PROTEINS BINDING NKG2D, CD16 AND CLEC12A

SEQUENCE LISTING

This application incorporates by reference in its entirety the Computer Readable Form (CRF) of a Sequence Listing in ASCII text format. The Sequence Listing text file is entitled “14247-540-999_SEQ_LISTING,” was created on May 3, 2021, and is 291,326 bytes in size.

FIELD OF THE INVENTION

The present application relates to multispecific binding proteins that bind NKG2D, CD16 and CLEC12A on a cell, pharmaceutical compositions comprising such proteins, and therapeutic methods using such proteins and pharmaceutical compositions, including for the treatment of cancer.

BACKGROUND

Despite substantial research efforts, cancer continues to be a significant clinical and financial burden in countries across the globe. According to the World Health Organization (WHO), it is the second leading cause of death. Surgery, radiation therapy, chemotherapy, biological therapy, immunotherapy, hormone therapy, stem-cell transplantation, and precision medicine are among the existing treatment modalities. Despite extensive research in these areas, a highly effective, curative solution, particularly for the most aggressive cancers, has yet to be identified. Furthermore, many of the existing anti-cancer treatment modalities have substantial adverse side effects.

Cancer immunotherapies are desirable because they are highly specific and can facilitate destruction of cancer cells using the patient's own immune system. Fusion proteins such as bi-specific T-cell engagers are cancer immunotherapies described in the literature that bind to tumor cells and T-cells to facilitate destruction of tumor cells. Antibodies that bind to certain tumor-associated antigens have been described in the literature. See, e.g., WO 2016/134371 and WO 2015/095412.

Natural killer (NK) cells are a component of the innate immune system and make up approximately 15% of circulating lymphocytes. NK cells infiltrate virtually all tissues and were originally characterized by their ability to kill tumor cells effectively without the need for prior sensitization. Activated NK cells kill target cells by means similar to cytotoxic T cells—i.e., via cytolytic granules that contain perforin and granzymes as well as via death receptor pathways. Activated NK cells also secrete inflammatory cytokines such as IFN-γ and chemokines that promote the recruitment of other leukocytes to the target tissue.

NK cells respond to signals through a variety of activating and inhibitory receptors on their surface. For example, when NK cells encounter healthy self-cells, their activity is inhibited through activation of the killer-cell immunoglobulin-like receptors (KIRs). Alternatively, when NK cells encounter foreign cells or cancer cells, they are activated via their activating receptors (e.g., NKG2D, NCRs, DNAM1). NK cells are also activated by the constant region of some immunoglobulins through CD16 receptors on their surface. The overall sensitivity of NK cells to activation depends on the sum of stimulatory and inhibitory signals. NKG2D is a type-II transmembrane protein that is expressed by essentially all natural killer cells where NKG2D serves as an activating receptor. NKG2D is also be found on T cells where it acts as a costimulatory receptor. The ability to modulate NK cell function via NKG2D is useful in various therapeutic contexts including malignancy.

C-type lectin domain family 12 member A (CLEC12A), also known as C-type lectin-like molecule-1 (CLL-1) or myeloid inhibitory C-type lectin-like receptor (MICL), is a member of the C-type lectin/C-type lectin-like domain (CTL/CTLD) superfamily. Members of this family share a common protein fold and have diverse functions, such as cell adhesion, cell-cell signaling, glycoprotein turnover, and roles in inflammation and immune response. CLEC12A, a type II transmembrane glycoprotein, is overexpressed in over 90% of acute myeloid leukemia patient on leukemic stem cells, but not on normal haematopoietic cells.

Despite many efforts undertaken by several biotech and pharma companies, development of specific CLEC12A targeted biologics is hindered by the absence of antibodies with good developability characteristics. The challenges in discovering CLEC12A antibodies may be attributed to the complexity of the antigen. CLEC12A is a monomeric heavily glycosylated protein with six potential N-glycosylation sites within an extracellular domain having 201 amino acids. Four out of six N-glycosylation sites are clustered in the membrane proximal domain of the molecule and are likely to be involved in presentation of the target on the cell surface. Variations in the glycosylation status of CLEC12A on the surface of different cell types has been reported (Marshall et al., (2006) Eur J Immunol. 36(8):2159-69).

Therefore, there remains a need in the field for new and useful proteins that bind CLEC12A for use in treatment of cancer.

SUMMARY

The present application provides multispecific binding proteins that bind to the NKG2D receptor and CD16 receptor on natural killer cells, and CLEC12A. Such proteins can engage more than one kind of NK-activating receptor, and may block the binding of natural ligands to NKG2D. In certain embodiments, the proteins can agonize NK cells in humans. In some embodiments, the proteins can agonize NK cells in humans and in other species such as rodents and cynomolgus monkeys. Formulations containing any one of the proteins disclosed herein; cells containing one or more nucleic acids expressing the proteins, and methods of enhancing tumor cell death using the proteins are also provided.

Accordingly, in one aspect, the present application provides a protein comprising:

(a) a first antigen-binding site that binds NKG2D;

(b) a second antigen-binding site that binds CLEC12A; and

wherein the second antigen-binding site that binds CLEC12A comprises:

(i) a heavy chain variable domain (VH) comprising complementarity-determining region 1 (CDR1), complementarity-determining region 2 (CDR2), and complementarity-determining region 3 (CDR3) comprising the amino acid sequences of SEQ ID NOs: 137, 272, and 273, respectively; and a light chain variable domain (VL) comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively;

(ii) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively;

(iii) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 251, 196, and 246, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively;

(v) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 221, 166, and 223, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 240, 226, and 179, respectively;

(v) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 206, 166, and 241, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 245, 239, and 179, respectively;

(vi) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 221, 236, and 223, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 152, 226, and 179, respectively;

(vii) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 159, 170, and 183, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 186, 188, and 189, respectively;

(viii) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 197, 202, and 207, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 211, 212, and 214, respectively;

(ix) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 220, 222, and 257, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 224, 225, and 227, respectively; or

(x) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 230, 231, and 232, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 233, 234, and 235, respectively.

In some embodiments, the second antigen-binding site that binds CLEC12A comprises a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 272, and 273, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In some embodiments, the second antigen-binding site that binds CLEC12A comprises a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively.

In certain embodiments, the VH comprises an amino acid sequence at least 90% identical to SEQ ID NO:335, and the VL comprises an amino acid sequence at least 90% identical to SEQ ID NO:336. In certain embodiments, the VH comprises an amino acid sequence at least 90% identical to SEQ ID NO:270, and the VL comprises an amino acid sequence at least 90% identical to SEQ ID NO:271. In certain embodiments, the VH comprises an amino acid sequence at least 90% identical to SEQ ID NO:178, and the VL comprises an amino acid sequence at least 90% identical to SEQ ID NO:289. In certain embodiments, the VH comprises the amino acid sequence of SEQ ID NO:335, and the VL comprises the amino acid sequence of SEQ ID NO:336. In certain embodiments, the VH comprises the amino acid sequence of SEQ ID NO:270, and the VL comprises the amino acid sequence of SEQ ID NO:271. In certain embodiments, the VH comprises the amino acid sequence of SEQ ID NO:178, and the VL comprises the amino acid sequence of SEQ ID NO:289. In certain embodiments, the VH and the VL comprise the amino acid sequences of SEQ ID NOs: 143 and 144; 147 and 144; 244 and 144; 178 and 144; 256 and 144; 143 and 163; 143 and 167; 143 and 171; or 174 and 175, respectively.

In some embodiments, the second antigen-binding site is present as a single-chain fragment variable (scFv), and wherein the scFv comprises an amino acid sequence selected from SEQ ID NOs: 274, 145, 146, 149, 150, 153, 154, 156, 157, 160, 161, 164, 165, 168, 169, 172, 173, 176, 177, 180, and 181. In certain embodiments, the scFv comprises the amino acid sequence of SEQ ID NO:274. In certain embodiments, the scFv comprises the amino acid sequence of SEQ ID NO:180.

In some embodiments, the second antigen-binding site that binds CLEC12A comprises a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 251, 196, and 246, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively.

In certain embodiments, the second antigen-binding site that binds CLEC12A comprises:

(i) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 195, 196, 187, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, 201, respectively;

(ii) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 187, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively; or

(iii) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 213, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively.

In certain embodiments, the VH comprises an amino acid sequence at least 90% identical to SEQ ID NO:148, and the VL comprises an amino acid sequence at least 90% identical to SEQ ID NO:203.

In certain embodiments, the VH comprises the amino acid sequence of SEQ ID NO:148, and the VL comprises the amino acid sequence of SEQ ID NO:203. In certain embodiments, the VH and the VL comprise the amino acid sequences of SEQ ID NOs: 162 and 203; 210 and 203; 162 and 218; 148 and 218; or 210 and 218, respectively. In some embodiments, second the antigen-binding site is present as an scFv, and wherein the scFv comprises an amino acid sequence selected from SEQ ID NOs: 184, 185, 204, 205, 208, 209, 215, 216, 252, 253, 254, and 255. In certain embodiments, the scFv comprises the amino acid sequence of SEQ ID NO:204.

In certain embodiments, the second antigen-binding site of a preceding embodiment binds CLEC12A in a glycosylation independent manner.

In another aspect, the present application provides a protein comprising:

(a) a first antigen-binding site that binds NKG2D;

(b) a second antigen-binding site that binds CLEC12A; and

wherein the second antigen-binding site binds CLEC12A in a glycosylation independent manner.

In some embodiments of a protein disclosed herein, the second antigen-binding site binds human CLEC12A with a dissociation constant (K_D) smaller than or equal to 20 nM as measured by surface plasmon resonance (SPR). In certain embodiments, the second antigen-binding site binds human CLEC12A with a K_Dsmaller than or equal to 1 nM as measured by SPR. In certain embodiments, the second antigen-binding site binds human CLEC12A comprising a K244Q substitution.

In some embodiments of a protein disclosed herein, the protein comprises an antibody Fc domain or a portion thereof sufficient to bind CD16.

In some embodiments of a protein disclosed herein, the first antigen-binding site that binds NKG2D is a Fab fragment, and the second antigen-binding site that binds CLEC12A is an scFv.

In some embodiments of a protein disclosed herein, the first antigen-binding site that binds NKG2D is an scFv, and the second antigen-binding site that binds CLEC12A is a Fab fragment.

In some embodiments of a protein disclosed herein, the protein further comprises an additional antigen-binding site that binds CLEC12A. In certain embodiments, the first antigen-binding site that binds NKG2D is an scFv, and the second and the additional antigen-binding sites that bind CLEC12A are each a Fab fragment. In certain embodiments, the first antigen-binding site that binds NKG2D is an scFv, and the second and the additional antigen-binding sites that bind CLEC12A are each an scFv. In certain embodiments, the amino acid sequences of the second and the additional antigen-binding sites are identical. In certain embodiments, the amino acid sequences of the second and the additional antigen-binding sites are different.

In another aspect, the present application provides a protein comprising:

(a) a first antigen-binding site that binds NKG2D;

(b) a second antigen-binding site that binds CLEC12A; and

wherein the first antigen-binding site that binds NKG2D is a Fab fragment, and the second antigen-binding site that binds CLEC12A is an scFv.

In some embodiments of a protein disclosed herein, the scFv that binds NKG2D is linked to an antibody constant domain or a portion thereof sufficient to bind CD16, via a hinge comprising Ala-Ser or Gly-Ser. In certain embodiments, the hinge further comprises an amino acid sequence Thr-Lys-Gly.

In some embodiments of a protein disclosed herein, each scFv that binds CLEC12A is linked to an antibody constant domain or a portion thereof sufficient to bind CD16, via a hinge comprising Ala-Ser or Gly-Ser. In certain embodiments, the hinge further comprises an amino acid sequence Thr-Lys-Gly.

In some embodiments of a protein disclosed herein, within the scFv that binds NKG2D, the heavy chain variable domain of the scFv forms a disulfide bridge with the light chain variable domain of the scFv. In certain embodiments, the disulfide bridge is formed between C44 of the heavy chain variable domain and C100 of the light chain variable domain, numbered under the Kabat numbering scheme

In some embodiments of a protein disclosed herein, within each scFv that binds CLEC12A, the heavy chain variable domain of the scFv forms a disulfide bridge with the light chain variable domain of the scFv. In certain embodiments, the disulfide bridge is formed between C44 of the heavy chain variable domain and C100 of the light chain variable domain, numbered under the Kabat numbering scheme.

In some embodiments of a protein disclosed herein, within the scFv that binds NKG2D, the heavy chain variable domain is linked to the light chain variable domain via a flexible linker. In certain embodiments, the flexible linker comprises (G₄S)₄.

In some embodiments of a protein disclosed herein, within each scFv that binds CLEC12A, the heavy chain variable domain is linked to the light chain variable domain via a flexible linker. In certain embodiments, the flexible linker comprises (G₄S)₄.

In some embodiments of a protein disclosed herein, within the scFv that binds NKG2D, the heavy chain variable domain is positioned at the C-terminus of the light chain variable domain.

In some embodiments of a protein disclosed herein, within each scFv that binds CLEC12A, the heavy chain variable domain is positioned at the C-terminus of the light chain variable domain.

In some embodiments of a protein disclosed herein, within the scFv that binds NKG2D, the heavy chain variable domain is positioned at the N-terminus of the light chain variable domain.

In some embodiments of a protein disclosed herein, within each scFv that binds CLEC12A, the heavy chain variable domain is positioned at the N-terminus of the light chain variable domain.

In some embodiments of a protein disclosed herein, the Fab fragment that binds NKG2D is not positioned between an antigen-binding site and the Fc or the portion thereof.

In some embodiments of a protein disclosed herein, no Fab fragment that binds CLEC12A is positioned between an antigen-binding site and the Fc or the portion thereof

In some embodiments of a protein disclosed herein, the first antigen-binding site that binds NKG2D comprises a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 81, 82, and 112, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively. In certain embodiments, the first antigen-binding site that binds NKG2D comprises:

(i) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 81, 82, and 97, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively; or

(ii) a VH comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 81, 82, and 84, respectively; and a VL comprising CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively.

In some embodiments of a protein disclosed herein, the VH of the first antigen-binding site comprises an amino acid sequence at least 90% identical to SEQ ID NO:95, and the VL of the first antigen-binding site comprises an amino acid sequence at least 90% identical to SEQ ID NO:85. In certain embodiments, the VH of the first antigen-binding site comprises the amino acid sequence of SEQ ID NO:95, and the VL of the first antigen-binding site comprises the amino acid sequence of SEQ ID NO:85.

In some embodiments of a protein disclosed herein, the antibody Fc domain is a human IgG1 antibody Fc domain. In some embodiments, the antibody Fc domain or the portion thereof comprises an amino acid sequence at least 90% identical to SEQ ID NO:118. In certain embodiments, at least one polypeptide chain of the antibody Fc domain comprises one or more mutations, relative to SEQ ID NO:118, at one or more positions selected from Q347, Y349, L351, S354, E356, E357, K360, Q362, S364, T366, L368, K370, N390, K392, T394, D399, S400, D401, F405, Y407, K409, T411, and K439, numbered according to the EU numbering system. In certain embodiments, at least one polypeptide chain of the antibody Fc domain comprises one or more mutations, relative to SEQ ID NO:118, selected from Q347E, Q347R, Y349S, Y349K, Y349T, Y349D, Y349E, Y349C, L351K, L351D, L351Y, S354C, E356K, E357Q, E357L, E357W, K360E, K360W, Q362E, S364K, S364E, S364H, S364D, T366V, T366I, T366L, T366M, T366K, T366W, T366S, L368E, L368A, L368D, K370S, N390D, N390E, K392L, K392M, K392V, K392F, K392D, K392E, T394F, D399R, D399K, D399V, S400K, S400R, D401K, F405A, F405T, Y407A, Y407I, Y407V, K409F, K409W, K409D, T411D, T411E, K439D, and K439E, numbered according to the EU numbering system. In certain embodiments, one polypeptide chain of the antibody heavy chain constant region comprises one or more mutations, relative to SEQ ID NO:118, at one or more positions selected from Q347, Y349, L351, S354, E356, E357, K360, Q362, S364, T366, L368, K370, K392, T394, D399, S400, D401, F405, Y407, K409, T411 and K439; and the other polypeptide chain of the antibody heavy chain constant region comprises one or more mutations, relative to SEQ ID NO:136, at one or more positions selected from Q347, Y349, L351, S354, E356, E357, S364, T366, L368, K370, N390, K392, T394, D399, D401, F405, Y407, K409, T411, and K439, numbered according to the EU numbering system. In certain embodiments, one polypeptide chain of the antibody heavy chain constant region comprises K360E and K409W substitutions relative to SEQ ID NO:118; and the other polypeptide chain of the antibody heavy chain constant region comprises Q347R, D399V and F405T substitutions relative to SEQ ID NO:136, numbered according to the EU numbering system. In certain embodiments, one polypeptide chain of the antibody heavy chain constant region comprises a Y349C substitution relative to SEQ ID NO:118; and the other polypeptide chain of the antibody heavy chain constant region comprises an S354C substitution relative to SEQ ID NO:136, numbered according to the EU numbering system.

In another aspect, the present application provides a protein comprising:

(a) a first polypeptide comprising the amino acid sequence of SEQ ID NO:259;

(b) a second polypeptide comprising the amino acid sequence of SEQ ID NO:260; and

In another aspect, the present application provides a protein comprising:

(a) a first polypeptide comprising the amino acid sequence of SEQ ID NO:276;

(b) a second polypeptide comprising the amino acid sequence of SEQ ID NO:260; and

In another aspect, the present application provides a pharmaceutical composition comprising a protein disclosed herein and a pharmaceutically acceptable carrier.

In another aspect, the present application provides a pharmaceutical formulation comprising a protein disclosed herein and a buffer at pH 7.0 or higher. In some embodiments, the pH of the formulation is between 7.0 and 8.0. In some embodiments, the buffer comprises citrate. In certain embodiments, the concentration of citrate is 10 to 20 mM. In some embodiments, the buffer further comprises phosphate. In some embodiments, the buffer further comprises a sugar or sugar alcohol. In certain embodiments, the sugar or sugar alcohol is a disaccharide. In certain embodiments, the disaccharide is sucrose. In some embodiments, the concentration of the sugar or sugar alcohol in the pharmaceutical formulation is 200 to 300 mM. In certain embodiments, the concentration of the sugar or sugar alcohol in the pharmaceutical formulation is about 250 mM. In some embodiments, the buffer further comprises a nonionic surfactant. In certain embodiments, the nonionic surfactant comprises a polysorbate. In certain embodiments, the polysorbate is polysorbate 80. In some embodiments, the concentration of polysorbate 80 in the pharmaceutical formulation is 0.005% to 0.05%. In certain embodiments, the concentration of polysorbate 80 in the pharmaceutical formulation is about 0.01%. In some embodiments, the concentration of NaCl, if any, is about 1 mM or lower in the pharmaceutical formulation. In some embodiments, the concentration of the protein is 10 to 200 mg/mL. In certain embodiments, the concentration of the protein is about 20 mg/mL.

In some embodiments, the pharmaceutical formulation comprises about 20 mM citrate, about 250 mM sucrose, and about 0.01% polysorbate 80, at pH 7.0. In certain embodiments, the pharmaceutical formulation comprises about 20 mg/mL of the protein, about 20 mM potassium phosphate, about 10 mM sodium citrate, and about 125 mM sodium chloride, at pH 7.8.

In another aspect, the present application provides a cell comprising one or more nucleic acids encoding a protein disclosed herein.

In another aspect, the present application provides a method of enhancing tumor cell death, the method comprising exposing the tumor cell and a natural killer cell to an effective amount of a protein, a pharmaceutical composition, or a pharmaceutical formulation disclosed herein.

In another aspect, the present application provides a method of treating cancer, the method comprising administering to a subject in need thereof an effective amount of a protein, a pharmaceutical composition, or a pharmaceutical formulation disclosed herein. In some embodiments, the cancer is a hematologic malignancy. In certain embodiments, the hematologic malignancy is selected from the group consisting of acute myeloid leukemia (AML), myelodysplastic syndrome (MDS), acute lymphoblastic leukemia (ALL), myeloproliferative neoplasms (MPNs), lymphoma, non-Hodgkin lymphomas, and classical Hodgkin lymphoma.

In some embodiments, the AML is selected from undifferentiated acute myeloblastic leukemia, acute myeloblastic leukemia with minimal maturation, acute myeloblastic leukemia with maturation, acute promyelocytic leukemia (APL), acute myelomonocytic leukemia, acute myelomonocytic leukemia with eosinophilia, acute monocytic leukemia, acute erythroid leukemia, acute megakaryoblastic leukemia (AMKL), acute basophilic leukemia, acute panmyelosis with fibrosis, and blastic plasmacytoid dendritic cell neoplasm (BPDCN). In certain embodiments, the AML is characterized by expression of CLL-1 on the AML leukemia stem cells (LSCs). In certain embodiments, the LSCs further express a membrane marker selected from CD34, CD38, CD123, TIM3, CD25, CD32, and CD96.

In some embodiments, the AML is a minimal residual disease (MRD). In certain embodiments, the MRD is characterized by the presence or absence of a mutation selected from FLT3-ITD ((Fms-like tyrosine kinase 3)-internal tandem duplications (ITD)), NPM1 (Nucleophosmin 1), DNMT3A (DNA methyltransferase gene DNMT3A), and IDH (Isocitrate dehydrogenase 1 and 2 (IDH1 and IDH2)).

In certain embodiments, the MDS is selected from MDS with multilineage dysplasia (MDS-MLD), MDS with single lineage dysplasia (MDS-SLD), MDS with ring sideroblasts (MDS-RS), MDS with excess blasts (MDS-EB), MDS with isolated del(5q), and MDS, unclassified (MDS-U). In certain embodiments, the MDS is a primary MDS or a secondary MDS.

In some embodiments, the ALL is selected from B-cell acute lymphoblastic leukemia (B-ALL) and T-cell acute lymphoblastic leukemia (T-ALL). In some embodiments, the MPN is selected from polycythaemia vera, essential thrombocythemia (ET), and myelofibrosis. In some embodiments, the non-Hodgkin lymphoma is selected from B-cell lymphoma and T-cell lymphoma. In some embodiments, the lymphoma is selected from chronic lymphocytic leukemia (CLL), lymphoblastic lymphoma (LPL), diffuse large B-cell lymphoma (DLBCL), Burkitt lymphoma (BL), primary mediastinal large B-cell lymphoma (PMBL), follicular lymphoma, mantle cell lymphoma, hairy cell leukemia, plasma cell myeloma (PCM) or multiple myeloma (MM), mature T/NK neoplasms, and histiocytic neoplasms. In certain embodiments, the cancer expresses CLEC12A.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a representation of a heterodimeric, multispecific antibody, e.g., a trispecific binding protein (TriNKET). Each arm can represent either the NKG2D-binding domain, or the binding domain corresponding to a tumor-associated antigen. In some embodiments, the NKG2D binding domain and the tumor-associated antigen binding domains can share a common light chain.

FIG. 2A-FIG. 2E illustrate five exemplary formats of a multispecific binding protein, e.g., a trispecific binding protein (TriNKET). As shown in FIG. 2A, either the NKG2D-binding domain or the tumor-associated antigen binding domain can take the scFv format (left arm). An antibody that contains a NKG2D-targeting scFv, a tumor-associated antigen targeting Fab fragment, and a heterodimerized antibody constant region is referred herein as the F3-TriNKET. An antibody that contains a tumor-associated antigen targeting scFv, a NKG2D-targeting Fab fragment, and a heterodimerized antibody constant region/domain that binds CD16 is referred herein as the F3′-TriNKET (FIG. 2E). As shown in FIG. 2B, both the NKG2D-binding domain and tumor-associated antigen binding domain can take the scFv format. FIGS. 2C to 2D are illustrations of an antibody with three antigen-binding sites, including two antigen-binding sites that bind the tumor-associated antigen, and the NKG2D-binding site fused to the heterodimerized antibody constant region. These antibody formats are referred herein as F4-TriNKET. FIG. 2C illustrates that the two tumor-associated antigen binding sites are in the Fab fragment format, and the NKG2D binding site in the scFv format. FIG. 2D illustrates that the tumor-associated antigen binding sites are in the scFv format, and the NKG2D binding site is in the scFv format. FIG. 2E represents a trispecific antibody (TriNKET) that contains a tumor-targeting scFv, a NKG2D-targeting Fab fragment, and a heterodimerized antibody constant region/domain (“CD domain”) that binds CD16. The antibody format is referred herein as F3′-TriNKET. In certain exemplary multispecific binding proteins, heterodimerization mutations on the antibody constant region include K360E and K409W on one constant domain; and Q347R, D399V and F405T on the opposite constant domain (shown as a triangular lock-and-key shape in the CD domains). The bold bar between the heavy and the light chain variable domains of the Fab fragments represents a disulfide bond.

FIG. 3 is a representation of a TriNKET in the Triomab form, which is a trifunctional, bispecific antibody that maintains an IgG-like shape. This chimera consists of two half antibodies, each with one light and one heavy chain, that originate from two parental antibodies. Triomab form may be a heterodimeric construct containing ½ of rat antibody and ½ of mouse antibody.

FIG. 4 is a representation of a TriNKET in the KiH Common Light Chain form, which involves the knobs-into-holes (KIHs) technology. KiH is a heterodimer containing 2 Fab fragments binding to target 1 and 2, and an Fc stabilized by heterodimerization mutations. TriNKET in the KiH format may be a heterodimeric construct with 2 Fab fragments binding to target 1 and target 2, containing two different heavy chains and a common light chain that pairs with both heavy chains.

FIG. 5 is a representation of a TriNKET in the dual-variable domain immunoglobulin (DVD-Ig™) form, which combines the target-binding domains of two monoclonal antibodies via flexible naturally occurring linkers, and yields a tetravalent IgG-like molecule. DVD-Ig™ is a homodimeric construct where variable domain targeting antigen 2 is fused to the N-terminus of a variable domain of Fab fragment targeting antigen 1. DVD-Ig™ form contains normal Fc.

FIG. 6 is a representation of a TriNKET in the Orthogonal Fab fragment interface (Ortho-Fab) form, which is a heterodimeric construct that contains 2 Fab fragments binding to target 1 and target 2 fused to Fc. Light chain (LC)-heavy chain (HC) pairing is ensured by orthogonal interface. Heterodimerization is ensured by mutations in the Fc.

FIG. 7 is a representation of a TriNKET in the 2-in-1 Ig format.

FIG. 8 is a representation of a TriNKET in the ES form, which is a heterodimeric construct containing two different Fab fragments binding to target 1 and target 2 fused to the Fc. Heterodimerization is ensured by electrostatic steering mutations in the Fc.

FIG. 9 is a representation of a TriNKET in the Fab Arm Exchange form: antibodies that exchange Fab fragment arms by swapping a heavy chain and attached light chain (half-molecule) with a heavy-light chain pair from another molecule, resulting in bispecific antibodies. Fab Arm Exchange form (cFae) is a heterodimer containing 2 Fab fragments binding to target 1 and 2, and an Fc stabilized by heterodimerization mutations.

FIG. 10 is a representation of a TriNKET in the SEED Body form, which is a heterodimer containing 2 Fab fragments binding to target 1 and 2, and an Fc stabilized by heterodimerization mutations.

FIG. 11 is a representation of a TriNKET in the LuZ-Y form, in which a leucine zipper is used to induce heterodimerization of two different HCs. The LuZ-Y form is a heterodimer containing two different scFabs binding to target 1 and 2, fused to Fc. Heterodimerization is ensured through leucine zipper motifs fused to C-terminus of Fc.

FIG. 12 is a representation of a TriNKET in the Cov-X-Body form.

FIG. 13A-FIG. 13B are representations of TriNKETs in the κλ-Body forms, which are heterodimeric constructs with two different Fab fragments fused to Fc stabilized by heterodimerization mutations: one Fab fragment targeting antigen 1 contains kappa LC, and the second Fab fragment targeting antigen 2 contains lambda LC. FIG. 13A is an exemplary representation of one form of a κλ-Body; FIG. 13B is an exemplary representation of another κλ-Body.

FIG. 14 is an Oasc-Fab heterodimeric construct that includes Fab fragment binding to target 1 and scFab binding to target 2, both of which are fused to the Fc domain. Heterodimerization is ensured by mutations in the Fc domain.

FIG. 15 is a DuetMab, which is a heterodimeric construct containing two different Fab fragments binding to antigens 1 and 2, and an Fc that is stabilized by heterodimerization mutations. Fab fragments 1 and 2 contain differential S—S bridges that ensure correct light chain and heavy chain pairing.

FIG. 16 is a CrossmAb, which is a heterodimeric construct with two different Fab fragments binding to targets 1 and 2, and an Fc stabilized by heterodimerization mutations. CL and CH1 domains, and VH and VL domains are switched, e.g., CH1 is fused in-line with VL, and CL is fused in-line with VH.

FIG. 17 is a Fit-Ig, which is a homodimeric construct where Fab fragment binding to antigen 2 is fused to the N-terminus of HC of Fab fragment that binds to antigen 1. The construct contains wild-type Fc.

FIG. 18A-FIG. 18P are a set of traces showing Bio-Layer Interferometry (BLI) profiles of antibodies collected from the murine hybridomas supernatants binding to hCLEC12A. Antibodies include 9E04 (FIG. 18A), 9F11 (FIG. 18B), 11E02 (FIG. 18C), 12F08 (FIG. 18D), 13E01 (FIG. 18E), 15A10 (FIG. 18F), 15D11 (FIG. 18G), 16B08 (FIG. 18H), 23D06 (FIG. 18I), 23A05 (FIG. 18J), 30A09 (FIG. 18K), 6D07 (FIG. 18L), 12B06 (FIG. 18M), 20G10 (FIG. 18N), 30H07 (FIG. 18O), and 32A03 (FIG. 18P).

FIG. 19A-FIG. 19P are a set of sensorgrams showing SPR profiles of antibodies collected from the murine mAb subclones binding to hCLEC12A (FIG. 19A-FIG. 19D, and FIG. 19I-FIG. 19L) and cCLEC12A (FIG. 19E-FIG. 19H, and FIG. 19M-FIG. 19P).

FIG. 20 is a line graph showing binding of purified CLEC12A subclones to PL21 AML cell line.

FIG. 21A-FIG. 21F are a set of sensorgrams showing SPR profiles of antibodies 16B8.C8 and 9F11.B7 binding to glycosylated (FIG. 21A and FIG. 21D), de-glycosylated (FIG. 21B and FIG. 21E), and de-sialylated (FIG. 21C and FIG. 21F) hCLEC12A.

FIG. 22A-FIG. 22S are a set of sensorgrams showing SPR profiles of TriNKETs derived from antibody 16B8.C8.

FIG. 23 is a line graph showing binding of hCLEC12A -targeting TriNKETs F3′-1304, F3′-1295 and a control CLEC12A-TriNKET to hCLEC12A-expressing cell line RMA-hCLEC12A.

FIG. 24 is a line graph showing NK cell-mediated lysis of CLEC12A-expressing cancer cell line HL60 in the presence of F3′-1295 and a control CLEC12A-TriNKET.

FIG. 25A-FIG. 25B are line graphs showing differential scanning colorimetry (DSC) analysis of TriNKETs F3′-129.

FIG. 26A-FIG. 26H are a set of sensorgrams showing SPR profiles of TriNKETs F3′-1295 (FIGS. 26A, 26B, 26E, and 26F) and F3′-1602 (FIGS. 26C, 26D, 26G, and 26H).

FIG. 27A-FIG. 27D are line graphs showing binding of hCLEC12A -targeting TriNKETs F3′-1295, F3′-1602, and a control CLEC12A-TriNKET to hCLEC12A-expressing cell line Ba/F3 (FIG. 27A), wild-type Ba/F3 (FIG. 27B), cancer line HL60 (FIG. 27C), and cancer line PL21 (FIG. 27D).

FIG. 28A-FIG. 28B are line graphs showing NK cell-mediated lysis of CLEC12A-expressing cancer cell lines PL21 (FIG. 28A) and HL60 (FIG. 28B) in the presence of F3′-1295, F3′-1602, and a control CLEC12A-TriNKET.

FIG. 29A-FIG. 29B are line graphs showing isoelectric point (pI) determined by capillary isoelectric focusing (cIEF) of F3′-1295 (FIG. 29A) and F3′-1602 (FIG. 29B).

FIG. 30A-FIG. 30D show flow cytometry based polyspecificity reagent (PSR) analysis of Rituximab (FIG. 30A), Trastuzumab (FIG. 30B), F3′-1602 TriNKET (FIG. 30C) and F3′-1295 TriNKET (FIG. 30D).

FIG. 31A-FIG. 31D is a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 31A and FIG. 31B) and F3-TriNKET AB0010 (FIG. 31C and FIG. 31D-).

FIG. 32A-FIG. 32C are line graphs showing binding of hF3′-1602 TriNKET and F3-TriNKET AB0010 to Ba/F3 expressing hCLEC12A (FIG. 32A), cancer line HL60 (FIG. 32B),and wild-type Ba/F3 (FIG. 32C).

FIG. 33A-FIG. 33B are line graphs showing NK cell-mediated lysis of CLEC12A-expressing cancer cell line PL21 using KHYG-CD16V cells (FIG. 33A) and primary NK cells (FIG. 33B) in the presence of hF3′-1602 TriNKET and F3-TriNKET AB0010.

FIG. 34A-FIG. 34B are line graphs showing binding of TriNKETs derived from 9F11.B7 and hF3′-1602 TriNKET and F3-TriNKET AB0010 to Ba/F3 expressing hCLEC12A (FIG. 34A) and cancer line HL60 (FIG. 34B).

FIG. 35 are line graphs showing NK cell-mediated lysis of CLEC12A-expressing cancer cell line HL60 in the presence of TriNKETs derived from 9F11.B7.

FIG. 36A-FIG. 36B are line graphs showing NK cell-mediated lysis of CLEC12A-expressing cancer cell line HL60 using KHYG-CD16V cells (FIG. 36A) and primary NK cells (FIG. 36B) in the presence of hF3′-1602 TriNKET and AB0192.

FIG. 37A-FIG. 37B are line graphs showing NK cell-mediated lysis of CLEC12A-expressing cancer cell line PL21 (FIG. 37A) and THP-1 (FIG. 37B) in the presence of hF3′-1602 TriNKET and parental mAb.

FIG. 38A-FIG. 38B are line graphs showing binding of hF3′-1602 TriNKET to cancer cell line HL60 (FIG. 38A) and PL21 (FIG. 38B).

FIG. 39A-FIG. 39C are line graphs showing binding of CLEC12A-targeting TriNKETs, parental mAbs, and AB0237 NKG2D dead arm variant TriNKET to Ba/F3 expressing hCLEC12A (FIG. 39A), and cancer cell lines HL60 (FIG. 39B) and PL21 (FIG. 39C).

FIG. 40A-FIG. 40B are bar graphs showing surface retention of CLEC12A-targeting TriNKETs and parental mAbs to cell lines PL21 (FIG. 40A) and HL60 (FIG. 40B).

FIG. 41A-FIG. 41B are line graphs showing NK cell-mediated lysis of CLEC12A-expressing cancer cell line PL21 (FIG. 41A) and HL60 (FIG. 41B) in the presence of hF3′-1602 TriNKET, AB0192, and parental mAbs.

FIG. 42A-FIG. 42B are line graphs showing NK cell-mediated lysis of CLEC12A-expressing cancer cell line PL21 (FIG. 42A) and HL60 (FIG. 42B) in the presence of hF3′-1602 TriNKET.

FIG. 43 are line graphs showing CD8-T cell-mediated lysis of CLEC12A-expressing cancer cell line PL21 in the presence of hF3′-1602 TriNKET, parental mAb, or an NKG2D-dead variant thereof.

FIG. 44A-FIG. 44I are a set of histograms showing binding of hF3′-1602 TriNKET (grey), parental mAb (white), and control IgG (black) to blood cells. Blood cells include granulocytes (FIG. 44A), monocytes (FIG. 44B), B cells (FIG. 44C), NK cells (FIG. 44D), NKT cells (FIG. 44E), Dendritic cells (FIG. 44F), CD4+ cells (FIG. 44G), CD8+ cells (FIG. 44H), and Lin-CD34+ cells (FIG. 44I).

FIG. 45A-FIG. 45H are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 45A-FIG. 45D) and trastuzumab control (FIG. 45E-FIG. 45H) binding to recombinant human CD64.

FIG. 46A-FIG. 46P are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 46A-FIG. 45H) and trastuzumab control (FIG. 46I-FIG. 46P) binding to CD32a H131.

FIG. 47A-FIG. 47P are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 47A-FIG. 47H) and trastuzumab control (FIG. 47I-FIG. 47P) binding to human CD32a R131 allele (FcγRIIa R131).

FIG. 48A-FIG. 48H are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 48A-FIG. 48D) and trastuzumab control (FIG. 48E-FIG. 48H) binding to human CD16a high affinity allele V158 (FcγRIIIa V158).

FIG. 49A-FIG. 49P are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 49A-FIG. 49H) and trastuzumab control (FIG. 49I-FIG. 49P) binding to human CD16a low affinity allele F158 (FcγRIIIa F158).

FIG. 50A-FIG. 50P are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 50A-FIG. 50H) and trastuzumab control (FIG. 50I-FIG. 50P) binding to human CD32b (FcγRIIb).

FIG. 51A-FIG. 51P are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 51A-FIG. 51H) and trastuzumab control (FIG. 51I-FIG. 51P) binding to human CD16b (FcγRIIIb).

FIG. 52A-FIG. 52P are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 52A-FIG. 52H) and trastuzumab control (FIG. 52I-FIG. 52P) binding to cynomolgus (cyno) CD16.

FIG. 53A-FIG. 53P are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 53A-FIG. 53H) and trastuzumab control (FIG. 53I-FIG. 53P) binding to human FcRn as measured at pH 6.0.

FIG. 54A-FIG. 54P are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET (FIG. 54A-FIG. 54H) and trastuzumab control (FIG. 54I-FIG. 54P) binding to cyno FcRn as measured at pH 6.0.

FIG. 55A-FIG. 55D are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET binding to human CLEC12A.

FIG. 56A-FIG. 56H are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET binding to human NKG2D.

FIG. 57A-FIG. 57H are a set of sensorgrams showing SPR profiles of hF3′-1602 TriNKET binding to recombinant cyno NKG2D (cNKG2D) extracellular domain (ECD).

FIG. 58A-FIG. 58B are line graphs showing target binding as determined by SPR for occupancy of one TriNKET arm on binding of the other TriNKET arm.

FIG. 59A-FIG. 59B are line graphs showing target binding as determined by SPR for synergy of NKG2D TriNKET (FIG. 59A) arm and CD16a (FIG. 59B).

FIG. 60A-FIG. 60E show that no non-specific binding was observed in any of the unrelated recombinant targets (FIG. 60B, FIG. 60C), whereas positive control (CLEC12A) showed binding of 50 RU at 100 nM concentration (FIG. 60A). Specificity was also tested against proteins expressed on the surface of the Ba/F3 cell line. No non-nonspecific binding of F3′-1602 at a concentration of 333 nM to the Ba/F3 parental cell line was observed (FIG. 60E), whereas Ba/F3 cell lines engineered to express CLEC12A, used as a positive control, showed a significant shift by flow cytometry (FIG. 60D).

FIG. 61A-FIG. 61C shows flow cytometry based polyspecificity reagent (PSR) analysis of Rituximab (FIG. 61A), Trastuzumab (FIG. 61B), and F3′-1602 TriNKET (FIG. 61C).

FIG. 62 are bar graphs showing relative binding (Z score) of F3′-1602 TriNKET at 1 to human CLEC12A in comparison to the entire human proteome microarray

FIG. 63A-FIG. 63C show models of hydrophobicity patches in hF3′-1602 TriNKET CLEC12A-binding arm.

FIG. 64A-FIG. 64E are bar graphs based on models of CDR-length (FIG. 64A), surface hydrophobicity (FIG. 64B), and surface charge (FIG. 64C-FIG. 64E) in hF3′-1602 TriNKET CLEC12A-binding arm.

FIG. 65A-FIG. 65C show models of hydrophobicity patches in hF3′-1602 TriNKET NKG2D-binding arm.

FIG. 66A-FIG. 66E are bar graphs based on models of CDR-length (FIG. 66A), surface hydrophobicity (FIG. 66B), and surface charge (FIG. 66C-FIG. 66E) in hF3′-1602 TriNKET NKG2D-binding arm.

FIG. 67 is a line graph showing Hydrophobic Interactions Chromatography (HIC) analysis of hF3′-1602 TriNKET.

FIG. 68 is a line graph showing pI as determined by cIEF for hF3′-1602 TriNKET.

FIG. 69A-FIG. 69I are a set of flow cytometry plots demonstrating sequence liability-remediated variants of hF3′-1602 binding to hCLEC12A.

FIG. 70 are line graphs showing NK cell-mediated lysis of CLEC12A-expressing cancer cell line PL21 in the presence of hF3′-1602 TriNKET and liability-remediated variants.

FIG. 71A-FIG. 71B are a set of sensorgrams showing the binding of hCLEC12A at three-fold serial dilution (300 nM to 0.41 nM) to F3′-1602 in the HST formulation. The black overlays represent the 1:1 kinetic fit of the raw data.

FIG. 72A-FIG. 72D is a set of graphs showing the binding of F3′-1602 in the HST formulation to hNKG2D. FIG. 72A and FIG. 72B represent the raw data of F3′-1602 injected over captured mFc-hNKG2D (9.77 nM to 5 μM). The black overlays represent the 1:1 kinetic fit of the raw data. FIG. 72C and FIG. 72D represent steady state response measurements plotted against the corresponding analyte concentrations. The vertical lines represent the steady state K_Dvalue.

FIG. 73A-FIG. 73B are a set of sensorgrams showing the binding of F3′-1602 in the HST formulation to hCD16a V158.

FIG. 74 is a graph showing the percentage lysis of PL21 cells by NKG2D- and CD16a-expressing KHYG-1-CD16a cells in the presence of stressed and control samples of F3′-1602 in the HST formulation.

FIG. 75A-FIG. 75C are a set of extracted ion chromatograms of CDRH3 peptide in control (FIG. 75A) and stressed F3′-1602 (FIG. 75B and FIG. 75C).

FIG. 76A-FIG. 76B are a set of sensorgrams showing the binding of hCLEC12A to F3′-1602 in the phosphate/citrate/NaCl formulation. The black overlays represent the 1:1 kinetic fit of the raw data.

FIG. 77A-FIG. 77D are a set of graphs showing the binding of F3′-1602 in the phosphate/citrate/NaCl formulation to hNKG2D. FIGS. 77A and 77B represent the raw data of F3′-1602 injected over captured mFc-hNKG2D (9.77 nM to 5 μM). The black overlays represent the 1:1 kinetic fit of the raw data. FIGS. 77C and 77D represent steady state response measurements plotted against the corresponding analyte concentrations. The vertical lines represent the steady state K_Dvalue.

FIG. 78A-FIG. 78B are a set of sensorgrams showing the binding of F3′-1602 in the phosphate/citrate/NaCl formulation to hCD16a V158.

FIG. 79 is a graph showing the percentage lysis of PL21 cells by NKG2D- and CD16a-expressing KHYG-1-CD16a cells in the presence of stressed and control samples of F3′-1602 in the phosphate/citrate/NaCl formulation.

FIG. 80A-FIG. 80B are a set of sensorgrams showing the binding of hCLEC12A at three-fold serial dilution (300 nM to 0.41 nM) to F3′-1602 after incubation in the CST formulation at 40° C. for four weeks. The black overlays represent the 1:1 kinetic fit of the raw data.

FIG. 81A-FIG. 81C are a set of sensorgrams showing the binding of hCLEC12A at three-fold serial dilution (300 nM to 0.41 nM) to F3′-1602 after incubation in the CST formulation at 2-8° C. and at 25° C. for four weeks. The black overlays represent the 1:1 kinetic fit of the raw data.

FIG. 82A-FIG. 82D are a set of graphs showing the binding of F3′-1602 in the CST formulation to hNKG2D. FIGS. 82A and 82B represent the raw data of F3′-1602 injected over captured mFc-hNKG2D (9.77 nM to 5 μM). The black overlays represent the 1:1 kinetic fit of the raw data. FIGS. 82C and 82D represent steady state response measurements plotted against the corresponding analyte concentrations. The vertical lines represent the steady state K_Dvalue.

FIG. 83A-FIG. 83B are a set of sensorgrams showing the binding of F3′-1602 in the CST formulation to hCD16a V158.

FIG. 84 is a graph showing the percentage lysis of PL21 cells by NKG2D- and CD16a-expressing KHYG-1-CD16a cells in the presence of stressed and control samples of F3′-1602 in the CST formulation.

FIG. 85 is a cIEF profile of F3′-1602 after incubation under pH 5.0 stress.

FIG. 86 is a cIEF profile of F3′-1602 after incubation under pH 8.0 stress.

FIG. 87A-FIG. 87B are a set of sensorgrams showing the binding of hCLEC12A at three-fold serial dilution (300 nM to 0.41 nM) to F3′-1602 after pH 5 stress. The black overlays represent the 1:1 kinetic fit of the raw data.

FIG. 88A-FIG. 88B are a set of sensorgrams showing the binding of hCLEC12A at three-fold serial dilution (300 nM to 0.41 nM) to F3′-1602 after pH 8 stress. The black overlays represent the 1:1 kinetic fit of the raw data.

FIG. 89A-FIG. 89D are a set of graphs showing the binding of F3′-1602 to hNKG2D after pH 5 stress. FIG. 89A and FIG. 89B represent the raw data of F3′-1602 injected over captured mFc-hNKG2D (9.77 nM to 5 μM). The black overlays represent the 1:1 kinetic fit of the raw data. FIGS. 89C and 89D represent steady state response measurements plotted against the corresponding analyte concentrations. The vertical lines represent the steady state K_Dvalue.

FIG. 90A-FIG. 90D are a set of graphs showing the binding of F3′-1602 to hNKG2D after pH 8 stress. FIGS. 90A and 90C represent the raw data of F3′-1602 injected over captured mFc-hNKG2D (9.77 nM to 5 μM). The black overlays represent the 1:1 kinetic fit of the raw data. FIGS. 90B and 90D represent steady state response measurements plotted against the corresponding analyte concentrations. The vertical lines represent the steady state K_Dvalue.

FIG. 91 is a graph showing the percentage lysis of PL21 cells by NKG2D- and CD16a-expressing KHYG-1-CD16a cells in the presence of F3′-1602 after pH 5 stress.

FIG. 92 is a graph showing the percentage lysis of PL21 cells by NKG2D- and CD16a-expressing KHYG-1-CD16a cells in the presence of F3′-1602 after pH 8 stress.

FIG. 93A-FIG. 93B are a set of sensorgrams showing the binding of hCLEC12A to F3′-1602 after forced oxidation stress. The black overlays represent the 1:1 kinetic fit of the raw data.

FIG. 94A-FIG. 94D are a set of graphs showing the binding of F3′-1602 to hNKG2D after forced oxidation stress. FIG. 94A and FIG. 94C represent the raw data of F3′-1602 injected over captured mFc-hNKG2D (9.77 nM to 5 μM). The black overlays represent the 1:1 kinetic fit of the raw data. FIG. 94B and FIG. 94D represent steady state response measurements plotted against the corresponding analyte concentrations. The vertical lines represent the steady state K_Dvalue.

FIG. 95 is a graph showing the percentage lysis of PL21 cells by NKG2D- and CD16a-expressing KHYG-1-CD16a cells in the presence of F3′-1602 after forced oxidation stress.

FIG. 96A-FIG. 96B are a set of sensorgrams showing the binding of hCLEC12A to F3′-1602 after low pH hold. The black overlays represent the 1:1 kinetic fit of the raw data.

FIG. 97A-FIG. 97D are a set of graphs showing the binding of F3′-1602 to hNKG2D after low pH hold. FIG. 97A and FIG. 97C represent the raw data of F3′-1602 injected over captured mFc-hNKG2D (9.77 nM to 5 μM). The black overlays represent the 1:1 kinetic fit of the raw data. FIG. 97B and FIG. 97D represent steady state response measurements plotted against the corresponding analyte concentrations. The vertical lines represent the steady state K_Dvalue.

FIG. 98A-FIG. 98B are a set of sensorgrams showing the binding of F3′-1602 to hCD16a V158 after low pH hold.

FIG. 99 is a graph showing the percentage lysis of PL21 cells by NKG2D- and CD16a-expressing KHYG-1-CD16a cells in the presence of F3′-1602 after low pH hold.

FIG. 100 is a graph showing the solubility of F3′-1602 in 10 mM sodium acetate buffer pH 5.0 in PEG-6000.

FIG. 101 is a graph showing the percentage of F3′-1602 monomer content as a function of concentration.

FIG. 102A-FIG. 102B are graphs showing the percentage lysis of PL-21 cells by NKG2D- and CD16a-expressing KHYG-1-CD16a cells in the presence of F3′-format TriNKET F3′-1602 (FIG. 102A) and F3-format TriNKET AB0010 (FIG. 102B) after thermal stress conditions.

FIG. 103A-FIG. 103C shows an illustration of a yeast display affinity maturation library construction. FIG. 103A: Substitution of pET1596 scFv CDRH3 with BamHI restriction enzyme site. FIG. 103B: Insertion of CDRH3 mutated library oligos at the previously introduced BamHI restriction site. FIG. 103C: Electroporation of BamHI linearized plasmid and library oligos into yeast followed by homologous recombination in yeast.

FIG. 104A-FIG. 104H show the discovery process from mouse immunization to AB0089. FIG. 104A-FIG. 104D represent results of CLEC12A binding by surface plasmon resonance (SPR). FIG. 104E-FIG. 104H represent binding to CLEC12A+ PL-21 AML cancer cells. Binding affinities and EC50 for each variant are indicated. BM, backmutation.

FIG. 105A-FIG. 105E shows FACS analysis of CDRH3 yeast library 17. First round of selection of affinity maturating yeast display library sorted at 1 nM CLEC12A-His (FIG. 105A). Second round of selection sorted at 1 nM CLEC12A-His (FIG. 105B). Third round of selection sorted at 0.1 nM CLEC12A-His (FIG. 105C). Binding of library 17-R3 displayed on yeast cells to 1 nM CLEC12A (FIG. 105D). Binding of original clone pET1596 displayed on yeast cells at 10 nM CLEC12A (FIG. 105E).

FIG. 106A-FIG. 106D shows FACS analysis of CDRH2 yeast library 30. Starting yeast display library which was generated by randomizing CDRH2 built on the background of the best clone identified from library 17 (GDYGDSLDY (SEQ ID NO:322) CDRH3) (FIG. 106A). Sorting was performed at 1 nM CLEC12A-His for both rounds. Flow cytometry analysis after round 1 (FIG. 106B). and round 2 (FIG. 106C) selection with CLEC12A-His, respectively. Binding of pET1596 displayed on yeast cells at 10 nM CLEC12A-His (FIG. 106D).

FIG. 107A-FIG. 107D is a comparison of the improved affinity clone (FIGS. 107A and 107D) identified from Library 30 to pET1596 (FIGS. 107B and 107D) and AB0237 (FIGS. 107C and 107D).

FIG. 108 is a schematic representation of AB0089. Heterodimerization mutations and an engineered interchain disulfide bridge ensure efficient chain pairing of the IgG1 Fc. The scFv is derived from the humanized and sequence liability corrected CLEC12A mAb 16B8.C8. The scFv is stabilized by an engineered disulfide bridge between VH and VL.

FIG. 109A-FIG. 109C is an X-ray crystal structure of EW-RVT Fc heterodimer. Superposition of the overall structure of the EW-RVT Fc heterodimer with the human wild-type (WT) IgG1 Fc homodimer (PDB ID: 3AVE) (FIG. 109A). Close-up view of W-VT interaction pair regions (FIG. 109B). Close-up view of the E-R interaction pair regions (FIG. 109C).

FIG. 110 shows the crystal structure of EW-RVTS-S heterodimeric Fc with a close-up view of the asymmetric disulfide bond formed between 349C(CH3A) and 354C(CH3B) residues.

FIG. 111A-FIG. 111C show a ribbon diagram model of the CLEC12A binding scFv in three different orientations (upper panel) and their corresponding surface charge distribution of the same orientation (lower panel). Three orientations are shown: both façades (FIG. 111A: front view; FIG. 111B: back view) and the antigen-engaging surface (FIG. 111C: top view).

FIG. 112A-FIG. 112B show the analyses of CDR length (FIG. 112A) and surface hydrophobicity patches (FIG. 112B) of the CLEC12A-targeting arm of AB0089. The arrow points to the line where CLEC12A targeting scFv of AB0089 stands in reference to advanced clinical stage mAbs. The two inner dotted lines indicate 2 SD (>95% of reference molecules within this region). The two outer most dotted lines indicate 3 SD (>99.7% of reference molecules within this region).

FIG. 113A-FIG. 113C show the analyses of patches containing positive (FIG. 113A) and negative charges (FIG. 113B), as well as the symmetry (FIG. 113C) of these charges around the CDR regions. Surface charge and charge symmetry analysis of CLEC12A targeting arm of AB0089. The arrow points to the line where CLEC12A targeting scFv of AB0089 stands in reference to advanced clinical stage mAbs. In each plot, there are two dashed lines—one closer and the other further to the solid line. The dashed line closer to the solid line indicates 2 SD (>95% of reference molecules within this region), whereas the dashed line further to the solid line indicates 3 SD (>99.7% of reference molecules within this region).

FIG. 114A-FIG. 114C show NKG2D-binding Fab with ribbon diagrams of three different orientations: both façades (FIG. 114A: front view; FIG. 114B: back view) and the antigen-engaging surface (FIG. 114C: top view).

FIG. 115A-FIG. 115B show the CDR length (FIG. 115A) and patches of surface hydrophobicity (FIG. 115B) analyses of the NKG2D-targeting arm of AB0089. The arrow points to the line where NKG2D targeting arm of AB0089 stands in reference to advanced clinical stage mAbs. The two inner dotted lines indicate 2 SD (>95% of reference molecules within this region). The two outer most dotted lines indicate 3 SD (>99.7% of reference molecules within this region).

FIG. 116A-FIG. 116C show the analyses of the CDR length (FIG. 116A) patches of surface hydrophobicity (FIG. 116B) and patches of negative charge (FIG. 116C) of NKG2D targeting arm of AB0089. The arrows indicate the position of the NKG2D targeting arm of AB0089 in reference to a database of 377 late-stage therapeutic antibodies. In each plot, there are two dashed lines—one closer and the other further to the solid line. The dashed line closer to the solid line indicates 2 SD (>95% of reference molecules within this region), whereas the dashed line further to the solid line indicates 3 SD (>99.7% of reference molecules within this region).

FIG. 117A-FIG. 117D show human acceptor frameworks used in AB0089 compared to other clinical stage therapeutic mAbs. Benchmarking of frameworks used in AB0089 against human germlines frequently used in clinical stage antibodies database containing 400+ mAbs from phase I+.

FIG. 118A-FIG. 118B show a cartoon representation of AB0089 indicating chain composition of the molecule (FIG. 118A) and the positions of chain H, S, and L of AB0089 on the Treg-adjusted EpiMatrix protein immunogenicity scale (FIG. 118B).

FIG. 119 shows a TriNKET purification process flow diagram.

FIG. 120 shows a Mab Select Protein A capture elution chromatogram.

FIG. 121 shows a SEC analysis of Protein A eluate, pH-adjusted and filtered for IEX loading.

FIG. 122 shows a SDS-PAGE analysis of AB0089 Protein A load, flow through, and eluate. Based on absorbance readings, 15 μl was loaded in the “Protein A load” and “Protein A FT” lanes, 2 μg was loaded in the “Protein A Eluate lane”. The major band at ˜120 kDa corresponds to AB0089.

FIG. 123 shows a Poros XS cation exchange chromatogram.

FIG. 124 shows a SDS-PAGE analysis of AB0089 CIEX load and elution fractions. Based on absorbance readings, lanes contain ˜2.0 μg of protein. The major band at ˜120 kDa is AB0089.

FIG. 125 shows reduced (“R”) and non-reduced (“NR”) SDS-PAGE of AB0089 Lot AB0089-002. The three chains (S, H, and L) of the molecule are labeled in the R lane. F3′ in the NR lane refers to AB0089.

FIG. 126 shows SEC analysis of purified AB0089 lot AB0089-002 and indicates 100% monomer as determined by integrated peak area.

FIG. 127 shows intact mass spectrometry analysis of AB0089 lot AB0089-002. LC-MS analysis of the intact AB0089 showed an observed mass (126,130.0 Da) that agrees well with the theoretical mass of 126,129.1 Da.

FIG. 128A-FIG. 128B show CE-SDS analysis of AB0089, R (FIG. 128A) and NR (FIG. 128B) CE-SDS analysis of AB0089.

FIG. 129 shows a charge profile of AB0089 by cIEF, indicating that AB0089 showed a major peak at a pI of 8.52±0.0

FIG. 130 shows a chromatogram of the hydrophobicity analysis of AB0089 by HIC, indicating that AB0089 is highly pure and homogenous.

FIG. 131A-FIG. 131C shows the thermal stability of AB0089 as assessed by DSC analysis in three different formulations: PBS pH 7.4 (FIG. 131A), HST pH 6.0 (FIG. 131B), and CST, pH 7.0 (FIG. 131C).

FIG. 132 shows a predicted disulfide map of AB0089.

FIG. 133A-FIG. 133B show the extracted ion chromatogram (XICs) for the engineered disulfide pair in the scFv (non-reduced and reduced; FIG. 133A) and the most intense charge state for that peptide pair (FIG. 133B).

FIG. 134A-FIG. 134B show identification of CH3-CH3 intermolecular Fc engineered disulfide.

FIG. 135A-FIG. 135D show binding of human CLEC12A to AB0089 tested by SPR.

FIG. 136A-FIG. 136B show that AB0089 (FIG. 136A) and tepoditamab (FIG. 136B) based molecule binding of CLEC12A conforms to a two-state model.

FIG. 137A-FIG. 137D show FACS assessment of AB0089 binding to isogenic cell lines expressing CLEC12A, and AML cancer cell lines (PL-21 and HL-60).

FIG. 138A-FIG. 138C show binding of AB0089 to a genetic variant CLEC12A (Q244) (FIG. 138B) compared to CLEC12A (K244) WT (FIG. 138A) and cyno CLEC12A (cCLEC12A) (FIG. 138C).

FIG. 139A-FIG. 139C show AB0089 binding to differentially glycosylated CLEC12A: hCLEC12A untreated (FIG. 139A), hCLEC12A de-sialylated (FIG. 139B) and hCLEC12A fully deglycosylated (FIG. 139C).

FIG. 140A-FIG. 140F show binding of AB0089 to human NKG2D tested by SPR. FIG. 140A-FIG. 140C sensorgrams represent raw data. Black overlays represent the 1:1 kinetic fit. FIG. 140D-FIG. 140F state affinity fit. Data from four independent Biacore channels are shown. The line represents the steady state KD value.

FIG. 141A-FIG. 141F show binding of AB0089 and trastuzumab to recombinant human CD16a (V158).

FIG. 142A-FIG. 142H show binding of AB0089 and trastuzumab to recombinant human CD64 (FcγRI).

FIG. 143A-FIG. 143G show binding of AB0089 and trastuzumab to cynomolgus CD64 (FcγRI).

FIG. 144A-FIG. 144P show binding of AB0089 and trastuzumab to human CD32a H131 (FcγRIIa H131).

FIG. 145A-FIG. 145P show binding of AB0089 and trastuzumab to human CD32a R131 (FcγRIIa R131).

FIG. 146A-FIG. 146P show binding of AB0089 and trastuzumab to recombinant human CD16a F158 (FcγRIIIa F158).

FIG. 147A-FIG. 147L show binding of AB0089 and trastuzumab to recombinant cynomolgus CD16 (FcγRIII).

FIG. 148A-FIG. 148P show binding of AB0089 and trastuzumab to recombinant human CD32b (FcγRII).

FIG. 149A-FIG. 149L show binding of AB0089 and trastuzumab to recombinant human CD16b (FcγRIIIb).

FIG. 150A-FIG. 150P show binding of AB0089 and trastuzumab to recombinant human FcRn.

FIG. 151A-FIG. 151L show binding of AB0089 and trastuzumab to recombinant cynomolgus FcRn.

FIG. 152 shows that simultaneous engagement of AB0089 to CD16a and NKG2D results in avid binding.

FIG. 153A-FIG. 153B show AB0089 simultaneous co-engagement of CLEC12A and NKG2D binding arms.

FIG. 154 is a summary of a comparison of the binding epitope of AB0089 in relation to AB0237, AB0192 and three CLEC12A binding reference TriNKETs by SPR. AB0089 interaction with hCLEC12A was blocked by AB0237, and two control molecules, cFAE-A49 CLL1-Merus and cFAE-A49 Tepoditamab, but not by Genentech-h6E7 based molecule, and illustrates that AB0089 binding epitope overlaps with Merus-CLL1 and tepoditamab.

FIG. 155A-FIG. 155B show potency of AB0089 in KHYG-1-CD16aV mediated cytotoxicity assay with HL60 and PL21 cells.

FIG. 156 shows that AB0089 potently enhances primary NK mediated cell lysis of cancer cells, outperforming parental mAb.

FIG. 157A-FIG. 157D shows a schematic representation of a PSR assay (FIG. 157A), and PSR binding AB0089 (FIG. 157D) in comparison with antibodies with known PSR binding (Rituximab; FIG. 157B) and an antibody with no known non-specificity (Trastuzumab) (FIG. 157C).

FIG. 158 shows the relative binding (Z score) of AB0089 at 1 μg/ml to human CLEC12A in comparison to the entire human proteome microarray. Plot depicting the relative binding Z-score (y-axis) vs. the top 24 identified binding partners of AB0089 (x-axis). hCLEC12a is plotted at position 1 with a Z-score of 150.61.

FIG. 159 shows SEC analysis of AB0089 after 4 weeks at 40° C. in HST, pH 6.0 compared to control sample, demonstrating that AB0089 exhibited high stability.

FIG. 160A-FIG. 160B shows CE-SDS analysis of AB0089 after 4 weeks at 40° C. in HST, pH 6.0, compared to control. Non-reduced (NR) (FIG. 160A) and reduced (R) (FIG. 160B).

FIG. 161A-FIG. 161F shows that AB0089 is stable after 4 weeks at 40° C. in HST, pH 6.0 with no effect on binding to hCLEC12A (FIG. 161A and FIG. 161B), hNKG2D (FIGS. 161C and 161D), or hCD16aV (FIG. 161E and FIG. 161F).

FIG. 162 shows AB0089 potency after 4 weeks at 40° C. in HST, pH 6.0.

FIG. 163 shows that AB0089 demonstrated high stability after 4 weeks of incubation in CST, pH 7.0, at 40° C., compared to control sample as measured by SEC analysis.

FIG. 164A-FIG. 164B show that no fragmentation of AB0089 was observed by CE-SDS after 4 weeks at 40° C. in CST, pH 7.0 compared to control sample. NR (FIG. 164A) and R (FIG. 164B).

FIG. 165A-FIG. 165F shows that AB0089 is stable after 4 weeks at 40° C. in CST, pH 7.0, with no effect on binding to hCLEC12A (FIG. 165A and FIG. 165B), hNKG2D (FIG. 165C and FIG. 165D), or hCD16aV (FIG. 165E and FIG. 165F).

FIG. 166 shows that long term thermal stress did not affect potency of AB0089 after 4 weeks at 40° C. in CST, pH 7.0 compared to control.

FIG. 167A-FIG. 167C are a set of extracted ion chromatograms of the tryptic peptide encompassing CDRH3 of CLEC12a targeting arm of AB0089, which showed no evidence of aspartic acid isomerization in the peptide encompassing the CLEC12A binding CDRH3 was observed, based on manual inspection of peak shape.

FIG. 168 is an SEC analysis showing that there is no difference in monomer content between oxidized AB0089 and control sample after forced oxidation.

FIG. 169A-FIG. 169B show that no fragmentation of AB0089 was observed by CE-SDS after forced oxidation compared to control sample. NR (FIG. 169A) and R (FIG. 169B).

FIG. 170A-FIG. 170F show that oxidative stress had no significant effect on the active protein content or on the affinities for hCLEC12A (FIG. 170A and 170B), hNKG2D (FIGS. 170C and 170D), and hCD16Av (FIGS. 170E and 170F).

FIG. 171 shows that AB0089 is stable in forced oxidation.

FIG. 172 is an SEC analysis showing that AB0089 is stable under high pH stress (pH 8.0, 40° C., 2 weeks).

FIG. 173A-FIG. 173B is an CE-SDS analysis showing that AB0089 exhibited low levels of degradation AB0089 after long term high pH stress (pH 8, 40° C., 2 weeks). NR (FIG. 173A) and R (FIG. 173B).

FIG. 174 is a cIEF showing that AB0089 exhibited an acidic shift after long term high pH exposure (pH 8, 40° C., 2 weeks).

FIG. 175A-FIG. 175F show that there was no meaningful difference in the amount of active protein or in the affinities for hCLEC12A (FIGS. 175A and 175B), hNKG2D (FIGS. 175C and 175D), or hCD16aV (FIGS. 175E and 175F) between stressed and control samples after long term high pH 8.0 stress (pH 8, 40° C., 2 weeks).

FIG. 176 shows that the potency of AB0089 was reduced approximately 2-fold after long term high pH stress (pH 8, 40° C., 2 weeks).

FIG. 177 is an SEC analysis showing that AB0089 is highly stable under low pH stress (pH 5.0, 40° C., 2 weeks).

FIG. 178A-FIG. 178B is a CE-SDS analysis of AB0089 after long term low pH stress (pH 5.0 40° C., 2 weeks) showing that low levels of AB0089 degradation were detected by non-reduced (FIG. 178A) and reduced (FIG. 178B) CE-SDS.

FIG. 179 is a cIEF analysis showing that AB0089 exposed to low pH stress shows slight acidic shift in charge profile.

FIG. 180A-FIG. 180F shows that there was no meaningful difference in active protein content or in AB0089 affinities for hCLEC12A (FIGS. 180A and 180B), hNKG2D (FIGS. 180C and 180D), or hCD16aV (FIGS. 180E and 180F) after long term low pH stress (pH 5.0, 40° C., 2 weeks).

FIG. 181 shows that the potency of AB0089 remained similar to potency of the control sample after long term low pH stress (pH 5.0 40° C., 2 weeks).

FIGS. 182A and 182B show an extracted ion chromatogram of the tryptic peptide encompassing CDRH3 of CLEC12A targeting arm of AB0089, which showed no evidence of aspartic acid isomerization in the peptide encompassing the CLEC12A binding CDRH3 after long term exposure to low pH (pH 5.0), based on manual inspection of peak shape.

FIG. 183 is an SEC analysis showing that AB0089 is stable after six cycles of freeze/thaw.

FIG. 184A-FIG. 184B show a CE-SDS analysis of AB0089 after six freeze/thaw cycles showing that no loss of purity was detected by reduced (FIG. 184A) or non-reduced CE-SDS (FIG. 184B).

FIG. 185A-FIG. 185B show an SEC analysis showing that AB0089 is stable after agitation stress (300 rpm at room temperature in a deep well plate for 3 days).

FIG. 186A-FIG. 186B show an SEC analysis showing that AB0089 was able to be concentrated to >175 mg/ml with minimal loss in percent monomer.

FIG. 187 is a graph showing that AB0089 is amendable to concentration (up to 178 mg/ml) without loss of monomer.

FIG. 188A-FIG. 188B show an SEC analysis of Protein A eluate pre- and post-low pH hold showing that AB0089 is highly stable during low pH hold/viral inactivation conditions used in biologics manufacturing.

FIG. 189 shows that AB0089 charge profile remains unaltered after low pH hold.

FIG. 190A-FIG. 190F show that the affinities of AB0089 for hCLEC12a, hNKG2D, and CD16aV were not affected by low pH hold.

FIG. 191 shows that the potency of AB0089 was unaltered after the low pH hold step.

FIG. 192A-FIG. 192B show an SEC analysis showing that there was no change in profile of percent monomer was observed after the low pH hold.

FIG. 193 shows that low pH hold did not have a significant effect on the charge profile of AB0089.

FIG. 194 shows that AB0089 strongly enhances human NK cytotoxicity of PL21 AML cells. Purified human NK cells were rested overnight and, the following day, were co-cultured with labeled PL21 target cells for DELFIA assay. Dose-titrations of CLEC12A TriNKET AB0089 or parental mAb AB0305 were prepared starting at 10 nM and added to the co-cultures. Assays were performed with a total of six healthy NK cell donors; shown is a representative result. Specific lysis was plotted against test article concentration, and data were fit to a 4-parameter non-linear regression model to generate potency values. Data points represent mean±standard deviation.

FIG. 195 shows that AB0192 strongly enhances human NK cytotoxicity of HL60 AML cells. Purified human NK cells were rested overnight and, the following day, were co-cultured with labeled HL60 target cells for DELFIA assay. Dose-titrations of CLEC12A TriNKET AB0089 or parental mAb AB0305 were prepared starting at 10 nM and added to the co-cultures. Assays were performed with a total of four healthy NK cell donors; shown is a representative result. Specific lysis was plotted against test article concentration, and data were fit to a 4-parameter non-linear regression model to generate potency values. Data points represent mean±standard deviation.

FIG. 196 shows that AB0192 strongly enhances human NK cytotoxicity of PL21 AML cells. Purified human NK cells were rested overnight and, the following day, were co-cultured with labeled PL21 target cells for DELFIA assay. Dose-titrations of CLEC12A TriNKET AB0192 or parental mAb AB0306 were prepared starting at 10 nM and added to the co-cultures. Assays were performed with a total of six healthy NK cell donors; shown is a representative result. Specific lysis was plotted against test article concentration, and data were fit to a 4-parameter non-linear regression model to generate potency values. Data points represent mean±standard deviation.

FIG. 197 shows that AB0192 strongly enhances human NK cytotoxicity of HL60 AML cells. Purified human NK cells were rested overnight and, the following day, were co-cultured with labeled HL60 target cells for DELFIA assay. Dose-titrations of CLEC12A TriNKET AB0192 or parental mAb AB0306 were prepared starting at 10 nM and added to the co-cultures. Assays were performed with a total of four healthy NK cell donors; shown is a representative result. Specific lysis was plotted against test article concentration, and data were fit to a 4-parameter non-linear regression model to generate potency values. Data points represent mean±standard deviation.

FIG. 198 shows that AB0089 activity is not influenced by the presence of soluble MIC-A. Purified human NK cells were rested overnight and, the following day, were set up in co-culture wells with DELFIA labeled PL21 target cells. Dose-titrations of AB0089 and AB0305 were prepared, and soluble MIC-A-Fc was added to 20 ng/mL into a titration series of each test article. Dose-titrations were fit to a 4-parameter non-linear regression model. Data points represent mean±standard deviation.

FIG. 199 shows that AB0089 induces IFNγ production and degranulation by human NK cells against PL21 AML cells. Frozen PBMCs were thawed and co-cultured with PL21 target cells in the presence of a dose-titration of AB0089, AB0305 or F3′-palivizumab. After incubation, cells were prepared for FACS analysis of CD107a degranulation and intracellular accumulation of IFNγ. The figure shown is representative of the three PBMC donors tested. The percentage of IFNγ+CD107+ NK cells induced was plotted against concentration, and data were fit to a 4-parameter non-linear regression model to generate potency values. Data points represent mean±standard deviation.

FIG. 200 shows that AB0089 induces IFNγ production and degranulation by human NK cells against HL60 AML cells. Frozen PBMCs were thawed and co-cultured with HL60 target cells in the presence of a dose-titration of AB0089, AB0305 or F3′-palivizumab. After incubation, cells were prepared for FACS analysis of CD107a degranulation and intracellular accumulation of IFNγ. The figure shown is representative of the three PBMC donors tested. The percentage of IFNγ+CD107+ NK cells induced was plotted against concentration, and data were fit to a 4-parameter non-linear regression model to generate potency values. Data points represent mean±standard deviation.

FIG. 201 shows that CLEC12A TriNKETs induce M0-macrophage ADCP activity against PL21 AML cells. In vitro generated M0 macrophages were labeled with Cell Trace Violet and co-cultured with Cell Trace CFSE-labeled PL21 target cells at an 8:1 E:T ratio. After two-hours, cells were prepared for flow cytometry analysis. Double-positive Cell Trace Violet/CFSE cells were considered phagocytic events. Assays were conducted with macrophages derived from three independent donors with the figure shown being a representative result. Percentage phagocytosis of target cells was plotted against test article concentration, and data were fit to a 4-parameter non-linear regression model to generate potency values. Data points represent mean±standard deviation.

FIG. 202 shows that AB0089 does not stimulate CDC against PL21 AML cells. AB0089 was incubated with BATDA-labeled PL21 cells in cell culture media containing 5% human serum for 60 minutes. Human IgG1 isotype control served as a negative control, while anti-MHC class I clone W6/32 was used as a positive control. Specific lysis of target cells was plotted against concentration, and data were fit to a 4-parameter non-linear regression model to generate potency values. Data points represent mean±standard deviation.

FIG. 203A-FIG. 203C show binding of CLEC12A-TriNKETs to cell lines expressing CLEC12A. Dose-response binding of CLEC12A TriNKETs and their parental mAbs were analyzed on (FIG. 203A) Ba/F3 cells overexpressing human CLEC12A. The human cancer cell lines (FIG. 203B) HL60 and (FIG. 203C) PL21 with endogenous expression of CLEC12A were used to corroborate results with the transfected cell line. Figures shown are representative of results from three independent experiments, which are summarized in Table 201.

FIG. 204A-FIG. 204B show CLEC12A surface retention on AML cell lines exposed to CLEC12A TriNKETs. CLEC12A surface retention on (FIG. 204A) HL60 and (FIG. 204B) PL21 cells was measured after exposure to test articles at 40 nM for two hours at 37° C. Test articles were used at saturation concentrations and detected using a human IgG Fc-specific secondary antibody. Staining intensity after two hours at 37° C. was normalized to staining intensity of the same test article on cells incubated for two hours on ice to derive the percentage of CLEC12A retained on cell surfaces. Data bars represent the average of duplicate wells and error bars represent standard deviation. Figures shown are representative of results from three independent experiments, which are summarized in Table 202.

FIG. 205 shows AB0089 binding normalized to AB0237 after two hours at 37° C. AB0089 and AB0237 binding were compared on HL60 cells after a two-hour incubation on ice or at 37° C. Histograms represent the following: (1st from top to bottom) AB0237 two-hours 37° C., (2nd from top to bottom) AB0237 two-hour on ice, (3rd from top to bottom) AB0089 two-hours at 37° C., (4th from top to bottom) AB0089 two-hours on ice. Binding of all test articles was detected with an anti-human IgG Fc fluorophore-conjugated antibody.

FIG. 206A-FIG. 206J show AB0089 binds CLEC12A+ blasts across 10 primary AML patient samples. Primary AML bone marrow and PBMC samples were obtained from commercial vendors, see Table 3. Samples were prepared for flow cytometry analysis according to the antibody panel in Table 1. Histograms represent staining on AML blasts as follows: 4th from top to bottom=hIgG1 isotype control, 3rd from top to bottom=AB0305-CF568, 2nd from top to bottom=AB0089-CF568, 1st from top to bottom=anti-CLEC12 50C1-CF568.

FIG. 207A-FIG. 207J show AB0089 binds CLEC12A+ CD34+CD38− cells across 10 primary AML patient samples. Primary AML bone marrow and PBMC samples were obtained from commercial vendors, see Table 200. Samples were prepared for flow cytometry analysis according to the antibody panel in Table 298. Samples were further gated on CD34+CD38− subpopulation to enrich for leukemic stem cells. Histograms represent staining on CD34+CD38− blasts as follows: 4th from top to bottom=hIgG1 isotype control, 3rd from top to bottom=AB0305-CF568, 2nd from top to bottom=AB0089-CF568, 1st from top to bottom=anti-CLEC12 50C1-CF568.

FIG. 208A-FIG. 208F show the path from immunization to AB0192. Upper panel represents results of CLEC12A binding by SPR. Lower panel shows binding to CLEC12A+ PL-21 AML cancer cells.

FIG. 209 shows a schematic representation of AB0192. Proprietary heterodimerization mutations and an engineered interchain disulfide bridge ensure efficient chain pairing of the IgG1 Fc. The scFv is derived from the humanized and sequence liability corrected CLEC12A mAb 9F11.B7. The scFv is stabilized by an engineered disulfide bridge between VH and VL.

FIG. 210 shows the amino acid sequences of the three polypeptide chains that make up AB0192. Chain S: anti-CLEC12A scFv-CH2-CH3 (humanized sequence, CDRs are bold not underlined, back mutations bold and underlined, engineered scFv disulfide in italic and bold, and Fc mutations for heterodimerization and the engineered CH3 disulfide are underlined not bold). Chain H: anti-NKGD VH-CH1-CH2-CH3 (fully human, CDRs are bold not underlined, Fc mutations for heterodimerization and the engineered CH3 disulfide are underlined not bold). Chain L: anti-NKGD VL-CL (fully human, CDRs are bold not underlined).

FIG. 211A-FIG. 211D show human acceptor frameworks used in AB0192 compared to other clinical stage therapeutic mAbs. Benchmarking of frameworks used in AB0192 against human germlines frequently used in clinical stage antibodies database containing 400+ mAbs from phase I+.

FIG. 212A-FIG. 212C show a ribbon diagram model of the CLEC12A binding scFv in three different orientations (upper panel) and their corresponding surface charge distribution of the same orientation (lower panel). The charge distribution of anti-FIG. CLEC12A scFv is polarized (“top view”, lower panel), with negatively-charged residues populated predominately within CDRH3 and CDRL2. Three orientations are shown: both façades (FIG. 212A: front view; FIG. 212B: back view) and the antigen-engaging surface (FIG. 212C: top view).

FIG. 213A-FIG. 213B show the analysis of CDR length and surface hydrophobicity patches of the CLEC12A-targeting arm of AB0192. The arrow indicates where CLEC12A targeting scFv of AB0192 stands in reference to advanced clinical stage mAbs. The inner dotted lines indicate 2 SD (>95% of reference molecules within this region). The outermost dotted lines indicate 3 SD (>99.7% of reference molecules within this region).

FIG. 214A-FIG. 214C show the analysis of patches containing positive and negative charges, as well as the symmetry of these charges around the CDR regions. The arrow indicates where CLEC12A targeting scFv of AB0192 stands in reference to advanced clinical stage mAbs. In each plot, there are two dashed lines—one closer and the other further to the solid line. The dashed line closer to the solid line indicates 2 SD (>95% of reference molecules within this region), whereas the dashed line further to the solid line indicates 3 SD (>99.7% of reference molecules within this region).

FIG. 215A-FIG. 215C show a ribbon diagrams with three different orientations of the NKG2D-binding Fab arm (upper panel) and their corresponding surface charge distribution of the same orientation (lower panel). Three orientations are shown: both façades (FIG. 215A: front view; FIG. 215B: back view) and the antigen-engaging surface (FIG. 215C: top view).

FIG. 216A-FIG. 216B show the CDR length (FIG. 216A) and patches of surface hydrophobicity (FIG. 216B) analyses of NKG2D-targeting arm of AB0192. The arrow indicates where NKG2D targeting Fab of AB0192 stands in reference to advanced clinical stage mAbs. The inner dotted lines indicate 2 SD (>95% of reference molecules within this region). The outermost dotted lines indicate 3 SD (>99.7% of reference molecules within this region).

FIG. 217A-FIG. 217C shows the analyses of patches containing positive (FIG. 217A) and negative charges (FIG. 217B), as well as the symmetry of these charges around the CDR regions (FIG. 217C). The arrows indicate the position of the NKG2D targeting Fab of AB0192 stands in reference to a database of 377 late-stage therapeutic antibodies. In each plot, there are two dashed lines—one closer and the other further to the solid line. The dashed line closer to the solid line indicates 2 SD (>95% of reference molecules within this region), whereas the dashed line further to the solid line indicates 3 SD (>99.7% of reference molecules within this region).

FIG. 218A-FIG. 218B show the position of AB0192 chains on the Treg-adjusted EpiMatrix protein immunogenicity scale. FIG. 218A shows a cartoon representation of AB0192; FIG. 218B shows positions of chain H, S, and L of AB0192 on Treg-Adjusted Immunogenicity Protein Scale.

FIG. 219A-FIG. 219D show a schematic representation of the AB0192 2 step platform purification. FIG. 219A and FIG. 219B: SDS-PAGE analysis of AB0192 before and after first step of purification process (Protein A capture). FIG. 219C: SEC analysis of AB0192 after protein A step and after cIEX step. FIG. 219D: Reduced and non-reduced SDS-PAGE of purified AB0192.

FIG. 220 shows the intact mass analysis of AB0192 (lot AB0192-005) using LC-MS analysis. AB0192 theoretical mass=126,429.5 Da. AB0192 Lot AB0192-005 observed mass=126,429.9 Da.

FIG. 221 shows a SEC (Superdex 200) analysis of purified AB0192 (lot AB0192-005), which showed >99% monomer as determined by integrated peak area.

FIG. 222A-FIG. 222B show the purity of AB0192 (lot AB0192-005) by non-reduced conditions (“NR”; FIG. 222A) and reduced conditions (“R”; FIG. 222B) by CE-SDS.

FIG. 223 shows the cIEF profile of AB0192 (lot AB0192-005).

FIG. 224 shows the hydrophobicity analysis of AB0192 by HIC.

FIG. 225A-FIG. 225C show DSC analysis of AB0192 in three different buffer systems. PBS pH 7.4 (FIG. 225A), HST, pH 6.0 (FIG. 225B) and CST, pH 7.0 (FIG. 225C).

FIG. 226 shows the predicted disulfide map for AB0192.

FIG. 227A-FIG. 227B show the extracted ion chromatogram (XICs) for the engineered disulfide pair in the scFv (non-reduced and reduced; FIG. 227A) and the most intense charge state for that peptide pair (FIG. 227B).

FIG. 228A-FIG. 228B show the existence of the CH3-CH3 intermolecular Fc engineered disulfide.

FIG. 229A-FIG. 229C show binding of AB0192 to human CLEC12A tested by SPR.

FIG. 230A-FIG. 230C show binding of AB0192 to a genetic variant of human CLEC12A (Q244) (FIG. 230B) compared to CLEC12A (K244) WT (FIG. 230A) and cyno CLEC12A tested by SPR (FIG. 230C) .

FIG. 231A-FIG. 231C show binding of AB0192 to differentially glycosylated CLEC12A tested by SPR.

FIG. 232A-FIG. 232D show binding of AB0192 to CLEC12A+ isogenic and HL60 and PL21 AML cells.

FIG. 233A-FIG. 233F show binding of AB0192 to human NKG2D tested by SPR.

FIG. 234A-FIG. 234F show binding of AB0192 to hCD16a V158 allele.

FIG. 235 shows simultaneous binding of AB0192 to CD16a and NKG2D results in avid binding. AB0192 injected over NKG2D (middle line, as indicated on right side of graph), CD16a (bottom line, as indicated on right side of graph) or mixed, CD16a and NKG2D, (top line, as indicated on right side of graph) surfaces. Note that the surface has not been optimized for a maximum avidity effect.

FIG. 236A-FIG. 236B show sequential saturation of captured AB0192 with hCLEC12A and hNKG2D. Targets were sequentially injected over AB0192 captured on a Biacore chip. FIG. 236A represents binding of hCLEC12A (100 nM) followed by binding of hNKG2D (7 μM) to captured AB0192. FIG. 236B depicts the reverse order of target binding (human NKG2D binding followed by hCLEC12A). Relative binding stoichiometries (relative to binding of each target to unoccupied captured AB0192) are shown in Table 218.

FIG. 237A-FIG. 237H show binding of AB0192 and trastuzumab to recombinant human CD64 (FcγRI). A difference in apparent maximum binding responses between the molecules is due to their difference in molecular weight.

FIG. 238A-FIG. 238H show binding of AB0192 and trastuzumab to recombinant human CD64 (FcγRI). A difference in apparent maximum binding responses between the molecules is due to their difference in molecular weight

FIG. 239A-FIG. 239P show binding of AB0192 and trastuzumab to recombinant human CD32a (FcγRIIa) H131 allele. FIG. 239A-FIG. 239D and FIG. 239I-FIG. 239H and FIG. 239M-FIG. 239P for each molecule represent raw binding sensorgrams while FIG. 239E-FIG. 239H and FIG. 239M-FIG. 239P represent fitting to a steady state affinity model. A difference in apparent maximum binding responses between the molecules is due to a difference in molecular weight.

FIG. 240A-FIG. 240P show binding of AB0192 and trastuzumab to recombinant human CD32a (FcγRIIa) R131 allele. Top panel for each molecule represent raw binding sensorgrams while the bottom panels represent fitting to a steady state affinity model. A difference in apparent maximum binding responses between the molecules is due to a difference in molecular weight.

FIG. 241A-FIG. 241P show binding of AB0192 and trastuzumab to recombinant human CD16a (FcγRIIIa) F158 allele. FIG. 241A-FIGS. 241D and 241I-FIG. 241L for each molecule represent raw binding sensorgrams while FIG. 241E-FIG. 241H and FIG. 241M-FIG. 241P represent fitting to a steady state affinity model. A difference in apparent maximum binding responses between the molecules is due to a difference in molecular weight.

FIG. 242A-FIG. 242L show binding of AB0192 and trastuzumab to recombinant cynomolgus CD16 (FcγRIII). FIGS. 242A, 242B, 242E, 242G, 242H, and 242K for each molecule represent raw binding sensorgrams while FIG. 242C, FIG. 242D, FIG. 242F, FIG. 242I, FIG. 242J, and FIG. 242L represent fitting to a steady state affinity model. A difference in apparent maximum binding responses between the molecules is due to a difference in molecular weight.

FIG. 243A-FIG. 243P show binding of AB0192 and trastuzumab to recombinant human CD32b (FcγRIIb). FIG. 243A-FIG. 243D and FIG. 243I-FIG. 243L for each molecule represent raw binding sensorgrams while FIG. 243E-FIG. 243H and FIG. 243M-FIG. 243P represent fitting to a steady state affinity model. A difference in apparent maximum binding responses between the molecules is due to a difference in molecular weight.

FIG. 244A-FIG. 244L show binding of AB0192 and trastuzumab to recombinant human CD16b (FcγRIIIb). FIGS. 244A, 244B, 244E, 244G, 244H and 244K for each molecule represent raw binding sensorgrams while FIG. 244C, FIG. 244D, FIG. 244F, FIG. 244I, FIG. 244J, and FIG. 244L represent fitting to a steady state affinity model. A difference in apparent maximum binding responses between the molecules is due to a difference in molecular weight.

FIG. 245A-FIG. 245P show binding of AB0192 and trastuzumab to recombinant human FcRn. FIG. 245A-FIG. 245D and FIG. 245I-FIG. 245L for each molecule represent raw binding sensorgrams while FIG. 245E-FIG. 245H and FIG. 245M-FIG. 245P represent fitting to a steady state affinity model. A difference in apparent maximum binding responses between the molecules is due to a difference in molecular weight.

FIG. 246A-FIG. 246L show binding of AB0192 and trastuzumab to recombinant cynomolgus FcRn. FIGS. 246A, 246B, 246E, 246G, 246H, and 246K for each molecule represent raw binding sensorgrams while FIG. 242C, FIG. 242D, FIG. 242F, FIG. 242I, FIG. 242J, and FIG. 242L represent fitting to a steady state affinity model. A difference in apparent maximum binding responses between the molecules is due to a difference in molecular weight.

FIG. 247 shows Epitope binning of AB0192 against other CLEC12A binding TriNKETs.

FIG. 248 shows potency of AB0192 in primary NK mediated cytotoxicity assay compared to corresponding mAb control.

FIG. 249A and FIG. 249B show non-specificity analysis of AB0192 in PSR assay. FIG. 249A: schematic representation of a PSR assay; FIG. 249B: PSR binding AB0089 (FIG. 249D) in comparison with antibodies with known PSR binding (Ixekizumab; FIG. 249B) and an antibody with no known non-specificity (Trastuzumab; FIG. 249C).

FIG. 250 shows the relative Z score of AB0089 in HuProt™ human proteome microarray assay. Plot depicting the relative binding Z-score (y-axis) vs. the top 24 identified binding partners of AB0192 (x-axis). hCLEC12a is plotted at position 1 with a Z-score of 150.06.

FIG. 251A-FIG. 251B show an SEC analysis of AB0192 after 4 weeks at 40° C. in HST, pH6.0, compared to control.

FIG. 252A-FIG. 252B show a CE-SDS analysis of AB0192 after 4 weeks at 40° C. in HST, pH 6.0, compared to control. Non-reduced conditions are shown in FIG. 252A. Reduced conditions are shown in FIG. 252B.

FIG. 253A-FIG. 253F show that AB0192 is stable after 4 weeks 40° C. stress in HST, pH 6.0: no effect on CLEC12A, NKG2D or CD16a binding.

FIG. 254 shows that there was no difference in potency between the control and stressed samples of AB0192 after 4 weeks at 40° C. in HST, pH 6.0 in KHYG-1-CD16a mediated cytotoxicity assay.

FIG. 255A-FIG. 255B show minimal aggregation of AB0192 after 4 weeks at 40° C. in CST, pH 7.0, as measured by SEC analysis.

FIG. 256A-FIG. 256B show minimal fragmentation of AB0192 after 4 weeks at 40° C. in CST, pH 7.0 compared to control, as measured by CE-SDS analysis.

FIG. 257A-FIG. 257F show that AB0192 is stable after 4 weeks 40° C. stress in CST, pH 7.0: no effect on CLEC12A, NKG2D or CD16a binding.

FIG. 258 shows that there was no difference in potency between the control and stressed samples of AB0192 after 4 weeks at 40° C. in CST, pH 7.0, in KHYG-1-CD16a cytotoxicity assay.

FIG. 259A-FIG. 259B show an SEC analysis of AB0192 after forced oxidation stress compared to control, and indicated no significant decrease in monomer content.

FIG. 260A-FIG. 260B show a CE-SDS analysis of AB0192 after forced oxidation compared to control, and indicated that purity remained high.

FIG. 261A-FIG. 261F show that AB0192 is stable after oxidation stress (pH 5.0, 40° C., 14 days): no effect on CLEC12A, NKG2D or CD16a binding. Slight differences in the maximum binding signal reflect differences in capture levels.

FIG. 262 shows that the potency of AB0192 after forced oxidation in KHYG-1-CD16a mediated cytotoxicity assay was similar to the control sample.

FIG. 263A-FIG. 263B show an SEC analysis of AB0192 after long term low pH exposure (pH 5.0, 40° C., 14 days), and indicated no loss of monomer.

FIG. 264A-FIG. 264B show a CE-SDS analysis of AB0192 after long term low pH stress (pH 5.0, 40° C., 14 days), and indicated that the differences in purity detected by non-reduced and reduced CE-SDS were less than 1%. Non-reduced conditions are shown in FIG. 264A. Reduced conditions are shown in FIG. 264B.

FIG. 265A-FIG. 265F show that AB0192 is stable after low pH stress (pH 5.0, 40° C., 14 days): no effect on CLEC12A, NKG2D or CD16a binding.

FIG. 266 shows a charge analysis of AB0192 after long term low pH stress (pH 5.0, 40° C., 14 days) by cIEF, with control and pH 5 stressed samples showing no difference.

FIG. 267 shows that the potency of AB0192 after long term pH 5.0 stress was unchanged relative to control in KHYG-1-CD16a cytotoxicity assay.

FIG. 268A-FIG. 268B show an SEC analysis of AB0192 after long term high pH stress (pH 8.0, 40° C., 14 days), which indicated that there was a 1.6% loss of monomer.

FIG. 269A-FIG. 269B show a CE-SDS analysis of AB0192 after long term high pH stress (pH 8.0, 40° C., 14 days). Minor differences in purity were detected by non-reduced (FIG. 269A) and reduced (FIG. 269B) conditions.

