pardon Els et al. (2014) Nature Protocols). Phage display selection was performed using standard protocols (Silacci et al., (2005) Proteomics, 5, 2340-50; Zhao et al., (2014) PLoS One, 9, e111339). Briefly, in the 1st and 2nd round of panning, 10 mg / ml of immobilized human CD16A-V158-Alpaca IgG2b in immunotubes was used to enrich human CD16A (V158)-specific binders. Immunotubes were blocked for 1 h with 5% milk (w / v) in PBS supplemented with 1% Tween 20 (MPBST). After washing with PBST (PBS buffer supplemented with 0.05% Tween 20), 1 x 1011 13 (1st round) or 2 x 1011 12 (2nd round) phage were initially depleted for 1 h in MPBST against human CD16B-His6 (NA2) and then incubated for 1 h with the antigen. After washing with PBST, bound phage were eluted with 100 mM triethylamine (Sigma-Aldrich). Eluted phage were used to infect exponentially growing E. coli TG1 bacteria and plated on 2 x YT (yeast extract tryptone)-agar plates supplemented with 2% glucose and 100 pg / mL ampicillin. After three rounds of selection, individual clones were picked and phage-containing supernatants were prepared using standard protocols. Phage ELISA was used to screen for anti-human CD16A antibodies.

[0313] For phage ELISA, Maxisorp immunoplates were coated with recombinant protein human CD16A-His6 (V158) as antigen and blocked with 5% milk (w / v) in PBS buffer. Phage supernatants were blocked for 30 min with MPBST and added to the wells of the ELISA plate for 1 h. After washing with PBST, bound phage were detected using HRP-conjugated anti-M13 antibody (GE Healthcare) and 3,3',5,5'-tetramethylbenzidine substrate (cat. no. 00-4201-56, eBioscience, USA).

[0314] Positive clones from phage ELISA were sequenced and recovered. Six anti-CD16 AVHH variants were constructed by fusing their open reading frame with a C-terminal human IgGl mf Fc (SEQ ID NO: 107) tag eukaryotic expression vector. Using the MAX titer protocol, plasmids were transfected into ExpiCHO-s cells (Thermofisher Scientific). Fc-tagged VHH variants (VHH-Fc) were purified by MabSelect SuRe (Cytiva), followed by SPHP column (Cytiva). Final product buffer was exchanged to DPBS by UF / DF and stored at -80°C for later use, including binding assays.

[0315] For antigen ELISA, Maxisorp immunoplates were coated with antigens (human CD16A (V158), human CD16A (F158), human CD16B (NA1), human CD16B (NA2), human CD16B (SH), or cynomolgus CD16) and blocked with 3% BSA (w / v) in PBS buffer (blocking buffer). Monoclonal VHH-Fc antibodies were blocked with blocking buffer for 30 min and added to the wells of the ELISA plate for 1 h. After washing with PBST, bound antibodies were detected using HRP-conjugated anti-human IgG antibody (Sigma, A0170) and 3,3’,5,5’-tetramethylbenzidine substrate (Cat. No: 00-4201-56, eBioscience, USA).

[0316] For flow cytometry, NK92mi / CD16A V158 cells, NK92mi / CD16B (NA1) cells, and NK92mi / CD16B (NA2) cells (10 5 cells / well) were incubated with different concentrations of IgG-like antibodies, followed by binding with Alexa Fluro-647 labeled anti-human IgG Fc antibody (Cat. No: 409320, BioLegend, USA). Cell fluorescence was quantified using flow cytometer (Guava easyCyte 8HT, Merck-Millipore, USA). TM

[0317] ​Following the above disclosed procedures, 44 positive clones were sequenced and recovered, and one representative positive anti-CD16A variant, BG523P (VHH AA SEQ ID NO: 112, VHH DNA SEQ ID NO: 113), was obtained from six VHH-Fc fusion clones. The binding affinity of BG523P to CD16A was also confirmed by antigen ELISA. The results of ELISA and FACS analysis of BG523P compared to the positive control LS21 are shown in Tables 10 to 12 and Figure 10 and 11A More specifically, the FACS binding data of BG523P to the NK92mi / CD16A cell line compared to the human CD16A specific binding agent LS21 (SEQ ID NO: 108, patent EP1888645B1) as a positive control are given in Table 10, which shows that BG523P has specific binding to NK92mi / CD16A cells. Figure 10 It is shown that BG523P exhibited higher binding to human CD16A, CD16B (NA1) and cynomolgus monkey CD16 at 1 mg / ml compared to LS21. Figure 11A It was shown that BG523P specifically binds to NK92mi / CD16A cells. Figure 11B and 11C It was shown that at high concentrations, BG523P exhibited weak binding to NK92mi / CD16B. Although BG523P exhibited slight binding activity to CD16B (NA1) in ELISA and FACS assays ( Figure 10 , Table 11), whose binding affinity was significantly reduced (calculated EC compared with CD16A binding in FACS assay). 50 This result suggests that BG523P can selectively bind to CD16A rather than CD16B. Figure 10 BG523P was also shown to bind little to the CD16B SH allotype. Considering that the major variants of human CD16B are the NA1 and NA2 allotypes and the frequency of the SH allotype is rare and reported to be less than 0.05 in Caucasians, the binding properties to the CD16B SH allotype were not further characterized.