FIG. 270A-FIG. 270F show that AB0192 is stable after high pH stress: no effect on CLEC12A, NKG2D or CD16a binding.

FIG. 271 shows a charge analysis of AB0192 after long term high pH stress (pH 8.0, 40° C., 14 days) by cIEF, with an acidic shift in the pH 8 stressed sample relative to the control sample.

FIG. 272 shows the potency of AB0192 after long term exposure to high pH 8.0 stress was not significantly affected, as assessed by KHYG-1-CD16aV cytotoxicity assay.

FIG. 273A-FIG. 273B show an SEC analysis of AB0192 after 6 cycles of freeze/thaw compared to control, which indicated no loss in monomer.

FIG. 274A-FIG. 274B show a CE-SDS analysis of AB0192 after freeze thaw stress, which indicated no loss of purity by either reduced (FIG. 274A) or non-reduced (FIG. 274B) conditions.

FIG. 275A-FIG. 275B show an SEC analysis of AB0192 after agitation stress compared to control, which indicated no loss of monomer.

FIG. 276A-FIG. 276B show a CE-SDS analysis of AB0192 after agitation stress compared to control, which indicated no increase in fragmentation detected by reduced (FIG. 276A) or non-reduced (FIG. 276B).

FIG. 277A-FIG. 277B show an SEC analysis of pH hold sample vs control, which indicated that no change in profile was observed after low pH hold.

FIG. 278 shows a cIEF analysis of AB0192 after low pH hold compared to control, which indicated that the low pH hold did not have a measurable effect on the charge profile of AB0192.

FIG. 279A-FIG. 279F show that AB0192 is stable after low pH hold: no effect on CLEC12A, NKG2D or CD16a binding. Slight differences in the maximum binding signal reflect differences in capture levels.

FIG. 280 shows the potency of AB0192 after low pH hold in KHYG1-CD16a mediated cytotoxicity assay, which indicated that the potency of AB0192 was unaltered after the low pH hold step.

FIG. 281A-FIG. 281B show binding of AB0089 to hCLEC12A pre and post low pH hold. Slight differences in absolute maximal signals reflect difference in capture levels of AB0089.

FIG. 282 shows AB0089 is stable after low pH hold. No difference between potencies of low pH hold and control samples was observed.

DETAILED DESCRIPTION

The present application provides multispecific binding proteins that bind the NKG2D receptor and CD16 receptor on natural killer cells, and the tumor-associated antigen. In some embodiments, the multispecific proteins further include an additional antigen-binding site that binds the tumor-associated antigen. The application also provides pharmaceutical compositions comprising such multispecific binding proteins, and therapeutic methods using such multispecific proteins and pharmaceutical compositions, for purposes such as treating autoimmune diseases and cancer. Various aspects of the multispecific binding proteins described in the present application are set forth below in sections; however, aspects of the multispecific binding proteins described in one particular section are not to be limited to any particular section.

To facilitate an understanding of the present application, a number of terms and phrases are defined below.

The terms “a” and “an” as used herein mean “one or more” and include the plural unless the context is inappropriate.

As used herein, the term “antigen-binding site” refers to the part of the immunoglobulin molecule that participates in antigen binding. In human antibodies, the antigen binding site is formed by amino acid residues of the N-terminal variable (“V”) regions of the heavy (“H”) and light (“L”) chains. Three highly divergent stretches within the V regions of the heavy and light chains are referred to as “hypervariable regions” which are interposed between more conserved flanking stretches known as “framework regions,” or “FR.” Thus the term “FR” refers to amino acid sequences which are naturally found between and adjacent to hypervariable regions in immunoglobulins. In a human antibody molecule, the three hypervariable regions of a light chain and the three hypervariable regions of a heavy chain are disposed relative to each other in three-dimensional space to form an antigen-binding surface. The antigen-binding surface is complementary to the three-dimensional surface of a bound antigen, and the three hypervariable regions of each of the heavy and light chains are referred to as “complementarity-determining regions,” or “CDRs.” In certain animals, such as camels and cartilaginous fish, the antigen-binding site is formed by a single antibody chain providing a “single domain antibody.” Antigen-binding sites can exist in an intact antibody, in an antigen-binding fragment of an antibody that retains the antigen-binding surface, or in a recombinant polypeptide such as an scFv, using a peptide linker to connect the heavy chain variable domain to the light chain variable domain in a single polypeptide.

The term “tumor-associated antigen” as used herein means any antigen including but not limited to a protein, glycoprotein, ganglioside, carbohydrate, lipid that is associated with cancer. Such antigen can be expressed on malignant cells or in the tumor microenvironment such as on tumor-associated blood vessels, extracellular matrix, mesenchymal stroma, or immune infiltrates. In preferred embodiments of the present disclosure, the terms “tumor-associated antigen” refers to CLEC12A.

As used herein, the terms “subject” and “patient” refer to an organism to be treated by the methods and compositions described herein. Such organisms preferably include, but are not limited to, mammals (e.g., murines, simians, equines, bovines, porcines, canines, felines, and the like), and more preferably include humans.

As used herein, the term “effective amount” refers to the amount of a compound (e.g., a compound of the present application) sufficient to effect beneficial or desired results. An effective amount can be administered in one or more administrations, applications or dosages and is not intended to be limited to a particular formulation or administration route. As used herein, the term “treating” includes any effect, e.g., lessening, reducing, modulating, ameliorating or eliminating, that results in the improvement of the condition, disease, disorder, and the like, or ameliorating a symptom thereof.

As used herein, the term “pharmaceutical composition” refers to the combination of an active agent with a carrier, inert or active, making the composition especially suitable for diagnostic or therapeutic use in vivo or ex vivo.

As used herein, the term “pharmaceutically acceptable carrier” refers to any of the standard pharmaceutical carriers, such as a phosphate buffered saline solution, water, emulsions (e.g., such as an oil/water or water/oil emulsions), and various types of wetting agents. The compositions also can include stabilizers and preservatives. For examples of carriers, stabilizers and adjuvants, see e.g., Martin, Remington's Pharmaceutical Sciences, 15th Ed., Mack Publ. Co., Easton, Pa. [1975].

As used herein, the term “pharmaceutically acceptable salt” refers to any pharmaceutically acceptable salt (e.g., acid or base) of a compound described in the present application which, upon administration to a subject, is capable of providing a compound described in this application or an active metabolite or residue thereof. As is known to those of skill in the art, “salts” of the compounds described in the present application may be derived from inorganic or organic acids and bases. Exemplary acids include, but are not limited to, hydrochloric, hydrobromic, sulfuric, nitric, perchloric, fumaric, maleic, phosphoric, glycolic, lactic, salicylic, succinic, toluene-p-sulfonic, tartaric, acetic, citric, methanesulfonic, ethanesulfonic, formic, benzoic, malonic, naphthalene-2-sulfonic, benzenesulfonic acid, and the like. Other acids, such as oxalic, though not in themselves pharmaceutically acceptable, may be employed in the preparation of salts useful as intermediates in obtaining the compounds described in the application and their pharmaceutically acceptable acid addition salts.

Exemplary bases include, but are not limited to, alkali metal (e.g., sodium) hydroxides, alkaline earth metal (e.g., magnesium) hydroxides, ammonia, and compounds of formula NW4⁺, wherein W is C_1-4alkyl, and the like.

Exemplary salts include, but are not limited to: acetate, adipate, alginate, aspartate, benzoate, benzenesulfonate, bisulfate, butyrate, citrate, camphorate, camphorsulfonate, cyclopentanepropionate, digluconate, dodecylsulfate, ethanesulfonate, fumarate, flucoheptanoate, glycerophosphate, hemisulfate, heptanoate, hexanoate, hydrochloride, hydrobromide, hydroiodide, 2-hydroxyethanesulfonate, lactate, maleate, methanesulfonate, 2-naphthalenesulfonate, nicotinate, oxalate, palmoate, pectinate, persulfate, phenylpropionate, picrate, pivalate, propionate, succinate, tartrate, thiocyanate, tosylate, undecanoate, and the like. Other examples of salts include anions of the compounds described in the present application compounded with a suitable cation such as Na⁺, NH₄⁺, and NW₄⁺ (wherein W is a C_1-4alkyl group), and the like.

For therapeutic use, salts of the compounds described in the present application are contemplated as being pharmaceutically acceptable. However, salts of acids and bases that are non-pharmaceutically acceptable may also find use, for example, in the preparation or purification of a pharmaceutically acceptable compound.

As used herein, CLEC12A (also known as CLL-1, DCAL-2, MICL, and CD371) refers to the protein of Uniprot Accession No. Q5QGZ9 and related isoforms and orthologs.

Throughout the description, where compositions are described as having, including, or comprising specific components, or where processes and methods are described as having, including, or comprising specific steps, it is contemplated that, additionally, there are compositions described in the present application that consist essentially of, or consist of, the recited components, and that there are processes and methods according to the present application that consist essentially of, or consist of, the recited processing steps.

As a general matter, compositions specifying a percentage are by weight unless otherwise specified. Further, if a variable is not accompanied by a definition, then the previous definition of the variable controls.

Abbreviation
Convention

A49M-I
Human anti-NKG2D antibody clone A49M-I

ADCC
Antibody dependent cellular cytotoxicity

ADCP
Antibody dependent cellular phagocytosis

AML
Acute myeloid leukemia

BM
Back mutation (murine residues introduced to human

mAb sequence)

cCLEC12A
Cynomolgus monkey CLEC12A

CD16
Low affinity immunoglobulin gamma Fc region

receptor III-A

CD16a
Fc gamma receptor III

CDC
Complement dependent cytotoxicity

CDR
complementarity determining region

CF SE
Carboxyfluorescein succinimidyl ester

CHO
Chinese Hamster Ovary

cIEF
capillary isoelectric focusing

CIEX
Cation Ion Exchange Chromatography

CLEC12A
C-type lectin domain family 12 member A

CLEC12A
C-type lectin domain family 12 member A, major

variant with K244, also known as CLEC12A (K244)

or CLEC12A WT

CLEC12AL
CLEC12A-Ligand

CMP
Common myeloid precursor

ConA
Concanavalin A

CST
20 mM sodium citrate, 250 mM sucrose, 0.01%

Tween-80, pH 7.0

DI water
Distilled water

DSC
Differential Scanning Calorimetry

EC50
effective concentration that gives half-maximal

response

Fab
fragment antigen-binding

FACS
Fluorescence activated cell sorting

Fc
fragment crystallizable region

FcγR
Fc-gamma receptor

GMP
Granulocyte monocyte precursor

FIBS
Hepes buffered saline

hcFAE-
Duobody TriNKET based on CLL1 antibody from

A49.CLL1-
Merus

Merus

hcFAE-
Duobody TriNKET based on h6E7 antibody from

49.h6E7
Genentech

hcFAE-
Duobody TriNKET based on CLEC12A binding arm

A49.tepoditamab
of tepoditamab from Merus

hCLEC12A
Human CLEC12A

HCT
Hematopoietic cell transfer

HI-FBS
Heat inactivated fetal bovine serum

BMWS
high molecular weight species

HPLC
high performance liquid chromatography

HST
20 mM histidine, 250 mM sucrose, 0.01%

polysorbate 80

HTFR
Homogenous time resolved fluorescence

IgG1
immunoglobulin G1

ITIM
Immunoreceptor tyrosine-based inhibitory motif

k_a
association rate constant

K_D
equilibrium dissociation constant

k_d
dissociation rate constant

LC
liquid chromatography

LMWS
low molecular weight species

LSC
Leukemic stem cells

mAb
Monoclonal antibody

MEP
Megakaryocyte erythroid precursor

MFI
Median Fluorescence Intensity

MHC
Major histocompatibility complex

MIC-A
MHC class I chain-related protein A

MS
mass spectrometry

MS/MS
tandem mass spectrometry

MSU
Monosodium urate

NK cell
Natural killer cell

NKG2D
NKG2-D type II integral membrane protein

PBS
phosphate buffered saline

PBSF
Phosphate Buffered Saline (0.1% Bovine Serum

Albumin (BSA))

pI
isoelectric point

PI3K
phosphatidylinositol 3′ kinase

PMT
Photomultiplier tube

PS-80
polysorbate-80, Tween-80

PSR
Poly Specific Reagent

RU
resonance units, a measure of mass reported by SPR

ScFv
single-chain variable fragment

SDS-CE
sodium dodecyl sulfate capillary electrophoresis

SEC
size exclusion chromatography

SH2
Src homology 2 domain

SHP1
Src homology region 2 domain-containing

phosphatase-1

SHP2
Src homology region 2 domain-containing

phosphatase-2

SPR
surface plasmon resonance

TAA
Tumor associated antigen

TriNKET
Trispecific Natural Killer cell Engaging Therapy

UHPLC
ultra-high performance liquid chromatography

WT
Wild-type

I. Proteins

The present application provides multispecific binding proteins that bind to the NKG2D receptor and CD16 receptor on natural killer cells, and CLEC12A. The multispecific binding proteins are useful in the pharmaceutical compositions and therapeutic methods described herein. Binding of the multispecific binding proteins to the NKG2D receptor and CD16 receptor on a natural killer cell enhances the activity of the natural killer cell toward destruction of tumor cells expressing the tumor antigen. Binding of the multispecific binding proteins to tumor antigen expressing tumor cells brings these cells into proximity with the natural killer cell, which facilitates direct and indirect destruction of the tumor cells by the natural killer cell. Multispecific binding proteins that bind NKG2D, CD16, and another target are disclosed in International Application Publication Nos. WO2018148445 and WO2019157366, which are not incorporated herein by reference. Further description of some exemplary multispecific binding proteins is provided below.

The first component of the multispecific binding protein is an antigen-binding site that binds to NKG2D receptor-expressing cells, which can include but are not limited to NK cells, γδ T cells and CD8⁺ αβ T cells. Upon NKG2D binding, the multispecific binding proteins may block natural ligands, such as ULBP6 and MICA, from binding to NKG2D and activating NK cells.

The second component of the multispecific binding proteins is an antigen-binding site that binds CLEC12A, a tumor-associated antigen. The tumor-associated antigen expressing cells may be found, for example, in solid tumor indications such as gastric cancer, esophageal cancer, lung cancer, breast cancer (e.g., triple-negative breast cancer), colorectal cancer, ovarian cancer, esophageal cancer, uterine cancer, cervical cancer, head and neck cancer, glioma, and pancreatic cancer, or in certain hematological malignancies such as acute myeloid leukemia.

The third component of the multispecific binding proteins is an antibody Fc domain or a portion thereof or an antigen-binding site that binds to cells expressing CD16, an Fc receptor on the surface of leukocytes including natural killer cells, macrophages, neutrophils, eosinophils, mast cells, and follicular dendritic cells.

An additional antigen-binding site of the multispecific binding proteins may bind the same tumor-associated antigen. In certain embodiments, the first antigen-binding site that binds NKG2D is an scFv, and the second and the additional antigen-binding sites that bind the tumor-associated antigen are each a Fab fragment. In certain embodiments, the first antigen-binding site that binds NKG2D is an scFv, and the second and the additional antigen-binding sites that bind the tumor-associated antigen are each an scFv. In certain embodiments, the first antigen-binding site that binds NKG2D is a Fab fragment, and the second and the additional antigen-binding sites that bind the tumor-associated antigen are each an scFv. In certain embodiments, the first antigen-binding site that binds NKG2D is a Fab, and the second and the additional antigen-binding sites that bind the tumor-associated antigen are each a Fab fragment.

The antigen-binding sites may each incorporate an antibody heavy chain variable domain and an antibody light chain variable domain (e.g., arranged as in an antibody, or fused together to form an scFv), or one or more of the antigen-binding sites may be a single domain antibody, such as a V_HH antibody like a camelid antibody or a V_NARantibody like those found in cartilaginous fish.

In some embodiments, the second antigen-binding site incorporates a light chain variable domain having an amino acid sequence identical to the amino acid sequence of the light chain variable domain present in the first antigen-binding site.

The multispecific binding proteins described herein can take various formats. For example, one format is a heterodimeric, multispecific antibody including a first immunoglobulin heavy chain, a first immunoglobulin light chain, a second immunoglobulin heavy chain and a second immunoglobulin light chain (FIG. 1). The first immunoglobulin heavy chain includes a first Fc (hinge-CH2-CH3) domain, a first heavy chain variable domain and optionally a first CH1 heavy chain domain. The first immunoglobulin light chain includes a first light chain variable domain and optionally a first light chain constant domain. The first immunoglobulin light chain, together with the first immunoglobulin heavy chain, forms an antigen-binding site that binds NKG2D. The second immunoglobulin heavy chain comprises a second Fc (hinge-CH2-CH3) domain, a second heavy chain variable domain and optionally a second CH1 heavy chain domain. The second immunoglobulin light chain includes a second light chain variable domain and optionally a second light chain constant domain. The second immunoglobulin light chain, together with the second immunoglobulin heavy chain, forms an antigen-binding site that binds CLEC12A. In some embodiments, the first Fc domain and second Fc domain together are able to bind to CD16 (FIG. 1). In some embodiments, the first immunoglobulin light chain is identical to the second immunoglobulin light chain.

Another exemplary format involves a heterodimeric, multispecific antibody including a first immunoglobulin heavy chain, a second immunoglobulin heavy chain and an immunoglobulin light chain (e.g., FIG. 2A). In some embodiments, the first immunoglobulin heavy chain includes a first Fc (hinge-CH2-CH3) domain fused via either a linker or an antibody hinge to a single-chain variable fragment (scFv) composed of a heavy chain variable domain and light chain variable domain, which pair and bind NKG2D or bind CLEC12A. The second immunoglobulin heavy chain includes a second Fc (hinge-CH2-CH3) domain, a second heavy chain variable domain and a CH1 heavy chain domain. The immunoglobulin light chain includes a light chain variable domain and a light chain constant domain. In some embodiments, the second immunoglobulin heavy chain pairs with the immunoglobulin light chain and binds to NKG2D or binds the tumor-associated antigen, with the proviso that when the first Fc domain is fused to an scFv that binds NKG2D, the second immunoglobulin heavy chain paired with the immunoglobulin light chain binds the tumor-associated antigen, but not NKG2D, and vice versa. In some embodiments, the scFv in the first immunoglobulin heavy chain binds CLEC12A; and the heavy chain variable domain in the second immunoglobulin heavy chain and the light chain variable domain in the immunoglobulin light chain, when paired, bind NKG2D (e.g., FIG. 2E). In some embodiments, the scFv in the first immunoglobulin heavy chain binds NKG2D; and the heavy chain variable domain in the second immunoglobulin heavy chain and the light chain variable domain in the immunoglobulin light chain, when paired, bind CLEC12A. In some embodiments, the first Fc domain and the second Fc domain together are able to bind to CD16 (e.g., FIG. 2A).

Another exemplary format involves a heterodimeric, multispecific antibody including a first immunoglobulin heavy chain, and a second immunoglobulin heavy chain (e.g., FIG. 2B). In some embodiments, the first immunoglobulin heavy chain includes a first Fc (hinge-CH2-CH3) domain fused via either a linker or an antibody hinge to a single-chain variable fragment (scFv) composed of a heavy chain variable domain and light chain variable domain, which pair and bind NKG2D, or bind CLEC12A. In some embodiments, the second immunoglobulin heavy chain includes a second Fc (hinge-CH2-CH3) domain fused via either a linker or an antibody hinge to a single-chain variable fragment (scFv) composed of a heavy chain variable domain and light chain variable domain which pair and bind NKG2D, or bind the tumor-associated antigen, with the proviso that when the first Fc domain is fused to an scFv that binds NKG2D, the second Fc domain fused to an scFv binds the tumor-associated antigen, but not NKG2D, and vice versa. In some embodiments, the first Fc domain and the second Fc domain together are able to bind to CD16 (e.g., FIG. 2B).

In some embodiments, the single-chain variable fragment (scFv) described above is linked to the antibody constant domain via a hinge sequence. In some embodiments, the hinge comprises amino acids Ala-Ser or Gly-Ser. In some embodiments, the hinge connects an scFv that binds NKG2D and the antibody heavy chain constant domain comprises amino acids Ala-Ser. In some embodiments, the hinge connects an scFv that binds the tumor-associated antigen and the antibody heavy chain constant domain comprises amino acids Gly-Ser. In some other embodiments, the hinge comprises amino acids Ala-Ser and Thr-Lys-Gly. The hinge sequence can provide flexibility of binding to the target antigen, and balance between flexibility and optimal geometry.

In some embodiments, the single-chain variable fragment (scFv) described above includes a heavy chain variable domain and a light chain variable domain. In some embodiments, the heavy chain variable domain forms a disulfide bridge with the light chain variable domain to enhance stability of the scFv. For example, a disulfide bridge can be formed between the C44 residue of the heavy chain variable domain and the C100 residue of the light chain variable domain, the amino acid positions numbered under Kabat. In some embodiments, the heavy chain variable domain is linked to the light chain variable domain via a flexible linker. Any suitable linker can be used, for example, the (G₄S)₄linker ((GlyGlyGlyGlySer)₄(SEQ ID NO:119)). In some embodiments of the scFv, the heavy chain variable domain is positioned at the N-terminus of the light chain variable domain. In some embodiments of the scFv, the heavy chain variable domain is positioned at the C terminus of the light chain variable domain.

The multispecific binding proteins described herein can further include one or more additional antigen-binding sites. The additional antigen-binding site(s) may be fused to the N-terminus of the constant region CH2 domain or to the C-terminus of the constant region CH3 domain, optionally via a linker sequence. In certain embodiments, the additional antigen-binding site(s) takes the form of a single-chain variable region (scFv) that is optionally disulfide-stabilized, resulting in a tetravalent or trivalent multispecific binding protein. For example, a multispecific binding protein includes a first antigen-binding site that binds NKG2D, a second antigen-binding site that binds CLEC12A, an additional antigen-binding site that binds the tumor-associated antigen, and an antibody constant region or a portion thereof sufficient to bind CD16 or a fourth antigen-binding site that binds CD16. Any one of these antigen binding sites can either take the form of a Fab fragment or an scFv, such as an scFv described above.

In some embodiments, the additional antigen-binding site binds a different epitope of the tumor-associated antigen from the second antigen-binding site. In some embodiments, the additional antigen-binding site binds the same epitope as the second antigen-binding site. In some embodiments, the additional antigen-binding site comprises the same heavy chain and light chain CDR sequences as the second antigen-binding site. In some embodiments, the additional antigen-binding site comprises the same heavy chain and light chain variable domain sequences as the second antigen-binding site. In some embodiments, the additional antigen-binding site has the same amino acid sequence(s) as the second antigen-binding site. In some embodiments, the additional antigen-binding site comprises heavy chain and light chain variable domain sequences that are different from the heavy chain and light chain variable domain sequences of the second antigen-binding site. In some embodiments, the additional antigen-binding site has an amino acid sequence that is different from the sequence of the second antigen-binding site. In some embodiments, the second antigen-binding site and the additional antigen-binding site bind different tumor-associated antigens. In some embodiments, the second antigen-binding site and the additional antigen-binding site binds different antigens. Exemplary formats are shown in FIG. 2C and FIG. 2D. Accordingly, the multispecific binding proteins can provide bivalent engagement of the tumor-associated antigen. Bivalent engagement of the tumor-associated antigen by the multispecific proteins can stabilize the tumor-associated antigen on the tumor cell surface and enhance cytotoxicity of NK cells towards the tumor cells. Bivalent engagement of the tumor-associated antigen by the multispecific proteins can confer stronger binding of the multispecific proteins to the tumor cells, thereby facilitating stronger cytotoxic response of NK cells towards the tumor cells, especially towards tumor cells expressing a low level of the tumor-associated antigen.

The multispecific binding proteins can take additional formats. In some embodiments, the multispecific binding protein is in the Triomab form, which is a trifunctional, bispecific antibody that maintains an IgG-like shape. This chimera consists of two half antibodies, each with one light and one heavy chain, that originate from two parental antibodies.

In some embodiments, the multispecific binding protein is the KiH form, which involves the knobs-into-holes (KiHs) technology. The KiH involves engineering C_H3 domains to create either a “knob” or a “hole” in each heavy chain to promote heterodimerization. The concept behind the “Knobs-into-Holes (KiH)” Fc technology was to introduce a “knob” in one CH3 domain (CH3A) by substitution of a small residue with a bulky one (e.g., T366W_CH3Ain EU numbering). To accommodate the “knob,” a complementary “hole” surface was created on the other CH3 domain (CH3B) by replacing the closest neighboring residues to the knob with smaller ones (e.g., T366S/L368A/Y407V_CH3B). The “hole” mutation was optimized by structured-guided phage library screening (Atwell S, Ridgway J B, Wells J A, Carter P., Stable heterodimers from remodeling the domain interface of a homodimer using a phage display library, J. Mol. Biol. (1997) 270(1):26-35). X-ray crystal structures of KiH Fc variants (Elliott J M, Ultsch M, Lee J, Tong R, Takeda K, Spiess C, et al., Antiparallel conformation of knob and hole aglycosylated half-antibody homodimers is mediated by a CH2-CH3 hydrophobic interaction. J. Mol. Biol. (2014) 426(9):1947-57; Mimoto F, Kadono S, Katada H, Igawa T, Kamikawa T, Hattori K. Crystal structure of a novel asymmetrically engineered Fc variant with improved affinity for FcγRs. Mol. Immunol. (2014) 58(1):132-8) demonstrated that heterodimerization is thermodynamically favored by hydrophobic interactions driven by steric complementarity at the inter-CH3 domain core interface, whereas the knob-knob and the hole-hole interfaces do not favor homodimerization owing to steric hindrance and disruption of the favorable interactions, respectively.

In some embodiments, the multispecific binding protein is in the dual-variable domain immunoglobulin (DVD-Ig™) form, which combines the target binding domains of two monoclonal antibodies via flexible naturally occurring linkers, and yields a tetravalent IgG-like molecule.

In some embodiments, the multispecific binding protein is in the Orthogonal Fab interface (Ortho-Fab) form. In the ortho-Fab IgG approach (Lewis S M, Wu X, Pustilnik A, Sereno A, Huang F, Rick H L, et al., Generation of bispecific IgG antibodies by structure-based design of an orthogonal Fab interface. Nat. Biotechnol. (2014) 32(2):191-8), structure-based regional design introduces complementary mutations at the LC and HC_VH-CH1interface in only one Fab fragment, without any changes being made to the other Fab fragment.

In some embodiments, the multispecific binding protein is in the 2-in-1 Ig format. In some embodiments, the multispecific binding protein is in the ES form, which is a heterodimeric construct containing two different Fab fragments binding to targets 1 and target 2 fused to the Fc. Heterodimerization is ensured by electrostatic steering mutations in the Fc.

In some embodiments, the multispecific binding protein is in the κλ-Body form, which is a heterodimeric construct with two different Fab fragments fused to Fc stabilized by heterodimerization mutations: Fab fragment 1 targeting antigen 1 contains kappa LC, and Fab fragment 2 targeting antigen 2 contains lambda LC. FIG. 13A is an exemplary representation of one form of a κλ-Body; FIG. 13B is an exemplary representation of another κλ-Body.

In some embodiments, the multispecific binding protein is in Fab Arm Exchange form (antibodies that exchange Fab fragment arms by swapping a heavy chain and attached light chain (half-molecule) with a heavy-light chain pair from another molecule, which results in bispecific antibodies).

In some embodiments, the multispecific binding protein is in the SEED Body form. The strand-exchange engineered domain (SEED) platform was designed to generate asymmetric and bispecific antibody-like molecules, a capability that expands therapeutic applications of natural antibodies. This protein engineering platform is based on exchanging structurally related sequences of immunoglobulin within the conserved CH3 domains. The SEED design allows efficient generation of AG/GA heterodimers, whereas disfavoring homodimerization of AG and GA SEED CH3 domains. (Muda M. et al., Protein Eng. Des. Sel. (2011, 24(5):447-54)).

In some embodiments, the multispecific binding protein is in the LuZ-Y form, in which a leucine zipper is used to induce heterodimerization of two different HCs. (Wranik, B J. et al., J. Biol. Chem. (2012), 287:43331-9).

In some embodiments, the multispecific binding protein is in the Cov-X-Body form. In bispecific CovX-Bodies, two different peptides are joined together using a branched azetidinone linker and fused to the scaffold antibody under mild conditions in a site-specific manner. Whereas the pharmacophores are responsible for functional activities, the antibody scaffold imparts long half-life and Ig-like distribution. The pharmacophores can be chemically optimized or replaced with other pharmacophores to generate optimized or unique bispecific antibodies. (Doppalapudi V R et al., PNAS (2010), 107(52);22611-22616).

In some embodiments, the multispecific binding protein is in an Oasc-Fab heterodimeric form that includes Fab fragment binding to target 1, and scFab binding to target 2 fused to Fc. Heterodimerization is ensured by mutations in the Fc.

In some embodiments, the multispecific binding protein is in a DuetMab form, which is a heterodimeric construct containing two different Fab fragments binding to antigens 1 and 2, and Fc stabilized by heterodimerization mutations. Fab fragments 1 and 2 contain differential S—S bridges that ensure correct LC and HC pairing.

In some embodiments, the multispecific binding protein is in a CrossmAb form, which is a heterodimeric construct with two different Fab fragments binding to targets 1 and 2, fused to Fc stabilized by heterodimerization. CL and CH1 domains and VH and VL domains are switched, e.g., CH1 is fused in-frame with VL, and CL is fused in-frame with VH.

In some embodiments, the multispecific binding protein is in a Fit-Ig form, which is a homodimeric construct where Fab fragment binding to antigen 2 is fused to the N terminus of HC of Fab fragment that binds to antigen 1. The construct contains wild-type Fc.

Individual components of the multispecific binding proteins are described in more detail below.

NKG2D-Binding Site

Upon binding to the NKG2D receptor and CD16 receptor on natural killer cells, and CLEC12A, the multispecific binding proteins can engage more than one kind of NK-activating receptor, and may block the binding of natural ligands to NKG2D. In certain embodiments, the proteins can agonize NK cells in humans. In some embodiments, the proteins can agonize NK cells in humans and in other species such as rodents and cynomolgus monkeys. In some embodiments, the proteins can agonize NK cells in humans and in other species such as cynomolgus monkeys.

Table 1 lists peptide sequences of heavy chain variable domains and light chain variable domains that, in combination, can bind to NKG2D. In some embodiments, the heavy chain variable domain and the light chain variable domain are arranged in Fab format. In some embodiments, the heavy chain variable domain and the light chain variable domain are fused together to form an scFv.

The NKG2D binding sites listed in Table 1 can vary in their binding affinity to NKG2D, nevertheless, they all activate human NK cells.

Unless indicated otherwise, the CDR sequences provided in Table 1 are determined under Kabat numbering.

TABLE 1

Heavy chain variable region amino acid
Light chain variable region amino acid

Clones
sequence
sequence

ADI-27705
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYNSYPITFGGGTK

TLVTVSS
VEIK

(SEQ ID NO: 1)
(SEQ ID NO: 5)

CDR1 (SEQ ID NO: 2) - GSFSGYYWS

CDR2 (SEQ ID NO: 3) -

EIDHSGSTNYNPSLKS

CDR3 (SEQ ID NO: 4) -

ARARGPWSFDP

ADI-27724
QVQLQQWGAGLLKPSETLSLTCAVYGG
EIVLTQSPGTLSLSPGERATLSCRASQ

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
SVSSSYLAWYQQKPGQAPRLLIYGA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSRATGIPDRFSGSGSGTDFTLTISRL

VTAADTAVYYCARARGPWSFDPWGQG
EPEDFAVYYCQQYGSSPITFGGGTK

TLVTVSS
VEIK

(SEQ ID NO: 1)
(SEQ ID NO: 6)

ADI-27740
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

(A40)
SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSIGSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYHSFYTFGGGTK

TLVTVSS
VEIK

(SEQ ID NO: 1)
(SEQ ID NO: 7)

ADI-27741
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSIGSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQSNSYYTFGGGTK

TLVTVSS
VEIK

(SEQ ID NO: 1)
(SEQ ID NO: 8)

ADI-27743
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYNSYPTFGGGTK

TLVTVSS
VEIK

(SEQ ID NO: 1)
(SEQ ID NO: 9)

ADI-28153
QVQLQQWGAGLLKPSETLSLTCAVYGG
ELQMTQSPSSLSASVGDRVTITCRTS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSYLNWYQQKPGQPPKLLIYWA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
STRESGVPDRFSGSGSGTDFTLTISSL

VTAADTAVYYCARARGPWGFDPWGQG
QPEDSATYYCQQSYDIPYTFGQGTK

TLVTVSS
LEIK

(SEQ ID NO: 10)
(SEQ ID NO: 11)

ADI-28226
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

(C26)
SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYGSFPITFGGGTK

TLVTVSS
VEIK

(SEQ ID NO: 1)
(SEQ ID NO: 12)

ADI-28154
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTDFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQSKEVPWTFGQGT

TLVTVSS
KVEIK

(SEQ ID NO: 1)
(SEQ ID NO: 13)

ADI-29399
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYNSFPTFGGGTKV

TLVTVSS
EIK

(SEQ ID NO: 1)
(SEQ ID NO: 14)

ADI-29401
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSIGSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYDIYPTFGGGTKV

TLVTVSS
EIK

(SEQ ID NO: 1)
(SEQ ID NO: 15)

ADI-29403
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYDSYPTFGGGTK

TLVTVSS
VEIK

(SEQ ID NO: 1)
(SEQ ID NO: 16)

ADI-29405
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYGSFPTFGGGTKV

TLVTVSS
EIK

(SEQ ID NO: 1)
(SEQ ID NO: 17)

ADI-29407
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYQSFPTFGGGTKV

TLVTVSS
EIK

(SEQ ID NO: 1)
(SEQ ID NO: 18)

ADI-29419
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYSSFSTFGGGTKV

TLVTVSS
EIK

(SEQ ID NO: 1)
(SEQ ID NO: 19)

ADI-29421
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYESYSTFGGGTKV

TLVTVSS
EIK

(SEQ ID NO: 1)
(SEQ ID NO: 20)

ADI-29424
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYDSFITFGGGTKV

TLVTVSS
EIK

(SEQ ID NO: 1)
(SEQ ID NO: 21)

ADI-29425
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYQSYPTFGGGTK

TLVTVSS
VEIK

(SEQ ID NO: 1)
(SEQ ID NO: 22)

ADI-29426
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSIGSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYHSFPTFGGGTKV

TLVTVSS
EIK

(SEQ ID NO: 1)
(SEQ ID NO: 23)

ADI-29429
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSIGSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYELYSYTFGGGTK

TLVTVSS
VEIK

(SEQ ID NO: 1)
(SEQ ID NO: 24)

ADI-29447
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

(F47)
SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCQQYDTFITFGGGTKV

TLVTVSS
EIK

(SEQ ID NO: 1)
(SEQ ID NO: 25)

ADI-27727
QVQLVQSGAEVKKPGSSVKVSCKASGG
DIVMTQSPDSLAVSLGERATINCKSS

TFSSYAISWVRQAPGQGLEWMGGIIPIFG
QSVLYSSNNKNYLAWYQQKPGQPP

TANYAQKFQGRVTITADESTSTAYMELS
KLLIYWASTRESGVPDRFSGSGSGTD

SLRSEDTAVYYCARGDSSIRHAYYYYG
FTLTISSLQAEDVAVYYCQQYYSTPI

MDVWGQGTTVTVSS
TFGGGTKVEIK

(SEQ ID NO: 26)
(SEQ ID NO: 32)

CDR1- GTFSSYAIS (non-Kabat)(SEQ ID
CDR1 (SEQ ID NO: 33) -

NO: 27) or SYAIS (SEQ ID NO: 28)
KSSQSVLYSSNNKNYLA

CDR2 (SEQ ID NO: 29) -
CDR2 (SEQ ID NO: 34) -

GIIPIFGTANYAQKFQG
WASTRES

CDR3 - ARGDSSIRHAYYYYGMDV
CDR3 (SEQ ID NO: 35) -

(non-Kabat)(SEQ ID NO: 30) or
QQYYSTPIT

GDSSIRHAYYYYGMDV (SEQ ID NO: 31)

ADI-29443
QLQLQESGPGLVKPSETLSLTCTVSGGSI
EIVLTQSPATLSLSPGERATLSCRASQ

(F43)
SSSSYYWGWIRQPPGKGLEWIGSIYYSGS
SVSRYLAWYQQKPGQAPRLLIYDAS

TYYNPSLKSRVTISVDTSKNQFSLKLSSV
NRATGIPARFSGSGSGTDFTLTISSLE

TAADTAVYYCARGSDRFHPYFDYWGQG
PEDFAVYYCQQFDTWPPTFGGGTKV

TLVTVSS
EIK

(SEQ ID NO: 36)
(SEQ ID NO: 42)

CDR1- GSISSSSYYWG (non-Kabat)(SEQ
CDR1 (SEQ ID NO: 43) -

ID NO: 37) or SSSYYWG (SEQ ID NO: 38)
RASQSVSRYLA

CDR2 (SEQ ID NO: 39) -
CDR2 (SEQ ID NO: 44) -

SIYYSGSTYYNPSLKS
DASNRAT

CDR3 - ARGSDRFHPYFDY (non-Kabat)
CDR3 (SEQ ID NO: 45) -

(SEQ ID NO: 40) or GSDRFHPYFDY (SEQ
QQFDTWPPT

ID NO: 41)

ADI-29404
QVQLQQWGAGLLKPSETLSLTCAVYGG
DIQMTQSPSTLSASVGDRVTITCRAS

(F04)
SFSGYYWSWIRQPPGKGLEWIGEIDHSG
QSISSWLAWYQQKPGKAPKLLIYKA

STNYNPSLKSRVTISVDTSKNQFSLKLSS
SSLESGVPSRFSGSGSGTEFTLTISSL

VTAADTAVYYCARARGPWSFDPWGQG
QPDDFATYYCEQYDSYPTFGGGTKV

TLVTVSS (SEQ ID NO: 1)
EIK

(SEQ ID NO: 46)

ADI-28200
QVQLVQSGAEVKKPGSSVKVSCKASGG
DIVMTQSPDSLAVSLGERATINCESS

TFSSYAISWVRQAPGQGLEWMGGIIPIFG
QSLLNSGNQKNYLTWYQQKPGQPP

TANYAQKFQGRVTITADESTSTAYMELS
KPLIYWASTRESGVPDRFSGSGSGTD

SLRSEDTAVYYCARRGRKASGSFYYYY
FTLTISSLQAEDVAVYYCQNDYSYP

GMDVWGQGTTVTVSS
YTFGQGTKLEIK

(SEQ ID NO: 47)
(SEQ ID NO: 49)

CDR1 (SEQ ID NO: 27) - GTFSSYAIS
CDR1 (SEQ ID NO: 50) -

CDR2 (SEQ ID NO: 29) -
ESSQSLLNSGNQKNYLT

GIIPIFGTANYAQKFQG
CDR2 (SEQ ID NO: 34) - WASTRES

CDR3 (SEQ ID NO: 48) -
CDR3 (SEQ ID NO: 51) - QNDYSYPYT

ARRGRKASGSFYYYYGMDV

ADI-29379
QVQLVQSGAEVKKPGASVKVSCKASGY
EIVMTQSPATLSVSPGERATLSCRAS

(E79)
TFTSYYMHWVRQAPGQGLEWMGIINPS
QSVSSNLAWYQQKPGQAPRLLIYGA

GGSTSYAQKFQGRVTMTRDTSTSTVYM
STRATGIPARFSGSGSGTEFTLTISSL

ELSSLRSEDTAVYYCARGAPNYGDTTHD
QSEDFAVYYCQQYDDWPFTFGGGT

YYYMDVWGKGTTVTVSS
KVEIK

(SEQ ID NO: 52)
(SEQ ID NO: 58)

CDR1 (SEQ ID NO: 53) - YTFTSYYMH
CDR1 (SEQ ID NO: 59) -

(non-Kabat) or SYYMH (SEQ ID NO: 54)
RASQSVSSNLA

CDR2 (SEQ ID NO: 55) -
CDR2 (SEQ ID NO: 60) - GASTRAT

IINPSGGSTSYAQKFQG
CDR3 (SEQ ID NO: 61) - QQYDDWPFT

CDR3 - ARGAPNYGDTTHDYYYMDV

(non-Kabat)(SEQ ID NO: 56) or

GAPNYGDTTHDYYYMDV (SEQ ID

NO: 57)

ADI-29463
QVQLVQSGAEVKKPGASVKVSCKASGY
EIVLTQSPGTLSLSPGERATLSCRASQ

(F63)
TFTGYYMHWVRQAPGQGLEWMGWINP
SVSSNLAWYQQKPGQAPRLLIYGAS

NSGGTNYAQKFQGRVTMTRDTSISTAY
TRATGIPARFSGSGSGTEFTLTISSLQ

MELSRLRSDDTAVYYCARDTGEYYDTD
SEDFAVYYCQQDDYWPPTFGGGTK

DHGMDVWGQGTTVTVSS
VEIK

(SEQ ID NO: 62)
(SEQ ID NO: 68)

CDR1 - YTFTGYYMH (non-Kabat)(SEQ ID
CDR1 (SEQ ID NO: 59) -

NO: 63) or GYYMH (SEQ ID NO: 64)
RASQSVSSNLA

CDR2 (SEQ ID NO: 65) -
CDR2 (SEQ ID NO: 60) - GASTRAT

WINPNSGGTNYAQKFQG
CDR3 (SEQ ID NO: 69) - QQDDYWPPT

CDR3 - ARDTGEYYDTDDHGMDV (non-

Kabat)(SEQ ID NO: 66) or

DTGEYYDTDDHGMDV (SEQ ID NO: 67)

ADI-27744
EVQLLESGGGLVQPGGSLRLSCAASGFT
DIQMTQSPSSVSASVGDRVTITCRAS

(A44)
FSSYAMSWVRQAPGKGLEWVSAISGSG
QGIDSWLAWYQQKPGKAPKLLIYA

GSTYYADSVKGRFTISRDNSKNTLYLQM
ASSLQSGVPSRFSGSGSGTDFTLTISS

NSLRAEDTAVYYCAKDGGYYDSGAGD
LQPEDFATYYCQQGVSYPRTFGGGT

YWGQGTLVTVSS
KVEIK

(SEQ ID NO: 70)
(SEQ ID NO: 75)

CDR1 - FTFSSYAMS (non-Kabat)(SEQ ID
CDR1 (SEQ ID NO: 76) -

NO: 71) or SYAMS (SEQ ID NO: 115)
RASQGIDSWLA

CDR2 (SEQ ID NO: 72) -
CDR2 (SEQ ID NO: 77) - AASSLQS

AISGSGGSTYYADSVKG
CDR3 (SEQ ID NO: 78) - QQGVSYPRT

CDR3 - AKDGGYYDSGAGDY (non-Kabat)

(SEQ ID NO: 73) or DGGYYDSGAGDY

(SEQ ID NO: 74)

ADI-27749
EVQLVESGGGLVKPGGSLRLSCAASGFT
DIQMTQSPSSVSASVGDRVTITCRAS

(A49)
FSSYSMNWVRQAPGKGLEWVSSISSSSS
QGISSWLAWYQQKPGKAPKLLIYAA

YIYYADSVKGRFTISRDNAKNSLYLQMN
SSLQSGVPSRFSGSGSGTDFTLTISSL

SLRAEDTAVYYCARGAPMGAAAGWFD
QPEDFATYYCQQGVSFPRTFGGGTK

PWGQGTLVTVSS
VEIK

(SEQ ID NO: 79)
(SEQ ID NO: 85)

CDR1 - FTFSSYSMN (SEQ ID NO: 80)(non-
CDR1 (SEQ ID NO: 86) -

Kabat) or SYSMN (SEQ ID NO: 81)
RASQGISSWLA

CDR2 (SEQ ID NO: 82) -
CDR2 (SEQ ID NO: 77) - AASSLQS

SISSSSSYIYYADSVKG
CDR3 (SEQ ID NO: 87) - QQGVSFPRT

CDR3- ARGAPMGAAAGWFDP (SEQ ID

NO: 83)(non-Kabat) or

GAPMGAAAGWFDP (SEQ ID NO: 84)

scFv (VL-VH) with Q44C in VH and G100C in VL, linker italicized:

DIQMTQSPSSVSASVGDRVTITCRASQGISSWLAWYQQKPGKAPKLLIYAASSLQS

GVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQQGVSFPRTFGCGTKVEIKGGGGSG

GGGSGGGGSGGGGSEVQLVESGGGLVKPGGSLRLSCAASGFTFSSYSMNWVRQA

PGKCLEWVSSISSSSSYIYYADSVKGRFTISRDNAKNSLYLQMNSLRAEDTAVYYC

ARGAPMGAAAGWFDPWGQGTLVTVSS (SEQ ID NO: 88)

ADI-29378
QVQLVQSGAEVKKPGASVKVSCKASGY
EIVLTQSPATLSLSPGERATLSCRASQ

(E78)
TFTSYYMHWVRQAPGQGLEWMGIINPS
SVSSYLAWYQQKPGQAPRLLIYDAS

GGSTSYAQKFQGRVTMTRDTSTSTVYM
NRATGIPARFSGSGSGTDFTLTISSLE

ELSSLRSEDTAVYYCAREGAGFAYGMD
PEDFAVYYCQQSDNWPFTFGGGTK

YYYMDVWGKGTTVTVSS
VEIK

(SEQ ID NO: 89)
(SEQ ID NO: 92)

CDR1 - YTFTSYYMH (SEQ ID NO: 53)
CDR1 (SEQ ID NO: 93) -

(non-Kabat) or SYYMH (SEQ ID NO: 54)
RASQSVSSYLA

CDR2 (SEQ ID NO: 55) -
CDR2 (SEQ ID NO: 44) - DASNRAT

IINPSGGSTSYAQKFQG
CDR3 (SEQ ID NO: 94) - QQSDNWPFT

CDR3 - AREGAGFAYGMDYYYMDV

(SEQ ID NO: 90)(non-Kabat) or

EGAGFAYGMDYYYMDV (SEQ ID NO: 91)

A49MI
EVQLVESGGGLVKPGGSLRLSCAASGFT
DIQMTQSPSSVSASVGDRVTITCRAS

FSSYSMNWVRQAPGKGLEWVSSISSSSS
QGISSWLAWYQQKPGKAPKLLIYAA

YIYYADSVKGRFTISRDNAKNSLYLQMN
SSLQSGVPSRFSGSGSGTDFTLTISSL

SLRAEDTAVYYCARGAPIGAAAGWFDP
QPEDFATYYCQQGVSFPRTFGGGTK

WGQGTLVTVSS (SEQ ID NO: 95)
VEIK

CDR1: FTFSSYSMN (SEQ ID NO: 80)(non-
(SEQ ID NO: 85)

Kabat) or SYSMN (SEQ ID NO: 81)
CDR1 (SEQ ID NO: 86) -

CDR2: SISSSSSYIYYADSVKG
RASQGISSWLA

(SEQ ID NO: 82)
CDR2 (SEQ ID NO: 77) - AASSLQS

CDR3: ARGAPIGAAAGWFDP (SEQ ID
CDR3 (SEQ ID NO: 87) - QQGVSFPRT

NO: 96)(non-Kabat) or GAPIGAAAGWFDP

(SEQ ID NO: 97)

A49MQ
EVQLVESGGGLVKPGGSLRLSCAASGFT
DIQMTQSPSSVSASVGDRVTITCRAS

FSSYSMNWVRQAPGKGLEWVSSISSSSS
QGISSWLAWYQQKPGKAPKLLIYAA

YIYYADSVKGRFTISRDNAKNSLYLQMN
SSLQSGVPSRFSGSGSGTDFTLTISSL

SLRAEDTAVYYCARGAPQGAAAGWFDP
QPEDFATYYCQQGVSFPRTFGGGTK

WGQGTLVTVSS
VEIK

(SEQ ID NO: 98)
(SEQ ID NO: 85)

CDR1: FTFSSYSMN (SEQ ID NO: 80)(non-
CDR1 (SEQ ID NO: 86) -

Kabat) or SYSMN (SEQ ID NO: 81)
RASQGISSWLA

CDR2: SISSSSSYIYYADSVKG
CDR2 (SEQ ID NO: 77) - AASSLQS

(SEQ ID NO: 82)
CDR3 (SEQ ID NO: 87) - QQGVSFPRT

CDR3 - ARGAPQGAAAGWFDP (SEQ ID

NO: 99)(non-Kabat) or GAPQGAAAGWFDP

(SEQ ID NO: 100)

A49ML
EVQLVESGGGLVKPGGSLRLSCAASGFT
DIQMTQSPSSVSASVGDRVTITCRAS

FSSYSMNWVRQAPGKGLEWVSSISSSSS
QGISSWLAWYQQKPGKAPKLLIYAA

YIYYADSVKGRFTISRDNAKNSLYLQMN
SSLQSGVPSRFSGSGSGTDFTLTISSL

SLRAEDTAVYYCARGAPLGAAAGWFDP
QPEDFATYYCQQGVSFPRTFGGGTK

WGQGTLVTVSS
VEIK

(SEQ ID NO: 101)
(SEQ ID NO: 85)

CDR1: FTFSSYSMN (SEQ ID NO: 80)(non-
CDR1 (SEQ ID NO: 86) -

Kabat) or SYSMN (SEQ ID NO: 81)
RASQGISSWLA

CDR2: SISSSSSYIYYADSVKG
CDR2 (SEQ ID NO: 77) - AASSLQS

(SEQ ID NO: 82)
CDR3 (SEQ ID NO: 87) - QQGVSFPRT

CDR3 - ARGAPLGAAAGWFDP (SEQ ID

NO: 102)(non-Kabat) or

GAPLGAAAGWFDP (SEQ ID NO: 103)

A49MF
EVQLVESGGGLVKPGGSLRLSCAASGFT
DIQMTQSPSSVSASVGDRVTITCRAS

FSSYSMNWVRQAPGKGLEWVSSISSSSS
QGISSWLAWYQQKPGKAPKLLIYAA

YIYYADSVKGRFTISRDNAKNSLYLQMN
SSLQSGVPSRFSGSGSGTDFTLTISSL

SLRAEDTAVYYCARGAPFGAAAGWFDP
QPEDFATYYCQQGVSFPRTFGGGTK

WGQGTLVTVSS
VEIK

(SEQ ID NO: 104)
(SEQ ID NO: 85)

CDR1: FTFSSYSMN (SEQ ID NO: 80)(non-
CDR1 (SEQ ID NO: 86) -

Kabat) or SYSMN (SEQ ID NO: 81)
RASQGISSWLA

CDR2: SISSSSSYIYYADSVKG
CDR2 (SEQ ID NO: 77) - AASSLQS

(SEQ ID NO: 82)
CDR3 (SEQ ID NO: 87) - QQGVSFPRT

CDR3 - ARGAPFGAAAGWFDP (SEQ ID

NO: 105)(non-Kabat) or

GAPFGAAAGWFDP (SEQ ID NO: 106)

A49MV
EVQLVESGGGLVKPGGSLRLSCAASGFT
DIQMTQSPSSVSASVGDRVTITCRAS

FSSYSMNWVRQAPGKGLEWVSSISSSSS
QGISSWLAWYQQKPGKAPKLLIYAA

YIYYADSVKGRFTISRDNAKNSLYLQMN
SSLQSGVPSRFSGSGSGTDFTLTISSL

SLRAEDTAVYYCARGAPVGAAAGWFDP
QPEDFATYYCQQGVSFPRTFGGGTK

WGQGTLVTVSS
VEIK

(SEQ ID NO: 107)
(SEQ ID NO: 85)

CDR1: FTFSSYSMN (SEQ ID NO: 80)(non-
CDR1 (SEQ ID NO: 86) -

Kabat) or SYSMN (SEQ ID NO: 81)
RASQGISSWLA

CDR2: SISSSSSYIYYADSVKG
CDR2 (SEQ ID NO: 77) - AASSLQS

(SEQ ID NO: 82)
CDR3 (SEQ ID NO: 87) - QQGVSFPRT

CDR3- ARGAPVGAAAGWFDP (SEQ ID

NO: 108)(non-Kabat) or

GAPVGAAAGWFDP (SEQ ID NO: 109)

A49-
EVQLVESGGGLVKPGGSLRLSCAASGFT
DIQMTQSPSSVSASVGDRVTITCRAS

consensus
FSSYSMNWVRQAPGKGLEWVSSISSSSS
QGISSWLAWYQQKPGKAPKLLIYAA

YIYYADSVKGRFTISRDNAKNSLYLQMN
SSLQSGVPSRFSGSGSGTDFTLTISSL

SLRAEDTAVYYCARGAPXGAAAGWFDP
QPEDFATYYCQQGVSFPRTFGGGTK

WGQGTLVTVSS, wherein X is M, L, I, V,
VEIK

Q, or F
(SEQ ID NO: 85)

(SEQ ID NO: 110)
CDR1 (SEQ ID NO: 86) -

CDR1: FTFSSYSMN (SEQ ID NO: 80)(non-
RASQGISSWLA

Kabat) or SYSMN (SEQ ID NO: 81)
CDR2 (SEQ ID NO: 77) - AASSLQS

CDR2: SISSSSSYIYYADSVKG
CDR3 (SEQ ID NO: 87) - QQGVSFPRT

(SEQ ID NO: 82)

CDR3- ARGAPXGAAAGWFDP (SEQ ID

NO: 111)(non-Kabat) or

GAPXGAAAGWFDP (SEQ ID NO: 112),

wherein X is M, L, I, V, Q, or F

NKG2D
QVQLVESGGGLVKPGGSLRLSCAASGFT
QSALTQPASVSGSPGQSITISCSGSSS

binder in U.S.
FSSYGMHWVRQAPGKGLEWVAFIRYDG
NIGNNAVNWYQQLPGKAPKLLIYY

Pat. No.
SNKYYADSVKGRFTISRDNSKNTLYLQM
DDLLPSGVSDRFSGSKSGTSAFLAIS

9,273,136
NSLRAEDTAVYYCAKDRGLGDGTYFDY
GLQSEDEADYYCAAWDDSLNGPVF

WGQGTTVTVSS (SEQ ID NO: 113)
GGGTKLTVL (SEQ ID NO: 114)

NKG2D
QVHLQESGPGLVKPSETLSLTCTVSDDSI
EIVLTQSPGTLSLSPGERATLSCRASQ

binder in U.S.
SSYYWSWIRQPPGKGLEWIGHISYSGSA
SVSSSYLAWYQQKPGQAPRLLIYGA

Pat. No.
NYNPSLKSRVTISVDTSKNQFSLKLS
SVTSSRATGIPDRFSGSGSGTDFTLTISRL

7,879,985
AADTAVYYCANWDDAFNIWGQGTMVT
EPEDFAVYYCQQYGSSPWTFGQGTK

VSS (SEQ ID NO: 116)
VEIK (SEQ ID NO: 117)

In certain embodiments, the first antigen-binding site that binds NKG2D (e.g., human NKG2D) comprises an antibody heavy chain variable domain (VH) that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the VH of an antibody disclosed in Table 1, and an antibody light chain variable domain (VL) that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the VL of the same antibody disclosed in Table 1. In certain embodiments, the first antigen-binding site comprises the heavy chain CDR1, CDR2, and CDR3 and the light chain CDR1, CDR2, and CDR3, determined under Kabat (see Kabat et al., (1991) Sequences of Proteins of Immunological Interest, NIH Publication No. 91-3242, Bethesda), Chothia (see, e.g., Chothia C & Lesk A M, (1987), J. Mol. Biol. 196: 901-917), MacCallum (see MacCallum R M et al., (1996) J. Mol. Biol. 262: 732-745), or any other CDR determination method known in the art, of the VH and VL sequences of an antibody discloses in Table 1. In certain embodiments, the first antigen-binding site comprises the heavy chain CDR1, CDR2, and CDR3 and the light chain CDR1, CDR2, and CDR3 of an antibody disclosed in Table 1.

In certain embodiments, the first antigen-binding site that binds to NKG2D comprises a heavy chain variable domain derived from SEQ ID NO:1, such as by having an amino acid sequence at least 90% (e.g., at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:1, and/or incorporating amino acid sequences identical to the CDR1 (SEQ ID NO:2), CDR2 (SEQ ID NO:3), and CDR3 (SEQ ID NO:4) sequences of SEQ ID NO:1. The heavy chain variable domain related to SEQ ID NO:1 can be coupled with a variety of light chain variable domains to form an NKG2D binding site. For example, the first antigen-binding site that incorporates a heavy chain variable domain related to SEQ ID NO:1 can further incorporate a light chain variable domain selected from the sequences derived from SEQ ID NOs: 5, 6, 7, 8, 9, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, and 46. For example, the first antigen-binding site incorporates a heavy chain variable domain with amino acid sequences at least 90% (e.g., at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:1 and a light chain variable domain with amino acid sequences at least 90% (e.g., at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to any one of the sequences selected from SEQ ID NOs: 5, 6, 7, 8, 9, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, and 46.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:26, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:32. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 27 or 28, 29, and 30 or 31, respectively (e.g., SEQ ID NOs: 27, 29, and 30, respectively, or SEQ ID NOs: 28, 29, and 31, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 33, 34, and 35, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 27 or 28, 29, and 30 or 31, respectively (e.g., SEQ ID NOs: 27, 29, and 30, respectively, or SEQ ID NOs: 28, 29, and 31, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 33, 34, and 35, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:36, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:42. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 37 or 38, 39, and 40 or 41, respectively (e.g., SEQ ID NOs: 37, 39, and 40, respectively, or SEQ ID NOs: 38, 39, and 41, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 43, 44, and 45, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 37 or 38, 39, and 40 or 41, respectively (e.g., SEQ ID NOs: 37, 39, and 40, respectively, or SEQ ID NOs: 38, 39, and 41, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 43, 44, and 45, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:47, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:49. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 27, 29, and 48, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 50, 34, and 51, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 27, 29, and 48, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 50, 34, and 51, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:52, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:58. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 53 or 54, 55, and 56 or 57, respectively (e.g., SEQ ID NOs: 53, 55, and 56, respectively, or SEQ ID NOs: 54, 55, and 57, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 59, 60, and 61, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 53 or 54, 55, and 56 or 57, respectively (e.g., SEQ ID NOs: 53, 55, and 56, respectively, or SEQ ID NOs: 54, 55, and 57, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 59, 60, and 61, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:62, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:68. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 63 or 64, 65, and 66 or 67, respectively (e.g., SEQ ID NOs: 63, 65, and 66, respectively, or SEQ ID NOs: 64, 65, and 67, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 59, 60, and 69, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 63 or 64, 65, and 66 or 67, respectively (e.g., SEQ ID NOs: 63, 65, and 66, respectively, or SEQ ID NOs: 64, 65, and 67, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 59, 60, and 69, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:89, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:92. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 53 or 54, 55, and 90 or 91, respectively (e.g., SEQ ID NOs: 53, 55, and 90, respectively, or SEQ ID NOs: 54, 55, and 91, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 93, 44, and 94, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 53 or 54, 55, and 90 or 91, respectively (e.g., SEQ ID NOs: 53, 55, and 90, respectively, or SEQ ID NOs: 54, 55, and 91, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 93, 44, and 94, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:70, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:75. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 71 or 115, 72, and 73 or 74, respectively (e.g., SEQ ID NOs: 71, 72, and 73, respectively, or SEQ ID NOs: 115, 72, and 74, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 76, 77, and 78, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 71 or 115, 72, and 73 or 74, respectively (e.g., SEQ ID NOs: 71, 72, and 73, respectively, or SEQ ID NOs: 115, 72, and 74, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 76, 77, and 78, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:79, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:85. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 83 or 84, respectively (e.g., SEQ ID NOs: 80, 82, and 83, respectively, or SEQ ID NOs: 81, 82, and 84, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 83 or 84 respectively (e.g., SEQ ID NOs: 80, 82, and 83, respectively, or SEQ ID NOs: 81, 82, and 84, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:95, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:85. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 96 or 97, respectively (e.g., SEQ ID NOs: 80, 82, and 96, respectively, or SEQ ID NOs: 81, 82, and 97, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 96 or 97, respectively (e.g., SEQ ID NOs: 80, 82, and 96, respectively, or SEQ ID NOs: 81, 82, and 97, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:98, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:85. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 99 or 100, respectively (e.g., SEQ ID NOs: 80, 82, and 99, respectively, or SEQ ID NOs: 81, 82, and 100, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 99 or 100, respectively (e.g., SEQ ID NOs: 80, 82, and 99, respectively, or SEQ ID NOs: 81, 82, and 100, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:101, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:85. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 102 or 103, respectively (e.g., SEQ ID NOs: 80, 82, and 102, respectively, or SEQ ID NOs: 81, 82, and 103, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 102 or 103, respectively (e.g., SEQ ID NOs: 80, 82, and 102, respectively, or SEQ ID NOs: 81, 82, and 103, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:104, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:85. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 105 or 106, respectively (e.g., SEQ ID NOs: 80, 82, and 105, respectively, or SEQ ID NOs: 81, 82, and 106, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 105 or 106, respectively (e.g., SEQ ID NOs: 80, 82, and 105, respectively, or SEQ ID NOs: 81, 82, and 106, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:107, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:85. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 108 or 109, respectively (e.g., SEQ ID NOs: 80, 82, and 108, respectively, or SEQ ID NOs: 81, 82, and 109, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 108 or 109, respectively (e.g., SEQ ID NOs: 80, 82, and 108, respectively, or SEQ ID NOs: 81, 82, and 109, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively.

In certain embodiments, the first antigen-binding site that binds NKG2D comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:110, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:85. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 111 or 112, respectively (e.g., SEQ ID NOs: 80, 82, and 111, respectively, or SEQ ID NOs: 81, 82, and 112, respectively). In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively. In certain embodiments, the first antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 80 or 81, 82, and 111 or 112, respectively (e.g., SEQ ID NOs: 80, 82, and 111, respectively, or SEQ ID NOs: 81, 82, and 112, respectively); and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 86, 77, and 87, respectively.

The multispecific binding proteins can bind to NKG2D-expressing cells, which include but are not limited to NK cells, γδ T cells and CD8⁺ αβ T cells. Upon NKG2D binding, the multispecific binding proteins may block natural ligands, such as ULBP6 and MICA, from binding to NKG2D and activating NK cells.

The multispecific binding proteins binds to cells expressing CD16, an Fc receptor on the surface of leukocytes including natural killer cells, macrophages, neutrophils, eosinophils, mast cells, and follicular dendritic cells. A protein of the present disclosure binds to NKG2D with an affinity of K_Dof 2 nM to 120 nM, e.g., 2 nM to 110 nM, 2 nM to 100 nM, 2 nM to 90 nM, 2 nM to 80 nM, 2 nM to 70 nM, 2 nM to 60 nM, 2 nM to 50 nM, 2 nM to 40 nM, 2 nM to 30 nM, 2 nM to 20 nM, 2 nM to 10 nM, about 15 nM, about 14 nM, about 13 nM, about 12 nM, about 11 nM, about 10 nM, about 9 nM, about 8 nM, about 7 nM, about 6 nM, about 5 nM, about 4.5 nM, about 4 nM, about 3.5 nM, about 3 nM, about 2.5 nM, about 2 nM, about 1.5 nM, about 1 nM, between about 0.5 nM to about 1 nM, about 1 nM to about 2 nM, about 2 nM to 3 nM, about 3 nM to 4 nM, about 4 nM to about 5 nM, about 5 nM to about 6 nM, about 6 nM to about 7 nM, about 7 nM to about 8 nM, about 8 nM to about 9 nM, about 9 nM to about 10 nM, about 1 nM to about 10 nM, about 2 nM to about 10 nM, about 3 nM to about 10 nM, about 4 nM to about 10 nM, about 5 nM to about 10 nM, about 6 nM to about 10 nM, about 7 nM to about 10 nM, or about 8 nM to about 10 nM. In some embodiments, NKG2D-binding sites bind to NKG2D with a K_Dof 10 to 62 nM.

CLEC12A Binding Site

The tumor associated antigen-binding site of the multispecific binding protein disclosed herein comprises a heavy chain variable domain and a light chain variable domain. Table 2 lists some exemplary sequences of heavy chain variable domains and light chain variable domains that, in combination, can bind to these tumor associated antigens.

In one aspect, the present disclosure provides multispecific binding proteins that bind to the NKG2D receptor and CD16 receptor on natural killer cells, and CLEC12A. Table 2 lists some exemplary sequences of heavy chain variable domains and light chain variable domains that, in combination, can bind to CLEC12A. CDR sequences are identified under Kabat numbering as indicated.

TABLE 2

Sequences of Exemplary Antigen-Binding Sites that Bind CLEC12A.

Clone
VH
VL

16B8.C8
EVQLQESGPGLVQPSQSLSITCT
DIQMNQSPSSLSASLGDTIAITCHA

VSGFSLTNYGLHWVRQSPGKG
SQNINFWLSWYQQKPGNIPKLLIY

LEWLGVIWSGGKTDYNTPFKS
EASNLHTGVPSRFSGSGSGTRFTLT

RLSISKDISKNQVFFKMNSLQP
ISSLQPEDIATYYCQQSHSYPLTFG

NDTAIYFCAKYDYDDSLDYWG
QGTKLEIK

QGTSVTVSS
[SEQ ID NO: 136]

[SEQ ID NO: 135]
CDR1: HASQNINFWLS [SEQ ID

CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8 in
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKIPKLLIY

scFv-1292 /
EWIGVIWSGGKTDYNPSLKSRV
EASNLHTGVPSRFSGSGSGTRFTLT

scFv-1301
TISKDTSKNQFSLKLSSVQAND
ISSLQPEDIATYYCQQSHSYPLTFG

(back
TAVYYCAKYDYDDSLDYWGQ
QGTKLEIK

mutations in
GTLVTVSS
[SEQ ID NO: 144]

VH and VL
[SEQ ID NO: 143]
CDR1: HASQNINFWLS [SEQ ID

underlined)
CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

scFv of
scFv -1292 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIG

16B8.C8
VIWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQANDTAVYYCAK

scFv-1292 /
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMT

scFv-1301
QSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNL

HTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKL

EIK [SEQ ID NO: 145]

scFv -1301 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIY

EASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETLS

LTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRV

TISKDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVT

VSS

[SEQ ID NO: 146]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKIPKLLIY

in scFv-1293/
EWIGVIWSGGKTDYNPSLKSRV
EASNLHTGVPSRFSGSGSGTRFTLT

scFv-1302
TISVDTSKNQFSLKLSSVQAND
ISSLQPEDIATYYCQQSHSYPLTFG

(back
TAVYYCAKYDYDDSLDYWGQ
QGTKLEIK

mutations in
GTLVTVSS
[SEQ ID NO: 144]

VH and VL
[SEQ ID NO: 147]
CDR1: HASQNINFWLS [SEQ ID

underlined)
CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

scFv of
scFv -1293 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIG

16B8.C8
VIWSGGKTDYNPSLKSRVTISVDTSKNQFSLKLSSVQANDTAVYYCAK

scFv-1293/
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMT

scFv-1302
QSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNL

HTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKL

EIK

[SEQ ID NO: 149]

scFv -1302 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIY

EASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETLS

LTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRV

TISVDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVT

VSS

[SEQ ID NO: 150]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKIPKLLIY

in scFv-1294 /
EWIGVIWSGGKTDYNPSLKSRV
EASNLHTGVPSRFSGSGSGTRFTLT

scFv-1303
TISKDTSKNQFSLKLSSVTANDT
ISSLQPEDIATYYCQQSHSYPLTFG

(back
AVYYCAKYDYDDSLDYWGQG
QGTKLEIK

mutations in
TLVTVSS
[SEQ ID NO: 144]

VH and VL
[SEQ ID NO: 244]
CDR1: HASQNINFWLS [SEQ ID

underlined)
CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

scFv of
scFv -1294 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIG

16B8.C8
VIWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVTANDTAVYYCAK

scFv-1294 /
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMT

scFv-1303
QSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNL

HTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKL

EIK

[SEQ ID NO: 153]

scFv -1303 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIY

EASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETLS

LTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRV

TISKDTSKNQFSLKLSSVTANDTAVYYCAKYDYDDSLDYWGQGTLVT

VSS

[SEQ ID NO: 154]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKIPKLLIY

in scFv-1295 /
EWIGVIWSGGKTDYNPSLKSRV
EASNLHTGVPSRFSGSGSGTRFTLT

scFv-1304
TISKDTSKNQFSLKLSSVQAAD
ISSLQPEDIATYYCQQSHSYPLTFG

(back
TAVYYCAKYDYDDSLDYWGQ
QGTKLEIK

mutations in
GTLVTVSS
[SEQ ID NO: 144]

VH and VL
[SEQ ID NO: 178]
CDR1: HASQNINFWLS [SEQ ID

underlined)
CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

scFv of
scFv -1295 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIG

16B8.C8
VIWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQAADTAVYYCAK

scFv- 1295 /
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMT

scFv-1304
QSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNL

HTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKL

EIK

[SEQ ID NO: 156]

scFv -1304 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIY

EASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETLS

LTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRV

TISKDTSKNQFSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLVT

VSS

[SEQ ID NO: 157]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKIPKLLIY

in scFv-1296 /
EWIGVIWSGGKTDYNPSLKSRV
EASNLHTGVPSRFSGSGSGTRFTLT

scFv-1305
TISKDTSKNQFSLKLSSVQAND
ISSLQPEDIATYYCQQSHSYPLTFG

(back
TAVYYCARYDYDDSLDYWGQ
QGTKLEIK

mutations in
GTLVTVSS
[SEQ ID NO: 144]

VH and VL
[SEQ ID NO: 256]
CDR1: HASQNINFWLS [SEQ ID

underlined)
CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

scFv of
scFv -1296 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIG

16B8.C8
VIWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQANDTAVYYCAR

scFv-1296 /
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMT

scFv-1305
QSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNL

HTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKL

EIK

[SEQ ID NO: 160]

scFv -1305 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIY

EASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETLS

LTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRV

TISKDTSKNQFSLKLSSVQANDTAVYYCARYDYDDSLDYWGQGTLVT

VSS

[SEQ ID NO: 161]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKAPKLLIY

in scFv-1297 /
EWIGVIWSGGKTDYNPSLKSRV
EASNLHTGVPSRFSGSGSGTRFTLT

scFv-1306
TISKDTSKNQFSLKLSSVQAND
ISSLQPEDIATYYCQQSHSYPLTFG

(back
TAVYYCAKYDYDDSLDYWGQ
QGTKLEIK

mutations in
GTLVTVSS
[SEQ ID NO: 163]

VH and VL
[SEQ ID NO: 143]
CDR1: HASQNINFWLS [SEQ ID

underlined)
CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

scFv of
scFv -1297 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIG

16B8.C8
VIWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQANDTAVYYCAK

scFv-1297 /
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMT

scFv-1306
QSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASN

LHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTK

LEIK

[SEQ ID NO: 164]

scFv -1306 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLI

YEASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTF

GCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETL

SLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSR

VTISKDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLV

TVSS

[SEQ ID NO: 165]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKIPKLLIY

in scFv-1298 /
EWIGVIWSGGKTDYNPSLKSRV
EASNLHTGVPSRFSGSGSGTDFTLT

scFv-1307
TISKDTSKNQFSLKLSSVQAND
ISSLQPEDIATYYCQQSHSYPLTFG

(back
TAVYYCAKYDYDDSLDYWGQ
QGTKLEIK

mutations in
GTLVTVSS
[SEQ ID NO: 167]

VH and VL
[SEQ ID NO: 143]
CDR1: HASQNINFWLS [SEQ ID

underlined)
CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

scFv of
scFv -1298 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIG

16B8.C8
VIWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQANDTAVYYCAK

scFv-1298 /
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMT

scFv-1307
QSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNL

HTGVPSRFSGSGSGTDFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKL

EIK

[SEQ ID NO: 168]

scFv -1307 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIY

EASNLHTGVPSRFSGSGSGTDFTLTISSLQPEDIATYYCQQSHSYPLTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETLS

LTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRV

TISKDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVT

VSS

[SEQ ID NO: 169]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKIPKLLIY

in scFv-1299 /
EWIGVIWSGGKTDYNPSLKSRV
EASNLHTGVPSRFSGSGSGTRFTLT

scFv-1308
TISKDTSKNQFSLKLSSVQAND
ISSLQPEDFATYYCQQSHSYPLTFG

TAVYYCAKYDYDDSLDYWGQ
QGTKLEIK

GTLVTVSS
[SEQ ID NO: 171]

[SEQ ID NO: 143]
CDR1: HASQNINFWLS [SEQ ID

CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

scFv of
scFv -1299 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIG

16B8.C8
VIWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQANDTAVYYCAK

scFv-1299 /
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMT

scFv-1308
QSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNL

HTGVPSRFSGSGSGTRFTLTISSLQPEDFATYYCQQSHSYPLTFGCGTKL

EIK

[SEQ ID NO: 172]

scFv -1308 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIY

EASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDFATYYCQQSHSYPLTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETLS

LTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRV

TISKDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVT

VSS

[SEQ ID NO: 173]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKAPKLLIY

in scFv-1300/
EWIGVIWSGGKTDYNPSLKSRV
EASNLHTGVPSRFSGSGSGTDFTLT

scFv-1309
TISVDTSKNQFSLKLSSVTAADT
ISSLQPEDFATYYCQQSHSYPLTFG

AVYYCARYDYDDSLDYWGQG
QGTKLEIK

TLVTVSS
[SEQ ID NO: 175]

[SEQ ID NO: 174]
CDR1: HASQNINFWLS [SEQ ID

CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

scFv of
scFv -1300 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIG

16B8.C8
VIWSGGKTDYNPSLKSRVTISVDTSKNQFSLKLSSVTAADTAVYYCAR

scFv-1300/
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMT

scFv-1309
QSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASN

LHTGVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQQSHSYPLTFGCGTK

LEIK

[SEQ ID NO: 176]

scFv -1309 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLI

YEASNLHTGVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQQSHSYPLTF

GCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETL

SLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSR

VTISVDTSKNQFSLKLSSVTAADTAVYYCARYDYDDSLDYWGQGTLV

TVSS

[SEQ ID NO: 177]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKAPKLLIY

in scFv-1602/
EWIGVIWSGGKTDYNPSLKSRV
EASNLHTGVPSRFSGSGSGTRFTLT

scFv-2061
TISKDTSKNQFSLKLSSVQAAD
ISSLQPEDIATYYCQQSHSYPLTFG

TAVYYCAKYDYDDSLDYWGQ
GGTKLEIK

GTLVTVSS [SEQ ID NO: 178]
[SEQ ID NO: 289]

CDR1: GFSLTNY [SEQ ID
CDR1: HASQNINFWLS [SEQ ID

NO: 137]
NO: 140]

CDR2: WSGGK [SEQ ID NO: 138]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR3: YDYDDSLDY [SEQ ID
CDR3: QQSHSYPLT [SEQ ID

NO: 139]
NO: 142]

scFv of
scFv -1602 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIG

16B8.C8
VIWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQAADTAVYYCAK

scFv-1602/
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMT

scFv-2061
QSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASN

LHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTK

LEIK

[SEQ ID NO: 180]

scFv-2061 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLI

YEASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTF

GCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETL

SLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSR

VTISKDTSKNQFSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLV

TVSS

[SEQ ID NO: 181]

Humanized
QLQLQESGPGLVKPSETLSLTCT
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8
VSGFSLTNYGLHWIRQPPGKGL
SQNINFWLSWYQQKPGKX₁PKLLI

consensus
EWIGVIWSGGKTDYNPSLKSRV
YEASNLHTGVPSRFSGSGSGTX₂FT

TISX₁DTSKNQFSLKLSSVX₂AX₃
LTISSLQPEDX₃ATYYCQQSHSYPLT

DTAVYYCAX₄YDYDDSLDYWG
FGX₄GTKLEIK

QGTLVTVSS
where X₁ is A or I, X₂ is D or R, X₃ is F

where X₁ is V or K, X₂ is T or Q,
or I, and X₄ is Q or G

X₃ is A or N, and X₄ is R or K
[SEQ ID NO: 151]

[SEQ ID NO: 182]
CDR1: HASQNINFWLS [SEQ ID

CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: WSGGK [SEQ ID NO: 138]
CDR3: QQSHSYPLT [SEQ ID

CDR3: YDYDDSLDY [SEQ ID
NO: 142]

NO: 139]

Humanized
QLQLQESGPGLVKPSETLSLTC
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8 in
TVSGFSLTNYGLHWIRQPPGKG
SQNINFWLSWYQQKPGKAPKLLIY

AB0305/
LEWIGVIWVGGATDYNPSLKS
EASNLHTGVPSRFSGSGSGTDFTLT

AB5030
RVTISVDTSKNQFSLKLSSVQA
ISSLQPEDFATYYCQQSHSYPLTFG

(Cysteine
ADTAVYYCAKGDYGDTLDYW
SGTKLEIK

heterodimeri-
GQGTLVTVSS
[SEQ ID NO: 271]

zation
[SEQ ID NO: 270]
DIQMTQSPSSLSASVGDRVTITCHA

mutations are
QLQLQESGPGLVKPSETLSLTCT
SQNINFWLSWYQQKPGKAPKLLIY

underlined.
VSGFSLTNYGLHWIRQPPGKCL
EASNLHTGVPSRFSGSGSGTDFTLT

Such
EWIGVIWVGGATDYNPSLKSRV
ISSLQPEDFATYYCQQSHSYPLTFG

mutations can
TISVDTSKNQFSLKLSSVQAAD

CGTKLEIK

facilitate
TAVYYCAKGDYGDTLDYWGQ
[SEQ ID NO: 336]

formation of a
GTLVTVSS
CDR1: HASQNINFWLS [SEQ ID

disulfide
[SEQ ID NO: 335]
NO: 140]

bridge
CDR1: GFSLTNY [SEQ ID
CDR2: EASNLHT [SEQ ID NO: 141]

between the
NO: 137]
CDR3: QQSHSYPLT [SEQ ID

VH and VL
CDR2: WVGGA [SEQ ID NO: 272]
NO: 142]

of the scFv.)
CDR3: GDYGDTLDY [SEQ ID

NO: 273]

scFv of
AB0305 (VH-VL):

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWI

16B8.C8
GVIWVGGATDYNPSLKSRVTISVDTSKNQFSLKLSSVQAADTAVYYC

AB0305/
AKGDYGDTLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQ

AB5030
MTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEA

SNLHTGVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQQSHSYPLTFGCG

TKLEIK

[SEQ ID NO: 274]

AB5030 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLI

YEASNLHTGVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQQSHSYPLTF

GCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETL

SLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWVGGATDYNPSLKSR

VTISVDTSKNQFSLKLSSVQAADTAVYYCAKGDYGDTLDYWGQGTL

VTVSS

[SEQ ID NO: 275]

Humanized
QLQLQESGPGLVKPSETLSLTC
DIQMTQSPSSLSASVGDRVTITCHA

16B8.C8 in
TVSGFSLTNYGLHWIRQPPGKG
SQNINFWLSWYQQKPGKAPKLLIY

AB0147/
LEWIGVILSGGWTDYNPSLKSR
EASNLHTGVPSRFSGSGSGTRFTLT

AB7410
VTISKDTSKNQFSLKLSSVQAA
ISSLQPEDIATYYCQQSHSYPLTFGS

DTAVYYCAKGDYGDALDYWG
GTKLEIK

QGTLVTVSS
[SEQ ID NO: 295]

[SEQ ID NO: 294]
CDR1: HASQNINFWLS [SEQ ID

CDR1: GFSLTNY [SEQ ID
NO: 140]

NO: 137]
CDR2: EASNLHT [SEQ ID NO: 141]

CDR2: LSGGW [SEQ ID NO: 296]
CDR3: QQSHSYPLT [SEQ ID

CDR3: GDYGDALDY [SEQ ID
NO: 142]

NO: 279]

scFv of
AB0147(VH-VL)

humanized
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWI

16B8.C8
GVILSGGWTDYNPSLKSRVTISKDTSKNQFSLKLSSVQAADTAVYYC

AB0147/
AKGDYGDALDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQ

AB7410
MTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEA

SNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCG

TKLEIK

[SEQ ID NO: 280]

AB7410 (VL-VH)

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLI

YEASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTF

GCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGPGLVKPSETL

SLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVILSGGWTDYNPSLKSR

VTISKDTSKNQFSLKLSSVQAADTAVYYCAKGDYGDALDYWGQGTL

VTVSS

[SEQ ID NO: 200]

9F11.B7
EVQLQESGGGLVQPGGSRKLS
DIKMTQSPSSMYASLGERVTITCK

CAASGFTFNSFGMHWVRQAPE
ASQDIYNYLSWFQLKPGKSPRPLIY

KGLEWVAFISSGSTSIYYANTV
RANILVSGVPSKFSGSGSGQDYSLT

KGRFTISRDNPKNTLFLQMTSL
INSLEYEDLGIYYCLQFDAFPFTFG

RSEDTAMYYCARDGYPTGGA
SGTKLEIK

MDYWGQGTSVTVSS [SEQ ID
[SEQ ID NO: 194]

NO: 193]
CDR1: KASQDIYNYLS [SEQ ID

CDR1: GFTFNSF [SEQ ID
NO: 198]

NO: 192]
CDR2: RANILVS [SEQ ID NO: 199]

CDR2: SSGSTS [SEQ ID NO: 196]
CDR3: LQFDAFPFT [SEQ ID

CDR3: DGYPTGGAMDY [SEQ
NO: 201]

ID NO: 187]

Humanized

EVQLVESGGGVVQPGGSLRLSC
DIQMTQSPSSLSASVGDRVTITCKA

9F11.B7 in
AASGFTFNSFGMHWVRQAPGK
SQDIYNYLSWFQQKPGKAPKPLIY

AB0191 /
GLEWVAFISSGSTSIYYANTVK
RANILVSGVPSRFSGSGSGQDYTFT

AB0185
GRFTISRDNSKNTLYLQMNSLR
ISSLQPEDIATYYCLQFDAFPFTFGS

(back
AEDTAVYYCARDGYPTGGAMD
GTKLEIK

mutations in
YWGQGTSVTVSS
[SEQ ID NO: 203]

VH and VL
[SEQ ID NO: 162]
CDR1: KASQDIYNYLS [SEQ ID

underlined)
CDR1: GFTFNSF [SEQ ID
NO: 198]

NO: 192]
CDR2: RANILVS [SEQ ID NO: 199]

CDR2: SSGSTS [SEQ ID NO: 196]
CDR3: LQFDAFPFT [SEQ ID

CDR3: DGYPTGGAMDY [SEQ
NO: 201]

ID NO: 187]

scFv of
AB0191 (VH-VL):

humanized
EVQLVESGGGVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKCLE

9F11.B7
WVAFISSGSTSIYYANTVKGRFTISRDNSKNTLYLQMNSLRAEDTAVY

AB0191 /
YCARDGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSGGGG

AB0185
SDIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLI

YRANILVSGVPSRFSGSGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTF

GCGTKLEIK

[SEQ ID NO: 184]

AB0185 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIY

RANILVSGVPSRFSGSGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGGGVVQPGGSL

RLSCAASGFTFNSFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKG

RFTISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQ

GTSVTVSS

[SEQ ID NO: 185]

Humanized

EVQLVESGGGVVQPGGSLRLSC
DIQMTQSPSSLSASVGDRVTITCKA

9F11.B7 in
AASGFTFNAFGMHWVRQAPG
SQDIYNYLSWFQQKPGKAPKPLIY

AB0192/
KGLEWVAFISSGSTSIYYANTV
RANILVSGVPSRFSGSGSGQDYTFT

AB0186
KGRFTISRDNSKNTLYLQMNSL
ISSLQPEDIATYYCLQFDAFPFTFGS

(back
RAEDTAVYYCARDGYPTGGAM
GTKLEIK

mutations in
DYWGQGTSVTVSS
[SEQ ID NO: 203]

VH and VL
[SEQ ID NO: 148]
CDR1: KASQDIYNYLS [SEQ ID

underlined)
CDR1: GFTFNAF [SEQ ID
NO: 198]

NO: 195]
CDR2: RANILVS [SEQ ID NO: 199]

CDR2: SSGSTS [SEQ ID NO: 196]
CDR3: LQFDAFPFT [SEQ ID

CDR3: DGYPTGGAMDY [SEQ
NO: 201]

ID NO: 187]

scFv of
AB0192 (VH-VL):

humanized
EVQLVESGGGVVQPGGSLRLSCAASGFTFNAFGMHWVRQAPGKCLE

9F11.B7
WVAFISSGSTSIYYANTVKGRFTISRDNSKNTLYLQMNSLRAEDTAVY

AB0192/
YCARDGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSGGGG

AB0186
SDIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLI

YRANILVSGVPSRFSGSGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTF

GCGTKLEIK

[SEQ ID NO: 204]

AB0186 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIY

RANILVSGVPSRFSGSGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGGGVVQPGGSL

RLSCAASGFTFNAFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVK

GRFTISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWG

QGTSVTVSS

[SEQ ID NO: 205]

Humanized

EVQLVESGGGVVQPGGSLRLSC
DIQMTQSPSSLSASVGDRVTITCKA

9F11.B7 in
AASGFTFNSFGMHWVRQAPGK
SQDIYNYLSWFQQKPGKAPKPLIY

AB0193/
GLEWVAFISSGSTSIYYANTVK
RANILVSGVPSRFSGSGSGQDYTFT

AB0187
GRFTISRDNSKNTLYLQMNSLR
ISSLQPEDIATYYCLQFDAFPFTFGS

(back
AEDTAVYYCARSGYPTGGAMD
GTKLEIK

mutations in
YWGQGTSVTVSS
[SEQ ID NO: 203]

VH and VL
[SEQ ID NO: 210]
CDR1: KASQDIYNYLS [SEQ ID

underlined)
CDR1: GFTFNSF [SEQ ID
NO: 198]

NO: 192]
CDR2: RANILVS [SEQ ID NO: 199]

CDR2: SSGSTS [SEQ ID NO: 196]
CDR3: LQFDAFPFT [SEQ ID

CDR3: SGYPTGGAMDY [SEQ ID
NO: 201]

NO: 213]

scFv of
AB0193 (VH-VL):

humanized
EVQLVESGGGVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKCLE

9F11.B7
WVAFISSGSTSIYYANTVKGRFTISRDNSKNTLYLQMNSLRAEDTAVY

AB0193/
YCARSGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSGGGG

AB0187
SDIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLI

YRANILVSGVPSRFSGSGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTF

GCGTKLEIK

[SEQ ID NO: 208]

AB0187 (VL-VH):

DIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIY

RANILVSGVPSRFSGSGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGGGVVQPGGSL

RLSCAASGFTFNSFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKG

RFTISRDNSKNTLYLQMNSLRAEDTAVYYCARSGYPTGGAMDYWGQ

GTSVTVSS

[SEQ ID NO: 209]

Humanized

EVQLVESGGGVVQPGGSLRLSC
DIKMTQSPSSLSASVGDRVTITCKA

9F11.B7 in
AASGFTFNSFGMHWVRQAPGK
SQDIYNYLSWFQQKPGKAPKPLIY

AB0194/
GLEWVAFISSGSTSIYYANTVK
RANILVSGVPSRFSGSGSGQDYTLT

AB0188
GRFTISRDNSKNTLYLQMNSLR
ISSLQPEDIATYYCLQFDAFPFTFGS

(back
AEDTAVYYCARDGYPTGGAMD
GTKLEIK

mutations in
YWGQGTSVTVSS
[SEQ ID NO: 218]

VH and VL
[SEQ ID NO: 162]
CDR1: KASQDIYNYLS [SEQ ID

underlined)
CDR1: GFTFNSF [SEQ ID
NO: 198]

NO: 192]
CDR2: RANILVS [SEQ ID NO: 199]

CDR2: SSGSTS [SEQ ID NO: 196]
CDR3: LQFDAFPFT [SEQ ID

CDR3: DGYPTGGAMDY [SEQ
NO: 201]

ID NO: 187]

scFv of
AB0194 (VH-VL):

humanized
EVQLVESGGGVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKCLE

9F11.B7
WVAFISSGSTSIYYANTVKGRFTISRDNSKNTLYLQMNSLRAEDTAVY

AB0194/
YCARDGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSGGGG

AB0188
SDIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLI

YRANILVSGVPSRFSGSGSGQDYTLTISSLQPEDIATYYCLQFDAFPFTF

GCGTKLEIK

[SEQ ID NO: 215]

AB0188 (VL-VH):

DIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIY

RANILVSGVPSRFSGSGSGQDYTLTISSLQPEDIATYYCLQFDAFPFTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGGGVVQPGGSL

RLSCAASGFTFNSFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKG

RFTISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQ

GTSVTVSS

[SEQ ID NO: 216]

Humanized

EVQLVESGGGVVQPGGSLRLSC
DIKMTQSPSSLSASVGDRVTITCKA

9F11.B7 in
AASGFTFNAFGMHWVRQAPG
SQDIYNYLSWFQQKPGKAPKPLIY

AB0195/
KGLEWVAFISSGSTSIYYANTV
RANILVSGVPSRFSGSGSGQDYTLT

AB0189
KGRFTISRDNSKNTLYLQMNSL
ISSLQPEDIATYYCLQFDAFPFTFGS

(back
RAEDTAVYYCARDGYPTGGAM
GTKLEIK

mutations in
DYWGQGTSVTVSS [SEQ ID
[SEQ ID NO: 218]

VH and VL
NO: 148]
CDR1: KASQDIYNYLS [SEQ ID

underlined)
CDR1: GFTFNAF [SEQ ID
NO: 198]

NO: 195]
CDR2: RANILVS [SEQ ID NO: 199]

CDR2: SSGSTS [SEQ ID NO: 196]
CDR3: LQFDAFPFT [SEQ ID

CDR3: DGYPTGGAMDY [SEQ
NO: 201]

ID NO: 187]

scFv of
AB0195 (VH-VL):

humanized
EVQLVESGGGVVQPGGSLRLSCAASGFTFNAFGMHWVRQAPGKCLE

9F11.B7
WVAFISSGSTSIYYANTVKGRFTISRDNSKNTLYLQMNSLRAEDTAVY

AB0195/
YCARDGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSGGGG

AB0189
SDIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLI

YRANILVSGVPSRFSGSGSGQDYTLTISSLQPEDIATYYCLQFDAFPFTF

GCGTKLEIK

[SEQ ID NO: 252]

AB0189 (VL-VH):

DIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIY

RANILVSGVPSRFSGSGSGQDYTLTISSLQPEDIATYYCLQFDAFPFTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGGGVVQPGGSL

RLSCAASGFTFNAFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVK

GRFTISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWG

QGTSVTVSS

[SEQ ID NO: 253]

Humanized

EVQLVESGGGVVQPGGSLRLSC
DIKMTQSPSSLSASVGDRVTITCKA

9F11.B7 in
AASGFTFNSFGMHWVRQAPGK
SQDIYNYLSWFQQKPGKAPKPLIY

AB0196/
GLEWVAFISSGSTSIYYANTVK
RANILVSGVPSRFSGSGSGQDYTLT

AB0190
GRFTISRDNSKNTLYLQMNSLR
ISSLQPEDIATYYCLQFDAFPFTFGS

(back
AEDTAVYYCARSGYPTGGAMD
GTKLEIK

mutations in
YWGQGTSVTVSS
[SEQ ID NO: 218]

VH and VL
[SEQ ID NO: 210]
CDR1: KASQDIYNYLS [SEQ ID

underlined)
CDR1: GFTFNSF [SEQ ID
NO: 198]

NO: 192]
CDR2: RANILVS [SEQ ID NO: 199]

CDR2: SSGSTS [SEQ ID NO: 196]
CDR3: LQFDAFPFT [SEQ ID

CDR3: SGYPTGGAMDY [SEQ ID
NO: 201]

NO: 213]

scFv of
AB0196 (VH-VL):

humanized
EVQLVESGGGVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKCLE

9F11.B7
WVAFISSGSTSIYYANTVKGRFTISRDNSKNTLYLQMNSLRAEDTAVY

AB0196/
YCARSGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSGGGG

AB0190
SDIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLI

YRANILVSGVPSRFSGSGSGQDYTLTISSLQPEDIATYYCLQFDAFPFTF

GCGTKLEIK

[SEQ ID NO: 254]

AB0190 (VL-VH):

DIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIY

RANILVSGVPSRFSGSGSGQDYTLTISSLQPEDIATYYCLQFDAFPFTFG

CGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGGGVVQPGGSL

RLSCAASGFTFNSFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKG

RFTISRDNSKNTLYLQMNSLRAEDTAVYYCARSGYPTGGAMDYWGQ

GTSVTVSS

[SEQ ID NO: 255]

Humanized
EVQLVESGGGVVQPGGSLRLSC
DIX₁MTQSPSSLSASVGDRVTITCK

9F11.B7
AASGFTFNX₁FGMHWVRQAPG
ASQDIYNYLSWFQQKPGKAPKPLI

consensus
KGLEWVAFISSGSTSIYYANTV
YRANILVSGVPSRFSGSGSGQDYT

KGRFTISRDNSKNTLYLQMNSL
X₂TISSLQPEDIATYYCLQFDAFPFT

RAEDTAVYYCARX₂GYPTGGA
FGSGTKLEIK

MDYWGQGTSVTVSS
where X₁ is Q or K and X₂ is F or L

where X₁ is S or A and X₂ is D or S
[SEQ ID NO: 250]

[SEQ ID NO: 249]
CDR1: KASQDIYNYLS [SEQ ID

CDR1: GFTFNXF where X is S or
NO: 198]

A [SEQ ID NO: 251]
CDR2: RANILVS [SEQ ID NO: 199]

CDR2: SSGSTS [SEQ ID NO: 196]
CDR3: LQFDAFPFT [SEQ ID

CDR3: XGYPTGGAMDY where
NO: 201]

X is D or S [SEQ ID NO: 246]

30A9.E9
EVQLQESGPGLVQPSQSLSITCT
DIVMTQSPSSLAVTAGEKVTMRCK

VSGFSLTSFGVHWVRQSPGKG
SSQSLLWNVNQNNYLLWYQQKQ

LEWLGVIWSGGSTDSNAAFISR
GQPPKLLIYGASIRESWVPDRFTGS

LTITKDNSKSQVFFKMNSLQAT
GSGTDFTLTISNVHVEDLAVYYCQ

DTAIYYCARSYFAMDYWGQGT
HNHGSFLPYTFGGGTKLEIK

SVSVSS
[SEQ ID NO: 248]

[SEQ ID NO: 247]
CDR1: KSSQSLLWNVNQNNYLL

CDR1: GFSLTSF [SEQ ID
[SEQ ID NO: 240]

NO: 221]
CDR2: GASIRES [SEQ ID NO: 226]

CDR2: WSGGS [SEQ ID NO: 166]
CDR3: QHNHGSFLPYT [SEQ ID

CDR3: SYFAMDY [SEQ ID
NO: 179]

NO: 223]

23A5.H8
QVQLRQSGPGLVQPSQSLSITC
GIVMTQSPSSLAVTAGEKVTMRCK

TVSGFSLTSYGVHWVRQSPGK
SSQSLLWSVNQNNYLLWYQQKQG

GLEWLGVMWSGGSTDYNAAF
QPPKLLIYGASIRQSWVPDRFTGSG

MSRLSISKDNSKSQVFFTMNSL
SGTDFTLSISNVHAEDLAVYYCQH

QADDTAIYYCARTHFGMDYW
NHGSFLPYTFGGGTKLEIK

GQGTPVTVSS
[SEQ ID NO: 243]

[SEQ ID NO: 242]
CDR1: KSSQSLLWSVNQNNYLL

CDR1: GFSLTSY [SEQ ID
[SEQ ID NO: 245]

NO: 206]
CDR2: GASIRQS [SEQ ID NO: 239]

CDR2: WSGGS [SEQ ID NO: 166]
CDR3: QHNHGSFLPYT [SEQ ID

CDR3: THFGMDY [SEQ ID
NO: 179]

NO: 241]

20D6.H8
EVQLQESGPGLVQPSQSLSITCT
GIVMTQSPSSLAVTAGEKVTMRCK

VSGFSLTSFGIHWVRQSPGKGL
SSQSLLWNVNQNNYLVWYQQKQ

EWLGVIWSGGNTDSNAAFISRL
GQPPKLLIYGASIRESWVPDRFTGS

SITKDISKSQVFFKMNSLQVTD
GSGTDFTLTISNVHAEDLAVYYCQ

TAIYYCARSYFAMDYWGQGTS
HNHGSFLPYTFGGGTKLEIK

VTVSS
[SEQ ID NO: 237]

[SEQ ID NO: 238]
CDR1: KSSQSLLWNVNQNNYLV

CDR1: GFSLTSF [SEQ ID
[SEQ ID NO: 152]

NO: 221]
CDR2: GASIRES [SEQ ID NO: 226]

CDR2: WSGGN [SEQ ID NO: 236]
CDR3: QHNHGSFLPYT [SEQ ID

CDR3: SYFAMDY [SEQ ID
NO: 179]

NO: 223]

15A10.G8
EVQLQESGAELVRSGASIKLSC
EVLLTQSPAIIAASPGEKVTITCSAR

AASAFNIKDYFIHWVRQRPDQ
SSVSYMSWYQQKPGSSPKIWIYGI

GLEWIGWIDPENDDTEYAPKFQ
SKLASGVPARFSGSGSGTYFSFTIN

DKATMTADTSSNTAYLQLSSLT
NLEAEDVATYYCQQRSYYPFTFGS

SADTAVYYCNALWSRGGYFD
GTKLEIK

YWGQGTTLTVSS
[SEQ ID NO: 158]

[SEQ ID NO: 155]
CDR1: SARSSVSYMS [SEQ ID

CDR1: AFNIKDY [SEQ ID
NO: 186]

NO: 159]
CDR2: GISKLAS [SEQ ID NO: 188]

CDR2: DPENDD [SEQ ID
CDR3: QQRSYYPFT [SEQ ID

NO: 170]
NO: 189]

CDR3: LWSRGGYFDY [SEQ ID

NO: 183]

13E1.A4
EVQLQESGPELEKPGASVRISC
SVLMTQTPLSLPVSLGDRASISCRS

KASGYSFTAYNMNWVKQSNG
SQGIVHINGNTYLEWYLQKPGQSP

KSLEWIGNIDPSYGDATYNQKF
KLLIYKVSNRFSGVPDRFSGSGSGT

KGKATLTVDKSSSTAYMQLKS
DFTLKISRVEAEDLGVYYCFQGSH

LTSEDSAVYYCARDNYYGSGY
VPWTFGGGTKLEIK

FDYWGQGTTLTVSS
[SEQ ID NO: 191]

[SEQ ID NO: 190]
CDR1: RSSQGIVHINGNTYLE [SEQ

CDR1: GYSFTAY [SEQ ID
ID NO: 211]

NO: 197]
CDR2: KVSNRFS [SEQ ID NO: 212]

CDR2: DPSYGD [SEQ ID
CDR3: FQGSHVPWT [SEQ ID

NO: 202]
NO: 214]

CDR3: DNYYGSGYFDY [SEQ ID

NO: 207]

12F8.H7
EVQLQESGAELVRSGASVKLSC
GIVMTQAPLTLSVTIGQPASISCKS

TVSGFNIKDYYMHWVKQRPEQ
SQSLLDSDGKTFLNWFLQRPGQSP

GLEWIGWIDPENGDTENVPKFQ
KRLISLVSKLDSGVPDRFTGSGSGT

GKATMTADTSSNTAYLQLRSL
DFTLKLSRVEPEDLGVYYCWQGT

TSEDTAVYYCKSYYYDSSSRY
HFPYTFGGGTKLEIK

VDVWGAGTTVTVSS
[SEQ ID NO: 219]

[SEQ ID NO: 217]
CDR1: KSSQSLLDSDGKTFLN [SEQ

CDR1: GFNIKDY [SEQ ID
ID NO: 224]

NO: 220]
CDR2: LVSKLDS [SEQ ID NO: 225]

CDR2: DPENGD [SEQ ID
CDR3: WQGTHFPYT [SEQ ID

NO: 222]
NO: 227]

CDR3: YYYDSSSRYVDV [SEQ

ID NO: 257]

9E4.B7
EVQLQESGAELMKPGASVKISC
GIVMTQSPASLSASVGETVTITCRA

RTTGYTFSTYWIEWVKQRPGR
GENIHSYLAWYQQKQGKSPQLLV

GPEWIGELFPGNSDTTLNEKFT
YNAKTLAEGVPSRFSGSGSGTQFS

GKATFTADSSSNTAYMQLSSLT
LKINSLQPEDFGSYYCQHHYGTPR

SEDSAVYYCARSGYYGSSLDY
TFGGGTKLEIK

WGQGTTLTVSS
[SEQ ID NO: 229]

[SEQ ID NO: 228]
CDR1: RAGENIHSYLA [SEQ ID

CDR1: GYTFSTY [SEQ ID
NO: 233]

NO: 230]
CDR2: NAKTLAE [SEQ ID NO: 234]

CDR2: FPGNSD [SEQ ID NO: 231]
CDR3: QHHYGTPRT [SEQ ID

CDR3: SGYYGSSLDY [SEQ ID
NO: 235]

NO: 232]

In certain embodiments, the second antigen-binding site that binds CLEC12A (e.g., human CLEC12A) comprises an antibody heavy chain variable domain (VH) that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the VH of an antibody disclosed in Table 2 , and an antibody light chain variable domain (VL) that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the VL of the same antibody disclosed in Table 2 . In certain embodiments, the second antigen-binding site comprises the heavy chain CDR1, CDR2, and CDR3 and the light chain CDR1, CDR2, and CDR3, determined under Kabat (see Kabat et al., (1991) Sequences of Proteins of Immunological Interest, NIH Publication No. 91-3242, Bethesda), Chothia (see, e.g., Chothia C & Lesk A M, (1987), J Mol Biol 196: 901-917), MacCallum (see MacCallum R M et al., (1996) J Mol Biol 262: 732-745), or any other CDR determination method known in the art, of the VH and VL sequences of an antigen-binding site disclosed in Table 2. In certain embodiments, the second antigen-binding site comprises the heavy chain CDR1, CDR2, and CDR3 and the light chain CDR1, CDR2, and CDR3 of an antibody disclosed in Table 2.