[0318] Table 10. FACS-based binding of VHHs to NK92mi / CD16A cell line

[0319] BG523P LS21 Human IgG EMAX 342.0 504.8 7.55 EC 50 (μg / mL) 0.242 1.041 0.01122

[0320] Table 11. FACS-based binding of VHHs to the NK92mi / CD16B (NA1) cell line

[0321] BG523P LS21 Human IgG EMAX 149.5 ～-2874 7.456 EC 50 (μg / mL) 9.518 ～155157 10.30

[0322] Table 12. FACS-based binding of VHHs to NK92mi / CD16B (NA2) cell line

[0323] BG523P LS21 Human IgG EMAX 14.82 7.823 7.655 EC 50 (μg / mL) 28.90 0.1363 0.3139

[0324] Example 9. Humanization of anti-human CD16A VHH BG523P

[0325] For humanization of BG523P, the human immunoglobulin gene database in the IMGT (http: / / www.imgt.org / IMGT_vquest / share / textes / index.html) and NCBI (http: / / www.ncbi.nlm.nih.gov / igblast / ) websites were blasted for sequences sharing high homology with the cDNA sequence of the variable region of BG523P. Human IGVH genes that are present at high frequency in the human antibody repertoire (Glanville 2009 PNAS 106:20216-20221) and share high homology with BG523P were selected as templates for humanization.

[0326] Humanization was performed by CDR grafting (Methods in Molecular Biology, Vol 248: Antibody Engineering, Methods and Protocols, Humana Press) and the humanized VHH variants were engineered as VHH-Fc using in-house developed expression vectors for later binding and biophysical stability analysis, etc. In the initial round of humanization, mutations in the framework regions from Camelid to human amino acid residues were guided by modeling the 3D structure and Camelid framework residues that are structurally important for maintaining the canonical structure of the CDRs were preserved in the first version of humanized BG523P. Among many variants, BG524P is the preferred humanized VHH with most Camelid residues preserved. Specifically, HCDR1 (SEQ ID NO: 109) and HCDR3 (SEQ ID NO: 111) of BG523P were grafted into the framework of human germline variable gene IGVH3-7 with 5 Camelid framework residues (F37, R45, V78, P84 and A94 by Kabat numbering) preserved, while one mutation was introduced in HCDR2 to remove a potential isomerization site. The sequences of BG524P are provided as SEQ ID NOs: 109, 114, 111 and 115-116 in Table 23.

[0327] Using an in-house developed expression vector with an Fc region containing human IgGl variant (SEQ ID NO: 107) with easily adaptable subcloning sites, the humanized BG523P variant was fused to the N-terminus of Fc as a VHH-Fc format. Expression and production of the humanized BG523P VHH-Fc antibody was achieved by transfecting the construct into ExpiCHO-s cells and purified by using a protein A column. Purified VHH-Fc antibodies were concentrated to 0.5-5 mg / mL in PBS and stored in aliquots at -80 °C freezer.

[0328] For affinity determination, VHH-Fc antibodies were captured by anti-human Fc surface and used in affinity determination based on surface plasmon resonance (SPR) technology. The results of SPR determination of binding profile of anti-CD16A VHHs are summarized in Table 13. BG524P showed slightly improved binding affinity to CD16A 158V and CD16A 158F with dissociation constants of 0.08 nM and 0.08 nM, respectively, compared to BG523P. At the same time, BG524P maintained the selectivity for CD16A over CD16B as characterized by SPR.

[0329] Table 13. Comparison of binding affinity of anti-CD16A VHHs to human CD16A by SPR

[0330]

[0331] NK92mi / CD16A F158 cell line was used to evaluate the ability of anti-CD16A VHH-Fc antibodies to bind to native CD16A on live cells.

[0332] Live NK92mi / CD16A 158F cells were seeded in 96-well plates and incubated with serial dilutions of anti-CD16A VHH-Fc. Goat anti-human IgG was used as a secondary antibody to detect the binding of the antibody to the cell surface. EC 50 Values were determined by fitting the dose response data to a four-parameter logistic model of GraphPad Prism. As Figure 12A As shown in Table 14 and Figure 6, BG524P showed improved binding affinity to native CD16A 158F but showed reduced Emax.

[0333] Table 14. Comparison of binding affinity of anti-CD16A VHHs to human CD16A by FACS

[0334] VHH BG523P BG524P NK92mi / CD16A 158F binds EC 50 (nM) 15.06 4.42 NK92mi / CD16A 158F binding Emax (MFI) 128500 96964

[0335] To determine if humanized BG523P retains the optimal biophysical stability of BG523P, the melting temperature (Tm) and aggregation temperature (Tagg) of BG524P were determined and compared to those of BG523P. BG524P showed inferior Tm and Tagg compared to those of BG523P (Table 15).

[0336] Melting temperature (Tm) was determined using high-throughput MicroCal TM VP-Capillary DSC (Malvern Instruments, Northampton, MA). Using a scan rate of 60 °C / hr, temperature profiles of each protein (at 0.5 mg / mL, 350 μL) were obtained from 20 °C to 100 °C. The temperature profile of the buffer alone was subtracted from each protein sample. The results obtained show the values of the midpoint of the transition temperature (Tm) and the enthalpy of the calorimetric (ΔH) of the samples.

[0337] Aggregation temperature Tagg (°C) represents the colloidal stability of the sample and was obtained by monitoring the onset of aggregation by SLS266 using UNCLE TM (Unechained lab, Pleasanton, CA). The samples were loaded into the UNCLE and subjected to a temperature ramp from 15 °C to 95 °C. The back-reflection optics cannot detect the scattering of near-UV light by protein aggregates, and thus only the non-scattered light reaches the detector. Therefore, the decrease in back-reflection light is a direct measure of aggregation in the sample.

[0338] Table 15. Comparison of thermal stability and colloidal stability of anti-CD16A VHHs

[0339] VHH-Fc Tm (°C) Tagg (°C) BG523P 69.1 68.0 BG524P 59.7 54.5

[0340] BG524P was further engineered by introducing mutations in the CDRs and back mutations in the framework regions to improve biophysical properties, remove PTM sites, and restore the binding Emax of native CD16A for therapeutic use in humans.