In certain embodiments, the second antigen-binding site is derived from 16B8.C8. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:135, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:136. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively.

In certain embodiments, the second antigen-binding site is derived from scFv-1292 or scFv-1301. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:143, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:144. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:145 or 146.

In certain embodiments, the second antigen-binding site is derived from scFv-1293 or scFv-1302. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:147, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:144. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:149 or 150.

In certain embodiments, the second antigen-binding site is derived from scFv-1294 or scFv-1303. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:244, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:144. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:153 or 154.

In certain embodiments, the second antigen-binding site is derived from scFv-1295 or scFv-1304. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:178, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:144. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:156 or 157.

In certain embodiments, the second antigen-binding site is derived from scFv-1296 or scFv-1305. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:256, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:144. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:160 or 161.

In certain embodiments, the second antigen-binding site is derived from scFv-1297 or scFv-1306. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:143, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:163. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:164 or 165.

In certain embodiments, the second antigen-binding site is derived from scFv-1298 or scFv-1307. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:143, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:167. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:168 or 169.

In certain embodiments, the second antigen-binding site is derived from scFv-1299 or scFv-1308. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:143, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:171. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:172 or 173.

In certain embodiments, the second antigen-binding site is derived from scFv-1300 or scFv-1309. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:174, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:175. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:176 or 177.

In certain embodiments, the second antigen-binding site is derived from scFv-1602 or scFv-2061. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:178, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:289. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:180 or 181.

In certain embodiments, the second antigen-binding site is derived from humanized 16B.C8. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:182, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:151. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 138, and 139, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively.

In certain embodiments, the second antigen-binding site is derived from AB0305 or AB5030. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:270, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:271. In certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:335, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:336. In certain embodiments, the second antigen-binding site comprises a VH and a VL each comprises a cysteine heterodimerization mutation. Such heterodimerization mutation can facilitate formation of a disulfide bridge between the VH and VL of the scFv. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 272, and 273, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 272, and 273, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:274 or 275.

In certain embodiments, the second antigen-binding site is derived from AB0147 or AB7410. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:294, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:295. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 296, and 279, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 137, 296, and 279, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 140, 141, and 142, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:280 or 200.

In certain embodiments, the second antigen-binding site is derived from 9F11.B7. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:193, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:194. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 187, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 187, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively.

In certain embodiments, the second antigen-binding site is derived from AB0191 or AB0185. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:162, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:203. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 187, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 187, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:184 or 185.

In certain embodiments, the second antigen-binding site is derived from AB0192 or AB0186. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:148, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:203. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 195, 196, and 187, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 195, 196, and 187, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:204 or 205.

In certain embodiments, the second antigen-binding site is derived from AB0193 or AB0187. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:210, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:203. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 213, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 213, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:208 or 209.

In certain embodiments, the second antigen-binding site is derived from AB0194 or AB0188. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:162, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:218. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 187, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and187, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:215 or 216.

In certain embodiments, the second antigen-binding site is derived from AB0195 or AB0189. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:148, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:218. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 195, 196, and 187, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 195, 196, and187, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:252 or 253.

In certain embodiments, the second antigen-binding site is derived from AB0196 or AB0190. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:210, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:218. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 213, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 192, 196, and 213, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site is present as an scFv, wherein the scFv comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:254 or 255.

In certain embodiments, the second antigen-binding site is derived from humanized 9F11.B7. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:249, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:250. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 251, 196, and 246, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 251, 196, and 246, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 198, 199, and 201, respectively.

In certain embodiments, the second antigen-binding site is derived from 30A9.E9. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:247, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:248. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 221, 166, and 223, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 240, 226, and 179, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 221, 166, and 223, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 240, 226, and 179, respectively.

In certain embodiments, the second antigen-binding site is derived from 23A5.H8. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:242, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:243. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 206, 166, and 241, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 245, 239, and 179, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 206, 166, and 241, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 245, 239, and 179, respectively.

In certain embodiments, the second antigen-binding site is derived from 20D6.H8. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:238, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:237. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 221, 236, and 223, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 152, 226, and 179, respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 221, 236, and 223, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 152, 226, and 179, respectively.

In certain embodiments, the second antigen-binding site is derived from 15A10.G8. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:155, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:158. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 159, 170, and 183, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 186, 188, and189, and respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 159, 170, and 183, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 186, 188, and 189, respectively.

In certain embodiments, the second antigen-binding site is derived from 13E1.A4. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:190, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:191. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 197, 202, and 207, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 211, 212, and 214, and respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 197, 202, and 207, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 211, 212, and 214, respectively.

In certain embodiments, the second antigen-binding site is derived from 12F8.H7. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:217, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:219. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 220, 222, and 257, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 224, 225, and 227, and respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 220, 222, and 257, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 224, 225, and 227, respectively.

In certain embodiments, the second antigen-binding site is derived from 9E4.B7. For example, in certain embodiments, the second antigen-binding site comprises a VH that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to the amino acid sequence of SEQ ID NO:228, and a VL that comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to SEQ ID NO:229. In certain embodiments, the VH comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 230, 231, and 232, respectively. In certain embodiments, the VL comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 233, 234, and 235, and respectively. In certain embodiments, the second antigen-binding site comprises (a) a VH that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 230, 231, and 232, respectively; and (b) a VL that comprises CDR1, CDR2, and CDR3 comprising the amino acid sequences of SEQ ID NOs: 233, 234, and 235, respectively.

In certain embodiments, the second antigen-binding site that binds CLEC12A is an scFv. For example, in certain embodiments, the second antigen-binding site comprises the amino acid sequence of SEQ ID NO:145, 146, 149, 150, 153, 154, 156, 157, 160, 161, 164, 165, 168, 169, 172, 173, 176, 177, 180, 181, 184, 185, 204, 205, 208, 209, 215, 216, 252, 253, 254, 255, 274, 275, 280, or 281. In specific embodiments, the second antigen-binding site comprises the amino acid sequence of SEQ ID NO:180.

Alternatively, novel antigen-binding sites that can bind to CLEC12A can be identified by screening for binding to the amino acid sequence defined by SEQ ID NO:258, a variant thereof, a mature extracellular fragment thereof or a fragment containing a domain of CLEC12A.

SEQ ID NO: 258

MSEEVTYADLQFQNSSEMEKIPEIGKFGEKAPPAPSHVWRPAALFLTLLC

LLLLIGLGVLASMFHVTLKIEMKKMNKLQNISEELQRNISLQLMSNMNIS

NKIRNLSTTLQTIATKLCRELYSKEQEHKCKPCPRRWIWHKDSCYFLSDD

VQTWQESKMACAAQNASLLKINNKNALEFIKSQSRSYDYWLGLSPEEDST

RGMRVDNIINSSAWVIRNAPDLNNMYCGYINRLYVQYYHCTYKKRMICEK

MANPVQLGSTYFREA (Uniprot ID No. Q5QGZ9)

It is contemplated that in an scFv, a VH and a VL can be connected by a linker, e.g., (GlyGlyGlyGlySer)₄i.e. (G₄S)₄linker (SEQ ID NO:119). A skilled person in the art would appreciate that any of the other disclosed linkers (see, e.g., Table 10) may be used in an scFv having a VH and VL sequences disclosed herein (e.g., in Table 2).

In each of the foregoing embodiments, it is contemplated herein that the scFv, VH and/or VL sequences that bind CLEC12A may contain amino acid alterations (e.g., at least 1, 2, 3, 4, 5, or 10 amino acid substitutions, deletions, or additions) in the framework regions of the VH and/or VL without affecting their ability to CLEC12A. For example, it is contemplated herein that scFv, VH and/or VL sequences that bind CLEC12A may contain cysteine heterodimerization mutations, facilitating formation of a disulfide bridge between the VH and VL of the scFv.

In certain embodiments, the second antigen-binding site disclosed herein binds human CLEC12A with a K_D(i.e., dissociation constant) of 1 nM or lower, 5 nM or lower, 10 nM or lower, 15 nM or lower, or 20 nM or lower, as measured by surface plasmon resonance (SPR) (e.g., using the method described in Example 12) or by bio-layer interferometry (BLI), and/or binds CLEC12A from a body fluid, tissue, and/or cell of a subject. In certain embodiments, a second antigen-binding site disclosed herein has a K_d(i.e., off-rate, also called K_off) equal to or lower than 1×10⁻⁵, 1×10⁻⁴, 1×10⁻³, 5×10⁻³, 0.01, 0.02, or 0.05 l/s, as measured by SPR (e.g., using the method described in Example 12) or by BLI.

In certain embodiments, the second antigen-binding site derived from 15A10.G8 or 20D6.A8 binds cynomolgus CLEC12A with a K_D(i.e., dissociation constant) of 5 nM or lower, 10 nM or lower, 15 nM or lower, 20 nM or lower, or 30 nM or lower, as measured by surface plasmon resonance (SPR) (e.g., using the method described in Example 12) or by bio-layer interferometry (BLI), and/or binds CLEC12A from a body fluid, tissue, and/or cell of a subject. In certain embodiments, a second antigen-binding site disclosed herein has a K_d(i.e., off-rate, also called K_off) equal to or lower than 1×10⁻³, 5×10⁻³, 0.01, 0.02, or 0.03 l/s, as measured by SPR (e.g., using the method described in Example 12) or by BLI.

Variations in the glycosylation status of CLEC12A on the surface of different cell types has been reported (Marshall et al., (2006) Eur J lmmunol. 36(8):2159-69). CLEC12A expressed on the surface of AML cells from different patients may have different glycosylation patterns as well. Moreover, branched glycans can restrict accessibility to the protein component of CLEC12A, limiting diversity of available epitopes. Certain antigen-binding sites or the present application can overcome the glycosylation variations. In certain embodiments, a second antigen-binding site disclosed herein, e.g., an antigen-binding site derived from 16B8.C8, a humanized 16B8.C8, 9F11.B7, or a humanized 9F11.B7, binds CLEC12A (e.g., human CLEC12A) in a glycosylation independent manner, i.e., binds both a glycosylated CLEC12A and a de-glycosylated CLEC12A. In certain embodiments, the ratio of the K_Dat which the second antigen-binding site binds deglycosylated CLEC12A to the K_Dat which the second antigen-binding site binds glycosylated CLEC12A is within the range of 1:10 to 10:1 (e.g., within the range of 1:5 to 5:1, 1:3 to 3:1, or 1:2 to 2:1). In certain embodiments, the ratio is about 1:5, about 1:3, about 1:2, about 1:1.5, about 1:1, about 1.5:1, about 2:1, about 3:1, or about 5:1. In certain embodiments, the ratio is about 1:1. In another aspect, the instant disclosure provides a multispecific binding protein comprising (a) a first antigen-binding site that binds NKG2D; (b) a second antigen-binding site that binds CLEC12A; and (c) an antibody Fc domain or a portion thereof sufficient to bind CD16, or a third antigen-binding site that binds CD16; wherein the second antigen-binding site binds CLEC12A in a glycosylation independent manner.

CLEC12A containing a K244Q substitution is a polymorphic variant prevalent in about 30% of the human population. In some embodiments, the second antigen-binding site of a multispecific binding protein disclosed herein binds CLEC12A-K244Q. In certain embodiments, the ratio of the K_Dat which the second antigen-binding site binds CLEC12A-K244Q to the K_Dat which the second antigen-binding site binds wild-type CLEC12A is within the range of 1:2 to 2:1. In certain embodiments, the ratio is about 1:2, about 1:1.5, about 1:1, about 1.5:1, or about 2:1. In certain embodiments, the ratio is about 1:1.

In some embodiments, the multispecific binding proteins described herein bind to CLEC12A with an isoelectric point (pI) of about 8 to about 9. In some embodiments, the multispecific binding proteins described herein bind to CLEC12A with a pI of about 8.5 to about 8.9. In certain embodiments, the multispecific binding proteins described herein bind to CLEC12A with a pI of about 8.7.

In some embodiments, the F3′ format of a CLEC12A TriNKET described herein (e.g., hF3′-1602 TriNKET), has EC50 about 2-fold lower (e.g., about 1-fold lower, about 2-fold lower, about 3-fold lower, about 4 fold lower, about 5-fold lower, about 6-fold lower) than that of the corresponding F3 format TriNKET, e.g., AB0010, in binding to human CLEC12A on the surface of a cell, e.g., a cancer cell. In some embodiments, the F3′ format of a CLEC12A TriNKET described herein (e.g., hF3′-1602 TriNKET), has EC50 about 50% lower (e.g., about 25% lower, about 30% lower, about 40% lower, about 50% lower, about 55% lower, about 60% lower, about 70% lower, about 75% lower) than that of the corresponding F3 format TriNKET, e.g., AB0010, in binding to human CLEC12A on the surface of a cell, e.g., a cancer cell.

In certain embodiments, the second antigen-binding site competes for binding to CLEC12A with a corresponding antigen-binding site described above. In one aspect, the instant disclosure provides a multispecific binding protein comprising (a) a first antigen-binding site that binds NKG2D; (b) a second antigen-binding site that binds CLEC12A; and (c) an antibody Fc domain or a portion thereof sufficient to bind CD16, or a third antigen-binding site that binds CD16; wherein the second antigen-binding site competes with an antigen-binding site that binds CLEC12A as disclosed herein. In certain embodiments, the second antigen-binding site competes with an antigen-binding site derived from 16B8.C8 disclosed above for binding to CLEC12A. In one embodiment, the second antigen-binding site competes with 16B8.C8 for binding to CLEC12A. In certain embodiments, the second antigen-binding site competes with an antigen-binding site derived from a humanized 16B8.C8 disclosed above for binding to CLEC12A. In one embodiment, the second antigen-binding site competes with a humanized 16B8.C8 for binding to CLEC12A. In certain embodiments, the second antigen-binding site competes with an antigen-binding site derived from 9F11.B7 disclosed above for binding to CLEC12A. In one embodiment, the second antigen-binding site competes with 9F11.B7 for binding to CLEC12A. In certain embodiments, the second antigen-binding site competes with an antigen-binding site derived from a humanized 9F11.B7 disclosed above for binding to CLEC12A. In one embodiment, the second antigen-binding site competes with a humanized 9F11.B7 for binding to CLEC12A. In certain embodiments, the second antigen-binding site competes with an antigen-binding site derived from 12F8.H7, 13E1.A4, 15A10.E8, 20D6.H8, or 23A5.H8 disclosed above for binding to CLEC12A. In some embodiments, the second antigen-binding site competes with 12F8.H7, 13E1.A4, 15A10.E8, 20D6.H8, or 23A5.H8 for binding to CLEC12A.

Fc Domain

Within the Fc domain, CD16 binding is mediated by the hinge region and the CH2 domain. For example, within human IgG1, the interaction with CD16 is primarily focused on amino acid residues Asp 265-Glu 269, Asn 297-Thr 299, Ala 327-Ile 332, Leu 234-Ser 239, and carbohydrate residue N-acetyl-D-glucosamine in the CH2 domain (see, Sondermann et al., Nature, 406 (6793):267-273). Based on the known domains, mutations can be selected to enhance or reduce the binding affinity to CD16, such as by using phage-displayed libraries or yeast surface-displayed cDNA libraries, or can be designed based on the known three-dimensional structure of the interaction. Accordingly, in certain embodiment, the antibody Fc domain or the portion thereof comprises a hinge and a CH2 domain.

The assembly of heterodimeric antibody heavy chains can be accomplished by expressing two different antibody heavy chain sequences in the same cell, which may lead to the assembly of homodimers of each antibody heavy chain as well as assembly of heterodimers. Promoting the preferential assembly of heterodimers can be accomplished by incorporating different mutations in the CH3 domain of each antibody heavy chain constant region as shown in U.S. Ser. Nos. 13/494,870, 16/028,850, 11/533,709, 12/875,015, 13/289,934, 14/773,418, 12/811,207, 13/866,756, 14/647,480, and 14/830,336. For example, mutations can be made in the CH3 domain based on human IgG1 and incorporating distinct pairs of amino acid substitutions within a first polypeptide and a second polypeptide that allow these two chains to selectively heterodimerize with each other. The positions of amino acid substitutions illustrated below are all numbered according to the EU index as in Kabat.

In one scenario, an amino acid substitution in the first polypeptide replaces the original amino acid with a larger amino acid, selected from arginine (R), phenylalanine (F), tyrosine (Y) or tryptophan (W), and at least one amino acid substitution in the second polypeptide replaces the original amino acid(s) with a smaller amino acid(s), chosen from alanine (A), serine (S), threonine (T), or valine (V), such that the larger amino acid substitution (a protuberance) fits into the surface of the smaller amino acid substitutions (a cavity). For example, one polypeptide can incorporate a T366W substitution, and the other can incorporate three substitutions including T366S, L368A, and Y407V.

An antibody heavy chain variable domain described in the application can optionally be coupled to an amino acid sequence at least 90% identical to an antibody constant region, such as an IgG constant region including hinge, CH2 and CH3 domains with or without CH1 domain. In some embodiments, the amino acid sequence of the constant region is at least 90% identical to a human antibody constant region, such as a human IgG1 constant region, an IgG2 constant region, IgG3 constant region, or IgG4 constant region. In one embodiment, the antibody Fc domain or a portion thereof sufficient to bind CD16 comprises an amino acid sequence at least 90% (e.g., at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) identical to wild-type human IgG1 Fc sequence DKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYV DGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTIS KAKGQPREPQVYTLPPSRDELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPP VLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG (SEQ ID NO:118). In some other embodiments, the amino acid sequence of the constant region is at least 90% identical to an antibody constant region from another mammal, such as rabbit, dog, cat, mouse, or horse.

In some embodiments, the antibody constant domain linked to the scFv or the Fab fragment is able to bind to CD16. In some embodiments, the protein incorporates a portion of an antibody Fc domain (for example, a portion of an antibody Fc domain sufficient to bind CD16), wherein the antibody Fc domain comprises a hinge and a CH2 domain (for example, a hinge and a CH2 domain of a human IgG1 antibody), and/or amino acid sequences at least 90% identical to amino acid sequence 234-332 of a human IgG antibody.

One or more mutations can be incorporated into the constant region as compared to human IgG1 constant region, for example at Q347, Y349, L351, S354, E356, E357, K360, Q362, S364, T366, L368, K370, N390, K392, T394, D399, S400, D401, F405, Y407, K409, T411 and/or K439. Exemplary substitutions include, for example, Q347E, Q347R, Y349S, Y349K, Y349T, Y349D, Y349E, Y349C, T350V, L351K, L351D, L351Y, S354C, E356K, E357Q, E357L, E357W, K360E, K360W, Q362E, S364K, S364E, S364H, S364D, T366V, T366I, T366L, T366M, T366K, T366W, T366S, L368E, L368A, L368D, K370S, N390D, N390E, K392L, K392M, K392V, K392F, K392D, K392E, T394F, T394W, D399R, D399K, D399V, S400K, S400R, D401K, F405A, F405T, Y407A, Y407I, Y407V, K409F, K409W, K409D, K409R, T411D, T411E, K439D, and K439E.

In certain embodiments, mutations that can be incorporated into the CH1 of a human IgG1 constant region may be at amino acid V125, F126, P127, T135, T139, A140, F170, P171, and/or V173. In certain embodiments, mutations that can be incorporated into the Cκ of a human IgG1 constant region may be at amino acid E123, F116, S176, V163, S174, and/or T164.

Alternatively, amino acid substitutions could be selected from the following sets of substitutions shown in Table 3.

TABLE 3

First Polypeptide
Second Polypeptide

Set 1
S364E/F405A
Y349K/T394F

Set 2
S364H/D401K
Y349T/T411E

Set 3
S364H/T394F
Y349T/F405A

Set 4
S364E/T394F
Y349K/F405A

Set 5
S364E/T411E
Y349K/D401K

Set 6
S364D/T394F
Y349K/F405A

Set 7
S364H/F405A
Y349T/T394F

Set 8
S364K/E357Q
L368D/K370S

Set 9
L368D/K370S
S364K

Set 10
L368E/K370S
S364K

Set 11
K360E/Q362E
D401K

Set 12
L368D/K370S
S364K/E357L

Set 13
K3705
S364K/E357Q

Set 14
F405L
K409R

Set 15
K409R
F405L

Alternatively, amino acid substitutions could be selected from the following sets of substitutions shown in Table 4.

TABLE 4

First Polypeptide
Second Polypeptide

Set 1
K409W
D399V/F405T

Set 2
Y349S
E357W

Set 3
K360E
Q347R

Set 4
K360E/K409W
Q347R/D399V/F405T

Set 5
Q347E/K360E/K409W
Q347R/D399V/F405T

Set 6
Y349S/K409W
E357W/D399V/F405T

Alternatively, amino acid substitutions could be selected from the following sets of substitutions shown in Table 5.

TABLE 5

First Polypeptide
Second Polypeptide

Set 1
T366K/L351K
L351D/L368E

Set 2
T366K/L351K
L351D/Y349E

Set 3
T366K/L351K
L351D/Y349D

Set 4
T366K/L351K
L351D/Y349E/L368E

Set 5
T366K/L351K
L351D/Y349D/L368E

Set 6
E356K/D399K
K392D/K409D

Alternatively, at least one amino acid substitution in each polypeptide chain could be selected from Table 6.

TABLE 6

First Polypeptide
Second Polypeptide

L351Y, D399R, D399K, S400K,
T366V, T366I, T366L, T366M,

S400R, Y407A, Y407I, Y407V
N390D, N390E, K392L,

K392M, K392V, K392F

K392D, K392E, K409F,

K409W, T411D and T411E

Alternatively, at least one amino acid substitution could be selected from the following sets of substitutions in Table 7, where the position(s) indicated in the First Polypeptide column is replaced by any known negatively-charged amino acid, and the position(s) indicated in the Second Polypeptide Column is replaced by any known positively-charged amino acid.

TABLE 7

First Polypeptide
Second Polypeptide

K392, K370, K409, or K439
D399, E356, or E357

Alternatively, at least one amino acid substitution could be selected from the following set in Table 8, where the position(s) indicated in the First Polypeptide column is replaced by any known positively-charged amino acid, and the position(s) indicated in the Second Polypeptide Column is replaced by any known negatively-charged amino acid.

TABLE 8

First Polypeptide
Second Polypeptide

D399, E356, or E357
K409, K439, K370, or K392

Alternatively, amino acid substitutions could be selected from the following sets in Table 9.

TABLE 9

First Polypeptide
Second Polypeptide

T350V, L351Y, F405A, and Y407V
T350V, T366L, K392L, and T394W

Alternatively, or in addition, the structural stability of a hetero-multimeric protein may be increased by introducing S354C on either of the first or second polypeptide chain, and Y349C on the opposing polypeptide chain, which forms an artificial disulfide bridge within the interface of the two polypeptides.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at position T366, and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of T366, L368 and Y407.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of T366, L368 and Y407, and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at position T366.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of E357, K360, Q362, 5364, L368, K370, T394, D401, F405, and T411 and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of Y349, E357, S364, L368, K370, T394, D401, F405 and T411.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of Y349, E357, S364, L368, K370, T394, D401, F405 and T411 and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of E357, K360, Q362, S364, L368, K370, T394, D401, F405, and T411.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of L351, D399, S400 and Y407 and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of T366, N390, K392, K409 and T411.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of T366, N390, K392, K409 and T411 and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of L351, D399, S400 and Y407.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of Q347, Y349, K360, and K409, and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of Q347, E357, D399 and F405.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of Q347, E357, D399 and F405, and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of Y349, K360, Q347 and K409.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of K370, K392, K409 and K439, and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of D356, E357 and D399.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of D356, E357 and D399, and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of K370, K392, K409 and K439.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of L351, E356, T366 and D399, and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of Y349, L351, L368, K392 and K409.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of Y349, L351, L368, K392 and K409, and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region at one or more positions selected from the group consisting of L351, E356, T366 and D399.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by an S354C substitution and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by a Y349C substitution.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by a Y349C substitution and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by an S354C substitution.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by K360E and K409W substitutions and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by Q347R, D399V and F405T substitutions.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by Q347R, D399V and F405T substitutions and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by K360E and K409W substitutions.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by a T366W substitutions and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by T366S, T368A, and Y407V substitutions.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by T366S, T368A, and Y407V substitutions and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by a T366W substitution.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by T350V, L351Y, F405A, and Y407V substitutions and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by T350V, T366L, K392L, and T394W substitutions.

In some embodiments, the amino acid sequence of one polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by T350V, T366L, K392L, and T394W substitutions and wherein the amino acid sequence of the other polypeptide chain of the antibody constant region differs from the amino acid sequence of an IgG1 constant region by T350V, L351Y, F405A, and Y407V substitutions.

Exemplary Multispecific Binding Proteins

Listed below are examples of TriNKETs comprising an antigen-binding site that binds CLEC12A and an antigen-binding site that binds NKG2D each linked to an antibody constant region, wherein the antibody constant regions include mutations that enable heterodimerization of two Fc chains.

Exemplary CLEC12A-targeting TriNKETs are contemplated in the F3′ and F3 formats. As described above, in the F3′ format, the antigen-binding site that binds the tumor-associated antigen is an scFv and the antigen-binding site that binds NKG2D is a Fab. In the F3 format, the antigen-binding site that binds the tumor-associated antigen is a Fab, and the antigen-binding site that binds NKG2D is an scFv. In each TriNKET, the scFv may comprise substitution of Cys in the VH and VL regions, facilitating formation of a disulfide bridge between the VH and VL of the scFv.

The VH and VL of the scFv can be connected via a linker, e.g., a peptide linker. In certain embodiments, the peptide linker is a flexible linker. Regarding the amino acid composition of the linker, peptides are selected with properties that confer flexibility, do not interfere with the structure and function of the other domains of the proteins described in the present application, and resist cleavage from proteases. For example, glycine and serine residues generally provide protease resistance. In certain embodiments, the VL is linked N-terminal or C-terminal to the VH via a (GlyGlyGlyGlySer)₄((G₄S)₄) linker (SEQ ID NO:119).

The length of the linker (e.g., flexible linker) can be “short,” e.g., 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12 amino acid residues, or “long,” e.g., at least 13 amino acid residues. In certain embodiments, the linker is 10-50, 10-40, 10-30, 10-25, 10-20, 15-50, 15-40, 15-30, 15-25, 15-20, 20-50, 20-40, 20-30, or 20-25 amino acid residues in length.

In certain embodiments, the linker comprises or consists of a (GS)_n(SEQ ID NO:120), (GGS)_n(SEQ ID NO:121), (GGGS)_n(SEQ ID NO:122), (GGSG)_n(SEQ ID NO:123), (GGSGG)_n(SEQ ID NO:124), and (GGGGS)_n(SEQ ID NO:125) sequence, wherein n is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20. In certain embodiments, the linker comprises or consists of an amino acid sequence selected from SEQ ID NO:119 and SEQ ID NO:126-134, as listed in Table 10.

TABLE 10

SEQ ID
Amino Acid Sequence

SEQ ID NO: 126
GSGSGSGSGSGSGSGSGSGS

SEQ ID NO: 127
GGSGGSGGSGGSGGSGGSGGSGGSGGSGGS

SEQ ID NO: 128
GGGSGGGSGGGSGGGSGGGSGGGSGGGSGG

GSGGGSGGGS

SEQ ID NO: 129
GGSGGGSGGGSGGGSGGGSGGGSGGGSGGG

SGGGSGGGSG

SEQ ID NO: 130
GGSGGGGSGGGGSGGGGSGGGGSGGGGSGG

GGSGGGGSGGGGSGGGGSGG

SEQ ID NO: 131
GGGGSGGGGSGGGGSGGGGSGGGGSGGGGS

GGGGSGGGGSGGGGSGGGGS

SEQ ID NO: 119
GGGGSGGGGSGGGGSGGGGS

SEQ ID NO: 132
GGGGSGGGGSGGGGS

SEQ ID NO: 133
GGGGSGGGGSGGGGSGGGGSGGGGSGGGGS

GGGGSGGGGSGGGGSGGGGSGGGGSGGGGS

GGGGSGGGGSGGGGSGGGGSGGGGSGGGGS

GGGGSGGGGS

SEQ ID NO: 134
GGSGGGGSGGGGSGGGGSGGGGSGGGGSGG

GGSGGGGSGGGGSGGGGSGGGGSGGGGSGG

GGSGGGGSGGGGSGGGGSGGGGSGGGGSGG

GGSGGGGSGG

In the F3′-TriNKETs, the tumor-associated antigen-binding scFv is linked to the N-terminus of an Fc via a Gly-Ser linker. The Ala-Ser or Gly-Ser linker is included at the elbow hinge region sequence to balance between flexibility and optimal geometry. In certain embodiments, an additional sequence Thr-Lys-Gly can be added N-terminal or C-terminal to the Ala-Ser or Gly-Ser sequence at the hinge.

As used herein to describe these exemplary TriNKETs, an Fc includes an antibody hinge, CH2, and CH3. In each exemplary TriNKET, the Fc domain linked to an scFv comprises the mutations of Q347R, D399V, and F405T, and the Fc domain linked to a Fab comprises matching mutations K360E and K409W for forming a heterodimer. The Fc domain linked to the scFv further includes an S354C substitution in the CH3 domain, which forms a disulfide bond with a Y349C substitution on the Fc linked to the Fab. These substitutions are bold in the sequences described in this subsection.

For example, a TriNKET of the present disclosure is F3′-1602 TriNKET, also referred to herein as “hF3′-1602 TriNKET” and “AB0237”. F3′-1602 TriNKET includes (a) a CLEC12A-binding scFv sequence derived from scFv-1602 of Table 2, in the orientation of VH positioned N-terminal to VL, linked to an Fc domain and (b) an NKG2D-binding Fab fragment derived from A49MI including a heavy chain portion comprising a heavy chain variable domain and a CH1 domain, and a light chain portion comprising a light chain variable domain and a light chain constant domain, wherein the CH1 domain is connected to the Fc domain. F3′-1602 TriNKET includes three polypeptides scFv-1602-VH-VL-Fc, A49MI-VH-CH1-Fc, and A49MI-VL-CL.

scFv-1602-VH-VL-Fc (“Chain S”)

(SEQ ID NO: 259)

QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGV

IWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQAADTAVYYCAKYDY

DDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGS

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYE

ASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGC

GTKLEIK

GS

DKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHED

PEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYK

CKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVK

GFYPSDIAVEWESNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQG

NVFSCSVMHEALHNHYTQKSLSLSPG

A49MI-VH-CH1-Fc (“Chain H”)

(SEQ ID NO: 260)

EVQLVESGGGLVKPGGSLRLSCAASGFTFSSYSMNWVRQAPGKGLEWVSS

ISSSSSYIYYADSVKGRFTISRDNAKNSLYLQMNSLRAEDTAVYYCARGA

PIGAAAGWFDPWGQGTLVTVSS

ASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGV

HTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEP

KSCDKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVS

HEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGK

EYKCKVSNKALPAPIEKTISKAKGQPREPQVCTLPPSRDELTENQVSLTC

LVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSWLTVDKSRW

QQGNVFSCSVMHEALHNHYTQKSLSLSPG

A49MI-VL-CL (“Chain L”)

(SEQ ID NO: 261)

DIQMTQSPSSVSASVGDRVTITCRASQGISSWLAWYQQKPGKAPKLLIYA

ASSLQSGVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQQGVSFPRTFGG

GTKVEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKV

DNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQG

LSSPVTKSFNRGEC

scFv-1602-VH-VL-Fc represents the full sequence of a CLEC12A binding scFv linked to an Fc domain via a hinge comprising Gly-Ser. The Fc domain linked to the scFv includes Q347R, D399V, and F405T substitutions for heterodimerization and an S354C substitution for forming a disulfide bond with a Y349C substitution in A49MI-VH-CH1-Fc as described below. The scFv has the amino acid sequence of scFv-1602, which includes a heavy chain variable domain of scFv-1602 connected to the N-terminus of a light chain variable domain of scFv-1602 via a (G₄S)₄linker. scFv-1602-VH-VL-Fc comprises substitution of Cys in the VH and VL regions at G44 and S100, facilitating formation of a disulfide bridge between the VH and VL of the scFv.

A49MI-VH-CH1-Fc represents the heavy chain portion of the Fab fragment, which comprises a heavy chain variable domain (SEQ ID NO:95) of NKG2D-binding A49MI and a CH1 domain, connected to an Fc domain. The Fc domain in A49MI-VH-CH1-Fc includes a Y349C substitution in the CH3 domain, which forms a disulfide bond with an S354C substitution on the Fc in scFv-1602-VH-VL-Fc. In A49MI-VH-CH1-Fc, the Fc domain also includes K360E and K409W substitutions for heterodimerization with the Fc in scFv-1602-VH-VL-Fc.

A49MI-VL-CL represents the light chain portion of the Fab fragment comprising a light chain variable domain of NKG2D-binding A49MI (SEQ ID NO:85) and a light chain constant domain.

The DNA sequences encoding the polypeptides scFv-1602-VH-VL-Fc, A49MI-VH-CH1-Fc, and A49MI-VL-CL are provided in Table 260.

TABLE 260

DNA Sequences of F3′-1602 TriNKET

Poly-

pep-

tide
DNA Sequence

scFv-
ATGGACATGAGGGTACCAGCTCAGCTGCTGGGCCTGCTGCTGC

1602-
TTTGGCTTCCTGGCGCTAGATGCCAATTGCAGCTGCAAGAATC

VH-VL-
CGGACCAGGGCTCGTAAAACCTTCTGAAACCCTCTCCCTCACT

Fc
TGCACTGTGTCTGGATTCTCTCTCACGAACTACGGGTTGCATT

GGATCCGACAGCCGCCTGGCAAGTGTCTTGAGTGGATTGGCGT

TATCTGGTCAGGAGGGAAGACGGATTATAATCCAAGTCTGAA

ATCCAGAGTGACAATTTCTAAGGACACGTCCAAAAATCAGTTT

AGCCTCAAACTTTCCTCAGTTCAGGCTGCTGATACAGCGGTGT

ACTATTGCGCAAAGTACGACTATGATGACAGTTTGGACTACTG

GGGACAGGGTACACTGGTAACAGTGAGTTCTGGAGGCGGCGG

TAGTGGAGGCGGGGGAAGTGGTGGCGGAGGGAGTGGAGGAG

GGGGCAGTGATATTCAAATGACGCAGAGCCCCAGTTCCCTTTC

AGCTAGCGTTGGCGACAGAGTAACGATAACGTGTCACGCTTCC

CAGAACATTAATTTTTGGCTTAGCTGGTATCAACAGAAACCGG

GAAAAGCACCCAAGCTGCTTATCTATGAAGCGTCCAATCTCCA

CACGGGTGTCCCATCAAGATTTTCTGGATCAGGCAGTGGCACA

CGATTCACACTGACCATTAGTTCCTTGCAGCCCGAGGATATCG

CAACTTACTACTGTCAGCAAAGTCACTCTTACCCGTTGACATT

CGGATGTGGAACTAAACTGGAGATTAAGGGATCCGATAAGAC

CCACACCTGTCCTCCATGTCCTGCTCCAGAACTGCTCGGCGGA

CCTTCCGTGTTCCTGTTTCCTCCAAAGCCTAAGGACACCCTGAT

GATCAGCAGGACCCCTGAAGTGACCTGCGTGGTGGTGGATGT

GTCTCACGAGGACCCAGAAGTGAAGTTCAATTGGTACGTGGA

CGGCGTGGAAGTGCACAACGCCAAGACCAAGCCTAGAGAGGA

ACAGTACAACAGCACCTACAGAGTGGTGTCCGTGCTGACCGT

GCTGCACCAGGATTGGCTGAACGGCAAAGAGTACAAGTGCAA

GGTGTCCAACAAGGCCCTGCCTGCTCCTATCGAGAAAACCATC

AGCAAGGCCAAGGGCCAGCCTCGCGAGCCTAGAGTGTATACC

TTGCCTCCATGCCGGGACGAGCTGACCAAGAATCAGGTGTCCC

TGACCTGCCTGGTCAAGGGCTTCTACCCTTCCGATATCGCCGT

GGAATGGGAGAGCAATGGCCAGCCAGAGAACAACTACAAGA

CCACACCTCCTGTGCTGGTGTCCGACGGCAGCTTTACCCTGTA

CAGCAAGCTGACAGTGGACAAGAGCAGATGGCAGCAGGGCA

ACGTGTTCTCCTGCAGCGTGATGCACGAGGCCCTGCACAACCA

CTACACCCAGAAAAGCCTGAGCCTGTCTCCTGGCTAA

[SEQ ID NO: 267]

A49MI-
ATGGGCTGGTCCTGCATCATCCTGTTTCTGGTGGCCACAGCCA

VH-
CAGGCGTGCACTCTGAGGTGCAGCTGGTTGAATCTGGCGGCG

CH1-Fc
GACTTGTGAAGCCTGGCGGATCTCTGAGACTGAGCTGTGCCGC

CAGCGGCTTCACCTTTAGCAGCTACAGCATGAACTGGGTCCGA

CAGGCCCCTGGCAAAGGCCTTGAATGGGTGTCCAGCATCAGC

AGCAGCTCCAGCTACATCTACTACGCCGACAGCGTGAAGGGC

AGATTCACCATCAGCCGGGACAACGCCAAGAACAGCCTGTAC

CTGCAGATGAACTCCCTGAGAGCCGAGGACACCGCCGTGTAC

TATTGTGCTAGAGGCGCCCCTATTGGAGCCGCCGCTGGATGGT

TCGATCCTTGGGGACAGGGAACCCTGGTCACCGTTTCTTCTGC

CTCTACAAAGGGCCCCAGCGTTTTCCCACTGGCTCCTAGCAGC

AAGAGCACAAGCGGAGGAACAGCTGCCCTGGGCTGTCTGGTC

AAGGACTACTTTCCTGAGCCTGTGACCGTGTCCTGGAACAGCG

GAGCACTGACTAGCGGCGTGCACACATTTCCAGCCGTGCTGCA

AAGCAGCGGCCTGTACTCTCTGAGCAGCGTCGTGACAGTGCCT

AGCAGCTCTCTGGGCACCCAGACCTACATCTGCAATGTGAACC

ACAAGCCTAGCAACACCAAGGTGGACAAGAAGGTGGAACCCA

AGAGCTGCGACAAGACCCACACCTGTCCTCCATGTCCTGCTCC

AGAACTGCTCGGCGGACCTTCCGTGTTCCTGTTTCCTCCAAAG

CCTAAGGACACCCTGATGATCAGCAGAACCCCTGAAGTGACC

TGCGTGGTGGTGGATGTGTCTCACGAGGACCCCGAAGTGAAG

TTCAATTGGTACGTGGACGGCGTGGAAGTGCACAACGCCAAG

ACCAAGCCTAGAGAGGAACAGTACAACAGCACCTACAGAGTG

GTGTCCGTGCTGACAGTGCTGCACCAGGATTGGCTGAACGGCA

AAGAGTACAAGTGCAAGGTGTCCAACAAGGCCCTGCCTGCTC

CTATCGAGAAAACCATCAGCAAGGCCAAGGGCCAGCCTCGCG

AACCTCAAGTCTGTACACTGCCTCCTAGCCGGGATGAGCTGAC

CGAGAATCAGGTGTCCCTGACCTGTCTCGTGAAGGGCTTCTAC

CCCTCCGATATCGCCGTGGAATGGGAGAGCAATGGCCAGCCA

GAGAACAACTACAAGACAACCCCTCCTGTGCTGGACAGCGAC

GGCTCATTCTTCCTGTACAGCTGGCTGACCGTGGACAAGTCCA

GATGGCAGCAGGGCAACGTGTTCTCCTGCAGCGTGATGCACG

AGGCCCTGCACAACCACTACACCCAGAAGTCCCTGTCTCTGAG

CCCAGGCTAA

[SEQ ID NO: 268]

A49MI-
ATGGACATGAGAGTTCCAGCTCAGCTGCTGGGCCTGCTGCTGC

VL-CL
TTTGGCTTCCTGGCGCTAGATGCGACATCCAGATGACACAGAG

CCCCAGCTCCGTGTCTGCCTCTGTGGGAGACAGAGTGACCATC

ACCTGTAGAGCCAGCCAGGGCATCTCTTCTTGGCTGGCCTGGT

ATCAGCAGAAGCCTGGCAAGGCCCCTAAGCTGCTGATCTATGC

CGCTAGCTCTCTGCAGTCTGGCGTGCCCTCTAGATTTTCTGGCA

GCGGCTCTGGCACCGACTTCACCCTGACCATATCTAGCCTGCA

GCCTGAGGACTTCGCCACCTACTATTGTCAGCAGGGCGTGTCC

TTTCCACGGACCTTTGGCGGCGGAACAAAGGTGGAAATCAAG

CGGACAGTGGCCGCTCCTAGCGTGTTCATCTTTCCACCTAGCG

ACGAGCAGCTGAAGTCCGGCACAGCCTCTGTTGTGTGCCTGCT

GAACAACTTCTACCCCAGAGAAGCCAAGGTGCAGTGGAAGGT

GGACAATGCCCTGCAGAGCGGCAACAGCCAAGAGAGCGTGAC

AGAGCAGGACAGCAAGGACTCCACCTACAGCCTGAGCAGCAC

CCTGACACTGAGCAAGGCCGACTACGAGAAGCACAAAGTGTA

CGCCTGCGAAGTGACCCACCAGGGCCTTTCTAGCCCTGTGACC

AAGAGCTTCAACCGGGGCGAGTGTTGA

[SEQ ID NO: 269]

In another example, a TriNKET of the present disclosure is hF3′-2061 TriNKET, also referred to herein as F3′2061 TriNKET. F3′-2061 TriNKET includes (a) a CLEC12A-binding scFv sequence derived from scFv-2061 of Table 2, in the orientation of VH positioned C-terminal to VL, linked to an Fc domain and (b) an NKG2D-binding Fab fragment derived from A49MI including a heavy chain portion comprising a heavy chain variable domain and a CH1 domain, and a light chain portion comprising a light chain variable domain and a light chain constant domain, wherein the CH1 domain is connected to the Fc domain. F3′-2061 TriNKET includes three polypeptides: scFv-2061-VL-VH-Fc, A49MI-VH-CH1-Fc, and A49MI-VL-CL.

scFv-2061-VL-VH-Fc

(SEQ ID NO: 262)

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYE

ASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGC

GTKLEIK

GGGGSGGGGSGGGGSGGGGS

QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGV

IWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQAADTAVYYCAKYDY

DDSLDYWGQGTLVTVSSGS

DKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHED

PEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYK

CKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVK

GFYPSDIAVEWESNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQG

NVFSCSVMHEALHNHYTQKSLSLSPG

scFv-2061-VL-VH-Fc represents the full sequence of a tumor-associated antigen binding scFv linked to an Fc domain via a hinge comprising Gly-Ser. The Fc domain linked to the scFv includes Q347R, D399V, and F405T substitutions for heterodimerization and an S354C substitution for forming a disulfide bond with a Y349C substitution in A49MI-VH-CH1-Fc as described below. The scFv has the amino acid sequence of scFv-2061, which includes a heavy chain variable domain of scFv-2061 connected to the C-terminus of a light chain variable domain of scFv-2061 via a (G₄S)₄linker. It is contemplated that scFv-2061-VL-VH-Fc may comprise substitution of Cys in the VH and VL regions, facilitating formation of a disulfide bridge between the VH and VL of the scFv.

A49MI-VH-CH1-Fc represents the heavy chain portion of the Fab fragment, which comprises a heavy chain variable domain (SEQ ID NO:95) of NKG2D-binding A49MI and a CH1 domain, connected to an Fc domain. The Fc domain in A49MI-VH-CH1-Fc includes a Y349C substitution in the CH3 domain, which forms a disulfide bond with an S354C substitution on the Fc in scFv-2061-VL-VH-Fc. In A49MI-VH-CH1-Fc, the Fc domain also includes K360E and K409W substitutions for heterodimerization with the Fc in scFv-2061-VL-VH-Fc.

A49MI-VL-CL represents the light chain portion of the Fab fragment comprising a light chain variable domain of NKG2D-binding A49MI (SEQ ID NO:85) and a light chain constant domain.

In another example, a TriNKET of the present disclosure is AB0089 TriNKET. AB0089 TriNKET includes (a) a CLEC12A-binding scFv sequence derived from AB0305 of Table 2, in the orientation of VH positioned N-terminal to VL, linked to an Fc domain and (b) an NKG2D-binding Fab fragment derived from A49MI including a heavy chain portion comprising a heavy chain variable domain and a CH1 domain, and a light chain portion comprising a light chain variable domain and a light chain constant domain, wherein the CH1 domain is connected to the Fc domain. AB0089 TriNKET includes three polypeptides AB0305-VH-VL-Fc, A49MI-VH-CH1-Fc, and A49MI-VL-CL.

AB0305-VH-VL-Fc

(SEQ ID NO: 276)

QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGV

IWVGGATDYNPSLKSRVTISVDTSKNQFSLKLSSVQAADTAVYYCAKGDY

GDTLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGS

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYE

ASNLHTGVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQQSHSYPLTFGC

GTKLEIK

GS

DKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHED

PEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYK

CKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVK

GFYPSDIAVEWESNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQG

NVFSCSVMHEALHNHYTQKSLSLSPG

AB0305-VH-VL-Fc represents the full sequence of a tumor-associated antigen binding scFv linked to an Fc domain via a hinge comprising Gly-Ser. The Fc domain linked to the scFv includes Q347R, D399V, and F405T substitutions for heterodimerization and an S354C substitution for forming a disulfide bond with a Y349C substitution in A49MI-VH-CH1-Fc as described below. The scFv includes a heavy chain variable domain of AB0305 connected to the N-terminus of a light chain variable domain of AB0305 via a (G₄S)₄linker. AB0305-VH-VL-Fc comprises substitution of Cys in the VH and VL regions at G44 and S100, facilitating formation of a disulfide bridge between the VH and VL of the scFv.

A49MI-VH-CH1-Fc represents the heavy chain portion of the Fab fragment, which comprises a heavy chain variable domain (SEQ ID NO:95) of NKG2D-binding A49MI and a CH1 domain, connected to an Fc domain. The Fc domain in A49MI-VH-CH1-Fc includes a Y349C substitution in the CH3 domain, which forms a disulfide bond with an S354C substitution on the Fc in AB0305-VH-VL-Fc. In A49MI-VH-CH1-Fc, the Fc domain also includes K360E and K409W substitutions for heterodimerization with the Fc in AB0305-VH-VL-Fc.

A49MI-VL-CL represents the light chain portion of the Fab fragment comprising a light chain variable domain of NKG2D-binding A49MI (SEQ ID NO:85) and a light chain constant domain.

In another example, a TriNKET of the present disclosure is AB0147 TriNKET. AB0147 TriNKET includes (a) a CLEC12A-binding scFv sequence derived from AB0304 of Table 2, in the orientation of VH positioned N-terminal to VL, linked to an Fc domain and (b) an NKG2D-binding Fab fragment derived from A49MI including a heavy chain portion comprising a heavy chain variable domain and a CH1 domain, and a light chain portion comprising a light chain variable domain and a light chain constant domain, wherein the CH1 domain is connected to the Fc domain. AB0147 TriNKET includes three polypeptides AB0304-VH-VL-Fc, A49MI-VH-CH1-Fc, and A49MI-VL-CL.

AB0304-VH-VL-Fc

(SEQ ID NO: 277)

QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGV

ILSGGWTDYNPSLKSRVTISKDTSKNQFSLKLSSVQAADTAVYYCAKGDY

GDALDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGS

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYE

ASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGC

GTKLEIK

GS

DKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHED

PEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYK

CKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVK

GFYPSDIAVEWESNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQG

NVFSCSVMHEALHNHYTQKSLSLSPG

AB0304-VH-VL-Fc represents the full sequence of a tumor-associated antigen binding scFv linked to an Fc domain via a hinge comprising Gly-Ser. The Fc domain linked to the scFv includes Q347R, D399V, and F405T substitutions for heterodimerization and an S354C substitution for forming a disulfide bond with a Y349C substitution in A49MI-VH-CH1-Fc as described below. The scFv includes a heavy chain variable domain of AB0304 connected to the N-terminus of a light chain variable domain of AB0304 via a (G₄S)₄linker. AB0304-VH-VL-Fc comprises substitution of Cys in the VH and VL regions at G44 and S100, facilitating formation of a disulfide bridge between the VH and VL of the scFv.

A49MI-VH-CH1-Fc represents the heavy chain portion of the Fab fragment, which comprises a heavy chain variable domain (SEQ ID NO:95) of NKG2D-binding A49MI and a CH1 domain, connected to an Fc domain. The Fc domain in A49MI-VH-CH1-Fc includes a Y349C substitution in the CH3 domain, which forms a disulfide bond with an S354C substitution on the Fc in AB0037-VH-VL-Fc. In A49MI-VH-CH1-Fc, the Fc domain also includes K360E and K409W substitutions for heterodimerization with the Fc in AB0304-VH-VL-Fc.

A49MI-VL-CL represents the light chain portion of the Fab fragment comprising a light chain variable domain of NKG2D-binding A49MI (SEQ ID NO:85) and a light chain constant domain.

In another example, a TriNKET of the present disclosure is AB0192 TriNKET. AB0192 TriNKET includes (a) a CLEC12A-binding scFv sequence derived from AB0192 of Table 2, in the orientation of VH positioned N-terminal to VL, linked to an Fc domain and (b) an NKG2D-binding Fab fragment derived from A49MI including a heavy chain portion comprising a heavy chain variable domain and a CH1 domain, and a light chain portion comprising a light chain variable domain and a light chain constant domain, wherein the CH1 domain is connected to the Fc domain. AB0192 TriNKET includes three polypeptides AB0192-VH-VL-Fc, A49MI-VH-CH1-Fc, and A49MI-VL-CL.

AB0192-VH-VL-Fc

(SEQ ID NO: 278)

EVQLVESGGGVVQPGGSLRLSCAASGFTFNAFGMHWVRQAPGKCLEWVAF

ISSGSTSIYYANTVKGRFTISRDNSKNTLYLQMNSLRAEDTAVYYCARDG

YPTGGAMDYWGQGTSVTVSS

GGGGSGGGGSGGGGSGGGGS

DIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYR

ANILVSGVPSRFSGSGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTFGC

GTKLEIK

GS

DKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHED

PEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYK

CKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVK

GFYPSDIAVEWESNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQG

NVFSCSVMHEALHNHYTQKSLSLSPG

AB0192-VH-VL-Fc represents the full sequence of a tumor-associated antigen binding scFv linked to an Fc domain via a hinge comprising Gly-Ser. The Fc domain linked to the scFv includes Q347R, D399V, and F405T substitutions for heterodimerization and an S354C substitution for forming a disulfide bond with a Y349C substitution in A49MI-VH-CH1-Fc as described below. The scFv includes a heavy chain variable domain of AB0192 connected to the N-terminus of a light chain variable domain of AB0192 via a (G₄S)₄linker. AB0192-VH-VL-Fc comprises substitution of Cys in the VH and VL regions at G44 and S100, facilitating formation of a disulfide bridge between the VH and VL of the scFv.

A49MI-VH-CH1-Fc represents the heavy chain portion of the Fab fragment, which comprises a heavy chain variable domain (SEQ ID NO:95) of NKG2D-binding A49MI and a CH1 domain, connected to an Fc domain. The Fc domain in A49MI-VH-CH1-Fc includes a Y349C substitution in the CH3 domain, which forms a disulfide bond with an S354C substitution on the Fc in AB0192-VH-VL-Fc. In A49MI-VH-CH1-Fc, the Fc domain also includes K360E and K409W substitutions for heterodimerization with the Fc in AB0192-VH-VL-Fc.

A49MI-VL-CL represents the light chain portion of the Fab fragment comprising a light chain variable domain of NKG2D-binding A49MI (SEQ ID NO:85) and a light chain constant domain.

In another example, a TriNKET of the present disclosure is hF3′-1295 TriNKET, also referred to herein as F3′-1295 TriNKET. F3′-1295 TriNKET includes (a) a CLEC12A-binding scFv sequence derived from scFv-1295 of Table 2, in the orientation of VH positioned N-terminal to VL, linked to an Fc domain and (b) an NKG2D-binding Fab fragment derived from A49MI including a heavy chain portion comprising a heavy chain variable domain and a CH1 domain, and a light chain portion comprising a light chain variable domain and a light chain constant domain, wherein the CH1 domain is connected to the Fc domain. F3′-1295 TriNKET includes three polypeptides scFv-1295-VH-VL-Fc, A49MI-VH-CH1-Fc, and A49MI-VL-CL.

scFv-1295-VH-VL-Fc

(SEQ ID NO: 281)

QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGV

IWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQAADTAVYYCAKYDY

DDSLDYWGQGTLVTVSS

GGGGSGGGGSGGGGSGGGGS

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYE

ASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGC

GTKLEIK

GS

DKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHED

PEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYK

CKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVK

GFYPSDIAVEWESNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQG

NVFSCSVMHEALHNHYTQKSLSLSPG

scFv-1295-VH-VL-Fc represents the full sequence of a tumor-associated antigen binding scFv linked to an Fc domain via a hinge comprising Gly-Ser. The Fc domain linked to the scFv includes Q347R, D399V, and F405T substitutions for heterodimerization and an S354C substitution for forming a disulfide bond with a Y349C substitution in A49MI-VH-CH1-Fc as described below. The scFv includes a heavy chain variable domain of scFv-1295 connected to the N-terminus of a light chain variable domain of scFv-1295 via a (G₄S)₄linker. scFv-1295-VH-VL-Fc comprises substitution of Cys in the VH and VL regions at G44 and S100, facilitating formation of a disulfide bridge between the VH and VL of the scFv.

A49MI-VH-CH1-Fc represents the heavy chain portion of the Fab fragment, which comprises a heavy chain variable domain (SEQ ID NO:95) of NKG2D-binding A49MI and a CH1 domain, connected to an Fc domain. The Fc domain in A49MI-VH-CH1-Fc includes a Y349C substitution in the CH3 domain, which forms a disulfide bond with an S354C substitution on the Fc in AB0192-VH-VL-Fc. In A49MI-VH-CH1-Fc, the Fc domain also includes K360E and K409W substitutions for heterodimerization with the Fc in scFv-1295-VH-VL-Fc.

A49MI-VL-CL represents the light chain portion of the Fab fragment comprising a light chain variable domain of NKG2D-binding A49MI (SEQ ID NO:85) and a light chain constant domain.

In another example, a TriNKET of the present disclosure is F3 1602 TriNKET, also referred to herein as AB0010. F3 1602 TriNKET includes (a) an NKG2D-binding scFv sequence derived from scFv-A49MI of Table 1, in the orientation of VH positioned N-terminal to VL, linked to an Fc domain and (b) a CLEC12A-binding Fab fragment derived from scFv-1602 including a heavy chain portion comprising a heavy chain variable domain and a CH1 domain, and a light chain portion comprising a light chain variable domain and a light chain constant domain, wherein the CH1 domain is connected to the Fc domain. F3 1602 TriNKET includes three polypeptides scFv-A49MI-VH-VL-Fc, 1602-VH-CH1-Fc, and 1602-VL-CL.

scFv-A49MI-VH-VL-Fc

(SEQ ID NO: 286)

DIQMTQSPSSVSASVGDRVTITCRASQGISSWLAWYQQKPGKAPKLLIYA

ASSLQSGVPSRFSGSGSGTDFTLTISSLQPEDFATYYCQQGVSFPRTFGG

GTKVEIKGGGGSGGGGSGGGGSGGGGS

EVQLVESGGGLVKPGGSLRLSCAASGFTFSSYSMNWVRQAPGKGLEWVSS

ISSSSSYIYYADSVKGRFTISRDNAKNSLYLQMNSLRAEDTAVYYCARGA

PIGAAAGWFDPWGQGTLVTVSS

GS

DKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHED

PEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYK

CKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVK

GFYPSDIAVEWESNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQG

NVFSCSVMHEALHNHYTQKSLSLSPG

1602-VH-CH1-Fc

(SEQ ID NO: 287)

QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKGLEWIGV

IWSGGKTDYNPSLKSRVTISKDTSKNQFSLKLSSVQAADTAVYYCAKYDY

DDSLDYWGQGTLVTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYF

PEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYIC

NVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGGPSVFLFPPKPKDT

LMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTY

RVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVCT

LPPSRDELTENQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDS

DGSFFLYSWLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG

1602-VL-CL

(SEQ ID NO: 288)

DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYE

ASNLHTGVPSRFSGSGSGTRFTLTISSLQPEDIATYYCQQSHSYPLTFGG

GTKLEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKV

DNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQG

LSSPVTKSFNRGEC

scFv-A49MI-VH-VL-Fc represents the full sequence of NKG2D binding scFv A49MI linked to an Fc domain via a hinge comprising Gly-Ser. The Fc domain linked to the scFv includes Q347R, D399V, and F405T substitutions for heterodimerization and an S354C substitution for forming a disulfide bond with a Y349C substitution in A49MI-VH-CH1-Fc as described below. The scFv includes a heavy chain variable domain of scFv-1295 connected to the N-terminus of a light chain variable domain of scFv-A49MI via a (G₄S)₄linker. scFv-A49MI-VH-VL-Fc comprises substitution of Cys in the VH and VL regions at G44 and S100, facilitating formation of a disulfide bridge between the VH and VL of the scFv.

1602-VH-CH1-Fc represents the heavy chain portion of the Fab fragment, which comprises a heavy chain variable domain of CLEC12A-binding 1602 and a CH1 domain, connected to an Fc domain. The Fc domain in 1602-VH-CH1-Fc includes a Y349C substitution in the CH3 domain, which forms a disulfide bond with an S354C substitution on the Fc in scFv-A49MI-VH-VL-Fc. In 1602-VH-CH1-Fc, the Fc domain also includes K360E and K409W substitutions for heterodimerization with the Fc in scFv-A49MI-VH-VL-Fc.

1602-VL-CL represents the light chain portion of the Fab fragment comprising a light chain variable domain of CLEC12A-binding 1602 and a light chain constant domain.

In certain embodiments, a TriNKET of the present disclosure is identical to one of the exemplary TriNKETs described above that includes the EW-RVT Fc mutations, except that the Fc domain linked to the NKG2D-binding Fab fragment comprises the substitutions of Q347R, D399V, and F405T, and the Fc domain linked to the tumor-associated antigen binding scFv comprises matching substitutions K360E and K409W for forming a heterodimer. In certain embodiments, a TriNKET of the present disclosure is identical to one of the exemplary TriNKETs described above that includes the KiH Fc mutations, except that the Fc domain linked to the NKG2D-binding Fab fragment comprises the “hole” substitutions of T366S, L368A, and Y407V, and the Fc domain linked to the tumor-associated antigen-binding scFv comprises the “knob” substitution of T366W for forming a heterodimer.

In certain embodiments, a TriNKET of the present disclosure is identical to one of the exemplary TriNKETs described above, except that the Fc domain linked to the NKG2D-binding Fab fragment includes an S354C substitution in the CH3 domain, and the Fc domain linked to the tumor-associated antigen-binding scFv includes a matching Y349C substitution in the CH3 domain for forming a disulfide bond.

A skilled person in the art would appreciate that during production and/or storage of proteins, N-terminal glutamate (E) or glutamine (Q) can be cyclized to form a lactam (e.g., spontaneously or catalyzed by an enzyme present during production and/or storage). Accordingly, in some embodiments where the N-terminal residue of an amino acid sequence of a polypeptide is E or Q, a corresponding amino acid sequence with the E or Q replaced with pyroglutamate is also contemplated herein.

A skilled person in the art would also appreciate that during protein production and/or storage, the C-terminal lysine (K) of a protein can be removed (e.g., spontaneously or catalyzed by an enzyme present during production and/or storage). Such removal of K is often observed with proteins that comprise an Fc domain at its C-terminus. Accordingly, in some embodiments where the C-terminal residue of an amino acid sequence of a polypeptide (e.g., an Fc domain sequence) is K, a corresponding amino acid sequence with the K removed is also contemplated herein.

The multispecific proteins described above can be made using recombinant DNA technology well known to a skilled person in the art. For example, a first nucleic acid sequence encoding the first immunoglobulin heavy chain can be cloned into a first expression vector; a second nucleic acid sequence encoding the second immunoglobulin heavy chain can be cloned into a second expression vector; a third nucleic acid sequence encoding the immunoglobulin light chain can be cloned into a third expression vector; and the first, second, and third expression vectors can be stably transfected together into host cells to produce the multimeric proteins.

To achieve the highest yield of the multispecific protein, different ratios of the first, second, and third expression vector can be explored to determine the optimal ratio for transfection into the host cells. After transfection, single clones can be isolated for cell bank generation using methods known in the art, such as limited dilution, ELISA, FACS, microscopy, or Clonepix.

The present application also provides a nucleic acid encoding a polypeptide that is a part of a multispecific protein provided herein. In some embodiments, a multispecific protein comprises three polypeptides. In some embodiments, a multispecific protein comprises (a) a first polypeptide comprising the amino acid sequence of SEQ ID NO:276; (b) a second polypeptide comprising the amino acid sequence of SEQ ID NO:260; and (c) a third polypeptide comprising the amino acid sequence of SEQ ID NO:261. In certain embodiments, the present application provides a nucleic acid encoding each of the three polypeptides that are part of the multispecific protein. In certain embodiments, provided herein is a nucleic acid that encodes a first polypeptide comprising the amino acid sequence of SEQ ID NO:276. In certain embodiments, provided herein is a nucleic acid that encodes a second polypeptide comprising the amino acid sequence of SEQ ID NO:260. In certain embodiments, provided herein is a nucleic acid that encodes a a third polypeptide comprising the amino acid sequence of SEQ ID NO:261.

In another aspect, the present application provides a cell comprising one or more nucleic acids encoding a polypeptide that is part of a multispecific protein provided herein. In some embodiments, a cell comprises one or more nucleic acids each encoding one or more of the three polypeptides that are part of the multispecific protein. In certain embodiments, provided herein is a cell comprising a nucleic acid that encodes a first polypeptide comprising the amino acid sequence of SEQ ID NO:276. In certain embodiments, provided herein is a cell comprising a nucleic acid that encodes a second polypeptide comprising the amino acid sequence of SEQ ID NO:260. In certain embodiments, provided herein is a cell comprising a nucleic acid that encodes a a third polypeptide comprising the amino acid sequence of SEQ ID NO:261.

Also provided herein are methods of producing or making a multispecific protein provided herein. In some embodiments, the method comprises: (a) culturing a host cell under conditions suitable for expression of the protein, wherein the host cell comprises one or more nucleic acids each encoding one or more polypeptides that is part of the multispecific protein; and (b) isolating and purifying the protein. In some embodiments, the method comprises: (a) culturing a host cell under conditions suitable for expression of the protein, wherein the host cell comprises a first nucleic acid encoding a first polypeptide comprising the amino acid sequence of SEQ ID NO:276, a second nucleic acid encoding a second polypeptide comprising the amino acid sequence of SEQ ID NO:260, and a third nucleic acid encoding a third polypeptide comprising the amino acid sequence of SEQ ID NO:261; and (b) isolating and purifying the protein.

Clones can be cultured under conditions suitable for bio-reactor scale-up and maintained expression of the multispecific protein. The multispecific proteins can be isolated and purified using methods known in the art including centrifugation, depth filtration, cell lysis, homogenization, freeze-thawing, affinity purification, gel filtration, ion exchange chromatography, hydrophobic interaction exchange chromatography, and mixed-mode chromatography.

II. Characteristics of the Multispecific Proteins

The multispecific proteins described herein include an NKG2D-binding site, a tumor-associated antigen binding site that binds CLEC12A, and an antibody Fc domain or a portion thereof sufficient to bind CD16, or an antigen-binding site that binds CD16. In some embodiments, the multispecific proteins contains an additional antigen-binding site that binds to the same tumor-associated antigen (CLEC12A), as exemplified in the F4-TriNKET format (e.g., FIGS. 2C and 2D).

In some embodiments, the multispecific proteins display similar thermal stability to the corresponding monoclonal antibody, i.e., a monoclonal antibody containing the same tumor-associated antigen binding site as the one incorporated in the multispecific proteins.

In some embodiments, the multispecific proteins simultaneously bind to cells expressing NKG2D and/or CD16, such as NK cells, and cells expressing CLEC12A, such as certain tumor cells. Binding of the multispecific proteins to NK cells can enhance the activity of the NK cells toward destruction of the tumor-associated antigen expressing tumor cells. It has been reported that NK cells exhibit more potent cytotoxicity against target cells that are stressed (see Chan et al., (2014) Cell Death Differ. 21(1):5-14). Without wishing to be bound by theory, it is hypothesized that when NK cells are engaged to a population of cells by a TriNKET, the NK cells may selectively kill the target cells that are stressed (e.g., malignant cells and cells in a tumor microenvironment). This mechanism could contribute to increased specificity and reduced toxicity of TriNKETs, making it possible to selectively clear the stressed cells even if expression of the tumor-associated antigen is not limited to the desired target cells.

In some embodiments, the multispecific proteins induce antibody dependent cellular phagocytosis (ADCP). In some embodiments, the multispecific proteins induce antibody dependent cellular cytotoxicity (ADCC)-like activity. In some embodiments, the multispecific proteins induce ADCC activity. In some embodiments, the multispecific proteins do not induce complement dependent cytotoxicity (CDC).

In some embodiments, the multispecific proteins bind to the tumor-associated antigen with a similar affinity to the corresponding the anti-tumor-associated antigen monoclonal antibody (i.e., a monoclonal antibody containing the same tumor-associated antigen binding site as the one incorporated in the multispecific proteins). In some embodiments, the multispecific proteins are more effective in killing the tumor cells expressing the tumor-associated antigen than the corresponding monoclonal antibodies.

In certain embodiments, the multispecific proteins described herein, which include a binding site for a tumor-associated antigen, activate primary human NK cells when co-culturing with cells expressing that tumor-associated antigen. NK cell activation is marked by the increase in CD107a degranulation and IFN-γ cytokine production. Furthermore, compared to a corresponding anti-tumor-associated antigen monoclonal antibody, the multispecific proteins can show superior activation of human NK cells in the presence of cells expressing the tumor-associated antigen.

In some embodiments, the multispecific proteins described herein, which include a binding site for the tumor-associated antigen, enhance the activity of rested and IL-2-activated human NK cells when co-culturing with cells expressing the tumor-associated antigen.

In some embodiments, compared to the corresponding monoclonal antibody that binds to the tumor-associated antigen, the multispecific proteins offer an advantage in targeting tumor cells that express medium and low levels of the tumor-associated antigen.

In some embodiments, the bivalent F4 format of the TriNKETs (i.e., TriNKETs include an additional antigen-binding site that binds to the tumor-associated antigen) improve the avidity with which the TriNKETs bind to the tumor-associated antigen, which in effect stabilize expression and maintenance of high levels of the tumor-associated antigen on the surface of the tumor cells. In some embodiments, the F4-TriNKETs mediate more potent killing of tumor cells than the corresponding F3-TriNKETs or F3′-TriNKETs.

III. Pharmaceutical Formulations

The present disclosure also provides pharmaceutical formulations that contain a therapeutically effective amount of a multispecific binding protein disclosed herein (e.g., F3′-1602). The pharmaceutical formulation comprises one or more excipients and is maintained at a certain pH. The term “excipient,” as used herein, means any non-therapeutic agent added to the formulation to provide a desired physical or chemical property, for example, pH, osmolarity, viscosity, clarity, color, isotonicity, odor, sterility, stability, rate of dissolution or release, adsorption, or penetration.

Excipients and pH

The one or more excipients in the pharmaceutical formulation of the multispecific binding proteins described in the present application comprises a buffering agent. The term “buffering agent,” as used herein, refers to one or more components that when added to an aqueous solution is able to protect the solution against variations in pH when adding acid or alkali, or upon dilution with a solvent. In addition to phosphate buffers, glycinate, carbonate, citrate, histidine buffers and the like can be used, in which case, sodium, potassium or ammonium ions can serve as counterion.

In certain embodiments, the buffer or buffer system comprises at least one buffer that has a buffering range that overlaps fully or in part with the range of pH 7.0-9.0. In certain embodiments, the buffer has a pKa of about 7.0±0.5. In certain embodiments, the buffer has a pKa of about 7.5±0.5.

In certain embodiments, the buffer comprises a citrate buffer. In certain embodiments, the citrate is present at a concentration of 5 to 100 mM, 10 to 100 mM, 15 to 100 mM, 20 to 100 mM, 5 to 50 mM, 10 to 50 mM, 15 to 100 mM, 20 to 100 mM, 5 to 25 mM, 10 to 25 mM, 15 to 25 mM, 20 to 25 mM, 5 to 20 mM, 10 to 20 mM, or 15 to 20 mM. In certain embodiments, the citrate is present at a concentration of 5 mM, 10 mM, 15 mM, 20 mM, 25 mM, or 50 mM. In certain embodiments, the citrate is present at a concentration of 20 mM.

In certain embodiments, the buffer comprises or further comprise a phosphate buffer. In certain embodiments, the phosphate is present at a concentration of 5 to 100 mM, 10 to 100 mM, 15 to 100 mM, 20 to 100 mM, 5 to 50 mM, 10 to 50 mM, 15 to 100 mM, 20 to 100 mM, 5 to 25 mM, 10 to 25 mM, 15 to 25 mM, 20 to 25 mM, 5 to 20 mM, 10 to 20 mM, or 15 to 20 mM. In certain embodiments, the phosphate is present at a concentration of 5 mM, 10 mM, 15 mM, 20 mM, 25 mM, or 50 mM. In certain embodiments, the phosphate is present at a concentration of 10 mM. It is understood that phosphate is not readily compatible with a lyophilized presentation. Accordingly, in certain embodiments, the concentration of phosphate in the pharmaceutical formulation, if any, is equal to or lower than 5 mM, 4 mM, 3 mM, 2 mM, 1 mM, 0.5 mM, 0.1 mM, 50 μM, 10 μM, 5 μM, or 1 μM. In certain embodiments, the concentration of phosphate in the pharmaceutical formulation is below the detection limit. In certain embodiments, no phosphate is added when preparing the pharmaceutical formulation.

The pharmaceutical formulation of the multispecific binding proteins described in the present application may have a pH above 7.0. For example, in certain embodiments, the pharmaceutical formulation has a pH of 7.0 to 8.0 (i.e., 7.5±0.5), 7.0 to 7.8 (i.e., 7.4±0.4), 7.0 to 7.6 (i.e., 7.3±0.3), 7.0 to 7.4 (i.e., 7.2±0.2), 7.0 to 7.2 (i.e., 7.1±0.1), 7.1 to 7.3 (i.e., 7.2±0.1), 7.2 to 7.4 (i.e., 7.3±0.1), 7.3 to 7.5 (i.e., 7.4±0.1), 7.4 to 7.6 (i.e., 7.5±0.1), 7.5 to 7.7 (i.e., 7.6±0.1), 7.6 to 7.8 (i.e., 7.7±0.1), 7.7 to 7.9 (i.e., 7.8±0.1), or 7.8 to 8.0 (i.e., 7.9±0.1). In certain embodiments, the pharmaceutical formulation has a pH of 7.0, 7.1, 7.2, 7.3, 7.4, 7.5, 7.6, 7.7, 7.8, 7.9, or 8.0. In certain embodiments, the pharmaceutical formulation has a pH of 7.0. Under the rules of scientific rounding, a pH greater than or equal to 6.95 and smaller than or equal to 7.05 is rounded as 7.0.

The one or more excipients in the pharmaceutical formulation of the multispecific binding proteins described in the present application further comprises a sugar or sugar alcohol. Sugars and sugar alcohols are useful in pharmaceutical formulations as a thermal stabilizer. In certain embodiments, the pharmaceutical formulation comprises a sugar, for example, a monosaccharide (glucose, xylose, or erythritol), a disaccharide (e.g., sucrose, trehalose, maltose, or galactose), or an oligosaccharide (e.g., stachyose). In specific embodiments, the pharmaceutical formulation comprises sucrose. In certain embodiments, the pharmaceutical composition comprises a sugar alcohol, for example, a sugar alcohol derived from a monosaccharide (e.g., mannitol, sorbitol, or xylitol), a sugar alcohol derived from a disaccharide (e.g., lactitol or maltitol), or a sugar alcohol derived from an oligosaccharide. In specific embodiments, the pharmaceutical formulation comprises sorbitol.

The amount of the sugar or sugar alcohol contained within the formulation can vary depending on the specific circumstances and intended purposes for which the formulation is used. In certain embodiments, the pharmaceutical formulation comprises 50 to 300 mM, 50 to 250 mM, 100 to 300 mM, 100 to 250 mM, 150 to 300 mM, 150 to 250 mM, 200 to 300 mM, 200 to 250 mM, or 250 to 300 mM of the sugar or sugar alcohol. In certain embodiments, the pharmaceutical formulation comprises 50 mM, 75 mM, 100 mM, 125 mM, 150 mM, 200 mM, 250 mM, or 300 mM of the sugar or sugar alcohol. In specific embodiments, the pharmaceutical formulation comprises 250 mM of the sugar or sugar alcohol (e.g., sucrose or sorbitol).

The one or more excipients in the pharmaceutical formulation disclosed herein further comprises a surfactant. The term “surfactant,” as used herein, refers to a surface active molecule containing both a hydrophobic portion (e.g., alkyl chain) and a hydrophilic portion (e.g., carboxyl and carboxylate groups). Surfactants are useful in pharmaceutical formulations for reducing aggregation of a therapeutic protein. Surfactants suitable for use in the pharmaceutical formulations are generally non-ionic surfactants and include, but are not limited to, polysorbates (e.g. polysorbates 20 or 80); poloxamers (e.g. poloxamer 188); sorbitan esters and derivatives; Triton; sodium laurel sulfate; sodium octyl glycoside; lauryl-, myristyl-, linoleyl-, or stearyl-sulfobetadine; lauryl-, myristyl-, linoleyl- or stearyl-sarcosine; linoleyl-, myristyl-, or cetyl-betaine; lauramidopropyl-cocamidopropyl-, linoleamidopropyl-, myristamidopropyl-, palmidopropyl-, or isostearamidopropylbetaine (e.g., lauroamidopropyl); myristamidopropyl-, palmidopropyl-, or isostearamidopropyl-dimethylamine; sodium methyl cocoyl-, or disodium methyl oleyl-taurate; and the MONAQUAT™ series (Mona Industries, Inc., Paterson, N.J.), polyethylene glycol, polypropyl glycol, and copolymers of ethylene and propylene glycol (e.g., Pluronics, PF68 etc.). In certain embodiments, the surfactant is a polysorbate. In certain embodiments, the surfactant is polysorbate 80.

The amount of a non-ionic surfactant contained within the pharmaceutical formulation of the multispecific binding proteins described in the present application may vary depending on the specific properties desired of the formulation, as well as the particular circumstances and purposes for which the formulations are intended to be used. In certain embodiments, the pharmaceutical formulation comprises 0.005% to 0.5%, 0.005% to 0.2%, 0.005% to 0.1%, 0.005% to 0.05%, 0.005% to 0.02%, 0.005% to 0.01%, 0.01% to 0.5%, 0.01% to 0.2%, 0.01% to 0.1%, 0.01% to 0.05%, or 0.01% to 0.02% of the non-ionic surfactant (e.g., polysorbate 80). In certain embodiments, the pharmaceutical formulation comprises 0.005%, 0.01%, 0.02%, 0.03%, 0.04%, 0.05%, 0.06%, 0.07%, 0.08%, 0.09%, 0.1%, 0.15%, 0.2%, 0.25%, 0.3%, 0.35%, 0.4%, 0.45%, or 0.5% of the non-ionic surfactant (e.g., polysorbate 80).

In certain embodiments, the pharmaceutical formulation is isotonic. An “isotonic” formulation is one which has essentially the same osmotic pressure as human blood. Isotonic formulations generally have an osmotic pressure from about 250 to 350 mOsmol/kgH₂O. Isotonicity can be measured using a vapor pressure or ice-freezing type osmometer. In certain embodiments, the osmolarity of the pharmaceutical formulation is 250 to 350 mOsmol/kgH₂O. In certain embodiments, the osmolarity of the pharmaceutical formulation is 300 to 350 mOsmol/kgH₂O.

Substances such as sugar, sugar alcohol, and NaCl can be included in the pharmaceutical formulation for desired osmolarity. It is understood that NaCl is not readily compatible with a lyophilized presentation. Accordingly, in certain embodiments, the concentration of NaCl in the pharmaceutical formulation, if any, is equal to or lower than 10 mM, 9 mM, 8 mM, 7 mM, 6 mM, 5 mM, 4 mM, 3 mM, 2 mM, 1 mM, 0.5 mM, 0.1 mM, 50 μM, 10 μM, 5 μM, or 1 μM. In certain embodiments, the concentration of NaCl in the pharmaceutical formulation is below the detection limit. In certain embodiments, no NaCl salt is added when preparing the pharmaceutical formulation.

The pharmaceutical formulation of the multispecific binding proteins described in the present application may further comprise one or more other substances, such as a bulking agent or a preservative. A “bulking agent” is a compound which adds mass to a lyophilized mixture and contributes to the physical structure of the lyophilized cake (e.g., facilitates the production of an essentially uniform lyophilized cake which maintains an open pore structure). Illustrative bulking agents include mannitol, glycine, polyethylene glycol and sorbitol. The lyophilized formulations of the multispecific binding proteins described in the present application may contain such bulking agents. A preservative reduces bacterial action and may, for example, facilitate the production of a multi-use (multiple-dose) formulation.

Exemplary Formulations

In certain embodiments, the pharmaceutical formulation of the multispecific binding proteins described in the present application comprises the multispecific binding protein in a buffer at pH 7.0 or higher (e.g., at pH 7.0 to 8.0).

In certain embodiments, the pharmaceutical formulation comprises 10 to 200 mg/mL of the multispecific binding protein, 10 to 25 mM of citrate, 200 to 300 mM of a sugar or sugar alcohol (e.g., sucrose), and 0.005% to 0.05% of a polysorbate (e.g., polysorbate 80), at pH 7.0 to 8.0. In certain embodiments, the pharmaceutical formulation comprises 10 to 200 mg/mL of the multispecific binding protein, 20 mM of citrate, 250 mM of a sugar or sugar alcohol (e.g., sucrose), and 0.01% of a polysorbate (e.g., polysorbate 80), at pH 7.0 to 8.0. In certain embodiments, the pharmaceutical formulation comprises 10 to 200 mg/mL of the multispecific binding protein, 20 mM of citrate, 250 mM of a sugar or sugar alcohol (e.g., sucrose), and 0.01% of a polysorbate (e.g., polysorbate 80), at pH 7.0 to 7.5 (e.g., pH 7.0).

In certain embodiments, the pharmaceutical formulation comprises 10 to 50 mg/mL of the multispecific binding protein, 10 to 25 mM of citrate, 200 to 300 mM of a sugar or sugar alcohol (e.g., sucrose), and 0.005% to 0.05% of a polysorbate (e.g., polysorbate 80), at pH 7.0 to 8.0. In certain embodiments, the pharmaceutical formulation comprises 10 to 50 mg/mL of the multispecific binding protein, 20 mM of citrate, 250 mM of a sugar or sugar alcohol (e.g., sucrose), and 0.01% of a polysorbate (e.g., polysorbate 80), at pH 7.0 to 8.0. In certain embodiments, the pharmaceutical formulation comprises 10 to 50 mg/mL of the multispecific binding protein, 20 mM of citrate, 250 mM of a sugar or sugar alcohol (e.g., sucrose), and 0.01% of a polysorbate (e.g., polysorbate 80), at pH 7.0 to 7.5 (e.g., pH 7.0).

In certain embodiments, the pharmaceutical formulation comprises 10 to 200 mg/mL of the multispecific binding protein, 5 to 20 mM of citrate, 5 to 20 mM of phosphate, and 100 to 150 mM of NaCl, at pH 7.0 to 8.0. In certain embodiments, the pharmaceutical formulation comprises 10 to 200 mg/mL of the multispecific binding protein, 10 mM of citrate, 10 mM of phosphate, and 125 mM of NaCl, at pH 7.0 to 8.0. In certain embodiments, the pharmaceutical formulation comprises 10 to 200 mg/mL of the multispecific binding protein, 10 mM of citrate, 10 mM of phosphate, and 125 mM of NaCl, at pH 7.5 to 8.0 (e.g., pH 7.8).

In certain embodiments, the pharmaceutical formulation comprises 10 to 50 mg/mL of the multispecific binding protein, 5 to 20 mM of citrate, 5 to 20 mM of phosphate, and 100 to 150 mM of NaCl, at pH 7.0 to 8.0. In certain embodiments, the pharmaceutical formulation comprises 10 to 50 mg/mL of the multispecific binding protein, 10 mM of citrate, 10 mM of phosphate, and 125 mM of NaCl, at pH 7.0 to 8.0. In certain embodiments, the pharmaceutical formulation comprises 10 to 50 mg/mL of the multispecific binding protein, 10 mM of citrate, 10 mM of phosphate, and 125 mM of NaCl, at pH 7.5 to 8.0 (e.g., pH 7.8).

Stability of the Multispecific Binding Protein

The pharmaceutical formulations of the multispecific binding proteins described in the present application exhibit high levels of stability. A pharmaceutical formulation is stable when the multispecific binding protein within the formulation retains an acceptable degree of physical property, chemical structure, and/or biological function after storage under defined conditions.

Stresses during manufacture, storage, and use of the pharmaceutical formulation include but are not limited to heat stress (e.g., incubation at 40° C. for 4 weeks), freeze/thaw stress (e.g., after 6 freeze/thaw cycles), agitation stress (e.g., stirring at 60 rpm at 4° C. for 18 hours), low pH stress (e.g., incubation at pH 5.0 at 40° C. for 2 weeks), high pH stress (e.g., incubation at pH 8.0 at 40° C. for 2 weeks), oxidative stress (e.g., forced oxidation in 0.02% hydrogen peroxide at room temperature for 24 hours), or low pH hold (e.g., incubation at pH 3.3 at room temperature for 1.5 hours). In certain embodiments, the multispecific binding protein is stable after exposure to one or more of these stresses.