[0341] In summary, the better engineered versions of the humanized monoclonal antibodies, BG525P (SEQ ID NOs: 109-111 and 117-118) and BG526P (SEQ ID NOs: 109, 114, 111, and 119-120) were derived from the mutagenesis process as described above, and both retain the binding affinity to CD16A, the selectivity over CD16B, and the optimal biophysical stability of the parental clone, as characterized in detail (Tables 16-18 and Figure 12B ).

[0342] Table 16. Comparison of the binding affinity of anti-CD16A VHHs to human CD16A by SPR

[0343]

[0344] Table 17. Comparison of the binding affinity of anti-CD16A VHHs to human CD16A by FACS

[0345] VHH BG523P BG525P BG526P NK92mi / CD16A 158F binds EC 50 (nM) 10.6 10.3 9.1 NK92mi / CD16A 158F binding Emax (MFI) 9915 9708 9197

[0346] Table 18. Comparison of the thermal and colloidal stability of anti-CD16A VHHs

[0347] VHH-Fc BG523P BG525P BG526P Tm (°C) 69.1 69.1 68.1 Tagg (°C) 68 68.1 65.2

[0348] Example 10. Anti-CD16A VHH binding to native CD16B

[0349] To assess the ability of anti-CD16A VHHs to bind to native CD16B on live cells, NK92mi cells were engineered to overexpress human CD16B NA1 or NA2. Live NK92mi / CD16B cells were seeded in 96-well plates and incubated with 300 nM of anti-CD16A VHH-Fc. Goat anti-human IgG was used as a secondary antibody to detect the binding of anti-CD16A VHH-Fc to the cell surface. The binding signal of humanized VHH-Fc to CD16B was similar to or lower than those of the parental clones, as shown in Table 19 and significantly lower than the corresponding binding signal to CD16A (Table 17, Figure 13 ). Figures 12A-12B

[0350] Table 19. Comparison of anti-CD16A VHHs to parental clones for binding to human CD16B by FACS

[0351]

[0352] Example 11. Human IgG competition for VHH binding to native CD16A

[0353] ​To assess the effect of human IgG competition on the ability of anti-CD16A VHH-Fc to bind native CD16A on live cells, FACS-based assays were performed in the presence or absence of human IgG. Live NK92mi / CD16A cells were seeded in 96-well plates and incubated at 37°C with serial dilutions of biotinylated anti-CD16A VHH-Fc alone or with 10 mg / ml of human IgG1 antibody CB6 (anti-SARS-Covid19 antibody) (SEQ ID NOs: 121-122). Streptavidin-AF647 was used as a secondary antibody to detect the binding of biotinylated anti-CD16A VHH-Fc to the cell surface. EC 50 Values were determined by fitting the dose response data to a four-parameter logistic model in GraphPad Prism. As Figures 14A-14C As shown in Table 20 and Figure 19, the presence of human IgG similarly affected the binding of BG525P and BG526P to CD16A 158F compared to the binding of the parental BG523P.

[0354] Table 20. Human IgG competition of binding affinity of BG523P-derived anti-CD16A VHH to human CD16A 158F by FACS

[0355]

[0356]

[0357] Example 12. Binding affinity of humanized CD16A to cynomolgus CD16 by SPR

[0358] For affinity determination, VHH-Fc were captured by an anti-human Fc surface and used in an affinity determination based on the surface plasmon resonance (SPR) technology. The results of the SPR determination of the binding profile of anti-CD16A VHH-Fc are summarized in Table 21. Humanized anti-CD16A VHH-Fc retained cross-reactivity to cynomolgus CD16.

[0359] Table 21. Binding affinity of anti-CD16A VHH to cynomolgus CD16 by SPR

[0360]

[0361] Table 22. List of CD16A and CD16B sequences

[0362]

[0363]

[0364] Table 23. Anti-CD16A VHH sequence list

[0365]

[0366]

[0367]

[0368]

[0369] Table 24. Amino acid and DNA sequences of MUC1xCD16A multispecific antibodies

[0370]

[0371]

[0372]

[0373]

[0374]

[0375] Table 25. Summary table of various anti-human MUC1 reference antibodies

[0376]

[0377]

[0378] Example 13. Generation and production of MUC1 -targeting antibodies

[0379] Generation and production of MUC1xCD16A multispecific antibody BG1222P

[0380] For construction of MUC1xCD16A multispecific antibody BG1222P, anti-CD16A VHH BG526P and anti-MUC1 antibody huBG219P-E39 were assembled in a “2+1” IgG-like bispecific fashion, facilitating knob-into-hole (KiH) mutations for Fc dimerization Figure 15). Specifically, the tandem BG526P VHH with a 2G4S linker (GGGGSGGGGS, SEQ ID NO: 72) in between was fused N-terminally to the hinge region of a human IgGl constant region carrying a T366W (EU numbering) mutation for the“knob” mutation, C220S (EU numbering) for removal of free cysteines and M252Y / S254T / T256E (chain 1, SEQ ID NO: 143-144) for half-life extension. For the MUC1 binding arm, the humanized MUC1 antibody huBG219P-E39 VH region was fused to the constant region of human IgGl carrying T366S / L368A / Y407V (EU numbering) for the“hole” mutation and M252Y / S254T / T256E (chain 2, SEQ ID NO: 145-146) for half-life extension. The light chain of BG1222P was made from the humanized MUC1 antibody huBG219P-E39 VL region fused to the constant region of human kappa chain (chain 3, SEQ ID NO: 147-148). Those constructs were made using either in-house developed expression vectors or pcDNA3.4 with easy-to-adapt subcloning sites.

[0381] All three plasmids were co-transfected into ExpiCHO-s cells (Thermofisher Scientific). The plasmid ratio was optimized to facilitate downstream purification procedures by improving the purity of the starting material. The bispecific antibody was first captured with MabSelect SuRe Lx (Cytiva) and further polished by two ion exchange columns Capto S ImpAct and Capto Q ImpRes (Cytiva) to remove most impurities and aggregates. The final product buffer was exchanged into DPBS or histidine buffer using G25 desalting columns (Cytiva) and stored at -80°C.

[0382] Generation of various anti-human MUC1 reference antibodies

[0383] Various anti-human MUC1 reference antibody sequences were extracted from public literature and patents as summarized in Table 25. The above-mentioned reference antibodies were constructed with in-house IgGl / Ck eukaryotic expression vectors.

[0384] For antibodies including the above reference antibodies and a non-fucosylated version of huBG219P-E39-AF (hereafter referred to with the suffix -AF), if used in the assay, were produced with the ExpiCHO transient expression system (Thermofisher Scientific). For inhibition of fucosylation, 2F-Peracetyl-fucose (cat# 344827, EMD Millipore) was added to the growth medium at a final concentration of 100 mM prior to inoculation. Conditioned media were harvested and non-fucosylated antibodies were purified using MabSelect SuRe columns (Cytiva) followed by SPHP columns (Cytiva). All purified antibodies were buffer exchanged to DPBS via UF / DF and stored at -80 °C in small aliquots for later assays.

[0385] Example 14. Binding kinetics and affinity determination of MUC1xCD16A multispecific antibodies

[0386] For affinity determination, an assay based on surface plasmon resonance (SPR) technology was developed to characterize the binding affinity of bispecific antibodies. Briefly, netrAvidin was immobilized onto the surface of a CM5 chip, then biotin-labeled anti-human Fc VHH was flowed over the surface and captured by the immobilized netrAvidin. MUC1xCD16A multispecific antibodies were captured by the anti-human Fc VHH / netrAvidin complex on the chip surface, and serial dilutions of CD16A or hMUC1-SEA-mFc (human MUC1-SEA domain linked to mouse IgG2a Fc) were flowed over the surface and the binding response was calculated by subtracting the RU from a reference flow cell that was not injected with bispecific antibodies. The results of the SPR determination of the binding profile of MUC1xCD16A multispecific antibodies are summarized in Table 26. The multispecific antibody BG1222P showed high binding affinity to the targets human CD16A (158V and 158F) and hMUC1-SEA.

[0387] Table 26. SPR binding affinity of multispecific antibodies to human CD16A and hMUC1-SEA-mFc

[0388] Analyte k on (M -1 s -1 )]]> k off (s -1 )]]> K D (nM) CD16A 158V 8.86E+05 4.22E-05 4.77E-11 CD16A 158F 5.79E+05 8.52E-05 1.47E-10 hMUC1-SEA-mFc 1.17E+06 3.35E-05 2.86E-11

[0389] Example 15. Determination of binding affinity of MUC1xCD16A multispecific antibodies

[0390] Binding affinity of MUC1xCD16A multispecific antibodies to MUC1 -expressing cancer cell lines

[0391] The binding affinity of purified BG1222P to MUC1 -expressing cancer cell lines and NK92mi was determined by FACS. Briefly, MUC1 -expressing tumor cell line T47D was incubated with serially diluted purified BG1222P, washed, and incubated with anti-human IgG secondary antibody conjugated to APC. After incubation and washing, fluorescence was measured by flow cytometry. The binding affinity of BG1222P to T47D is shown in Table 27 and Figure 16 Table 28 below. The results indicate that BG1222P has high binding affinity to MUC1 -expressing cancer cell lines.

[0392] Table 27. Cell binding affinity of MUC1xCD16A multispecific antibody BG1222P to T47D

[0393] Antibody Cell Line EC 50 (nM) BG1222P T47D 28.07

[0394] Binding affinity of MUC1xCD16A multispecific antibodies to NK92mi / CD16A F158 and NK92mi / CD16A V158 cells

[0395] To assess the binding activity of BG1222P and huBG219P-E39-AF antibodies to CD16A expressed on live cells, NK92mi (ATCC) cells were engineered to overexpress human CD16A (NK92mi / CD16A F158 and NK92mi / CD16A V158) by co-transducing expression plasmids containing CD16A (F158 or V158 alleles) and FcRy cDNAs. Purified and biotinylated bispecific antibodies were serially diluted and incubated with NK92mi / CD16A F158 or NK92mi / CD16A V158 cells at 37°C for 45 min in the presence or absence of 10 mg / ml human IgG. After two washes with FACS buffer, diluted Alexa Fluor 647 streptavidin (Invitrogen #S32357) was added and incubated with cells at 4°C for 60 min in the dark. After two washes with FACS buffer, cells were resuspended with FACS buffer and acquired on a BD FACSCelesta. Titration curves were generated using a nonlinear fit sigmoidal dose response from GraphPad, and EC 50 Table 28 below. The results indicate that BG1222P has high binding affinity to MUC1 -expressing cancer cell lines. Figure 17BG1222P exhibited strong binding affinity to F158 and V158 human CD16A overexpressing cell lines. Human IgG competition attenuated the binding Emax of huBG219P-E39-AF, but not BG1222P. The results indicate that the CD16A binding affinity of BG1222P can be better maintained compared to huBG219P-E39-AF in circulation or tumors where there is a high level of IgG. The CD16A binding activity of BG1222P is less attenuated by IgG competition compared to that for huBG219P-E39-AF.