Stability can be measured by determining the percentage of the multispecific binding protein in the formulation that remains in a native conformation after storage for a defined amount of time at a defined temperature. The percentage of a protein in a native conformation can be determined by, for example, size exclusion chromatography (e.g., size exclusion high performance liquid chromatography), where a protein in the native conformation is not aggregated (eluted in a high molecular weight fraction) or degraded (eluted in a low molecular weight fraction). In certain embodiments, less than 0.5%, 0.6%, 0.7%, 0.8%, 0.9%, or 1% of the multispecific binding protein forms a high molecular weight complex (i.e., having a higher molecular weight than the native protein), as determined by size-exclusion chromatography, after one or more of the stresses disclosed herein. In certain embodiments, less than 0.5%, 0.6%, 0.7%, 0.8%, 0.9%, or 1% of the multispecific binding protein is degraded (i.e., having a lower molecular weight than the native protein), as determined by size-exclusion chromatography, after one or more of the stresses disclosed herein.

Stability can also be measured by determining the percentage of multispecific binding protein present in a more acidic fraction (“acidic form”) relative to the main fraction of protein (“main charge form”). Though not wishing to be bound by theory, deamidation of a protein may cause it to become more negatively charged and thus more acidic relative to the non-deamidated protein (see, e.g., Robinson, Protein Deamidation, (2002) PNAS 99(8):5283-88). The percentage of the acidic form of a protein can be determined by ion exchange chromatography (e.g., cation exchange high performance liquid chromatography) or imaged capillary isoelectric focusing (icIEF). In certain embodiments, at least 30%, 40%, 50%, 60%, 70%, 80%, or 90% of the multispecific binding protein in the pharmaceutical formulation is in the main charge form, after one or more of the stresses disclosed herein. In certain embodiments, no more than 10%, 20%, 30%, 40%, or 50% of the multispecific binding protein in the pharmaceutical formulation is in an acidic form, after one or more of the stresses disclosed herein.

Stability can also be measured by determining the purity of the multispecific binding protein by electrophoresis after denaturing the protein with sodium dodecyl sulfate (SDS). The protein sample can be denatured in the presence or absence of an agent that reduces protein disulfide bonds (e.g., β-mercaptoethanol). In certain embodiments, the purity of the multispecific binding protein in the pharmaceutical formulation, as measured by capillary electrophoresis after denaturing the protein sample under reducing conditions (e.g., in the presence of β-mercaptoethanol), is at least 95%, 96%, 97%, 98%, or 99%, after one or more of the stresses disclosed herein. In certain embodiments, the purity of the multispecific binding protein in the pharmaceutical formulation, as measured by capillary electrophoresis after denaturing the protein sample under reducing conditions (e.g., in the presence of β-mercaptoethanol), is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%, after one or more of the stresses disclosed herein. In certain embodiments, the purity of the multispecific binding protein in the pharmaceutical formulation, as measured by capillary electrophoresis after denaturing the protein sample under non-reducing conditions, is at least 95%, 96%, 97%, 98%, or 99%, after one or more of the stresses disclosed herein. In certain embodiments, the purity of the multispecific binding protein in the pharmaceutical formulation, as measured by capillary electrophoresis after denaturing the protein sample under non-reducing conditions, is at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%, after one or more of the stresses disclosed herein.

Stability can also be measured by determining the parameters of a protein solution by dynamic light scattering. The Z-average and polydispersity index (PDI) values indicate the average diameter of particles in a solution and these measures increase when aggregates are present in the solution. The monomer % Pd value indicates the spread of different monomers detected, where lower values indicate a monodispere solution, which is preferred. The monomer size detected by DLS is useful in confirming that the main population is monomer and to characterize any higher order aggregates that may be present. In certain embodiments, the Z-average value of the pharmaceutical formulation does not increase by more than 5%, 10%, or 15%, after one or more of the stresses disclosed herein. In certain embodiments, the PDI value of the pharmaceutical formulation does not increase by more than 10%, 20%, 30%, 40%, or 50%, after one or more of the stresses disclosed herein.

In addition to the assessment of the parameters above, stability can also be measured by determining the protein's functional characters. The multispecific binding proteins disclosed herein bind CLEC12A, NKG2D, and CD16. According, in certain embodiments, the binding affinity of the multispecific binding protein to CLEC12A (e.g., human CLEC12A) after one or more of the stresses disclosed herein, as measured by dissociation constant (K_D) or maximal binding response (capture level), is not lower than that in the control sample (not exposed to the stress) by more than 1%, 2%, 3%, 4%, 5%, 10%, 20%, or 30%. In certain embodiments, the binding affinity of the multispecific binding protein to NKG2D (e.g., human NKG2D) after one or more of the stresses disclosed herein, as measured by dissociation constant (K_D) or maximal binding response (capture level), is not lower than that in the control sample (not exposed to the stress) by more than 1%, 2%, 3%, 4%, 5%, 10%, 20%, or 30%. In certain embodiments, the binding affinity of the multispecific binding protein to CD16 (e.g., human CD16a V158) after one or more of the stresses disclosed herein, as measured by dissociation constant (K_D) or maximal binding response (capture level), is not lower than that in the control sample (not exposed to the stress) by more than 1%, 2%, 3%, 4%, 5%, 10%, 20%, or 30%.

The multispecific binding protein disclosed herein has the ability to mediate cytotoxicity against a CLEC12A-expressing tumor cell by an NK cell. In certain embodiments, the ability of the multispecific binding protein to mediate such cytotoxicity after one or more of the stresses disclosed herein, as measured by EC50 of the multispecific binding protein in a mixed culture cytotoxicity assay, is not greater than that in the control sample (not exposed to the stress) by more than 1%, 2%, 3%, 4%, 5%, 10%, 20%, or 30%. In certain embodiments, the ability of the multispecific binding protein to mediate such cytotoxicity after one or more of the stresses disclosed herein, as measured by the maximum lysis of the target cells in a mixed culture cytotoxicity assay, is not greater than that in the control sample (not exposed to the stress) by more than 1%, 2%, 3%, 4%, 5%, 10%, 20%, or 30%.

Exemplary methods to determine stability of the multispecific binding protein in the pharmaceutical formulation are described in Examples 13 and 14 of the present disclosure. Additionally, stability of the protein can be assessed by measuring the binding affinity of the multispecific binding protein to its targets or the biological activity of the multispecific binding protein in certain in vitro assays, such as the NK cell activation assays and cytotoxicity assays described in WO 2018/152518.

Dosage Forms

The pharmaceutical formulation can be prepared and stored as a liquid formulation or a lyophilized form. In certain embodiments, the pharmaceutical formulation is a liquid formulation for storage at 2-8° C. (e.g., 4° C.) or a frozen formulation for storage at −20° C. or lower. The sugar or sugar alcohol in the formulation is used as a lyoprotectant.

Prior to pharmaceutical use, the pharmaceutical formulation can be diluted or reconstituted in an aqueous carrier is suitable for the route of administration. Other exemplary carriers include sterile water for injection (SWFI), bacteriostatic water for injection (BWFI), a pH buffered solution (e.g., phosphate-buffered saline), sterile saline solution, Ringer's solution, or dextrose solution. For example, when the pharmaceutical formulation is prepared for intravenous administration, the pharmaceutical formulation can be diluted in a 0.9% sodium chloride (NaCl) solution. In certain embodiments, the diluted pharmaceutical formulation is isotonic and suitable for administration by intravenous infusion.

The pharmaceutical formulation comprises the multispecific binding protein at a concentration suitable for storage. In certain embodiments, the pharmaceutical formulation comprises the multispecific binding protein at a concentration of 10-200 mg/mL, 10-150 mg/mL, 10-100 mg/mL, 10-50 mg/mL, 10-40 mg/mL, 10-30 mg/mL, 10-25 mg/mL, 10-20 mg/mL, 10-15 mg/mL, 20-200 mg/mL, 20-150 mg/mL, 20-100 mg/mL, 20-50 mg/mL, 20-40 mg/mL, 20-30 mg/mL, 20-25 mg/mL, 50-200 mg/mL, 50-150 mg/mL, 50-100 mg/mL, 100-200 mg/mL, 100-150 mg/mL, or 150-200 mg/mL. In certain embodiments, the pharmaceutical formulation comprises the multispecific binding protein at a concentration of 10 mg/mL, 15 mg/mL, 20 mg/mL, 25 mg/mL, 30 mg/mL, 40 mg/mL, 50 mg/mL, 60 mg/mL, 70 mg/mL, 80 mg/mL, 90 mg/mL, 100 mg/mL, 150 mg/mL, or 200 mg/mL.

In certain embodiments, the pharmaceutical formulation is packaged in a container (e.g., a vial, bag, pen, or syringe). In certain embodiments, the formulation may be a lyophilized formulation or a liquid formulation. In certain embodiments, the amount of multispecific binding protein in the container is suitable for administration as a single dose. In certain embodiments, the amount of multispecific binding protein in the container is suitable for administration in multiple doses. In certain embodiments, the pharmaceutical formulation comprises the multispecific binding protein at an amount of 0.1 to 2000 mg. In certain embodiments, the pharmaceutical formulation comprises the multispecific binding protein at an amount of 1 to 2000 mg, 10 to 2000 mg, 20 to 2000 mg, 50 to 2000 mg, 100 to 2000 mg, 200 to 2000 mg, 500 to 2000 mg, 1000 to 2000 mg, 0.1 to 1000 mg, 1 to 1000 mg, 10 to 1000 mg, 20 to 1000 mg, 50 to 1000 mg, 100 to 1000 mg, 200 to 1000 mg, 500 to 1000 mg, 0.1 to 500 mg, 1 to 500 mg, 10 to 500 mg, 20 to 500 mg, 50 to 500 mg, 100 to 500 mg, 200 to 500 mg, 0.1 to 200 mg, 1 to 200 mg, 10 to 200 mg, 20 to 200 mg, 50 to 200 mg, 100 to 200 mg, 0.1 to 100 mg, 1 to 100 mg, 10 to 100 mg, 20 to 100 mg, 50 to 100 mg, 0.1 to 50 mg, 1 to 50 mg, 10 to 50 mg, 20 to 50 mg, 0.1 to 20 mg, 1 to 20 mg, 10 to 20 mg, 0.1 to 10 mg, 1 to 10 mg, or 0.1 to 1 mg. In certain embodiments, the pharmaceutical formulation comprises the multispecific binding protein at an amount of 0.1 mg, 1 mg, 2 mg, 5 mg, 10 mg, 20 mg, 30 mg, 40 mg, 50 mg, 60 mg, 70 mg, 80 mg, 90 mg, 100 mg, 150 mg, 200 mg, 250 mg, 300 mg, 400 mg, 500 mg, 600 mg, 700 mg, 800 mg, 900 mg, 1000 mg, 1500 mg, or 2000 mg.

IV. Therapeutic Applications

The present application provides methods for treating autoimmune disease or cancer using a multispecific binding protein described herein and/or a pharmaceutical composition (e.g., pharmaceutical formulation) described herein. The methods may be used to treat a variety of cancers expressing CLEC12A.

The therapeutic method can be characterized according to the cancer to be treated. The cancer to be treated can be characterized according to the presence of a particular antigen expressed on the surface of the cancer cell.

Cancers characterized by the expression of CLEC12A, include, without limitation, solid tumor indications such as gastric cancer, esophageal cancer, lung cancer, triple-negative breast cancer, colorectal cancer, and head and neck cancer.

It is contemplated that the protein, conjugate, cells, and/or pharmaceutical compositions of the present disclosure can be used to treat a variety of cancers, not limited to cancers in which the cancer cells or the cells in the cancer microenvironment express CLEC12A.

The therapeutic method can be characterized according to the cancer to be treated. For example, in certain embodiments, the cancer is acute myeloid leukemia, multiple myeloma, diffuse large B cell lymphoma, thymoma, adenoid cystic carcinoma, gastrointestinal cancer, renal cancer, breast cancer, glioblastoma, lung cancer, ovarian cancer, brain cancer, prostate cancer, pancreatic cancer, or melanoma. In some embodiments, the cancer is a hematologic malignancy or leukemia. In certain embodiments, the cancer is acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), myelodysplasia, myelodysplastic syndromes, acute T-lymphoblastic leukemia, or acute promyelocytic leukemia, chronic myelomonocytic leukemia, or myeloid blast crisis of chronic myeloid leukemia.

In certain embodiments, the AML is a minimal residual disease (MRD). In certain embodiments, the MRD is characterized by the presence or absence of a mutation selected from CLEC12A-ITD ((Fms-like tyrosine kinase 3)-internal tandem duplications (ITD)), NPM1 (Nucleophosmin 1), DNMT3A (DNA methyltransferase gene DNMT3A), and IDH (Isocitrate dehydrogenase 1 and 2 (IDH1 and IDH2)). In certain embodiments, the MDS is selected from MDS with multilineage dysplasia (MDS-MLD), MDS with single lineage dysplasia (MDS-SLD), MDS with ring sideroblasts (MDS-RS), MDS with excess blasts (MDS-EB), MDS with isolated del(5q), and MDS, unclassified (MDS-U). In certain embodiments, the MDS is a primary MDS or a secondary MDS.

In certain embodiments, the ALL is selected from B-cell acute lymphoblastic leukemia (B-ALL) and T-cell acute lymphoblastic leukemia (T-ALL). In certain embodiments, the MPN is selected from polycythaemia vera, essential thrombocythemia (ET), and myelofibrosis. In certain embodiments, the non-Hodgkin lymphoma is selected from B-cell lymphoma and T-cell lymphoma. In certain embodiments, the lymphoma is selected from chronic lymphocytic leukemia (CLL), lymphoblastic lymphoma (LPL), diffuse large B-cell lymphoma (DLBCL), Burkitt lymphoma (BL), primary mediastinal large B-cell lymphoma (PMBL), follicular lymphoma, mantle cell lymphoma, hairy cell leukemia, plasma cell myeloma (PCM) or multiple myeloma (MM), mature T/NK neoplasms, and histiocytic neoplasms.

In certain embodiments, the cancer is a solid tumor. In certain other embodiments, the cancer is brain cancer, bladder cancer, breast cancer, cervical cancer, colon cancer, colorectal cancer, endometrial cancer, esophageal cancer, leukemia, lung cancer, liver cancer, melanoma, ovarian cancer, pancreatic cancer, prostate cancer, rectal cancer, renal cancer, stomach cancer, testicular cancer, or uterine cancer. In yet other embodiments, the cancer is a vascularized tumor, squamous cell carcinoma, adenocarcinoma, small cell carcinoma, melanoma, glioma, neuroblastoma, sarcoma (e.g., an angiosarcoma or chondrosarcoma), larynx cancer, parotid cancer, biliary tract cancer, thyroid cancer, acral lentiginous melanoma, actinic keratoses, acute lymphocytic leukemia, acute myeloid leukemia, adenoid cystic carcinoma, adenomas, adenosarcoma, adenosquamous carcinoma, anal canal cancer, anal cancer, anorectum cancer, astrocytic tumor, Bartholin gland carcinoma, basal cell carcinoma, biliary cancer, bone cancer, bone marrow cancer, bronchial cancer, bronchial gland carcinoma, carcinoid, cholangiocarcinoma, chondrosarcoma, choroid plexus papilloma/carcinoma, chronic lymphocytic leukemia, chronic myeloid leukemia, clear cell carcinoma, connective tissue cancer, cystadenoma, digestive system cancer, duodenum cancer, endocrine system cancer, endodermal sinus tumor, endometrial hyperplasia, endometrial stromal sarcoma, endometrioid adenocarcinoma, endothelial cell cancer, ependymal cancer, epithelial cell cancer, Ewing's sarcoma, eye and orbit cancer, female genital cancer, focal nodular hyperplasia, gallbladder cancer, gastric antrum cancer, gastric fundus cancer, gastrinoma, glioblastoma, glucagonoma, heart cancer, hemangioblastomas, hemangioendothelioma, hemangiomas, hepatic adenoma, hepatic adenomatosis, hepatobiliary cancer, hepatocellular carcinoma, Hodgkin's disease, ileum cancer, insulinoma, intraepithelial neoplasia, intraepithelial squamous cell neoplasia, intrahepatic bile duct cancer, invasive squamous cell carcinoma, jejunum cancer, joint cancer, Kaposi's sarcoma, pelvic cancer, large cell carcinoma, large intestine cancer, leiomyosarcoma, lentigo maligna melanomas, lymphoma, male genital cancer, malignant melanoma, malignant mesothelial tumors, medulloblastoma, medulloepithelioma, meningeal cancer, mesothelial cancer, metastatic carcinoma, mouth cancer, mucoepidermoid carcinoma, multiple myeloma, muscle cancer, nasal tract cancer, nervous system cancer, neuroepithelial adenocarcinoma nodular melanoma, non-epithelial skin cancer, non-Hodgkin's lymphoma, oat cell carcinoma, oligodendroglial cancer, oral cavity cancer, osteosarcoma, papillary serous adenocarcinoma, penile cancer, pharynx cancer, pituitary tumors, plasmacytoma, pseudosarcoma, pulmonary blastoma, rectal cancer, renal cell carcinoma, respiratory system cancer, retinoblastoma, rhabdomyosarcoma, sarcoma, serous carcinoma, sinus cancer, skin cancer, small cell carcinoma, small intestine cancer, smooth muscle cancer, soft tissue cancer, somatostatin-secreting tumor, spine cancer, squamous cell carcinoma, striated muscle cancer, submesothelial cancer, superficial spreading melanoma, T cell leukemia, tongue cancer, undifferentiated carcinoma, ureter cancer, urethra cancer, urinary bladder cancer, urinary system cancer, uterine cervix cancer, uterine corpus cancer, uveal melanoma, vaginal cancer, verrucous carcinoma, VIPoma, vulva cancer, well differentiated carcinoma, or Wilms tumor.

In certain other embodiments, the cancer is non-Hodgkin's lymphoma, such as a B-cell lymphoma or a T-cell lymphoma. In certain embodiments, the non-Hodgkin's lymphoma is a B-cell lymphoma, such as a diffuse large B-cell lymphoma, primary mediastinal B-cell lymphoma, follicular lymphoma, small lymphocytic lymphoma, mantle cell lymphoma, marginal zone B-cell lymphoma, extranodal marginal zone B-cell lymphoma, nodal marginal zone B-cell lymphoma, splenic marginal zone B-cell lymphoma, Burkitt lymphoma, lymphoplasmacytic lymphoma, hairy cell leukemia, or primary central nervous system (CNS) lymphoma. In certain other embodiments, the non-Hodgkin's lymphoma is a T-cell lymphoma, such as a precursor T-lymphoblastic lymphoma, peripheral T-cell lymphoma, cutaneous T-cell lymphoma, angioimmunoblastic T-cell lymphoma, extranodal natural killer/T-cell lymphoma, enteropathy type T-cell lymphoma, subcutaneous panniculitis-like T-cell lymphoma, anaplastic large cell lymphoma, or peripheral T-cell lymphoma.

IV. Combination Therapy

Another aspect of the present application provides for combination therapy. A multispecific binding protein described herein can be used in combination with additional therapeutic agents to treat autoimmune disease or to treat cancer.

Exemplary therapeutic agents that may be used as part of a combination therapy in treating autoimmune inflammatory diseases are described in Li et al. (2017) Front. Pharmacol., 8:460, and include, for example, non-steroidal anti-inflammatory drugs (NSAIDs) (e.g., COX-2 inhibitors), glucocorticoids (e.g., prednisone/prednisolone, methylprednisolone, and the fluorinated glucocorticoids such as dexamethasone and betamethasone), disease-modifying antirheumatic drugs (DMARDs) (e.g., methotrexate, leflunomide, gold compounds, sulfasalazine, azathioprine, cyclophosphamide, antimalarials, D-penicillamine, and cyclosporine), anti-TNF biologics (e.g., infliximab, etanercept, adalimumab, golimumab, Certolizumab pegol, and their biosimilars), and other biologics targeting CTLA-4 (e.g., abatacept), IL-6 receptor (e.g., tocilizumab), IL-1 (e.g., anakinra), Thl immune responses (IL-12/IL-23) (e.g., ustekinumab), Th17 immune responses (IL-17) (e.g., secukinumab) and CD20 (e.g., rituximab).

Exemplary therapeutic agents that may be used as part of a combination therapy in treating cancer include, for example, radiation, mitomycin, tretinoin, ribomustin, gemcitabine, vincristine, etoposide, cladribine, mitobronitol, methotrexate, doxorubicin, carboquone, pentostatin, nitracrine, zinostatin, cetrorelix, letrozole, raltitrexed, daunorubicin, fadrozole, fotemustine, thymalfasin, sobuzoxane, nedaplatin, cytarabine, bicalutamide, vinorelbine, vesnarinone, aminoglutethimide, amsacrine, proglumide, elliptinium acetate, ketanserin, doxifluridine, etretinate, isotretinoin, streptozocin, nimustine, vindesine, flutamide, drogenil, butocin, carmofur, razoxane, sizofilan, carboplatin, mitolactol, tegafur, ifosfamide, prednimustine, picibanil, levamisole, teniposide, improsulfan, enocitabine, lisuride, oxymetholone, tamoxifen, progesterone, mepitiostane, epitiostanol, formestane, interferon-alpha, interferon-2 alpha, interferon-beta, interferon-gamma (IFN-γ), colony stimulating factor-1, colony stimulating factor-2, denileukin diftitox, interleukin-2, luteinizing hormone releasing factor and variations of the aforementioned agents that may exhibit differential binding to its cognate receptor, or increased or decreased serum half-life.

An additional class of agents that may be used as part of a combination therapy in treating cancer is immune checkpoint inhibitors. Exemplary immune checkpoint inhibitors include agents that inhibit one or more of (i) cytotoxic T lymphocyte-associated antigen 4 (CTLA4), (ii) programmed cell death protein 1 (PD1), (iii) PDL1, (iv) LAG3, (v) B7-H3, (vi) B7-H4, and (vii) TIM3. The CTLA4 inhibitor ipilimumab has been approved by the United States Food and Drug Administration for treating melanoma.

Yet other agents that may be used as part of a combination therapy in treating cancer are monoclonal antibody agents that target non-checkpoint targets (e.g., herceptin) and non-cytotoxic agents (e.g., tyrosine-kinase inhibitors).

Yet other categories of anti-cancer agents include, for example: (i) an inhibitor selected from an ALK Inhibitor, an ATR Inhibitor, an A2A Antagonist, a Base Excision Repair Inhibitor, a Bcr-Abl Tyrosine Kinase Inhibitor, a Bruton's Tyrosine Kinase Inhibitor, a CDC7 Inhibitor, a CHK1 Inhibitor, a Cyclin-Dependent Kinase Inhibitor, a DNA-PK Inhibitor, an Inhibitor of both DNA-PK and mTOR, a DNMT1 Inhibitor, a DNMT1 Inhibitor plus 2-chloro-deoxyadenosine, an HDAC Inhibitor, a Hedgehog Signaling Pathway Inhibitor, an IDO Inhibitor, a JAK Inhibitor, a mTOR Inhibitor, a MEK Inhibitor, a MELK Inhibitor, a MTH1 Inhibitor, a PARP Inhibitor, a Phosphoinositide 3-Kinase Inhibitor, an Inhibitor of both PARP1 and DHODH, a Proteasome Inhibitor, a Topoisomerase-II Inhibitor, a Tyrosine Kinase Inhibitor, a VEGFR Inhibitor, and a WEE1 Inhibitor; (ii) an agonist of OX40, CD137, CD40, GITR, CD27, HVEM, TNFRSF25, or ICOS; and (iii) a cytokine selected from IL-12, IL-15, GM-CSF, and G-CSF.

Proteins of the present application can also be used as an adjunct to surgical removal of the primary lesion.

The amount of multispecific binding protein and additional therapeutic agent and the relative timing of administration may be selected in order to achieve a desired combined therapeutic effect. For example, when administering a combination therapy to a patient in need of such administration, the therapeutic agents in the combination, or a pharmaceutical composition or compositions comprising the therapeutic agents, may be administered in any order such as, for example, sequentially, concurrently, together, simultaneously and the like. Further, for example, a multispecific binding protein may be administered during a time when the additional therapeutic agent(s) exerts its prophylactic or therapeutic effect, or vice versa.

V. Pharmaceutical Compositions

The present disclosure also features pharmaceutical compositions that contain a therapeutically effective amount of a protein described herein. The composition can be formulated for use in a variety of drug delivery systems. One or more physiologically acceptable excipients or carriers can also be included in the composition for proper formulation. Suitable formulations for use in the present disclosure are found in Remington's Pharmaceutical Sciences, Mack Publishing Company, Philadelphia, Pa., 17th ed., 1985. For a brief review of methods for drug delivery, see, e.g., Langer (Science 249:1527-1533, 1990).

The intravenous drug delivery formulation of the present disclosure may be contained in a bag, a pen, or a syringe. In certain embodiments, the bag may be connected to a channel comprising a tube and/or a needle. In certain embodiments, the formulation may be a lyophilized formulation or a liquid formulation. In certain embodiments, the formulation may be freeze-dried (lyophilized) and contained in about 12-60 vials. In certain embodiments, the formulation may be freeze-dried and 45 mg of the freeze-dried formulation may be contained in one vial. In certain embodiments, the about 40 mg to about 100 mg of freeze-dried formulation may be contained in one vial. In certain embodiments, freeze-dried formulation from 12, 27, or 45 vials are combined to obtain a therapeutic dose of the protein in the intravenous drug formulation. In certain embodiments, the formulation may be a liquid formulation and stored as about 250 mg/vial to about 1000 mg/vial. In certain embodiments, the formulation may be a liquid formulation and stored as about 600 mg/vial. In certain embodiments, the formulation may be a liquid formulation and stored as about 250 mg/vial.

The protein could exist in a liquid aqueous pharmaceutical formulation including a therapeutically effective amount of the protein in a buffered solution forming a formulation.

These compositions may be sterilized by conventional sterilization techniques, or may be sterile filtered. The resulting aqueous solutions may be packaged for use as-is, or lyophilized, the lyophilized preparation being combined with a sterile aqueous carrier prior to administration. The pH of the preparations typically will be between 3 and 11, more preferably between 5 and 9 or between 6 and 8, and most preferably between 7 and 8, such as 7 to 7.5. The resulting compositions in solid form may be packaged in multiple single dose units, each containing a fixed amount of the above-mentioned agent or agents. The composition in solid form can also be packaged in a container for a flexible quantity.

In certain embodiments, the present disclosure provides a formulation with an extended shelf life including a protein of the present disclosure, in combination with mannitol, citric acid monohydrate, sodium citrate, disodium phosphate dihydrate, sodium dihydrogen phosphate dihydrate, sodium chloride, polysorbate 80, water, and sodium hydroxide.

In certain embodiments, an aqueous formulation is prepared including a protein of the present disclosure in a pH-buffered solution. The buffer may have a pH ranging from about 4 to about 8, e.g., from about 4.5 to about 6.0, or from about 4.8 to about 5.5, or may have a pH of about 5.0 to about 5.2. Ranges intermediate to the above recited pH's are also intended to be part of this disclosure. For example, ranges of values using a combination of any of the above recited values as upper and/or lower limits are intended to be included. Examples of buffers that will control the pH within this range include acetate (e.g., sodium acetate), succinate (such as sodium succinate), gluconate, histidine, citrate and other organic acid buffers.

In certain embodiments, the formulation includes a buffer system which contains citrate and phosphate to maintain the pH in a range of about 4 to about 8. In certain embodiments the pH range may be from about 4.5 to about 6.0, or from about pH 4.8 to about 5.5, or in a pH range of about 5.0 to about 5.2. In certain embodiments, the buffer system includes citric acid monohydrate, sodium citrate, disodium phosphate dihydrate, and/or sodium dihydrogen phosphate dihydrate. In certain embodiments, the buffer system includes about 1.3 mg/mL of citric acid (e.g., 1.305 mg/mL), about 0.3 mg/mL of sodium citrate (e.g., 0.305 mg/mL), about 1.5 mg/mL of disodium phosphate dihydrate (e.g., 1.53 mg/mL), about 0.9 mg/mL of sodium dihydrogen phosphate dihydrate (e.g., 0.86 mg/mL), and about 6.2 mg/mL of sodium chloride (e.g., 6.165 mg/mL). In certain embodiments, the buffer system includes about 1 to about 1.5 mg/mL of citric acid, about 0.25 to about 0.5 mg/mL of sodium citrate, about 1.25 to about 1.75 mg/mL of disodium phosphate dihydrate, about 0.7 to about 1.1 mg/mL of sodium dihydrogen phosphate dihydrate, and about 6.0 to about 6.4 mg/mL of sodium chloride. In certain embodiments, the pH of the formulation is adjusted with sodium hydroxide.

A polyol, which acts as a tonicifier and may stabilize the antibody, may also be included in the formulation. The polyol is added to the formulation in an amount which may vary with respect to the desired isotonicity of the formulation. In certain embodiments, the aqueous formulation may be isotonic. The amount of polyol added may also be altered with respect to the molecular weight of the polyol. For example, a lower amount of a monosaccharide (e.g., mannitol) may be added, compared to a disaccharide (such as trehalose). In certain embodiments, the polyol which may be used in the formulation as a tonicity agent is mannitol. In certain embodiments, the mannitol concentration may be about 5 to about 20 mg/mL. In certain embodiments, the concentration of mannitol may be about 7.5 to about 15 mg/mL. In certain embodiments, the concentration of mannitol may be about 10 to about 14 mg/mL. In certain embodiments, the concentration of mannitol may be about 12 mg/mL. In certain embodiments, the polyol sorbitol may be included in the formulation.

A detergent or surfactant may also be added to the formulation. Exemplary detergents include nonionic detergents such as polysorbates (e.g., polysorbates 20, 80 etc.) or poloxamers (e.g., poloxamer 188). The amount of detergent added is such that it reduces aggregation of the formulated antibody and/or minimizes the formation of particulates in the formulation and/or reduces adsorption. In certain embodiments, the formulation may include a surfactant which is a polysorbate. In certain embodiments, the formulation may contain the detergent polysorbate 80 or Tween 80. Tween 80 is a term used to describe polyoxyethylene (20) sorbitanmonooleate (see Fiedler, Lexikon der Hifsstoffe, Editio Cantor Verlag Aulendorf, 4th ed., 1996). In certain embodiments, the formulation may contain between about 0.1 mg/mL and about 10 mg/mL of polysorbate 80, or between about 0.5 mg/mL and about 5 mg/mL. In certain embodiments, about 0.1% polysorbate 80 may be added in the formulation.

In embodiments, the protein product of the present disclosure is formulated as a liquid formulation. The liquid formulation may be presented at a 10 mg/mL concentration in either a USP/Ph Eur type I 50R vial closed with a rubber stopper and sealed with an aluminum crimp seal closure. The stopper may be made of elastomer complying with USP and Ph Eur. In certain embodiments vials may be filled with 61.2 mL of the protein product solution in order to allow an extractable volume of 60 mL. In certain embodiments, the liquid formulation may be diluted with 0.9% saline solution.

In certain embodiments, the liquid formulation of the disclosure may be prepared as a 10 mg/mL concentration solution in combination with a sugar at stabilizing levels. In certain embodiments the liquid formulation may be prepared in an aqueous carrier. In certain embodiments, a stabilizer may be added in an amount no greater than that which may result in a viscosity undesirable or unsuitable for intravenous administration. In certain embodiments, the sugar may be disaccharides, e.g., sucrose. In certain embodiments, the liquid formulation may also include one or more of a buffering agent, a surfactant, and a preservative.

In certain embodiments, the pH of the liquid formulation may be set by addition of a pharmaceutically acceptable acid and/or base. In certain embodiments, the pharmaceutically acceptable acid may be hydrochloric acid. In certain embodiments, the base may be sodium hydroxide.

In addition to aggregation, deamidation is a common product variant of peptides and proteins that may occur during fermentation, harvest/cell clarification, purification, drug substance/drug product storage and during sample analysis. Deamidation is the loss of NH₃from a protein forming a succinimide intermediate that can undergo hydrolysis. The succinimide intermediate results in a 17 dalton mass decrease of the parent peptide. The subsequent hydrolysis results in an 18 dalton mass increase. Isolation of the succinimide intermediate is difficult due to instability under aqueous conditions. As such, deamidation is typically detectable as 1 dalton mass increase. Deamidation of an asparagine results in either aspartic or isoaspartic acid. The parameters affecting the rate of deamidation include pH, temperature, solvent dielectric constant, ionic strength, primary sequence, local polypeptide conformation and tertiary structure. The amino acid residues adjacent to Asn in the peptide chain affect deamidation rates. Gly and Ser following an Asn in protein sequences results in a higher susceptibility to deamidation.

In certain embodiments, the liquid formulation of the present disclosure may be preserved under conditions of pH and humidity to prevent deamination of the protein product.

The aqueous carrier of interest herein is one which is pharmaceutically acceptable (safe and non-toxic for administration to a human) and is useful for the preparation of a liquid formulation. Illustrative carriers include sterile water for injection (SWFI), bacteriostatic water for injection (BWFI), a pH buffered solution (e.g., phosphate-buffered saline), sterile saline solution, Ringer's solution or dextrose solution.

A preservative may be optionally added to the formulations herein to reduce bacterial action. The addition of a preservative may, for example, facilitate the production of a multi-use (multiple-dose) formulation.

Intravenous (IV) formulations may be the preferred administration route in particular instances, such as when a patient is in the hospital after transplantation receiving all drugs via the IV route. In certain embodiments, the liquid formulation is diluted with 0.9% Sodium Chloride solution before administration. In certain embodiments, the diluted drug product for injection is isotonic and suitable for administration by intravenous infusion.

In certain embodiments, a salt or buffer components may be added in an amount of 10 mM-200 mM. The salts and/or buffers are pharmaceutically acceptable and are derived from various known acids (inorganic and organic) with “base forming” metals or amines. In certain embodiments, the buffer may be phosphate buffer. In certain embodiments, the buffer may be glycinate, carbonate, citrate buffers, in which case, sodium, potassium or ammonium ions can serve as counterion.

A protein of the present disclosure could exist in a lyophilized formulation including the proteins and a lyoprotectant. The lyoprotectant may be sugar, e.g., disaccharides. In certain embodiments, the lyoprotectant may be sucrose or maltose. The lyophilized formulation may also include one or more of a buffering agent, a surfactant, a bulking agent, and/or a preservative.

The amount of sucrose or maltose useful for stabilization of the lyophilized drug product may be in a weight ratio of at least 1:2 protein to sucrose or maltose. In certain embodiments, the protein to sucrose or maltose weight ratio may be of from 1:2 to 1:5.

In certain embodiments, the pH of the formulation, prior to lyophilization, may be set by addition of a pharmaceutically acceptable acid and/or base. In certain embodiments the pharmaceutically acceptable acid may be hydrochloric acid. In certain embodiments, the pharmaceutically acceptable base may be sodium hydroxide.

Before lyophilization, the pH of the solution containing a protein of the present disclosure may be adjusted between 6 to 8. In certain embodiments, the pH range for the lyophilized drug product may be from 7 to 8.

In certain embodiments, a “bulking agent” may be added. A “bulking agent” is a compound which adds mass to a lyophilized mixture and contributes to the physical structure of the lyophilized cake (e.g., facilitates the production of an essentially uniform lyophilized cake which maintains an open pore structure). Illustrative bulking agents include mannitol, glycine, polyethylene glycol and sorbitol. The lyophilized formulations of the multispecific binding proteins described in the present application may contain such bulking agents.

In certain embodiments, the lyophilized drug product may be constituted with an aqueous carrier. The aqueous carrier of interest herein is one which is pharmaceutically acceptable (e.g., safe and non-toxic for administration to a human) and is useful for the preparation of a liquid formulation, after lyophilization. Illustrative diluents include sterile water for injection (SWFI), bacteriostatic water for injection (BWFI), a pH buffered solution (e.g., phosphate-buffered saline), sterile saline solution, Ringer's solution or dextrose solution.

In certain embodiments, the lyophilized drug product of the current disclosure is reconstituted with either Sterile Water for Injection, USP (SWFI) or 0.9% Sodium Chloride Injection, USP. During reconstitution, the lyophilized powder dissolves into a solution.

In certain embodiments, the lyophilized protein product of the instant disclosure is constituted to about 4.5 mL water for injection and diluted with 0.9% saline solution (sodium chloride solution).

Actual dosage levels of the active ingredients in the pharmaceutical compositions of multispecific binding proteins described in this application may be varied so as to obtain an amount of the active ingredient which is effective to achieve the desired therapeutic response for a particular patient, composition, and mode of administration, without being toxic to the patient.

The specific dose can be a uniform dose for each patient, for example, 50-5000 mg of protein. Alternatively, a patient's dose can be tailored to the approximate body weight or surface area of the patient. Other factors in determining the appropriate dosage can include the disease or condition to be treated or prevented, the severity of the disease, the route of administration, and the age, sex and medical condition of the patient. Further refinement of the calculations necessary to determine the appropriate dosage for treatment is routinely made by those skilled in the art, especially in light of the dosage information and assays disclosed herein. The dosage can also be determined through the use of known assays for determining dosages used in conjunction with appropriate dose-response data. An individual patient's dosage can be adjusted as the progress of the disease is monitored. Blood levels of the targetable construct or complex in a patient can be measured to see if the dosage needs to be adjusted to reach or maintain an effective concentration. Pharmacogenomics may be used to determine which targetable constructs and/or complexes, and dosages thereof, are most likely to be effective for a given individual (Schmitz et al., Clinica Chimica Acta 308: 43-53, 2001; Steimer et al., Clinica Chimica Acta 308: 33-41, 2001).

In general, dosages based on body weight are from about 0.01 μg to about 100 mg per kg of body weight, such as about 0.01 μg to about 100 mg/kg of body weight, about 0.01 μg to about 50 mg/kg of body weight, about 0.01 μg to about 10 mg/kg of body weight, about 0.01 μg to about 1 mg/kg of body weight, about 0.01 μg to about 100 μg/kg of body weight, about 0.01 μg to about 50 μg/kg of body weight, about 0.01 μg to about 10 μg/kg of body weight, about 0.01 μg to about 1 μg/kg of body weight, about 0.01 μg to about 0.1 μg/kg of body weight, about 0.1 μg to about 100 mg/kg of body weight, about 0.1 μg to about 50 mg/kg of body weight, about 0.1 μg to about 10 mg/kg of body weight, about 0.1 μg to about 1 mg/kg of body weight, about 0.1 μg to about 100 μg/kg of body weight, about 0.1 μg to about 10 μg/kg of body weight, about 0.1 μg to about 1 μg/kg of body weight, about 1 μg to about 100 mg/kg of body weight, about 1 μg to about 50 mg/kg of body weight, about 1 μg to about 10 mg/kg of body weight, about 1 μg to about 1 mg/kg of body weight, about 1 μg to about 100 μg/kg of body weight, about 1 μg to about 50 μg/kg of body weight, about 1 μg to about 10 μg/kg of body weight, about 10 μg to about 100 mg/kg of body weight, about 10 μg to about 50 mg/kg of body weight, about 10 μg to about 10 mg/kg of body weight, about 10 μg to about 1 mg/kg of body weight, about 10 μg to about 100 μg/kg of body weight, about 10 μg to about 50 μg/kg of body weight, about 50 μg to about 100 mg/kg of body weight, about 50 μg to about 50 mg/kg of body weight, about 50 μg to about 10 mg/kg of body weight, about 50 μg to about 1 mg/kg of body weight, about 50 μg to about 100 μg/kg of body weight, about 100 μg to about 100 mg/kg of body weight, about 100 μg to about 50 mg/kg of body weight, about 100 μg to about 10 mg/kg of body weight, about 100 μg to about 1 mg/kg of body weight, about 1 mg to about 100 mg/kg of body weight, about 1 mg to about 50 mg/kg of body weight, about 1 mg to about 10 mg/kg of body weight, about 10 mg to about 100 mg/kg of body weight, about 10 mg to about 50 mg/kg of body weight, about 50 mg to about 100 mg/kg of body weight.

Doses may be given once or more times daily, weekly, monthly or yearly, or even once every 2 to 20 years. Persons of ordinary skill in the art can easily estimate repetition rates for dosing based on measured residence times and concentrations of the targetable construct or complex in bodily fluids or tissues. Administration of the multispecific binding proteins described in the present application could be intravenous, intraarterial, intraperitoneal, intramuscular, subcutaneous, intrapleural, intrathecal, intracavitary, by perfusion through a catheter or by direct intralesional injection. This may be administered once or more times daily, once or more times weekly, once or more times monthly, and once or more times annually.

The description above provides multiple aspects and embodiments of the multispecific binding proteins described in the application. The patent application specifically contemplates all combinations and permutations of the aspects and embodiments. The use of any and all examples, or exemplary language herein, for example, “such as” or “including,” is intended merely to illustrate better the multispecific binding proteins described in the present application, and does not pose a limitation on the scope of the disclosure, unless so expressly stated. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the multispecific binding proteins described in the present application.

EXAMPLES

The following examples are merely illustrative and are not intended to limit the scope or content of the multispecific binding proteins described in the present application in any way.

Example 1
Characterization of Supernatants of Selected Hybridoma Clones

CLEC12A-specific antibodies were generated by immunizing BALB/c mice with hCLEC12A-His fusion protein. Supernatants of 16 hybridomas were assessed for CLEC12A binding by Bio-layer Interferometry (BLI) binding using an OctetRed384 (ForteBio). These 16 hybridomas were further analyzed for binding to human and cynomolgus CLEC12A expressed on the cell surface of isogenic RMA cells; binding to cynomolgus CLEC12A was not observed. Estimated kinetic parameters are presented in Table 11 and binding traces are shown in FIGS. 18A-18P. Nine clones were selected for further study. The ability of these nine clones to bind hCLEC12A-His RMA and CLEC12A-expressing cancer cell lines U937 and PL21 was further analyzed by high resolution surface plasmon resonance (SPR). The experiment was performed at 37° C. to mimic physiological temperature using a Biacore 8K instrument.

TABLE 11

Kinetic parameters and affinities of CLEC12A-His

binding to the antibodies produced from candidate hybridomas

SPR at 37° C.
Cell Binding MFI

Test
Binning

K_D
RMA-
k_a

articles
profile
k_a(1/Ms)
k_d(1/s)
(nM)
hCLEC12A
(1/Ms)
k_d(1/s)

9E04
competitor
247.0
3.51 × 10⁻²
9E04
competitor
247.0
3.51 × 10⁻²

9F11
competitor
549.0
5.57 × 10⁻³
9F11
competitor
549.0
5.57 × 10⁻³

11E02
—
779.0
7.28 × 10⁻³
11E02
—
779.0
7.28 × 10⁻³

12F08
unique
76.7
6.23 × 10⁻³
12F08
unique
76.7
6.23 × 10⁻³

13E01
unique
178.0
7.50 × 10⁻³
13E01
unique
178.0
7.50 × 10⁻³

15A10
unique
79.5
2.23 × 10⁻³
15A10
unique
79.5
2.23 × 10⁻³

15D11
—
372.0
6.64 × 10⁻³
15D11
—
372.0
6.64 × 10⁻³

16B08
competitor
Heterogeneous binding
819.0
16B08
competitor

20D06
unique
819.0
5.25 × 10⁻²
20D06
unique
819.0
5.25 × 10⁻²

23A05
unique
90.2
4.49 × 10⁻²
23A05
unique
90.2
4.49 × 10⁻²

30A09
—
Heterogeneous binding
74.7
30A09
—

6D07
—
Non binder
247.0
6D07
—

12B06
—
Non binder
178.0
12B06
—

20G10
—
Non binder
82.2
20G10
—

30H07
—
Non binder
76.3
30H07
—

32A03
—
Non binder
71.4
32A03
—

Binning of hybridoma fusions compared to reference mAbs was performed by BLI using OctetRed384 (ForteBio). Briefly, hybridoma supernatants were loaded onto anti-mouse IgG capture sensor tips for 15 minutes and equilibrated for 5 minutes in PBSF. Sensors were dipped into 200 nM hCLEC12A-His and allowed to associate for 180 seconds followed by dipping into 100 nM reference CLEC12A mAb. The increase in response units indicated that the hybridoma was a non-competitor to the reference mAb, whereas no increase in signal indicated that the hybridoma did compete with the reference mAb. The VH and VL sequences of these reference antibodies are provided in Table 12.

TABLE 12

Reference antibodies

Anti-CLEC12A

mAbs
Source
Sequence ID
Epitope

Merus-CLL1
Merus
VH [SEQ ID NO: 263]
unknown

US
EVQLVQSGAEVKKPGASVKVSCKASGY

2014/0120096A1
TFTSYYMHWVRQAPGQGLEWMGIINPS

GGSTSYAQKFQGRVTMTRDTSTSTVYM

ELSSLRSEDTAVYYCARGNYGDEFDYW

GQGTLVTVSS

VL[SEQ ID NO: 264]

DIQMTQSPSSLSASVGDRVTITCRASQSIS

SYLNWYQQKPGKAPKLLIYAASSLQSGV

PSRFSGSGSGTDFTLTISSLQPEDFATYYC

QQSYSTPPTFGQGTKVEIK

Genentech-h6E7
Genentech US
VH [SEQ ID NO: 265]
C-type

2016/0075787A1
DIQMTQSPSSLSASVGDRVTITCRASQSV
lectin-like

STSSYNYMHWYQQKPGKPPKLLIKYASN
domain,

LESGVPSRFSGSGSGTDFTLTISSLQPEDF
residues

ATYYCQHSWEIPLTFGQGTKVEIK
142-158

VL [SEQ ID NO: 266]

EVQLVQSGAEVKKPGASVKVSCKASGY

SFTDYYMHWVRQAPGQGLEWIGRINPY

NGAAFYSQNFKDRVTLTVDTSTSTAYLE

LSSLRSEDTAVYYCAIERGADLEGYAMD

YWGQGTLVTVSS

Binning analysis demonstrated that antibodies produced from five of the hybridomas, namely 12F8.G3, 13E01, 15A10.G8, 20D6.A8, and 23A5.D4, did not compete with the reference antibodies for binding to hCLEC12A-His. Binding of hybridomas to isogenic human CLEC12A (hCLEC12A) and cross-reactivity with cynomolgus monkey CLEC12A (cCLEC12A) were also evaluated by measuring the binding of the antibodies to isogenic RMA cells expressing CLEC12A and the U937 AML cancer cell line.

Briefly, RMA cells were transducted with a retroviral vector encoding cCLEC12A or hCLEC12A. Binding of the α-CLEC12A mAbs from crude hybridoma harvests to the hCLEC12A or cCLEC12A isogenic cell lines, as well as CLEC12A+ U937 (ATCC catalog number CRL-1593.2) cancer cell lines, was performed as follows. 100,000 RMA, REH or SEM cells were added per well of a 96 well round bottom plate. Cells were spun down and the pellet was gently dissociated by vortexing. 100 μL of Zombie live/dead dye (PBS+1:2000 dye) were added per well and incubated in the dark at room temperature for 20 minutes. Cells were washed with 200 μL of FACS buffer (PBS+2% FBS). 50 μL of hybridoma supernatants were added to the washed cells and the mixtures were incubated for 30 minutes on ice in the dark. Cells were washed twice with FACS buffer and then 50 μL of anti-mouse Fc-PE secondary reagent (1:200 dilution) were added and incubated for 20 minutes on ice in the dark. After the incubation, the cells were washed and then fixed with 50 μL of 4% paraformaldehyde for 15 minutes on ice. The fixed cells were washed with FACS buffer, resuspended in 200 μL FACS buffer, and stored at 4° C. until ready for acquisition. The samples of cells resuspended in FACS buffer were run on BD FACSCelesta equipped with an HTS (high throughput sampler) to determine the binding affinities of the antibodies to isogenic RMA cells expressing CLEC12A and the U937 AML cancer cell line.

The binding affinities of the hybridoma supernatants to PL21 AML cancer cells (DSMZ catalog number ACC536), a human AML cell line reported to express CLEC12A, were also measured. As shown in Table 11, nine of the clones displayed binding affinity to cancer cells expressing hCLEC12A. Binding to cCLEC12A was not observed.

Example 2
Analysis of Purified Anti-CLEC12A Murine Antibodies

In this Example, kinetic parameters and binding affinities of the purified anti-CLEC12A murine antibodies were analyzed. Based on the analysis described in Example 1, eight hybridomas (9F11.B7, 12F8.G3, 16B8.C8, 15A10.G8, 20D6.A8, 9E4.B7, 13E1.A4, and 23A5.D4) were selected for subcloning and sequencing. Each subclone was purified from the hybridoma culture, and binding to hCLEC12A-His and cCLEC12A-His was assessed by SPR as shown in FIG. 19. The data from these experiments are shown in FIG. 19A. Antibodies 9E4.B7, 9F11.B7, 12F8.G3, and 16B8.C8 bound to hCLEC12A only (FIG. 19A); whereas antibodies 13E1.A4, 15A10.G8, 20D6.A8, and 23A5.D4 bound to both hCLEC12A and cCLEC12A (FIG. 19B). Kinetic constants and binding affinities of hCLEC12A and cCLEC12A to purified murine subcloned mAbs are provided in Table 13.

TABLE 13

Kinetic parameters and affinities of hCLEC12A binding to purified murine

subclones

Human CLEC12A-His
Cyno CLEC12A-His

Test Article
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

9E4.B7
Low affinity heterogenous interaction
No binding

9F11.B7
1.14 × 10⁶
1.57 × 10⁻³
1.4
No binding

12F8.G3
3.97 × 10⁵
1.39 × 10⁻³
3.5
No binding

16B8.C8*
4.05 × 10⁶
4.44 × 10⁻⁵
0.01
No binding

13E1.A4
Low affinity heterogenous interaction
Low affinity heterogenous interaction

15A10.G8
3.01 × 10⁵
4.94 × 10⁻⁴
1.6
1.96 × 10⁵
6.59 × 10⁻⁴
3.3

20D6.A8
1.31 × 10⁷
9.50 × 10⁻²
7.2
9.16 × 10⁵
2.46 × 10⁻²
26

23A5.D4
Low affinity heterogenous interaction
Low affinity heterogenous interaction

The ability of the eight purified subcloned mAbs to bind to CLEC12A expressing cells was assessed by FACS analysis with human CLEC12A-positive PL21 cancer cell line. As shown in FIG. 20, 9E4.B7, 9F11.B7 and 16B8.C8 all bound to the PL21 cell line with sub-nanomolar EC50 values; however, only 9F11.B7 and 16B8.C8 satisfied both recombinant protein binding and cell binding criteria, as 9E4.B7 demonstrated low affinity heterogenous binding to recombinant hCLEC12A-His (Table 14). Although 15A10.G8, 13E1.A4, 20D6.A8 and 23A5.D4 showed binding to recombinant human and cyno CLEC12A in the SPR assay, these mAbs failed to recognize cancer cells, suggesting conformational differences in the binding epitope between recombinant and cell surface expressed CLEC12A. As demonstrated, both Merus-CLL1 and Genentech-h6E7 mAbs bound to PL21 with significantly inferior EC50 values as compared to the novel CLEC12A hybridoma clones.

TABLE 14

Cell binding confirmation of purified mouse

mAbs to human PL21 cell line

Test article
EC₅₀(nM)
Max MFI

9E4.B7
0.64
186

9F11.B7
0.56
217

12F8.G3
Non binder
n/a*

16B8.C8
0.18
300

13E1.A4
Non binder
n/a

15A10.G8
Non binder
n/a

20D6.A8
Non binder
n/a

23A5.D4
Non binder
n/a

Merus-CLL1^#
2.02
201

Genentech-h6E7^#
5
265

To assess if mAbs bind to CLEC12A in a glycosylation independent manner, binding of clones 16B8.C8 and 9F11.B7 to glycosylated (FIGS. 21A and 21D), de-sialylated (FIGS. 21C and 21F) and PNGase-treated hCLEC12A (FIGS. 21B and 21E) was assessed by SPR. As demonstrated by the sensorgrams and in Table 15, both 16B8.C8 and 9F11.B7 bound to de-sialylated and deglycosylated versions of hCLEC12A without loss of affinity, suggesting that the antibody interactions with hCLEC12A are not affected by glycosylation status of the target.

TABLE 15

Kinetic parameters and affinities of 16B8.C8 and 9F11.B7

to differentially glycosylated hCLEC12A by SPR.

Test article
Analyte
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

Murine 16B8. mAb
Glycosylated
4.05 × 10⁶
4.70 × 10⁻⁵
0.012

hCLEC12A

Murine 16B8 mAb
De-sialylated
5.25 × 10⁶
3.62 × 10⁻⁵
0.007

hCLEC12A

Murine 16B8 mAb
De-glycosylated
4.26 × 10⁶
4.45 × 10⁻⁵
0.011

hCLEC12A

Murine F3′-9F11
Glycosylated
1.15 × 10⁶
1.11 × 10⁻³
0.96

hCLEC12A

Murine F3′-9F11
De-sialylated
1.58 × 10⁶
9.80 × 10⁻⁴
0.62

hCLEC12A

Murine F3′-9F11
De-glycosylated
1.11 × 10⁶
3.74 × 10⁻⁴
0.34

hCLEC12A

Example 3
Putative Sequence Liability Analysis

Potential sequence liabilities in CDRs (identified under Chothia) of the 16B8.C8 and 9F11.B7 antibodies were examined. The following potential liabilities were considered: M (potential oxidation site); NG, NS and NT sequence motif (potential deamidation site); DG, DS and DT sequence motif (potential isomerization site); DP sequence motif (potential site for chemical hydrolysis). The results are summarized in Table 16.

TABLE 16

Putative sequence liabilities in the CDRs of selected murine mAbs

Clone ID
Potential sequence liability motif
Location

16B8.C8
DS (Isomerization site)
CDRH3

9F11.B7
M (Oxidation site)
CDRH3

DG (Isomerization site)
CDRH3

NS (Deamidation site)
CDRH1

Variants of these antibodies were designed to remove the putative sequence liability motifs.

Example 4
Humanization of Subclones

Based on the data collected regarding kinetics and affinity for recombinant hCLEC12A protein, binding to cell surface expressed hCLEC12A, and binding to AML cancer cell lines, two mouse hybridoma subclones, namely 16B8.C8 and 9F11.B7, were selected for humanization.

The 16B8.C8 antibody was humanized. Back mutations were introduced in the framework regions to create variants having the VH and VL sequences of scFv-1292 to scFv-1309, and scFv-1602 and scFv-2061.

The 9F11.B7 antibody was humanized. Back mutations were introduced in the framework regions to create variants AB0186 to AB0196.

Example 5
CLEC12A TriNKET Selection

The following criteria were applied in choosing the optimal TriNKET candidate, which led to selection of F3′-1602 TriNKET as the lead candidate:

- Potent enhancement of NK activity against AML cancer cells
- Potency greater than the corresponding monoclonal antibody
- Binding to multiple AML cell lines
- High affinity for recombinant hCLEC12A
- Low affinity for human NKG2D
- Cross-reactivity with cynomolgus monkey (cyno) targets
- Good developability profile:
- Thermostability of scFv≥65° C.
- isoelectric point (pI) between 8-9
- Hydrophobicity Interaction Chromatography (HIC) retention time similar to well behaved therapeutic mAbs
- High specificity for human CLEC12A
- Clean polyspecific reagent (PSR) profile
- Transient expression titers similar to mAb

All 18 humanized scFv variants of clone 16B8.C8 were combined with an A49MI NKG2D Fab arm to produce humanized F3′ CLEC12A TriNKETs. Binding affinities of the 18 semi-purified TriNKETs for hCLEC12A were assessed by SPR in screening. Binding signals for eight F3′-TriNKETs were less than 5 RU (approximately 15% of the expected signal ran in this assay) and, therefore, these TriNKETs were considered as nonbinders to hCLEC12A. The remaining ten F3′-TriNKETs bound to hCLEC12A with <10 nM affinity, as shown in FIGS. 22A-22S and Table 17. However, a potential N-glycosylation site was involuntarily introduced in eight out of the ten F3′-TriNKETs by introducing murine back mutation N in position H85. Only constructs F3′-1295 and F3′-1304 (containing A in position H85) did not present the N-glycosylation sequence liability; therefore, only these two constructs were carried forward for further characterization.

TABLE 17

Kinetics and affinities of hCLEC12A binding to semi-purified

humanized F3′ TriNKETs.

Number of back

scFv
mutations and residues
N-glycosylation
k_a
k_d
K_D

Test article
orientations
reverted to human
sequence liability site
(M⁻¹s⁻¹)
(s⁻¹)
(nM)

F3′-1292
VH-VL
7 BM
Yes
4.02 × 10⁵
1.30 × 10⁻³
3.22

F3′-1293
VH-VL
6 BM, K(H71)V*
Yes
4.58 × 10⁵
1.75 × 10⁻³
3.81

F3′-1294
VH-VL
6 BM, Q(H83)T
Yes
Binding signal < 5 RU

F3′-1295
VH-VL
6 BM, N(H85)A
No
5.20 × 10⁵
1.53 × 10⁻³
2.94

F3′-1296
VH-VL
6 BM, K(H94)R
Yes
8.74 × 10⁵
1.17 × 10⁻³
1.34

F3′-1297
VH-VL
6 BM, I(L43)A
Yes
5.04 × 10⁵
1.41 × 10⁻³
2.79

F3′-1298
VH-VL
6 BM, R(L70)D
Yes
6.44 × 10⁵
1.85 × 10⁻³
2.87

F3′-1299
VH-VL
6 BM, I(L83)F
Yes
4.26 × 10⁵
1.21 × 10⁻³
2.83

F3′-1300
VH-VL
No BM
No
Binding signal < 5 RU

F3′-1301
VL-VH
7 BM
Yes
Binding signal < 5 RU

F3′-1302
VL-VH
6 BM, K(H71)V
Yes
Binding signal < 5 RU

F3′-1303
VL-VH
6 BM, Q(H83)T
Yes
Binding signal < 5 RU

F3′-1304
VL-VH
6 BM, N(H85)A
No
4.80 × 10⁵
1.30 × 10⁻³
2.70

F3′-1305
VL-VH
6 BM, K(H94)R
Yes
Binding signal < 5 RU

F3′-1306
VL-VH
6 BM, I(L43)A
Yes
9.48 × 10⁶
4.89 × 10⁻²
5.16

F3′-1307
VL-VH
6 BM, R(L70)D
Yes
9.14 × 10⁵
1.00 × 10⁻³
1.10

F3′-1308
VL-VH
6 BM, I(L83)F
Yes
Binding signal < 5 RU

F3′-1309
VL-VH
No BM
No
Binding signal < 5 RU

m16B8.C8**
Mouse mAb

Yes
3.03 × 10⁶
2.35 × 10⁻⁵
0.008

*In the “residues reverted to human,” the first letter is a murine residue, and the letter and number in the parentheses indicate positions of reverted residues in the heavy chain and light chain, and the last letter is the human residue, e.g., in K(H71)V, murine K in position 71 of the heavy chain is replaced by human residue V.

**murine parental mAb 16B8.C8 was fully purified.

F3′-1295 and F3′-1304 TriNKETs were tested for binding isogenic to hCLEC12A+RMA cells, as shown in FIG. 23, and using the methods as described in Example 5 or 6. Binding of F3′-1295 to cell-surface expressed hCLEC12A was comparable to hcFAE-A49.CLL1-Merus control TriNKET, derived from the Merus antibody described above in Example 1, whereas binding of F3′-1304 to cell-surface expressed hCLEC12A was poorer compared to the hcFAE-A49.CLL1-Merus control multispecific binding protein.

Potency of F3′-1295 was assessed in the primary NK cell mediated cytotoxicity assay with HL-60 AML cells, as shown in FIG. 24, and using the methods as described in Example 8. F3′-1295 induced potent lysis of HL60 AML cells. Cell killing EC50 was similar, and maximum percentage of cell lysis was higher than control hcFAE-A49.CLL1-Merus TriNKET (Table 18).

TABLE 18

Potency of F3′-1295 in primary NK cells mediated cytotoxicity assay.

Test articles
EC50 (nM)
Max lysis (%)

F3′-1295
1.1
59

hcFAE-A49.CLL1-Merus
1.1
43

Thermostability of F3′-1295 was evaluated by differential scanning calorimetry (DSC). To perform DSC, briefly, TriNKETs were diluted to 0.5 mg/mL with PBS. 325 μL were added to a 96-well deep well plate along with a matching buffer blank. Thermograms were generated using a MicroCal PEAQ DSC (Malvern, Pa.). Temperature was ramped from 20-100° C. at 90° C./hour. Raw thermograms were background subtracted, the baseline model was splined, and data were fitted using a non-two state model.

The melting temperature of the first transition (T_m1) for F3′-1295 was 60.9° C., which is 4° C. lower than pre-defined criteria of ≥65° C. for the T_mof the first TriNKET transition. In the process of humanization, it is possible that the introduction of the murine residue to human framework (back mutation) may have had an impact on the thermal stability. Therefore, by reverting the back mutations to human residues thermal stability of F3′-1295 may improve. Therefore, the thermal stability of the ten F3′variants was assessed by DSC regardless of the presence or absence of the N-glycosylation sequence liability, as shown in Table 19. The first transition T_m(attributed to the scFv domain) of F3′-1297 and F3′-1306 TriNKETs was significantly higher (66.5° C. to 67.5° C.) than the other F3′ TriNKETs. Moreover, the T_onsetof these two molecules was significantly better compared to other F3′ TriNKETs tested. The sole difference between these two F3′-TriNKETs is the orientation of scFv; F3′-1297 (VH-VL orientation) and F3′-1306 (VL-VH orientation).

To determine if the human residue in position (L43) was responsible for an increase in thermostability, in both F3′-1306 and F3′-1297, the mouse residue I(L43) was reverted back to the original human framework residue A(L43). Moreover, the murine Ile in position L43 of F3′-1295 was replaced with human Ala, and then was produced recombinantly. The new second generation humanized TriNKET derived from F3′-1295 was named F3′-1602, or F3′-1602 TriNKET.

TABLE 19

DSC analysis of purified F3′-TriNKETs in PBS.

Mouse residues

scFv
changed to
T_onset
T_m1
T_m2
T_m3
T_m4
T_m5

Test articles
orientation
human
(° C.)
(° C.)
(° C.)
(° C.)
(° C.)
(° C.)

F3′-1292
VH-VL
none
51.4
60.0
67.6
76.4
81.6
83.7

F3′-1293
VH-VL
K(H71)V ^#
51.2
59.8
66.9
76.5
81.6
83.7

F3′-1295
VH-VL
N(H85)A
52.5
60.9
67.1
76.6
81.6
83.4

F3′-1296
VH-VL
K(H94)R
50.5
59.7
67.1
76.3
81.6
83.7

F3′-1297
VH-VL
I(L43)A
56.4
66.5*
76.0
81.5
83.6

F3′-1298
VH-VL
R(L70)D
48.7
59.2
66.5
76.0
81.7
83.8

F3′-1299
VH-VL
I(L83)F
49.6
59.8
66.1
76.0
81.8
83.8

F3′-1304
VL-VH
N(H85)A
50.5
60.1
65.5
76.3
81.7
83.5

F3′-1306
VL-VH
I(L43)A
58.5
67.5
76.2
81.6
83.7

F3′-1307
VL-VH
R(L70)D
50.5
60.2
66.0
76.0
81.8
83.9

^# Chothia position (mouse residue to human residue)

*Overlaps with Tm₂

Thermostability of F3′-1295 and second generation F3′-1602 TriNKETs was evaluated by DSC in PBS (FIGS. 25A-25B). Based on the T_mof the first transition, the scFv of F3′-1602 was thermally stabilized by 7.5° C. compared to F3′-1295, shown in Table 20. This stabilization is attributed to the Ile/Ala mutations. Thus, F3′-1602 met the T_m1cut-off at ≥65° C.

TABLE 20

DSC analysis of F3'-1295 and F3'-1602 TriNKETs in PBS.

Test article
T_onset(° C.)
T_m1(° C.)
T_m2(° C.)
T_m3(° C.)
T_m4(° C.)

F3′-1295
52.5
60.9
67.1
76.6
81.6

F3′-1602
58.5
66.8
76.0
81.5
83.3

To understand if the I(L43)A substitution had an effect on the TriNKET affinity for hCLEC12A, binding of hCLEC12A to F3′-1295 and F3′-1602 was determined by SPR (Biacore) at 37° C. (FIGS. 26A-26H), and using the methods as described in Example 1. The kinetic constants and equilibrium binding affinities are listed in Table 21. Both TriNKETs have very similar off rates, but F3′-1602 displays about 2-fold lower K_Ddue to its faster on rate. Therefore, the I(L43)A substitution had an effect on the K_D.

TABLE 21

Kinetic parameters and affinities of F3′-1295 and F3′-1602

binding to hCLEC12A by SPR.

Test

article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

F3′1295
hCLEC12A-His
4.11 × 10⁵
5.87 × 10⁻⁴
1.43

hCLEC12A-His
4.15 × 10⁵
5.89 × 10⁻⁴
1.42

hCLEC12A-His
4.19 × 10⁵
6.05 × 10⁻⁴
1.44

hCLEC12A-His
4.16 × 10⁵
5.96 × 10⁻⁴
1.43

Average ± StDev
(4.15 ±
(5.94 ±
1.43 ± 0.01

0.03) × 10⁵
0.09) × 10⁻⁴

F3′1602
hCLEC12A-His
8.54 × 10⁵
4.94 × 10⁻⁴
0.58

hCLEC12A-His
8.33 × 10⁵
4.94 × 10⁻⁴
0.59

hCLEC12A-His
8.63 × 10⁵
4.89 × 10⁻⁴
0.57

hCLEC12A-His
8.26 × 10⁵
4.99 × 10⁻⁴
0.60

Average ± StDev
(8.44 ±
(4.94 ±
0.57 ± 0.01

0.17) × 10⁵
0.04) × 10⁻⁴

F3′-1295 and F3′-1602 were tested for their ability to bind isogenic Ba/F3 cells expressing human CLEC12A, in comparison to the parental Ba/F3 cells (as shown in FIG. 27A, and the data listed in Table 22) using the methods as described in Example 5 or 6. F3′-1602 was able to bind hCLEC12A+Ba/F3 with about 2-fold lower EC50 value but with similar maximum binding MFI compared to F3′-1295. No binding to the parental Ba/F3 cell lacking expression of CLEC12A was detected by either F3′-1295 or F3′-1602 (FIG. 27B), suggesting high specificity of TriNKET binding to hCLEC12A.

TABLE 22

EC50 and max MFI of F3′-TriNKETs binding to hCLEC12A +

Ba/F3 isogenic cell line.

hCLEC12A +
Ba/F3 parental

Ba/F3 cell line
cell line

Test article
EC50 (nM)
Max MFI
EC50 (nM)
Max MFI

F3′-1295
26.1
48040
No binding

F3′-1602
14.4
50160
No binding

hcFAE-A49.CLL-Merus
14.3
39740
No binding

F3′-1295 and F3′-1602 were further tested for their ability to bind HL60 (FIG. 27C) and PL21 (FIG. 27D) AML cancer cell lines, and using the methods as described in Example 5 or 6. F3′-1602 was able to bind to both cell lines with lower EC50 values but with similar maximum binding MFI as compared to F3′-1295, as shown in Table 23.

Potency of F3′-1295 and F3′-1602 TriNKETs was assessed in a KHYG-1-CD16aV mediated cytotoxicity assay, the results of which are shown in FIG. 28. Both F3′-1602 and F3′-1295 induced potent lysis of PL21 (FIG. 28A) and HL60 (FIG. 28B) AML cancer cells, and both display similar EC50 values and % max lysis, shown in Table 23.

TABLE 23

Potency of F3'-1294 and F3'1602 in KHYG-1CD16aV

mediated cytotoxicity assay.

PL21 cancer cell line
HL60 cancer cell line

EC50
Max cell
EC50
Max cell

Test articles
(nM)
lysis (%)
(nM)
lysis (%)

F3′-1295
1.23
22
1.15
48

F3′-1602
1.87
25
0.90
49

hcFAE-A49.CLL-Merus
2.97
10
1.83
27

Hydrophobic properties of F3′-1295 and F3′-1602 TriNKETs were evaluated by analytical hydrophobic interaction chromatography (HIC), the results of which are shown in Table 24. Trastuzumab (Roche) was used as a representative control of a therapeutic antibody. The retention times (RT) for both TriNKETs were similar to the RT of Trastuzumab, suggesting that predicted aggregation propensity of both TriNKETs is low.

TABLE 24

HIC retention times for F3'-1295 and F3'-1602

TriNKETs compared to Trastuzumab.

Absolute retention
ΔRT relative to

Test article
time (min)
Trastuzumab (min)

F3′-1295
9.7
+0.6

F3′-1602
9.4
+0.3

Trastuzumab
9.1
0.0

The experimental isoelectric point (pI) of F3′-1295 (FIG. 29A) and F3′-1602 (FIG. 29B) was assessed by capillary isoelectric focusing (cIEF), as described in Example 15. Both constructs showed nearly identical pIs. The major peak of F3′-1295 has a pI of 8.73 and that of F3′-1602 has a pI of 8.81 (Table 25).

TABLE 25

pI of F3′-1295 and F3′-1602 determined by cIEF.

Test article
pI (main peak)

F3′-1295
8.73

F3′-1602
8.81

Off-target effects of a drug need to be evaluated when developing protein therapeutics. A flow cytometry based polyspecificity reagent (PSR) assay allows for determination of antibodies that have a higher probability to bind non-specifically. F3′-1295 and F3′-1602 were tested for non-specific binding to a preparation of detergent solubilized CHO cell membrane proteins in the PSR assay, using the methods as described in Example 12. Both humanized F3′-1602 and F3′-1295 did not bind to PSR (no signal shift to the right) and showed very similar profiles as the PSR control Trastuzumab, demonstrated in FIGS. 30A-30D, suggesting high specificity of the TriNKETs. Rituximab was used as a positive control in this assay. F3′-1602 was selected for further comparison with additional TriNKETs described.

The effect of different TriNKET formats was tested experimentally for alterations in efficacy. To obtain CLEC12A TriNKET in the F3 format, humanized 16B8.C8 was converted using the same humanization variant that was used F3′-1602, into Chains VH-CH1-Fc and VL-CL, whereas NKG2D A49M-I mAb was converted into VH-VL-Fc. The protein received the designation AB0010 or F3-1602.

Binding affinities of human CLEC12A-His for AB0010 (F3 TriNKET format) and F3′-1602 (F3′ TriNKET format) were assessed by SPR (Biacore) at 37° C. The kinetic constants and equilibrium binding affinities are shown in Table 26, and the raw data and fits are shown in FIGS. 31A-31D. Overall binding affinities of both constructs were within 2-fold difference, with AB0010 forming slightly a stronger complex with CLEC12A compared to F3′-1602, as evidenced by the K_D.

TABLE 26

Kinetic parameters and affinities of CLEC12A binding to F3′

(F3′-1602) and F3 (AB0010) TriNKETs.

Format
Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

F3′
F3′1602
hCLEC12A-His
7.91 × 10⁵
9.52 × 10⁻⁴
1.20

F3′
F3′1602
hCLEC12A-His
7.68 × 10⁵
9.13 × 10⁻⁴
1.19

F3′
Average
7.79 × 10⁵
9.33 × 10⁻⁴
1.20

F3
AB0010
hCLEC12A-His
7.15 × 10⁵
5.08 × 10⁻⁴
0.71

F3
AB0010
hCLEC12A-His
7.04 × 10⁵
5.41 × 10⁻⁴
0.77

F3
Average
7.10 × 10⁵
5.25 × 10⁻⁴
0.74

TriNKETs of different formats, F3′-1602 (F3′) and AB0010 (F3), were tested for their abilities to bind Ba/F3 cells expressing human CLEC12A (FIG. 32A) and AML cancer cell line (FIG. 32B). The EC50 value for F3′-1602 was superior over AB0010 in HL-60 cells, about 2-fold decreased (Table 27). Neither AB0010 nor F3′-1602 showed any binding to the parental Ba/F3 cells, demonstrating high specificity of binding to CLEC12A (FIG. 32C).

TABLE 27

Binding EC50 and max MFI of F3′-1602 and

AB0010 to hCLEC12A+ Ba/F3 and HL60 AML cells.

hCLEC12A+ Ba/F3

Ba/F3

cell line
HL60
parent cell line

Test
Max
EC50
Max
EC50
Max
EC50

Format
article
MFI
(nM)
MFI
(nM)
MFI
(nM)

F3′
F3′-1602
36920
3.20
3820
1.1
No binding

F3
AB0010
54827
6.13
6064
4.2
No binding

The potency of AB0010 and F3′-1602 TriNKETs was assessed in KHYG-1-CD16aV mediated cytotoxicity assay, shown in FIG. 33A. F3′-1602 induced lysis of HL60 and PL21 AML cancer cells with a 2-fold better EC50 value than AB0010 (Table 28). Both F3 (AB0010) and F3′ (F3′-1602) TriNKETs showed superiority over corresponding humanized parental mAb (AB0037-001).

TABLE 28

Potency of F3′-1602 and AB0010 in KHYG-1-CD16aV

mediated cytotoxicity assay.

PL21 AML cancer cell line

Format
Test articles
EC50 (nM)
Max cell lysis (%)

F3
AB0010
2.99
21.75

F3′
F3′-1602
1.16
20.97

mAb
AB0037-001
N/A
N/A

The potencies of AB0010 and F3′-1602 TriNKETs were assessed in a human primary NK cell mediated cytotoxicity assay (FIG. 33B). F3′-1602 (F3′ format) induced lysis of PL21 AML cancer cells with a 2-fold lower EC50 value than AB0010 (F3 format) (Table 29).

TABLE 29

Potency of F3'1602 and AB0010 in primary NK cell mediated

cytotoxicity assay.

PL21 cancer cell lines

Format
Test article
EC50 (nM)
Max cell lysis (%)

F3'
F3'-1602
0.13
45.85

F3
AB0010
0.30
43.82

It was observed in both KHYG-1 mediated and primary NK cell mediated cytotoxic assays that F3′-1602 (F3′-format) demonstrated about 2-fold better EC50 than AB0010 (F3-format), whereas the overall maximum percentage of killing was not affected. In accelerated stability and forced degradation studies AB0010 was less structurally stable compared to F3′-1602 (F3′-1602) (see Example 16). Therefore, the CLEC12A TriNKET in the F3′ format (F3′-1602) was chosen to move forward for further study.

For generation of 9F11.B7 TriNKETs, all 12 scFv variants of 9F11.B7 were combined with A49M-I NKG2D Fab arm to produce humanized F3′ TriNKETs. Affinities of the 10 semi-purified (Protein A) TriNKETs for hCLEC12A were assessed by SPR in the screening mode shown in Table 30. Three (AB0190, AB0193 and AB0196) 9F11.B7 based F3′-TriNKETs showed heterogenous binding and could not be fitted to a 1:1 kinetic model with high confidence. The binding kinetics of the remaining 7 TriNKETs were similar to the chimeric parent mouse 9F11.B7 mAb, suggesting that neither humanization nor conversion of Fab to scFv affected the affinity for hCLEC12A.

TABLE 30

Kinetics and affinities of hCLEC12A binding to semi-purified,

humanized F3' TriNKETs.

k_a
k_d
K_D

Test article
(M⁻¹s⁻¹)
(s⁻¹)
(nM)
Comments

AB0185
7.26 × 10⁵
1.01 × 10⁻³
1.40

AB0186
1.12 × 10⁶
1.17 × 10⁻³
1.04

AB0188
5.69 × 10⁵
9.01 × 10⁻⁴
1.58
Low heterodimer yield

AB0189
1.04 × 10⁶
1.27 × 10⁻³
1.21

AB0190
1.55 × 10⁷
1.77 × 10⁻²
1.14
Estimated/heterogeneous

binding

AB0191
7.28 × 10⁵
1.26 × 10⁻³
1.73

AB0192
1.09 × 10⁶
1.38 × 10⁻³
1.26

AB0193
1.97 × 10⁷
2.07 × 10⁻²
1.04
Estimated/heterogeneous

binding

AB0195
7.66 × 10⁵
1.29 × 10⁻³
1.68

AB0196
Heterogenous weak binding; data

insufficient for estimation

m9F11-hIgG1
9.50 × 10⁵
1.02 × 10⁻³
1.07

Binding of 9 fully purified humanized 9F11.B7 TriNKETs to hCLEC12A+Ba/F3 and HL60 AML cancer cell lines, shown in FIG. 34A and FIG. 34B, was tested by FACS. AB0190, AB0193 and AB0196 showed inferior binding EC50 values for both cell lines (Table 31), which correlates with poor behavior in SPR described in Table 30. EC50 values for the remaining 6 clones were similar, showing correlation with the SPR data.

TABLE 31

Binding EC50 and max MFI of 9F11.B7 based F3'-TriNKETs

to hCLEC12A + Ba/F3 isogenic and HL60 AML cell lines

hCLEC12A + Ba/F3
HL60 cancer

isogenic cell line
cell line

Max

Max

Test

binding
EC50
binding

Article
EC50 (nM)
MFI
(nM)
MFI

AB0185
0.7
8025
0.4
372

AB0186
0.7
8160
0.5
412

AB0189
1.3
8760
0.7
420

AB0190
2.9
5139
N/A
n/a

AB0191
0.6
7941
0.6
420

AB0192
0.7
10158
0.6
503

AB0193
2.9
7732
19
345

AB0195
0.6
10775
0.7
551

AB0196
2.7
7427
135
650

F3'-1602
2.0
14712
1.6
924

Potencies of 9F11.B7-based TriNKETs were assessed in KHYG-1-CD16aV cell mediated cytotoxicity assay (FIG. 35 and Table 32). AB0190, AB0193 and AB0196 displayed minimal activity. AB0189 and AB0192 showed the highest potency among other TriNKET candidates. AB0192 TriNKET emerged as the lead TriNKET from 9F11.B7 humanization efforts.

TABLE 32

Potency of humanized 9F11.B7 F3'-TriNKETs in

KHYG-1CD16aV mediated cytotoxicity assay.

HL60 AML cancer cell line

Test article
EC50 (nM)
Max cell lysis (%)

AB0185
2.0
25

AB0186
2.7
19

AB0189
0.6
18

AB0190
n/a
n/a

AB0191
3.1
22

AB0192
1.3
22

AB0193
n/a
n/a

AB0195
2.2
20

AB0196
n/a
n/a

The potency of AB0192 was assessed in a KHYG-1-CD16aV cell mediated cytotoxicity assay in comparison with the F3′-1602 (FIG. 36A). F3′-1602 showed improved efficacy over AB0192 both in EC50 (0.4 nM and 1.3 nM, respectively) and percentage maximum HL-60 cell lysis (41% and 22%, respectively).

The potencies of AB0192 and F3′-1602 TriNKETs were further assessed in a human primary NK cell mediated cytotoxicity assay (FIG. 36B). F3′-1602 and AB0192 induced lysis of HL60 AML cancer with similar EC50 and % max cell killing (Table 33).

AB0192 satisfies TriNKET pre-developability criteria: high affinity for hCLEC12A, efficient binding to AML cells, high potency in killing AML cells in primary NK cells mediated cytotoxicity assay, high thermostability, T_m1≥65° C., low hydrophobicity, 8.0<PI<9.0, expression level in transient transfection similar to monoclonal antibody, clean PSR (Table 33).

F3′-1602 demonstrated better potency , higher humanness, and fewer potential sequence liabilities compared to AB0192. Structural stability of F3′-1602 was confirmed in an extensive panel of accelerated stability and forced degradation studies, described in Example 15. Therefore, F3′-1602 was chosen for further development.

TABLE 33

Summary of biochemical characterization of F3'-1602 (F3'-1602)

and AB0192.

Properties
F3'-1602
AB0192

Number of back
5
7

mutations

% humanness
80.09 (VH),
82.3 (VH),

T20 score
84.67(VK)
81.2 (VK)

Potential seq
DS
M (oxidation site)

liabilities
(isomerization
in CDRH3

site) in
DG (isomerization site)

CDRH3
in CDRH3

NS (deamidation site)

in CDRH1

HIC retention time
+0.91
+0.97

relative to Humira (min)

pI (cIEF, major peak)
8.81
N/A

Thermal stability by DSC
66.8
65.0

(Tm1° C.) in PBS

Expression Titer
73.3
64.4

(ExpiCHO, mg/L)
50.2

EC50 (nM): Potency
0.4
1.3

(KHYG-CD16a

mediated, HL60 cell line)

EC50 (nM): Potency
0.29
0.25

(Primary NK cell

mediated, HL60 cell line)

Affinity to hCLEC12A
0.80
1.2

by SPR (K_Din nM)

PSR specificity
Clean
Clean

F3′-1602 and its corresponding humanized mAb (h16B8.C8) were assessed for cell killing by primary NK cell mediated cytotoxicity. F3′-1602 demonstrated potent killing of PL21 (FIG. 37A) and THP-1 (FIG. 37B) AML cancer cell lines, whereas the corresponding monoclonal antibody markedly reduced potency (Table 34).

TABLE 34

Potency of F3'-1602 in comparison with parental mAb in

primary NK mediated cytotoxicity assay.

F3'-1602
Parent mAb

Cell line
EC50 (nM)
Max lysis (%)
EC50 (nM)
Max lysis (%)

PL21
0.07
35
0.5
18

THP-1
0.7
30
NA
NA

Example 6
Assessment of TriNKET Binding to Cell Expressed Human Cancer Antigens

Isogenic Ba/F3 cells expressing hCLEC12A, and human cancer cell lines HL60 and PL21 expressing human CLEC12A were used to assess tumor antigen binding of CLEC12A targeted TriNKETs and mAbs.

AB0089 is an F3′ TriNKET derived from the same binding domain as parental mAb AB0305. AB0192 is an F3′ TriNKET derived from the same binding domain as parental mAb AB0192. AB0237 is an F3′ TriNKET, also referred to herein as hF3′-1602 or F3′-1602 TriNKET, and is derived from scFv-1602. AB0300 (also referred to as F3′-1602si or AB0237si) is also an F3′ TriNKET from the same sequence as AB0237 but bears mutations in the CH2 domain to eliminate Fc-gamma receptor binding. F3′ TriNKETs generally showed weaker binding compared to parental mAb.

To assess whether the antigen-binding site that binds NKG2D and the effector function of the Fc contributed to the cytotoxicity of AB0237, two TriNKET variants were constructed. The first variant contains mutations in the light chain variable domain of the A49 antigen-binding site that binds NKG2D. As a result, this variant does not bind human NKG2D. The amino acid sequence of the mutant light chain variable domain is DIQMTQSPSTLSASVGDRVTITCRASNSIS SWLAWYQQKPGKAPKLLIYEAS STKSGVPS RFSGSGSGTEFTLTISSLQPDDFATYYCQQYDDLPTFGGGTKVEIK (SEQ ID NO:282). The amino acid sequences of this first variant are otherwise identical to those of AB0237. The second variant contains mutations in the Fc domain. Specifically, each polypeptide chain of the Fc domain contains L234A/L235A/P329G substitutions (numbered under EU numbering system), which was reported to reduce the binding of the Fc to Fcγ receptors. The amino acid sequences of this second variant are otherwise identical to those of AB0237. F3′-palivizumab contains the same NKG2D and CD16 binder sequences as in F3′-1602 and an scFv that binds an irrelevant target not expressed in either of the two cancer cell lines tested. EC50 values are provided in Table 35.

TABLE 35

EC50 values and % max cell lysis in KHYG-1-CD16aV-mediated

potency assay.

PL-21
HL-60

Test article
EC50 nM
% Max lysis
EC50 nM
% Max lysis

F3'-1602
0.7
33
0.27
87

F3'-1602-Dead-
NA
2
2.5
33

2D

F3'-1602 si
NA
NA
NA
12

F3'-Palivizumab
NA
NA
NA
NA

The cells were then incubated with a fluorophore conjugated anti-human IgG secondary antibody and were analyzed by flow cytometry. The mean fluorescence intensity (MFI) values were normalized to secondary antibody only controls to obtain fold over background (FOB) values.

AB0237 demonstrates high affinity binding to the two AML cell lines tested, namely, HL60 (FIG. 38A) and PL21 (FIG. 38B). The EC50 for HL60 was 2.2 nM and the EC50 for PL21 was 1.2 nM.

For two of the three binders F3′ TriNKETs showed higher loading onto target cells compared to its parental mAb. Mutated variants of AB0237, AB0300 and AB0237 NKG2D dead arm, showed similar binding as their parental TriNKET AB0237 to Ba/F3-hCLEC12A (FIG. 39A), HL-60 (FIG. 39B), and PL-21 (FIG. 39C).

Example 7
Assessment of TriNKET Surface Retention to Cell Expressed Human Cancer Antigens

Human cancer cell lines HL60 and PL21 expressing human CLEC12A were used to assess surface retention of CLEC12A-TriNKETs after incubation with cells. PL21 or HL60 cells in duplicate plates were incubated with TriNKETs or mAbs as indicated. The first plate was moved to 37° C. for two hours, whereas the second plate was incubated on ice for 20 minutes. After the respective incubation times, cells were washed. Surface bound TriNKET or mAb was detected using a fluorescent secondary antibody for flow cytometry.

Surface retention was calculated as follows:

% surface retention=(sample MFI 2 hrs/sample MFI 20 min))*100%

FIG. 40A and FIG. 40B show cell surface retention of TriNKETs bound to CLEC12A+ cell lines, PL21 and HL60, respectively. Monovalent binding TriNKETs, AB0089, AB0147, AB0192 and AB0237 showed high surface retention at two hours on both PL21 and HL60 cells lines. For both PL21 and HL60, bivalent binding mAb AB0037 showed lower surface retention at two hours compared to TriNKETs.

Example 8
Human NK Cell Cytotoxicity Assay

Lysis of target cells was measured by the DELFIA cytotoxicity assay. Briefly, human cancer cell lines expressing CLEC12A were harvested from culture, washed with HBS, and resuspended in growth media at 10⁶/mL for labeling with BATDA reagent (Perkin Elmer AD0116). Manufacturer instructions were followed for labeling of the target cells. After labeling, cells were washed three times with HBS, and were resuspended at 0.5-1.0×10⁵/mL in culture media. 100 μl of BATDA labeled cells were added to each well of the 96-well plate. Monoclonal antibodies or TriNKETs against CLEC12A were diluted in culture media, and 50 μl of diluted mAb or TriNKET were added to each well.

To prepare NK cells, PBMCs were isolated from human peripheral blood buffy coats using density gradient centrifugation, washed, and prepared for NK cell isolation. NK cells were isolated using a negative selection technique with magnetic beads. Purity of isolated NK cells was typically >90% CD3−CD56+. Isolated NK cells were rested overnight and harvested from culture. The cells were then washed and resuspended at concentrations of 10⁵-2.0×10⁶/mL in culture media for an effector-to-target (E:T) ratio of 5:1. 50 μl of NK cells were added to each well of the plate for a total of 200 μl culture volume. The plate was incubated at 37° C. with 5% CO₂for 2-3 hours.

After the incubation, the plate was removed from the incubator and the cells were pelleted by centrifugation at 200×g for 5 minutes. 20 μl of culture supernatant were transferred to a clean microplate and 200 μl of room temperature europium solution (Perkin Elmer C135-100) were added to each well. The plate was protected from light and incubated on a plate shaker at 250 rpm for 15 minutes, then read using SpectraMax i3X instruments.

Spontaneous release of a substance that can form a fluorescent chelate with europium was measured in target cells incubated in the absence of NK cells. Maximum release of such a substance was measured in target cells lysed with 1% Triton-X. % Specific lysis was calculated as follows:

% Specific lysis=((Experimental release−Spontaneous release)/(Maximum release−Spontaneous release))*100%.

FIG. 41A and FIG. 41B show CLEC12A-TriNKET mediated killing of PL21 and HL60 target cells by rested human NK cells, respectively. AB0037 a CLEC12A targeted monoclonal antibody showed little enhancement of NK cell mediated lysis of PL21 and HL60 target cells. All three CLEC12A targeted TriNKETs (AB0237, AB0192 and AB0089) showed superior lysis of target cells compared to the CLEC12A targeted monoclonal antibody. EC50 values are shown in Table 36.

TABLE 36

Potency of F3'-1602 in comparison with parental mAb in

primary NK mediated cytotoxicity assay.

F3'-1602
Parent mAb

Cell line
EC50 (nM)
Max lysis (%)
EC50 (nM)
Max lysis (%)

PL-21
0.07
35
0.5
18

THP-1
0.7
30
NA
NA

To confirm the cytoxicity findings, a second NK cell line, KHYG-1-CD16aV was engineered to stably express CD16aV and NKG2D, and the assay was conducted as described above. Cytotoxicity of PL21 is shown in FIG. 42A; Cytotoxicity of HL60 is shown in FIG. 42B. The EC50 values and maximum lysis values were derived from the cell lysis curves by the GraphPad Prism software using four parameter logistic non-linear regression curve fitting model (Table 37). F3′-1602 shows subnanomolar EC50 and high efficiency maximum lysis in the KHYG-1 CD16A mediated cytotoxicity assay.

TABLE 37

Potency of F3'-1602 in KHYG-1-CD16aV mediated cytotoxicity assay.

Test articles
Cell line
EC50
% Max lysis

F3'-1602 #1
PL21
0.3
66

F3'-1602 #2
HL60
0.2
77

Example 9
Primary Human CD8+ T Cell Cytotoxicity Assay

Lysis of target cells was measured by the DELFIA cytotoxicity assay as described in Example 8, except that CD8+ T cells were used as immune effector cells. CD8+ T cells were prepared as follows: Human PBMCs were isolated from human peripheral blood buffy coats using density gradient centrifugation with Lymphoprep and SepMate 50, according to the manufacturer's instructions. Isolated PBMCs were stimulated with 1 μg/mL ConA (an IL-2 culture supplement) in culture media at 37° C. for 18 hours. ConA was then removed and PBMCs were cultured with 25 units/mL IL-2 at 37° C. for 4 days. CD8+ T cells were purified using a negative selection technique with magnetic beads (EasySep™ Human CD8+ T Cell Isolation Kit), according to the manufacturer's instructions. CD8+ T cells were cultured in media containing 10 ng/mL IL-15 at 37° C. for 6-13 days before using in the cytolysis assay.

As shown in FIG. 43, AB0237 TriNKET, but not parental AB0037 mAb, significantly enhanced the ability of CD8 T cell to lyse target cells. AB0237 TriNKET increased CD8+ T cell-mediated lysis of cells in a dose-dependent manner.

To assess whether the antigen-binding site that binds NKG2D and the effector function of the Fc contributed to the cytotoxicity of AB0237 TriNKET, a TriNKET variant was constructed. The variant contains mutations in the light chain variable domain of the A49 antigen-binding site that binds NKG2D. As a result, this variant does not bind human NKG2D. The amino acid sequence of the mutant light chain variable domain is DIQMTQSPSTLSASVGDRVTITCRASNSIS SWLAWYQQKPGKAPKLLIYEAS STKSGVPS RFSGSGSGTEFTLTISSLQPDDFATYYCQQYDDLPTFGGGTKVEIK (SEQ ID NO:282). The amino acid sequences of this first variant are otherwise identical to those of AB0237 TriNKET.

The ability of the variant to induce NK cell-mediated lysis of target cells was assessed using the DELFIA cytotoxicity assay described above. As shown in FIG. 43, the mutations in the NKG2D-targeting domain substantially reduced cytotoxicity. This result suggested that binding to NKG2D and binding to Fcγ receptors both contributed to the cytotoxic activity of AB0237 TriNKET.

Example 10
Assessment of TriNKET or mAb Binding to Whole Human Blood

The ability of AB0237 and mAb AB0037 to bind different types of blood cells was assessed. Briefly, human whole blood was incubated with AB0237 (grey), mAb AB0037 (white), or a human IgG1 isotype control antibody (black). The blood cells were analyzed by flow cytometry and binding of AB0237, mAb AB0037, or the isotype control antibody was detected using a fluorophore conjugated anti-human IgG secondary antibody.

As shown in FIGS. 44A-44I, staining was observed on monocytes, granulocytes, and a proportion of dendritic cells in whole blood for AB0237 and its parental mAb AB0037.

Example 11
Binding of F3′-1602 TriNKET to Fc Receptors: Label Free Surface Plasmon Resonance Characterization

The binding affinities of F3′-1602 TriNKET for recombinant human and cynomolgus Fcγ Receptors was determined and compared the binding affinities of an IgG1 isotype control, trastuzumab. Binding affinity of F3′-1602 TriNKET to human and cynomolgus monkey (cyno) FcRn at pH 6.0 and pH 7.4 was also determined.

FcγR Binding

Human and cyno CD64, high and low affinity alleles of CD16a (V158 and F158, respectively), two alleles of CD32a (H131 and R131), CD16b, CD32b and cynomolgus CD16 were captured via biotin on a primed Biacore chip, prepared according to the manufacturer's instructions. Prior to FcγR binding experiment requiring a high starting concentration, F3′-1602 TriNKET and trastuzumab, samples were buffer exchanged into 1× HBS-EP+ SPR running buffer by three consecutive dilution and concentration cycles with the use of a 0.5 mL concentrator unit, resulting in >300-fold dilution of original buffer components while maintaining a high concentration of protein. Protein concentration was measured at absorbance of A280 on a NanoDrop instrument with the use of sample appropriate theoretical molar extinction coefficients (198405 M⁻¹cm⁻¹for F3′-1602 TriNKET and standard IgG setting for trastuzumab).

Each FcγR binding experiment was performed with multiple cycles utilizing 2-fold serially diluted F3′-1602 TriNKET or trastuzumab. For no-signal reference, a 0 nM blank cycle at the start of each concentration series and receptor free chip surface were used. Each experimental cycle included association, dissociation and regeneration steps. All steps were executed at a flow rate of 30 μL/min and at 25° C. Analyte (F3′-1602 TriNKET or trastuzumab) concentrations, association and dissociation times, and regeneration parameters were FcγR-specific (Table 38).

TABLE 38

Parameters used in Biacore FcγR binding experiments

hCD32a

hCD64,
hCD16a
hCD16a
cyno
H131,

Parameter
cyno CD64
V158
F158
CD16
R131
hCD16b
hCD32b

Type of chip
CAP chip
Type S SA chip

FcγR capture
35-50
34-37
24-27
5-15
10-15

level, RU

Concentration
400-1.56
1500-0.023
6000-0.023
16000-125
24000-188

range, nM

Association
240
150
60
60

time, sec

Dissociation
600
300
60
150

time, sec

Surface
CAPture kit
5 sec pulse of 1-2 mM NaOH
5 sec of 10 mM Gly,

regeneration
regeneration

pH 3.0

The shape of the raw sensorgrams representing F3′-1602 TriNKET and trastuzumab binding recombinant human CD64 enabled fitting the data with a 1:1 kinetic fit model. The black traces representing the 1:1 kinetic fit closely follow the actual sensorgrams depicted in color in FIGS. 45A-45D.

The χ²error value calculated by the Biacore Insight evaluation software was ≤1.9% of calculated R_max, representing a good fit. Table 39 summarizes kinetic rates and human CD64 affinity values for F3′-1602 TriNKET and trastuzumab.

TABLE 39

Kinetic parameters and affinity values of F3'-1602 TriNKET

binding to human CD64.

Sample
k_a(1/Ms)
k_d(1/s)
K_D(nM)

F3'-1602 TriNKET
5.3 × 10⁴
2.7 × 10⁻⁴
5.2

5.2 × 10⁴
2.8 × 10⁻⁴
5.3

5.1 × 10⁴
2.6 × 10⁻⁴
5.2

5.2 × 10⁴
2.7 × 10⁻⁴
5.3

Average ± SD
(5.2 ± 0.1) × 10⁴
(2.7 ± 0.1) × 10⁻⁴
5.2 ± 0.1

trastuzumab
9.6 × 10⁴
2.3 × 10⁻⁴
2.4

9.6 × 10⁴
2.3 × 10⁻⁴
2.5

9.6 × 10⁴
2.3 × 10⁻⁴
2.5

9.2 × 10⁴
2.3 × 10⁻⁴
2.5

Average ± SD
(9.3 ± 0.1) × 10⁴
(2.3 ± 0.0) × 10⁻⁴
2.5 ± 0.1

Binding of F3′-1602 TriNKET to CD32a H131 was determined as described above and is shown in FIGS. 46A-46P. The data were fit with a steady state affinity fit model; affinity values are summarized in Table 40.

TABLE 40

Affinity values of F3'-1602 TriNKET and trastuzumab for

human CD32a H131.

K_D(μM)

Replicate
Replicate
Replicate
Replicate
Average ±

Sample
K_D(μM)
K_D(μM)
K_D(μM)
K_D(μM)
SD

F3'-1602
0.9
0.8
0.9
0.9
0.9 ± 0.0

TriNKET

trastuzumab
1.2
1.1
1.2
1.1
1.1 ± 0.1

Binding of F3′-1602 TriNKET to human CD32a R131 allele (FcγRIIa R131) was determined as described above and is shown in FIGS. 47A-47P. The data were fit with a steady state affinity fit model; affinity values are summarized in Table 41.

TABLE 41

Affinity values of F3'-1602 TriNKET and trastuzumab for

human CD32a R131.

K_D(nM)

Replicate
Replicate
Replicate
Replicate
Average ±

Sample
K_D(nM)
K_D(nM)
K_D(nM)
K_D(nM)
SD

F3'-1602
1.4
1.3
1.7
1.2
1.4 ± 0.2

TriNKET

trastuzumab
1.7
1.6
1.9
1.6
1.7 ± 0.2

Binding of F3′-1602 TriNKET to human CD16a high affinity allele V158 (FcγRIIIa V158) was determined as described above and is shown in FIGS. 48A-48H. The data were fit with a 1:1 kinetic fit model; kinetic parameters and affinity values are summarized in Table 42. The affinity of F3′-1602 TriNKET for both alleles is comparable (three-fold or less difference) to trastuzumab's affinity for the same receptors.

TABLE 42

Kinetic parameters and affinity values of human CD16a V158

for F3'-1602 TriNKET and trastuzumab.

Kinetic

Sample
k_a(1/Ms)
k_d(1/s)
Fit K_D(nM)

F3'-1602 TriNKET
9.6 × 10⁴
2.2 × 10⁻²
234.0

9.5 × 10⁴
2.2 × 10⁻²
235.0

9.3 × 10⁴
2.3 × 10⁻²
243.0

9.3 × 10⁴
2.2 × 10⁻²
241.0

Average ± SD
(9.4 ± 0.2) × 10⁴
(2.2 ± 0.0) × 10⁻²
238.3 ± 4.4

trastuzumab
1.7 × 10⁵
1.3 × 10⁻²
76.5

1.7 × 10⁵
1.3 × 10⁻²
76.0

1.3 × 10⁵
1.0 × 10⁻²
80.3

1.3 × 10⁵
1.0 × 10⁻²
77.7

Average ± SD
(1.5 ± 0.2) × 10⁵
(1.1 ± 0.1) × 10⁻²
77.6 ± 1.9

Binding of F3′-1602 TriNKET to human CD16a low affinity allele F158 (FcγRIIIa F158) was determined as described above and is shown in FIGS. 49A-49P. The data were fit with a steady state affinity fit model; affinity values are summarized in Table 43.

TABLE 43

Affinity of F3'-1602 TriNKET and trastuzumab for human

CD16a F158.

K_D(nM)

Replicate
Replicate
Replicate
Replicate
Average ±

Sample
K_D(nM)
K_D(nM)
K_D(nM)
K_D(nM)
SD

F3'-1602
1130.0
1120.0
1090.0
1090.0
1107.5 ±

TriNKET

20.6

trastuzumab
436.0
430.0
445.0
448.0
439.8 ±

8.3

Binding of F3′-1602 TriNKET to human CD32b (FcγRIIb) was determined as described above and is shown in FIGS. 50A-50P. The data were fit with a steady state affinity fit model; affinity values are summarized in Table 44.

TABLE 44

Affinity of human CD32b for F3'-1602 TriNKET and trastuzumab.

K_D(μM)

Replicate
Replicate
Replicate
Replicate
Average ±

Sample
K_D(μM)
K_D(μM)
K_D(μM)
K_D(μM)
SD

F3'-1602
7.2
6.1
6.9
6.1
6.6 ± 0.6

TriNKET

trastuzumab
7.0
6.7
7.2
6.6
6.9 ± 0.3

Binding of F3′-1602 TriNKET to human CD16b (FcγRIIIb) was determined as described above and is shown in FIGS. 51A-51P. The data were fit with a steady state affinity fit model; affinity values are summarized in Table 45.

TABLE 45

Affinity of human CD16b for F3’-1602 TriNKET and

trastuzumab.

K_D( μM)

Replicate
Replicate
Replicate
Replicate
Average ±

Sample
K_D(μM)
K_D(μM)
K_D(μM)
K_d(μM)
SD

F3’-1602
7.6
7.2
8.6
6.5
7.5 ± 0.8

TriNKET

trastuzumab
4.2
4.0
4.8
3.8
4.2 ± 0.4

Binding of F3′-1602 TriNKET to cyno CD16 was measured as described above and is shown in FIGS. 52A-52P. The data were fit with a steady state affinity fit model; affinity values are summarized in Table 46.

TABLE 46

Affinity values of F3’-1602 TriNKET and trastuzumab for

cynomolgus CD16.

K_D(nM)

Replicate
Replicate
Replicate
Replicate
Average ±

Sample
K_D(nM)
K_D(nM)
K_D(nM)
K_D(nM)
SD

F3’-1602
315.0
315.0
294.0
302.0
306.5 ± 10.3

TriNKET

trastuzumab
148.0
141.0
164.0
149.0
150.5 ± 9.7

Binding of F3′-1602 TriNKET to human FcRn was measured at pH 6.0 as described above and is shown in FIGS. 53A-53P. The data were fit with a steady state affinity fit model; affinity values are summarized in Table 47.

TABLE 47

Affinity values of F3’-1602 TriNKET and trastuzumab for human

FcRn at pH 6.0.

K_D(μM)

Replicate
Replicate
Replicate
Replicate
Average ±

Sample
K_D(μM)
K_D(μM)
K_D(μM)
K_D(μM)
SD

F3’-1602
1.3
1.1
1.5
1.6
1.4 ± 0.2

TriNKET

trastuzumab
1.8
1.3
1.3
1.2
1.4 ± 0.2

Binding of F3′-1602 TriNKET to cyno FcRn was measured at pH 6.0 as described above and is shown in FIGS. 54A-54P. The data were fit with a steady state affinity fit model; affinity values are summarized in Table 48.

TABLE 48

Affinity values of F3’-1602 TriNKET and trastuzumab for

cyno FcRn at pH 6.0.

K_D(μM)

Replicate
Replicate
Replicate
Replicate
Average ±

Sample
K_D(μM)
K_D(μM)
K_D(μM)
K_D(μM)
SD

F3’-1602
1.1
1.1
1.4
1.5
1.3 ± 0.2

TriNKET

trastuzumab
1.6
1.6
1.2
1.3
1.4 ± 0.2

Binding of F3′-1602 TriNKET to human and cyno FcRn was measured at pH 7.4 as described above. No quantifiable binding was observed.

F3′-1602 TriNKET and IgG1 isotype control trastuzumab do not demonstrate physiologically meaningful differences (less than 3-fold difference) and are comparable in their binding to human and cynomolgus CD64 (FcγRI), CD16 (FcγRIII) recombinant receptors tested (Table 49). F3′-1602 TriNKET and trastuzumab are similar (less than 1.2-fold difference) in their binding to human CD32 (FcγRII) receptors (Table 49). F3′-1602 TriNKET is similar to trastuzumab in its affinity for human and cynomolgus FcRn at pH 6.0. F3′-1602 TriNKET and trastuzumab did not demonstrate any detectable binding at the concentrations tested at pH 7.4 (Table 50).

TABLE 49

Summary of FcγRs affinities for F3′-1602 TriNKET and trastuzumab.

Human
Cyno
Human CD32a
Human CD16a
Human
Human
Cyno

CD64
CD64
K_D, μM
K_D, nM
CD32b
CD16b
CD16

Sample
K_D, nM
K_D, nM
H131
R131
V158
F158
K_D, μM
K_D, μM
K_D, nM

F3′-1602
5.2 ± 0.1
2.0 ± 0.0
0.9 ± 0.0
1.4 ± 0.2
238.2 ± 4.4
1107.5 ± 20.6
6.6 ± 0.6
7.5 ± 0.8
306.5 ± 10.3

TriNKET

trastuzumab
2.5 ± 0.1
0.9 ± 0.0
1.1 ± 0.1
1.7 ± 0.2
77.6 ± 1.9
439.8 ± 8.3
6.9 ± 0.3
4.2 ± 0.4
150.5 ± 9.7

TABLE 50

Summary of F3’-1602 TriNKET and trastuzumab binding human

and cynomolgus FcRn.

Human
Human
Cyno
Cyno

FcRn,
FcRn,
FcRn,
FcRn,

pH 6.0
pH 7.4
pH 6.0
pH 7.4

Sample
K_D, μM
K_D, μM
K_D, μM
K_D, μM

F3’-1602
1.4 ± 0.2
No
1.3 ± 0.2
No

TriNKET

quantifiable

quantifiable

binding

binding

trastuzumab
1.4 ± 0.2
No
1.4 ± 0.2
No

quantifiable

quantifiable

binding

binding

FcRn Binding

Recombinant human or cynomolgus monkey (cyno) FcRn was directly immobilized via amine coupling at a density of 169-216 RU. F3′-1602 TriNKET and commercial trastuzumab as an IgG1 assay control were buffer exchanged into MBS-EP+, pH 6.0 as described above before the SPR binding experiment. Each FcRn binding experiment consisted of multiple cycles utilizing varying concentrations of analytes. Two-fold dilution series (12 μM-0.023 μM) of F3′-1602 TriNKET and trastuzumab were used. As no-signal, negative controls, a 0 nM blank cycle was placed at the start of each concentration series, and a blank amine coupling modified chip surface lacking FcRn were used. Each experimental cycle included association, dissociation, and regeneration steps. The association step consisted of 60 sec analyte injection and was followed by a 60 sec dissociation step. Surface regeneration (60 sec injection of HBS-EP+ buffer, pH 7.4) completed each kinetic cycle. All steps were executed at a flow rate of 30 μL/min and at 25° C. MBS-EP+ buffer, pH 6.0 was used as a running buffer. The data were fitted to a steady state affinity model with Biacore Insight evaluation software. Chi²(calculated by the software) related to the calculated R_maxwas used to assess goodness of the fit.

Example 12
Molecular Format and Design, Structure, Affinity, Potency, Specificity and Cross-Reactivity Analysis of F3′-1602 TriNKET
Surface Plasmon Resonance (SPR)

Binding affinities of F3′-1602 TriNKET for recombinant human CLEC12A (hCLEC12A) or cyno CLEC12A (cCLEC12A) were measured by SPR using a Biacore 8K instrument at physiological temperature of 37° C. Briefly, human Fc specific antibodies were covalently immobilized at a density of about 8000-10000 resonance units (RU) on carboxy methyl dextran matrix of a CM5 biosensor chip to create an anti-hFc IgG chip. F3′-1602 TriNKET samples were injected on the anti-hFc IgG chip at a flow rate of 10 μL/min for 60 seconds to achieve an about 250 RU capture level. hCLEC12A-His or cCLEC12A-His was serially diluted (100 nM-0.14 nM) in three-fold dilutions with running buffer and injected at a flow rate of 30 μl/min over the captured test articles. Association was monitored for 300 seconds, and dissociation was monitored for 900 seconds. Surfaces were regenerated between cycles with three pulses of 10 mM glycine-HCl (pH 1.7) injected for 20 seconds at 100 μL/min.

The kinetic constants and equilibrium binding affinities are provided in Table 51, and raw data and fits are shown in FIGS. 55A-55D. The molecule binds strongly to CLEC12A and weakly to NKG2D and CD16a. The complex between F3′-1602 TriNKET and human CLEC12A is strong as evidenced from the slow rate dissociation constant 4.94±0.09×10⁻⁴s⁻¹. The equilibrium binding affinity K_Dfor F3′-1602 TriNKET was 0.59±0.01 nM.

TABLE 51

Comparison of kinetic parameters of F3’-1602 TriNKET to human

CLEC12A at pH 7.4 and pH 6.0.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

F3’-1602
hCLEC12A-
8.54 × 10⁵
4.94 × 10⁻⁴
0.58

TriNKET #1
His

F3’-1602
hCLEC12A-
8.33 × 10⁵
4.94 × 10⁻⁴
0.59

TriNKET #2
His

F3’-1602
hCLEC12A-
8.63 × 10⁵
4.89 × 10⁻⁴
0.57

TriNKET #3
His

F3’-1602
hCLEC12A-
8.26 × 10⁵
4.99 × 10⁻⁴
0.60

TriNKET #4
His

Average ±
hCLEC12A-
(8.44 ± 0.03) ×
(4.94 ± 0.09) ×
0.59 ± 0.01

Std Dev
His
10⁵
10⁻⁴

Further, polymorphic variant CLEC12A-K244Q is prevalent in about 30% of the population. Binding of F3′-1602 TriNKET to CLEC12A-K244Q was examined by SPR and compared its affinity to wild-type CLEC12A. As shown in Table 52, binding kinetics of F3′-1602 TriNKET for wild-type and the K244Q variant of CLEC12A are similar.

TABLE 52

Kinetic parameters and affinities of F3’-1602 TriNKET for human

CLEC12A-K244Q.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

F3’-l602
hCLEC12A WT
8.26 × 10⁵
4.96 × 10⁻⁴
0.60

TriNKET

F3’-1602
hCLEC12A-K244Q
5.65 × 10⁵
4.36 × 10⁻⁴
0.77

TriNKET

Binding affinities of F3′-1602 TriNKET for recombinant human NKG2D were measured by SPR using a Biacore 8K instrument at physiological temperature of 37° C. Briefly, mouse (mFc) specific antibodies were covalently immobilized at a density of about 8000-10000 RU on carboxy methyl dextran matrix of a CM5 biosensor chip to create an anti-mFc IgG chip. mFc-tagged human NKG2D was captured on the anti-mFc IgG chip at a flow rate of 10 μL/min for 60 seconds to achieve an about 35 RU capture level. F3′-1602 TriNKET was buffer exchanged into running buffer and serially diluted (5000 nM-9.78 nM) in two-fold dilutions with 1× HBS-EP+. F3′-1602 TriNKET was injected at a flow rate of 30 μl/min over the captured test articles. Association was monitored for 60 seconds, and dissociation was monitored for 60 seconds. Surfaces were regenerated between cycles with three pulses of 10 mM glycine-HCl (pH 1.7) injected for 20 seconds at 100 μL/min.

FIGS. 56A-56H show F3′-1602 TriNKET binding sensorgrams to human NKG2D. Two different fits were utilized to obtain the equilibrium affinity data: steady state affinity fit and kinetic fit. The kinetic constants and equilibrium affinity constants are shown in Table 54. The dissociation rate constant was (1.30±0.03)×10⁻¹s⁻¹. Equilibrium affinity constants (K_D) obtained by kinetics fit and steady state affinity fit were, 721±21 nM and 727±22 nM respectively.

TABLE 54

Kinetic parameters and affinities of F3’-1602 TriNKET for human

NKG2D measured by SPR.

Steady State

Test

Kinetics fit
Affinity fit

articles
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)
K_D(nM)

F3’-1602
hNKG2D
1.80 × 10⁵
1.26 × 10⁻¹
703
709

TriNKET

#1

F3’-1602
hNKG2D
1.84 × 10⁵
1.30 × 10⁻¹
705
709

TriNKET

#2

F3’-1602
hNKG2D
1.75 × 10⁵
1.31 × 10⁻¹
745
751

TriNKET

#3

F3’-1602
hNKG2D
1.80 × 10⁵
1.32. × 10⁻¹
732
740

TriNKET

#4

Average ±
hNKG2D
(1.80 ± 0.04) × 10⁵
(1.30 ± 0.03) × 10⁻¹
721 ± 21
727 ± 22

St Dev

Cross-reactivity of F3′-1602 TriNKET with cyno NKG2D was further investigated by SPR as described in Example 5 or 6. F3′-1602 TriNKET was tested for binding to recombinant cyno NKG2D (cNKG2D) extracellular domain (ECD) by SPR (Biacore) at 37° C. (FIGS. 57A-57H). NKG2D is a dimer in nature, therefore recombinant mFc-tagged cyno NKG2D dimer was used for this experiment. The binding affinities are shown in Table 55. Binding affinities were obtained from kinetic fit (K_D839±67 nM) and steady fit (K_D828±88 nM), and were similar to the binding affinity observed with hCLEC12A (K_D721±21 nM obtained with kinetic fit).

TABLE 55

Kinetic parameters and affinities of F3’-1602 TriNKET for cyno

NKG2D measured by SPR.

Steady State

Test

Kinetics fit
Affinity fit

articles
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)
K_D(nM)

F3’-1602
cNKG2D
1.77 × 10⁵
1.47 × 10⁻¹
831
829

TriNKET #1

F3’-1602
cNKG2D
1.64 × 10⁵
1.36 × 10⁻¹
825
828

TriNKET #2

F3’-1602
cNKG2D
1.56 × 10⁵
1.45 × 10⁻¹
931
935

TriNKET #3

F3’-1602
cNKG2D
1.76 × 10⁵
1.35 × 10⁻¹
769
719

TriNKET #4

Average ± St Dev
cNKG2D
(1.68 ± 0.1) × 10⁵
(1.41 ± 0.06) × 10⁻¹
839 ± 67
828 ± 88

Simultaneous Engagement of CLEC12A and NKG2D

F3′-1602 TriNKET was diluted in 1× HBS-EP+ buffer containing 0.1 mg/mL BSA and was captured on an anti-human Fc surface of CM5 chip at a flow rate 5 μL/min for 60 sec to achieve capture level of 150-250 RU. The net difference between baseline signal and the signal after completion of F3′-1602 TriNKET injection representing the amount of F3′-1602 captured was recorded. hCLEC12A-His (100 nM) or mFc-hNKG2D (7 μM) was injected over captured F3′-1602 TriNKET at 30 μL/min for 90 sec to reach saturation. This injection was immediately followed by an injection of a pre-incubated mixture of hCLEC12A-His (100 nM) and mFc-hNKG2D (7 μM) at a flow rate of 30 μL/min for 90 sec with the use of the A-B-A injection command in the Biacore 8K control software (the second target was pre-mixed with the first target to assure all binding sites for the first target are occupied). The chip was regenerated by two 20 second pulses of 10 mM Glycine (pH 1.7) at 100 μL/min. The experiment was conducted at 37° C. to mimic physiological temperature. An average relative binding stoichiometry of each target bound to captured F3′-1602 TriNKET that is already saturated with the other target (injected second) was expressed as a fraction of the full capacity binding to unoccupied F3′-1602 TriNKET.

Target binding sensorgrams as shown in FIGS. 58A-58B demonstrate that the occupancy status of the CLEC12A domain does not interfere with NKG2D binding and occupancy status of the NKG2D domain does not interfere with CLEC12A binding. Similarity in shapes of the respective sensorgram segments depicting binding of each target to free F3′-1602 TriNKET and F3′-1602 TriNKET that has been saturated with the other target suggests that the kinetic parameters are not meaningfully affected by the target occupancy status of the F3′-1602 TriNKET molecule. For instance, the shape of the CLEC12A binding segment of the sensorgrams is similar in both panels. Saturating concentration of the NKG2D was maintained throughout the entire experiment, represented in the lower panel due to the fast dissociation rate of the target. Additionally, the lack of any impact on relative stoichiometry of each target binding (when compared to binding to unoccupied F3′-1602) signifies full independence of NKG2D and CLEC12A binding sites on F3′-1602 TriNKET (Table 56).

TABLE 56

Relative binding stoichiometries of F3’-1602 TriNKET for CLEC12A

and NKG2D, expressed relative to binding of each target to unoccupied

captured F3’-1602 TriNKET.

CLEC12A,
NKG2D,

relative
relative

binding
binding

stoichiometry
stoichiometry

Target Bound to F3’-1602 TriNKET
1.00
1.00

unoccupied with another target

(injected first)

Target Bound to F3’-1602 TriNKET
1.21 ± 0.06
1.19 ± 0.04

saturated with another target

(injected second)

NKG2D-CD16a Synergy

Synergistic NKG2D and CD16a binding was evaluated by SPR. 233 RU of hNKG2D alone, 618 RU of CD16a F158 allele alone, and 797 RU of the mixture of hNKG2D and CD16a(F158) were amino-coupled to the surface of the CM5 Series S Biacore chip. 2.8 μM F3′-1602 TriNKET or trastuzumab were injected for 150 sec at 10 μL/min. Dissociation phase was observed for 180 seconds when regeneration was not needed and 1500 second when natural regeneration of the surface (almost complete dissociation of analyte) was needed between the cycles at the same flow rate.

As shown in FIG. 59, while F3′-1602 TriNKET binds with low affinity to CD16a or NKG2D alone, F3′-1602 TriNKET injected over the mixed surface manifests in avidity of simultaneous CD16a and NKG2D binding through much slower off-rate (FIG. 59A). Experimental control trastuzumab does not bind immobilized NKG2D and exhibits the same apparent off-rate when injected over both mixed (CD16a+NKG2D) and CD16a-only surfaces (FIG. 59B). The data indicate that F3′-1602 TriNKET avidly engages both CD16a and NKG2D.

Binding Unrelated Recombinant Proteins and Cell Binding Specificity

Specificity for CLEC12A was tested by SPR against five different unrelated proteins at concentrations as high as 500 nM and is shown in FIG. 60. No non-specific binding was observed to any of the unrelated recombinant targets (FIG. 60B, FIG. 60C), whereas positive control (CLEC12A) showed binding of 50 RU at 100 nM concentration (FIG. 60A). Specificity was also tested against proteins expressed on the surface of the Ba/F3 cell line. No non-nonspecific binding of F3′-1602 at a concentration of 333 nM to the Ba/F3 parental cell line was observed (FIG. 60E), whereas Ba/F3 cell lines engineered to express CLEC12A, used as a positive control, showed a significant shift by flow cytometry (FIG. 60D).

Non-Specific Binding to Polyspecific Reagent (PSR)

A flow cytometry based PSR assay allows the filtering out of antibodies that have a higher probability to bind non-specifically to unrelated proteins. As part of the developability assessment, F3′-1602 was tested for non-specific binding to a preparation of detergent solubilized membrane proteins in a PSR assay (FIGS. 61A-61C). PSR assay correlates well with cross-interaction chromatography, a surrogate for antibody solubility, as well as with baculovirus particle enzyme-linked immunosorbent assay, a surrogate for in vivo clearance (Xu et. al (2013). Addressing polyspecificity of antibodies selected from an in vitro yeast presentation system: a FACS-based, high-throughput selection and analytical tool. Protein engineering design and selection, 26, 663-670).

50 μL of 100 nM TriNKET or control mAb in PBSF were incubated with pre-washed 5 μL protein A dyna beads slurry (Invitrogen, catalog #10001D) for 30 minutes at room temperature. TriNKET or mAb bound magnetic beads were allowed to stand on a magnetic rack for 60 seconds and the supernatant was discarded. The bound beads were washed with 100 PBSF. Beads were incubated for 20 minutes on ice with 50 μL of biotinylated PSR reagent which was diluted 25-fold from the stock (Xu et. al., (2013) Protein engineering design and selection, 26, 663-670). Samples were put on the magnetic rack, supernatant discarded, and washed with 100 μL of PBSF. A secondary FACS reagent, to detect binding of biotinylated PSR reagent to TriNKETs or control mAbs, was made as follows: 1:250 μL of Streptavidin-PE (Biologend, catalog #405204) and 1:100 donkey anti-human Fc were combined in PBSF. To each sample, 100 μL of the secondary reagents were added and allowed to incubate for 20 minutes on ice. The beads were washed twice with 100 μL PBSF, and samples were analyzed on a FACS Celesta (BD). Two PSR controls, Rituximab (PSR positive) and Trastuzumab (PSR clean), were used in the assay.

High-Spec® Cross Reactivity Assay on HuProt™ Human Proteome Assays

To examine specificity of F3′-1602 TriNKET, a protein array technology was used. The HuProt™ human proteome microarray provides the largest database of individually purified human full length proteins on a single microscopic slide. An array consisting of 22,000 full-length human proteins are expressed in yeast S. cerevisiae, purified, and subsequently printed in duplicate on a mircoarray glass slide that allows thousands of interactions to be profiled in a high-throughput manner.

Specificity of F3′-1602 TriNKET was tested at 0.1 μg/mL and 1 μg/mL concentration against native Huprot human proteome array embedded on microslides at CDI laboratories (Baltimore, Md.), according to their standard procedure.

FIG. 62 shows the relative binding (Z score) of F3′-1602 TriNKET at 1 μg/mL to human CLEC12A in comparison to the entire human proteome microarray. Z score is the average binding score of two duplicates of a given protein. For comparison purpose, the top 24 proteins with residual background binding to F3′-1602 TriNKET were also provided in FIG. 62. Table 57 shows the Z and S scores of F3′-1602 TriNKET to human CLEC12A and the top 6 proteins from the microarray. S score is the difference of Z score of a given protein and the one rank next to it. If the S score of the top hit is >3 an antibody is considered to be highly specific to its target. Based on the Z and S score criteria, F3′-1602 TriNKET showed high specificity to human and lack of off-target binding in the HuProt™ human proteome assay.

TABLE 57

Z and S scores of F3’-1602 TriNKET in HuProt ™ human proteome

microarray assay.

Name
Rank
Protein-ID
Z score
S score

hCLEC12A
1

150.21
144.62

MFF
2
JHU15965.P1168B06
5.60
0.59

RPLP0
3
JHU16464.P173H05
5.01
0.12

PARS2
4
JHU06781.P071H01
4.89
0.062

HMGA1_frag
5
JHU16418.P173B08
4.83
0.27

THEMIS2_frag
6
JHU15201.P160B01
4.56
0.37

HAGHL
7
JHU08774.P092D12
4.19
0.01

Molecular Modeling

Anti-CLEC12A and anti-NKG2D binding arms of F3′-1602 TriNKET were compared with 377 post Phase I biotherapeutic molecules using Therapeutic Antibody Profiler (TAP) available at the SAbPred website. TAP used ABodyBuilder to generate a model for F3′-1602 TriNKET with side chains by PEARS. The CDRH3 was built by MODELLER due to its diversity.

Five different parameters were evaluated:

Total CDR length

Patches of surface hydrophobicity (PSH) across the CDR vicinity

Patches of positive charge (PPC) across the CDR vicinity

Patches of negative charge (PNC) across the CDR vicinity

Structural Fv charge symmetry parameter (sFvCSP)

These parameters of F3′-1602 TriNKET were then compared with the profile distributions of therapeutic antibodies to predict the developability and any potential issues that might cause downstream challenges.

FIGS. 63A-63C illustrate a ribbon diagram model of the CLEC12A binding scFv in three different orientations (upper panel) and their corresponding surface charge distribution of the same orientation (lower panel). The charge distribution of anti-CLEC12A scFv is polarized (“top view”, lower panel), with negatively-charged residues populated predominately within CDRH3 and CDRL2. The uneven distribution of electrostatic patches on the paratope is likely to be target-related and reflects the complementarity of charge distribution on its cognate epitope, which may likely contribute to the high affinity interaction between CLEC12A and the scFv of F3′-1602 TriNKET.

FIGS. 64A-64E show the CDR length and surface hydrophobicity analyses of scFv CLEC12A targeting arm of F3′-1602 TriNKET. The length of CDRs for the CLEC12A binding arm of F3′-1602 TriNKET are typical for late stage therapeutic antibodies.

The hydrophobicity of a monoclonal antibody is an important biophysical property relevant for its developability into a therapeutic. Hydrophobic patch analysis of the CDRs of an antibody is predictive of its behavior. The CLEC12A arm of F3′-1602 TriNKET has much lower hydrophobicity comparing to other reference molecules. Based on the modeling, there is no hydrophobic patch of significant size on the surface of CLEC12A-binding arm of F3′-1602 TriNKET.

The charge distribution of anti-CLEC12A scFv is polarized, with negatively-charged residues populated predominately within CDRH3 and CDRL2 (as shown in FIGS. 63A-63C). Although the positively charged patches and charge symmetry are within the norm, without wishing to be bound by theory, it is hypothesized that the distinct negatively charged patches on the paratope is target-related and reflects the complementarity of charge distribution on its cognate epitope, which may contribute to its high affinity interaction with CLEC12A.

The NKG2D-binding Fab arm of F3′-1602 TriNKET is shown in ribbon diagrams with three different orientations (FIGS. 65A-65C) and their corresponding surface charge distribution of the same orientation are provided in FIGS. 65A-65C (lower panel). In contrast to CLEC12A binding arm, the surface charge distribution on NKG2D A49M-I arm is more evenly distributed.

FIGS. 66A-66E show the CDR length and patches of surface hydrophobicity analyses of NKG2D-targeting arm of F3′-1602 TriNKET. Analysis was performed using TAP and was benchmarked 377 late-stage therapeutic antibodies. It appears that the CDR length, the hydrophobic properties, the positive/negative charge distributions, and the Fc charge symmetry of the NKG2D-targeting arm of F3′-1602 TriNKET are well within the norm of existing therapeutic antibodies. The modelling suggests that F3′-1602 TriNKET will not have developability complications during the manufacturing processes.

Hydrophobic Interactions Chromatography

The hydrophobicity prediction data was confirmed by investigating F3′-1602 TriNKET behavior with analytical Hydrophobic Interactions Chromatography (HIC), a technique that relies upon proteins with significant patches of exposed hydrophobic patches being more prone to aggregation. To perform HIC, briefly, injections of TriNKETs (5 μg of protein) were prepared in a 5:4 ratio of high salt buffer (100 mM sodium phosphate, 1.8 M ammonium sulfate, pH 6.5) to sample. Samples were analyzed using an Agilent 1260 Infinity II HPLC equipped with a Sepax Proteomix HIC Butyl-NP5 5 uM column held at 25° C. The gradient was run from 0% low salt buffer (100 mM sodium phosphate, pH 6.5) to 100% low salt buffer over 6.5 minutes at a flow rate of 1.0 mL/minute. Chromatograms were monitored at 280 nm. Retention times of F3′-1602 TriNKET on the analytical HIC column is shown in Table 58 and HIC profile in FIG. 67. Commercial trastuzumab (Roche) was used as an example of a well-behaved biologic and functioned as an internal control for the assay. F3′-1602 TriNKET has a retention time of 9.8 minutes, compared to 9.6 minutes for trastuzumab. Thus, experimental hydrophobicity analysis suggests that the hydrophobic properties of F3′-1602 TriNKET are acceptable for further development.

TABLE 58

HIC retention time of F3’-1602 TriNKET in comparison

with trastuzumab.

Test article
HIC retention time (min)

F3-1602 TriNKET
9.8

Trastuzumab
9.6

Capillary Isoelectric Focusing (cIEF)

Experimental pI of F3′-1602 was obtained by cIEF, as described in Example 15 (FIG. 68). Commercial grade Trastuzumab was included in the assay as an internal control. The cIEF profile of F3′-1602 TriNKET was typical for a monoclonal antibody, with a main peak at pI 8.80 (Table 59). The presence of minor amounts of acidic and basic species were also observed.

TABLE 59

pI of scFv-1602 F3’ by cIEF compared to trastuzumab.

Test article
pI (major peak)

scFv-1602 F3’
8.80

Trastuzumab
8.91

Immunogenicity Assessment

Immunogenicity assessment was performed using the EpiMatrix algorithm from EpiVax was used as described in Cohen et al. (2010) A method for individualizing the prediction of immunogenicity of protein vaccines and biologic therapeutics: individualized T cell epitope measure (iTEM). J. Biomed. Biotechnol. 961752. The T_regadjusted Epimatrix Protein Score, which ranges from −80 (no immunogenicity) to 80 (highly immunogenic), for the sequences of the three chains of scFv-1602 F3′ are VH-CH1-Fc: −33.39, VH-VL-Fc: −26.4 and VL-CL: −27.4. Thus, the predicted risk of immunogenicity for the scFv-1602 F3′ appears to be low.

Example 13
F3′-1602 TriNKET Analysis of Putative Sequence Liabilities

The amino acid sequences of the three polypeptide chains of F3′-1602 TriNKET was analyzed for putative sequence liabilities as described in Example 3. Putative sequence liabilities are shown in Table 60.

TABLE 60

Putative sequence liabilities in F3’-1602 TriNKET. CDRs are under

Chothia numbering.

VL-CL
VH-CH1-Fc
VH-VL-Fc

None
DP in CDRH3
DS in CDRH3

(potential chemical instability)
(potential aspartate isomerization)

To examine the liabilities, accelerated stability (4 weeks at 40° C.) and forced degradation studies were performed. The DP in the CDRH3 of VH-CH1-Fc was not modified in accelerated stability studies or forced degradation studies, and therefore, no further analysis was necessary. However, it was observed that modification of the DS in the CDRH3 in the scFv led to a reduction in CLEC12A binding and decreased potency of F3′-1602 TriNKET in accelerated stability studies, as described in Example 14.

To replace the DS in CDRH3, yeast display was performed to identify alternative sequence motifs without the DS site. Three variants, namely YDYDDALDY (SEQ ID NO:283) YDYDDILDY (SEQ ID NO:284) and YDYDDLLDY (SEQ ID NO:285), were identified that showed binding to hCLEC12A albeit with weaker binding signal compared to the parent scFv (YDYDDSLDY) (SEQ ID NO:139), while variants YDYDDVLDY (SEQ ID NO:290), YDYDDTLDY (SEQ ID NO:291), and YDYDESLDY (SEQ ID NO:292) were identified as non-binders. Based on the binding analysis only variants YDYDDALDY ((SEQ ID NO:283), AB0053) and YDYDDILDY ((SEQ ID NO:284), AB0085) were considered for mammalian production and further characterization. Binding of AB0053 and AB0085 to hCLEC12A-His was characterized using Surface Plasmon Resonance (SPR) at 37° C. Both mutations introduced to remove DS sequence liability had an effect on the binding to hCLEC12A-His (Table 61). FIGS. 69A-69I demonstrate FACS analysis of binding to hCLEC12A-his using yeast libraries generated to remove the sequence liability in CDRH3. The data showed complete loss of binding in the DS to DI engineered variant (AB0085). Introduction of DA in place of DS (AB0053) did not eliminate binding completely but lead to apparent heterogeneity in binding sensorgram. Following SPR analysis, potency of the liability-remediated variants (AB0053 and AB0085) were tested in KHYG-1-CD16aV mediated cytotoxicity assay (FIG. 70, Table 62). AB0085 lost ability to kill AML cells as determined by significantly reduced percentage of maximum killing. EC50 for AB0053 construct increased ˜3-fold compared to F3′-1602 TriNKET (AB0237) control, whereas percentage of maximum killing was not affected. This finding is in line with the 3-fold change in overall affinity for hCLEC12A established in the SPR experiment (Table 61).

Binding affinities of F3′-1602 TriNKET to recombinant human CLEC12A were measured by SPR at 37° C. using a Biacore 8K instrument. Briefly, human Fc specific antibodies were covalently immobilized at a density ˜8000-10000 resonance units (RU) on carboxy methyl dextran matrix of a CM5 biosensor chip via amine-coupling chemistry to create an anti-hFc IgG chip. F3′-1602 TriNKET samples were captured on the anti-hFc IgG chip at a concentration of 1.5 μg/mL at a flow rate of 10 μL/min for 60 seconds to achieve ˜150-250 RU capture level. hCLEC12A-His was serially diluted (100 nM-0.046 nM) in three-fold dilutions with HBS-EP+ buffer (1×; 10 mM HEPES, 150 mM NaCl, 3 mM EDTA, 0.05% P20 pH 7.4) with 0.1 mg/mL bovine serum albumin (BSA) and injected at a flow rate of 30 μl/min over the captured test articles. Association was monitored for 300 seconds and dissociation was monitored for 600 seconds. Surfaces were regenerated between cycles with three pulses of 10 mM glycine-HCl, pH 1.7 injected for 20 seconds at 100 μL/min. HBS-EP+ (1×) with 0.1 mg/mL BSA buffer was used throughout the experiment. Data were analyzed using Biacore 8K Insight Evaluation software (GE Healthcare).

TABLE 61

Kinetic parameters and affinities of DS engineered clones for

hCLEC12A for CLEC12A.

Test
k_a

Article
(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)
Notes

AB0053
1.63 × 10⁶*
k_d1= 1.57 × 10⁻² and
2.41
Two-state

k_d2= 9.49 × 10⁻⁴

fit

AB0085
No Binding

F3’-1602
1.11 × 10⁶
8.95 × 10⁻⁴
0.80
Classical

(AB0237)

1:1 fit

*k_a2value was insignificant compared to the k_a1value. Therefore, only k_a1value is shown.

TABLE 62

Comparison of potency of DS engineered TriNKETs

in KHYG-1-CD16aV cytotoxicity assay.

PL-21 cells

Test Article
EC50 (nM)
Max lysis (%)

F3’-1602 TriNKET (AB0237)
0.12
98.96

AB0053
0.39
105.10

AB0085
N/D*
N/D

*N/D not determined

As the library approach produced AB0053 as the only hit, and that hit demonstrated kinetics of binding to hCLEC12A significantly different from F3′-1602 TriNKET and inferior potency in the cytotoxicity assay, it was concluded that the DS motif is necessary for maintaining architecture of CDRH3 and therefore cannot be effectively removed. The DS liability was instead controlled with formulation, as described in Example 14.

Example 14
Formulations of F3′-1602

In Example 13, sequence liability sites were found to be critical for the ability of F3′-1602 to bind CLEC12A and therefore cannot be removed. This example was designed to assess the formulations that may maintain the stability of F3′-1602 despite the presence of the sequence liabilities.

Briefly, three formulations were tested for F3′-1602:

(1) 20 mM histidine, 250 mM sucrose, and 0.01% polysorbate 80, at pH of 6.0, referred to herein as the “HST formulation.” F3′-1602 was diafiltered into the HST buffer and the recovered sample was passed through a 0.2 μm filter. The final concentration of F3′-1602 was 19.7 mg/mL, as measured by absorbance at 280 nm (A280).
(2) 20 mM potassium phosphate, 10 mM sodium citrate, and 125 mM sodium chloride, at pH 7.8, referred to herein as the “phosphate/citrate/NaCl formulation.” F3′-1602 was buffer switched into the phosphate/citrate/NaCl formulation buffer using an Amicon Ultra 0.5 mL 30K device and Zeba Desalting Columns, and the recovered sample was passed through a 0.2 μm filter. The final concentration of F3′-1602 was 20.6 mg/mL, as measured by A280.
(3) 20 mM citrate, 250 mM sucrose, and 0.01% polysorbate 80, at pH 7.0, referred to herein as the “CST formulation.” F3′-1602 was diafiltered into the CST formulation buffer and the recovered sample was passed through a 0.2 μm filter. The final concentration of F3′-1602 was 20 mg/mL, as measured by A280.

Stability of F3′-1602 in the HST Formulation

An accelerated stability study was carried out in the HST buffer at 40° C. for four weeks. Size exclusion chromatography (SEC) was used to monitor the formation of high molecular weight species (HMWS) and low molecular weight species (LMWS) as a function of storage conditions. Briefly, 50 μg of test material was injected onto an Agilent 1260 Infinity II high pressure liquid chromatography (HPLC) instrument with 1260 Quat Pump, 1260 Vialsampler, 1260 VWD. The sample was separated on a Tosoh TSKgel G3000SWxl, 7.8 mm I.D.×30 cm, 5 μm column. SEC running buffer was PBS, pH 7.0, flowing at 0.50 ml/min. Absorbance was monitored at both 214 and 280 nm, peak areas were manually integrated and the percent of HMWS, LMWS, and monomer were reported.

At the onset of the study, F3′-1602 was analyzed by SEC and was 99.6% monomer. After four weeks at 40° C., monomer decreased to 93.1%. LMWS were not detected in the control and increased to 5.4% after four weeks at 40° C. HMWS were present in the control at 0.4% and increased to 1.5% after four weeks at 40° C. (Table 63). These data suggest that F3′-1602 degrades primarily through the formation of LMW species (fragmentation).

TABLE 63

Summary of SEC % monomer, HMW, and LMW

(accelerated stability in the HST formulation)

Test article
Monomer (%)
HMW (%)
LMW (%)

Control
99.6
0.4
ND

1 week
99.0
0.7
0.3

2 weeks
98.5
1.0
0.5

3 weeks
97.8
1.5
0.8

4 weeks
93.1
1.5
5.4

ND: none detected

The binding affinities of F3′-1602 samples to recombinant human CLEC12A were measured by SPR at 37° C. using a Biacore 8K instrument. Briefly, human Fc specific antibodies were covalently immobilized at a density ˜8000-10000 resonance units (RU) on carboxy methyl dextran matrix of a CM5 biosensor chip via amine-coupling chemistry to create an anti-hFc IgG chip. F3′-1602 samples were captured on the anti-hFc IgG chip at a concentration of 1.5 μg/mL at a flow rate of 10 μL/min for 60 seconds to achieve ˜150-250 RU capture level. hCLEC12A-His was serially diluted (100 nM-0.046 nM) in three-fold dilutions with HBS-EP+ (1×) with 0.1 mg/mL bovine serum albumin (BSA) buffer and injected at a flow rate of 30 μl/min over the captured test articles. Association was monitored for 300 seconds and dissociation was monitored for 600 seconds. Surfaces were regenerated between cycles with three pulses of 10 mM glycine-HCl, pH 1.7 injected for 20 seconds at 100 μL/min. HBS-EP+ (1×) with 0.1 mg/mL BSA buffer was used throughout the experiment. Data were analyzed using Biacore 8K Insight evaluation software (GE Healthcare). The kinetic rate constants and overall affinity were obtained using 1:1 kinetic fit binding model (FIGS. 71A-71B).

After four weeks of incubation at 40° C. in the HST formulation, binding of F3′-1602 to human CLEC12A was decreased (FIG. 71B). The difference in association and dissociation rate constants between the control and 40° C. stressed sample are within the experimental variability of the assay (Table 64); however, the maximal binding response expressed in resonance units (RU) decreased more than fivefold after the incubation at 40° C. (FIG. 71B). The nominal capture of F3′-1602 protein was the same for both, stressed and control, samples. Under such circumstances, R_maxis roughly proportional to the active concentration of F3′-1602 captured on the Biacore chip. This result in turn indicates that after thermal stress in the HST formulation, more than 80% of F3′-1602 molecules lost the ability to bind CLEC12A (Table 64).

TABLE 64

Summary of kinetic parameters and binding affinity of F3’-1602 for hCLEC12A

(accelerated stability in the HST formulation)

Capture

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)
level (RU)

F3’-1602 Control
hCLEC12A
7.91 × 10⁵
8.73 × 10⁻⁴
1.11
83.8

F3’-1602 in the HST
hCLEC12A
1.24 × 10⁶
2.14 × 10⁻³
1.69
84.2

formulation, 40° C. 4 wks

Results are an average of n = 2 replicates

The binding affinities of F3′-1602 samples to recombinant human NKG2D were measured by SPR at 37° C. using a Biacore 8K instrument. Briefly, mFc specific antibodies were covalently immobilized at a density of ˜8000-10000 RU on carboxy methyl dextran matrix of a CM5 biosensor chips via amine coupling chemistry to create an anti-mFc IgG chip. mFc-tagged human NKG2D was captured on the anti-mFc IgG chip at a concentration of 0.2 μg/mL at a flow rate of 10 μL/min for 60 seconds to achieve ˜35 RU capture level. F3′-1602 was buffer exchanged into HBS-EP+ (1×) buffer and serially diluted (5000 nM-9.78 nM) in two-fold dilutions with HBS-EP+. Analyte (F3′-1602) was injected at a flow rate of 30 μL/min over the captured test articles. Association was monitored for 60 seconds and dissociation was monitored for 60 seconds. Surfaces were regenerated between cycles with three pulses of 10 mM glycine-HCl, pH 1.7 injected for 20 seconds at 100 μL/min. HBS-EP+ (1×) buffer was used throughout the experiment. Data were analyzed using Biacore 8K Insight evaluation software (GE Healthcare). The binding affinity and kinetic rates were obtained with the use of both 1:1 kinetic and steady-state affinity model fits (FIGS. 72A-72D).

After four weeks of incubation at 40° C. in the HST formulation, binding of F3′-1602 to human NKG2D was unaltered (FIGS. 72B and 72D, Table 65). The difference in association and dissociation rate constants as well as overall affinity for NKG2D between the control and 40° C. stressed sample were within the experimental variability of the assay. Maximal binding response did not change significantly. The results indicate that F3′-1602 maintained the ability to bind NKG2D under these thermal stress conditions.

TABLE 65

Summary of kinetic parameters and binding

affinity of F3′-1602 for hNKG2D (accelerated stability in the HST formulation)

1:1 Kinetic
Steady
Capture

Fit K_D
state K_D
level

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
(nM)
(RU)

F3′-1602 Control
hNKG2D
1.70 × 10⁵
1.29 × 10⁻¹
760
771
50.1

F3′-1602 in the
hNKG2D
1.29 × 10⁵
9.86 × 10⁻²
762
776
47.4

HST formulation,

40° C., 4 wks

The binding affinities of F3′-1602 samples to biotinylated recombinant CD16a V158 (FcγRIIIa V158) receptor were measured by SPR. Briefly, FcγRIIIa V158 was captured on a SA Series S Biacore chip prior to the kinetic evaluation experiment to achieve capture ˜10-25 RU. Trastuzumab, a well characterized IgG1 biologic drug, was used as an experimental control. The evaluation used injections of varying concentrations of ligand (F3′-1602 or trastuzumab) titrated over the captured receptor (1500 nM-11.7 nM in 1:1 serial dilution and 0 nM). Each injection cycle consisted of association, dissociation and regeneration steps. Association consisted of 150 seconds of ligand injection, dissociation was monitored for 300 seconds and then the surface was regenerated between cycles with 2 mM sodium hydroxide injected for 5 seconds. All steps were executed at a flow rate of 30 μL/min. 1× HBS-EP+ buffer solution was used as a running/sample dilution buffer. Data were analyzed using Biacore 8K Insight evaluation software (GE Healthcare). The data were fit with a 1:1 kinetic model (FIGS. 73A-73B).

After four weeks of incubation at 20 mg/mL, 40° C. in the HST formulation, binding of F3′-1602 to human CD16a V158 was unaltered (FIG. 73A and Table 66). The difference in binding affinity between the control and 40° C. stressed sample as well as maximal binding response (R_max) were within the experimental variability of the assay. The results indicate that F3′-1602 retained affinity for human CD16a under these accelerated storage conditions.

TABLE 66

Summary of kinetic parameters and binding affinity of F3’-1602 for hNKG2D

(accelerated stability in the HST formulation)

Capture

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)
level (RU)

F3’-1602 Control
hCD16aV
(1.13 ± 0.04) × 10⁵
(2.40 ± 0.09) × 10⁻²
212 ± 1
13.7

F3’-1602 in the HST
hCD16aV
(8.24 ± 0.78) × 10⁴
(1.65 ± 0.06) × 10⁻²
201 ± 12
14.2

formulation, 40° C., 4 wks

Results are an average of n = 3 replicates

The effect of accelerated stability on F3′-1602 function was also assessed in a cytotoxicity assay using KHYG-1-CD16a cells engineered to express CD16a in addition to NKG2D. CLEC12A⁺ human cancer cell lines (PL-21) were labeled with BATDA reagent. After labeling, cells were washed and resuspended in primary cell culture media. BATDA labeled cells, F3′-1602, and rested KHYG-1-CD16V cells were added to the wells of a 96-well plate. Additional wells were prepared for maximum lysis of target cells by addition of 1% Triton-X. Spontaneous release was monitored from wells with only BATDA-labeled cells. After three hours of culture, the cells were pelleted, the culture supernatant was transferred to a clean microplate, and room temperature europium solution was added to each well. The plate was protected from light and incubated on a plate-shaker at 250 rpm for 15 minutes. Plates were read using a SpectraMax i3X instrument. The % Specific lysis was calculated as follows:

% Specific lysis=((Experimental release−Spontaneous release)/(Maximum release−Spontaneous release))*100%

As shown in FIG. 74 and Table 67, the ability of F3′-1602 to kill PL21 CLEC12A⁺ cancer cells decreased by approximately 7-fold after 4 weeks at 40° C. in the HST formulation, whereas max % lysis decreased only slightly (5%).

TABLE 67

EC50 and maximum lysis of CLEC12A + PL21 cells by F3’-1602

(accelerated stability in the HST formulation)

Test article
EC50 (nM)
Max. lysis (%)

F3’-1602 Control
0.17
63

F3’-1602 in the HST
1.21
58

formulation, 40° C., 4 wks

The significant loss of binding to hCLEC12A after 40° C. incubation and the corresponding loss in potency led to an investigation into the root cause of this behavior. UHPLC-MS/MS peptide mapping revealed that an aspartic acid in the anti-CLEC12A scFv (CDR-H3) was undergoing isomerization (FIG. 75B).

F3′-1602 CDRH3, as defined under Chothia, contains four aspartic acids (DYDDSLDY [SEQ ID NO:293]). MS/MS was used to determine which of them was undergoing isomerization by monitoring the succinimide intermediate. Conversion to isoaspartic acid does not result in a mass difference, whereas the succinimide intermediate corresponds to a loss of water (approximately 18 Da). This mass shift can be used to identify the aspartic acid which is isomerizing. MS/MS spectra of the unmodified chymotryptic peptide and the corresponding succinimide variant showed that the aspartic acid at position H99 (Chothia) in the scFv VH (DYDDSLDY [SEQ ID NO:293]) was responsible for the isomerization.

Stability of F3′-1602 in the Phosphate/Citrate/NaCl Formulation

Attempts to remove the liability by mutagenesis revealed that the aspartic acid at position H99 of the CLEC12A-binding scFv is an important determinant of CDR structure and cannot be easily replaced without the loss of binding to the target. Therefore, limited pre-formulation development was performed to identify a formulation with a more basic pH since isomerization rate is pH dependent. The pH dependence was confirmed by stressing F3′-1602 in citrate and phosphate buffer covering pH 5.5-7.5. After 4 weeks at 40° C. in phosphate buffer (pH 6.5/7.0/7.5), F3′-1602 showed a pH-dependent loss of binding to CLEC12A. Interestingly, there was no trend observed in citrate buffer (pH 5.5/6.0/6.5). Additional formulations were designed based on these results, and common excipients (NaCl, sucrose, EDTA) were screened for their ability to stabilize CLEC12A binding (Table 68). This screening indicated that high pH and NaCl had a protective effect on CLEC12A binding.

TABLE 68

Phosphate/Citrate formulation buffers

Buffer
Citrate
Phosphate
EDTA
Sucrose
NaCl

#
(mM)
(mM)
(mM)
(mM)
(mM)
pH

1
50
0
0
0
0
7.0

2
20
50
0
0
0
7.0

3
20
50
0
0
0
7.5

4
20
50
0
0
0
8.0

5
0
50
10
0
0
7.0

6
0
50
10
0
0
7.5

7
0
50
10
0
0
8.0

8
50
0
0
250
0
7.0

9
50
0
0
0
250
7.0

Based on these results, F3′-1602 was formulated in 20 mM potassium phosphate, 10 mM sodium citrate, 125 mM sodium chloride, pH 7.8 (the phosphate/citrate/NaCl formulation).

The formation of HMWS and LMWS was assessed by SEC as described in the “Stability of F3′-1602 in the HST formulation” subsection. F3′-1602 was incubated at 40° C. for 4 weeks. At the onset of the study, F3′-1602 was analyzed by SEC and was 97.9% monomer. After four weeks at 40° C., monomer decreased to 91.4%. LMWS increased from 1.8 to 5.0% after four weeks at 40° C. HMWS were present in the control at 0.3% and increased to 3.6% after four weeks at 40° C. (Table 69). These data suggest that in the phosphate/citrate/NaCl formulation, F3′-1602 degraded through both aggregation (HMWS) and fragmentation (LMWS).

TABLE 69

Summary of SEC % monomer, HMW, and LMW

(accelerated stability in the phosphate/citrate/NaCl formulation)

Test article
Monomer (%)
HMW (%)
LMW (%)

F3’-1602 control
97.9
0.3
1.8

F3’-1602, 4 weeks, 40° C.
91.4
3.6
5.0

Fragmentation and aggregation of intact F3′-1602 was assessed in the accelerated stability study by non-reduced sodium dodecyl sulfate capillary electrophoresis (SDS-CE). Briefly, F3′-1602 was diluted to 0.5 mg/mL with 1× Sample Buffer to achieve a final sample volume of 50 μL. Internal standard and iodoacetamide were added to samples, which were then incubated at 70° C. for 10 minutes in a heat block, followed by an ice bath for 5 minutes. Samples were transferred to a 96-well plate, centrifuged, and loaded into the instrument. Individual samples were loaded into the capillary for 40 seconds at 4600 volts and separated for 35 minutes at 5750 volts on a Maurice (ProteinSimple, San Jose, Calif.).

The F3′-1602 main peak purity decreased from 97.0% to 91.2% by the end of the 4-week incubation. There was a corresponding increase in LMWS from 3.0% to 8.4% (Table 70). There was a slight increase in HMWS (none detected to 0.4%) throughout the duration of the study as determined by non-reduced SDS-CE. The difference in migration time of the main peak in the overlay was within the experimental variability of this assay.

TABLE 70

Summary of non-reduced SDS-CE purity

(accelerated stability in the phosphate/citrate/NaCl formulation)

Test article
LMW (%)
Main (%)
HMW (%)

F3’-1602 control
3.0
97.0
ND

F3’-1602 4 weeks, 40° C.
8.4
91.2
0.4

ND, none detected

Fragmentation of the three individual polypeptide chains of F3′-1602 was assessed at each timepoint in the accelerated stability study by reduced SDS-CE. In this assay, purity was determined by the sum of the three chains. Over the course of the study, purity decreased from 97.4% to 95.2% (Table 71). This decrease in purity was the result of increased LMWS, which agrees with the SEC and non-reduced SDS-CE results. The difference in migration times of the main peaks in the overlay was within the experimental variability of this assay.

TABLE 71

Summary of reduced SDS-CE purity

(accelerated stability in the phosphate/citrate/NaCl formulation)

Light
Heavy
scFv-Fc

Test article
chain (%)
chain (%)
chain (%)
Purity (%)

F3’-1602 control
17.9
39.2
40.3
97.4

F3’-1602 4 weeks, 40° C.
20.3
39.4
35.5
95.2

The binding affinities of F3′-1602 samples to recombinant human CLEC12A were measured by SPR as described in the “Stability of F3′-1602 in the HST formulation” subsection (FIGS. 76A-76B). After four weeks of incubation in the phosphate/citrate/NaCl formulation, human CLEC12A binding was reduced to some extent (FIG. 76B). The difference in association and dissociation rate constants between the control and 40° C. stressed sample were within the experimental variability of the assay (Table 72), whereas the maximal binding response (Rmax) adjusted for capture differences decreased approximately 10% after incubation at 40° C. Such a loss was significantly less than more than 80% that was observed after 4 weeks of incubation in the HST formulation, indicating that the phosphate/citrate/NaCl formulation was able to stabilize F3′-1602 under accelerated (40° C.) conditions. Therefore, the isomerization of H99, as described in the “stability of F3′-1602 in the HST formulation” subsection, was substantially reduced by formulation development.

TABLE 72

Summary of kinetic parameters and binding affinity of F3’-1602 for hCLEC12A

(accelerated stability in the phosphate/citrate/NaCl formulation)

Capture

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)
level (RU)

F3’-1602 control
hCLEC12A
7.79 × 10⁵
9.33 × 10⁻⁴
1.20
160.6

F3’-1602 in the phosphate/Citrate/
hCLEC12A
8.29 × 10⁵
9.48 × 10⁻⁴
1.14
146.7

NaCl formulation, 40° C., 4 wks

Results are an average of n = 2 replicates

The binding affinities of F3′-1602 samples to recombinant human NKG2D were measured by SPR as described in the “Stability of F3′-1602 in the HST formulation” subsection (FIGS. 77A-77D). After four weeks of incubation at 20 mg/mL, 40° C. in the phosphate/citrate/NaCl formulation, binding of F3′-1602 to human NKG2D was unaltered (FIGS. 77B and 77D and Table 73). The difference in equilibrium binding affinity as well as Rmax between the control and 40° C. stressed sample were within the experimental variability of the assay. This indicates that F3′-1602 retained affinity for human NKG2D under these accelerated storage conditions.

TABLE 73

Summary of kinetic parameters and binding affinity of F3′-1602 for hNKG2D

(accelerated stability in the phosphate/citrate/NaCl formulation)

1:1 Kinetic
Steady
Capture

Fit K_D
state K_D
level

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
(nM)
(RU)

F3′-1602
hNKG2D
1.93 × 10⁵
1.53 × 10¹
790
791
34.5

Control

F3′-1602 in
hNKG2D
1.53 × 10⁵
1.33 × 10¹
869
872
31.4

the phosphate/

citrate/NaCl

formulation,

40° C., 4 wks

Results are an average of n = 2 replicates

The binding affinities of F3′-1602 samples to biotinylated recombinant CD16a V158 (FcγRIIIa V158) receptor were measured by SPR as described in the “Stability of F3′-1602 in the HST formulation” subsection (FIGS. 78A-78B). After four weeks of incubation at 20 mg/mL, 40° C. in the phosphate/citrate/NaCl formulation, binding of F3′-1602 to human CD16a V158 was unaltered (FIG. 78B and Table 74). The difference in binding affinity as well as apparent maximal binding response between the control and 40° C. stressed sample are within the experimental variability of the assay. This indicates that F3′-1602 retains affinity for human CD16a under these accelerated storage conditions.

TABLE 74

Summary of kinetic parameters and binding affinity of F3’-1602 for hCD16a V158

(accelerated stability in the phosphate/citrate/NaCl formulation)

Capture

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)
level (RU)

F3’-1602 Control^a
hCD16aV
(1.13 ± 0.04) × 10⁵
(2.40 ± 0.09) × 10⁻²
212 ± 1
13.7

F3’-1602 in the phosphate/citrate/
hCD16aV
9.67 × 10⁴
2.09 × 10⁻²
216
14.0

NaCl formulation, 40° C., 4 wks

The effect of accelerated stability on F3′-1602 function was also assessed in a cytotoxicity assay using NKG2D- and CD16a-expressing KHYG-1-CD16a cells, as described in the “Stability of F3′-1602 in the HST formulation” subsection. The ability of F3′-1602 to kill PL-21 CLEC12A+ cancer cells was unaltered after 4 weeks incubation at 40° C. in the phosphate/citrate/NaCl formulation at a concentration of 20 mg/mL (FIG. 79 and Table 75). This result was in stark contrast to what was observed after incubation in the HST formulation, and indicates that F3′-1602 was stabilized in 20 mM phosphate, 10 mM citrate, 125 mM NaCl, pH 7.8.

TABLE 75

EC50 and maximum lysis of CLEC12A + PL-21 cells by F3’-1602

(accelerated stability in the phosphate/citrate/NaCl formulation)

Test article
EC50 (nM)
Max. lysis (%)

F3’-1602 Control
0.26
90

F3’-1602 in the phosphate/citrate/
0.29
93

NaCl formulation, 40° C., 4 wks

Stability of F3′-1602 in the CST Formulation

Based on the stability data in the phosphate/citrate/NaCl formulation, it was concluded that the loss of activity at 40° C. could be minimized in a liquid formulation. Stability of F3′-1602 was further analyzed in a lyophilization-compatible liquid formulation containing sucrose as a lyoprotectant.

An accelerated stability study at 20 mg/mL was carried out in 20 mM citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 (the CST formulation) at refrigerated (2-8° C.) and accelerated (25 and 40° C.) conditions for four weeks. A full suite of analytics was carried out on the 40° C. stressed samples while only binding to CLEC12A (SPR) and potency were monitored for the 2-8° C. and 25° C. arms.

The formation of HMWS and LMWS was assessed by SEC as described in the “Stability of F3′-1602 in the HST formulation” subsection. At the onset of the study, F3′-1602 was analyzed by SEC and was 99.6% monomer. After four weeks at 40° C., monomer decreased to 98.6%. LMWS were not detected in the control and increased to 0.5% after four weeks at 40° C. HMWS were present in the control at 0.4% and increased to 1.0% after four weeks at 40° C. (Table 76). These data suggest that F3′-1602 was resistant to both aggregation and fragmentation under accelerated conditions in the CST formulation.

TABLE 76

Summary of SEC % monomer, HMW, and LMW

(accelerated stability in the CST formulation)

Test article
Monomer (%)
HMW (%)
LMW (%)

Control
99.6
0.4
ND

1 week
99.3
0.5
0.2

2 weeks
99.0
0.7
0.3

3 weeks
98.7
0.9
0.4

4 weeks
98.6
1.0
0.5

ND, none detected

Fragmentation of intact F3′-1602 was assessed at each timepoint in the accelerated stability study by non-reduced SDS-CE, as described in the “Stability of F3′-1602 in the phosphate/citrate/NaCl formulation” subsection. The F3′-1602 main peak purity decreased from 96.1% to 91.0% by the end of the 4-week incubation. There was a corresponding increase in LMWS from 3.9% to 9.0% (Table 77). There was no increase in HMWS throughout the duration of the study as determined by non-reduced SDS-CE. The difference in migration time of the main peak in the overlay was within the experimental variability of this assay.

TABLE 77

Summary of non-reduced SDS-CE purity

(accelerated stability in the CST formulation)

Test article
LMW (%)
Main (%)
HMW (%)

Control
3.9
96.1
ND

1 week
4.8
95.2
ND

2 weeks
6.6
93.4
ND

3 weeks
5.3
94.7
ND

4 weeks
9.0
91.0
ND

ND: none detected

Fragmentation of the three individual polypeptide chains of F3′-1602 was assessed at each timepoint in the accelerated stability study by reduced SDS-CE, as described in the “Stability of F3′-1602 in the phosphate/citrate/NaCl formulation” subsection. In this assay, purity was determined by the sum of the three chains. Over the course of the study, purity decreased from 98.3% to 97.7% (Table 78). The difference in migration times of the main peaks in the overlay was within the experimental variability of this assay.

TABLE 78

Summary of reduced SDS-CE purity

(accelerated stability in the CST formulation)

Test article
Purity (%)

Control
98.3

1 week
98.0

2 weeks
98.0

3 weeks
97.8

4 weeks
97.7

Purity was the sum of chain L, H, and S % area

The binding affinities of F3′-1602 samples to recombinant human CLEC12A were measured by SPR as described in the “Stability of F3′-1602 in the HST formulation” subsection (FIG. 80A). After four weeks of incubation at 20 mg/mL, 40° C. in the CST formulation, human CLEC12A binding was reduced (FIG. 80B and Table 79). The difference in association and dissociation rate constants between the control and 40° C. stressed sample are within the experimental variability of the assay, however apparent binding response decreased about 15-17% after the stress at 40° C. This indicates that 15-17% of F3′-1602 lost the ability to bind CLEC12A after 4 weeks under these thermal stress conditions. While the classical SPR binding assay is a good indicator of potential loss of activity when it comes to stark differences in binding (20% or more), it is not reliable enough to address the impact of the smaller changes in activity of the sample (especially at the low replicate number); other analytical methods are used to elucidate the changes with more scientific certainty if needed.

TABLE 79

Summary of kinetic parameters and binding affinity of F3’-1602 for hCLEC12A

(40° C. accelerated stability in the CST formulation)

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)
Capture level (RU)

F3’-1602 control
hCLEC12A
7.50 × 10⁵
7.35 × 10⁻⁴
0.98
221.3

F3’-1602 in the CST
hCLEC12A
7.84 × 10⁵
6.74 × 10⁻⁴
0.86
211.5

formulation, 40° C., 4 wks

Results are an average of n = 2 replicates

At the lower temperatures (refrigerated and 25° C.), no difference in affinity to hCLEC12A or apparent binding adjusted for differences in capture levels were observed after 4 weeks in the CST formulation (FIGS. 81A-81C, Table 80, and Table 81).

TABLE 80

Summary of kinetic parameters and binding affinity of

F3′-1602 for hCLEC12A (2-8° C. stability in the CST formulation)

Capture

KD
level

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
(RU)

F3′-1602 control
hCLEC12A
7.50 × 10⁵
7.35 × 10⁻⁴
0.98
221.3

F3′-1602 in the
hCLEC12A
6.72 × 10⁵
6.00 × 10⁻⁴
0.89
352.4

CST formulation,

2-8° C., 2 wk

F3′-1602 in the
hCLEC12A
7.76 × 10⁵
7.08 × 10⁻⁴
0.91
222.8

CST formulation,

2-8° C., 4 wk

Results are an average of n = 2 replicates

TABLE 81

Summary of kinetic parameters and binding affinity of F3′-1602

for hCLEC12A (25° C. stability in the CST formulation)

Capture

K_D
level

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
(RU)

F3′-1602 control
hCLEC12A
7.50 × 10⁵
7.35 × 10⁻⁴
0.98
221.3

F3′-1602 in the
hCLEC12A
8.24 × 10⁵
7.65 × 10⁻⁴
0.93
187.6

CST formulation,

25° C., 2 wk

F3′-1602 in the
hCLEC12A
7.88 × 10⁵
7.19 × 10⁻⁴
0.91
179.9

CST formulation,

25° C., 4 wk

Results are an average of n = 2 replicates

The binding affinities of F3′-1602 samples to recombinant human NKG2D were measured by SPR as described in the “Stability of F3′-1602 in the HST formulation” subsection (FIGS. 82A-82D). After four weeks of incubation at 20 mg/mL, 40° C. in the CST formulation, binding of F3′-1602 to human NKG2D was unaltered (FIGS. 82B and 82D and Table 82). The difference in equilibrium binding affinity as well as apparent Rmax between the control and 40° C. stressed sample were within the experimental variability of the assay. This indicates that F3′-1602 retained affinity for human NKG2D under these accelerated storage conditions.

TABLE 82

Summary of kinetic parameters and binding affinity of F3′-1602 for hNKG2D

(accelerated stability in the CST formulation)

1:1 Kinetic
Steady

Fit
state
Capture

K_D
K_D
level

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
(nM)
(RU)

F3′-1602
hNKG2D
(2.11 ± 0.06) × 10⁵
(1.39 ± 0.03) × 10¹
661 ± 9
664 ± 10
34.5

Control

F3′-1602 in
hNKG2D
(1.98 ± 0.11) × 10⁵
(1.40 ± 0.08) × 10¹
708 ± 45
711 ± 45
33.2

the CST

formulation,

40° C., 4 wks

Results are an average n = 3 measurements

The binding affinities of F3′-1602 samples to biotinylated recombinant CD16a V158 (FcγRIIIa V158) receptor were measured by SPR as described in the “Stability of F3′-1602 in the HST formulation” subsection (FIGS. 83A-83B). After four weeks of incubation at 20 mg/mL, 40° C. in the CST formulation, binding of F3′-1602 to human CD16a V158 was unaltered (FIG. 83B and Table 83). The difference in binding affinity and apparent R_maxbetween the control and 40° C. stressed sample are within the experimental variability of the assay. This indicates that F3′-1602 retains affinity for human CD16a under these accelerated storage conditions.

TABLE 83

Summary of kinetic parameters and affinity of F3′-1602

for hCD16a V158 (accelerated stability in the CST formulation)

Capture

K_D
level

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
(RU)

F3′-1602
hCD16aV
(1.13 ±
(2.40 ±
212 ±
13.7

Control

0.04) × 10⁵
0.09) × 10⁻²
1

F3′-1602, CST,
hCD16aV
(1.05 ±
(2.25 ±
213 ±
13.6

pH 7.0, 40° C.,

0.04) × 10⁵
0.09) × 10⁻²
3

4 wks

Results are an average of n = 3 replicates

TABLE 84

EC50 and maximum lysis of CLEC12A+ PL-21

cells by F3′-1602 (accelerated stability in CST pH 7.0)

EC50
Max.

Test article
(nM)
lysis (%)

F3′-1602 Control
0.51
78.6

F3′-1602 in the CST formulation, 40° C., 1 wk
0.47
77.7

F3′-1602 in the CST formulation, 40° C., 2 wk
0.52
74.7

F3′-1602 in the CST formulation, 40° C., 3 wk
0.47
74.9

F3′-1602 in the CST formulation, 40° C., 4 wk
0.53
74.9

Example 15
Manufacturability of F3′-1602 in the CST Formulation

Various conditions throughout manufacturing of F3′-1602 could affect its stability. This example was designed to assess the stability of F3′-1602 in the CST formulation through freeze/thaw, agitation, low and high pH, forced oxidation, and low pH hold. The methods of measuring protein integrity and function, where provided in Example 14, are not repeated in this example.

Freeze/Thaw

Stability during freeze/thaw cycles is important for biotherapeutics as process intermediates and bulk drug substance may be frozen to ensure stability between process steps. To create freeze/thaw stress, 8.0 mg F3′-1602 was buffer switched into 20 mM sodium citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 using Amicon Ultra 15 ml 30K devices and brought to 400 μL with the same buffer (˜20 mg/mL F3′-1602, concentration was confirmed by A280 with a Nanodrop One spectrophotometer). The concentrated sample was divided into 4×100 μL aliquots. The control was immediately stored at −80° C. The remaining three aliquots were subjected to freeze/thaw cycles at −80° C. in a Styrofoam container to slow the freezing process. At cycles 2, 4, and 6, an aliquot was removed and transferred to −80° C. storage until analysis.

The freeze/thaw stability of F3′-1602 was assessed by measuring protein concentration by A280 at the completion of the study. F3′-1602 concentration was 20.4 mg/mL in the control and 20.2 mg/mL after 6 freeze/thaw cycles indicating that there was no loss of protein due to the freeze/thaw stress.

To assess whether F3′-1602 is stable through the freeze/thaw cycles, test article was frozen and thawed six times and analyzed by SEC after cycles 2, 4, and 6. F3′-1602 at time 0 was 99.5% monomer and the % monomer remained unchanged through all six freeze/thaw cycles (Table 85). This result indicates that F3′-1602 was resistant to both aggregation and fragmentation during freeze/thaw stress.

TABLE 85

Summary of SEC % monomer, HMW, and LMW (freeze/thaw stability)

Test article
Monomer (%)
HMW (%)
LMW (%)

Control
99.5
0.5
ND

2 F/T cycles
99.8
0.2
ND

4 F/T cycles
99.8
0.2
ND

6 F/T cycles
99.8
0.2
ND

ND: none detected

Fragmentation of intact F3′-1602 was assessed in the freeze/thaw stability study by non-reduced SDS-CE. The F3′-1602 main peak purity decreased minimally (96.1% to 95.5%) after 6 freeze/thaw cycles. There was a corresponding increase in LMWS from 3.9% to 4.5% (Table 86). No HMWS was detected throughout the duration of the study. These results indicate that F3′-1602 was resistant to fragmentation in 20 mM citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0. The difference in migration time of the main peak in the overlay was within the experimental variability of this assay.

TABLE 86

Summary of non-reduced SDS-CE purity (freeze/thaw stability)

Test article
LMW (%)
Main (%)
HMW (%)

Control
3.9
96.1
ND

2 F/T cycles
4.3
95.7
ND

4 F/T cycles
4.4
95.6
ND

6 F/T cycles
4.5
95.5
ND

ND: none detected

Fragmentation of the 3 individual polypeptide chains of F3′-1602 was assessed in the freeze/thaw stability study by reduced SDS-CE. In this assay, purity was determined by the sum of the three chains. F3′-1602 purity was unchanged after 6 freeze/thaw cycles (Table 87). These results indicate that F3′-1602 was resistant to fragmentation in 20 mM citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0. The difference in migration times of the main peaks in the overlay was within the experimental variability of this assay.

TABLE 87

Summary of reduced SDS-CE purity (freeze/thaw stability)

Test article
Purity (%)

Control
98.3

2 F/T cycles
98.0

4 F/T cycles
98.0

6 F/T cycles
98.0

Purity was the sum of chain L, H, and S percent area

Agitation

Stability during agitation is important for biotherapeutics as process intermediates and bulk drug substance may experience shear stress during processing steps. To create agitation stress, 8.0 mg F3′-1602 was buffer switched into 20 mM sodium citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 using Amicon Ultra 15 mL 30K devices and brought to 400 μL with the same buffer (˜20 mg/mL F3′-1602, concentration was confirmed by A280 with a Nanodrop One spectrophotometer). The concentrated sample was transferred into glass vial with stir bar, placed on stir plate and stirred at 60 rpm at 4° C. for 18 hours. After the agitation, the sample was stored at −80° C. until analysis.

The agitation stability of F3′-1602 was assessed. After agitation, there was visible protein precipitation. The sample was spun down, and supernatant was used in all characterization assays. Protein concentration in the supernatant was assessed by A280. F3′-1602 concentration was 20.4 mg/mL in the control and 16.8 mg/mL after agitation stress.

To determine if F3′-1602 is stable during agitation stress, test article was analyzed by SEC before and after agitation. F3′-1602 control was 99.5% monomer and the % monomer remained unchanged after agitation (Table 88). This indicates that F3′-1602 was resistant to both aggregation and fragmentation during agitation.

TABLE 88

Summary of SEC % monomer, HMW, and LMW (agitation stability)

Test article
HMW (%)
Monomer (%)
LMW (%)

Control
0.5
99.5
ND

Post-agitation
0.7
99.3
ND

ND: none detected

Fragmentation of intact F3′-1602 was assessed after agitation by non-reduced SDS-CE. The F3′-1602 main peak purity decreased minimally (96.1% to 95.5%) after agitation. There was a corresponding increase in LMWS from 3.9% to 4.5% (Table 89). No BMWS was detected throughout the duration of the study. These results indicate that F3′-1602 was resistant to fragmentation in 20 mM citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0. The difference in migration time of the main peak in the overlay was within the experimental variability of this assay.

TABLE 89

Summary of SEC % monomer, HMW, and LMW (agitation stability)

Test article
LMW (%)
Main (%)
HMW (%)

Control
3.9
96.1
ND

Post-agitation
3.8
96.2
ND

ND: none detected

Fragmentation of the three individual polypeptide chains of F3′-1602 was assessed in the agitation study by reduced SDS-CE. In this assay, purity was determined by the sum of the three chains. F3′-1602 purity was unchanged after agitation stress (Table 90). These results indicate that F3′-1602 was resistant to fragmentation in 20 mM citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0. The difference in migration times of the main peaks in the overlay was within the experimental variability of this assay.

TABLE 90

Summary of reduced SDS-CE purity (agitation stability)

Test article
Purity (%)

Control
98.3

Post-agitation
98.2

Purity was the sum of chains L, H, and S percent area

Low and High pH

To create low pH stress, 1.0 mg of F3′-1602 was buffer switched into 20 mM sodium acetate, pH 5.0 using Amicon Ultra 0.5 mL 10K devices and brought to 1.0 mL with the same buffer (˜1 mg/mL F3′-1602, concentration was confirmed by A280 with a Nanodrop One spectrophotometer). The sample was split. Half was stored at −80° C. as a control and the other half was incubated at 40° C. for 2 weeks. After the incubation, the sample was stored at −80° C. until analysis.

To create high pH stress, 1.0 mg of F3′-1602 was buffer switched into 20 mM Tris, pH 8.3 using Amicon Ultra 0.5 mL 10K devices and brought to 1.0 mL with the same buffer (˜1 mg/mL F3′-1602, concentration was confirmed by A280 on the Nanodrop One). The sample was split. Half was stored at −80° C. as a control and the other half was incubated at 40° C. for 2 weeks. After the incubation, the sample was stored at −80° C. until analysis. The pKa of Tris will change as a result of temperature. At 40° C., the pH 8.3 buffer was ˜pH 8.0.

To assess the chemical stability of F3′-1602, test article was held at pH 5.0 and pH 8.0 for two weeks at 40° C. At low pH, typical chemical modifications are aspartic acid isomerization and fragmentation, while at high pH, the primary degradation pathways are deamidation and oxidation. To confirm the integrity of F3′-1602 post-incubation, control and stressed samples were analyzed by SEC (Table 91). The results indicate that F3′-1602 remained largely intact with low levels of fragmentation as would be expected under these harsh conditions.

TABLE 91

Summary of SEC % monomer, HMW, and LMW (pH 5 and 8 stress)

Test article
Monomer (%)
HMW (%)
LMW (%)

F3′-1602 pH 5 control
99.8
0.2
ND

F3′-1602 pH 5 stressed
99.4
0.2
0.3

F3′-1602 pH 8 control
99.0
1.0
ND

F3′-1602 pH 8 stressed
95.2
0.9
3.9

ND: none detected

Chemical modification of amino acid side chains can typically be observed at a global scale with capillary isoelectric focusing (cIEF). To assess the modification of F3′-1602 samples held at low pH and high pH, F3′-1602 was diluted to 1 mg/mL with MilliQ water, 15 μL of sample was added to 60 μL of master mix (water, methyl cellulose, Pharmalyte 3-10, arginine, pI markers 4.05 and 9.99), vortexed, and centrifuged briefly. 60 μL of sample was aspirated from the top of the solution and added to a 96-well plate and centrifuged before testing. During the experiment, the sample was separated for one minute at 1500 volts followed by 8 minutes at 3000 volts on a Maurice (ProteinSimple, San Jose, Calif.).

The cIEF profile of F3′-1602 after being held at pH 5.0 for two weeks showed an increase in basic species from 8.9 to 39.1% (FIG. 85 and Table 92). A similar increase in acidic species was detected by cIEF after F3′-1602 was held at pH 8.0. A corresponding decrease in the main peak intensity (from 51.4 to 29.0%) was also observed (FIG. 86 and Table 92). Without wishing to be bound by theory, it is possible that deamidation throughout the F3′-1602 sequence lead to this acidic shift.

TABLE 92

Summary of F3′-1602 % acidic,

main, and basic species (pH 5 and 8 stress)

Acidic
Main
Basic

Test article
(%)
(%)
(%)

pH 5.0, control
39.8
51.4
8.9

pH 5.0, 2 wks, 40° C.
32.0
29.0
39.1

pH 8.0, control
39.6
52.2
8.4

pH 8.0, 2 wks, 40° C.
66.7
24.6
8.8

To determine whether the pH stress induced changes observed by cIEF was due to the modification in the complementarity determining regions (CDRs), the same pH 5 and pH 8 stressed samples were analyzed by LC-MS/MS peptide map. Briefly, 50 μg of F3′-1602 was diluted with 6 M guanidine hydrochloride in 250 mM Tris, pH 8.5, and disulfide bonds were reduced with 25 mM dithiothreitol for 30 minutes at 40° C. in a thermomixer. The reduced cysteines were then alkylated with 50 mM iodoacetamide for 60 minutes at room temperature in the dark. The reduced and alkylated protein was buffer-switched into 50 mM Tris, pH 8.0 and digested with trypsin (1:25 enzyme:substrate) for 1 hour at 37° C. LC-MS/MS was carried out on a Thermo Vanquish UHPLC coupled to a QExactive+ mass spectrometer. Briefly, 5 μg of protein digest was separated on a Waters UPLC BEH Peptide C18 column (2.1×150 mm) over 150 minutes using a 2% to 40% acetonitrile gradient at 0.3 mL/min. Mass spectra were acquired from 230-2000 m/z in positive mode at 35,000 resolution for the full MS scan and 17,500 resolution for the Top10 MS/MS scans. The UPLC-MS/MS files were analyzed using the Byos PTM workflow (Protein Metrics). Peptides were identified using accurate mass and MS/MS searching the F3′-1602 sequence.

Based on the peptide map data, all but one of the CDRs in F3′-1602 were resistant to modification during low and high pH stress. Only CDRH3 of the CLEC12A-binding scFv-Fc chain showed evidence of modification (isomerization) after the pH 5 stress (Table 93). This modification is the same one identified in the accelerated (40° C.) stability study.

TABLE 93

Post-translational modifications in F3′-1602 CDRs (pH 5.0 and pH 8.0)

Relative abundance (%)

Sequence
Potential
pH 5
pH 8

Target
CDR
(Chothia)
liability
control
stressed
control
stressed

CLEC
H1
GFSLTNY (SEQ
None
NA
NA
NA
NA

12A

ID NO: 137)

H2
WSGGK (SEQ ID
None
NA
NA
NA
NA

NO: 138)

H3
YDYDDSLDY
Isomerization
ND
Observed
ND
ND

(SEQ ID NO: 139)

L1
HASQNINFWLS
None
NA
NA
NA
NA

(SEQ ID NO: 140)

L2
EASNLHT (SEQ
None
NA
NA
NA
NA

ID NO: 141)

L3
QQSHSYPLT
None
NA
NA
NA
NA

(SEQ ID NO: 142)

NKG2D
H1
GFTFSSY (SEQ
None
NA
NA
NA
NA

ID NO: 297)

H2
SSSSSY (SEQ ID
None
NA
NA
NA
NA

NO: 298)

H3
GAPIGAAAGW
Truncation
ND
ND
ND
ND

FDP (SEQ ID

NO: 97)

L1
RASQGISSWLA
None
NA
NA
NA
NA

(SEQ ID NO: 86)

L2
AASSLQS (SEQ
None
NA
NA
NA
NA

ID NO: 77)

L3
QQGVSFPRT
None
NA
NA
NA
NA

(SEQ ID NO: 87)

ND: none detected; NA: not applicable

The binding affinities of F3′-1602 samples to recombinant human CLEC12A were measured by SPR as described in Example 14 (FIGS. 87A-87B). After two weeks of incubation at 40° C. in 20 mM sodium acetate, pH 5.0, binding of F3′-1602 to human CLEC12A was decreased (FIG. 87B and Table 94). The difference in association and dissociation rate constants between the control and low pH 40° C. stressed sample were within the experimental variability of the assay, however, apparent maximal binding response of the stressed sample was about half of the control after pH 5 stress. This indicates that F3′-1602 had a reduced ability to bind CLEC12A under these conditions (FIGS. 88A-88B). After two weeks of incubation at 40° C. in 20 mM Tris, pH 8.0, binding of F3′-1602 to human CLEC12A was unaltered (FIG. 88B and Table 94). This indicates that F3′-1602 retained affinity for CLEC12A under these accelerated pH stress conditions.

TABLE 94

Summary of kinetic parameters and binding affinity

of F3′-1602 for hCLEC12A (pH 5.0 and pH 8.0 stress)

Capture

k_a
k_d
K_D
level

Test article
Target
(M⁻¹s⁻¹)
(s⁻¹)
(nM)
(RU)

F3′-1602 Control
hCLEC12A
7.99 × 10⁵
8.53 ×
1.07
76.2

10⁻⁴

F3′-1602, pH 5.0,
hCLEC12A
8.65 × 10⁵
8.71 ×
1.01
87.4

40° C., 2 wks

10⁻⁴

F3′-1602 Control
hCLEC12A
8.01 × 105
8.40 ×
1.1
86.9

10 − 4

F3′-1602, pH 8.0,
hCLEC12A
8.35 × 105
8.45 ×
1.0
82.8

40° C., 2 wks

10 − 4

Results are an average of n = 2 replicates

The binding affinities of F3′-1602 samples to recombinant human NKG2D were measured by SPR as described in Example 14 (FIGS. 89A-89D and FIGS. 90A-90D). After two weeks of incubation at 40° C. in 20 mM sodium acetate, pH 5.0, binding of F3′-1602 to human NKG2D was unaltered (FIGS. 89B and 89D and Table 95). This indicates that F3′-1602 retained affinity for NKG2D under these accelerated pH stress conditions. After two weeks of incubation at 40° C. in 20 mM tris, pH 8.0, binding of F3′-1602 to human NKG2D was unaltered (FIGS. 90B and 90D and Table 95). This indicates that F3′-1602 retained affinity for NKG2D under these accelerated pH stress conditions.

TABLE 95

Summary of kinetic parameters and binding affinity of

F3’-1602 for hNKG2D (pH 5.0 and pH 8.0 stress)

1:1 Kinetic
Steady
Capture

Fit K_D
state K_D
level

Test article
Target
k_a(M⁻¹s⁻¹))
k_d(s⁻¹)
(nM)
(nM)
(RU)

F3’-1602 Control
hNKG2D
1.89 × 10⁵
1.28 × 10⁻¹
678
684
51.9

F3’-1602, pH 5.0,
hNKG2D
1.88 × 10⁵
1.27 × 10⁻¹
675
677
46.2

40° C., 2 wks

F3’-1602 Control
hNKG2D
1.80 × 10⁵
1.25 × 10⁻¹
694
695
47.2

F3’-1602, pH 8.0,
hNKG2D
1.35 × 10⁵
1.22 × 10⁻¹
905
919
51.0

40° C., 2 wks

The effect of low and high pH stress on F3′-1602 function was assessed in a cytotoxicity assay using NKG2D- and CD16a-expressing KHYG-1-CD16a cells as described in Example 14. After pH 5.0 and 8.0 stress, F3′-1602 potency against PL-21 CLEC12A+ cancer cells reflected by EC50 and maximum lysis was reduced ˜2-fold after pH 5 stress (FIG. 91 and Table 96) and was unchanged after pH 8 stress (FIG. 92 and Table 96). Without wishing to be bound by theory, it is possible that the loss of potency after pH 5 stress was due to the same isomerization of the scFv CDR-H3 aspartic acid identified in the initial accelerated stability study. The differences in the pH 5 and pH 8 results further demonstrates that F3′-1602 was more stable at elevated pH.

TABLE 96

EC50 and maximum lysis of CLEC12A+

PL-21 cells by F3′-1602 (pH 5.0 and pH 8.0 stress)

Test article
EC50 (nM)
Maximum Lysis (%)

F3′-1602 Control
0.18
91%

F3′-1602 pH 5.0, 2 wks, 40° C.
0.34
93%

F3′-1602Control
0.19
95%

F3′-1602 pH 8.0, 2 wks, 40° C.
0.24
90%

Forced Oxidation

One of the most common stability indicating modifications of proteins is oxidation of methionine residues. To identify oxidation hot spots, 1.0 mg of F3′-1602 was diluted to 1 mg/mL with phosphate buffered saline (PBS). 3.3 μL of 3% hydrogen peroxide was added to a 0.5 mL aliquot of F3′-1602 (0.02% hydrogen peroxide). Oxidation of F3′-1602 proceeded for 24 hours at room temperature in the dark. After 24 hours, the sample was buffer switched into PBS, pH 7.4 using Amicon Ultra 0.5 mL 10K MWCO devices following the manufacturer's instructions. The un-oxidized control (0.5 mL) was stored at −80° C.

After forced oxidation, there were no significant differences in F3′-1602% monomer, BMWS or LMWS (Table 97).

TABLE 97

Summary of SEC % monomer, HMW, and LMW (forced oxidation)

Test article
Monomer (%)
HMW (%)
LMW (%)

F3′-1602 Control
99.9
0.1
ND

F3′-1602 0.02% H₂O₂, 24 hrs
99.9
0.1
ND

ND: none detected

After forced oxidation, the purity of F3′-1602 by non-reduced SDS-CE was decreased by less than 1% (Table 98). This indicates that F3′-1602 was resistant to fragmentation during forced oxidation. There were no new fragments detected after forced oxidation. The peak shape of the oxidized sample changed after the oxidative stress. Without wishing to be bound by theory, this may indicate that there was no longer a homogeneous population and that the high levels of oxidation resulted in a conformational change not present in the control sample of F3′-1602. The difference in migration times of the main peaks in the overlay was within the experimental variability of this assay.

TABLE 98

Summary of non-reduced SDS-CE purity (forced oxidation)

Test article
LMW (%)
Main (%)
HMW (%)

F3’-1602, Control
3.9
96.1
ND

F3’-1602, 0.02% H₂O₂, 24 hrs
4.7
95.2
ND

ND: none detected

After forced oxidation, the purity of F3′-1602 by reduced SDS-CE was unchanged (Table 99). This indicates that F3′-1602 was resistant to fragmentation during forced oxidation. There were no new fragments detected after forced oxidation. The difference in migration times of the main peaks in the overlay was within the experimental variability of this assay.