[0396] Table 28. Comparison of human CD16A overexpressing NK92mi binding affinity of MUC1xCD16A multispecific antibodies and MUC1 antibodies

[0397]

[0398] Example 16. Antibody-dependent cellular cytotoxicity activity of MUC1xCD16A multispecific antibodies

[0399] A nanoluc-release assay was established to determine the antibody-dependent cellular cytotoxicity activity of BG1222P and huBG219P-E39-AF antibodies against MUC1 + cell lines were engineered to express nanoluc in the cells by retroviral transduction and used as target cells. When target cells were lysed by effector cells, nanoluc was released in the culture media. Cytotoxicity was assessed by measuring nanoluc in the supernatant using the Nano-Glo Luciferase Assay Reagent Kit (Promega, Madison, Wis.). Briefly, effector and target cells with a 2: 1 E:T ratio were added to V-bottom 96-well plates with serial dilutions of bispecific antibodies in the presence or absence of 10 mg / ml human IgGl and co-cultured at 37°C for 20-24 hours. Specific lysis was determined using the following equation: % of specific lysis = [luminescence (sample) - luminescence (spontaneous)] / [luminescence (max) - luminescence (spontaneous)] x 100%. Luminescence (spontaneous) represents luminescence counts from supernatant of target cells without effector cells and antibodies. Luminescence (max) represents luminescence counts released after total cell lysis induced by the addition of Triton-X-100. As shown in Tables 29 and 30 and Figure 18A-19H shows that BG1222P induces lysis of MUC1 -expressing cell lines in a dose-dependent manner in the presence or absence of human IgG but does not induce MUC1 -negative cell line MDA-MB-453. The results indicate that the ADCC activity of BG1222P is superior to that of huBG219P-E39-AF.

[0400] Table 29. Comparison of antibody-dependent cellular cytotoxicity activity of MUC1xCD16A multispecific antibodies and MUC1 antibodies mediated by NK92mi / CD16A F158 cells

[0401]

[0402] Table 30. Comparison of antibody-dependent cellular cytotoxicity activity of MUC1xCD16A multispecific antibodies and MUC1 antibodies mediated by NK92mi / CD16A V158 cells

[0403]

[0404] Example 17. Cell lysis activity of MUC1xCD16A multispecific antibodies for MUC1 -expressing cells in a human whole blood assay

[0405] The nanoluc-release assay was also used to assess the cell lysis activity of BG1222P in human whole blood. Briefly, 100 μL / well human whole blood from healthy donors was mixed with target cells (2000 cells / well) described in Example 16 (T-47D / nanoluc, HCC827 / nanoluc, H358 / nanoluc, and MDA-MB-453 / nanoluc) and serial dilutions of bispecific antibodies in U-bottom 96-well plates. The total volume was 200 μL / well. After incubation at 37 °C for 18-20 hours, nanoluc released into the supernatant was measured using the Nano-Glo Luciferase Assay Reagent Kit. Specific lysis was determined using the equation described in Example 16.

[0406] To compare the cell lysis activity of the bispecific antibodies to that of the anti-MUC1 monoclonal IgGl antibodies, the following antibodies with engineered Fc (non-fucosylated, -AF) were generated with reference to published sequences: galantamine-AF, crizotinib-AF, HuVH-HMFG1-AF, MUC1 5F3-hFc-AF, and MUC1 3D1-hFc-AF. Removal of the core fucose of the Fc glycan (non-fucosylation) has been shown to highly increase the FcyRIIIa binding affinity and thus the cytotoxic activity of the antibodies. In the human whole blood assay as described above, these antibodies were compared to the bispecific antibodies using T-47D / nanoluc as target cells. As shown in Tables 31 and Figure 20 BG1222P exhibited a much stronger cell lysis activity in terms of EC 50 and Emaxcompared to the aforementioned anti-MUC1 non-fucosylated antibodies.

[0407] In addition to T47D cells, the cell lysis activity of BG1222P for target cells with different MUC1 expression levels was also evaluated in the human whole blood assay. As shown in Tables 32 and Figure 21 A-21D demonstrated that BG1222P specifically induced lysis of MUC1 expressing cell lines in a dose-dependent manner in human whole blood, but did not induce lysis of MUC1 negative cell line MDA-MB-453. The results indicated that the cell lysis activity of BG1222P was superior to that of huBG219P-E39-AF.

[0408] Table 31. Comparison of cell lysis activity of MUC1xCD16A multispecific antibodies and various MUC1 non-fucosylated antibodies in human whole blood

[0409]

[0410] Table 32. Comparison of cell lysis activity of MUC1xCD16A multispecific antibodies and MUC1 antibodies in human whole blood

[0411]

[0412] Example 18. Phagocytic activity of MUC1xCD16A multispecific antibodies

[0413] Human PBMC-derived M2 macrophages were used as effector cells to assess the phagocytic activity of BG1222P. The generation of M2 macrophages was performed according to the protocol described by Leidi et al. (Journal of immunology, (2009) 182(7), 4415-4422). Briefly, human PBMC (Sailybio) were cultured in 6-well plates (Corning) in complete RPMI1640 medium supplemented with 30 ng / ml human M-CSF (Peprotech) for 4 days. Adherent cells were retained by gently washing away non-adherent and loosely adherent cells, with half of the medium changed, and cultured for an additional 2-3 days. For M2 polarization, 10 ng / ml IL-10 (Peprotech) was added during the last 48h of culture.