TABLE 99

Summary of reduced SDS-CE purity (forced oxidation)

Chain L
Chain H
Chain S
Purity

Test article
(%)
(%)
(%)
(%)

F3’-1602, Control
17.7
40.1
40.5
98.3

F3’-1602, 0.02% H₂O₂, 24 hrs
17.8
40.0
40.4
98.2

In addition to determining the biophysical effects of forced oxidation by SEC, it is important to understand which amino acids are most susceptible to oxidation. There are no methionine residues in the F3′-1602 CDRs, however, several methionines are found in the framework and constant domains. Levels of oxidation were low at all sites in the control and only two sites showed significant increases in oxidation after exposure to hydrogen peroxide (Table 100). Low levels of oxidation have been observed in other therapeutic mAbs.

TABLE 100

Summary of site-specific percent oxidation in

F3′-1602 (forced oxidation)

Relative

abundance

SEQ ID

(%)

Peptide Sequence
NO
Chain
Position
Control
Oxidized

LSCAASGFTFSSYSMNWVR
299
H
34
0.6
0.8

NSLYLQMNSLR
300
H
83
0.2
0.3

DTLMISR
301
H/S*
257/283
2.3
71.3

WQQGNVFSCSVMHEALHNHYTQ
302
H/S*
433/454
1.5
47.4

K

DIQMTQSPSSVSASVGDR
303
L
4
0.3
0.4

YDYDDSLDYWGQGTLVTVSSGG
304
S
141
ND
ND

GGSGGGGSGGGGSGGGGSDIQM

TQSPSSLSASVGDRVTITCHAS

FQNINWLSWYQQKPGKAPK

*peptide is found in chain H and S;

ND, none detected

The binding affinities of forced oxidation stressed F3′-1602 samples to recombinant human CLEC12A were measured by SPR as described in Example 14 (FIGS. 93A-93B). After 24 hours of incubation at room temperature in presence of 0.02% hydrogen peroxide, kinetics and affinity of binding of F3′-1602 to human CLEC12A as well as apparent Rmax was unaltered (FIG. 93B and Table 101). This indicates that F3′-1602 retained binding to CLEC12A after forced oxidation and modifications of methionines in the constant regions and framework did not affect the affinity of the F3′-1602 for hCLEC12A.

TABLE 101

Summary of kinetic parameters and binding affinity of F3’-1602 for

hCLEC12A (forced oxidation)

Capture

k_a

K_D
level

Test article
Target
(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
(RU)

F3’-1602 Control
hCLEC12A
7.99 × 10⁵
8.58 × 10⁻⁴
1.1
78.6

F3’-1602 Forced
hCLEC12A
8.06 × 105
8.70 × 10⁻⁴
1.1
76.8

Oxidized

Results are an average of n = 2 replicates

The binding affinities of forced oxidation stressed F3′-1602 samples to recombinant human NKG2D were measured by SPR as described in Example 14 (FIGS. 94A-94D). After 24 hours of incubation at room temperature in presence of 0.02% hydrogen peroxide, kinetics and affinity of binding of F3′-1602 to human NKG2D was unaltered (FIGS. 94B and 94D and Table 102). This indicates that F3′-1602 retains binding to NKG2D after forced oxidation and modifications of methionines in the constant regions and framework do not affect the affinity of the F3′-1602 for hNKG2D.

TABLE 102

Summary of kinetic parameters and binding affinity of

F3’-1602 for hNKG2D (forced oxidation)

1:1 Kinetic
Steady
Capture

Fit K_D
state K_D
level

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
(nM)
(RU)

F3’-1602 Control
hNKG2D
(1.43 ± 0.03) × 10⁵
(1.07 ± 0.04) × 10⁻¹
750 ± 23
759 ± 23
44.1

F3’-1602, 0.02%
hNKG2D
(1.48 ± 0.06) × 10⁵
(1.07 ± 0.04) × 10⁻¹
722 ± 6
732 ± 7
48.2

H₂O₂, 24 hrs

The effect of forced oxidation stress on F3′-1602 function was assessed in a cytotoxicity assay using NKG2D- and CD16a-expressing KHYG-1-CD16a cells as described in Example 14. As shown in FIG. 95 and Table 103, after oxidative stress, F3′-1602 potency (EC50 and maximum lysis) was unchanged.

TABLE 103

EC50 and maximum lysis of CLEC12A+ PL-21 cells by F3’-1602

(forced oxidation)

Test article
EC50 (nM)
Maximum Lysis (%)

F3’-1602 Control
0.26
60%

F3’-1602 0.02% H₂O₂, 24 hrs
0.28
63%

Low pH Hold

Viral inactivation by low pH hold is an important step in biologics manufacturing. To assess the stability of F3′-1602 at low pH hold, F3′-1602 was captured on a MabSelect Sure Protein A column and eluted with 100 mM glycine, pH 3.0. Peak fractions were pooled, and the 12 mL of the protein A eluate was pH was adjusted to pH 3.3 with additional 100 mM glycine pH 3.0 buffer. The pH-adjusted eluate was held at room temperature. After 1.5 hours, the eluate was brought to neutral pH with 1.0 M Tris-HCl, pH 8.3. The pH-hold sample was further purified by cation exchange chromatography. The final product was buffer exchanged to PBS pH 7.4 for analysis.

After incubation at low pH, F3′-1602 was analyzed by SEC to determine if the hold caused aggregation. There was no increase in BMW and LMW content in F3′-1602 due to the low pH hold step (Table 104).

TABLE 104

Summary of SEC % monomer, HMW, and LMW (low pH hold)

Monomer
HMW
LMW

Test article
(%)
(%)
(%)

F3’-1602 Control, ProA eluate
77.4
11.5
11.1

F3’-1602 low pH hold, ProA eluate
86.4
3.4
10.2

After low pH hold, the purity of F3′-1602 by non-reduced SDS-CE was unchanged (Table 105). This indicates that F3′-1602 was resistant to fragmentation during low pH hold. There were no new fragments detected after the low pH hold. The difference in migration times of the main peaks in the overlay was within the experimental variability of this assay.

TABLE 105

Summary of non-reduced SDS-CE purity (low pH hold)

Test article
LMW (%)
Main (%)
HMW (%)

F3’-1602 Control
1.3
98.7
ND

F3’-1602 low pH hold
1.1
98.9
ND

ND, none detected

After low pH hold, the purity of F3′-1602 by reduced SDS-CE was unchanged (Table 106). This indicates that F3′-1602 was resistant to fragmentation during low pH hold. There were no new fragments detected after the low pH hold.

TABLE 106

Summary of reduced SDS-CE purity (low pH hold)

scFv-Fc

Light chain
Heavy chain
chain
Purity

Test article
(%)
(%)
(%)
(%)

F3’-1602 Control
18.1
39.9
40.4
98.4

F3’-1602 low pH hold
17.6
39.7
41.1
98.4

Chemical modification of amino acid side chains can typically be observed at a global scale with cIEF. The cIEF profiles of F3′-1602 control and low pH hold lot were visually very similar and the relative quantitation of acid, main, and basic species are all within 1% of each other, indicating that the low pH hold did not have a measurable effect on the charge profile of F3′-1602 (Table 107).

TABLE 107

Summary of F3’-1602 % acidic, main, and basic species

(low pH hold)

Acidic
Main
Basic

species
species
species

Test article
pI
(%)
(%)
(%)

F3’-1602 control
8.75
42.4
48.5
9.1

F3’-1602, low pH hold
8.80
42.7
49.1
8.2

F3′-1602 samples were analyzed by intact mass to monitor modification and fragmentation as a result of the low pH hold. The observed mass for both the control and low pH hold sample matched each other and the theoretical F3′-1602 mass (G0F/G0F glycoform: 126,360.4 Da). Low pH hold did not affect the mass of F3′-1602 (Table 108) and fragmentation was not observed.

TABLE 108

F3’-1602 theoretical and observed masses (low pH hold)

Delta
Mass

Theor. Mass
Obs. Mass
mass
accuracy

Test article
(Da)
(Da)
(Da)
(ppm)

F3’-1602 control
126,360.4
126,360.8
+0.4
3.2

F3’-1602 low pH hold
126,360.4
126,360.7
+0.3
2.4

Low pH hold did not induce any change in kinetics or affinity of F3′-1602 binding to human CLEC12A (FIG. 96B and Table 109). The difference in association and dissociation rate constants as well as apparent maximal binding response (Rmax) between the control and low pH hold sample were within the experimental variability of the assay (FIGS. 96A-96B). This indicates that F3′-1602 retained affinity for CLEC12A after a low pH hold step of biotherapeutic manufacturing.

TABLE 109

Summary of kinetic parameters and binding affinity of F3’-1602 for hCLEC12A

(low pH hold)

Capture

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)
level (RU)

F3’-1602
hCLEC12A
(6.78 ± 0.13) × 10⁵
(4.33 ± 0.20) × 10⁴
0.638 ± 0.028
176 ± 19

Control

F3’-1602
hCLEC12A
(7.15 ± 0.10) × 10⁵
(4.61 ± 0.03) × 10⁴
0.644 ± 0.008
157 ± 7

Low pH hold

Results are an average of n = 4 replicates

After low pH hold, human NKG2D binding was unaltered (FIGS. 97B and 97D and Table 110). The differences in steady state affinity and apparent maximal binding response affinity between the control and 40° C. stressed sample were within the experimental variability of the assay (FIGS. 97A-97D). This indicates that F3′-1602 retained affinity for human NKG2D after low pH hold.

TABLE 110

Summary of kinetic parameters and binding affinity of

F3’-1602 for hNKG2D (low pH hold)

1:1 Kinetic
Steady
Capture

Fit K_D
state K_D
level

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
(nM)
(RU)

F3’-1602 Control
hNKG2D
(2.06 ± 0.09) × 10⁵
(1.46 ± 0.08) × 10⁻¹
714 ± 62
710 ± 63
34.0

F3’-1602, low
hNKG2D
(2.07 ± 0.08) × 10⁵
(1.39 ± 0.02) × 10⁻¹
671 ± 20
672 ± 18
33.7

pH hold

Results are an average of n = 3 measurements

After low pH hold, binding of F3′-1602 to human CD16a V158 was unaltered (FIG. 98B and Table 111). The difference in binding affinity and maximal binding response between the control and low pH hold sample were within the experimental variability of the assay (FIGS. 98A-98B). This indicates that F3′-1602 retained affinity for human CD16a under these accelerated storage conditions.

TABLE 111

Summary of kinetic parameters and affinity of F3’-1602 for hCD16a

V158 (low pH hold)

Capture

Test

K_D
level

article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
(RU)

F3’-
hCD16aV
(1.13 ± 0.04) ×
(2.40 ± 0.09) ×
212 ± 1
13.7

1602

10⁵
10⁻²

Control

F3’-
hCD16aV
(1.09 ± 0.04) ×
(2.29 ± 0.04) ×
210 ± 3
14.1

1602,

10⁵
10⁻²

low pH

hold

Results are an average of n = 3 replicates

In addition to compromising biophysical stability and integrity, low pH could cause chemical degradation of amino acid side chains, in particular, aspartic acid isomerization. Chemical degradation of F3′-1602 in regions that are critical for function, if any, should manifest as reduced efficacy in a potency assay. Therefore, the potency of F3′-1602 against CLEC12A⁺ PL-21 cancer cells were tested after low pH hold in the KHYG1-CD16V bioassay. The potency of F3′-1602 was unaltered by the low pH hold (FIG. 99 and Table 112).

TABLE 112

EC50 and maximum lysis of CLEC12A+ PL-21 cells by F3’-1602

(low pH hold)

Test article
EC50 (nM)
Maximum Lysis (%)

F3’-1602 Control
0.27
61

F3’-1602 Low pH hold
0.15
54

Colloidal Stability

PEG precipitation was used to assess the colloidal stability of F3′-1602. Briefly, concentrated F3′-1602 (˜20 mg/mL) was diluted to 1 mg/mL in 100 μL aliquots (10 mM sodium acetate, pH 5.0) with final concentrations of PEG 6000 ranging from 0 to 30% (w/v). The aliquots were thoroughly mixed and placed at 2-8° C. overnight. The following day, the tubes were centrifuged to pellet any precipitation and the concentration of the supernatant was measured by A280 on a Lunatic (Unchained Labs).

The solubility of F3′-1602 in acetate buffer pH 5.0 with increasing concentration of PEG 6000 (0%-30%) was monitored by A280. The midpoint transition (Cm) in 10 mM sodium acetate, pH 5.0 was 24.5%, indicating that F3′-1602 has high solubility in this buffer (FIG. 100). As comparators, trastuzumab and adalimumab were also assessed in the PEG precipitation assay in 10 mM sodium acetate, pH 5.0. The midpoint transitions of trastuzumab and adalimumab, used as assay controls, were >30% and 11.7% PEG-6000, respectively.

F3′-1602 was then concentrated using 30 kDa molecular weight cutoff devices in PBS, pH 7.4. Concentration was determined by A280 on a Nanodrop and product quality was monitored by SEC. Test material was injected onto an Agilent 1260 Infinity II high pressure liquid chromatography instrument with 1260 Quat Pump, 1260 Vialsampler, 1260 VWD. The sample was separated on a Superdex200 Increase 10/300 GL column. SEC running buffer was PBS, pH 7.4, flowing at 0.60 mL/min. Absorbance was monitored at both 214 and 280 nm, peak areas were manually integrated and the percent of BMWS, LMWS, and monomer were reported.

F3′-1602 was concentrated to approximately 25, 50, 100, and 150 mg/mL in PBS, pH 7.4 and recovery was monitored by visual inspection, concentration (A280), and SEC. F3′-1602 was able to be concentrated to >150 mg/mL in PBS, pH 7.4 with minimal loss in percent monomer and no visible precipitation (FIG. 101 and Table 113).

TABLE 113

F3’-1602 HMWS, Monomer, and LMWS content as a function

of concentration

Concentration

(mg/mL)
HMWS (%)
Monomer (%)
LMWS (%)

4.1
ND
98.3
1.7

26.0
0.1
97.7
2.3

58.4
0.1
97.7
2.2

89.2
0.2
97.0
2.9

158.8
0.2
96.6
3.2

Example 16
Stability and Manufacturability of F3′-1602 and AB0010

F3′-1602 and AB0010 are TriNKETs in the F3′ and F3 formats, respectively, that have CLEC12A-binding sites derived from the same humanized anti-CLEC12A antibody. As described in Example 5 or 6, these two TriNKETs have similar binding affinity to CLEC12A. This example was designed to compare the stability and manufacturability of F3′-1602 and AB0010.

Briefly, F3′-1602 and AB0010 were formulated in 20 mM sodium phosphate, 10 mM sodium citrate, 125 mM NaCl, and 0.01% PS-80, at pH 7.8, as described in Example 14. The formulated proteins were subject to thermal stress at 40° C. for four weeks and their stabilities were assessed by the non-reduced SDS-CE, reduced SDS-CE, SEC, and cytotoxicity assays, as described in Example 14.

Data presented in Table 114 shows that the LMW fraction of the stressed AB0010 sample increased significantly more than that of the stressed F3′-1602 sample. Data presented in Table 115 shows that the purity of the stressed AB0010 sample decreased significantly more than that of the stressed F3′-1602 sample. These results indicate that AB0010 degraded significantly more than F3′-1602 under the thermal stress condition.

TABLE 114

Summary of non-reduced SDS-CE purity of F3′-1602 and AB0010

Test article
LMW (%)
Main (%)
HMW (%)

F3′-1602 Control
3.0
97.0
ND

F3′-1602 40° C. 4 wks
8.4
91.2
0.4

AB0010 Control
8.4
90.4
1.2

AB0010 40° C. 4 wks
27.8
71.4
0.9

ND, none detected

TABLE 115

Summary of reduced SDS-CE purity of F3’-1602 and AB0010

Test article
Purity (%)

F3’-1602 Control
97.4

F3’-1602 40° C. 4 wks
95.2

AB0010 Control
95.0

AB0010 40° C. 4 wks
89.3

As shown in Table 116, the amount of HMWS in the stressed AB0010 sample increased significantly more than that of the stressed F3′-1602 sample. This result suggests that AB0010 aggregated significantly more than F3′-1602 under the thermal stress conditions.

TABLE 116

Summary of SEC % monomer, HMW, and LMW of F3’-1602

and AB0010

Test article
HMWS (%)
Monomer (%)
LMWS (%)

F3’-1602 Control
0.3
97.9
1.8

F3’-1602 40° C. 4 wks
3.6
91.4
5.0

AB0010 Control
0.8
99.2
0.1

AB0010 40° C. 4 wks
13.8
81.7
4.5

As described in Example 14, sucrose is more compatible with a lyophilized presentation than NaCl. Therefore, F3′-1602 and AB0010 were also formulated in 20 mM sodium phosphate, 10 mM sodium citrate, 250 mM sucrose, and 0.01% polysorbate 80, at pH 7.8. The TriNKET proteins in each formulation were then subject to thermal stress at 40° C. for four weeks, and their ability to mediate NK cell-mediated cytotoxicity were assessed by the cytotoxicity assay as described in Example 14. As shown in FIGS. 102A and 102B and Table 117, the ability of F3′-1602 and AB0010 to mediate killing of PL21 CLEC12A⁺ cancer cells by NKG2D- and CD16a-expressing KHYG-1-CD16a cells was not significantly affected by the thermal stress.

TABLE 117

EC50 and maximum lysis of CLEC12A+ PL21 cells by

F3′-1602 or AB0010

Max.

EC50
lysis

Test article
(nM)
(%)

F3′-1602 Control
0.26
90

F3′-1602 40° C., 4 wks in 20 mM sodium phosphate,
0.34
93

10 mM sodium citrate, 250 mM sucrose, and 0.01%

polysorbate 80, at pH 7.8

F3′-1602 40° C., 4 wks in 20 mM sodium phosphate,
0.29
93

10 mM sodium citrate, 125 mM NaCl, and 0.01%

polysorbate 80, at pH 7.8

AB0010 Control
0.57
74

AB0010 40° C., 4 wks in 20 mM sodium phosphate,
0.75
74

10 mM sodium citrate, 250 mM sucrose, and 0.01%

polysorbate 80, at pH 7.8

AB0010 40° C., 4 wks in 20 mM sodium phosphate,
0.58
78

10 mM sodium citrate, 125 mM NaCl, and 0.01%

polysorbate 80, at pH 7.8

Example 17
Molecular Format, Design, Structure, and Characteristics of CLEC12A TriNKET AB0089
Objective

The objectives of this study are to describe the molecular format, design, structure, and characteristics of CLEC12A TriNKET AB0089. AB0089 is an improved version of AB0237, in which the primary sequence of the CLEC12A binding VH was optimized to remove the isomerization sequence liability while retaining affinity for CLEC12A and potency in cytotoxic lysis of CLEC12A+ AML cells. In addition, we aimed at reduction of the number of murine backmutations in the framework regions which increased humanness of the molecule. In particular, in this report we a) describe molecular format and design of AB0089; b) outline protein engineering efforts to remove sequence liability in CDRH3; c) describe protein engineering efforts to reduce number of murine backmutations; d) provide information about expression and purification of AB0089; e) describe basic biochemical and biophysical characterization of the molecule; f) determine the affinity of AB0089 for recombinant human targets (CLEC12A, NKG2D and CD16a); g) demonstrate AB0089 binding to AML cancer cell lines expressing human CLEC12A; h) demonstrate selectivity of AB0089; i) determine potency of AB0089 in killing AML cancer cells; j) assess binding epitope; k) assess binding to FcγR and FcR and compare with IgG1 mAb control; and 1) describe behavior of AB0089 in developability studies including accelerated stability, forced degradation, and manufacturability.

Study Design

During the developability assessment of CLEC12A specific AB0237 TriNKET, an isomerization liability in the CDRH3 was identified, which led to a reduction in binding to CLEC12A and potency under harsh stresses.Yeast display was used to remove the sequence liability and optimize CMC properties of the molecule. In addition, the framework of the molecule was further optimized to remove murine backmutations and replace them with the corresponding human residues, thus improving the humanness of the molecule. Properties of AB0089 were assessed by binding to recombinant hCLEC12A, binding to isogenic and cancer cells expressing CLEC12A as well as potency of the molecule against CLEC12A+ AML cancer cells. The biophysical and biochemical characteristics of AB0089 including thermal stability (DSC), hydrophobicity (HIC), and pI (cIEF) were determined. Molecular modeling was used to assess charge and hydrophobic patch distribution on the surface of AB0089 and to benchmark with clinical stage monoclonal antibodies. Specificity of binding was assessed by PSR, binding to cells not expressing the target of interest, and by HuProt™ Array. AB0089 was expressed in two different mammalian cell lines (Expi293 and ExpiCHO) and purified using Dragonfly Tx 2-step platform purification method. The binding properties of AB0089, including CLEC12A, NKG2D, CD16a (V158 and F158), an extended panel of Fcγ receptors (CD32a (H131 and R131), CD16b, CD32b, cynomolgus CD16, human and cynomolgus CD64), and FcRn (human and cynomolgus) were fully characterized. Lastly, AB0089 was assessed in a full developability panel of assays, including accelerated (40° C.) stability in a platform formulation (20 mM histidine, 250 mM sucrose, 0.01% Tween-80, pH 6.0) and 20 mM citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 buffers, forced degradation (long term pH 5, pH 8 stress, forced oxidation), and manufacturability (freeze/thaw, agitation, low pH hold).

Materials and Methods

Materials

Test Articles

TABLE 118

Summary of test articles.

Name
Lot
Expression host

AB0237
AB0237-002; 06Sep19AB
ExpiCHO

AB0089
AB0089-001
Expi293

AB0089
AB0089-002 (developability)
ExpiCHO

AB0089
AB0089-003 (low pH hold)
ExpiCHO

AB0089
AB0089-004 (low pH hold)
ExpiCHO

AB0192
AB0192-005
ExpiCHO

Protein Reagents

TABLE 119

Summary of protein reagents.

Name
Manufacturer
Lot

Human CLEC12A-His
Dragonfly Tx
16Oct18GL

Human CLEC12A-K244Q-His
Dragonfly Tx
30Apr18AC

Cyno CLEC12A-His
Dragonfly Tx
30May18AC

Biotinylated CLEC12A-His
Dragonfly Tx
17Apr19GL

mFc-hNKG2D
Dragonfly Tx
01Dec17AB

mFc-cNKG2D
Dragonfly Tx
28Sep18GL

humanized mAb 16B8.C8
Dragonfly Tx
AB0037-001

hcFAE-A49.CLL1-Merus
Dragonfly Tx
10May18DF

(duobody TriNKET)

hcFAE-A49.h6E7 (duobody
Dragonfly Tx
05Feb19DF

TriNKET)

hcFAE-A49.tepoditamab
Dragonfly Tx
27Aug19XL

(duobody TriNKET)

Trastuzumab
Roche
N3017H05

Adalimumab
AbbVie
83364XH07

Rituximab
Dragonfly Tx
19Apr19 MH

TABLE 120

Summary of recombinant Fc receptor proteins.

Reagent
Vendor
Catalog Number

Human CD64
ACRO Biosystems
FCA-H82E8

Cynomolgus CD64
ACRO Biosystems
FCA-C82E8

Human CD32a H131
ACRO Biosystems
CDA-H82E6

Human CD32a R131
ACRO Biosystems
CDA-H82E7

Human CD32b
ACRO Biosystems
CDB-H82E0

Human CD16a V158
Sino Biologicals
10389-H27H1-B

Human CD16a F158
Sino Biologicals
10389-H27H-B

Human CD16b
ACRO Biosystems
CDB-H82E4

Cynomolgus CD16
ACRO Biosystems
FC6-C82E0

Human FcRn
ACRO Biosystems
FCM-H5286

Cynomolgus FcRn
ACRO Biosystems
FCM-C5284

Methods

Cell Binding by Flow Cytometry

Binding of AB0089 to isogenic CLEC12A expressing cells and to PL-21 cancer cells was performed as described in detail in Example 19. The EC50 values were derived from the cell binding curves using GraphPad Prism software (four parameter logistic non-linear regression curve fitting model).

Yeast Library Construction and Evaluation

A walking mutagenesis strategy was employed using a synthetic CDRH3 oligonucleotide pool to construct a yeast display affinity library (referred to as Library 17) where each residue of the AB0237 CDRH3 (YDYDDSLDY, SEQ ID NO:139) was substituted with a randomization of all 20 amino acids, as a single site or in pairs (FIGS. 103A-103C).

First, a synthetic scFv double stranded DNA was designed by replacing the CDRH3 with a unique BamH1 restriction site. Then, the scFv was cloned into the yeast display plasmid (pYES-Aga2 vector) by Hifi cloning method, resulting in pYES-Aga2-scFv-BamH1 plasmid. DNA oligos with regions homologous to the vector were designed with the following randomizations introduced in the original CDRH3 (YDYDDSLDY, SEQ ID NO:139) to XDYDDSLDY (SEQ ID NO:305), YXYDDSLDY (SEQ ID NO:306), YDXDDSLDY (SEQ ID NO:307), YDYXDSLDY (SEQ ID NO:308), YDYDXSLDY (SEQ ID NO:309), YDYDDXLDY (SEQ ID NO:310), YDYDDSXDY (SEQ ID NO:311), YDYDDSLXY (SEQ ID NO:312), YDYDDSLDX (SEQ ID NO:313), XXYDDSLDY (SEQ ID NO:314), XDXDDSLDY (SEQ ID NO:315), XDYXDSLDY (SEQ ID NO:316), XDYDXSLDY (SEQ ID NO:317), XDYDDXLDY (SEQ ID NO:318), XDYDDSXDY (SEQ ID NO:319), XDYDDSLXY (SEQ ID NO:320), XDYDDSLDX (SEQ ID NO:321), where X=any amino acid) and pooled together. pYES-Aga2-scFv-BamHI vector was then digested with BamHI and this digested plasmid was electroporated with DNA oligo pools to yield the finalized yeast display Library 17. Through the process of homologous recombination, the yeast internally ligates the library oligos with the linearized plasmid, resulting in each yeast cell containing one plasmid with one CDRH3 mutated oligo. Following selection and characterization of Library 17 outputs, a similar affinity maturation strategy was employed targeting CDRH2 for walking mutagenesis in the background of a single optimized CDRH3. The library for this second round of affinity maturation is herein referred to as Library 30.

The library was assessed by FACS sorting after growing in D-UT media. 24 hours prior to sorting, the yeast were re-seeded to a final concentration of 0.5 O.D./ml (as measured by a visible light spectrophotometer), spun down, washed twice in GR-UT media, and then left to incubate overnight in the same media. On the day of the experiment, 3 O.D. of cells were taken be tested and placed in the well of a 96-well U-bottom plate. This was repeated up to the number of antigens desired to be tested. The cells were then washed twice with PBS-F (PBS+0.25% BSA), and then various concentrations of biotinylated CLEC12A-His were incubated for 30 min. Yeast cells were then stained with FITC labeled anti-Flag and streptavidin PE and sorted using the BD FACSAria for the highest affinity binders.

The final yeast libraries were analyzed by flow cytometry. The yeast library was initially seeded in D-UT media 24 hours prior to characterization. 24 hours prior to flow cytometry, the yeast were reseeded to a final concentration of 0.5 O.D./ml, spun down, washed twice in GR-UT media, and then left to incubate overnight in the same media. On the day of the experiment, 0.2 O.D. of cells were taken per antigen to be tested and placed in each well of a 96-well U-bottom plate. The cells were washed in PBS-F (PBS+0.25% BSA), and then biotinylated CLEC12A-His was added to a final concentration of 10 nM for 30 mins. The cells were washed again and non-biotinylated CLEC12A-His (final concentration 1 μM) was added for an additional 40 mins, with samples being pulled out every 20 minutes. Anti-Flag-FITC and streptavidin PE were added and incubated for 30 mins in the dark, on ice, washed twice, resuspended in PBS-F, and analyzed using the FACS Celesta.

Site-Directed Mutagenesis

The gBlock (double stranded DNA fragment) with DS to DT mutation encoding AB0089 scFv targeting CLEC12A arm was ordered from Integrated DNA Technologies (Coralville, Iowa). It was cloned into pcDNA3.4 vector following the protocol as described in below.

Codon Optimization

DNA codon optimization of the scFv portion of AB0089 Chain S was performed via Codon Optimization Tool from Integrated DNA Technologies (Coralville, Iowa), while the linker and Fc portion was codon optimized (as part of the vector backbone) using the Codon Optimization Tool from GeneArt Gene Synthesis (Thermo Fisher Scientific, Waltham, Mass., USA). DNA codon optimization for the expression of AB0089 Chain H and Chain L were performed using the Codon Optimization Tool from GeneArt Gene Synthesis (Thermo Fisher Scientific, Waltham, Mass., USA). The goal was to achieve high expression level in Expi293 cells (Thermo Fisher Scientific, Waltham, Mass., USA).

Cloning into Vectors for Transient Expression

AB0089 Chain H and L were synthesized and cloned into pcDNA3.4 vector using Thermo Fisher Scientific's GeneArt gene synthesis platform. The vector backbone for cloning AB0089 Chain S was prepared internally using the pcDNA3.4 vector (Thermo Fisher Scientific, Waltham, Mass., USA), by creating Kpnl and BamHI restriction sites for linearization and cloning using the NEBuilder HiFi Technology from New England Biolabs (Ipswich, Mass.). The general protocol for cloning using the NEBuilder HiFi Technology was followed. Briefly, the gBlocks (double stranded DNA fragments) expressing the gene of interest were cloned into the vector backbone, transformed using the TG-1 bacterial cells, DNA was harvested using the Zyppy Plasmid Miniprep Kit (Zymo Research, Irvine, Calif.) and the plasmid sequence was corroborated by Sanger sequencing (Genewiz, South Plainfield, N.J.). Once sequence corroborated, maxi-preps were performed to isolate plasmid DNA using the Nucleo Bond Xtra Maxi EF Kit (Macherey-Nagel, Bethlehem, Pa.).

Expression

Transient Expression in Expi293 Cells

Transient transfection of Expi293 cells was performed according to the general protocol provided by manufacturer (Life Technologies, Thermo Fisher Scientific, Waltham, Mass.) with the exception that PEI reagent was used for transfection. Expi293 cells were transfected with the 3 DNA chains created for transient expression of the 3 polypeptides that make up AB0089: pETseq0854 (Chain H), pETseq0424 (Chain L) and pETseq1739 (Chain S) using the plasmid ratio 4:1:1 to achieve a high percentage of heterodimer. Briefly, DNA-PEI complexes were created by mixing DNA and PEIpro Reagent (Polyplus Transfection; Illkirch, France) separately first in OptiMEM media (Life Technologies, Thermo Fisher Scientific, Waltham, Mass.), followed by mixing and subsequent addition to the Expi293 cells (>95% viability; VCD of 2.5×10⁶/mL). Following transfection, the cells were cultured in a humidified shaker-incubator at 37° C. and 8% CO₂. The cell supernatant was harvested 5 days later.

Transient Expression in ExpiCHO Cells

Transient transfections of ExpiCHO cells were performed according to the general protocol outlined by the manufacturer (Life Technologies, Thermo Fisher Scientific, Waltham, Mass.). ExpiCHO cells were transfected with the 3 DNA chains created for expression of the 3 polypeptides that make up AB0089: pETseq0854 (Chain H), pETseq0424 (Chain L) and pETseq1739 (Chain S) using the plasmid ratio 4:1:1 to achieve a high percentage of heterodimer. Briefly, DNA-Expifectamine complexes were created by mixing DNA and Expifectamine Reagent (Life Technologies) separately first in OptiPro media (Life Technologies), followed by mixing and subsequent addition to the ExpiCHO cells (>95% viability; VCD of 6×106/mL). Following transfection, the cells were cultured in a humidified shaker-incubator at 37° C. and 8% CO₂. ExpiCHO Feed and enhancer were added 18-20 hours later to boost protein expression, as per the protocol. The cell supernatant was harvested 9-11 days later, depending on cell viability (>70%).

Harvest

For harvest of AB0089, the cell culture supernatants were transferred to 500 mL or 1 L polycarbonate centrifuge bottles and spun at 7500 rpm for 35 mins at 4° C., followed by filtration using 0.2-micron filtration bottles. Below materials/reagents (vendor; catalog number) were used:

- Vi-CELL XR Cell Counter (Beckman Coulter)
- Vi-CELL 4 mL sample cups (Beckman Coulter; 08229NT2)
- Vi-Cell Reagent Pack (Beckman Coulter; B94987)
- 1L Polycarbonate Centrifuge bottle (Beckman Coulter; 366751)
- 500 mL Polycarbonate Centrifuge bottle (Beckman Coulter; 355649)
- Nalgene® Rapid-Flow™ Filter Units and Bottle Top Filters, PES Membrane, Sterile, Thermo Scientific, 1 L (VWR; 73520-986)
- Nalgene® Rapid-Flow™ Filter Units and Bottle Top Filters, PES Membrane, Sterile, Thermo Scientific, 500 mL (VWR; 73520-984)

Purification

Protein A Capture Chromatography

A 5.0 cm diameter SNAP column was packed to a bed height of 5.6 cm with Protein A MabSelect SuRe gel for a column volume of approximately 110 ml. Protein A method steps and buffers are shown in Table 121. Elution fractions were neutralized to approximately pH 7.0 using 1.0 M Tris, pH 8.3 (5-10% of elution volume). Below materials/reagents (vendor; catalog number) were used:

SNAP Column (Essential Life Solutions; S-50/125-PASS-OE-FP10)

MabSelect SuRe Protein A Gel (GE Healthcare; 175-43-802)

5× Protein A Binding Buffer (500 mM Sodium Phosphate, 750 mM Sodium Chloride pH 7.2) (Boston Bioproducts; C-6332)

1× Protein A Binding Buffer (100 mM Sodium Phosphate, 150 mM Sodium Chloride, pH 7.2) (Diluted 5× stock)

100 mM Glycine pH 3.0 (Boston Bioproducts; BB-95-100 mM)

1.0 M Tris, pH 8.3 (Boston Bioproducts; LT-783)

10.0 N Sodium Hydroxide (VWR; BDH7247-4)

0.1 N Sodium Hydroxide (Diluted 10.0N stock)

TABLE 121

Protein A MabSelect Sure chromatography method steps.

Volume
Flow Rate

Step
Buffer/Material
(CV)
(mL/min)

Equilibration
1× Protein A Binding Buffer
5
10

Sample Application
Conditioned Media
10

Wash
1× Protein A Binding Buffer
20
10

Elution
100 mM Glycine, pH 3.0
7
10

Strip
0.1 N Sodium Hydroxide
4
10

Re-Equilibration
1× Protein A Binding Buffer
5
10

Cation Exchange Chromatography

Cation Exchange chromatography (CIEX) was performed using Poros XS resin packed to a bed height of approximately 24.5 cm in a XK-26 column body for a final column volume of 130 ml. Equilibration and wash buffer (Buffer A) was 50 mM sodium acetate pH 5.0 with 10 mM sodium chloride. Elution buffer (Buffer B) was 50 mM sodium acetate pH 5.0 with 1.0 M sodium chloride.

The eluate from the Protein A column was pH- and conductivity-adjusted by diluting at least five-fold with Buffer A. After washing the loaded column, the protein was eluted with steps of increasing sodium chloride concentration. The IEX chromatography method steps are shown in Table 122.

TABLE 122

Cation exchange chromatography method steps.

%
Volume
Flow Rate

Step
Buffer
B
(CV)
(mL/min)

Equilibration
50 mM Sodium Acetate, 10 mM NaCl, pH 5.0
0
7
15.0

Sample
Neutralized ProA EL Buffer, Diluted ~10× with
0
NA
15.0

Application
IEX EQ buffer

Wash
50 mM Sodium Acetate, 10 mM NaCl, pH 5.0
0
5
15.0

Elution Step
Buffer A: 50 mM NaOAc, 10 mM NaCl, pH 5.0
10
4
15.0

Elution Step
Buffer B: 50 mM NaOAc, 1 M NaCl, pH 5.0
15
15
15.0

Elution Step

18
15
15.0

Elution Step

20
15
15.0

Elution Step

30
7
15.0

Elution Step

100
7
15.0

Strip/Sanitization
0.1 N NaOH
0
5
15.0

Buffer Exchange by G25 Chromatography and Diafiltration

The IEX product was buffer exchanged to PBS pH 7.4 (ThermoFisher Scientific, P/N 10010023) by gel filtration chromatography using Sephadex™ G25 Fine resin (GE Healthcare, P/N 17-0032-01) packed in a XK-26 column to a bed height of 30 cm (160 ml column volume). Alternatively, the product was buffer exchanged and concentrated by discontinuous diafiltration using Amicon Ultra 15 centrifugal filters (Millipore, P/N: UFC903024).

SDS-PAGE

NuPAGE 4-12% Bis-Tris gels were run with MES running buffer. Wells were loaded with 2-3 μg of sample for both non-reduced and reduced preparations. Reduced samples were treated with 20 mM DTT and heated at 70° C. for 10 minutes. Gel power supply settings were constant 200V for 50 minutes. Protein bands were visualized by staining with Novex SimplyBlue SafeStain according to manufacturer's instructions. Gels were scanned on an Azure Biosystems C600 imager. The below materials/reagents (vendor; catalog number) were used:

- XCell SureLock Mini-Cell (Thermo Fisher; EI0001)
- PowerEase™ 300 W Power Supply (Thermo Fisher; PS0301)
- Isotemp Digital Heat Block (Fisher Scientific; 88-860-022)
- NuPAGE MES SDS Running Buffer (20×) (Thermo Fisher; NP0002)
- SimplyBlue SafeStain (Thermo Fisher; LC6065)
- NuPAGE LDS Sample Buffer (4×) (Thermo Fisher; NP0007)
- Novex Sharp Prestained Protein Standard (Thermo Fisher; LC5800)
- 2-MERCAPTOETHANOL (MP Biomedical; 194705)
- Azure c600 imager (Azure Biosystems)

Analytical Size Exclusion Chromatography (Superdex 200)

Analytical size exclusion chromatography was performed with a Superdex 200 Increase 10/300 column installed on an Agilent 1260 HPLC system. Mobile phase was PBS pH 7.4 (ThermoFisher Scientific, P/N 10010023). Flow rate was 0.5 ml/min with run time of 45 minutes. Absorbance was measured at 214 nm and 280 nm. Peaks were integrated by the Open Lab CDS Data Analysis Software Version 2.3 using the GC/LC Area Percent Method. Peaks below the threshold of detection were integrated by manually defining the baseline.

Intact Mass

10 μg of AB0089 was diluted to 0.25 μg/μL with 0.1% formic acid and 1 μg was injected onto a MabPac 2.1×100 mM column and analyzed using a Thermo Vanquish UHPLC coupled to a QExactive+ mass spectrometer with the Biopharma option. Spectra were acquired in high-mass range mode at 17,500 resolution. Data was analyzed using the Sliding Window ReSpect algorithm in the BioPharma Finder data analysis software package. The below materials/reagents (vendor; catalog number) were used: 0.1% formic acid in water (Honeywell; LC452-2.5) and MAbPac RP 2.1×100 mm, 4 μm column (Thermo Scientific; 088647).

Endotoxin

Endotoxin readings were performed on an Endosafe® nexgen MCS cartridge reader (Charles River Laboratories) and cartridges (Charles River Laboratories P/N PTS2001F). Samples were prepared for assay by diluting in either endotoxin-free water or sterile PBS pH 7.4 (ThermoFisher Scientific, P/N 10010023). An assay was considered valid if all assay performance parameters and controls are within specification.

Hydrophobic Interaction Chromatography (HIC)

Injections of AB0089 (5 μg) were prepared in a 5:4 ratio of high salt buffer (100 mM sodium phosphate, 1.8 M ammonium sulfate, pH 6.5) to sample. Samples were analyzed using an Agilent 1260 Infinity II HPLC equipped with a Sepax Proteomix HIC Butyl-NP5 5 uM column held at 25° C. The gradient runs from 0% low salt buffer (100 mM sodium phosphate, pH 6.5) to 100% low salt buffer over 6.5 minutes at a flow rate of 1.0 mL/minute. Chromatograms were monitored at 280 nm. The below materials/reagents (vendor; catalog number) were used:

Proteomix HIC Butyl-NP5 5 uM Column, 4.6×35 mm (Sepax; 431NP5-4603)

Sodium Phosphate Monobasic (Sigma Aldrich; RES20906-A702X)

Sodium Phosphate Dibasic (Fisher Bioreagents; BB332-500)

Ammonium Sulfate, >99.0% (Sigma; A4418-500G)

Sodium Hydroxide, 10.0 N (Avantor Materials; 5000-03)

Hydrochloric Acid, 10.0 N (Ricca Chemical Company; 3770-32)

Non-Reduced CE-SDS

AB0089 was diluted to 0.5 mg/mL with 1× sample buffer to achieve a final sample volume of 50 μL. Internal standard and iodoacetamide were added to samples, which were then incubated at 70° C. for 10 minutes in a heat block, followed by an ice bath for 5 minutes. Samples were transferred to a 96-well plate, centrifuged, and loaded into the instrument. Individual samples were loaded into the capillary for 40 seconds at 4600 volts and separated for 35 minutes at 5750 volts on a Maurice instrument (ProteinSimple, San Jose, Calif.). The below materials/reagents (vendor; catalog number) were used:

25× Internal Standard (ProteinSimple; 046-016)

CE-SDS Plus Cartridge (ProteinSimple; 090-157)

Iodoacetamide (Sigma-Aldrich; 16125-5G)

Maurice CE-SDS PLUS Application Kit Reagents (Bottom Running Buffer, Separation Matrix, Wash Solution, Conditioning Solution 1 & 2, 1× Sample Buffer) (ProteinSimple; 046-571)

Top Running Buffer (ProteinSimple; 046-384)

Reduced CE-SDS

AB0089 was diluted to 0.5 mg/mL with 1× sample buffer to achieve a final sample volume of 50 μL. Internal standard and beta-mercaptoethanol (Bio-Rad, #161-0710) were added to samples, which were then incubated at 70° C. for 10 minutes in a heat block, followed by an ice bath for 5 minutes. Samples were transferred to a 96-well plate, centrifuged, and loaded into the instrument. Individual samples were loaded into the capillary for 40 seconds at 4600 volts and separated for 31.2 minutes at 5750 volts on a Maurice instrument (ProteinSimple, San Jose, Calif.).

Capillary Isoelectric Focusing (cIEF)

AB0089 was diluted to 1 mg/mL with MilliQ water, 15 μL of sample was added to 60 μL of master mix (water, methyl cellulose, Pharmalyte 3-10, arginine, pI markers 4.05 and 9.99), vortexed, and centrifuged briefly. 60 μL of sample was aspirated from the top of the solution and added to a 96-well plate and centrifuged before testing. The sample was separated for one minute at 1500 volts followed by 8 minutes at 3000 volts on a Maurice instrument (ProteinSimple, San Jose, Calif.). The below materials/reagents (vendor; catalog number) were used:

0.5% Methyl Cellulose (ProteinSimple; 102730)

1% Methyl Cellulose (ProteinSimple; 101876)

500 mM arginine (ProteinSimple; 042-691)

Anolyte (ProteinSimple; 102727)

Catholyte (ProteinSimple; 102728)

cIEF Cartridge (ProteinSimple; 090-101)

Fluorescence calibration standard (ProteinSimple; 046-025)

Pharmalytes 3-10 (Sigma Aldrich; P1522-25ML)

pI Marker 4.05 (ProteinSimple; 046-029)

pI Marker 9.99 (ProteinSimple; 046-034)

System suitability mix (ProteinSimple; 046-027)

System suitability peptide panel (ProteinSimple; 046-026)

Differential Scanning Calorimetry (DSC)

AB0089 was prepared at 0.5 mg/mL in 1× PBS (Gibco, #10010-031) or alternative formulations. 325 μL was added to a 96-well deep well plate along with a matching buffer blank. Thermograms were generated using a MicroCal PEAQ DSC (Malvern, Pa.). Temperature was ramped from 20-100° C. at 60° C./hour. Raw thermograms were background subtracted, the baseline model was spline, and data were fitted using a non-two state model.

SPR Instrument and Reagents

Biacore 8K (Instrument ID: 2222251, GE Healthcare Life Sciences) was used to run the kinetic measurements and Biacore 8K Insight Evaluation Software was used for analysis of experimental data. The below materials/reagents (vendor; catalog number) were used:

10 mM sodium acetate, pH 5.0 (Cytiva; BR100351)

10 mM Glycine, pH 1.7 (Boston BioProducts; BB-2413D)

10× HBS-EP+ (Cytiva; BR1006-69)

1× HBS-EP+ Prepared in house from 10× stock solution

1× MBS-EP+, pH6.0 (Prepared in house)

BSA Fraction V (7.5%) (Gibco; 15260-037)

NaCl (Fisher Scientific; S271-1)

NaOH (Cytiva; BR100358)

Amine coupling kit (Cytiva; BR-1000-50)

Mouse antibody capture kit (Cytiva; BR100838)

CM5 Chip (Cytiva; 29-1496-03)

Series S Sensor Chip SA (Cytiva; BR-1005-31)

SA Chip (Cytiva; BR100531)

Biotin CAPture kit (Cytiva; 28-9202-34)

Goat anti-human IgG Fc cross-absorbed secondary antibody (Invitrogen; 31125)

Micro Bio-Spin 30 columns (Bio-Rad; 732-6223)

Amicon Ultra 0.5 mL Centrifugal Filters 10,000 NMWL (Millipore Sigma; UFC501096)

CLEC12A Affinity by SPR

Binding affinities of AB0089 to recombinant human CLEC12A (Dragonfly Tx) were measured by SPR using a Biacore 8K instrument. Briefly, human Fc specific antibodies were covalently immobilized at a density ˜8000-10000 resonance units (RU) on carboxy methyl dextran matrix of a CM5 biosensor chip via amine-coupling chemistry to create an anti-hFc IgG chip. AB0089 samples were captured on the anti-hFc IgG chip at a concentration of 1.5 μg/mL at a flow rate of 10 μL/min for 60 seconds to achieve approximately 150 RU capture level. hCLEC12A-His was serially diluted (100 nM-0.14 nM) in three-fold dilutions SPR running buffer and injected at a flow rate of 30 μl/min over the captured test articles. Association was monitored for 300 seconds and dissociation was monitored for 600 seconds. Surfaces were regenerated between cycles with three pulses of 10 mM glycine-HCl, pH 1.7 injected for 20 seconds at 100 μl/min. HBS-EP+ (1×) with 0.1 mg/ml BSA was used as running buffer throughout the experiment. Experiments were performed at physiological temperature of 37° C. To obtain kinetic rate constants and overall binding affinity, double-referenced data were fit to a two-state kinetic model using Biacore 8K Insight Evaluation software (Cytiva). Justification for the use of the two-state model is provided below.The goodness of the fit between the fitted curve and the experimental data were expressed by the evaluation software as Chi².

NKG2D Affinity by SPR

Binding affinity of AB0089 for recombinant human NKG2D (Dragonfly Tx) was measured by SPR using a Biacore 8K instrument. Briefly, mFc specific antibodies were covalently immobilized at a density of ˜8000-10000 RU on carboxy methyl dextran matrix of a CM5 biosensor chip via amine coupling chemistry to create an anti-mFc IgG chip. mFc-tagged human NKG2D was captured on the anti-mFc IgG chip at a concentration of 0.2 μg/mL at a flow rate 10 μL/min for 60 seconds to achieve approximately 30 RU capture level. AB0089 was buffer exchanged into HBS-EP+ (1×) buffer and serially diluted (5000 nM-9.77 nM) in two-fold dilutions with HBS-EP+. Analyte (AB0089) was injected at a flow rate of 30 μl/min over the captured test articles. Association was monitored for 60 seconds and dissociation was monitored for 60 seconds. Surfaces were regenerated between cycles with three pulses of 10 mM glycine-HCl, pH 1.7 injected for 20 seconds at 100 μl/min. HBS-EP+ (1×) was used as running buffer throughout the experiment. Experiments were performed at physiological temperature of 37° C. Double-referenced data were fit to a 1:1 kinetic and a Steady State interaction models using Biacore 8K Insight Evaluation software (GE Healthcare) with global fit analysis in the 1:1 kinetic fit model. Since the samples had been buffer exchanged into the running buffer prior to each experiment, offset for a steady state fit was set to constant (0). The goodness of the fit was expressed by the evaluation software as Chi²that was considered in the context of R_maxvalue.

Synergy of Simultaneous Binding of CD16a and NKG2D

Synergistic NKG2D and CD16a binding was evaluated by SPR. hNKG2D alone (12.3 μg/mL), CD16a F158 allele alone (9 μg/mL) and the mixture of hNKG2D and CD16a F158 at the same concentration each were amine-coupled to the surface of the CM5 Series S Biacore chip for 1 minute. 1.8 μM AB0089 or trastuzumab were injected for 150 sec at 10 μL/min. Dissociation phase was observed for 180 sec when regeneration was not needed and 1800 sec when natural regeneration of the surface (almost complete dissociation of analyte) was needed between the cycles at the same flow rate. 1× HBS-EP+ buffer was used as running and sample dilution buffer.

Co-Engagement (CLEC12A and NKG2D)

AB0089 was diluted in 1× HBS-EP+ buffer containing 0.1 mg/ml BSA and was captured on an anti-human Fc surface of CM5 chip at a flow rate 5 μL/min for 60 sec to achieve capture level of 150-250 RU. The net difference between baseline signal and the signal after completion of AB0089 injection representing the amount of AB0089 captured was recorded. hCLEC12A-His (200 nM) or mFc-hNKG2D (7 μM) was injected over captured AB0089 at 20 μl/min for 90 sec to reach saturation. This injection was immediately followed by an injection of pre-incubated mixture of hCLEC12A-His (200 nM) and mFc-hNKG2D (7 μM) at a flow rate of 20 μl/min for 90 sec with the use of the A-B-A injection command in the Biacore 8K control software (second target was pre-mixed with the first target to assure all binding sites for the first target are occupied). The chip was regenerated by two 20 sec pulses of 10 mM Glycine (pH 1.7) at 100 μL/min. The experiment was conducted at 37° C. and 1× HBS-EP+ buffer containing 0.1 mg/ml BSA was used as a running buffer. Binding of each antigen, expressed in RU, was recorded as the net difference between the baseline signal prior and after the injection of individual antigen. An average relative binding ratio of each target bound to AB0089 unoccupied with another target (injected first) was assigned a value of 1.0. An average relative binding stoichiometry of each target bound to captured AB0089 that is already saturated with the other target (injected second) was expressed as a fraction of the full capacity binding to unoccupied AB0089.

Epitope Binning

Anti-CLEC12A antibody or TriNKET was directly immobilized to the active flow cell of a Series S CM5 sensor chip via amino coupling at 3 μg/mL concentration for a final density of 370-1070 RU. A classic sandwich assay run at 25° C. was used for epitope binning. The first injection was a 120 second association of 200 nM CLEC12A-His followed directly by 200 nM of the sandwiching antibody/TriNKET. The sandwiching antibody/TriNKET had a 120 seconds association time and 120 seconds dissociation time. Both analyte and sandwiching molecule injections were run at 30 μL/min. A 20 second pulse of 10 mM glycine pH 1.7 injected at 100 μL/min was used to regenerate the chip surface after dissociation. The first cycle included buffer injections for both the analyte and sandwiching molecule. The second cycle included 200 nM hCLEC12A-His analyte followed by a buffer injection for sandwiching molecule. Subsequent cycles included 200 nM hCLEC12A-His analyte injection followed by 200 nM sandwiching molecule. 1× HBS-EP+ running buffer was used throughout experiment. Sensorgrams were visualized using Biacore Insight Evaluation software.

FcγR Binding

Human and cynomolgus CD64, high and low affinity alleles of CD16a (V158 and F158, respectively), two alleles of CD32a (H131 and R131), CD16b, CD32b and cynomolgus CD16 were captured via biotin on a primed (according to the manufacturers instruction) Biacore chip. Table 123 summarizes respective capture levels and types of the Biacore chips used. Due to irreversibility of streptavidin-biotin interaction, the capture step for all receptors except FcγRI (CD64) was performed just once prior to the start of the binding experiment. hCD64 and cyno CD64 capture on the CAP chip surface was performed during each cycle in accordance with the manufacturer's instructions for the Biotin CAPture kit.

Prior to any FcγR binding experiment requiring a high starting concentration, AB0089 and trastuzumab, samples were buffer exchanged into 1× HBS-EP+ SPR running buffer by three consecutive dilution/concentration cycles with the use of a 0.5 mL concentrator unit resulting in >300-fold dilution of original buffer components while maintaining high concentration of protein. Protein concentration was measured at A280 on a NanoDrop instrument and with the use of sample appropriate theoretical molar extinction coefficients (196790 M⁻¹cm⁻¹for AB0089 and standard IgG setting for trastuzumab).

Each FcγR binding experiment consisted of multiple cycles utilizing 2-fold serially diluted AB0089 or trastuzumab. 0 nM blank cycle placed at the start of each concentration series and receptor free chip surface were used for referencing. Each experimental cycle included association, dissociation and regeneration steps. Analyte (AB0089 or trastuzumab) concentrations, association and dissociation times, and regeneration parameters were FcγR-specific (Table 123). All steps were executed at a flow rate of 30 μL/min and at 25° C.

TABLE 123

Parameters used in SPR FcγR binding experiments.

hCD64,
hCD16a
hCD16a
cyno
hCD32a H131,

Parameter
cyno CD64
V158
F158
CD16
R131
hCD16b
hCD32b

Type of chip
CAP chip
Type S SA chip

FcγR capture level, RU
35-50
10-27
10-27
7-25
10-23

Concentration range, nM
400-1.56
1500-0.023
6000-0.023
16000-125
24000-188

Association time, sec
240
150
60
60

Dissociation time, sec
600
300
60
150

Surface regeneration
CAPture kit regeneration
5 sec pulse of 1-2 mM NaOH

A series of 2-5 blank experimental cycles with the use of experiment appropriate running buffer in place of analyte was performed prior to sample analysis to stabilize the chip surface. Raw data was fitted to either a 1:1 kinetic model or a steady state affinity model using Biacore Insight evaluation software. Chi²(calculated by the software) related to either calculated R_maxor apparent maximal binding response was used to assess goodness of the fit. All reported values were rounded to one decimal place after the calculation.

FcRn Binding

Recombinant human or cynomolgus monkey (cyno) FcRn was captured on a surface of an SA series S chip at a concentration of 0.25 μg/mL for 1 minute. AB0089 and commercial trastuzumab (IgG1 assay control) were buffer exchanged into MBS-EP+, pH 6.0 prior to the SPR binding experiment. Each FcRn binding experiment consisted of multiple cycles utilizing varying concentrations of analytes. Two-fold dilution series (12 μM-0.023 μM) of AB0089 and trastuzumab were used. 0 nM blank cycle placed at the start of each concentration series and a blank amine coupling modified chip surface lacking FcRn were used for referencing. Each experimental cycle included association, dissociation and regeneration steps. The association step consisted of 60 sec analyte injection and was followed by a 60 sec dissociation step. Surface regeneration (60 sec injection of HBS-EP+ buffer, pH 7.4) completed each kinetic cycle. All steps were executed at a flow rate of 30 μL/min and at 25° C. MBS-EP+ buffer, pH 6.0 was used as a running buffer. The data were fitted to a steady state affinity model with Biacore Insight evaluation software. Chi²(calculated by the software) related to the calculated R_maxwas used to assess goodness of the fit. All reported values were rounded to one decimal place.

KHYG-1-CD16aV Mediated Cytotoxicity Assay

The potency of AB0089 was tested with KHYG-1-CD16aV cells. The EC50 and maximum lysis values were derived from the cell lysis curves using GraphPad Prism software (four parameter logistic non-linear regression curve fitting model).

Primary NK Cells Mediated Cytotoxicity Assay

The potency of AB0089 was tested with primary NK cells. The EC50 and maximum lysis values were derived from the cell lysis curves using GraphPad Prism software (four parameter logistic non-linear regression curve fitting model).

Non-Specific Binding to Polyspecific Reagent (PSR)

50 μL of 100 nM TriNKET or control mAb in PBSF were incubated with pre-washed 5 μL protein A dyna beads slurry for 30 mins at room temperature. TriNKETs or mAb bound magnetic beads were allowed to stand on a magnetic rack for 60 seconds and the supernatant was discarded. The bound beads were washed with 100 μL PBSF. Beads were incubated for 20 minutes on ice with 50 μL of biotinylated PSR reagent which was diluted 25-fold from the stock (Xu et al., (2013) Protein engineering design and selection, 26, 663-670). Samples were put on the magnetic rack, supernatant discarded, and washed with 100 μL of PBSF. A secondary FACS reagent, to detect binding of biotinylated PSR reagent to TriNKETs or control mAbs, was made as follows: 1:250 μL of Streptavidin-PE and 1:100 donkey anti-human Fc were combined in PBSF. To each sample, 100 μL of the secondary reagents were added and allowed to incubate for 20 minutes on ice. The beads were washed twice with 100 μL PBSF. Samples were analyzed on a BDFACS Celesta. Two PSR controls, Rituximab (PSR positive) and Trastuzumab (PSR negative), were used in the assay. The below materials/reagents (vendor; catalog number) were used: PBSF (1× PBS pH 7.4+0.25% BSA) (Dragonfly Tx), Dyna beads (Invitrogen; 10001D), Streptavidin-PE (Biolegend; 405204), and Donkey anti-human Fc Ab (Jackson ImmunoResearch; 709-096-149).

High-Spec® Cross Reactivity Assay Using HuProt™ Arrays

The specificity of AB0089 was tested at 0.1 μg/ml and 1 μg/ml concentration against native HuProt human proteome array immobilized on microslides. The assay was performed at CDI laboratories (Baltimore, Md.).

Immunogenicity Assessment

EpiMatrix program from EpiVax was used for all calculations as described in Cohen et al. (2013) Mol Cancer Ther, 12, 2748-59.

Molecular Modeling

Molecular modelling was performed using the SAbPred website (opig.stats.ox.ac.uk/webapps/newsabdab/sabpred/). CLEC12A and NKG2D binding arms were compared with 377 post Phase I biotherapeutic molecules using Therapeutic Antibody Profiler (TAP). TAP used ABodyBuilder to generate a model for AB0089 with side chains by PEARS. The CDRH3 was built by MODELLER, since this loop is the most diverse of all CDRs.

Five different parameters were evaluated:

- 1. Total CDR length
- 2. Patches of surface hydrophobicity (PSH) across the CDR vicinity
- 3. Patches of positive charge (PPC) across the CDR vicinity
- 4. Patches of negative charge (PNC) across the CDR vicinity
- 5. Structural Fv charge symmetry parameter (sFvCSP) for CLEC12A binding scFv and NKG2D binding Fab

Developability

Accelerated (40° C.) Stability in HST, pH 6.0

AB0089 lot AB0089-002 was concentrated and diafiltered into 20 mM histidine, 250 mM sucrose, 0.01% Tween-80, pH 6.0 (HST, pH 6.0) using an Amicon Ultral5 30K device. The concentration of the recovered sample was determined by A280 and the sample was diluted to 20 mg/ml with HST buffer. A 0.2 ml aliquot was immediately frozen at −80° C. as the control and two 0.15 ml aliquots were stored at 40° C. in the dark. An aliquot was removed after 2 and 4 weeks and immediately frozen at −80° C. until analysis. The below materials/reagents were used:

Amicon Ultra15 Centrifugal Filters (30K MWCO) (Millipore; UFC903024)

Hydrochloric acid, 10N (RICCA; 3770-32)

L-Histidine (MP Biomedicals; 101954)

Sodium hydroxide, 10N (JT Baker; 5000-03)

Sucrose (VWR; M117-500G)

Tween-80 (VWR; 0442-1L)

Accelerated (40° C.) Stability in CST, pH 7.0

AB0089 lot AB0089-002 was concentrated and diafiltered into 20 mM sodium citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 using an Amicon Ultral5 30K device. The concentration of the recovered sample was determined by A280 and the sample was diluted to 20 mg/ml with CST buffer. A 0.4 ml aliquot was immediately frozen at −80° C. as the control and four 0.15 ml aliquots were stored at 40° C. in the dark. An aliquot was removed weekly and immediately frozen at −80° C. until analysis. The below materials/reagents were used:

Amicon Ultra15 Centrifugal Filters (30K MWCO) (Millipore; UFC903024)

Hydrochloric acid, 10N (RICCA; 3770-32)

Sodium Citrate Dihydrate (Fisher Chemical; 5466)

Sodium hydroxide, 10N (JT Baker; 5000-03)

Sucrose (VWR; M117-500G)

Tween-80 Solution (10%) (Sigma: PB192-10ML)

Forced Oxidation

AB0089 lot AB0089-002 was diluted to 1 mg/ml with phosphate buffered saline (PBS). 6.7 μL of 3% hydrogen peroxide was added to a 1.0 ml aliquot of AB0089 to obtain a final concentration of 0.02% hydrogen peroxide. Oxidation of AB0089 proceeded for 24 hours at room temperature in the dark. After 24 hours, the sample was buffer switched into PBS, pH 7.4 using Amicon Ultra 0.5 ml 10K MWCO devices following the manufacturer's instructions. The un-oxidized control (0.5 ml) was stored at −80° C. The below materials/reagents were used: 30% hydrogen peroxide (VWR; BDH7690-1), Gibco PBS pH 7.4 (1×) (ThermoFisher; 10010-031), and Zeba Desalting Columns (Thermo Scientifc; 89890).

pH 5 Stress

Concentrated AB0089 lot AB0089-002 was diluted to 1.0 mg/ml with 20 mM sodium acetate, pH 5.0. Concentration was corroborated by A280 with a Nanodrop One spectrophotometer. The sample was split. Half was stored at −80° C. as a control and the other half was incubated at 40° C. for 2 weeks. After the incubation, the sample was stored at −80° C. until analysis. Sodium acetate buffer (1 M, pH 5.0) (Boston Bioproducts; BB-60) was used.

pH 8 Stress

Concentrated AB0089 lot AB0089-002 was diluted to 1.0 mg/ml with 20 mM Tris, pH 8.3. Concentration was corroborated by A280 on the Nanodrop One spectrophotometer. The sample was split. Half was stored at −80° C. as a control and the other half was incubated at 40° C. for 2 weeks. After the incubation, the sample was stored at −80° C. until analysis. The pKa of Tris is temperature dependent. Tris, pH 8.3 measured at 25° C. is expected to have a pH of ˜8.0 at 40° C. Tris buffer (1 M, pH 8.3) (Boston Bioproducts; LT-783) was used.

Freeze/Thaw

Three 0.1 ml aliquots of 20 mg/ml AB0089 lot AB0089-002 in CST, pH 7.0 buffer were subjected to freeze/thaw cycles at −80° C. in a Styrofoam container to slow the freezing process. At cycles 2, 4, and 6, an aliquot was removed and transferred to −80° C. storage until analysis.

Agitation

0.5 ml aliquots of 10 mg/ml AB0089 lot AB0089-002 in 20 mM sodium citrate, 250 mM sucrose, pH 7.0 or 20 mM sodium citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 were added to a deep well plate (Thermo, #278752), the place was sealed, covered with foil and shaken at room temperature at 300 rpm for 3 days. After 3 days the plate was spun down and analyzed by A280 (concentration), A340 (turbidity), and SEC (aggregation).

Low pH Hold

AB0089 was captured on a Mab Select Sure Protein A column and eluted with 100 mM glycine, pH 3.0. Peak fractions were pooled, and 12 ml of the protein A eluate was pH was adjusted to pH 3.3 with 100 mM glycine pH 3.0 buffer. The pH-adjusted eluate was held at room temperature. After 1.5 hours, the eluate was brought to neutral pH with 1.0 M Tris-HCl, pH 8.3. The pH-hold sample was further purified by cation exchange chromatography. The final product was buffer exchanged to PBS pH 7.4 for analysis and received designation AB0089 lot AB0089-003.

High Concentration

AB0089 Lot AB0089-002 was concentrated using 30 kDa molecular weight cutoff devices in PBS, pH 7.4. Concentration was determined by A280 on a Nanodrop and product quality was monitored by SEC as described above.

Size Exclusion Chromatography (TSKgel G3000SWxl)

The following method was used for assessment of developability samples. 50 μg of test material was injected onto an Agilent 1260 Infinity II high pressure liquid chromatography (HPLC) instrument with 1260 Quat Pump, 1260 Vialsampler, 1260 VWD. The sample was separated on a Tosoh TSKgel G3000SWxl, 7.8 mm I.D.×30 cm, 5 μm column. SEC running buffer was PBS, pH 7.0, flowing at 0.50 ml/min. Absorbance was monitored at both 214 and 280 nm, peak areas were manually integrated and the percent of high molecular weight species (HMWS), low molecular weight species (LMWS), and monomer were reported. The below materials/reagents (vendor; catalog number) were used: gel filtration standards (Bio-Rad; 1511901), SEC buffer (20 mM Sodium Phosphate Monobasic Monohydrate/Sodium Phosphate Dibasic Heptahydrate, 0.3 M Sodium Chloride, pH 7.0) (Boston BioProducts; C-5194D), and TSKgel G3000SWxl, 7.8 mm×30 cm, 5 μm column (Tosoh; 08541).

LC-MS/MS Tryptic Peptide Mapping

50 μg of AB0089 was diluted with 6 M guanidine hydrochloride in 250 mM Tris, pH 8.5 and disulfide bonds were reduced with 25 mM dithiothreitol for 30 minutes at 40° C. in a thermomixer. The reduced cysteines were then alkylated with 50 mM iodoacetamide for 60 minutes at room temperature in the dark. The reduced and alkylated protein was buffer-switched into 50 mM Tris, pH 8.0 and digested with trypsin (1:25 enzyme:substrate) for 1 hour at 37° C. LC-MS/MS was carried out on a Thermo Vanquish UHPLC coupled to a QExactive+ mass spectrometer. Briefly, 5 μg of protein digest was separated on a Waters UPLC BEH Peptide C18 column (2.1×150 mm) over 150 minutes using a 2% to 40% acetonitrile gradient at 0.3 ml/min. Mass spectra were acquired from 230-2000 m/z in positive mode at 35,000 resolution for the full MS scan and 17,500 resolution for the Top10 MS/MS scans. The UPLC-MS/MS files were analyzed using the Byos PTM workflow (Protein Metrics). Peptides were identified using accurate mass and MS/MS searching the AB0089 sequence. Settings for peptide identification were as follows:

Precursor Mass Tolerance 6.00 ppm

Fragment Mass Tolerance 20.00 ppm

Cleavage Site(s) RK, C-terminal

Digestion Specificity Fully specific

Missed Cleavages 2

Modifications Carbamidomethyl/+57.021464 @ C, fixed

DTT/+151.994915 @ C

Oxidation/15.994915 @ M, W

Dethiomethyl/−48.003371 @ M

Deamidated/+0.984016 @ N, Q

pyro-Glu/−17.026549 @ NTerm Q

pyro-Glu/−18.010565 @ NTerm E

Carbamyl/+43.005814 @ NTerm, K, R

Acetyl/+42.010565 @ Protein NTerm

Ammonia-loss/−17.026549 @ N

Dioxidation/+31.989829 @ W

Kynurenine/+3.994915 @ W

N-glycan 50 common

The following materials/reagents (vendor; catalog number) were used:

0.1% formic acid in acetonitrile (JT Baker; LC441-2.5)

0.1% formic acid in water (JT Baker; LC452-2.5)

Formic acid (Thermo Scientific; 28905)

Guanidine HCl (Alfa Aesar; A13543)

Hydrochloric acid, 6 N (Avantor; 4103-01)

No-weigh dithiothreitol (Thermo Scientific; 20291)

No-weigh iodoacetamide (Thermo Scientific; 90034)

Sodium hydroxide, 6 N (Avantor; 5672-02)

Tris Base (Avantor; 4109-01)

Trypsin (Thermo Scientific; 90057)

UPLC BEH Peptide C18 column (2.1×150 mm) (Waters; 186003556)

Zeba spin desalting columns (Thermo Scientific; 89882)

Results

Discovery

Hybridoma subclone 16B8.C8 (from which sequence optimized AB0089 and its predecessor AB0237 were derived) was discovered through immunization of BALB/c mice with recombinant human CLEC12A-His protein (FIGS. 104A-104H). A detailed description of the immunization process, antibody screening, assessment of recombinant and cell surface expressed CLEC12A binding, and epitope binning are described.

During the developability assessment of AB0237, an aspartic acid isomerization liability was identified in CDRH3 of the CLEC12A binding scFv. Attempts to remove this liability by single amino acid substitution or by targeted yeast display approaches (randomization of “DS” or “S”) resulted in low expression yield and significant loss of binding to CLEC12A, suggesting that this motif may be important for maintaining CDR architecture. Therefore, an affinity maturation effort focusing on optimization of CDRH3 and CDRH2 was undertaken to further improve affinity for CLEC12A, and then the DS sequence was replaced by DT on the background of the optimized CDRH3. In addition, we sought to improve the humanness of the molecule by reducing the number of murine backmutations in the framework. We hypothesized that only one backmutation in position H83 was required, while other backmutations were not obligatory for maintaining CDRs architecture. The optimized scFv framework that contained a single backmutation was named pET1596 and was used as the background for all additional engineering. Later, the (H94)K backmutation was brought back to provide additional stabilization of the molecule. These engineering efforts resulted in the discovery of the AB0089 molecule, which retained high affinity for human CLEC12A and potency towards AML cancer cell lines, while both removing the isomerization liability and improving humanness.

Sequence Optimization

CDRH3 Focused Randomized Affinity Maturation Library

A yeast display affinity maturation library (referred to as Library 17) was created successfully by mutating CDRH3 residues (YDYDDSLDY, SEQ ID NO:139) of pET1596 at each position one at a time or by changing two amino acid residues to all 20 amino acids as described above. To enrich for scFvs that have higher affinity towards hCLEC12A in library 17, three rounds of selections were carried out with biotinylated CLEC12A-His at 1 nM (1^stand 2^ndround of selection; FIGS. 105A-105B) and 0.1 nM (3^rdround of selection; FIG. 105C). The affinities between the parental clone pET1596 and the library clones were compared and could be accurately discriminated, allowing a quantitative selection in real time (FIGS. 105B-105D). After 3 rounds of FACS sorting to isolate clones, one clone that contained 2 amino acid differences compared to the parental clone (GDYGDSLDY) (SEQ ID NO:322) exhibited higher binding affinity for CLEC12A than the parental clone (FIGS. 105D-105E).

CDRH3 Focused Combinatory Affinity Maturation Library

Outcomes from the CDRH3 focused library demonstrated an improvement in affinity, however further improvement was highly desirable. Thus, Library 30 was created by incorporating the CDRH2 combinatory library (WSGGK (SEQ ID NO:138) to XSGGK (SEQ ID NO:323), WXGGK (SEQ ID NO:324), WSXGK (SEQ ID NO:325), WSGXK (SEQ ID NO:326), WSGGX (SEQ ID NO:327), XXGGK (SEQ ID NO:328), XSXGK (SEQ ID NO:329), XSGXK (SEQ ID NO:330), XSGGX (SEQ ID NO:331), WXXGK (SEQ ID NO:332), WXGXK (SEQ ID NO:333), WXGGX (SEQ ID NO:334), where X=any amino acid) into the optimized CDRH3 backbone. The goal was to engineer and select binders with better affinity than the parental clone (pET1596) or CDRH3 optimized variant. This created a library with a randomized CDRH2 while retaining a fixed CDRH3. Two rounds of FACS sorting were performed on this affinity maturation yeast display library (Library 30) to enrich for high affinity binders (FIGS. 106A-106D).

After 2 rounds of FACS sorting, one clone with changes in CDRH2 from WSGGK (SEQ ID NO:138) to WVGGA (SEQ ID NO:272) on the GDYGDSLDY (SEQ ID NO:322) CDRH3 background had the highest affinity for CLEC12A (FIGS. 107A-107D). This clone showed a significant improvement in CLEC12A affinity compared to parental pET1596 and AB0237.

Remediation of DS Sequence Liability

Once the affinity matured clone was identified, site-directed mutagenesis was used to produce a rational variant (CDRH3: GDYGDSLDY (SEQ ID NO:322) to GDYGDTLDY (SEQ ID NO:273)), resulting in a CLEC12A targeting scFv that does not contain the isomerization liability, while maintaining high affinity binding to the target. The comparison of the CDRs of the optimized AB0089 molecule and AB0237 is shown in Table 124.

TABLE 124

Comparison of CLEC12A binding

CDRs in AB0089 and AB0237.

TriNKETs
CDRH1
CDRH2
CDRH3

AB0237
GFSLTNY
WSGGK
YDYDDSLDY

(SEQ ID
(SEQ ID
(SEQ ID

NO: 137)
NO: 138)
NO: 139)

AB0089
GFSLTNY
WVGGA

G
DYGDTLDY

(SEQ ID
(SEQ ID
(SEQ ID

NO: 137)
NO: 272)
NO: 273)

Sequence differences obtained as a result of affinity maturation are underlined. Isomerization liability correction obtained by site-directed mutagenesis (DS to DT) is shown in green.

Based on the structural modeling, we decided to re-introduce a murine residue in position H94 to provide additional support for the CDR architecture. AB0089 contains only 2 murine residues in position H83 and H94 in the VH of the CLEC12A binding scFv, while there are 5 murine backmutations in AB0237. The rest of the backmutations were reverted to human germline, thus improving the humanness of AB0089 (Table 125).

TABLE 125

Murine backmutations in the frameworks of AB0089 and AB0237.

Chain
AB0089
AB0237

VH

(H71)K

VH
(H83)Q
(H83)Q

VH
(H94)K
(H94)K

VL

(L70)R

VL

(L83)I

Molecular Format and Design

AB0089 is a heterodimeric tri-functional antibody that redirects NK and T cell activity against CLEC12A-expressing cancer cells. AB0089 contains three chains: Chain H, Chain L and Chain S. The schematic representation of the AB0089 molecule is provided in FIG. 108. Chain H is the heavy chain of A49M-I NKG2D-targeting antibody that was selected for having low affinity while being a potent agonist of NKG2D receptors on NK and cytotoxic T cells. The M-I mutation was introduced to eliminate Met102 residue in the CDRH3 that is sensitive to oxidation and therefore may present a potential liability. Chain L is the light chain of the NKG2D-targeting antibody A49M-I. Chain S contains a scFv that binds with high affinity to CLEC12A on AML cancer cells. Both scFv and Fab domains are fused to a human IgG1 Fc. Both chains bear several mutations in the CH3 domain introduced to ensure stable heterodimerization of two Fc monomers. There are seven such mutations in the IgG1 Fc of the AB0089: three mutations in Chain H and four mutations in Chain S. Two of these mutations are involved in the formation of a stabilizing S—S bridge (Y349C and S354C) between Chains H and Chain S (EU nomenclature). The scFv (VH-VL orientation) in Chain S is stabilized by two mutations introducing a S—S bridge ((H44)C and (L100)C, Chothia nomenclature). Chain S also contains a (G₄S)₄non-immunogenic linker between VH and VL of the scFv. The amino acid sequence of the CLEC12A-targeting scFv is based on the sequence of humanized 16B8.C8 antibody, obtained from hybridoma (Green Mountain Antibodies). The amino acid sequence of the NKG2D Fab is based on the fully human antibody A49M-I, obtained using human yeast display technology (Adimab).

Amino Acid Sequence

The amino acid sequences of the three polypeptide chains that make up AB0089 are shown below. CDRs are defined using Chothia nomenclature and are represented in Table 126. The C-terminal K, commonly processed and cleaved in IgG1, is deleted in Chain S and Chain H to prevent heterogeneity.

Chain S: CLEC12A targeting scFv-CH2-CH3 (humanized sequence, CDRs are italicized, back mutations are double underlined, engineered scFv disulfide in lower case letters, and Fc mutations for heterodimerization and the engineered CH3 disulfide are single underlined).

(SEQ ID NO: 276)

QLQLQESGPGLVKPSETLSLTCTVS custom-character

GLHWIRQPPGKcLE

WIGVI custom-character

TDYNPSLKSRVTISVDTSKNQFSLKLSSV custom-character

AADTAV

YYCA custom-character

WGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSD

IQMTQSPSSLSASVGDRVTITC custom-character

WYQQKPGKAPKLL

IY custom-character

GVPSRFSGSGSGTDFTLTISSLQPEDFATYYC custom-character

FGcGTKLEIKGSDKTHTCPPCPAPELLGGPSVFLFPPKPKDT

LMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQY

NSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQ

PREPRVYTLPPCRDELTKNQVSLTCLVKGFYPSDIAVEWESNGQPE

NNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQGNVFSCSVMHEALHN

HYTQKSLSLSPG

Chain H: NKG2D targeting VH-CH1-CH2-CH3 (fully human, CDRs are italicized, Fc mutations for heterodimerization and the engineered CH3 disulfide are underlined).

(SEQ ID NO: 260)

EVQLVESGGGLVKPGGSLRLSCAAS custom-character

SMNWVRQAPGKGLEWVS

SI custom-character

IYYADSVKGRFTISRDNAKNSLYLQMNSLRAEDTAVYYCAR

custom-character

WGQGTLVTVSSASTKGPSVFPLAPSSKSTSGGTAAL

GCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSS

SLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGGPSV

FLEPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKT

KPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISK

AKGQPREPQVCTLPPSRDELTENQVSLTCLVKGFYPSDIAVEWESNGQP

ENNYKTTPPVLDSDGSFFLYSWLTVDKSRWQQGNVFSCSV

MHEALHNHYTQKSLSLSPG

Chain L: NKG2D targeting VL-CL (fully human,

CDRs are italicized).

(SEQ ID NO: 261)

DIQMTQSPSSVSASVGDRVTITC custom-character

WYQQKPGKAPKLLIY

custom-character

GVPSRFSGSGSGTDFTLTISSLQPEDFATYYC custom-character

F

GGGTKVEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQ

WKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEV

THQGLSSPVTKSFNRGEC

TABLE 126

AB0089 CDRs sequences.

Variable
Frame-

AB0089
region
work
CDR1
CDR2
CDR3

Chain
VH
VH4-59
GFSLTNY
WVGGA
GDYGDTLDY

S

(SEQ ID
(SEQ ID
(SEQ ID

NO: 137)
NO: 272)
NO: 273)

Chain
VL
VK1-12
HASQNIN
EASNLHT
QQSHSYPLT

S

FWLS
(SEQ ID
(SEQ ID

(SEQ ID
NO: 141)
NO: 142)

NO: 140)

Chain
VH
VH3-21
GFTFSSY
SSSSY
GAPIGAAAG

H

(SEQ ID
(SEQ ID
WFDP (SEQ

NO: 297)
NO: 298)
ID NO: 97)

Chain
VL
VK1-2
RASQGIS
AASSLQS
QQGVSFPRT

L

SWLA
(SEQ ID
(SEQ ID

(SEQ ID
NO: 77)
NO: 87)

NO: 86)

CDRs are identified by Chothia

Analysis of Potential Sequence Liabilities in CDRs

Chain L (NKG2D binding light chain) has no predicted liabilities. Chain H (NKG2D binding heavy chain) has a DP in CDRH3. This motif is not affected by forced degradation (as in other constructs, data not shown). As such, this sequence was not altered in AB0089. Chain S (CLEC12A binding scFv-Fc chain) has no predicted liabilities.

Description of Heterodimerization Mutations

EU-numbering convention was used to annotate the amino acid residues in the Fc. Chain H contains the following mutations:

- K360E—Heterodimerization mutation in the CH3 domain of the Fc region.
- K409W—Heterodimerization mutation in the CH3 domain of the Fc region.
- Y349C—Inter-subunit disulfide stabilization mutation in the CH3 domain of the Fc region.

The K360E/K409W mutations in the Fc region of Chain H are designed to render heterodimer-favoring interactions with Chain S due to hydrophobic/steric complementarity plus long-range electrostatic interactions (Choi et al., (2013) Mol Cancer Ther, 12, 2748-59; Choi et al., (2015) Mol Immunol, 65, 377-83). The mutation Y349C was introduced to generate an additional inter-chain disulfide bond Y349C-(CH3-ChainH)-S354C(CH3-ChainS) pair (Choi et al., (2015) Mol Immunol, 65, 377-83).

Chain S contains the following mutations:

- G44C—Disulfide stabilization mutation in VH region of the scFv.
- S100C—Disulfide stabilization mutation in VL region of the scFv.
- Q347R—Heterodimerization mutation in the CH3 domain of the Fc region.
- D399V—Heterodimerization mutation in the CH3 domain of the Fc region.
- F405T—Heterodimerization mutation in the CH3 domain of the Fc region.
- S354C—Inter-subunit disulfide stabilization mutation in the CH3 domain of the Fc region.

The mutations G44C and S100C were introduced to stabilize the scFv by an engineered disulfide bond between the VH and the VL. The disulfide stabilization improves structural and thermal stability of scFv. The Q347R/D399V/F405T mutations in the Fc region of Chain S were designed to render heterodimer-favoring interactions with Chain H due to hydrophobic/steric complementarity plus long-range electrostatic interactions (Choi et al, (2013) Mol Cancer Ther, 12, 2748-59). The mutation S354C was introduced to generate an additional inter-CH3 disulfide bond with a Y349C(CH3-ChainH)-S354C(CH3-ChainS) pair (Choi et al, (2015) Mol Immunol, 65, 377-83).

The heterodimerization of the Chain H and Chain S is based on five mutations, so called, EW-RVT, with the interaction pairs K409W(CH3A)-D399V/F405T(CH3B) (dubbed the W-VT pair) and K360E(CH3A)-Q347R(CH3B) (dubbed the E-R pair), which were designed to replace the conserved electrostatic interactions with asymmetric hydrophobic interactions and to add asymmetric long-range electrostatic interactions at the rim of the heterodimeric CH3 interface, respectively (Choi et al., (2013) Mol Cancer Ther, 12, 2748-59). Detailed analysis of the heterodimeric CH3A-CH3B inter-face revealed very few structural perturbations resulting from mutations at the interface compared to the native IgG1 Fc (FIGS. 109A-109C). Backbone and side-chain conformations of the residues adjacent to the interfacial mutations in Fc-EW-RVT are almost the same as those in Fc-WT, except for the side-chain substitutions of the mutated residues (FIG. 109A). The residues of W-VT mutation pairs, i.e., K409W(CH3A)-D399V(CH3B)/F405T(CH3B), were in close proximity to form complementary contacts in mostly buried CH3A-CH3B interfaces (FIG. 109B), suggesting that W-VT pairs strengthen the hydrophobic interface between CH3A-CH3B by forming optimal hydrophobic interactions. The residues of the E-R mutation pair, K360ECH3A-Q347RCH3B, were located sufficiently close at 3.45 Å to form a salt bridge unlike the wild-type K-Q pair (4.61 A) (FIG. 109C), contributing to the preferential formation of the heterodimer by strong electrostatic interactions. Thus, the mutated residues introduced to Fc-EW-RVT form non-covalent interactions in the CH3A-CH3B interface favoring heterodimer formation without significant structural perturbations to the overall CH3 domain structure.

By design the W-VT pair residues (K409W(CH3A)-D399V(CH3B)/F405T(CH3B)) and E-R pair residues (K360E(CH3A)-Q347R(CH3B)) are in the heterodimeric CH3 interface to form asymmetric, complementary hydrophobic interactions and electrostatic interactions at 3.45 Å, respectively, which preferentially favor the heterodimer. On the other hand, the respective homodimer formation was thermodynamically dis-favored due to the steric clash between 409WCH3A and F405CH3A residues as well as the repulsive electrostatic interactions between the 347R(CH3B) and K360(CH3B) residues. The loss of the wild-type electrostatic interaction residues of K409WT-D39WT also discourages homodimer formation. Additional stabilization of heterodimer is brought by the inter-CH3 disulfide-bonded EW-RVT(S—S) variant with Y349C(CH3A) and S354C(CH3B) mutations (FIG. 110). The X-ray structure of Fc-EW-RVT(S—S) has been described at 2.5 Å resolution (Choi et al., (2015) Mol Immunol, 65, 377-83) and was very similar to the known IgG1 structures.

Structural Modeling

In the absence of a crystal structure, structural models of the variable segments of CLEC12A and NKG2D binding arms were generated for comparison. Model building was performed using the SAbPred web site (opig.stats.ox.ac.uk/webapps/newsabdab/sabpred/).

Surface Charge Distribution and Hydrophobic Patch Analysis of CLEC12A Binding Arm of AB0089

FIGS. 111A-111C illustrate a ribbon diagram model of the CLEC12A binding scFv in three different orientations (upper panel) and their corresponding surface charge distribution of the same orientation (lower panel). The charge distribution of the anti-CLEC12A scFv is polarized (“top view”, lower panel), with negatively charged residues populated predominately within CDRH3 and CDRL2. The uneven distribution of electrostatic patches on the paratope is likely to be target-related and reflects the complementarity of charge distribution on its cognate epitope, which may contribute to the high affinity interaction between CLEC12A and the scFv of AB0089.

FIGS. 112A-112B show the analysis of CDR length and surface hydrophobicity patches of the CLEC12A-targeting arm of AB0089. Analyses were performed using therapeutic antibody profiles (TAP): opig.stats.ox.ac.uk/webapps/newsabdab/sabpred/tap . It compares the candidate in reference to 377 late-stage therapeutic antibodies. The length of CDRs for the CLEC12A-binding arm of AB0089 are within the normal range for late stage therapeutic antibodies. Hydrophobic patch analysis of the CDRs of an antibody is predictive of its developability. The theoretical hydrophobic properties of the CLEC12A binding arm of AB0089 are well within the norm of commercialized biotherapeutics and late-stage therapeutic antibody candidates.

FIGS. 113A-113C show the analyses of patches containing positive and negative charges, as well as the symmetry of these charges around the CDR regions. All three charts indicate that the charge distributions are within the norm compared with the population of therapeutic antibodies.

Surface Charge Distribution and Hydrophobic Patch Analysis of NKG2D Binding Arm of AB0089

The NKG2D-binding Fab is shown in ribbon diagrams with three different orientations (FIGS. 114A-114C, upper panel) and the corresponding surface charge distribution in the same orientation are provided (FIGS. 114A-114C, lower panel). The surface charge on the NKG2D binding arm is more evenly distributed than charges on the CLEC12A targeting arm.

FIGS. 115A-115B show the CDR length and patches of surface hydrophobicity analyses of the NKG2D-targeting arm of AB0089. It appears that the CDR length, hydrophobic properties, and positive/negative charge distribution are well within the norm of existing therapeutic antibodies.

FIGS. 116A-116C show the analyses of patches containing positive and negative charges, as well as the symmetry of these charges around the CDR regions. All three charts indicate that the charge distributions are within the norm comparing with the population of therapeutic antibodies.

Framework Assessment

The CLEC12A binding arm of AB0089 was discovered through mouse immunization. The frameworks chosen for CDR grafting are frequently used in clinical stage therapeutic monoclonal antibodies (FIGS. 117A-117D). The NKG2D binding arm is fully human, is based on a framework frequently used in advanced stage mAbs and does not contain any unusual residues or deviations from the germline.

Immunogenicity Prediction

Protein sequences of the three chains of AB0089 were examined for the presence of potentially immunogenic T cell epitopes and to rank the immunogenicity potential using the EpiMatrix algorithm (FIGS. 118A-118B). AB0089 scores favorably on the Treg adjusted Immunogenicity Protein Scale. The Treg adjusted EpiMatrix Protein Score for the sequences of the three chains of AB0089 are −33.39 (chain H), −27.13 (chain S), and −23.49 (chain L) (Table 127). The predicted risk of immunogenicity for the AB0089 appears to be low.

TABLE 127

EpiVax immunogenicity analysis of AB0089.

EpiMatrix
EpiMatrix
Treg adjusted

Protein Sequence
Length
Hits
Score
Epx Score

AB0089_CHAIN_H
451
207
5.99
−33.39

AB0089_CHAIN_L
214
98
11.17
−23.49

AB0089_CHAIN_S
472
202
−2.96
−27.13

Expression

Expression of AB0089 was achieved using Dragonfly Tx F3′ TriNKET platform purification method. Briefly, AB0089 was expressed by transient transfection of Expi293 and ExpiCHO cells with three plasmids encoding the NKG2D targeting light chain (chain L), the NKG2D targeting IgG heavy chain (chain H), and the CLEC12A targeting scFv-Fc fusion (chain S). For both Expi293 and ExpiCHO expression, Chain S:Chain H:Chain L were transfected at the plasmid ratio 4:1:1 to maximize the percentage of desired heterodimer product. This ratio was empirically obtained based on previous humanized TriNKETs and was not specifically optimized for AB0089.

Purification

Purification of AB0089 was performed using F3′ TriNKET platform purification methods developed by Dragonfly Therapeutics (FIG. 119). This purification process has been applied before to material expressed in Expi293, ExpiCHO and CHO-M cell cultures for many different TriNKETs. No optimization or extensive process development for AB0089 purification was performed. Two-step purification process was employed.

Two lots of AB0089 were produced in transiently transfected ExpiCHO and Expi293 cells. Table 128 summarizes the titer and final product purity for both lots. Purification of the AB0089 lot AB0089-002 expressed in ExpiCHO is described in detail below.

TABLE 128

Summary of AB0089 purification.

Lot
Expression host
Titer mg/L
Purity, % Monomer

AB0089-001
Expi293
16.8
>99

AB0089-002
ExpiCHO
68.3
>99

Protein A Capture

AB0089 was expressed in ExpiCHO cell line as described above. Approximately 5.7 liters of filtered supernatant was loaded on a Protein A column. Load flow rate was 10 ml/min (298 cm/h; 3.3 minute residence time). The capture step yield was approximately 752 mg. FIG. 120 shows the capture step chromatogram.

By SEC analysis, the Protein A eluate contains mainly desired heterodimeric AB0089 species (approximately 88.8%) eluting at 23.4 min (FIG. 121). Impurities include UMW aggregates and very little amount of LMW.

SDS-PAGE analysis of the Protein A load, flow-though and eluate demonstrate that capture of AB0089 was complete, based on absence of a AB0089 band in the flow through sample. The protein A eluate lane shows primarily TriNKET with additional bands corresponding to the HMW and LMW peaks observed by SEC (FIG. 122).

Cation Exchange Chromatography

The Protein A eluate was further purified by cation exchange chromatography using a Poros XS column (1.2 cm D×20 cm H) operated at a flow rate of 7 ml/min (371 cm/h; 3.2-minute residence time). The product began to elute with the 15% B elution step, when conductivity reached approximately 20 mS/cm (FIG. 123). SDS-PAGE analysis of the elution fractions shows highly pure product in the first five fractions (FIG. 124). Peak tail fractions were excluded from the IEX product pool as SDS-PAGE indicates the presence of contaminating product-related proteins.

The IEX fractions were concentrated resulting in approximately 401 mg of purified AB0089. The pooled fractions were buffer exchanged to PBS pH 7.4 by dialysis. The AB0089 yield after buffer exchange was approximately 391 mg at a final concentration of 9.66 mg/ml (AB0089 lot AB0089-002). Yields and recoveries for AB0089 lot AB0089-002 production are summarized in Table 129.

TABLE 129

Yields and recoveries of AB0089 purification and buffer exchange process.

Total
Step
Total

Step
Protein (mg)
Recovery (%)
Recovery (%)

MabSelect SuRe Capture
752
NA
NA

POROS XS Polish
401
53
53

Post Buffer Exchange
391
98
52

NA, not applicable

Characterization of AB0089 Lot AB0089-002 Used in Cyno Study

AB0089 lot AB0089-002 intended for use in cyno study was extensively characterized for purity, identity, binding to the targets and potency. All tested criteria met specification.

Purity by SDS-PAGE

SDS-PAGE of the final sample showed major bands were consistent with expected molecular weight for both reduced and non-reduced AB0089 (FIG. 125).

Purity by Analytical SEC (Superdex 200)

SEC analysis of purified AB0089 lot AB0089-002 showed 100% monomer as determined by integrated peak area (FIG. 126).

Identity by Intact Mass Analysis

The identity of AB0089 (lot AB0089-002) was corroborated by mass spectrometry (FIG. 127). LC-MS analysis of the intact AB0089 showed an observed mass (126,130.0 Da) that agrees well with the theoretical mass of 126,129.1 Da. The major observed glycosylation pattern (G0F/G0F) is typical for an Fc-containing molecule expressed in CHO cells.

Endotoxin

The endotoxin level of the final AB0089 lot AB0089-002 was 0.259 EU/mg. All assay parameters met specification.

Biochemical and Biophysical Properties

AB0089 was characterized by SEC, CE, cIEF, HIC, and DSC to corroborate that is has the proper purity and pre-developability characteristics of a good biological therapeutic.

Purity

Purify of AB0089 was determined by CE-SDS (FIGS. 128A-128B). Under non-reduced conditions the major impurities are method induced (free light chain and TriNKET—LC). Under reduced conditions the 3 expected chains (LC, HC, and scFv-Fc chain) are observed. The purity of AB0089 lot AB0089-002 determined by SEC, NR and R CE-SDS is summarized in Table 130.

TABLE 130

Purity of AB0089 by SEC and CE-SDS analysis.

NR CE-
R CE-

Test
SEC
SDS
SDS

article
HMW (%)
Monomer (%)
LMW (%)
Purity (%)
Purity (%)

AB0089
ND
100
ND
98.9
98.0

ND, none detected

Experimental pI and Charge Profile by cIEF

Charge profile of AB0089 was analyzed by cIEF (FIG. 129). AB0089 showed a major peak at a pI of 8.52±0.0. Several lesser abundant, overlapping acidic peaks and minor basic peaks are also observed.

Hydrophobicity Assessed by HIC.

The hydrophobic properties of AB0089 were assessed by HIC (FIG. 130, Table 131). The chromatogram revealed a single narrow peak, indicating that the molecule is highly pure and homogeneous. The HIC retention time was 10.2 minutes, which is within the range of well behaving marketed monoclonal antibodies (8.9-13.5 minutes, data not shown).

TABLE 131

Hydrophobicity assessment of AB0089 by HIC. Humira is used as an

internal control of well-behaved IgG1 mAb.

Test article
HIC retention time (min)

AB0089
10.2

Humira
8.9

Thermal Stability by DSC

The thermal stability of AB0089 was assessed by DSC in several buffers (PBS pH 7.4, HST pH 6.0, CST pH 7.0). AB0089 demonstrated high thermal stability in all buffers tested (FIGS. 131A-131C). T_m1 corresponding to the unfolding of scFv satisfied the ≥65° C. criteria utilized in thermostability assessment (Table 132).

TABLE 132

Thermal stability of AB0089 by DSC.

Test Article
Buffer
T_onset(° C.)
T_m1(° C.)
T_m2(° C.)
T_m3(° C.)
T_m4(° C.)

AB0089
PBS, pH 7.4
64.0
71.8
76.7
81.3
83.4

AB0089
HST, pH 6.0
61.2
69.0
78.7
84.3
86.3

AB0089
CST, pH 7.0
64.7
72.4
77.1
82.2
84.2

Disulfide Bond Assignment

AB0089 is an engineered molecule based on the backbone of a monoclonal IgG1 antibody. While a typical IgG1 contains 16 disulfide bonds, AB0089's F3′ format has only 15 disulfide bonds. Half of AB0089 is an anti-NKG2D half antibody, as such it contains the expected seven disulfide bonds (one in each IgG domain and the intermolecular disulfide connecting the LC to the HC). The other half of AB0089 is an anti-CLEC12A scFv-Fc fusion. The scFv contains three disulfide bonds (one in each anti-CLEC12A VH and VL domains and an engineered, stabilizing disulfide bridging the variable domains) and the Fc contains the two expected disulfides in the IgG domains. The heterodimer is covalently associated with two native IgG1 hinge disulfides and one additional intermolecular engineered disulfide connecting the Fab-Fc to the scFv-Fc in the CH3 domain. Together this results in 15 expected disulfide bonds in AB0089. The predicted disulfide map for AB0089 is shown in FIG. 132.

The disulfide bond connectivity of AB0089 was corroborated by LC-MS/MS peptide mapping analysis of a non-reduced digest. Disulfide bonded peptides were identified by MS/MS database searching and corroborated by comparing their intensities in the native and reduced digests. In the tryptic digestion, the VL and engineered disulfide pairs are connected by a long peptide that contains 2 cysteines (three peptides connected by two disulfide bonds, S7:S36:S22) that is not observed in the LC-MSMS data. To detect the engineered scFv stabilizing disulfide, AB0089 was digested with both trypsin and chymotrypsin. FIGS. 133A-133B show the extracted ion chromatogram (XICs) for the engineered disulfide pair in the scFv (non-reduced and reduced) and the most intense charge state for that peptide pair. The VL disulfide pair was not observed in the trypsin/chymotrypsin double digest, presumably because it is too small and hydrophilic to be retained on the column. All other disulfides were identified after digestion with only trypsin, including the engineered S—S bridge introduced to stabilize the Fc heterodimerization (FIGS. 134A-134B).

A summary of the observed disulfide linked peptides in AB0089 is shown in Table 133. All observed disulfide linked peptides were had high mass accuracy (<4 ppm), were reducible, and were sequence corroborated by MS/MS fragmentation.

TABLE 133

Disulfide linked peptides theoretical and experimental masses.

Mass

Tryptic
Theoretical
Experimental
accuracy

Domain
peptide
mass (Da)
mass (Da)
(ppm)

VL
L2:L7
4482.0784
4482.0786
0.4

CL
L12:L19
3555.749
3555.755
1.8

CL-CH1
L20-21:H19
1260.4863
1260.4861
−0.3

intermolecular

VH
H3:H10
3371.4686
3371.4666
2.5

CH1
H13:H14-15
7916.9194
7916.9282
1.5

Hinge
H20-21:S17-18
5454.7834
5454.7905
1.5

CH2
H23:H30
2328.0977
2328.1018
1.8

CH3-CH3
H36:S34-35
2757.383
2757.3855
1.6

intermolecular^a

CH3
H37:H41
4432.0675
4432.0687
0.3

scFv VH
S1-2:S8
5828.907
5828.9052
−0.1

scFv stabilizing^a
S7:S36^b
1011.4154
1011.4188
3.4

scFv VL
S10-11:S14
Not detected

CH2
S20:S27
non-unique, see H23:H30

CH3
S36:S41
3844.8236
3844.8284
1.2

^aEngineered disulfide, ^bTrypsin/chymotrypsin double digest

Binding Characteristics

Binding Affinity for Human CLEC12A

The affinity of AB0089 for human CLEC12A was determined by SPR at 37° C. AB0089 binds with high affinity to human CLEC12A (FIGS. 135A-135D and Table 134). However, the binding demonstrates heterogeneity and does not follow typical 1:1 binding kinetics. An experiment described below and in FIG. 136 established a two-state binding mechanism and justified the use of a two-state model for fitting the data. Similar two-state reaction kinetics was observed for CLEC12A binding arm of tepoditamab (Merus).

TABLE 134

Kinetic parameters and binding affinity of AB0089 for hCLEC12A.

Test

article
k_a1(M⁻¹s⁻¹)
k_d1(s⁻¹)
k_a2(s⁻¹)
k_d2(s⁻¹)
K_D(nM)

AB0089
1.09 × 10⁷
4.26 × 10⁻²
3.41 × 10⁻³
7.28 × 10⁻⁴
0.69

AB0089
1.06 × 10⁷
4.49 × 10⁻²
3.49 × 10⁻³
7.48 × 10⁻⁴
0.75

AB0089
1.02 × 10⁷
4.32 × 10⁻²
3.45 × 10⁻³
7.41 × 10⁻⁴
0.75

AB0089
9.99 × 10⁶
4.32 × 10⁻²
3.39 × 10⁻³
7.49 × 10⁻⁴
0.78

Average ±
(1.04 ± 0.03) × 10⁷
(4.35 ± 0.12) × 10⁻²
(3.43 ± 0.04) × 10⁻³
(7.41 ± 0.10) × 10⁻⁴
0.74 ± 0.04

Std Dev.

A two-state binding model assumes two binding stages that are defined by two sets of kinetic constants:

1. formation of an initial CLEC12A-AB0089 complex

2. transition of the initial complex into a more stable, final complex

$A + B \to_{k_{d 1}}^{k_{a 1}} AB \to_{k_{d 2}}^{k_{a 2}} {AB}^{*}$

This mechanism implies that longer association times in an SPR experiment would lead to a higher ratio of the more stable complex on the surface and a slower apparent dissociation rate. Both AB0089 and a tepoditamab based molecule have an association time dependent apparent dissociation rate, thus corroborating the two-state interaction model (FIG. 136).

Binding to CLEC12A+Isogenic and Cancer Cell Lines

Assessment of AB0089 binding to isogenic cell lines expressing CLEC12A and AML cancer cells was monitored by FACS (FIGS. 137A-137D, Table 135). AB0089 showed high affinity for cell surface expressed CLEC12A and binds AML cell lines PL-21 and HL-60 with EC50 less than 1 nM. Specificity of AB0089 for CLEC12A was demonstrated by lack of binding to the parental Ba/F3 cells that do not express the target.

TABLE 135

Binding of AB0089 to isogenic and cancer cell

lines expressing CLEC12A.

Cell line
EC50 (nM)

Ba/F3-hCLEC12A
4.02

Ba/F3 parental
No binding

PL21
0.45

HL-60
0.74

Binding Affinity for Polymorphic Variant of Human CLEC12A (Q244)

Polymorphic variant CLEC12A (Q244) is prevalent in 30% of population. We tested binding of AB0089 to this genetic variant by SPR and compared its affinity to the major form of CLEC12A (K244) (FIGS. 138A-138C). Binding affinities of AB0089 for WT and Q244 variant of CLEC12A were similar (Table 136). Binding of AB0089 to cynomolgus monkey CLEC12A (cCLEC12A) was also tested. No binding was observed at 100 nM concentration, demonstrating that AB0089 is not cross-reactive to cynomolgus CLEC12A.

TABLE 136

Kinetic parameters and binding affinity of AB0089

for (K244) and (Q244) variants of human CLEC12A.

Test article
Target
k_al(M⁻¹s⁻¹)
k_dl(s⁻¹)
k_a2(s⁻¹)
k_d2(s⁻¹)
K_D(nM)

AB0089
hCLEC12A
(1.63 ± 0.51) × 10⁷
(3.98 ± 0.14) × 10⁻²
(2.95 ± 0.06) × 10⁻³
(9.62 ± 0.00) × 10⁻⁴
0.60 ± 0.26

(K244) WT

AB0089
hCLEC12A
(9.70 ± 0.45) × 10⁻⁶
(3.53 ± 0.13) × 10⁻³
(2.67 ± 0.06) × 10⁻³
(1.25 ± 0.02) × 10⁻³
1.17 ± 0.00

(Q244)

Binding to Differentially Glycosylated CLEC12A

Human CLEC12A is a heavily glycosylated protein (6 potential N-glycosylation sites with an extracellular domain (ECD) compromising of 201 amino acids). Variations in the glycosylation status of CLEC12A on the surface of different cell types is documented in the literature (Marshall et al., (2006) Eur. J. Immunol., 36: 2159-2169). CLEC12A expressed on the surface of AML cells from different patients may have different glycosylation patterns as well. To determine if AB0089 binds to different glyco-variants of human CLEC12A, the antigen was treated with neuraminidase (to remove terminal sialic acids) or PNGaseF (to remove N-linked glycans) and affinity was compared to untreated CLEC12A by SPR (FIGS. 139A-139C and Table 137). AB0089 bound to both fully glycosylated and de-glycosylated CLEC12A and is expected to be unaffected by heterogeneity in target glycan composition.

TABLE 137

Kinetic parameters and binding affinity of AB0089 for differentially glycosylated CLEC12A.

Test article
Target
k_al(M⁻¹s⁻¹)
k_dl(s⁻¹)
k_a2(s⁻¹)
k_d2(s⁻¹)
K_D(nM)

AB0089
Untreated
(1.63 ± 0.05) × 10⁷
(3.98 ± 0.01) × 10^-2
(2.95 ± 0.06) × 10⁻³
(9.62 ± 0.00) × 10⁻⁴
0.60 ± 0.26

CLEC12A

AB0089
De-sialylated
(2.09 ± 0.21) × 10⁷
(3.29 ± 0.18) × 10⁻²
(3.04 ± 0.06) × 10⁻³
(8.63 ± 0.24) × 10⁻⁴
0.35 ± 0.47

CLEC12A

AB0089
De-glycosylated
(8.96 ± 0.86) × 10⁶
(6.33 ± 0.28) × 10⁻³
(4.09 ± 0.13) × 10⁻³
(5.67 ± 0.10) × 10⁻⁴
0.09 ± 0.01

CLEC12A

Average of 3 replicates

Binding Affinity for Human NKG2D

AB0089 was tested for binding to human NKG2D ECD by SPR (Biacore) at 37° C. (FIGS. 140A-140F). NKG2D is a dimer in nature, therefore recombinant mFc-tagged NKG2D dimer was used for this experiment. The shape of the binding sensorgrams depicted in FIGS. 140A-140F enabled two different fitting models to calculate affinity and kinetic rates data:steady state affinity fit and 1:1 kinetic fit. The kinetic constants and equilibrium affinity values are shown in Table 138. AB0089 was designed to bind to human NKG2D with low affinity, most importantly with the fast rate of dissociation. The dissociation rate constant was 1.43±0.03×10⁻¹s⁻¹. Affinity values (K_D) obtained by fitting the data to a 1:1 kinetic and steady state affinity models were very similar, 881±10 nM and 884±11 nM respectively, which suggests high confidence in the measured parameters.

TABLE 138

Kinetic parameters and binding affinity of

AB0089 for human NKG2D.

Kinetics
Steady

Fit
State Fit

Test article
k_a(M⁻¹s⁻¹ )
k_d(s⁻¹)
K_D(nM)
K_D(nM)

AB0089
1.61 × 10⁵
1.41 × 10⁻¹
872
873

AB0089
1.59 × 10⁵
1.42 × 10⁻¹
892
894

AB0089
1.66 × 10⁵
1.46 × 10⁻¹
880
885

Average ±
(1.62 ± 0.03) ×
(1.43 ± 0.03) ×
881 ± 10
884 ± 11

Std Dev.
10⁵
10⁻¹

Binding Affinity for CD16a (FcγRIIIa)

One of the modalities of AB0089 mechanism of action is through engaging human CD16a (FcγRIIIa) that is expressed on NK cells via its Fc. Binding of AB0089 to CD16a (V158 allele) was evaluated by SPR and compared to the CD16a (V158) affinity for trastuzumab, an approved and marketed monoclonal antibody therapeutics of the same IgG1 isotype (FIGS. 141A-141F, Table 139). Binding of AB0089 to CD16a is comparable to the one of trastuzumab. A 3.8-fold difference in affinity is unlikely to be of physiological consequence and is also unlikely to impact AB0089 therapeutic effect since the latter relies on synergy of simultaneous engagement of two targets on the NK cells, CD16 and NKG2D. Additionally, a slightly weaker binding to CD16a may also be of benefit for the drug's safety.

TABLE 139

Kinetic parameters and binding affinity of

AB0089 for hCD16a (V158).

Test article
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0089
8.76 × 10⁴
2.94 × 10⁻²
336

AB0089
7.70 × 10⁴
2.64 × 10⁻²
343

AB0089
8.46 × 10⁴
2.72 × 10⁻²
321

Average ±
(8.31 ± 0.55) × 10⁴
(2.77 ± 0.16) × 10⁻²
333 ± 11

StDev

trastuzumab
1.31 × 10⁵
1.12 × 10⁻²
86

trastuzumab
1.22 × 10⁵
1.08 × 10⁻²
89

trastuzumab
1.30 × 10⁵
1.13 × 10⁻²
87

Average ±
(1.28 ± 0.05) × 10⁵
(1.11 ± 0.03) × 10⁻²
87 ± 1

StDev

Binding to Fcγ Receptors

Since AB0089 is an IgG1 derived biologic drug candidate it is important that it maintains appropriate Fc receptor binding properties. SPR data suggest that AB0089 binds human and cynomolgus Fcγ receptors with affinities comparable to trastuzumab, a marketed IgG1 biologic that served an experimental control (Table 140). Detailed description of each individual receptor binding is presented below.

TABLE 140

Summary of affinities of AB0089 and trastuzumab for

human and cynomolgus FcγRs.

Fcγ receptor
AB0089, K_D(nM)
trastuzumab, K_D(nM)

hFcγRI
5.1 ± 1.0
2.9 ± 0.4

cFcγRI

1.5 ± 0.3^a
1.0 ± 0.1

hFcγRIIa H131
900.4 ± 57.1
1045.1 ± 62.2

hFcγRIIa R131
1253.3 ± 269.2
1493.2 ± 265.9

hFcγRIIIa V158
333.4 ± 11.1
87.1 ± 1.4

hFcγRIIIa F158
1271.0 ± 24.6
331.4 ± 3.3

cFcγRIII^a
348.8 ± 2.4
128.8 ± 2.1

hFcγRIIb
6270.3 ± 1164.4
6270.3 ± 1031.7

hFcγRIIIb^a
8851.3 ± 1454.4
3872.4 ± 486.3

^an = 3, otherwise n = 4

Binding of AB0089 to Recombinant Human and Cynomolgus CD64 (FcγRI)

FIGS. 142A-142H and FIGS. 143A-143G depict binding of AB0089 to human and cynomolgus CD64, respectively. Table 141 demonstrates that AB0089 binds recombinant human and cynomolgus CD64 with affinities comparable (less than two-fold different) to trastuzumab.

TABLE 141

Binding of AB0089 and trastuzumab to

human and cynomolgus CD64.

Human FcγRI
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0089 Lot AB0089-002
4.4 × 10⁴
1.8 × 10⁻⁴
6.4

AB0089 Lot AB0089-002
4.6 × 10⁴
2.2 × 10⁻⁴
4.7

AB0089 Lot AB0089-002
4.6 × 10⁴
1.9 × 10⁻⁴
4.1

AB0089 Lot AB0089-002
4.6 × 10⁴
2.4 × 10⁻⁴
5.2

Average ± StDev
(4.6 ± 0.1) ×
(2.3 ± 0.4) ×
5.1 ± 1.0

10⁴
10⁻⁴

trastuzumab
9.0 × 10⁴
2.3 × 10⁻⁴
2.5

trastuzumab
8.6 × 10⁴
2.1 × 10⁻⁴
2.5

trastuzumab
8.3 × 10⁴
2.7 × 10⁻⁴
3.2

trastuzumab
8.8 × 10⁴
2.8 × 10⁻⁴
3.2

Average ± StDev
(8.1 ± 0.0) ×
(2.5 ± 0.3) ×
2.9 ± 0.4

10⁴
10⁻⁴

Cynomolgus FcγRI
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0089 Lot AB0089-002
5.5 × 10⁴
8.1 × 10⁻⁵
1.5

AB0089 Lot AB0089-002
5.4 × 10⁴
9.5 × 10⁻⁵
1.8

AB0089 Lot AB0089-002
5.5 × 10⁴
6.6 × 10⁻⁵
1.2

Average ± StDev
(5.5 ± 0.0) ×
(8.1 ± 1.5) ×
1.5 ± 0.3

10⁴
10⁻⁵

trastuzumab
1.3 × 10⁵
1.4 × 10⁻⁴
1.1

trastuzumab
1.3 × 10⁵
1.2 × 10⁻⁴
0.9

trastuzumab
1.3 × 10⁵
1.1 × 10⁻⁴
0.8

trastuzumab
1.3 × 10⁵
1.3 × 10⁻⁴
1.0

Average ± StDev
(1.3 ± 0.1) ×
(1.2 ± 0.2) ×
1.0 ± 0.1

10⁵
10⁻⁴

Binding of AB0089 to Recombinant Human CD32a (FcγRIIa)

FIGS. 144A-144P and FIGS. 145A-145P depict binding of AB0089 and trastuzumab to CD32a (H131 and R131 alleles), respectively, tested by SPR. There is no meaningful difference between affinities of both alleles for AB0089 and trastuzumab (Table 142).

TABLE 142

Binding of AB0089 and trastuzumab to

recombinant human CD32a (FcγRIIa).

K_D(nM)

Test article
FcgRIIa H131
FcgRIIa R131

AB0089 (lot AB0089-002)
953.6
1335.7

AB0089 (lot AB0089-002)
822.8
1024.1

AB0089 (lot AB0089-002)
930.1
1598.0

AB0089 (lot AB0089-002)
895.0
1055.3

Average ± StDev
900.4 ± 57.1
1253.3 ± 269.2

trastuzumab
1069.2
1488.2

trastuzumab
953.1
1381.0

trastuzumab
1090.2
1862.1

trastuzumab
1068.0
1241.4

Average ± StDev
1045.1 ± 62.2
1493.2 ± 265.9

Binding of AB0089 to Human CD16a (FcγR3a) and Cynomolgus CD16 (FcγR3)

Detailed description of AB0089 binding to high affinity CD16a (V158) is provided herein. Binding of AB0089 to the F158 allele of human CD16a and cynomolgus CD16 are presented in Table 143, FIGS. 146A-146P, and FIGS. 147A-147L. Binding of AB0089 to recombinant human CD16a (V158 and F158 alleles) and cynomolgus CD16 is comparable to a marketed IgG1 biologic drug trastuzumab. A 2.3-3.8 difference in affinity for CD16 proteins between AB0089 and trastuzumab is unlikely to be of physiological consequence and also unlikely to impact the therapeutic effect since the latter relies on the synergy of engagement of two targets on the NK cells (CD16a and NKG2D). Small difference in apparent maximum binding responses between trastuzumab and AB0089 is attributed to a difference in molecular weight. Additionally, some replicates may display slight variability in apparent maximum binding responses caused by slight variation of receptor capture levels. Note that while the shape of the sensorgrams representing binding to human CD16a F158 allele affords fitting the interactions to both, steady-state and 1:1 kinetic, models, affinity fit (steady-state) model was used for historic reasons.

TABLE 143

Binding of AB0089 and trastuzumab to recombinant

human CD16a F158 allele and cynomolgus CD16a.

Human FcγIIIIa

F158
Cyno FcγRIII

Test article
K_D(nM)
K_D(nM)

AB0089 (AB0089-002)
1250.9
351.3

AB0089 (AB0089-002)
1250.6
346.7

AB0089 (AB0089-002)
1282.2
348.3

AB0089 (AB0089-002)
1300.4

Average ± StDev
1271.0 ± 24.6
348.8 ± 2.4

trastuzumab
332.3
127.1

trastuzumab
326.6
128.1

trastuzumab
333.8
131.1

trastuzumab
333.1

Average ± StDev
331.4 ± 3.3
128.8 ± 2.1

Binding of AB0089 to Human CD32b (FcγR2b)

FIGS. 148A-148P depict binding to AB0089 and trastuzumab to CD32b tested by SPR. There is no meaningful difference between affinities of CD32b for AB0089 and trastuzumab (Table 144).

TABLE 144

Binding of AB0089 and trastuzumab to

CD32b (FcγR2b).

Test article
K_D(μM)

AB0089 (lot AB0089-002)
7.1

AB0089 (lot AB0089-002)
5.3

AB0089 (lot AB0089-002)
5.3

AB0089 (lot AB0089-002)
7.4

Average ± StDev
6.3 ± 1.2

trastuzumab
7.1

trastuzumab
5.6

trastuzumab
5.2

trastuzumab
7.2

Average ± StDev
6.3 ± 1.0

Binding of AB0089 to Human CD16b (FcγR3b)

FIGS. 149A-149L depict binding of AB0089 and trastuzumab to CD16b tested by SPR. AB0089 binds human CD16b (FcγR3b) comparably to trastuzumab (Table 145). 2.3-fold difference in affinity for the receptor is unlikely to be of physiological significance.

TABLE 145

Binding of AB0089 and trastuzumab to

recombinant human CD16b (FcγR3b).

Test Article
K_D(μM)

AB0089 (lot AB0089-002)
10.3

AB0089 (lot AB0089-002)
8.8

AB0089 (lot AB0089-002)
7.4

Average ± StDev
8.9 ± 1.5

trastuzumab
4.3

trastuzumab
3.9

trastuzumab
3.4

Average ± StDev
3.9 ± 0.5

Binding to FcRn

FIGS. 150A-150P and FIGS. 151A-151L depict AB0089 binding to human and cyno FcRn, respectively, tested by SPR. Affinities of AB0089 for human and cynomolgus FcRn are similar to those of trastuzumab, a marketed IgG1 biologic that served as an experimental control (Table 146).

TABLE 146

Summary of affinities of AB0089 and

trastuzumab for human and cyno FcRn.

AB0089,
trastuzumab,

FcRn
K_D(μM)
K_D(μM)

hFcRn, pH 6.0
1.1 ± 0.1
1.7 ± 0.2

cFcRn, pH 6.0^a
1.0 ± 0.1
1.5 ± 0.1

^an = 3, otherwise n = 4

Simultaneous Engagement of AB0089 to CD16a and NKG2D

AB0089 is a tri-functional IgG-based biologic with two binding modalities on the NK cells: CD16a binding is achieved through the Fc and NKG2D binding is achieved through A49M-I Fab. Both interactions are important for the optimal cytotoxic activity of AB0089. To demonstrate synergy of co-engagement of human CD16a and human NKG2D binding, we performed an SPR experiment where we qualitatively compared binding of AB0089 to NKG2D, CD16a and mixed NKG2D-CD16a Biacore chip surfaces. The affinity of AB0089 for human NKG2D and human CD16a are both low, however, binding to both targets simultaneously results in an avidity effect that manifests as a slower off-rate. Thus, AB0089 can avidly engage CD16a and NKG2D (FIG. 152).

Co-Engagement of CLEC12A and NKG2D

To determine if binding of one target interferes with binding of the other target to AB0089, the CLEC12A and NKG2D were sequentially injected over AB0089 captured on an anti-hFc IgG SPR chip (FIGS. 153A-153B). Target binding sensorgrams demonstrate that the occupancy status of the CLEC12A domain does not interfere with NKG2D binding (FIG. 153A) and vice versa (FIG. 153B). Similarity in shapes of the respective sensorgram segments depicting binding of each target to free AB0089 and AB0089 that has been saturated with the other target suggests that the kinetic parameters are not meaningfully affected by the target occupancy status of the AB0089 molecule. For instance, the shape of the CLEC12A binding segment of the sensorgrams is similar in both panels. Note that saturating concentration of the NKG2D had to be maintained throughout the entire experiment represented in lower panel due to fast dissociation rate of this target. Additionally, the lack of any impact on relative stoichiometry of each target binding (when compared to binding to unoccupied AB0089) signifies full independence of NKG2D and CLEC12A binding sites on AB0089 (Table 147). Therefore, AB0089 may successfully achieve simultaneous co-engagement of the tumor antigen and the NKG2D targeting arm.

TABLE 147

Relative binding stoichiometry of AB0089 for

CLEC12A and NKG2D.

Binding stoichiometries are expressed relative to

binding of each target to unoccupied captured AB0089.

CLEC12A,
NKG2D,

relative binding
relative binding

stoichiometry
stoichiometry

Target bound to AB0089
1.00
1.00

unoccupied with another

target (injected first)

Target bound to AB0089
1.16 ± 0.02
1.16 ± 0.06

saturated with another

target (injected second)

Epitope Binning

We compared binding epitope of AB0089 in relation to AB0237, AB0192 and three CLEC12A binding reference TriNKETs by SPR. cFAE-A49. Tepoditamab is based on CLEC12A binding arm of a molecule from Merus (WO_2014_051433_A1), cFAE-A49. CLL1-Merus is based on the clone 4331 from Merus (WO_2014_051433_A1) and cFAE-A49. h6E7 is based on h6E7 mAb from Genentech (h6E7) (US_2016_0075787_A1). AB0089 interaction with hCLEC12A was blocked by AB0237, and two control molecules, cFAE-A49. CLL1-Merus and cFAE-A49. Tepoditamab, but not by not Genentech-h6E7 based molecule (FIG. 154) suggesting that the binding epitope of AB0089 overlaps with the epitope of tepoditamab. The results indicated that both AB0089 and AB0237 share the same epitope. This footprint is different from AB0192 which additionally cross-blocks with cFAE-A49.h6E7.

Potency

Potency of AB0089 was tested using the KHYG-1-CD16aV cytotoxicity assay with a NK cell line engineered to stably express CD16aV and NKG2D (FIGS. 155A-155B and Table 148). AB0089 demonstrated sub-nanomolar potency in driving the lysis of HL60 and PL-21 AML cells. In addition AB0089 potently enhances cytotoxic lysis of PL21 AML cells mediated by purified human primary NK cells vastly outperforming the corresponding monoclonal antibody (FIG. 156 and Table 149). Additional cytotoxicity assessment of AB0089 is described in detail in Example 18.

TABLE 148

Potency of AB0089 in KHYG-1-CD16aV mediated

cytotoxicity assay with HL60 and PL21 cells.

Cancer cell
EC50 (nM)
Max killing (%)

HL60
0.27
111

PL21
0.16
83

TABLE 149

Potency of AB0089 in primary NK mediated

cytotoxic killing of PL21 cells.

Test article
EC50 (nM)
Max killing (%)

AB0089
0.35
20

AB0305 - parental mAb
1.85
5

Specificity

Non-Specific Binding Assessed by PSR

To assess the specificity of AB0089, a flow cytometry based PSR assay that measures binding to a preparation of detergent solubilized CHO cell membrane proteins was performed (FIGS. 157A-157D). The PSR assay has been demonstrated to correlate with orthogonal specificity assays such as cross-interaction chromatography, and the baculovirus particle enzyme-linked immunosorbent assay. Behavior in these assays strongly associates with antibody solubility and in vivo clearance and therefore serves as a good developability predictive factor (Xu et al., (2013) Protein engineering design and selection, 26, 663-670). AB0089 showed very similar profile to the low PSR control (Trastuzumab), suggesting high specificity and lack of non-specific binding to unrelated proteins (FIG. 157B). Rituximab was used as a positive control for PSR binding.

Specificity Assessment of AB0089 in HuProt™ Array

To directly examine the specificity of AB0089 we have explored a protein array technology. The HuProt™ human proteome microarray provides the largest database of individually purified human full-length proteins on a single microscopic slide. An array consisting of 22,000 full-length human proteins are expressed in S. cerevisiae yeast, purified, and are subsequently printed in duplicate on a microarray glass slide that allows thousands of interactions to be profiled in a high-throughput manner. FIG. 158 shows the relative binding (Z score) of AB0089 at 1 μg/ml to human CLEC12A in comparison to the entire human proteome microarray. For comparison purpose, the top 24 proteins with residual background binding to AB0089 were also provided. Table 150 shows the Z and S scores of AB0089 to human CLEC12A and the top 6 proteins from the microarray. S score is the difference of Z score of a given protein and the one rank next to it. Based on the Z and S score criteria, AB0089 showed high specificity to human CLEC12A and lack of off-target binding in the HuProt™ human proteome assay.

TABLE 150

Summary of the top hits by HuProt ™ array

Name
Rank
Protein-ID
Z score
S score

hCLEC12A
1
hCLEC12A
150.61
146.120

RPLP0
2
JHU16464.P173H05
4.49
0.051

POLR2E
3
JHU02080.P189F02
4.44
0.57

RPRD1A
4
JHU16457.P173E09
3.86
0.01

ZNF608
5
JHU13018.P136G04
3.85
0.10

ECDHDC1
6
JHU25034.P240F11
3.75
0.06

BC089418
7
JHU16002.P168E09
3.70
0.12

Developability Assessment

The stability and forced degradation of AB0089 was assessed under the following conditions:

Accelerated (40° C.) Stability

- 20 mg/ml AB0089 in a formulation HST, pH 6.0 (20 mM histidine, 250 mM sucrose, 0.01% Tween-80, pH 6.0), 40° C., 4 weeks (SEC, SPR, potency, CE-SDS, peptide map).
- 20 mg/ml AB0089 in CST, pH 7.0 (20 mM sodium citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0), 40° C., 4 weeks (SEC, SPR, potency, CE-SDS, peptide map).

Chemical Stability

- Oxidation: 1 mg/ml AB0089 in PBS with 0.02% hydrogen peroxide, room temperature, 24 hours (SEC, CE-SDS, SPR, potency, peptide map).
- pH 8.0: 1 mg/ml AB0089 in 20 mM Tris, 40° C., 2 weeks (cIEF, SPR, potency, peptide map).
- pH 5.0: 1 mg/ml AB0089 in 20 mM sodium acetate, 40° C., 2 weeks (cIEF, SPR, potency, peptide map).

Manufacturability

- Freeze/thaw: 20 mg/ml AB0089 in 20 mM sodium citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 (SEC, SDS-CE)
- Agitation: 10 mg/ml AB0089 in 20 mM sodium citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 (SEC, SDS-CE)
- High concentration: >150 mg/ml, PBS (SEC, A280)
- Low pH hold (mimicking conditions of viral clearance step of manufacturing process): pH 3.3, 1.5 hours, ProA eluate (SEC); fully purified AB0089 (cIEF, CE-SDS, SPR, potency)
- Low pH hold (to investigate isomerization): fully purified AB0089 was diluted with ProA elution buffer and pH adjusted to 3.5. The sample was held for 1 hour at room temperature, neutralized with Tris buffer and buffer exchanged into PBS, pH 7.4. before additional analysis (SEC, cIEF, CE-SDS, SPR, potency, peptide map).

Accelerated (40° C.) Stability

To assess the stability of AB0089, the molecule was staged at protein concentration of 20 mg/ml at 40° C. for 4 weeks under two different buffer conditions (1) in 20 mM histidine, 250 mM sucrose, 0.01% PS-80, pH 6.0 (HST), and (2) in 20 mM citrate, 250 mM sucrose, 0.01% PS-80, pH 7.0 (CST). Stability in both buffers was assessed by SEC, CE, SPR and potency.

Stability in HST, pH 6.0

AB0089 demonstrated high stability after 4 weeks of incubation in HST, pH 6.0, at 40° C. No aggregation and minimal loss of monomer (1.6%) was observed by SEC (FIG. 159 and Table 151). Minimal fragmentation was detected by R CE-SDS (FIGS. 160A-160B and Table 152). Additionally, there was no meaningful difference observed in the amount of active species, binding kinetics or affinities for hCLEC12A, hNKG2D and hCD16a between control and stressed samples (FIGS. 161A-161F and Table 153). Finally, no difference in potency between the control and stressed samples was detected (FIG. 162 and Table 154).

TABLE 151

SEC analysis of AB0089 after 2 and 4 weeks

incubation at 40° C. in HST, pH 6.0.

Monomer
LMWS
HMWS

Test Article
(%)
(%)
(%)

AB0089 control
99.5
0.5
0.0

AB0089, HST, pH6.0,
99.1
0.9
ND

40° C., 2 wks

AB0089, HST, pH6.0,
98.4
1.3
0.4

40° C., 4 wks

TABLE 152

NR and R CE-SDS analysis of AB0089 after 4 weeks at

40° C. in HST, pH 6.0.

NR CE
R CE

Test article
Purity (%)
Purity (%)

AB0089 control
98.2
94.3

AB0089, HST, pH6.0, 40° C., 4 wks
98.3
97.2

TABLE 153

Kinetic parameters and binding affinities of AB0089 for CLEC12A,

NKG2D and CD16a after 4 weeks at 40° C. in HST, pH 6.0.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0089 Control
hCLEC12A
k_a1(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
1.0 ± 0.0

(8.87 ± 0.53) × 10⁶
(3.49 ± 0.04) × 10⁻³
(4.20 ± 0.07) × 10⁻²
(9.37 ± 0.32) × 10⁻⁴

AB0089 Stressed
hCLEC12A
k_a1(M⁻¹S⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s^-1)
1.0 ± 0.1

(7.91 ± 0.42) × 10⁶
(3.69 ± 0.04) × 10⁻³
(4.24 ± 0.17) × 10⁻²
(8.56 ± 0.45) × 10⁻⁴

AB0089 Control
hNKG2D
(1.75 ± 0.12) × 10⁵
(1.35 ± 0.08) × 10⁻¹
772 ± 13

AB0089 Stressed
hNKG2D
(1.66 ± 0.03) × 10⁵
(1.32 ± 0.02) × 10⁻¹
796 ± 5

AB0089 Control
CD16aV
(8.53 ± 0.05) × 10⁴
(2.76 ± 0.02) × 10⁻²
323 ± 3

AB0089 Stressed
CD16aV
(7.93 ± 0.27) × 10⁴
(2.84 ± 0.07) × 10⁻²
359 ± 6

Average of 3 replicates

TABLE 154

AB0089 potency after 4 weeks at 40° C. in HST, pH 6.0, in

KHYG-1-CD16aV cytotoxicity assay.

Test article
EC50 (nM)
Max killing (%)

AB0089 control
0.16
83

AB0089, HST, pH 6.0, 40° C., 4 wks
0.17
84

Stability in CST pH 7.0

AB0089 demonstrated high stability after 4 weeks of incubation in CST, pH 7.0, at 40° C. A minimal loss monomer content (1.6%) was observed by SEC (FIG. 163 and Table 155). No fragmentation of AB0089 was observed by CE-SDS (FIGS. 164A-164B and Table 156). Additionally, there was no meaningful difference in active protein content or in binding affinities for hCLEC12A, hNKG2D and hCD16a between control and stressed samples (FIGS. 165A-165F and Table 157). Finally, long term thermal stress did not affect potency of AB0089 (FIG. 166 and Table 158).

TABLE 155

SEC analysis of AB0089 during weeks 1-4 at 40° C. in CST, pH 7.0.

Conc.
Monomer
LMWS
HMWS

Test Article
(mg/ml)
(%)
(%)
(%)

AB0089 control
22.0
99.1
0.5
0.5

AB0089, CST, pH7.0, 40° C.,
ND
98.2
1.2
0.6

1 wk

AB0089, CST, pH 7.0, 40° C.,

97.9
1.3
0.7

2 wks

AB0089, CST, pH 7.0, 40° C.,

97.7
1.5
0.8

3 wks

AB0089, CST, pH 7.0, 40° C.,
22.6
97.5
1.7
1.0

4 wks

ND, none detected

TABLE 156

NR and R CE-SDS analysis of AB0089 after

4 weeks at 40° C. in CST, pH 7.0.

Test article
NR CE Purity (%)
R CE Purity (%)

AB0089 control
98.6
92.7

AB0089, CST, pH 7.0, 40° C., 4 wks
96.2
96.0

TABLE 157

Kinetic parameters and binding affinities of AB0089 for CLEC12A,

NKG2D and CD16a after 4 weeks at 40° C. in CST, pH 7.0

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0089 Control
hCLEC12A
k_a1(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
1.0 ± 0.0

(8.87 ± 0.53) × 10⁶
(3.49 ± 0.04) × 10⁻³
(4.20 ± 0.07) × 10⁻²
(9.37 ± 0.32) × 10⁻⁴

AB0089 Stressed
hCLEC12A
k_a1(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
1.0 ± 0.0

(6.31 ± 0.32) × 10⁶
(3.93 ± 0.12) × 10⁻³
(3.73 ± 0.15) × 10⁻²
(8.00 ± 0.05) × 10⁻⁴

AB0089 Control
hNKG 2D
(1.75 ± 0.12) × 10⁵
(1.35 ± 0.08) × 10⁻¹
772 ± 13

AB0089 Stressed
hNKG2D
(1.60 ± 0.04) × 10⁵
(1.26 ± 0.06) × 10⁻⁴
791 ± 13

AB0089 Control
CD16aV
(8.53 ± 0.05) × 10⁴
(2.76 ± 0.02) × 10⁻²
323 ± 3

AB0089 Stressed
CD16aV
(8.31 ± 0.43) × 10⁴
(2.84 ± 0.07) × 10⁻²
342 ± 9

Average of 3 replicates

TABLE 158

Potency of AB0089 after 4 weeks at 40° C. in CST, pH 7.0 in

KHYG1-CD16aV cytotoxicity assay.

Test article
Buffer
EC50 (nM)
Max killing (%)

AB0089 control
CST, pH 7.0
0.13
87

AB0089 stressed
CST, pH 7.0
0.16
88

Peptide Map Analysis of Accelerated Stability Samples

To determine if the 40° C. stress induced modifications in the complementarity determining regions (CDRs), the control and 4-week stressed samples were analyzed by LC-MS/MS peptide mapping. Based on the peptide map data, all of the CDRs in AB0089 are resistant to modification during elevated temperature stress (Table 159). No evidence of aspartic acid isomerization in the peptide encompassing the CLEC12A binding CDRH3 was observed, based on manual inspection of peak shape (FIGS. 167A-167C).

TABLE 159

Post-translational modifications in AB0089 CDRs

after elevated temperature stress.

Relative

Abundance (%)

Sequence

Stressed
Stressed

Chain
CDR
(Chothia)
Modification
Control
(CST)
(HST)

S
H1
GFSLTNY (SEQ ID
Deamidation
0.7
1.7
0.9

NO: 137)

S
H2
WVGGA (SEQ ID
None
NA

NO: 272)

S
H3
GDYGDTLDY (SEQ
Isomerization
None detected

ID NO: 273)

S
Ll
HASQNINFWLS
Deamidation
0.3
1.0
0.4

(SEQ ID NO: 140)

S
L2
EASNLHT (SEQ ID
Deamidation
None detected

NO: 141)

S
L3
QQSHSYPLT (SEQ
None
NA

ID NO: 142)

H
H1
GFTFSSY (SEQ ID
None
NA

NO: 297)

H
H2
SSSSSY (SEQ ID
None
NA

NO: 298)

H
H3
GAPIGAAAGWFDP
Truncation
None detected

(SEQ ID NO: 97)

L
Ll
RASQGISSWLA
None
NA

(SEQ ID NO: 86)

L
L2
AASSLQS (SEQ ID
None
NA

NO: 77)

L
L3
QQGVSFPRT (SEQ
None
NA

ID NO: 87)

NA: not applicable

Chemical Stability

Forced Oxidation

To assess stability of AB0089 under oxidative stress, AB0089 was incubated with 0.02% hydrogen peroxide for 24 hours at room temperature in PBS. SEC analysis showed no difference in monomer content between oxidized AB0089 and control sample (FIG. 168 and Table 160). No increase in fragmentation was detected in AB0089 by both reduced and non-reduced CE-SDS (FIGS. 169A-169B and Table 161).

TABLE 160

SEC analysis of AB0089 after forced oxidation compared to control.

Concentration
Monomer
LMWS
HMWS

Test article
(mg/ml)
(%)
(%)
(%)

AB0089 control
1.1
99.3
0.4
0.3

AB0089 forced oxidation
1.0
99.1
0.5
0.4

TABLE 161

NR and R CE-SDS analysis of AB0089 after forced oxidation.

Test article
NR CE Purity (%)
R CE Purity (%)

AB0089 control
98.9
98.0

AB0089 forced oxidization
97.4
98.1

Site-specific oxidation of methionine was monitored by tryptic peptide mapping following forced oxidation. Significant levels of oxidation were only detected at two methionines in the Fc (Table 162). Our historic data show that the same residues are modified in trastuzumab at 61.6% and 39.1% under the same conditions.

TABLE 162

Summary of methionine oxidation in AB0089

befor e and after oxidative stress.

Relative

Abundance (%)

Sequence
Chain
Position
Control
Oxidized

LSCAASGFTFSSYSMNWVR
H
34
0.6
0.9

(SEQ ID NO: 300)

NSLYLQMNSLR (SEQ ID
H
83
0.4
0.9

NO: 301)

DTLMISR (SEQ ID NO:
H/S
257/278
2.1
66.7

302)

WQQGNVFSCSVMHEALHNH
H/S
433/454
1.3
37.8

YTQK (SEQ ID NO:

303)

DIQMTQSPSSVSASVGDR
L
4
0.5
1.1

(SEQ ID NO: 304)

GDYGDTLDYWGQGTLVTVS
S
141
3.1
3.5

SGGGGSGGGGSGGGGSGGG

GSDIQMTQSPSSLSASVGD

R (SEQ ID NO: 339)

Oxidation of tryptophan residues in AB0089 CDRs was also monitored in the same experiment. None of the tryptophan residues had a significant increase in oxidation after stress (Table 163).

TABLE 163

Summary of tryptophan oxidation in AB0089 CDRs

before and after oxidative stress.

Relative

Mass shift
Abundance (%)

Sequence
Chain
CDR
(Da)^a
Control
Oxidized

WVGGA (SEQ ID NO:
S
H2
+16/+32/+4
ND/0.1/0.2
ND/0.2/0.3

272)

HASQNINFWLS (SEQ
S
L1
+16/+32/+4
ND/0.2/0.2
ND/0.3/0.3

ID NO: 140)

GAPIGAAAGWFDP (SEQ
H
H3
+16/+32/+4
0.0/0.2/0.4
0.0/0.3/0.7

ID NO: 97)

RASQGISSWLA (SEQ
L
L1
+16/+32/+4
0.2/0.1/0.2
0.2/0.2/0.3

ID NO: 86)

^asingle oxidation (+16 Da), double oxidation (+32 Da), and kynurenine (+4 Da),

ND, none detected.

Oxidative stress had no significant effect on the active protein content or on the affinities for hCLEC12A, hNKG2D, and hCD16aV (FIGS. 170A-170F and Table 164). No significant effect of oxidation on the potency of AB0089 was observed (FIG. 171 and Table 165).

TABLE 164

Kinetic parameters and binding affinities of AB0089 for CLEC12A,

NKG2D and CD16A after forced oxidation compared to control.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0089 Control
hCLEC12A
k_a1(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
0.9 ± 0.0

(1.01 ± 0.04) × 10⁷
(3.46 ± 0.01) × 10⁻³
(4.76 ± 0.22) × 10⁻²
(8.37 ± 0.29) × 10⁻⁴

AB0089 Oxidized
hCLEC12A
k_a1(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
0.9 ± 0.0

(9.93 ± 0.11) × 10⁶
(3.47 ± 0.04) × 10⁻³
(4.81 ± 0.10) × 10⁻²
(8.37 ± 0.22) × 10⁻⁴

AB0089 Control
hNKG2D
(1.69 ± 0.04) × 10⁵
(1.35 ± 0.01) × 10⁻¹
797 ± 19

AB0089 Oxidized
hNKG2D
(1.71 ± 0.13) × 10⁵
(1.30 ± 0.07) × 10⁻¹
762 ± 13

AB0089 Control
hCD16aV
(8.64 ± 0.22) × 10⁴
(2.78 ± 0.05) × 10⁻²
322 ± 7

AB0089 Oxidized
hCD16aV
(8.65 ± 0.43) × 10⁴
(2.77 ± 0.05) × 10⁻²
320 ± 10

Average of 3 replicates

TABLE 165

AB0089 potency before and after forced oxidation in

KHYG1-CD16aV cytotoxicity assay.

Test article
EC50 (nM)
Max Killing (%)

AB0089 Control
0.11
84

AB0089 Oxidized
0.16
87

pH 8 Stress

Chemical stability of AB0089 was assessed by long term incubation at elevated pH (20 mM Tris, pH 8.0, 40° C., 2 weeks). AB0089 appears to be highly stable: only 1.2% loss of monomer was detected by SEC (FIG. 172 and Table 166).

TABLE 166

SEC analysis of AB0089 after long term high pH

stress (pH 8.0, 40° C., 2 wks).

Test article
Monomer (%)
LMWS (%)
HMWS (%)

AB0089 control
99.4
0.4
0.2

AB0089 pH 8 stressed
98.2
1.1
0.7

Reduced and non-reduced CE-SDS showed low levels of degradation, 1.7% and 3.9%, respectively (FIGS. 173A-173B and Table 167).

TABLE 167

NR and R CE-SDS analysis of AB0089 after long term high pH

stress (pH 8, 40° C., 2 wks).

Test article
NR CE Purity (%)
R CE Purity (%)

AB0089 control
98.6
98.2

AB0089 pH 8 stressed
96.9
94.3

After long term high pH stress, there was an acidic shift in the global charge profile detected by cIEF (FIG. 174 and Table 168). This acidic shift can be attributed to deamidation throughout the AB0089 sequence. A similar acidic shift was observed for trastuzumab after the same stress conditions (data not shown).

TABLE 168

AB0089 charge profile after long term high pH

exposure (pH 8, 40° C., 2 wks).

Test article
% acidic
% main
% basic

AB0089 control
43.1
51.4
5.5

AB0089 pH 8 stressed
66.6
29.0
4.4

There was no meaningful difference in the amount of active protein or in the affinities for hCLEC12A, hNKG2D, or hCD16aV between stressed and control samples (FIGS. 175A-175F and Table 169).

TABLE 169

Kinetic parameters and binding affinities of AB0089 for CLEC12A, NKG2D,

and CD16aV after long term high pH stress (pH 8, 40° C., 2 wks).

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0089 Control
hCLEC12A
k_a1(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
0.9 ± 0.0

(1.01 ± 0.04) × 10⁷
(3.46 ± 0.01) × 10⁻³
(4.76 ± 0.22) × 10⁻²
(8.37 ± 0.29) × 10⁻⁴

AB0089 pH 8 Stressed
hCLEC12A
k_a1(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
0.9 ± 0.1

(9.93 ± 0.05) × 10⁷
(3.36 ± 0.09) × 10⁻³
(4.90 ± 0.24) × 10⁻²
(8.63 ± 0.33) × 10⁻⁴

AB0089 Control
hNKG2D
(1.69 ± 0.04) × 10⁵
(1.35 ± 0.01) × 10⁻¹
797 ± 19

AB0089 pH 8 Stressed
hNKG2D
1.30 × 10⁵
1.27 × 10⁻¹
983

AB0089 Control
hCD16aV
(8.64 ± 0.22) × 10⁴
(2.78 ± 0.05) × 10⁻²
322 ± 7

AB0089 pH 8 Stressed
hCD16aV
(7.94 ± 0.28) × 10⁴
(2.79 ± 0.15) × 10⁻²
350 ± 11

Average of 3 replicates for all samples, except pH 8 stressed binding to hNKG2D (n = 2)

After long term high pH exposure, the potency of AB0089 was reduced approximately 2-fold (FIG. 176 and Table 170). Considering that orthogonal methods of analysis did not show any difference in binding or in the factive species content (FIGS. 175A-175F), it appears that these differences may be due to potency assay variability.

TABLE 170

AB0089 potency after long term high pH

exposure (pH 8, 40° C., 2 wks).

Test article
EC50 (nM)
Max killing (%)

AB0089 control
0.11
84

AB0089 pH 8 stressed
0.23
83

pH 5 Stress

Chemical stability of AB0089 was also assessed by long term incubation at low pH (20 mM sodium acetate, pH 5.0, 40° C., 2 weeks). DF appears to be highly stable: only 0.3% loss of monomer was detected by SEC (FIG. 177 and Table 171).

TABLE 171

SEC analysis of AB0089 after long term low pH

stress (pH 5.0, 40° C., 2 wks).

Test article
Monomer (%)
LMWS (%)
HMWS (%)

AB0089 control
99.3
0.4
0.2

AB0089 pH 5 stressed
99.0
0.8
0.2

After long term pH 5 stress, low levels of AB0089 degradation were detected by non-reduced and reduced CE-SDS: 1.0% and 0.3% loss of monomer content, respectively (FIGS. 178A-178B and Table 172).

TABLE 172

NR and R CE-SDS analysis of AB0089 after long term low pH

stress (pH 5.0, 40° C., 2 wks).

Test article
NR CE Purity (%)
R CE Purity (%)

AB0089 control
98.6
97.7

AB0089 pH 5 stressed
97.6
97.4

After long term pH 5 stress, there was a small acidic shift in the global charge profile detected by cIEF (FIG. 178A-178B and Table 173), with one acidic variant in particular increasing significantly.

TABLE 173

Charge profile of AB0089 after long term low pH

stress (pH 5.0 40° C., 2 wks).

Test article
% acidic
% main
% basic

AB0089 control
37.8
56.4
5.8

AB0089 pH 5 stressed
47.3
45.9
6.8

After long term low pH exposure, there was no meaningful difference in active protein content or in AB0089 affinities for hCLEC12A, hNKG2D, or hCD16aV (FIGS. 180A-180F and Table 174).

TABLE 174

Kinetic parameters and binding affinity of AB0089 for CLEC12A,

NKG2D, and CD16aV after high pH stress (pH 5.0, 40° C., 2 wks).

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0089 Control
hCLEC12A
k_a1(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
0.9 ± 0.0

(1.01 ± 0.04) × 10⁷
(3.46 ± 0.01) × 10⁻³
(4.76 ± 0.22) × 10⁻²
(8.37 ± 0.29) × 10⁻⁴

AB0089 pH 5 stressed
hCLEC12A
k_al(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
1.0 ± 0.0

(8.62 ± 0.02) × 10⁶
(3.47 ± 0.04) × 10⁻³
(4.27 ± 0.22) × 10⁻²
(8.40 ± 0.09) × 10⁻⁴

AB0089 Control
hNKG2D
(1.69 ± 0.04) × 10⁵
(1.35 ± 0.01) × 10⁻¹
797 ± 19

AB0089 pH 5 Stressed
hNKG2D
1.87 × 10⁵
1.31 × 10⁻¹
703

AB0089 Control
CD16aV
(8.64 ± 0.22) × 10⁴
(2.78 ± 0.05) × 10⁻²
322 ± 7

AB0089 pH 5 Stressed
CD16aV
(7.54 ± 0.36) × 10⁴
(2.89 ± 0.05) × 10⁻²
383 ± 12

Average of 3 replicates for all samples, except pH 5 stressed sample binding to hNKG2D (n = 2)

Potency of AB0089 after long term pH 5 exposure remained similar to potency of the control sample (FIG. 181 and Table 175).

TABLE 175

AB0089 potency after long term low pH stress

(pH 5.0 40° C., 2 wks) in KHYG-1-CD16aV

cytotoxicity assay.

EC50
Max killing

Test article
(nM)
(%)

AB0089 control
0.11
84

AB0089 pH 5 stressed
0.16
86

Peptide Mapping Analysis of the pH 5 and pH 8 Stressed Material

To determine if the pH stress induced changes observed by cIEF are due to the modification in the complementarity determining regions (CDRs), pH 5 and pH 8 stressed samples were analyzed by LC-MS/MS peptide mapping. Based on the peptide map data, all of the CDRs in AB0089 are resistant to modification during low and high pH stress (Table 176). No evidence of aspartic acid isomerization in the peptide encompassing the CDRH3 of the CLEC12A targeting arm of AB0089 was observed, based on manual inspection of peak shape (FIGS. 182A-182B).

TABLE 176

Post-translational modifications in AB0089 CDRs after

long term exposure to low and high pH (pH 5.0 and pH 8.0).

Relative

abundance (%)

Sequence

pH 5
pH 8

Chain
CDR
(Chothia)
Modification
control
stressed
control
stressed

S
H1
GFSLTNY (SEQ ID
Deamidation
0.2
0.6
0.4
1.4

NO: 137)

S
H2
WVGGA (SEQ ID
None
NA

NO: 272)

S
H3
GDYGDTLDY (SEQ
Isomerization
None detected

ID NO: 273)

S
L1
HASQNINFWLS
Deamidation
0.1
0.1
0.2
0.4

(SEQ ID NO: 140)

S
L2
EASNLHT (SEQ ID
Deamidation
None detected

NO: 141)

S
L3
QQSHSYPLT (SEQ
None
NA

ID NO: 142)

H
H1
GFTFSSY (SEQ ID
None
NA

NO: 297)

H
H2
SSSSSY (SEQ ID
None
NA

NO: 298)

H
H3
GAPIGAAAGWFDP
Truncation
None detected

(SEQ ID NO: 97)

L
L1
RASQGISSWLA
None
NA

(SEQ ID NO: 86)

L
L2
AASSLQS (SEQ ID
None
NA

NO: 77)

L
L3
QQGVSFPRT (SEQ
None
NA

ID NO: 87)

NA: not applicable

Manufacturability

Freeze/Thaw Stability

Aliquots of AB0089 (20 mg/ml in CST pH 7.0) were frozen and thawed 6 times by placing the vials in Styrofoam container (to slow the freezing rate) and placing the container at −80° C. After 6 cycles of freeze/thaw AB0089 showed no loss in monomer by SEC. Protein concentration measured by A280 remained the same (FIG. 183 and Table 177).

TABLE 177

SEC analysis of AB0089 after freeze/thaw stress.

Conc.
Monomer
LMWS
HMWS

Test article
(mg/ml)
(%)
(%)
(%)

AB0089 control
22.3
99.0
0.7
0.3

AB0089 after 6 F/T cycles
22.1
98.9
0.8
0.3

After 6 freeze/thaw cycles no loss of purity was detected by reduced or non-reduced CE-SDS (FIGS. 184A-184B and Table 178).

TABLE 178

Purity by NR and R CE-SDS analysis of

AB0089 after freeze/thaw stress.

NR CE
R CE

Purity
Purity

Test article
(%)
(%)

AB0089 control
98.4
93.2

AB0089 after 6 F/T cycles
98.3
93.7

Agitation

AB0089 (at 10 mg/ml in 20 mM sodium citrate, 250 mM sucrose, pH 7.0, with or without 0.01% Tween-80) was shaken at 300 rpm at room temperature in a deep well plate for 3 days. After agitation stress, there was no loss of monomer detected by SEC (FIGS. 185A-185B and Table 179), no loss of protein concentration (A280), and no increase in turbidity (A340) was observed.

TABLE 179

SEC analysis of AB0089 after agitation stress.

0.01%
Conc.
Monomer

Test article
Tween-80
(mg/ml)
(%)

AB0089 control
+
9.86
99.5

AB0089 agitation
+
9.97
99.7

AB0089 control
−
9.81
99.4

AB0089 agitation
−
9.93
99.1

High Concentration

AB0089 was concentrated to approximately 10, 15, 25, 50, 100, 150 and 175 mg/ml in PBS, pH 7.4 and recovery was monitored by visual inspection, concentration (A280), and SEC. For SEC analysis, samples were injected undiluted (injection volume was normalized to the concentration). AB0089 was able to be concentrated to >175 mg/ml with minimal loss in percent monomer (FIGS. 186A-186B, FIG. 187, Table 180) and no visible precipitation.

TABLE 180

SEC analysis of AB0089 at different concentrations.

Concentration
%
%

(mg/ml)
Monomer
HMW

9.7
99.7
0.3

14.7
99.7
0.3

26.2
99.7
0.3

53.4
99.6
0.4

94.2
99.6
0.4

143.5
99.6
0.4

178.4
99.5
0.5

Low pH Hold

Two different methods were used to assess the behavior of AB0089 during low pH hold (mock viral inactivation). The first method mimicked the manufacturing process, where the eluate from the first chromatography step (Protein A) was held at low pH before additional purification. The second method incubated fully purified AB0089 under low pH conditions. This method was used to assess potential low pH induced chemical modifications of the CDRs.

Low pH Hold Exposure of Protein A Eluate

To determine if AB0089 is resistant to low pH hold, AB0089 Protein A eluate was adjusted to pH 3.5 and held for 2 hours at room temperature. After the hold period, the Protein A eluate was neutralized with 1.0 M Tris, pH 8.3 to achieve neutral pH. Analytical SEC was performed to determine if there were any changes in profile or aggregate content pre- and post-low pH exposure (FIGS. 188A-188B). No change in the SEC profile of AB0089 was observed after low pH hold compared to “no-hold” control sample, suggesting that AB0089 is highly stable during low pH hold/viral inactivation conditions used in biologics manufacturing.

Protein was further processed through the second step of DF platform purification method and analyzed using a panel of additional assays in comparison with purified protein that was not subjected to low pH hold. Chemical modification of amino acid side chains can typically be observed at a global scale with cIEF. The cIEF profiles of AB0089 control and low pH hold lot are visually very similar and the relative quantitation of acidic, main, and basic species are all within 5% of each other. This indicates that the low pH hold did not have a measurable effect on the charge profile of AB0089 (FIG. 189 and Table 181).

TABLE 181

AB0089 charge profile after low pH hold.

%
%
%

Test article
pI
acidic
main
basic

AB0089 control
8.5
43.7
50.8
5.5

AB0089 low pH hold
8.6
39.5
54.3
6.2

The affinities of AB0089 for hCLEC12a, hNKG2D, and CD16aV were not affected by low pH hold (FIGS. 190A-190F and Table 182). No difference in active protein content was observed. The difference in binding affinity and maximal binding response between the control and low pH hold sample were within the experimental variability of the assay. This indicates that AB0089 retains affinity for all three targets after low pH hold.

TABLE 182

Kinetic parameters and binding affinities of AB0089 for

CLEC12A, NKG2D, and CD16aV after low pH hold.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0089 Control
hCLEC12A
k_a1(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
0.7 ± 0.0

(1.04 ± 0.03) × 10⁷
(3.43 ± 0.04) × 10⁻³
(4.35 ± 0.12) × 10⁻²
(7.41 ± 0.10) × 10⁻⁴

AB0089 low pH hold
hCLEC12A
k_a1(M⁻¹s⁻¹)
k_a2(s⁻¹)
k_d1(s⁻¹)
k_d2(s⁻¹)
0.7 ± 0.0

(1.11 ± 0.07) × 10⁷
(3.50 ± 0.03) × 10⁻³
(4.65 ± 0.05) × 10⁻²
(7.49 ± 0.03) × 10⁻⁴

AB0089 Control
hNKG2D
(1.62 ± 0.03) × 10⁵
(1.43 ± 0.03) × 10⁻¹
881 ± 10

AB0089 low pH hold
hNKG2D
(1.59 ± 0.07) × 10⁵
(1.50 ± 0.05) × 10⁻⁴
941 ± 45

AB0089 Control
CD16aV
(8.31 ± 0.55) × 10⁴
(2.77 ± 0.16) × 10⁻²
333 ± 11

AB0089 low pH hold
CD16aV
(8.73 ± 0.22) × 10⁴
(2.74 ± 0.05) × 10⁻²
314 ± 8

Results are an average of 3 replicates

Potency of AB0089 subjected to low pH hold and potency of control sample were tested against CLEC12A+ HL-60 cancer cells in the KHYG1-CD16aV bioassay. The potency of AB0089 was unaltered after the low pH hold step (FIG. 191 and Table 183).

TABLE 183

Potency of AB0089 after low pH hold in

KHYG-1-CD16aV cytotoxicity assay.

EC50
Max killing

Test article
(nM)
(%)

AB0089 control
0.11
95

AB0089 low pH hold
0.11
89

Low pH Hold Exposure of Fully Purified Protein

An additional low pH hold study was carried out to investigate if the DT motif in the CDRH3 (KGDYGDTLDY) of the CLEC12A targeting scFv would isomerize during a viral inactivation step. Low pH hold material was prepared from fully purified AB0089 material using the following steps:

- 1. AB0089 was diluted 1:1 with 0.1 M glycine, pH 3.0 to mimic ProA elution
- 2. The pH was adjusted to 3.5±0.1 with 2 M acetic acid (to mimic low pH hold conditions)
- 3. The low pH hold was carried out for 60 min at room temperature
- 4. The pH was neutralized to ˜7.5 with 1 M Tris, pH 8.0
- 5. AB0089 was buffer exchanged into PBS, pH 7.4

The final low pH hold material (AB0089 lot AB0089-004) was compared to control (AB0089 lot AB0089-002) using a panel of biological assays: SEC, cIEF, SPR binding to CLEC12A, peptide map, and potency. Analytical SEC was performed to determine if there were any changes in profile or aggregate content pre- and post-low pH exposure (FIGS. 192A-192B and Table 184). No change in profile of % monomer was observed after the low pH hold.

TABLE 184

SEC analysis of AB0089 after low pH hold

compared to control.

Monomer
LMWS
HMWS

Test article
(%)
(%)
(%)

AB0089 control
99.7
ND
0.3

AB0089 low pH hold
99.7
ND
0.3

ND, none detected

As was seen in the first low pH hold study, the cIEF profiles of AB0089 control and low pH hold lot are visually very similar and the relative quantitation of acidic, main, and basic species are all within 6% of each other. Hence, low pH hold did not have a significant effect on the charge profile of AB0089 (FIG. 193 and Table 185).

TABLE 185

Charge profile of AB0089 after low pH hold.

%
%
%

Test article
pI
acidic
main
basic

AB0089 control
8.6
42.5
51.1
6.5

AB0089 low pH hold
8.6
38.0
56.8
5.4

The affinity of AB0089 for hCLEC12a was unaltered compared to control (FIGS. 281A-281B and Table 186). The difference in binding affinity and maximal binding response between the control and low pH hold sample are within the experimental variability of the assay. This indicates that there was no impact of low pH incubation on the binding of the CLEC12A binding arm with target antigen.

TABLE 186

Summary of kinetic parameters and binding affinities of AB0089 for hCLEC12A

after low pH hold compared to control.

Test article
k_a1(M⁻¹s⁻¹)
k_d1(s⁻¹)
k_a2(s⁻¹)
k_d2(s⁻¹)
K_D(nM)

AB0089
(1.03 ± 0.03) × 10⁷
(2.72 ± 0.11) × 10⁻²
(3.59 ± 0.05) × 10⁻³
(8.56 ± 0.32) × 10⁻⁴
0.51 ± 0.02

control

AB0089
(1.02 ± 0.06) × 10⁷
(2.76 ± 0.10) × 10⁻²
(3.60 ± 0.04) × 10⁻³
(8.62 ± 0.11) × 10⁻⁴
0.53 ± 0.02

low pH hold

Results are an average of 3 replicates

The potency of AB0089 after low pH hold against CLEC12A+ HL-60 cancer cells was assessed in the KHYG1-CD16V bioassay. No difference between potencies of low pH hold and control samples was observed (FIG. 282 and Table 187).

TABLE 187

Potency AB0089 after low pH hold in the

KHYG1-CD16aV cytotoxicity assay.

Maximum

EC50
Lysis

Test article
(nM)
(%)

AB0089 control (AB0089-002)
0.04
99

AB0089 low pH hold (AB0089-004)
0.03
98

To determine if the low pH hold induced changes in the modification in the CDRs, the control and low pH hold samples were analyzed by LC-MS/MS peptide mapping. Based on the peptide map data, all of the CDRs in AB0089 are resistant to modification during low pH hold (Table 188). No evidence of aspartic acid isomerization was observed, based on manual inspection of peak shape.

TABLE 188

AB0089 are resistant to post-translational

modifications after low pH hold.

Relative

abundance (%)

Sequence
Potential

low

Chain
CDR
(Chothia)
liability
control
pH hold

S
H1
GFSLTNY (SEQ ID
Deamidation
0.7
1.1

NO: 137)

S
H2
WVGGA (SEQ ID
None
NA

NO: 272)

S
H3
GDYGDTLDY (SEQ
Isomerization
ND
ND

ID NO: 273)

S
L1
HASQNINFWLS (SEQ
Deamidation
0.2
0.1

ID NO: 140)

S
L2
EASNLHT (SEQ ID
Deamidation
ND
ND

NO: 141)

S
L3
QQSHSYPLT (SEQ ID
None
NA

NO: 142)

H
H1
GFTFSSY (SEQ ID
None
NA

NO: 297)

H
H2
SSSSSY (SEQ ID NO:
None
NA

298)

H
H3
GAPIGAAAGWFDP
Truncation
ND
ND

(SEQ ID NO: 97)

L
L1
RASQGISSWLA (SEQ
None
NA

ID NO: 86)

L
L2
AASSLQS (SEQ ID
None
NA

NO: 77)

L
L3
QQGVSFPRT (SEQ ID
None
NA

NO: 87)

NA, not applicable

Taken together, these results demonstrate that AB0089 is an improved version of AB0192 with an optimized CMC profile. The molecular format and structure of AB0089 meets the conceptual requirements for a new multi-specific therapeutic modality that simultaneously engages CLEC12A, NKG2D and CD16a to treat AML. AB0089 is potent and has a favorable developability profile with no sequence liabilities or significant post-translational modifications observed under any stresses. AB0089 binds to an epitope on CLEC12a similar to tepoditamab (Merus).

Example 18
AB0089 Cytotoxicity and Mechanism of Action
Objectives

Measure the activity of AB0089 in cell-based assays in vitro using physiologically relevant conditions to characterize AB0089's modes of action.

Study Design

The primary mechanism of action for AB0089 was tested using in vitro assays systems with human NK cells and CLEC12A+ cancer cells. Other effector functions aside from NK cell-mediated cytolysis were also paired with CLEC12A+ cancer cells to test additional mechanisms of action for AB0089. Macrophages were effector cells for ADCP assays and human serum was used as a source of complement to test CDC activity.

Materials and Methods

Primary Cells and Cell Lines

Human cancer cell lines were obtained from research cell banks. Below cell lines (item number; vendor) were used: PL21 (ACC-536; DSMZ) and HL60 (CCL-240; ATCC).

Human blood was obtained from Biological Specialty Corporation (225-11-04). PBMCs were isolated by density gradient centrifugation, and NK cells were purified using negative depletion with magnetic beads (StemCell CAT#17955). The manufacturer's instructions were followed for purification of NK cells from PBMCs. Frozen NK cells were purchased from BioIVTHUMAN-HL65-U-200429 CD14+ monocytes were obtained from BioIVT (HMN370652, HMN370653, HMN370654) Frozen PBMC were purified in-house (see above) and frozen for later use.

Cell Culture Materials

Below items (vendor; item number) were used: RPMI-1640 (Corning; 10040CV), DMEM (Corning; 10031CV), Heat inactivated FBS (ThermoFisher; 16140-071), GlutaMax I (ThermoFisher; 35050-061), Penn/Strep (ThermoFisher; 151450-122), 2-Mercaptoethanol (MP Biomedicals; 02194705), IL-2 (PeproTech; 200-02), CFSE (ThermoFisher; C34554), Cell Trace Violet (ThermoFisher; C34557), rhM-CSF (R&D Systems; 216-MC/CF), TrypLE (ThermoFisher; 12604-013), BFA (Biolegend; 420601), Monensin (BioLegend; 420701), and Human serum (Sigma Aldrich; 51764-5X1ML).

Assay Materials

Below items (vendor; item number) were used: Hepes buffered saline (Alfa Aesar; J67502), BATDA reagent (Perkin Elmer; C136-100), TC round bottom 96-well plate (Corning; 3799), Europium solution (Perkin Elmer; C135-100), DELFIA yellow microplates 96-well (Perkin Elmer; AAAND-0001), Fixation buffer (BioLegend; 420801), Zombie NIR (BioLegend; 77184), and Anti-WIC class I clone W6/32 (Invitrogen; HLA-C1).

Below test articles (description) were used: AB0237 (CLEC12A-Targeted TriNKET with Fab NKG2D binding (A49), and scFv CLEC12A binding domain (1602)), AB0300 (AB0237 with LALAPG mutation in CH2 domain), AB0037 mAb (Humanized 1602 anti-CLEC12A mAb), AB0089 (CLEC12A-Targeted TriNKET with Fab NKG2D binding (A49) and scFv CLEC12A binding domain (1739)), AB0305 mAb (Humanized 1739 anti-CLEC12A mAb), AB0192 (CLEC12A-Targeted TriNKET with Fab NKG2D binding (A49), and scFv CLEC12A binding domain (1797)), AB0306 mAb (Humanized 1797 anti-CLEC12A mAb), and F3′-palivizumab (Non-Targeting TriNKET with Fab NKG2D binding (A49), and scFv binding domain from palivizumab).

Below instruments (serial number) were used: Attune NxT (2AFC228271119), BD Celesta (H66034400085), BD Celesta (H66034400160), Spectramax i3x (363704226) and Spectramax i3x (363703638).

Preparation of Reagents

RPMI primary cell culture media was prepared starting with RPMI-1640. 10% HI-FBS, 1× GlutaMax, 133 penn/strep, and 50 μM 2-mercaptoethanol were added for complete media. RPMI cell culture media was prepared starting with RPMI-1640. 10% HI-FBS, 1× GlutaMax, and 1× penn/strep, were added for complete media. 1× HBS was prepared by diluting 1 part of 5× HBS into 4 parts of DI water. Solution was sterile filtered before use.

DELFIA Cytotoxicity Assay

Labeling Target Cells:

Briefly, CLEC12A+ target cells were pelleted, and washed with 1× HBS. Target cells were resuspended in pre-warmed RPMI primary cell culture media at 10⁶cells/mL. BATDA reagent (bis(acetoxymethyl) 2,2′:6′,2″-terpyridine-6,6″-dicarboxylate) was diluted 1:400 into the cell suspension. Cells were mixed and incubated at 37° C. with 5% CO₂for 15 minutes. The labeled target cells were washed 3× with 1× HBS and resuspended at 5×10⁴cells/mL in RPMI primary cell culture media.

Preparing the Assay Plate:

Rested human NK cells were removed from culture and pelleted, the cells were resuspended in RPMI primary cell culture media at 0.5×10⁶cells/mL. 4× test articles were prepared in RPMI primary cell culture media. In a round bottom TC 96-well plate, 100 μl of labeled target cells, 50 μl of 4× TriNKET/mAb, and 50 μl of effector cells were added. Control wells for background were prepared by pelleting labeled target cells, and 100 μl of the supernatant was added to background wells, containing 100 μl of RPMI primary cell culture media. Spontaneous release wells were prepared by adding 100 μl of labeled target cells to wells containing 100 μl of RPMI primary cell culture media. Maximum release wells were prepared by adding 100 μl of labeled target cells to wells containing 80 μl of RPMI primary cell culture media and 20 μl of 10% TritonX-100 solution. The assay plate was incubated at 37° C. with 5% CO₂for 2-3 hours.

Reading the Assay Plate:

The assay plate was removed from the incubator and the plate was centrifuged to pellet cells. 20 μl of supernatant was removed from each well and transferred to a clean 96-well yellow DELFIA assay plate. 200 μl of room temperature Europium solution was added to each well and the plate was placed on a plate shaker 15 minutes at 250 RPM. The plate was read using an HTRF cartridge in a SpectraMax i3x with the following PMT and optics settings:

Integration time 0.4 ms
Excitation time 0.05 ms
Number of pulses 100
Measurement delay 0.25 ms
Read height 2.36 mm

Data Analysis:

The mean of the background samples was calculated and subtracted from all samples. Specific lysis was calculated as follows:

% Specific lysis=(sample−spontaneous)/(max−spontaneous)*100%

Data were fit to a four-parameter non-linear regression model using Prism GraphPad 7.0 or 8.0.

IFNγ and CD107a Activation Assays

Plate Setup:

Human cancer cell lines expressing CLEC12A were harvested from culture, and cells were adjusted to 2×10⁶cells/mL. Test articles were diluted in culture media. Rested NK cells or PBMCs were harvested from culture, cells were washed, and were resuspended at 2-4×10⁶cells/mL in culture media. IL-2 and fluorophore conjugated anti-CD107a were added to the NK cells or PBMCs for the activation culture. Brefeldin-A (BFA) and monensin were diluted into culture media to block protein transport out of the cell for intracellular cytokine staining. Into a 96-well plate, 50 μl of tumor targets, mAbs/TriNKETs, BFA/monensin, and NK cells were added for a total culture volume of 200 μl. Plate was cultured for 4 hours before samples were prepared for FACS analysis.

Preparation of Samples for FACS:

Following the 4-hour activation culture, cells were prepared for analysis by flow cytometry using fluorophore conjugated antibodies listed in Table 189. CD107a and IFNγ staining was analyzed in CD3-CD56+ populations to assess NK cell activation.

TABLE 18

FACS NK activation panel

Dilution

Fluorophore
Target
factor

BV421
IFNy
50

BV510
CD8
100

BV605
dump

BV785
CD3
100

PE-D594
CD56
100

PE-Cy7
CD4
100

APC
CD107a
50

A700
CD45
50

NIR
live-dead
1000

TruStain FcX
50

BV buffer
50

mono block
50

Cells of interest were identified using FSC vs. SSC plot, and an appropriately shaped gate was drawn around the cells. Within the gated cells, doublet cells were removed by viewing FSC-H vs FSC-A plot. Within the single cell population, live cells were gated. Within live cells, NK cells were identified as CD56+CD3−. CD107a degranulation and intracellular (IC) IFNγ were analyzed within the NK cell population.

ADCP Assays

Primary human CD14+ monocytes are obtained from BioIVT. Monocytes were isolated from peripheral blood via negative selection yielding >93% purity. Monocytes were cultured for 6 days in 50 ng/ml rhM-CSF (R&D systems), replenishing rhM-CSF every two days. After 6 days, macrophages were detached from the culture flask(s) with TrypLE Express (Gibco) and labeled with Celltrace Violet (2 μM, ThermoFisher). Target cells were labeled with CFSE (2 μM, ThermoFisher). In round-bottom 96-well plates, target cells (1×10⁴cells/well) are first incubated with the indicated concentrations of each test article for 30 minutes. Labeled macrophages (8×10⁴cells/well) are then added to the appropriate wells and plates were incubated for an additional 2 hours. Cells were stained with Zombie NIR Dye (Biolegend) for live dead discrimination and fixed. Phagocytosis is quantified by Celesta HTS (BD Biosciences), with data analysis in FlowJo 10.5.3. Data were gated on single, live cells, and % phagocytosis was calculated as follows: % phagocytosis=100×(count CFSE+, Celltrace Violet+ cells)/(total count CFSE+ cells). Data were graphed and analyzed using GraphPad Prism.

CDC Assays

CDC was measured in a DELFIA cytotoxicity assay. PL21 cancer cell targets were labeled with BATDA labeling reagent. Labeled target cells were then incubated with the indicated concentrations of TriNKETs or mAbs for 30 min, to allow for opsonization. Human serum was added to a 5% final concentration. The plate was then incubated for 60 minutes before termination of assay. 20 μl of culture supernatant was transferred to DELFIA yellow plate. 200 μl europium solution was added per well and incubated at room temperature for 15 minutes with shaking in the dark. Fluorescence was then measured on the Spectramax plate reader.

Results and Discussion

Rested Human NK Cell Lysis of CLEC12A+ Target Cells

The activities of AB0089 and AB0192 was characterized using rested primary human NK cells from healthy donors in co-culture with either human CLEC12A+ AML line PL21 or HL60. AB0089 and AB0192 were compared to their parental hIgG1 mAbs AB0305 and AB0306, respectively. FIG. 194 shows a representative plot of PL21 leukemia target cell lysis by rested human NK cells in the presence of AB0089. AB0089 showed dose-dependent cell lysis enhancement activity that was more potent and reached a higher maximum lysis than its parental mAb AB0305. The results of similar cytotoxicity assays using NK cells derived from a total of six donors are summarized in Table 190. Across the assays, the potency AB0089 in enhancing NK cell lysis of PL21 target cells had mean EC50 value 0.25±0.24 nM.

TABLE 190

AB0089 EC50 and % Max lysis values for

rested NK cell lysis of PL21 cells

(Values represent the mean of assays with

a total of six donors standard deviation)

Molecule
EC50 (nM)
% Max lysis

AB0089
0.25 ± 0.24
17 ± 10

AB0305
ND
ND

ND = Not determined

FIG. 195 shows a representative plot of HL60 leukemia target cell lysis by rested primary human NK cells in the presence of AB0089. Again, AB0089 showed more potent dose-dependent lysis and higher specific lysis overall than its parental mAb AB0305. The results of assays using NK cells derived from four healthy donors yielded a mean EC50 of 0.09±0.05 nM as summarized in Table 191.

TABLE 191

AB0089 EC50 and % Max lysis values for

rested NK cells lysis of HL60 cells

(Values represent the mean of assays with a

total of four donors ± the standard deviation)

EC50
%

Molecule
(nM)
Max lysis

AB0089
0.09 ± 0.05
47 ± 14

AB0305
ND
ND

ND = Not determined

FIG. 196 shows a representative plot of PL21 leukemia target cell lysis by rested human NK cells in the presence of AB0192. AB0192 showed dose-dependent cell lysis enhancement activity that was more potent and reached a higher maximum lysis than its parental mAb AB0306. The results of similar cytotoxicity assays using NK cells derived from a total of six healthy donors are summarized in Table 192. Across the assays, the potency AB0192 in enhancing NK cell lysis of PL21 target cells had mean EC50 value 0.61±0.60 nM.

TABLE 192

AB0192 EC50 and % Max lysis values for

rested NK cell lysis of PL21 cells

(Values represent the mean of assays with a

total of six donors ± the standard deviation)

EC50
%

Molecule
(nM)
Max lysis

AB0192
0.61 ± 0.60
18 ± 13

AB0306
ND
ND

ND = Not determined

FIG. 197 shows a representative plot of HL60 leukemia target cell lysis by rested primary human NK cells in the presence of AB0192. Again, AB0192 showed more potent dose-dependent lysis and higher specific lysis overall than its parental mAb. The results of assays using NK cells derived from four healthy donors yielded a mean EC50 of 0.30±0.27 nM as summarized in Table 193.

TABLE 193

AB0192 EC50 and % Max lysis values for

rested NK cells lysis of HL60 cells

(Values represent the mean of assays with

a total of four donors ± the standard deviation)

Molecule
EC50 (nM)
% Max lysis

AB0192
0.30 ± 0.27
50 ± 14

AB0306
ND
ND

ND = Not determined

AB0089 Activity in the Presence of Soluble MIC-A

We also tested the activity of AB0089 in the presence of soluble NKG2D ligand. For these assays, we selected to use a recombinant version of the NKG2D ligand MIC-A. MIC-A has broad expression across cancer indications and is known to shed from the cell surface resulting in accumulation in patient serum. Soluble MIC-A-Fc was added to a NK cell cytolysis assay system at 20 ng/mL, a physiologically relevant serum concentration found in cancer patients. FIG. 198 shows the dose-response curves of AB0089 and its parental mAb AB0305 in a primary NK cell cytolysis assay against PL21 target cells in the absence and presence of soluble MIC-A. The addition of MIC-A had no effect on the potency or maximum lysis achieved by AB0089 (see Table 194 for EC50 and max lysis values).

TABLE 194

AB0089 EC50 and % Max lysis values for lysis

of PL21 cells by NK cells with sMIC-A-Fc

EC50
%

Molecule
(nM)
Max lysis

AB0089
0.32
15

AB0089 + sMIC-A-Fc
0.26
13

AB0305
ND
ND

AB0305 + sMIC-A-Fc
ND
ND

ND = Not determined

AB0089 Stimulation of NK Cell IFNγ Production and Degranulation

In addition to direct lysis of target cells, NK cells also produce cytokines upon activation. Thus, we wanted to examine alternative readouts of NK activation in addition to target cell lysis. We chose to analyze IFNγ production and CD107a degranulation from NK cells co-cultured with CLEC12A+ AML cells in the presence of AB0089. Rested NK cells within PBMCs in co-culture with PL21 and HL60 target cells showed little basal induction of CD107a degranulation or intracellular IFNγ accumulation after four hours (FIG. 199 and FIG. 200, respectively). Addition of AB0089 to the co-cultures resulted in robust induction of degranulation and IFNγ production in a dose-responsive manner. In contrast, neither parental mAb AB0305 nor non-CLEC12A-targeting TriNKET F3′-palivizumab showed a robust increase in CD107a+IFNγ+ NK cells. Assays were performed with three independent PBMC donors in co-cultures with PL21 or HL60 target cells; the results are summarized in Tables 195 and 196, respectively.

TABLE 195

Summary of NK cell induction of IFNγ and

CD107a in co-culture with PL21 cells

(Values represent the mean of assays with a

total of three donors ± the standard deviation)

Molecule
EC50 (nM)
% Max

AB0089
0.08 ± 0.05
15.3 ± 4.0

AB0305
ND
ND

F3′-palivizumab
ND
ND

ND = Not determined

TABLE 196

Summary of NK cell induction of IFNγ and

CD107a in co-culture with HL60 cells

(Values represent the mean of assays with a

total of three donors ± the standard deviation)

Molecule
EC50 (nM)
% Max

AB0089
0.05 ± 0.03
15.0 ± 4.6

AB0305
ND
ND

F3′-palivizumab
ND
ND

ND = Not determined

AB0089 Induces ADCP Activity

The Fc domain of a human IgG1 antibody can mediate three different types of effector functions.

One type of Fc-mediated effector function is ADCC, which is carried out by engagement of CD16 on NK cells; NK cell stimulation has been extensively characterized for AB0089. A second Fc-mediated effector function is ADCP, where macrophages attack and engulf cells coated with antibody. For AB0089 and alternative CLEC12A TriNKETs, we wanted to assess their ability to induce ADCP of opsonized target cells. We utilized an in vitro assay system with M0 macrophages, derived from culturing purified CD14+ monocytes with M-CSF, as effector cells. CLEC12A+ target cells were labeled with Cell Trace CFSE dye, opsonized with test articles and co-cultured with Cell Trace Violet-labeled M0 macrophages. Phagocytosis was analyzed by flow cytometry as Cell Trace Violet+ Cell Trace CFSE+ (double-positive) events.

AB0089 and other CLEC12A TriNKETs AB0237 and AB0192 all enhanced phagocytosis of PL21 AML cells by M0 macrophages. Parental mAbs for each TriNKET also all showed an ability to induce M0 macrophage phagocytosis of opsonized target cells, but to a lower maximum level compared to their respective corresponding TriNKET (FIG. 201). AB0237si, which bears mutations in the CH2 domain to silence Fcγ-receptor binding, served as a negative control. As expected, AB0237si failed to mediate ADCP of opsonized target cells. Results using M0 macrophages derived from three different donors are summarized in Table 197.

TABLE 197

Summary of EC50 and % Max values for ADCP activity

(Values represent the mean of assays with a total

of three donors ± the standard deviation.)

Molecule
EC50 (nM)
% Max

AB0237si
ND
ND

AB0037
0.05 ± 0.06
11 ± 9

AB0237
0.08 ± 0.07
12 ± 8

AB0305
0.02 ± 0.01
10 ± 7

AB0089
0.03 ± 0.02
15 ± 10

AB0306
ND
ND

AB0192
0.03 ± 0.02
15 ± 7

ND = Not determined

AB0089 does not have CDC Activity

A third effector function of human IgG1 isotype antibodies is initiation of the complement cascade, leading to complement dependent cytotoxicity (CDC). AB0089 is constructed using a human IgG1 Fc domain; therefore, we wanted to understand the ability of AB0089 to stimulate CDC activity. We used PL21 cells as targets and 5% human serum as a source of complement factors. Neither AB0089 nor its parental mAb AB0305 stimulated complement-mediated killing of PL21 AML cells (FIG. 202). In contrast, a positive control antibody against MHC class I (clone W6/32) showed dose-dependent lysis of PL21 target cells in the presence of human serum, corroborating that the serum has active complement factors. Cell characterization analysis indicated similar expression levels of CLEC12A and MHC class I on PL21 target cells.

AB0089 demonstrated potent activity in a variety of physiologically relevant contexts. AB0089 was first tested in CLEC12A+ target cell killing assays using rested primary human NK cells. We chose to test killing of CLEC12A+ target cells derived from AML, being the indication that overexpresses CLEC12A in the clinic. For these experiments, we selected HL60 and PL21 as representative AML cell lines.

AB0089 enhanced primary NK cell cytolysis of both target cell lines and outperformed its parental hIgG1 antibody in enhancing cytolysis. We also examined the impact of soluble NKG2D ligand on the activity of AB0089 and found that physiological levels of soluble MIC-A-Fc did not alter the activity of AB0089 in cell lysis.

In addition to cell lysis, AB0089 also stimulated cytokine production from NK cells co-cultured with CLEC12A+ target cells. AB0089 enhanced intracellular accumulation of IFNγ as well as NK cell degranulation in co-cultures of human PBMCs with PL21 and HL60 AML cells. EC50 values for IFNγ and CD107a induction by AB0089 were consistent with EC50 values generated in cell lysis assays.

Finally, we investigated additional mechanisms of action for AB0089. AB0089 was demonstrated to engage macrophages to mediate ADCP of AB0089-opsonized target cells. In contrast, AB0089 does not mediate CDC of AB0089-opsonized target cells.

AB0089 was demonstrated to have potent activity against CLEC12A+ cancer cells through various effectors, including NK cells and macrophages. Moreover, the activity of AB0089 was maintained in the presence of a physiologically-relevant concentration of soluble NKG2D ligand MIC-A.

Example 19
AB0089 Binding to CLEC12A+ Cell Lines and Primary AML
Objectives

Measure the binding of AB0089 to human CLEC12A-expressing cells. Quantify CLEC12A surface retention following incubation with AB0089. Demonstrate binding of AB0089 to primary AML patient samples and compare to CLEC12A expression pattern.

Study Design

Representative cell lines with positive CLEC12A surface expression were used for characterization of AB0089 cell binding. Primary AML patient samples were obtained from Creative Bioarray and BioIVT. Binding of AB0089 and parental mAb AB0305 were compared to a commercially available CLEC12A antibody.

Materials and Methods

Primary Cells and Cell Lines

Human cancer cell lines were obtained from research cell banks. Below cell lines (item number; vendor) were used: PL21 (ACC-536; DSMZ), HL60 (CCL-240; ATCC), Ba/F3-hCLEC12A (made in-house) and Ba/F3 (available from DSMZ ACC-300).

Human CLEC12A was introduced into Ba/F3 parental cells by retroviral transduction.

Cell Culture Materials

Assay Materials

Below items (vendor; item number) were used: Hepes buffered saline (Alfa Aesar; J67502), BATDA reagent (Perkin Elmer; C136-100), TC round bottom 96-well plate (Corning; 3799), Europium solution (Perkin Elmer; C135-100), and DELFIA yellow microplates 96-well (Perkin Elmer; AAAND-0001).

Below test articles (description) were used: AB0237 (CLEC12A-Targeted TriNKET with Fab NKG2D binding (A49), and scFv CLEC12A binding domain (1602)), AB0300 (AB0237 with LALAPG mutation in CH2 domain), AB0237-Dead-2D (AB0237, NKG2D binding domain mutated to ablate binding), AB0037 mAb (Humanized 1602 anti-CLEC12A mAb), AB0089 (CLEC12A-Targeted TriNKET with Fab NKG2D binding (A49) and scFv CLEC12A binding domain (1739)), AB0305 mAb (Humanized 1739 anti-CLEC12A mAb), AB0192 (CLEC12A-Targeted TriNKET with Fab NKG2D binding (A49), and scFv CLEC12A binding domain (1797)), AB0306 mAb (Humanized 1797 anti-CLEC12A mAb), hcFAE-A49.h6E7 (CLEC12A-Targeted TriNKET with Fab NKG2D binding (A49), and Fab CLEC12A binding domain (Genentech 6E7)) and hcFAE-A49.CLL1-Merus (CLEC12A-Targeted TriNKET with Fab NKG2D binding (A49), and Fab CLEC12A binding domain (Merus CLL-1 binder)).

Below instruments (serial number) were used: Attune NxT (2AFC228271119), BD Celesta (H66034400085), and BD Celesta (H66034400160).

Preparation of Reagents

RPMI primary cell culture media was prepared starting with RPMI-1640. 10% HI-FBS, 1× GlutaMax, 1× penn/strep, and 50 μM 2-mercaptoethanol were added for complete media. RPMI cell culture media was prepared starting with RPMI-1640. 10% HI-FBS, 1× GlutaMax, and 1× penn/strep, were added for complete media. 1× HBS was prepared by diluting 1 part of 5× HBS into 4 parts of DI water. Solution was sterile filtered before use.

Binding and Detection of Surface Bound mAb/TriNKET

100,000 cells were seeded per well into a 96-well plate for FACS staining. Cells were washed with PBS. Cells were incubated in a 1:2000 dilution of live/dead dye in PBS for 15 minutes, then washed with FACS buffer. TriNKETs or mAbs were diluted into FACS buffer, and 50 μl containing diluted TriNKET or mAb was added to cells. Cells were incubated on ice for 20 minutes. After the incubation, cells were washed with FACS buffer. Anti-human IgG-Fc secondary antibody was diluted into FACS buffer, and 50 μl was added per well for detection of the bound TriNKETs or mAbs. Cells were incubated for 20 minutes on ice, then washed with FACS buffer. 50 μl of fixation buffer was added to each well, and cells were incubated for 10 minutes at room temperature. Cells were washed with FACS buffer and resuspended in FACS buffer for analysis with BD FACS Celesta SN #H66034400085, or BD FACS Celesta SN #H66034400160. Cells of interest were identified using FSC vs. SSC plot, and an appropriately shaped gate was drawn around the cells. Within the gated cells, doublet cells were removed by viewing FSC-H vs FSC-A plot. Within the single cell population, live cells were gated. Within the live gate, the mean fluorescence intensity (MFI) of each sample and the secondary-only control was calculated. Fold-over-background (FOB) was calculated as the ratio of test article MFI over secondary-only background MFI. Data were fit to a four-parameter non-linear regression curve using GraphPad Prism 7.0.