[0414] Target cells (T-47D and MDA-MB-453) were labeled with carboxyfluorescein succinimidyl ester (CFSE) (Life Technologies) according to the manufacturer's instructions. A 2:1 ratio of target cells and M2 macrophages were plated with bispecific antibodies in the presence or absence of 10 mg / ml human IgG in U-bottom 96-well plates. After 2h incubation at 37°C, cells were stained with anti-CD11b-BV421 and subjected to flow cytometry. The percentage of macrophages that underwent antibody-dependent cellular phagocytosis of target cells was determined by FACS of double positive cells (CFSE+ and CD11b+) after gating on CD11b+ M2 macrophages. As shown in Table 33 and Figure 22 A-22B demonstrated that BG1222P showed better phagocytic activity than huBG219P-E39-AF in a dose-dependent manner in MUC1 -expressing cell line T47D with or without the addition of human IgG. BG1222P did not exhibit activity on MUC1 -negative cell line MDA-MB-453.

[0415] Table 33. Comparison of phagocytic activity of MUC1xCD16A multispecific antibodies and MUC1 antibodies mediated by human macrophage M2

[0416]

[0417] Example 19. NK fratricide of MUC1xCD16A multispecific antibodies

[0418] A flow cytometry-based assay was established to determine the NK fratricidal activity of BG1222P. Primary NK cells were isolated from PBMCs of healthy donors using an NK cell isolation kit from Miltenyi Biotec (Germany) according to the manufacturer's instructions. The isolated NK cells were cultured with serial dilutions of BG1222P in V-bottom 96-well plates. Daratumumab, which is known to exhibit NK fratricidal activity at the cellular level and in patients, was used as a positive control. After incubation for 5 hours at 37°C, cells were stained with anti-CD3-BV421, Annexin V-FITC, 7-AAD and anti-CD56-AF647. The percentage of apoptotic and dead NK cells was determined by FACS of Annexin V-positive and double-positive cells (Annexin V+ and 7-AAD+) after gating on CD3-CD56+ NK cells. As Figure 23 As shown in A-23B, BG1222P did not show fratricidal activity in NK cells from either donor. In contrast, daratumumab exhibited dose-dependent cytotoxic activity against NK cells from both donors.

[0419] Example 20. Pharmacokinetic Profile of MUC1xCD16A Multispecific Antibody in Cynomolgus Monkeys

[0420] Blood samples were collected from cynomolgus monkeys at 0, 0.5, 1, 4, 8, 1, 3, 7, 10, 14, 21, and 28 days after intravenous infusion of 5 mg / kg or 25 mg / kg BG1222P, followed by centrifugation (4°C, 3000×g, 15 min) to separate serum. The concentration of BG1222P was measured by an in-house developed ELISA ligand binding method. Briefly, HuMUC1-SEA-mFc was used as a capture reagent, and biotin-labeled CD16A-V158 with a His tag was used as a detection reagent for BG1222P. The pharmacokinetic profiles and parameters obtained were as follows: Figure 24 and shown in Table 34. In the 5 mg / kg dosing group, BG1222P was below the lower limit of quantification (0.78 μg / ml) on the 21st day after dosing. In the 5 mg / kg dosing group, anti-drug antibodies (ADA) in the serum were detected from the 10th day, indicating a potential impact on the pharmacokinetic curve. In the 25 mg / kg dosing group, BG1222P showed much lower clearance and minimal ADA impact. Even in these cases, after the 5-25 mg / kg dose, a relatively long terminal elimination phase half-life of BG1222P in the range of 4.25 to 10.2 days was observed. The clearance rate in the range of 4.3-9.1 days indicates that BG1222P is slowly cleared from the body. The volume of distribution (V z) is close to the physiological serum volume in cynomolgus monkeys, indicating that BG1222P is mainly located in the serum volume.

[0421] Table 34. Pharmacokinetic parameters of MUC1xCD16A multispecific antibodies in cynomolgus monkeys following intravenous infusion

[0422]

[0423] Table 35. Sequences of amino acid linkers of scFv

[0424] SEQ ID NO Annotation Sequence SEQ ID NO: 70 AA linker GGGGSGGGGSGGGGSGGGGS SEQ ID NO: 71 AA linker GGGGSGGGGSGGGGS SEQ ID NO: 72 AA linker GGGGSGGGGS

[0425] Example 21. MUC1xCD16A Multispecific Antibody Shows Reduced Interference by Soluble MUC1

[0426] The effect of the presence of soluble MUC1 on the specific binding of the MUC1xCD16A multispecific antibody to MUC1-expressing cells was determined by competitive FACS assay. Briefly, human MUC1-expressing cells were incubated with 30, 3, or 0.3 μg / ml BG1222P in the presence of serially diluted soluble MUC1 (Shanghai Linc-Bio Science Co. LTD). After washing and incubation with an anti-human IgG secondary antibody, fluorescence was measured by flow cytometry. The IC of soluble MUC1 blocking BG1222P binding to MUC1-expressing cells was 0.1%. 50 The values ​​are shown in Table 36, and the blocking curves are Figures 25A-25C The profiles indicate that HuVH-HMFG binding to MUC1 expressing cells can be easily perturbed at high, medium and low antibody concentrations (i.e., 30, 3, 0.3 μg / ml), but BG1222P binding is only slightly perturbed at low antibody concentration (0.3 μg / ml) ( Figures 25A-25C ). Overall, the profiles in FIG25 indicate that binding of BG1222P to MUCl expressing cells showed significantly reduced interference by soluble MUCl compared to HuVH-HMFG1, which targets the membrane distal portion of MUCl.

[0427] Table 36. IC of Soluble MUC1 Blocking Activity 50

[0428]

[0429] (NA: Not available. Figures 25A-25C As shown in the binding curve, IC 50 The data cannot be well fitted and extracted due to extremely low noise.)

Claims

1. A multispecific antibody or antigen-binding fragment thereof, comprising a first antigen-binding domain that specifically binds to human MUC1 and a second antigen-binding domain that specifically binds to human CD16A. 2 . The multispecific antibody or antigen-binding fragment of claim 1 , wherein the first antigen-binding domain has high selectivity over human CD16B.

3. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the first antigen-binding domain that specifically binds to human MUC1 comprises: (i) A heavy chain variable region comprising (a) a heavy chain complementarity determining region 1 (HCDR1) of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) a light chain complementarity determining region 1 (LCDR1) of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 28, and (f) LCDR3 of SEQ ID NO: 29; (ii) A heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 65, and (f) LCDR3 of SEQ ID NO: 29; (iii) comprising a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 4, (b) HCDR2 of SEQ ID NO: 5, and (c) HCDR3 of SEQ ID NO: 6, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 7, (e) LCDR2 of SEQ ID NO: 8, and (f) LCDR3 of SEQ ID NO: 9; or (iv). A heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 14, (b) HCDR2 of SEQ ID NO: 15, and (c) HCDR3 of SEQ ID NO: 16, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 17, (e) LCDR2 of SEQ ID NO: 18, and (f) LCDR3 of SEQ ID NO:

19.

4. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the first antigen-binding domain comprises: (i) a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 30, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 31; (ii) a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 61, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 62; (iii) a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 61, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 66; (iv) a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 61, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 68; (v) a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 10, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 11; or (vi). a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 20, and a light chain variable region (VL) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:

21.

5. The multispecific antibody or antigen-binding fragment of any of the preceding claims, wherein one, two, three, four, five, six, seven, eight, nine or ten amino acids have been inserted, deleted or substituted within one or more of SEQ ID NOs: 30, 31, 61, 62, 66, 68, 10, 11, 20 and 21.

6. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the first antigen-binding domain comprises: (i) A heavy chain variable region (VH) comprising SEQ ID NO: 30, and a light chain variable region (VL) comprising SEQ ID NO: 31; (ii) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 62; (iii) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 66; (iv) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 68; (v) a heavy chain variable region (VH) comprising SEQ ID NO: 10, and a light chain variable region (VL) comprising SEQ ID NO: 11; or (vi). A heavy chain variable region (VH) comprising SEQ ID NO: 20, and a light chain variable region (VL) comprising SEQ ID NO:

21.

7. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the second antigen-binding domain that specifically binds to human CD16A comprises: (i) A heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 109, (b) HCDR2 of SEQ ID NO: 110, and (c) HCDR3 of SEQ ID NO: 111; or (ii). A heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 109, (b) HCDR2 of SEQ ID NO: 114, and (c) HCDR3 of SEQ ID NO:

111.

8. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the second antigen-binding domain comprises: (i) A heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 112; (ii) a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 115; (iii) a heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO: 117; or (iv). A heavy chain variable region (VH) comprising an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% identical to SEQ ID NO:

119.

9. The multispecific antibody or antigen-binding fragment of any of the preceding claims, wherein one, two, three, four, five, six, seven, eight, nine or ten amino acids have been inserted, deleted or substituted within one or more of SEQ ID NOs: 112, 115, 117 and 119.

10. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the second antigen-binding domain comprises: (i) A heavy chain variable region (VH) comprising SEQ ID NO: 112; (ii) A heavy chain variable region (VH) comprising SEQ ID NO: 115; (iii) a heavy chain variable region (VH) comprising SEQ ID NO: 117; or (iv). A heavy chain variable region (VH) comprising SEQ ID NO:

119.

11. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein: (i) The first antigen-binding domain that specifically binds to human MUC1 comprises: a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 28, and (f) LCDR3 of SEQ ID NO: 29; and the second antigen-binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 110; and (c) HCDR3 of SEQ ID NO: 111; (ii). The first antigen-binding domain that specifically binds to human MUC1 comprises: a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 65, and (f) LCDR3 of SEQ ID NO: 29; and the second antigen-binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 110; and (c) HCDR3 of SEQ ID NO: 111; (iii). The first antigen-binding domain that specifically binds to human MUC1 comprises: a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 4, (b) HCDR2 of SEQ ID NO: 5, and (c) HCDR3 of SEQ ID NO: 6, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 7, (e) LCDR2 of SEQ ID NO: 8, and (f) LCDR3 of SEQ ID NO: 9; and the second antigen-binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 110; and (c) HCDR3 of SEQ ID NO: 111; (iv) comprising a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 14, (b) HCDR2 of SEQ ID NO: 15, and (c) HCDR3 of SEQ ID NO: 16, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 17, (e) LCDR2 of SEQ ID NO: 18, and (f) LCDR3 of SEQ ID NO: 19; and the second antigen-binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 110; and (c) HCDR3 of SEQ ID NO: 111; (v) the first antigen-binding domain that specifically binds to human MUC1 comprises: a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 28, and (f) LCDR3 of SEQ ID NO: 29; and the second antigen-binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 114; and (c) HCDR3 of SEQ ID NO: 111; (vi) the first antigen-binding domain that specifically binds to human MUC1 comprises: a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 65, and (f) LCDR3 of SEQ ID NO: 29; and the second antigen-binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 114; and (c) HCDR3 of SEQ ID NO: 111; (vii). The first antigen-binding domain that specifically binds to human MUC1 comprises: a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 4, (b) HCDR2 of SEQ ID NO: 5, and (c) HCDR3 of SEQ ID NO: 6, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 7, (e) LCDR2 of SEQ ID NO: 8, and (f) LCDR3 of SEQ ID NO: 9; and the second antigen-binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 114; and (c) HCDR3 of SEQ ID NO: 111; or (viii). A heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 14, (b) HCDR2 of SEQ ID NO: 15, and (c) HCDR3 of SEQ ID NO: 16, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 17, (e) LCDR2 of SEQ ID NO: 18, and (f) LCDR3 of SEQ ID NO: 19; and a second antigen-binding domain that specifically binds to human CD16A comprises: (a) HCDR1 of SEQ ID NO: 109; (b) HCDR2 of SEQ ID NO: 114; and (c) HCDR3 of SEQ ID NO:

111.

12. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein: (i) The first antigen-binding domain that specifically binds to human MUC1 comprises: a) a heavy chain variable region (VH) comprising SEQ ID NO: 30, and a light chain variable region (VL) comprising SEQ ID NO: 31; b) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 62; c) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 66; d) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 68; e) a heavy chain variable region (VH) comprising SEQ ID NO: 10, and a light chain variable region (VL) comprising SEQ ID NO: 11; or f) a heavy chain variable region (VH) comprising SEQ ID NO: 20, and a light chain variable region (VL) comprising SEQ ID NO: 21; (ii) The second antigen-binding domain that specifically binds to human CD16A comprises: a) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 109, (b) HCDR2 of SEQ ID NO: 110, (c) HCDR3 of SEQ ID NO: 111; or b) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 109, (b) HCDR2 of SEQ ID NO: 114, and (c) HCDR3 of SEQ ID NO:

111.

13. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein: (i) The first antigen-binding domain that specifically binds to human MUC1 comprises: a) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 28, and (f) LCDR3 of SEQ ID NO: 29; b) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 24, (b) HCDR2 of SEQ ID NO: 25, and (c) HCDR3 of SEQ ID NO: 26, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 27, (e) LCDR2 of SEQ ID NO: 65, and (f) LCDR3 of SEQ ID NO: 29; c) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 4, (b) HCDR2 of SEQ ID NO: 5, and (c) HCDR3 of SEQ ID NO: 6, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 7, (e) LCDR2 of SEQ ID NO: 8, and (f) LCDR3 of SEQ ID NO: 9; or d) a heavy chain variable region comprising (a) HCDR1 of SEQ ID NO: 14, (b) HCDR2 of SEQ ID NO: 15, and (c) HCDR3 of SEQ ID NO: 16, and a light chain variable region comprising (d) LCDR1 of SEQ ID NO: 17, (e) LCDR2 of SEQ ID NO: 18, and (f) LCDR3 of SEQ ID NO: 19; (ii) The second antigen-binding domain that specifically binds to human CD16A comprises: a) a heavy chain variable region (VH) comprising SEQ ID NO: 112; b) a heavy chain variable region (VH) comprising SEQ ID NO: 115; c) a heavy chain variable region (VH) comprising SEQ ID NO: 117; or d) a heavy chain variable region (VH) comprising SEQ ID NO:

119.

14. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein: (i) The first antigen-binding domain that specifically binds to human MUC1 comprises: a) a heavy chain variable region (VH) comprising SEQ ID NO: 30, and a light chain variable region (VL) comprising SEQ ID NO: 31; b) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 62; c) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 66; d) a heavy chain variable region (VH) comprising SEQ ID NO: 61, and a light chain variable region (VL) comprising SEQ ID NO: 68; e) a heavy chain variable region (VH) comprising SEQ ID NO: 10, and a light chain variable region (VL) comprising SEQ ID NO: 11; or f) a heavy chain variable region (VH) comprising SEQ ID NO: 20, and a light chain variable region (VL) comprising SEQ ID NO: 21; (ii) The second antigen-binding domain that specifically binds to human CD16A comprises: a) a heavy chain variable region (VH) comprising SEQ ID NO: 112; b) a heavy chain variable region (VH) comprising SEQ ID NO: 115; c) a heavy chain variable region (VH) comprising SEQ ID NO: 117; or d) a heavy chain variable region (VH) comprising SEQ ID NO:

119.

15. The multispecific antibody or antigen-binding fragment of any preceding claim, which is a monoclonal antibody, a chimeric antibody, a humanized antibody, a human engineered antibody, a single-chain antibody (scFv), a Fab fragment, a Fab' fragment, or a F(ab')2 fragment.

16. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the multispecific antibody is a bispecific antibody.

17. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the multispecific antibody is BG1222P (SEQ ID NO: 143, SEQ ID NO: 145, and SEQ ID NO: 147).

18. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the antibody or antigen-binding fragment thereof has antibody-dependent cellular cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC).

19. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the antibody or antigen-binding fragment thereof has reduced glycosylation or no glycosylation or is hypofucosylated.

20. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the antibody or antigen-binding fragment thereof comprises an increased bisecting GlcNac structure.

21. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the Fc domain is IgG1 with reduced effector function.

22. The multispecific antibody or antigen-binding fragment of any one of the preceding claims, wherein the Fc domain is IgG4.

23. A pharmaceutical composition comprising the multispecific antibody or antigen-binding fragment thereof according to any one of the preceding claims and a pharmaceutically acceptable carrier.

24. The pharmaceutical composition of claim 23, comprising histidine / histidine HCl, trehalose dihydrate, and polysorbate 20.

25. An isolated nucleic acid encoding the multispecific antibody or antigen-binding fragment of any one of claims 1 to 22. A vector comprising the nucleic acid of claim 25 .

27. A host cell comprising the nucleic acid of claim 25 or the vector of claim 26.

28. A method for producing a multispecific antibody or an antigen-binding fragment thereof, comprising culturing the host cell of claim 27 and recovering the antibody or antigen-binding fragment from the culture.

Images