Surface Retention Assays

100,000 cells were seeded per well into duplicate 96-well plates for FACS staining. Cells were washed with FACS buffer. Test articles were diluted to 40 nM and were added to each of the duplicate plates. The first plate was incubated on ice for 2 hours, while the second plate was moved to 37° C. for two hours. After the incubation, cells were washed with PBS and cells were incubated in a 1:2000 dilution of live/dead dye in PBS for 15 minutes, then washed with FACS buffer. Anti-human IgG-Fc secondary antibody was diluted into FACS buffer, and 50 μl was added per well for detection of the bound TriNKETs or mAbs. Cells were incubated for 20 minutes on ice, then washed with FACS buffer. 50 μl of fixation buffer was added to each well, and cells were incubated for 10 minutes at room temperature. Cells were washed with FACS buffer and resuspended in FACS buffer for analysis with BD FACS Celesta SN #H66034400085, or BD FACS Celesta SN #H66034400160. Cells of interest were identified using FSC vs. SSC plot, and an appropriately shaped gate was drawn around the cells. Within the gated cells, doublet cells were removed by viewing FSC-H vs FSC-A plot. Within the single cell population, live cells were gated. Within the live gate, the MFI of each sample was calculated. hIgG1 isotype control MFI was subtracted from all test article MFIs. Surface retention was then calculated as follows:

% surface retention=(2-hour 37° C. MFI/2-hour ice MFI)*100

Analysis of AB0089 Binding in Primary AML Samples

Frozen AML patient samples were thawed at 37° C. and used immediately in binding assays. Test articles, hIgG1 isotype control and commercial reagent anti-CLEC12A antibody clone 50C1 were labeled using the Biotium Mix-n-Stain CF568 labeling kit; manufacturer's instructions were followed. AML samples were stained with a cocktail of antibodies as described in Tables 198 and 199. Samples were analyzed using an Attune NxT cytometer equipped with CytKick Max autosampler. Raw data was analyzed using FlowJo X software.

TABLE 198

AML patient sample flow cytometry panel

Dilution

Fluorophore
Antigen
factor
Vendor
Cat #

BV421
CD38
100
BioLegend
368514

BV510
CD33
100
BioLegend
366610

BV605
CD15
100
BioLegend
323032

BV650
CD14
100
BioLegend
301836

BV785
CD45
50
BioLegend
368528

FITC
CD34
50
StemCell
60013F1

PE or CF568
CLEC12A

See Table 200

PE-Cy7
CD117
100
BioLegend
313212

APC
CD13
100
BioLegend
301706

NIR
live-dead
1000
BioLegend
423105

TruStain FcX
25
BioLegend
422302

BV buffer
25
BD
566349

mono block
25
BioLegend
426103

TABLE 199

CLEC12A antibodies and labeled test articles

Molecule
lot #

hIgG1
EXP20000921-RMR 18 Mar. 2020

AB0305
EXP200001542-RMR 22 May 2020

AB0089
EXP200001542-RMR 22 May 2020

AB0306
EXP200001542-RMR 22 May 2020

AB0192
EXP200001542-RMR 22 May 2020

50C1
EXP200001542-RMR 22 May 2020

TABLE 200

Primary AML patient samples

AML

Sample ID
Tissue
subtype
Vendor

19217S
PBMC
M3
Creative Bioarray

19224S
PBMC
M1
Creative Bioarray

19233S
PBMC
M1
Creative Bioarray

19246S
PBMC
M5a
Creative Bioarray

19258S
Bone marrow
M5b
Creative Bioarray

Y1724
PBMC
M4
Creative Bioarray

W5299
PBMC
M2
Creative Bioarray

W5300
PBMC
M3
Creative Bioarray

W5313
Bone marrow
M1
Creative Bioarray

W5294
PBMC
M0
Creative Bioarray

Results and Discussion

CLEC12A TriNKETs Bind Cell Lines Expressing Human CLEC12A

Binding of CLEC12A TriNKETs and their parental mAbs were tested using three cell lines expressing human CLEC12A (FIGS. 203A-203C). Mouse Ba/F3 cells were engineered to express human CLEC12A, while PL21 and HL60 are human AML cell lines with endogenous CLEC12A expression.

As comparators, two tool TriNKETs were generated using CLEC12A binding domains from Genentech (h6E7) and Merus (CLL1-Merus) known in the public domain. Duobody TriNKETs with Fab CLEC12A binding domains were paired with Fab NKG2D binder A49. Duobody TriNKET hcFAE-A49.h6E7 is derived from the Genentech binder, while hcFAE-A49.CLL1-Merus is derived from the Merus binder.

Lead TriNKET AB0089 and alternative CLEC12A TriNKETs AB0237 and AB0192 all bound CLEC12A expressing cells with nanomolar potency (Table 4). AB0089 had reduced potency but reached a higher maximum FOB compared to its parental mAb AB0305 on the three cell lines tested. AB0237 demonstrated comparable binding to its parental mAb AB0037, yielding potency values within 2-fold of each other for all three cell lines. Variants of AB0237 AB0237-Dead 2D and AB0300 bearing mutations in the NKG2D and CD16 binding domains, respectively, showed similar cell binding compared to AB0237. In contrast, AB0192 showed decreased potency and maximum FOB compared to its parental mAb AB0306.

In comparison to tool TriNKETs, AB0089 showed comparable cell loading to hcFAE-A49.CLL1-Merus and higher cell loading than hcFAE-A40.h6E7. In addition, AB0237 binding CLEC12A cell lines was more potent than hcFAE-A49.CLL1-Merus and comparable in potency to hcFAE-h6E7. All EC50 binding potencies and maximum FOB binding are summarized in Table 201.

TABLE 201

Summary of cell binding values

Average EC50 values in nM and mean FOB

with ± standard deviation from three independent experiments.

HL60
PL21
Ba/F3-hCLEC12A

EC50 nM
FOB
EC50 nM
FOB
EC50 nM
FOB

AB0089
1.08 ± 0.24
70 ± 28
0.34 ± 0.13
16 ± 8
2.28 ± 0.59
434 ± 258

AB0305
0.12 ± 0.01
53 ± 16
0.12 ± 0.01
15 ± 9
0.67 ± 0.20
279 ± 182

AB0192
0.61 ± 0.15
42 ± 14
0.37 ± 0.08
12 ± 5
1.33 ± 0.41
325 ± 242

AB0306
0.26 ± 0.08
47 ± 10
0.18 ± 0.05
12 ± 5
0.68 ± 0.09
303 ± 205

AB0237
1.01 ± 0.14
66 ± 18
0.52 ± 0.09
15 ± 6
1.54 ± 0.35
341 ± 249

AB0037
0.52 ± 0.12
51 ± 14
0.52 ± 0.09
13 ± 7
1.40 ± 0.24
301 ± 173

AB0237 Dead-2D
1.19 ± 0.15
59 ± 14
0.63 ± 0.14
14 ± 6
2.35 ± 0.62
357 ± 233

AB0300
1.50 ± 0.19
65 ± 20
1.58 ± 0.14
18 ± 6
2.92 ± 0.57
368 ± 249

hcFAE-A49.h6E7
1.33 ± 0.22
31 ± 10
0.55 ± N/A
10 ± 5
2.87 ± 0.56
318 ± 212

hcFAE-A49.CLL1-Merus
2.79 ± 1.04
59 ± 24
1.13 ± 0.14
15 ± 10
4.45 ± 0.78
369 ± 186

CLEC12A-TriNKET Surface Retention

Internalization of CLEC12A following antibody ligation has been described in the literature. Leveraging on this phenomenon, antibody-drug conjugates targeting CLEC12A have been developed that depend on internalization of conjugated drugs to kill target cells. TriNKET activity, however, requires cell surface retention of CLEC12A in complex with TriNKETs to allow for NK cell engagement and effector activity against CLEC12A+ cancer cells. To better understand CLEC12A retention on cell surfaces in the presence of CLEC12A TriNKETs, we examined the binding of saturating concentrations of each TriNKET and parental mAb to CLEC12A+ cell lines after a two-hour incubation at 37° C. compared to incubations on ice.

When bound to cells for 2 hours at 37° C., 90% or more of cell surface CLEC12A was retained with each of the TriNKETs. FIGS. 204A-204B show examples of surface retention for AB0089, AB0192, AB0237 and their parental antibodies on HL60 and PL21 AML lines. The data from three independent experiments are summarized in Table 202. AB0089 appeared to demonstrate superior surface retention compared to other CLEC12A TriNKETs and their parental mAbs.

However, AB0089 showed a lower binding signal after incubation on ice compared to AB0237 but had a similar signal after incubation at 37° C. for two hours, which may have led to inflated surface retention values for AB0089 (FIG. 205). Nevertheless, these data suggest temperature-dependent binding of AB0089 to reach equilibrium, which could result from two-step binding mechanism suggested by kinetic binding data (Example 17).

By contrast, AB0192 and its parental mAb AB0306, which originates from a different parental murine mAb, showed similar surface retention on HL60 and PL21 cells. These data indicate that surface retention of CLEC12A may be dependent upon whether bound monovalently (as is the case for F3′ TriNKETs) versus bivalently (as mAbs do) as well as the epitope of the specific binder.

TABLE 202

Summary of CLEC12A cell surface retention

Values indicate the average percent surface

retention for a given test article after two

hours at 37° C. relative to the same test

article after two hours on ice in three

experiments ± standard deviation.

HL60
PL21

AB0089
147 ± 13
124 ± 12

AB0305
90 ± 7
92 ± 2

AB0192
103 ± 8
101 ± 10

AB0306
114 ± 4
106 ± 9

AB0237
106 ± 4
91 ± 9

AB0037
74 ± 1
79 ± 3

AB0089 Binds CLEC12A on Human AML Patient Samples

Primary bone marrow and PBMC samples from a total of 10 AML patients were obtained (see Table 200 for sample information) to examine binding of AB0089 and its parental mAb AB0305 on primary AML cells. We compared binding patterns of fluorophore-labeled test articles AB0089 and AB0305 to a commercially available anti-CLEC12A antibody clone 50C1 (Lahoud et al, The Journal of Immunology. 2009; 182:7587-7594). Among the 10 donors, the percentage of CLEC12A+ cells in the AML blast gate varied widely from ˜40-95%. We analyzed AML patient samples from different FAB subtypes, which might contribute to the variation in CLEC12A expression and proportion observed.

AB0089 and AB0305 demonstrate a comparable binding pattern to anti-CLEC12A clone 50C1 across the 10 AML patient samples tested (FIGS. 206A-206J). We further gated blasts to examine the CD34+CD38− subpopulation, which is enriched for leukemic stem cells, from each sample. Again, AB0089 and AB0305 had similar staining on CD34+CD38− blasts as clone 50C1 (FIGS. 207A-207J). Overall, these data corroborated that AB0089 is able to bind CLEC12A on primary AML blasts as well as on the CD34+CD38− subset that is enriched for LSCs.

AB0089 and other CLEC12A TriNKETs bound to Ba/F3 cells ectopically expressing human CLEC12A as well as PL21 and HL60 AML cells expressing endogenous CLEC12A in a dose-responsive manner with similar nanomolar potencies across the lines. Despite reports in the literature of CLEC12A internalization following antibody ligation, AB0089 and both alternative CLEC12A TriNKETs AB0237 and AB0192 showed high cell surface retention of CLEC12A after two hours of incubation with HL60 and PL21 CLEC12A+ cell lines. Finally, AB0089 and its parental mAb AB0305 bound similarly to a commercial reagent anti-CLEC12A antibody clone 50C1 on bulk and CD34+CD38− blasts across primary AML patient samples from 10 donors. Together, these data support potent binding of AB0089 to cells expressing CLEC12A and indicates that AB0089 binds to primary AML blasts.

Example 20
AB0089 CLEC12A TrinKET
Objectives

The objectives of this study are to describe the molecular format, design, structure, and characteristics of CLEC12A TriNKET AB0192. In particular, in this report we a) describe the molecular format and design of AB0192, b) provide information about expression and purification of AB0192, c) describe basic biochemical and biophysical characterization of the molecule, d) determine the affinity of AB0192 for recombinant human targets (CLEC12A, NKG2D and CD16a), e) demonstrate AB0192 binding to isogenic and AML cancer cell lines expressing human CLEC12A, f) demonstrate selectivity of AB0192, g) determine potency of AB0192 in killing AML cancer cells, h) assess binding epitope, i) assess binding to FcγR and FcRn and compare with IgG1 mAb control, j) describe behavior of AB0192 in developability studies including accelerated stability, forced degradation, and manufacturability.

Study Design

Murine anti-human CLEC12A mAb (9F11.B7) was discovered by mouse immunization. 9F11.B7 was humanized, converted to a scFv and combined with A49M-I NKG2D Fab binding arm to produce an F3′-TriNKET (AB0192). AB0192 was assessed by binding to recombinant human and cyno CLEC12A, binding to isogenic and cancer cells expressing CLEC12A as well as potency of the molecule against CLEC12A+ AML cancer cells. The biophysical and biochemical characteristics of AB0192 including thermal stability (DSC), hydrophobicity (HIC), and pI (cIEF) were determined. Specificity of binding was assessed by PSR, binding to cells not expressing the target of interest and by HuProt™ Array. AB0192 was expressed in two different mammalian cell lines (Expi293 and ExpiCHO) and purified using DragonflyTx 2-step platform purification method. Binding properties of AB0192, including CLEC12A, NKG2D, CD16a (V158 and F158), an extended panel of Fcγ receptors (CD32a (H131 and R131), CD16b, CD32b, cynomolgus CD16, human and cynomolgus CD64), and FcRn (human and cynomolgus) was fully characterized. Binding epitope was characterized in binning experiment with CLEC12A binding molecules from Merus and Genentech. Lastly, AB0192 was assessed in a full developability panel, including accelerated (40° C.) stability in the Dragonfly Tx platform formulation (20 mM histidine, 250 mM sucrose, 0.01% Tween-80, pH 6.0) and 20 mM citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 buffer, forced degradation (long term pH 5, pH 8 stress, forced oxidation), and manufacturability (freeze/thaw, agitation, low pH hold).

Materials and Methods

Methods used to produce and characterize AB0192 are similar to the ones used in Example 17.

Test Articles

AB0192-002, (Benchling record name AB0192), humanized TriNKET expressed in Expi293 (lot AB0192-002).

AB0192-005, (Benchling record name AB0192), humanized TriNKET expressed in ExpiCHO (lot AB0192-005).

Protein Reagents

TABLE 203

Summary of protein reagents.

Name
Manufacturer
Lot

Human CLEC12A-His
Dragonfly Tx
16 Oct. 2018 GL

Human CLEC12A-K244Q-His
Dragonfly Tx
30 Apr. 2018 AC

Cyno CLEC12A-His
Dragonfly Tx
30 May 2018 AC

mFc-hNKG2D
Dragonfly Tx
1 Dec. 2017 AB

mFc-cNKG2D
Dragonfly Tx
28 Sep. 2018 GL

humanized mAb 16B8.C8
Dragonfly Tx
AB0037-001

hcFAE-A49.CLL1-Merus
Dragonfly Tx
10 May 2018 DF

(Duobody TriNKET)

hcFAE-A49.h6E7
Dragonfly Tx
5 Feb. 2019 DF

(Duobody TriNKET)

hcFAE-A49.tepoditamab
Dragonfly Tx
27 Aug. 2019 XL

(Duobody TriNKET)

Trastuzumab
Roche
N3017H05

Adalimumab
AbbVie
83364XH07

Rituximab
Dragonfly Tx
19 Apr. 2019 MH

Results

Discovery of AB0192 TriNKET

Hybridoma subclone 9F11.B7 was discovered through immunization of BALB/c mouse strain with recombinant human CLEC12A-his preformed at Green Mountain Antibodies (Burlington, Vt.).

Murine 9F11.B7 was humanized by grafting murine CDRs to the appropriate human framework (VH3-30*02 and VK1-33) without loss of affinity for recombinant CLEC12A or cell surface expressed target (FIGS. 208A-208F). 7 murine backmutations were introduced to maintain CDRs architecture. Humanized 9F11.B7 mAb was converted to a scFv and paired with NKG2D targeting A49 M-I Fab to produce F3′ TriNKET (AB00192). This TriNKET was extensively characterized and designated as a candidate for CLEC12A program AB0192.

Molecular Format and Design

AB0192 is a heterodimeric tri-functional antibody that redirects NK and T cell activity against CLEC12A-expressing cancer cells. AB0192 contains three chains: Chain H, Chain L and Chain S. The schematic representation of the AB0192 molecule is provided in FIG. 209. Chain H is the heavy chain of A49M-I that was selected for having low affinity while being a potent agonist of NKG2D receptors on NK and cytotoxic T cells. The M-I mutation was introduced to replace the Met102 residue in the CDRH3 that is sensitive to oxidation and therefore may present a liability. Chain L is the light chain of the anti-NKG2D-targeting antibody A49M-I. Chain S contains a scFv that binds with high affinity to CLEC12A on AML cancer cells. Both scFv and Fab domains are fused to a human IgG1 Fc that has native, unmodified sequence in the CH2 domain, where antibodies bind to Fc receptors. Both chains bear several mutations in the CH3 domain introduced to ensure stable heterodimerization of two Fc monomers. There are seven such mutations in the IgG1 Fc of the AB0192: three mutations in Chain H (K360E, K409W, Y349C) and four mutations in Chain S (Q347R, D399V, F405T and S354C). Two of these mutations are involved in the formation of a stabilizing S—S bridge (Y349C and S354C) between Chains H and Chain S (EU nomenclature). scFv (VH-VL orientation) in Chain S is stabilized by two mutations introducing a S—S bridge ((H44)C and (L100)C, Chothia nomenclature). Chain S also contains a (G4S)4 non-immunogenic linker between VH and VL of scFv. The amino acid sequence of the CLEC12A-targeting scFv is based on the sequence of humanized 9F11.B7 antibody, obtained from hybridoma (Green Mountain Antibodies, Burlington, Vt.). The amino acid sequence of NKG2D Fab is based on the fully human antibody A49M-I, obtained using human yeast display technology (Adimab, Lebanon, N.H.).

Description of Heterodimerization Mutants

EU-numbering convention was used to annotate the amino acid residues in the Fc.

Chain H contains the following mutations:

- K360E—Heterodimerization mutation in the CH3 domain of the Fc region.
- K409W—Heterodimerization mutation in the CH3 domain of the Fc region.
- Y349C—Inter-subunit disulfide stabilization mutation in the CH3 domain of the Fc region.

The K360E/K409W mutations in the Fc region of Chain H are designed to render heterodimer-favoring interactions with Chain S due to hydrophobic/steric complementarity plus long-range electrostatic interactions (Choi et al., (2013) Mol Cancer Ther 12(12):2748-59; Choi et al., (2015) Mol Immunol 65(2):377-83). The mutation Y349C was introduced to generate an additional inter-chain disulfide bond Y349C-(CH3-ChainH)-S354C(CH3-ChainS) pair (Choi et al., (2015) Mol Immunol 65(2):377-83.).

Chain S contains the following mutations:

- G44C—Disulfide stabilization mutation in VH region of the scFv.
- Q100C—Disulfide stabilization mutation in VL region of the scFv.
- Q347R—Heterodimerization mutation in the CH3 domain of the Fc region.
- D399V—Heterodimerization mutation in the CH3 domain of the Fc region.
- F405T—Heterodimerization mutation in the CH3 domain of the Fc region.
- S354C—Inter-subunit disulfide stabilization mutation in the CH3 domain of the Fc region.

The mutations G(H44)C and Q(L100)C were introduced to stabilize the scFv by an engineered disulfide bond between the VH and the VL. The disulfide stabilization improves structural and thermal stability of scFv. The Q347R/D399V/F405T mutations in the Fc region of Chain S were designed to render heterodimer-favoring interactions with Chain H due to hydrophobic/steric complementarity plus long-range electrostatic interactions (Choi et al, (2013) Mol Cancer Ther 12(12):2748-59). The mutation S354C was introduced to generate an additional inter-CH3 disulfide bond with a Y349C(CH3-ChainH)-S354C(CH3-ChainS) pair (Choi et al., (2015) Mol Immunol 65(2):377-83).

Amino Acid Sequence

The amino acid sequences of the three polypeptide chains that make up AB0192 are shown in FIG. 210. CDRs are outlined in Table 204.

TABLE 204

Sequences of AB0192 CDRs*.

Variable
Frame-

AB0192
region
work
CDR1
CDR2
CDR3

Chain
VH
VH3-30
GFTFNAF
SSGSTS
DGYPTGGA

S

(SEQ ID
(SEQ ID
MDY (SEQ

NO: 195)
NO: 196)
ID NO:

187)

Chain
VL
VK1-33
KASQDIYN
RAN1LVS
LQFDAFPF

S

YLS (SEQ
(SEQ ID
T (SEQ

ID NO:
NO: 199)
ID NO:

198)

201)

Chain
VH
VH3-21
GFTFSSY
SSSSY
GAPMIGAA

H

(SEQ ID
(SEQ ID
AGWFDP

NO: 297)
NO: 338)
(SEQ ID

NO: 339)

Chain
VL
VK1-12
RASQGISS
AASSLQS
QQGVSFPR

L

WA (SEQ
(SEQ ID
T (SEQ

ID NO:
NO: 77)
ID NO:

337)

87)

*CDRs are identified by Chothia

Analysis of Potential Sequence Liabilities

Chain L (NKG2D binding light chain) has no predicted liabilities. Chain H (NKG2D binding heavy chain) has a DP in CDRH3. This motif is not affected by forced degradation. As such, this sequence was not altered in AB0089. Chain S contains a Met (potential oxidation site) and DG (potential isomerization site) in CDRH3. No preemptive removal of M or DG in CDRH3 was attempted. We have carefully monitored modification of these residues during accelerated stability and forced degradation studies. No modification of Met or DG was observed under any stresses tested, as shown below.

Framework Assessment

The CLEC12A binding arm of AB0192 was discovered through mouse immunization. The frameworks chosen for humanization are frequently used in clinical stage therapeutic monoclonal antibodies (FIGS. 211A-211D). The NKG2D binding arm is fully human, is based on a framework frequently used in advanced stage mAbs and does not contain any unusual residues or deviations from the germline.

Backmutations

Molecular modelling was used to choose murine backmutations that help maintaining CDRs architecture (Table 205). Seven backmutations: 3 in the HC and 4 in the LC were introduced.

TABLE 205

Summary of murine backmutations introduced

in the human framework in AB0192.

Number

of back
Murine framework residues

Framework
mutations
(Chothia)

VH
3
E(H1), R(H94), S(H108)

VL
4
F(L36), P(L46), Q(L69), Y(L71)

Structure Modeling

In the absence of a crystal structure, structural models of the variable segments of CLEC12A and NKG2D binding arm were generated for comparison. Model building was performed using the SAbPred website (opig.stats.ox.ac.uk/webapps/newsabdab/sabpred/).

Surface Charge Distribution and Hydrophobic Patch Analysis of CLEC12A Binding Arm of AB0192

FIGS. 212A-212C illustrate a ribbon diagram model of the CLEC12A binding scFv in three different orientations (upper panel) and their corresponding surface charge distribution of the same orientation (lower panel). The charge distribution of anti-CLEC12A scFv is polarized (“top view”, lower panel), with negatively-charged residues populated predominately within CDRH3 and CDRL2. The uneven distribution of electrostatic patches on the paratope is likely to be target-related and reflects the complementarity of charge distribution on its cognate epitope, which may likely contribute to the high affinity interaction between CLEC12A and the scFv of AB0192.

FIGS. 213A-213B show the analysis of CDR length and surface hydrophobicity patches of the CLEC12A-targeting arm of AB0192. Analyses were performed using therapeutic antibody profiles (TAP): opig.stats.ox.ac.uk/webapps/newsabdab/sabpred/tap. It compares the candidate in reference to 377 late-stage therapeutic antibodies. The length of CDRs for the CLEC12A-binding arm of AB0192 are within typical for late stage therapeutic antibodies. Hydrophobic patch analysis of the CDRs of an antibody is predictive of its developability. The theoretical hydrophobic properties of the TAA arm of AB0192 are well within the norm of commercialized biotherapeutics and late-stage therapeutic antibody candidates.

FIGS. 214A-214C shows the analysis of patches containing positive and negative charges, as well as the symmetry of these charges around the CDR regions. All three charts indicate that the charge distributions are within the norm compared to advanced stage therapeutic antibodies.

Surface Charge Distribution and Hydrophobic Patch Analysis of NKG2D Binding Arm of AB0192

The NKG2D-binding Fab arm is shown in ribbon diagrams with three different orientations (FIGS. 215A-215C, upper panel) and their corresponding surface charge distribution of the same orientation are provided in FIGS. 215A-215C, lower panel). The surface charge distribution on NKG2D A49M-I arm is more evenly distributed than charges on CLEC12A targeting arm of the molecule.

FIGS. 216A-216B show the CDR length and patches of surface hydrophobicity analyses of NKG2D-targeting arm of AB0192. It appears that the CDR length, the hydrophobic properties, the positive/negative charge distributions of the NKG2D-targeting arm of AB0192 are well within the norm of existing therapeutic antibodies.

FIGS. 217A-217C show the analyses of patches containing positive and negative charges, as well as the symmetry of these charges around the CDR regions. All three charts indicate that the charge distributions are within the norm comparing with the population of therapeutic antibodies.

Immunogenicity Prediction

The protein sequences of three chains of AB0192 were examined for the presence of potentially immunogenic T cell epitopes using the EpiMatrix algorithm and to rank the immunogenicity potential (FIGS. 218A-218B). AB0192 scores favorably on the Treg adjusted Immunogenicity Protein Scale. The Treg adjusted Epimatrix Protein Score for the sequences of the three chains of AB0192 are chain H: −33.39, chain S: −19.94 and chain L: −23.49 (Table 206). The predicted risk of immunogenicity for the AB0192 appears to be low.

TABLE 206

EpiVax immunogenicity analysis of AB0192.

Treg

EpiMatrix
EpiMatrix
adjusted

Protein Sequence
Length
Hits
Score
Score

AB0192_CHAIN_H
451
207
5.99
−33.39

AB0192_CHAIN_L
214
98
11.17
−23.49

AB0192_CHAIN_S
475
227
12.13
−19.94

Expression

Expression of AB0192 was achieved using Dragonfly Tx F3′ TriNKET platform purification method. Briefly, AB0192 was expressed by transient transfection of Expi293 and ExpiCHO cells with three plasmids encoding the NKG2D targeting light chain (chain L), the NKG2D targeting IgG heavy chain (chain H), and the CLEC12A targeting scFv-Fc fusion (chain S). For both Expi293 and ExpiCHO expression, Chain S:Chain H:Chain L were transfected at the plasmid ratio 4:1:1 to maximize the percentage of desired heterodimer product. This ratio was empirically obtained based on previous humanized TriNKETs developed and was not specifically optimized for AB0192.

Purification

Purification of AB0192 was performed using 2 step F3′ TriNKET platform purification method. No optimization or extensive process development for AB0192 purification was performed. The schematic representation of AB0192 purification (lot AB0192-005) expressed in ExpiCHO is shown on FIGS. 219A-219D.

Two lots of AB0192 were produced in transiently transfected ExpiCHO and Expi293 cells. Table 207 summarizes the titer and final product purity for both lots.

TABLE 207

Summary of expression titers and final purity

of two AB0192 lots.

Expression
Titer
Purity,

AB0192 Lot
host
(mg/L)
% Monomer

AB0192-002
Expi293
16
99.5

AB0192-005
ExpiCHO
26
100.0

The identity of AB0192 was corroborated by mass spectrometry. LC-MS analysis of the intact AB0192 (lot AB0192-005) showed an observed mass (126,429.9 Da) (FIG. 220) that agrees well with the theoretical mass of 126,429.5 Da. The major observed glycosylation pattern (G0F/G0F) is typical for an Fc-containing molecule expressed in CHO cells.

Biochemical and Biophysical Properties

AB0192 was characterized by SEC, CE, cIEF, HIC, and DSC to corroborate that is has the proper purity and pre-developability characteristics of a good biological therapeutic.

Purity

Purity of AB0192 (lot AB0192-005) assessed by SEC, non-reduced and reduced CE-SDS is summarized Table 208.

TABLE 208

Purity of AB0192 by SEC and CE-SDS analysis.

NR
R

SEC
CE-SDS
CE-SDS

HMW
Monomer
LMW
Purity
Purity

Test article
(%)
(%)
(%)
(%)
(%)

AB0192
ND
100
ND
95
98.8

(lot AB0192-005)

Size Excluscion Chromatography

SEC analysis of purified AB0192 (lot AB0192-005) showed >99% monomer as determined by integrated peak area (FIG. 221).

Non-Reduced and Reduced CE-SDS

AB0192 has high purify by CE-SDS (FIGS. 222A-222B). Under non-reduced conditions (left panel) the major impurities are method induced (free light chain and TriNKET—LC). Under reduced conditions (right panel) three expected chains (light chain (L), heavy chain (H), and scFv-Fc chain (S)) are observed.

Experimental pI and Charge Profile by cIEF

cIEF profile of AB0192 (lot AB0192-005) is shown on (FIG. 223). AB0192 has a major peak at a pI of 8.9±0.0. Several lesser abundant, overlapping acidic peaks and minor basic peaks are also observed.

Hydrophobicity Assessed by HIC

Hydrophobicity of AB0192 was assessed using hydrophobic interactions chromatography (HIC) (FIG. 224 and Table 209). The chromatogram reveals a single narrow peak, indicating that the molecule is highly pure and homogeneous. Retention time for AB0192 is comparable with retention time of Humira (a well behaved therapeutic monoclonal antibody).

TABLE 209

Hydrophobicity assessment of AB0192 by HIC.

Humira is used as an internal control of

well-behaved IgG1 mAb.

Retention time

Test article
(min)

AB0192
9.9

Humira
8.9

Thermal Stability by DSC

The thermal stability of AB0192 was assessed by DSC in several buffers (PBS, pH 7.4, HST, pH 6.0, CST, pH 7.0). AB0192 demonstrated high thermal stability in all buffers tested (FIGS. 225A-225C, Table 210). Tm1 corresponding to the unfolding of scFv satisfied ≥65° C. criteria postulated in thermostability assessment.

TABLE 210

Thermal stability analysis of AB0192 in three different buffer systems by DSC.

Test Article
Buffer
T_onset(° C.)
T_m1(° C.)
T_m2(° C.)
T_m3(° C.)
T_m4(° C.)

AB0192
PBS, pH 7.4
60.0
67.3
76.5
81.4
83.3

AB0192
HST, pH 6.0
60.8
69.0
78.8
84.0
86.1

AB0192
CST, pH 7.0
61.2
68.0
77.4
82.0
83.9

Disulfide Bond Assignment

AB0192 is an engineered molecule based on the backbone of a monoclonal IgG1 antibody. While a typical IgG1 contains 16 disulfide bonds, AB0192's F3′ format has only 15 disulfide bonds. Half of AB0192 is an anti-NKG2D half antibody, as such it contains the expected seven disulfide bonds (one in each IgG domain and the intermolecular disulfide connecting the LC to the HC). The other half of AB0192 is an anti-CLEC12A scFv-Fc fusion. The scFv contains three disulfide bonds (one in each anti-CLEC12A VH and VL domains and an engineered, stabilizing disulfide bridging the variable domains) and the Fc contains the two expected disulfides in the IgG domains. The heterodimer is held together with two hinge disulfides and one additional intermolecular engineered disulfide connecting the Fab-Fc to the scFv-Fc in the CH3 domain. Together this results in 15 expected disulfide bonds in AB0192. The predicted disulfide map for AB0192 is shown in FIG. 226.

The disulfide bond connectivity of AB0192 was corroborated by LC-MS/MS peptide mapping analysis of a non-reduced digest. Disulfide bonded peptides were identified by MS/MS database searching and corroborated by comparing their intensities in the native and reduced digests. To observe the engineered scFv stabilizing disulfide, AB0192 was digested with both trypsin and chymotrypsin. FIGS. 227A-227B show the extracted ion chromatogram (XICs) for the engineered disulfide pair in the scFv (non-reduced and reduced) and the most intense charge state for that peptide pair. All other disulfides were identified after digestion with only trypsin. XICs in FIGS. 228A-228B corroborate the existence of the engineered S—S bridge introduced to stabilize the Fc heterodimerization. A summary of the observed disulfide linked peptides in AB0192 is shown in Table 211. All theoretical disulfide linked peptides were observed with high mass accuracy (<5 ppm), were reducible, and were sequence corroborated by MS/MS fragmentation.

TABLE 211

Disulfide linked peptides theoretical and experimental masses.

Theoretical

Mass

mass
Experimental
accuracy

Domain
Peptide
(Da)
mass (Da)
(ppm)

VL
L2:L7
4482.0784
4482.0791
0.5

CL
L12:L19
3555.7490
3555.7527
1.4

CL-CH1
L20-21:H19
1260.4863
1260.4862
0.2

inter-

molecular

VH
H3:H10
3371.4686
3371.4728
1.3

CH1
H13:H14-15
7916.9194
7916.9162
0.2

Hinge
H20-21:S20-21
5454.7834
5454.7899
1.6

CH2
H23:H30
2328.0977
2328.1015
1.5

CH3-CH3
H36:S37-38
2757.3830
2757.3866
1.7

inter-

molecular^a

CH3
H37:H41
4432.0675
4432.073
1.3

scFv VL
S26:S42-44^b
1747.8314
1747.8381
3.8

scFv VH
S2:S9
3359.4951
3359.4993
1.3

scFv
S8:S46^b
1011.4154
1011.4198
4.4

stabilizing^a

CH2
S23:S30
non-unique, see H23:H30

CH3
S39:S44
3844.8236
3844.8287
1.2

^aEngineered disulfide

^bTrypsin/chymotrypsin double digest

Binding Characteristics

Binding to Human CLEC12A

The affinity of AB0089 for human CLEC12A was determined by SPR at 37° C. AB0089 binds with high affinity to human CLEC12A (FIGS. 229A-229C, Table 212). Per TriNKET MOA, the molecule is designed to bind strongly to TAA and weakly to NKG2D and CD16a. The complex between AB0192 and human CLEC12A is strong as evidenced from the slow rate dissociation constant 1.40±0.09×10⁻³s⁻¹. The equilibrium binding affinity K_Dfor AB0192 is 1.81±0.16 nM.

TABLE 212

Kinetic parameters and binding affinity of AB0192

for hCLEC12A measured by SPR.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0192
hCLEC12A-His
7.87 × 10⁵
1.36 × 10⁻³
1.72

AB0192
hCLEC12A-His
7.57 × 10⁵
1.51 × 10⁻³
1.99

AB0192
hCLEC12A-His
7.85 × 10⁵
1.34 × 10⁻³
1.71

Average ± Std Dev.
hCLEC12A-His
(7.76 ± 0.17) × 10⁵
(1.40 ± 0.09) × 10⁻³
1.81 ± 0.16

Average of 3 replicates

Binding to Polymorphic Variant of Human CLEC12A (Q244)

Polymorphic variant CLEC12A (Q244) is prevalent in 30% of population. We tested binding of AB0192 to this genetic variant by SPR and compared its affinity to CLEC12A (K244) WT (FIGS. 230A-230C). Binding affinities of AB0192 for WT and (K244Q) variant of CLEC12A are within 2.5-fold (Table 213). Binding of AB0192 to cynomolgus monkey CLEC12A (cCLEC12A) was also tested. No quantifiable binding was observed at 100 nM concentration.

TABLE 213

Kinetic parameters and binding affinities for (K244) and (Q244)

variants of human CLEC12A.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0192
hCLEC12A (K244) WT
(6.98 ± 0.09) × 10⁵
(1.15 ± 0.01) × 10⁻³
1.65 ± 0.04

AB0192
hCLEC12A (Q244)
(8.24 ± 1.33) × 10⁵
(3.70 ± 0.38) × 10⁻³
4.51 ± 0.26

Binding to Differentially Glycosylated CLEC12A

Human CLEC12A is a heavily glycosylated protein (6 potential N-glycosylation sites with an ECD compromising of 201 amino acids). Variations in the glycosylation status of CLEC12A on the surface of different cell types is documented in the literature (Marshall et al., 2006) Eur. J. Immunol., 36: 2159-2169). CLEC12A expressed on the surface of AML cells from different patients may have different glycosylation pattern as well. To determine AB0089 binds to different glyco-variants of human CLEC12A, the antigen was treated with neuraminidase (to remove terminal sialic acids) or PNGase F (to remove N-linked glycans) and affinity was compared to untreated CLEC12A by SPR. AB0192 bound to both glycosylated and de-glycosylated variants of CLEC12A (FIGS. 231A-231C and Table 214) and is expected to be unaffected by heterogeneity in target glycan composition.

TABLE 214

Kinetic parameters and binding affinity of AB0192 for

differentially glycosylated CLEC12A.

Test

article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0192
untreated
(6.98 ± 0.09) ×
(1.15 ± 0.01) ×
1.65 ±

CLEC12A
10⁵
10⁻³
0.04

AB0192
De-sialylated
(1.01 ± 0.03) ×
(1.04 ± 0.00) ×
1.04 ±

CLEC12A
10⁶
10⁻³
0.03

AB0192
De-glycosylated
(1.31 ± 0.02) ×
(3.48 ± 0.11) ×
0.27 ±

CLEC12A
10⁶
10⁻⁴
0.01

Average of 3 replicates

Binding to CLEC12A+ Isogenic and Cancer Cell Lines

Assessment of AB0192 binding to both isogenic cell lines expressing CLEC12A and AML cancer cells was monitored by FACS (FIGS. 232-232D and Table 215). AB0192 showed high affinity for cell surface expressed CLEC12A on the isogenic Ba/F3-CLEC12A cells and binds AML cell lines PL-21 and HL-60 with EC50 of less than 1 nM. Specificity of AB0089 for CLEC12A was demonstrated by lack of binding to the parental Ba/F3 cells that do not express the target.

TABLE 215

Binding of AB0192 to isogenic Ba/F3-CLEC12A and

cancer cell lines expressing CLEC12A.

Cell line
EC50 (nM)

Ba/F3-hCLEC12A
0.91

Ba/F3 parental
No binding

PL21
0.43

HL-60
0.6

Binding to Human NKG2D

Binding of AB0192 to human NKG2D ECD by SPR at 37° C. (Biacore) (FIGS. 233A-233F). NKG2D is a dimer in nature, therefore recombinant mFc-tagged NKG2D dimer was used for this experiment. Affinity was determined using AB0192. FIGS. 233A-233F shows AB0192 binding sensorgrams to human NKG2D. Two different fits were utilized to obtain the equilibrium affinity data: steady state affinity fit and kinetic fit. The kinetic constants and equilibrium affinity constants are shown in Table 216. AB0192 was designed to bind to human NKG2D with low affinity, most importantly with the fast rate of dissociation. The dissociation rate constant was (1.46±0.06×10−1 s−1). Equilibrium affinity constants (KD) obtained by kinetics fit and steady state affinity fit were very similar, 779±10 nM and 781±9 nM, respectively, which suggests high confidence in the measured parameters.

TABLE 216

Kinetic parameters and affinities of AB0192 for

human NKG2D measured by SPR.

Kinetics
Steady

Fit K_D
State Fit

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
(nM)
K_D(nM)

AB0192
hNKG2D
1.86 × 10⁵
1.43 × 10⁻¹
770
770

AB0192
hNKG2D
1.96 × 10⁵
1.53 × 10⁻¹
778
783

AB0192
hNKG2D
1.81 × 10⁵
1.43 × 10⁻¹
790
788

Average ± Std Dev.
hNKG2D
(1.88 ± 0.08) × 10⁵
(1.46 ± 0.06) × 10⁻¹
779 ± 10
781 ± 9

Binding to Human CD16a (V158 Allele)

One of the modalities of AB0192 mechanism of action is through engaging human CD16a (FcγRIIIa) that is expressed on NK cells via its Fc. AB0192 binding to human CD16a was tested by SPR (FIGS. 234A-234F, Table 217). CD16a high affinity V158 allele was used for this experiment. Since the literature reported Fc receptor affinity values for IgG1 vary depending on the experimental design and sourcing of the receptors, trastuzumab (Herceptin) was included as a human IgG1 control. The affinity of AB0192 for both alleles is comparable (around two-fold different) to trastuzumab's affinity for the same receptor. We do not expect this difference to have any meaningful physiological significance.

TABLE 217

Kinetics and affinities of AB0192 for hCD16a (V158)

and trastuzumab measured by SPR.

Test article
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0192
1.03 × 10⁵
1.90 × 10⁻²
184

AB0192
1.16 × 10⁵
2.01 × 10⁻²
173

AB0192
1.16 × 10⁵
2.06 × 10⁻²
177

Average ± StDev
(1.12 ± 0.08) × 10⁵
(1.99 ± 0.08) × 10⁻²
178 ± 6

Trastuzumab
1.31 × 10⁵
1.12 × 10⁻²
86

Trastuzumab
1.22 × 10⁵
1.08 × 10⁻²
89

Trastuzumab
1.30 × 10⁵
1.13 × 10⁻²
87

Average ± StDev
(1.28 ± 0.05) × 10⁵
(1.11 ± 0.03) × 10⁻²
87 ± 1

Testing CD16a-NKG2D Synergy

AB0192 is a tri-functional IgG-based biologic with two binding modalities on the NK cells: CD16a binding is achieved through the Fc and NKG2D binding is achieved through A49M-I Fab. Both interactions are important for the optimal cytotoxic activity of AB0192. To demonstrate synergy of co-engagement of human CD16a and human NKG2D binding, we performed an SPR experiment where we qualitatively compared binding of AB0192 to NKG2D, CD16a and mixed NKG2D-CD16a Biacore chip surfaces. The affinity of AB0192 for human NKG2D and human CD16a are both low, however, binding to both targets simultaneously results in an avidity effect that manifests as a slower off-rate (FIG. 235). Thus, AB0192 can avidly engage CD16a and NKG2D.

Co-Engagement of CLEC12A and NKG2D

To determine if binding of one target interferes with binding of the other target to AB0192, CLEC12A and NKG2D were sequentially injected over AB0192 captured on an anti-hFc IgG SPR chip (FIGS. 236A-236B). Target binding sensorgrams demonstrate that the occupancy status of CLEC12A does not interfere with NKG2D binding (FIG. 236A) and vice versa (FIG. 236B). Similarity in the shapes of the respective sensorgrams segments depicting binding of each target to free AB0192 and AB0192 that has been saturated with the other target suggest that the kinetic parameters are not meaningfully affected by the target occupancy status of the AB0192 molecule. For instance, the shape of the CLEC12A binding segment of the sensorgrams is similar in both panels. Additionally, the lack of any impact on relative stoichiometry of each target binding (when compared to binding to unoccupied AB0192) signifies full independence of NKG2D and CLEC12A binding sites on AB0192 (Table 218). Therefore, AB0192 successfully achieves simultaneous co-engagement of the tumor antigen and the NKG2D targeting arm.

TABLE 218

Relative binding stoichiometries of AB0192 for CLEC12A and NKG2D.

Binding stoichiometries are expressed relative to binding of each target

to unoccupied captured AB0192.

CLEC12A,
NKG2D,

relative binding
relative binding

stoichiometry
stoichiometry

Target bound to AB0192
1.00
1.00

unoccupied with another target

(injected first)

Target bound to AB0192
1.03 ± 0.07
1.15 ± 0.20

saturated with another target

(injected second)

Binding to Fcγ Receptors

Since AB0192 is an IgG1 derived biologic drug candidate it is important that it maintains appropriate Fc receptor binding properties. SPR data suggest that AB0089 binds human and cynomolgus Fcγ receptors with affinities comparable to trastuzumab, a marketed IgG1 biologic that served an experimental control (Table 219). Detailed description of each individual receptor binding is presented below.

TABLE 219

Binding of AB0192 and trastuzumab binding recombinant

human and cynomolgus FcγRs.

AB0192
Trastuzumab, hIgG1 control

Feγ receptor
K_D(nM)
K_D(nM)

hFcγRI
5.2 ±
1.2
2.9 ±
0.4

cFcγRI
1.9 ±
0.3
1.0 ±
0.1

hFcγRIIa H131
1011.3 ±
63.3
1045.1 ±
62.2

hFcγRIIa R131
1398.8 ±
286.1
1493.2 ±
265.9

hFcγRIIIa V158^a
178.3 ±
5.5
87.1 ±
1.4

hFcγRIIIa F158
680.0 ±
69.6
331.4 ±
3.3

cFcγRIII^a
203.8 ±
1.0
128.8 ±
2.1

hFcγRIIb
6062.3 ±
827.0
6270.3 ±
1031.7

hFcγRIIIb^a
6014.8 ±
735.5
3872.4 ±
486.3

^an = 3, otherwise n = 4.

Binding to Human and Cynomolgus CD64 (FcγRI)

Table 220 demonstrates that AB0089 binds recombinant human and cynomolgus CD64 (FcγRI) with affinities comparable (less than two-fold different) to trastuzumab. FIGS. 237A-237H and FIGS. 238A-238H depict binding of AB0089 to human and cynomolgus CD64, respectively.

TABLE 220

Binding of AB0192 and trastuzumab to recombinant

human and cynomolgus CD64 (FcγRI)

Test article
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

Human CD64 (FcγRI)

AB0192 Lot AB0192-005
4.7 × 10⁴
2.9 × 10⁻⁴
6.1

AB0192 Lot AB0192-005
4.7 × 10⁴
2.6 × 10⁻⁴
5.5

AB0192 Lot AB0192-005
4.7 × 10⁴
2.6 × 10⁻⁴
5.5

AB0192 Lot AB0192-005
5.2 × 10⁴
1.8 × 10⁻⁴
3.5

Average ± StDev
(4.8 ± 0.3) 10⁴
(2.5 ± 0.5) × 10⁻⁴
5.2 ± 1.2

trastuzumab
9.0 × 10⁴
2.3 × 10⁻⁴
2.5

trastuzumab
8.6 × 10⁴
2.1 × 10⁻⁴
2.5

trastuzumab
8.3 × 10⁴
2.7 × 10⁻⁴
3.2

trastuzumab
8.8 × 10⁴
2.8 × 10⁻⁴
3.2

Average ± StDev
(8.1 ± 0.0) × 10⁴
(2.5 ± 0.3) × 10⁻⁴
2.9 ± 0.4

Cynomolgus CD64 (FcγRI)

AB0192 Lot AB0192-005
5.4 × 10⁴
1.2 × 10⁻⁴
2.2

AB0192 Lot AB0192-005
8.0 × 10⁴
1.2 × 10⁻⁴
1.5

AB0192 Lot AB0192-005
5.7 × 10⁴
1.1 × 10⁻⁴
1.9

AB0192 Lot AB0192-005
5.5 × 10⁴
1.2 × 10⁻⁴
2.1

Average ± StDev
(6.2 ± 1.3) × 10⁴
(1.2 ± 0.1) × 10⁻⁴
1.9 ± 0.3

trastuzumab
1.3 × 10⁵
1.4 × 10⁻⁴
1.1

trastuzumab
1.3 × 10⁵
1.2 × 10⁻⁴
0.9

trastuzumab
1.3 × 10⁵
1.1 × 10⁻⁴
0.8

trastuzumab
1.3 × 10⁵
1.3 × 10⁻⁴
1.0

Average ± StDev
(1.3 ± 0.1) × 10⁵
(1.2 ± 0.2) × 10⁻⁴
1.0 ± 0.1

Binding to Human CD32a (FcγRIIa)

There is no meaningful difference between affinities of both alleles of human CD32a for AB0089 and trastuzumab (Table 221). FIGS. 239A-239P and FIGS. 240A-240P depict binding of AB0089 and trastuzumab to CD32a (H131 and R131 alleles), respectively, tested by SPR.

TABLE 221

Binding of AB0192 and trastuzumab to recombinant human CD32a

(FcγRIIa) H131 and R131 alleles.

K_D(nM)

Test article
FcγRIIa H131
FcγRIIa R131

AB0192 Lot AB0192-005
1035.5
1376.1

AB0192 Lot AB0192-005
921.1
1286.2

AB0192 Lot AB0192-005
1068.0
1800.6

AB0192 Lot AB0192-005
1020.6
1132.4

Average ± StDev
1011.3 ± 63.3
1398.8 ± 286.1

trastuzumab
1069.2
1488.2

trastuzumab
953.1
1381.0

trastuzumab
1090.2
1862.1

trastuzumab
1068.0
1241.4

Average ± StDev
1045.1 ± 62.2
1493.2 ± 265.9

Binding to Human CD16a F158 (FcγRIIIa F158) and Cyno CD16

Detailed description of AB0192 binding CD16a V158 is provided in Section 6.9.6. Binding of AB0192 to the low affinity F158 allele of human FcγRIIIa and cynomolgus FcγRIII are presented in Table 222, FIG. 241A-241P and FIG. 242A-242L. The AB0192 binding affinity values for recombinant human CD16a (F158 allele) and cynomolgus CD16 are comparable, less than two-fold different in affinity, to the ones of trastuzumab (Table 222).

TABLE 222

Binding of AB0192 and trastuzumab to recombinant human CD16a

(FcγRIIIa) F158 allele and cynomolgus CD16 (FcγRIII).

Human FcγRIIIa F158
Cynomolgus FcγRIII

Test article
K_D(nM)
K_D(nM)

AB0192 Lot AB0192-005
772.9
202.7

AB0192 Lot AB0192-005
633.7
204.6

AB0192 Lot AB0192-005
693.3
204.1

AB0192 Lot AB0192-005
620.0
n/a

Average ± StDev
680.0 ± 69.6
203.8 ± 1.0

trastuzumab
332.3
127.1

trastuzumab
326.6
128.1

trastuzumab
333.8
131.1

trastuzumab
333.1
n/a

Average ± StDev
331.4 ± 3.3
128.8 ± 2.1

Binding to Human CD32b (FcγRIIb)

There is no meaningful difference between affinities of CD32b for AB0089 and trastuzumab (Table 223). FIGS. 243A-243P depict binding to AB0089 and trastuzumab to CD32b tested by SPR.

TABLE 223

Binding of AB0192 and trastuzumab to

recombinant human CD32b (FcγRIIb).

Test article
K_D(μM)

AB0192 Lot AB0192-005
6.5

AB0192 Lot AB0192-005
5.6

AB0192 Lot AB0192-005
5.2

AB0192 Lot AB0192-005
7.0

Average ± StDev
6.1 ± 0.8

trastuzumab
7.1

trastuzumab
5.6

trastuzumab
5.2

trastuzumab
7.2

Average ± StDev
6.3 ± 1.0

Binding to Human CD16b (FcγRIIIb)

AB0089 binds human CD16b (FcγR3b) with affinity comparable (less than two-fold different) to the one of trastuzumab (Table 224). FIGS. 244A-244P depict binding of AB0089 and trastuzumab to CD16b tested by SPR.

TABLE 224

Binding of AB0192 and trastuzumab to

recombinant human CD16b (FcγRIIIb).

K_D(μM)

AB0192 Lot AB0192-005
6.8

AB0192 Lot AB0192-005
6.0

AB0192 Lot AB0192-005
5.3

Average ± StDev
6.0 ± 0.7

trastuzumab
4.3

trastuzumab
3.9

trastuzumab
3.4

Average ± StDev
3.9 ± 0.5

Binding to Human and Cynomolgus FcRn

FIGS. 245A-245P and FIGS. 246A-246L depict AB0192 binding to human and cynomolgus FcRn, respectively, tested by SPR. Affinities of AB0192 for human and cynomolgus FcRn are similar to those of trastuzumab, a marketed IgG1 biologic that served as an experimental control (Table 225).

TABLE 225

Binding of AB0192 and trastuzumab to human and cynomolgus FcRn.

Human FcRn K_D(μM)
Cyno FcRn K_D(μM)

AB0192 Lot AB0192-005
1.5
1.5

AB0192 Lot AB0192-005
1.3
1.4

AB0192 Lot AB0192-005
1.5
1.2

AB0192 Lot AB0192-005
1.5
n/a

Average ± StDev
1.5 ± 0.1
1.4 ± 0.1

trastuzumab
1.8
1.6

trastuzumab
1.5
1.6

trastuzumab
2.0
1.5

trastuzumab
1.7
n/a

Average ± StDev
1.7 ± 0.2
1.5 ± 0.1

Epitope Binning

We compared binding epitope of AB0192 in relation to AB0237, AB0089 and three reference controls TriNKETs by SPR (FIG. 247). cFAE-A49. Tepoditamab is based on CLEC12A binding arm of tepoditamab from Merus (WO_2014_051433_A1), cFAE-A49.CLL1-Merus is based on the clone 4331 from Merus (WO_2014_051433_A1) and cFAE-A49.h6E7 is based on h6E7 mAb from Genentech (h6E7) (US_2016_0075787_A1) AB0192 interaction with hCLEC12A was blocked by all three control molecules cFAE-A49.Tepoditamab, cFAE.A49.h6E7 and cFAE-A49.CLL1-Merus suggesting the AB0192 binding epitope is broad and overlaps with both Merus and Genentech molecules. This footprint is different from AB0237 and AB0089 which cross-block only with Merus molecules. Thus, binding epitope for AB0192 is different from AB0089 and AB0237.

Potency

The ability of AB0192 to lyse human cancer cells expressing CLEC12A was determined in the human NK cells mediated cytotoxicity assay. AB0192 demonstrated high potency in killing PL21 AML cells and vastly outperformed corresponding monoclonal antibody (FIG. 248 and Table 226).

TABLE 226

Potency of AB0192 in primary NK mediated

cytotoxicity assay compared to corresponding

mAb control.

EC50
Max killing

Test article
(nM)
(%)

AB0192
1.72
24

Parental mAb (AB0306)
ND
ND

Specificity

Non-Specific Binding Assessed by PSR Assay

Non-specific binding of AB0192 was assessed by flow cytometry based PSR assay that measures binding to a preparation of detergent solubilized CHO cell membrane proteins (FIGS. 249A-249B). PSR assay is known to correlate well with cross-interaction chromatography, a surrogate for antibody solubility, as well as with baculovirus particle enzyme-linked immunosorbent assay, a surrogate for in vivo clearance and therefore serves as a good developability predictive factor. AB0192 showed no non-specific binding, similar to low PSR control (Trastuzumab), suggesting high specificity and lack on non-specific binding to unrelated proteins.

Specificity Assessment of AB0192 in HuProt™ Microarray Assay

To examine specificity of AB0192 we have explored a protein array technology. The HuProt™ human proteome microarray provides the largest database of individually purified human full-length proteins on a single microscopic slide. An array consisting of 22,000 full-length human proteins are expressed in yeast Cerevisiae, purified, and are subsequently printed in duplicate on a microarray glass slide that allows thousands of interactions to be profiled in a high-throughput manner. FIG. 250 and shows the relative binding (Z score) of AB0192 at 1 μg/ml to human CLEC12A in comparison to the entire human proteome microarray. Z score is the average binding score of two duplicates of a given protein. For comparison purpose, the top 24 proteins with residual background binding to AB0192 were also provided in FIG. 250. Table 227 shows the Z and S scores of AB0192 to human CLEC12A and the top 6 proteins from the microarray. S score is the difference of Z score of a given protein and the one rank next to it. Based on the Z and S score criteria, AB0192 showed high specificity to human and lack of off-target binding in the HuProt™ human proteome assay.

TABLE 227

Summary of the top hits by HuProt ™ array.

Name
Rank
Protein-ID
Z score
S score

hCLEC12A
1
hCLEC12A
150.06
144.04

MFF
2
JHU15965.P168B06
6.05
0.81

POLR2E
3
JHU02080.P189F02
5.24
0.22

HMGA1_frag
4
JHU16418.P173B08
5.02
0.12

PARS2
5
JHU06781.P071H01
4.90
0.24

THEMIS_frag
6
JHU15201.P160B01
4.66
0.23

ZNF608
7
JHU13018.P136G04
4.43
0.006

Developability Assessment

The stability and forced degradation of AB0192 was assessed under the following conditions:

Accelerated (40° C.) Stability

20 mg/ml in 20 mM histidine, 250 mM sucrose, 0.01% Tween-80, pH 6.0, 40° C., 4 weeks (SEC, SPR, potency, SDS-CE)
20 mg/ml in 20 mM sodium citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0, 40° C., 4 weeks (SEC, SPR, potency, SDS-CE)

Chemical Stability

Oxidation: 1 mg/ml in 0.02% hydrogen peroxide, PBS, room temp., 24 hours (SEC, CE-SDS, SPR, potency, peptide map)
pH 8.0: 1 mg/ml 20 mM Tris, 40° C., 2 weeks (cIEF, SPR, potency, peptide map)
pH 5.0: 1 mg/ml 20 mM sodium acetate, 40° C., 2 weeks (cIEF, SPR, potency, peptide map)

Manufacturability

Freeze/thaw: 20 mg/ml 6 cycles, 20 mM sodium citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 (SEC, SDS-CE)
Agitation: 20 mg/ml in 20 mM sodium citrate, 250 mM sucrose, 0.01% Tween-80, pH 7.0 (SEC, SDS-CE)
High concentration: >150 mg/ml, PBS (SEC, A280)
Low pH hold: pH 3.3, 1.0 hours, ProA elution buffer (SEC, cIEF, SDS-CE, intact mass, SPR, potency)

Accelerated (40° C.) Stability

To assess the stability of AB0192, the molecule was staged at 40° C. for 4 weeks under two different buffer conditions (1) 20 mg/ml in 20 mM histidine, 250 mM sucrose, 0.01% PS-80, pH 6.0 (HST), and (2) 20 mg/ml in 20 mM citrate, 250 mM sucrose, 0.01% PS-80, pH 7.0 (CST). Stability in both buffers was assessed by SEC, CE, SPR and potency.

Stability in HST, pH 6.0

After 4 weeks at 40° C., AB0192 in HST, pH 6.0 showed minimal aggregation or fragmentation by SEC (FIGS. 251A-251B, Table 228) or CE-SDS (FIGS. 252A-252B, Table 229). Loss of monomer was 1.6% by SEC and 1% by non-reduced CE. There is no meaningful difference in the active species content or the affinities for hCLEC12A, hNKG2D and hCD16a between control and stressed samples (FIGS. 253A-253F and Table 230). Finally, there was no difference in potency between the control and stressed samples tested in KHYG-1-CD16aV cytotoxicity assay (FIG. 254 and Table 231).

TABLE 228

SEC analysis of AB0192 after 2 and 4 weeks at 40° C.

in HST, pH 6.0.

Conc.
Monomer
LMWS
HMWS

Test article
(mg/ml)
(%)
(%)
(%)

AB0192 control
20.1
98.9
1.2
0.0

AB0192 2 wks., 40° C.
20.2
98.0
2.0
0.0

AB0192, 4 wks., 40° C.
21.3
97.3
2.8
0.0

TABLE 229

NR and R CE-SDS analysis of AB0192 after

4 weeks at 40° C. in HST pH 6.0.

NR CE
R CE

purity
purity

Test article
(%)
(%)

AB0192 control
99.1
97.5

AB0192, 4 wks., 40° C.
98.1
97.4

TABLE 230

Kinetic parameters and binding affinities of AB0089 for CLEC12A, NKG2D

and CD16a after 4 weeks at 40° C. in HST, pH 6.0, measured by SPR.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0192 control
hCLEC12A
(7.76 ± 0.17) × 10⁵
(1.40 ± 0.09) × 10⁻³
1.81 ± 0.2

AB0192, 4 wks., 40° C.
hCLEC12A
(7.76 ± 0.40) × 10⁵
(1.40 ± 0.07) × 10⁻³
1.81 ± 0.2

AB0192 control
hNKG2D
(2.08 ± 0.03) × 10⁵
(1.22 ± 0.01) × 10⁻¹
588 ± 6

AB0192, 4 wks., 40° C.
hNKG2D
(2.11 ± 0.35) × 10⁵
(1.27 ± 0.19) × 10⁻¹
602 ± 9

AB0192 control
CD16aV
(1.17 ± 0.02) × 10⁵
(1.99 ± 0.05) × 10⁻²
171 ± 2

AB0192, 4 wks., 40° C.
CD16aV
(1.03 ± 0.03) × 10⁵
(2.01 ± 0.03) × 10⁻²
194 ± 4

Average of n = 3 replicates

TABLE 231

Potency of AB0192 after 4 weeks at 40° C. in HST,

pH 6.0 in KHYG-1-CD16aV mediated cytotoxicity assay.

EC50
Max killing

Test article
Buffer
(nM)
(%)

AB0192 control
HST
3.3
21

AB0192, 40° C. 4 wks.
HST
1.6
17

Stability in CST, pH 7.0

After 4 weeks at 40° C., AB0192 in CST, pH 7.0, showed minimal aggregation (FIG. 255A-255B and Table 232) and fragmentation (FIG. 256A-256B and Table 233). 1.6% and 0.8% loss of monomer was observed by SEC and non-reduced CE-SDS, respectively. There was no meaningful difference in the active species content or in the affinities for hCLEC12A, hNKG2D and hCD16a between control and stressed samples (FIG. 257A-257F and Table 234). Finally, there was no difference in potency between the control and stressed samples in KHYG-1-CD16aV cytotoxicity assay (FIG. 258 and Table 235).

TABLE 232

SEC analysis of AB0192 after 1-4 weeks at 40° C. in CST, pH 7.0.

Conc.
Monomer
LMWS
HMWS

Test article
(mg/ml)
(%)
(%)
(%)

AB0192 control
21.3
99.0
1.1
ND

AB0192, 1 wk., 40° C.
21.4
98.1
1.5
0.4

AB0192, 2 wks., 40° C.
21.4
97.9
1.6
0.5

AB0192, 3 wks., 40° C.
21.7
97.8
1.7
0.5

AB0192, 4 wks., 40° C.
22.2
97.4
1.8
0.8

TABLE 233

NR and R CE-SDS analysis of AB0192 after

4 weeks at 40° C. in CST, pH 7.0.

NR CE
R CE

Conc.
purity
purity

Test article
(mg/ml)
(%)
(%)

AB0192 control
21.3
98.8
97.8

AB0192, 40° C. 4 wks.
22.4
98.0
97.0

TABLE 234

Kinetic parameters and binding affinities of AB0089 for CLEC12A, NKG2D

and CD16a after 4 weeks at 40° C. in CST, pH 7.0, measured by SPR.

Test article
Target
k_a(M⁻¹ s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0192 control
hCLEC12A
(7.76 ± 0.17) × 10⁵
(1.40 ± 0.09) × 10⁻³
1.81 ± 0.2

AB0192, 4 wks. 40° C.
hCLEC12A
(7.49 ± 0.22) × 10⁵
(1.47 ± 0.12) × 10⁻³
1.96 ± 0.2

AB0192 control
hNKG2D
(2.08 ± 0.03) × 10⁵
(1.22 ± 0.01) × 10⁻¹
588 ± 6

AB0192, 4 wks. 40° C.
hNKG2D
(1.85 ± 0.06) × 10⁵
(1.21 ± 0.04) × 10⁻¹
653 ± 10

AB0192 control
CD16aV
(1.17 ± 0.02) × 10⁵
(1.99 ± 0.05) × 10⁻²
171 ± 2

AB0192, 4 wks. 40° C.
CD16aV
(1.11 ± 0.07) × 10⁵
(1.94 ± 0.01) × 10⁻²
174 ± 10

Average of 3 replicates

TABLE 235

Potency of AB0192 after 4 weeks at 40° C. in CST, pH 7.0,

in KHYG-1-CD16aV mediated cytotoxicity assay.

EC50
Max killing

Test article
(nM)
(%)

AB0192 control
3.3
21

AB0192 40° C. 4 wks
1.4
19

Chemical Stability

Forced Oxidation

After 24 h of oxidative stress in 0.02% hydrogen peroxide AB0192 showed no significant decrease in monomer content (FIGS. 259A-259B and Table 236). Purity assessed by CE remained high (FIGS. 260A-260B and Table 237). Amount of active species and affinities for CLEC12A, NKG2D, and CD16aV were unaltered (FIGS. 261A-261F and Table 238), and potency was similar to the control sample (FIG. 262 and Table 239).

TABLE 236

SEC analysis of AB0192 after forced oxidation compared to control.

Conc.
%
LMWS
HMWS

Test article
(mg/ml)
Monomer
(%)
(%)

AB0192 control
1.1
99.4
0.7
ND

AB0192 oxidized
0.9
99.7
0.3
ND

TABLE 237

NR and R CE-SDS analysis of AB0192 after forced oxidation.

NR CE
R CE

Conc.
Purity
Purity

Test article
(mg/ml)
(%)
(%)

AB0192 control
1.1
99.1
97.8

AB0192 oxidized
0.9
98.9
97.7

TABLE 238

Kinetic parameters and binding affinities of AB0089 for CLEC12A,

NKG2D and CD16a after forced oxidation measured by SPR.

Test

article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0192 control
hCLEC12A
(7.71 ± 0.04) × 10⁵
(1.29 ± 0.02) × 10⁻³
1.68 ± 0.0

AB0192
hCLEC12A
(7.75 ± 0.07) × 10⁵
(1.32 ± 0.05) × 10⁻³
1.71 ± 0.1

AB0192 control
hNKG2D
(2.08 ± 0.05) × 10⁵
(1.24 ± 0.03) × 10⁻¹
598 ± 9

AB0192 oxidized
hNKG2D
(1.95 ± 0.02) × 10⁵
(1.19 ± 0.03) × 10⁻¹
612 ± 11

AB0192 control
CD16aV
(1.15 ± 0.07) × 10⁵
(2.02 ± 0.05) × 10⁻²
177 ± 7

AB0192 oxidized
CD16aV
(1.19 ± 0.04) × 10⁵
(1.92 ± 0.04) × 10⁻²
162 ± 3

Average of n = 3 replicates

TABLE 239

Potency of AB0192 after forced oxidation

stress assessed by KHYG-1-CD16aV

mediated cytotoxicity assay.

EC50
Max killing

Test article
(nM)
(%)

AB0192 control
1.9
18

AB0192 oxidized
2.6
18

Site-specific oxidation of methionine was monitored after forced oxidation by tryptic peptide mapping. Significant levels of oxidation were only detected at two methionines in the Fc, as expected (Table 240) as expected for an IgG1. Our historic data show that the same residues are modified in trastuzumab at 61.6% and 39.1% under the same conditions.

TABLE 240

Summary of methionine oxidation in AB0192

after oxidative stress.

SEQ

Relative

ID

Abundance (%)

NO
Sequence
Chain
Position
Control
Oxidized

299
LSCAASGFTFSSYSMNWVR
H
34
0.5
0.8

300
NSLYLQMNSLR
H
83
0.1
0.3

301
DTLMISR
H/S
257/281
1.4
64.7

302
WQQGNVFSCSVMHEALHNHYTQK
H/S
433/457
1.2
35.9

303
DIQMTQSPSSVSASVGDR
L
4
0.1
0.3

340
LSCAASGFTFNAFGMHWVR
S
34
0.8
0.9

341
NTLYLQMNSLR
S
83
0.1
0.2

342
DGYPTGGAMDYWGQGTSVTVSSG
S
107/110/144
0.2
0.9

GGGSGGGGSGGGGSGGGGSDIQM

TQSPSSLSASVGDR^a

^a% oxidation repor ted as the sum of oxidation on M107, W110, and M144

Oxidation of tryptophan residues in AB0192 CDRs was also monitored in the same experiment. None of the tryptophan residues had a significant increase in oxidation after stress (Table 241).

TABLE 241

Summary of tryptophan oxidation in AB0192

CDRs after oxidative stress.

Relative

Mass shift
Abundance (%)

Sequence
Chain
CDR
(Da)^a
Control
Oxidized

GAPIGAAAGWFDP
H
H3
+16/+32/+4
ND/0.2/0.2
ND/0.1/0.0

(SEQ ID NO: 97)

RASQGISSWLA
L
L1
+16/+32/+4
0.1/0.2/0.0
0.2/0.1/0.0

(SEQ ID NO: 86)

^asingle oxidation (+16 Da), double oxidation (+32 Da), and kyneurenine (+4 Da)

ND, none detected.

pH 5 Stress

The chemical stability of AB0192 was assessed at low pH (20 mM sodium acetate, pH 5.0, 40° C., 2 weeks). After 2 weeks at pH 5, no loss of monomer (FIGS. 263A-263B and Table 242) was observed by SEC. Differences in purity detected by non-reduced and reduced CE-SDS were less than 1% (FIGS. 264A-264B and Table 243). There were no meaningful differences in CLEC12A, NKG2D, or CD16aV affinity of the amount of active species observed (FIGS. 265A-265F and Table 242). A small acidic shift in the global charge profile was detected by cIEF (FIG. 266 and Table 245), however, this acidic shift did not affect AB0192 potency (FIG. 267 and Table 246).

TABLE 242

SEC analysis of AB0192 after long term low pH stress

(pH 5.0, 40° C., 14 days).

Conc,
Monomer
LMWS
HMWS

Test article
(mg/ml)
(%)
(%)
(%)

AB0192 control
1.2
99.3
0.7
ND

AB0192, pH 5 40° C., 2 wks.
1.2
99.4
0.6
ND

TABLE 244

Kinetic parameters and binding affinities of AB0089 for

CLEC12A, NKG2D and CD16a after long term low pH

stress (pH 5.0, 40° C., 14 days) measured by SPR.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0192 control
hCLEC12A
(7.71 ±
(1.29 ±
1.68 ± 0.0

0.04) × 10⁵
0.02) × 10⁻³

AB0192, pH 5
hCLEC12A
(7.87 ±
(1.30 ±
1.65 ± 0.0

40° C., 2 wks.

0.12) × 10⁵
0.03) × 10⁻³

AB0192 control
hNKG2D
(2.08 ±
(1.22 ±
588 ± 6

0.03) × 10⁵
0.01) × 10⁻¹

AB0192, pH 5
hNKG2D
(2.10 ±
(1.20 ±
574 ± 12

40° C., 2 wks.

0.06) × 10⁵
0.01) × 10⁻¹

AB0192 control
CD16aV
(1.17 ±
(1.99 ±
171 ± 2

0.02) × 10⁵
0.05) × 10⁻²

AB0192, pH 5
CD16aV
(1.08 ±
(2.08 ±
192 ± 2

40° C., 2 wks.

0.02) × 10⁵
0.04) × 10⁻²

Average of 3 replicates

TABLE 245

Charge profile of AB0192 after long term low pH stress

(pH 5.0, 40° C., 14 days).

Test article
% acidic
% main
% basic

AB0192 control
41.9
49.4
8.7

AB0192, pH 5 40° C., 2 wks.
49.8
41.4
8.8

TABLE 246

Potency of AB0192 after long term low pH stress (pH 5.0, 40° C.,

14 days) in KHYG-1-CD16aV cytotoxicity assay.

Test article
EC50 (nM)
Max killing (%)

AB0192 control
1.9
18

AB0192, pH 5 40° C., 2 wks.
2.0
18

pH 8 Stress

Chemical stability of AB0192 was also assessed by incubation at elevated pH (20 mM Tris, pH 8.0, 40° C., 2 weeks). After 2 weeks at pH 8, there was a 1.6% loss of monomer by SEC (FIGS. 268A-268B and Table 247). Minor differences in purity were detected by non-reduced and reduced CE-SDS (FIGS. 269A-269B and Table 248). There were no meaningful differences in CLEC12A, NKG2D, or CD16aV affinities or amount of active species observed (FIGS. 270A-270F and Table 249). A significant acidic shift in the global charge profile detected by cIEF was detected (FIG. 271 and Table 250), which can be attributed to deamidation throughout the molecule. However, this acidic shift did not have any significant effect on AB0192 potency (FIG. 272 and Table 251).

TABLE 247

SEC analysis of AB0192 after long term high pH stress

(pH 8.0, 40 C., 14 days)

Conc.
Monomer
LMWS
HMWS

Test article
(mg/ml)
(%)
(%)
(%)

AB0192 control
1.1
99.3
0.7
0.0

AB0192, pH 8 40° C., 2 wks.
1.2
97.7
1.6
0.7

TABLE 248

NR and R CE-SDS analysis of AB0192 after long term

high pH stress (pH 8.0, 40° C., 14 days).

Test article
NR CE Purity (%)
R CE Purity

AB0192 control
97.4
98.3

AB0192, pH 8 40° C., 2 wks.
95.7
94.7

TABLE 249

Kinetic parameters and binding affinity of AB0089 for CLEC12A,

NKG2D and CD16a after long term high pH stress

(pH 8.0, 40° C., 14 days) measured by SPR.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0192 control
hCLEC12A
(7.71 ±
(1.29 ±
1.68 ± 0.00

0.04) × 10⁵
0.02) × 10⁻³

AB0192, pH 8
hCLEC12A
(7.68 ±
(1.28 ±
1.66 ± 0.10

40° C., 2 wks.

0.17) × 10⁵
0.04) × 10⁻³

AB0192 control
hNKG2D
(2.08 ±
(1.22 ±
588 ± 6

0.03) × 10⁵
0.01) × 10⁻¹

AB0192, pH 8
hNKG2D
(1.56 ±
(1.21 ±
779 ± 45

40° C., 2 wks.

0.09) × 10⁵
0.04) × 10⁻¹

AB0192 control
CD16aV
(1.17 ±
(1.99 ±
171 ± 2

0.02) × 10⁵
0.05) × 10⁻²

AB0192, pH 8
CD16aV
(1.10 ±
(1.91 ±
173 ± 2

40° C., 2 wks.

0.04) × 10⁵
0.04) × 10⁻²

Average of 3 replicates

TABLE 250

AB0192 charge profile after long term high

pH stress (pH 8.0, 40° C., 14 days).

Test article
% acidic
% main
% basic

AB0192 control
42.7
48.6
8.7

AB0192, pH 8 40° C., 2 wks.
60.3
30.5
9.1

TABLE 251

Potency of AB0192 after long term high pH stress (pH 8.0, 40 C.,

14 days) assessed by KHYG-1-CD16aV cytotoxicity assay.

Test article
EC50 (nM)
Max killing (%)

AB0192 control
1.9
18

AB0192, pH 8 40° C., 2 wks.
3.0
18

Peptide Mapping Analysis of the pH 5 and pH 8 Stressed Material

To determine if the pH stress induced changes observed by clEF are due to the modification in the complementarity determining regions (CDRs), the same pH 5 and pH 8 stressed samples were analyzed by LC-MS/MS peptide mapping. Based on the peptide map data, all of the CDRs in AB0192 are resistant to modification during low and high pH stress (Table 252). No evidence of aspartic acid isomerization was observed, based on manual inspection of peak shape.

TABLE 252

Post-translational modifications in AB0192

CDRs after low and high pH stresses.

Relative

abundance (%)

Sequence
Potential
pH 5
pH 8

Chain
CDR
(Chothia)
liability
control
stressed
control
stressed

S
H1
GFTFNAF
Deamidation
NA

(SEQ ID

NO: 195)

S
H2
SSGSTS (SEQ
None
NA

ID NO: 196)

S
H3
DGYPTGGAM
Isomerization
None detected

DY (SEQ ID

NO: 187)

S
L1
KASQDIYNYL
Deamidation
NA

S (SEQ ID

NO: 198)

S
L2
RANILVS (SEQ
Deamidation
None detected

ID NO: 199)

S
L3
LQFDAFPFT
None
NA

(SEQ ID

NO: 201)

H
H1
GFTFSSY (SEQ
None
NA

ID NO: 297)

H
H2
SSSSSY (SEQ
None
NA

ID NO: 298)

H
H3
GAPIGAAAG
Truncation
None detected

WFDP (SEQ ID

NO: 97)

L
L1
RASQGISSWL
None
NA

A (SEQ ID

NO: 86)

L
L2
AASSLQS
None
NA

(SEQ ID NO:

77)

L
L3
QQGVSFPRT
None
NA

(SEQ ID NO:

87)

NA, not applicable

Manufacturability

Freeze/Thaw Stability

Aliquots of AB0192 (20 mg/ml in CST pH 7.0) were frozen and thawed 6 times by placing the vials in Styrofoam container (to slow the freezing rate) and placing the container at −80° C. Once frozen, the container was placed on a lab bench and the aliquots were allowed to thaw. After 6 cycles of freeze/thaw AB0192 showed no loss in monomer by SEC and protein concentration remained the same (FIGS. 273A-273B and Table 253).

TABLE 253

SEC analysis of AB0192 after freeze/thaw stress.

Test
Conc.
Monomer
LMWS
HMWS

article
(mg/ml)
(%)
(%)
(%)

AB0192 control
21.2
98.9
1.1
ND

AB0192 6 F/T
20.8
98.9
1.1
ND

After 6 freeze/thaw cycles no loss of purity was detected by either reduced or non-reduced CE-SDS (FIGS. 274A-274B and Table 254).

TABLE 254

NR and R CE-SDS analysis of AB0192 after freeze/thaw stress.

Test article
NR CE Purity (%)
R CE Purity (%)

AB0192 control
99.0
97.3

AB0192 6 F/T
99.1
97.7

Agitation

Agitation study was performed in CST, pH 7.0 buffer. AB0192 was incubated for 24 h at constant steering (60 rpm) in a glass vial with a micro steer bar. No loss of monomer was detected by SEC compared to the control sample (FIGS. 275A-275B and Table 255).

TABLE 255

SEC analysis of AB0192 after agitation stress.

Test
Conc.
Monomer
LMWS
BMWS

article
(mg/ml)
(%)
(%)
(%)

AB0192
21.2
98.9
1.1
ND

control

AB0192
21.4
98.7
1.0
0.3

agitation

ND, none detected

After 24 hours of stirring, there was no increase in fragmentation detected by reduced or non-reduced CE-SDS (FIGS. 276A-276B and Table 256).

TABLE 256

NR and R CE-SDS analysis of AB0192 after agitation stress.

Test article
NR CE Purity (%)
R CE Purity (%)

AB0192 control
99.0
97.3

AB0192 agitation
99.2
97.1

Low pH Hold

To determine if AB0192 is amendable to commonly used low pH hold protocol used for viral inactivation in biologics manufacturing, we performed low pH hold study. Incubation at low pH can potentially compromise biophysical stability and integrity of the molecule and could cause chemical degradation of amino acid side chains, in particular, aspartic acid isomerization. There is a potential sequence liability DS in CDRH3 of Chain S. Therefore, we have applied a wide range of assays to assess protein quality after low pH hold. The bulk of the AB0192 Protein A eluate was immediately adjusted to neutral pH, while a smaller sample was adjusted to pH 3.7 and held at room temperature for 1 hour to mimic a typical viral inactivation step. After the hold period, Protein A eluate was neutralized with 1.0 M Tris, pH 8.3 to achieve neutral pH. Analytical size exclusion chromatography was performed to determine if there were any changes in profile or aggregate content pre- and post-low pH exposure (FIGS. 277A-277B). No change in profile was observed after low pH hold. Both aliquots were further purified by second ion exchange chromatography and subjected to a panel of characterization assays.

Chemical modification of amino acid side chains can typically be observed at a global scale with cIEF. The cIEF profiles of AB0192 control and low pH hold lot are visually very similar and the relative quantitation of acid, main, and basic species are all within 2% of each other. Indicating that the low pH hold did not have a measurable effect on the charge profile of AB0192 (FIG. 278 and Table 257).

TABLE 257

Charge profile of AB0192 after low pH hold.

Test article
pI
% acidic
% main
% basic

AB0192 control
8.9
41.4
49.8
8.8

AB0192 low
8.9
43.1
48.2
8.7

pH hold

After the low pH hold step, affinities for hCLEC12a, hNKG2D, and CD16aV were unaltered (FIGS. 279A-279F and Table 258). This indicates that AB0192 retains affinity for all three targets under low pH hold conditions used as viral inactivation step in biologics manufacturing

TABLE 258

Kinetic parameters and binding affinities of AB0192 for CLEC12A,

NKG2D and CD16a after low pH hold measured by SPR.

Test article
Target
k_a(M⁻¹s⁻¹)
k_d(s⁻¹)
K_D(nM)

AB0192
hCLEC12A
6.85 × 10⁵
1.25 × 10⁻³
1.83

Control

AB0192 low
hCLEC12A
(6.90 ± 0.17) ×
(1.15 ± 0.01) ×
1.66 ±

pH hold

10⁵
10⁻³
0.05

AB0192
hNKG2D
(1.88 ± 0.08) ×
(1.46 ± 0.060) ×
779 ±

Control

10⁵
10⁻¹
10

AB0192 low
hNKG2D
(1.78 ± 0.08) ×
(1.42 ± 0.03) ×
796 ±

pH hold

10⁵
10⁻¹
47

AB0192
CD16aV
(1.12 ± 0.08) ×
(1.99 ± 0.08) ×
178 ±

Control

10⁵
10⁻²
6

AB0192 low
CD16aV
(1.18 ± 0.06) ×
(2.12 ± 0.09) ×
180 ±

pH hold

10⁵
10⁻²
15

Potency of low pH hold sample was further tested against CLEC12A+ PL-21 cancer cells in the KHYG1-CD16aV cytotoxicity bioassay. The potency of AB0192 was unaltered after the low pH hold step (FIG. 280 and Table 259).

TABLE 259

Potency of AB0192 after low pH hold in

KHYG1-CD16a mediated cytotoxicity assay.

Test article
EC50 (nM)
Max killing (%)

AB0192 control
2.5
18

AB0192 low pH hold
2.0
18

Taken together, these results demonstrate that the molecular format and structure of AB0192 meets the conceptual requirements for a new multi-specific therapeutic modality that simultaneously engages CLEC12A, NKG2D and CD16a to treat AML. AB0192 is potent and has a favorable developability profile with no sequence liabilities or significant post-translational modifications observed under any stresses. AB0192 binds to an epitope different from AB0237 and AB0089 TriNKETs.

INCORPORATION BY REFERENCE

Unless stated to the contrary, the entire disclosure of each of the patent documents and scientific articles referred to herein is incorporated by reference for all purposes.

Equivalents

The application may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The foregoing embodiments are therefore to be considered in all respects illustrative rather than limiting the application described herein. Scope of the application is thus indicated by the appended claims rather than by the foregoing description, and all changes that come within the meaning and range of equivalency of the claims are intended to be embraced therein.

SEQUENCES

SEQ

ID NO
Sequence

1
QVQLQQWGAGLLKPSETLSLTCAVYGGSFSGYYWSWIRQPPGKGLEWIGEIDHSGSTNYNPSLKSRVT

ISVDTSKNQFSLKLSSVTAADTAVYYCARARGPWSFDPWGQGTLVTVSS

2
GSFSGYYWS

3
EIDHSGSTNYNPSLKS

4
ARARGPWSFDP

5
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYNSYPITFGGGTKVEIK

6
EIVLTQSPGTLSLSPGERATLSCRASQSVSSSYLAWYQQKPGQAPRLLIYGASSRATGIPDRFSGSGS

GTDFTLTISRLEPEDFAVYYCQQYGSSPITFGGGTKVEIK

7
DIQMTQSPSTLSASVGDRVTITCRASQSIGSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYHSFYTFGGGTKVEIK

8
DIQMTQSPSTLSASVGDRVTITCRASQSIGSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQSNSYYTFGGGTKVEIK

9
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYNSYPTFGGGTKVEIK

10
QVQLQQWGAGLLKPSETLSLTCAVYGGSFSGYYWSWIRQPPGKGLEWIGEIDHSGSTNYNPSLKSRVT

ISVDTSKNQFSLKLSSVTAADTAVYYCARARGPWGFDPWGQGTLVTVSS

11
ELQMTQSPSSLSASVGDRVTITCRTSQSISSYLNWYQQKPGQPPKLLIYWASTRESGVPDRFSGSGSG

TDFTLTISSLQPEDSATYYCQQSYDIPYTFGQGTKLEIK

12
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYGSFPITFGGGTKVEIK

13
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TDFTLTISSLQPDDFATYYCQQSKEVPWTFGQGTKVEIK

14
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYNSFPTFGGGTKVEIK

15
DIQMTQSPSTLSASVGDRVTITCRASQSIGSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYDIYPTFGGGTKVEIK

16
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYDSYPTFGGGTKVEIK

17
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYGSFPTFGGGTKVEIK

18
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYQSFPTFGGGTKVEIK

19
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYSSFSTFGGGTKVEIK

20
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYESYSTFGGGTKVEIK

21
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYDSFITFGGGTKVEIK

22
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYQSYPTFGGGTKVEIK

23
DIQMTQSPSTLSASVGDRVTITCRASQSIGSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYHSFPTFGGGTKVEIK

24
DIQMTQSPSTLSASVGDRVTITCRASQSIGSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYELYSYTFGGGTKVEIK

25
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYDTFITFGGGTKVEIK

26
QVQLVQSGAEVKKPGSSVKVSCKASGGTFSSYAISWVRQAPGQGLEWMGGIIPIFGTANYAQKFQGRV

TITADESTSTAYMELSSLRSEDTAVYYCARGDSSIRHAYYYYGMDVWGQGTTVTVSS

27
GTFSSYAIS

28
SYAIS

29
GIIPIFGTANYAQKFQG

30
ARGDSSIRHAYYYYGMDV

31
GDSSIRHAYYYYGMDV

32
DIVMTQSPDSLAVSLGERATINCKSSQSVLYSSNNKNYLAWYQQKPGQPPKLLIYWASTRESGVPDRF

SGSGSGTDFTLTISSLQAEDVAVYYCQQYYSTPITFGGGTKVEIK

33
KSSQSVLYSSNNKNYLA

34
WASTRES

35
QQYYSTPIT

36
QLQLQESGPGLVKPSETLSLTCTVSGGSISSSSYYWGWIRQPPGKGLEWIGSIYYSGSTYYNPSLKSR

VTISVDTSKNQFSLKLSSVTAADTAVYYCARGSDRFHPYFDYWGQGTLVTVSS

37
GSISSSSYYWG

38
SSSYYWG

39
SIYYSGSTYYNPSLKS

40
ARGSDRFHPYFDY

41
GSDRFHPYFDY

42
EIVLTQSPATLSLSPGERATLSCRASQSVSRYLAWYQQKPGQAPRLLIYDASNRATGIPARFSGSGSG

TDFTLTISSLEPEDFAVYYCQQFDTWPPTFGGGTKVEIK

43
RASQSVSRYLA

44
DASNRAT

45
QQFDTWPPT

46
DIQMTQSPSTLSASVGDRVTITCRASQSISSWLAWYQQKPGKAPKLLIYKASSLESGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCEQYDSYPTFGGGTKVEIK

47
QVQLVQSGAEVKKPGSSVKVSCKASGGTFSSYAISWVRQAPGQGLEWMGGIIPIFGTANYAQKFQGRV

TITADESTSTAYMELSSLRSEDTAVYYCARRGRKASGSFYYYYGMDVWGQGTTVTVSS

48
ARRGRKASGSFYYYYGMDV

49
DIVMTQSPDSLAVSLGERATINCESSQSLLNSGNQKNYLTWYQQKPGQPPKPLIYWASTRESGVPDRF

SGSGSGTDFTLTISSLQAEDVAVYYCQNDYSYPYTFGQGTKLEIK

50
ESSQSLLNSGNQKNYLT

51
QNDYSYPYT

52
QVQLVQSGAEVKKPGASVKVSCKASGYTFTSYYMHWVRQAPGQGLEWMGIINPSGGSTSYAQKFQGRV

TMTRDTSTSTVYMELSSLRSEDTAVYYCARGAPNYGDTTHDYYYMDVWGKGTTVTVSS

53
YTFTSYYMH

54
SYYMH

55
IINPSGGSTSYAQKFQG

56
ARGAPNYGDTTHDYYYMDV

57
GAPNYGDTTHDYYYMDV

58
EIVMTQSPATLSVSPGERATLSCRASQSVSSNLAWYQQKPGQAPRLLIYGASTRATGIPARESGSGSG

TEFTLTISSLQSEDFAVYYCQQYDDWPFTFGGGTKVEIK

59
RASQSVSSNLA

60
GASTRAT

61
QQYDDWPFT

62
QVQLVQSGAEVKKPGASVKVSCKASGYTFTGYYMHWVRQAPGQGLEWMGWINPNSGGTNYAQKFQGRV

TMTRDTSISTAYMELSRLRSDDTAVYYCARDTGEYYDTDDHGMDVWGQGTTVTVSS

63
YTFTGYYMH

64
GYYMH

65
WINPNSGGTNYAQKFQG

66
ARDTGEYYDTDDHGMDV

67
DTGEYYDTDDHGMDV

68
EIVLTQSPGTLSLSPGERATLSCRASQSVSSNLAWYQQKPGQAPRLLIYGASTRATGIPARESGSGSG

TEFTLTISSLQSEDFAVYYCQQDDYWPPTFGGGTKVEIK

69
QQDDYWPPT

70
EVQLLESGGGLVQPGGSLRLSCAASGETFSSYAMSWVRQAPGKGLEWVSAISGSGGSTYYADSVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCAKDGGYYDSGAGDYWGQGTLVTVSS

71
FTFSSYAMS

72
AISGSGGSTYYADSVKG

73
AKDGGYYDSGAGDY

74
DGGYYDSGAGDY

75
DIQMTQSPSSVSASVGDRVTITCRASQGIDSWLAWYQQKPGKAPKLLIYAASSLQSGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQGVSYPRTFGGGTKVEIK

76
RASQGIDSWLA

77
AASSLQS

78
QQGVSYPRT

79
EVQLVESGGGLVKPGGSLRLSCAASGETFSSYSMNWVRQAPGKGLEWVSSISSSSSYIYYADSVKGRF

TISRDNAKNSLYLQMNSLRAEDTAVYYCARGAPMGAAAGWFDPWGQGTLVTVSS

80
FTFSSYSMN

81
SYSMN

82
SISSSSSYIYYADSVKG

83
ARGAPMGAAAGWFDP

84
GAPMGAAAGWFDP

85
DIQMTQSPSSVSASVGDRVTITCRASQGISSWLAWYQQKPGKAPKLLIYAASSLQSGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQGVSFPRTFGGGTKVEIK

86
RASQGISSWLA

87
QQGVSFPRT

88
DIQMTQSPSSVSASVGDRVTITCRASQGISSWLAWYQQKPGKAPKLLIYAASSLQSGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQGVSFPRTFGCGTKVEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGG

GLVKPGGSLRLSCAASGFTFSSYSMNWVRQAPGKCLEWVSSISSSSSYTYYADSVKGRFTISRDNAKN

SLYLQMNSLRAEDTAVYYCARGAPMGAAAGWFDPWGQGTLVTVSS

89
QVQLVQSGAEVKKPGASVKVSCKASGYTFTSYYMHWVRQAPGQGLEWMGIINPSGGSTSYAQKFQGRV

TMTRDTSTSTVYMELSSLRSEDTAVYYCAREGAGFAYGMDYYYMDVWGKGTTVTVSS

90
AREGAGFAYGMDYYYMDV

91
EGAGFAYGMDYYYMDV

92
EIVLTQSPATLSLSPGERATLSCRASQSVSSYLAWYQQKPGQAPRLLIYDASNRATGIPARFSGSGSG

TDFTLTISSLEPEDFAVYYCQQSDNWPFTFGGGTKVEIK

93
RASQSVSSYLA

94
QQSDNWPFT

95
EVQLVESGGGLVKPGGSLRLSCAASGFTFSSYSMNWVRQAPGKGLEWVSSISSSSSYTYYADSVKGRF

TISRDNAKNSLYLQMNSLRAEDTAVYYCARGAPIGAAAGWFDPWGQGTLVTVSS

96
ARGAPIGAAAGWFDP

97
GAPIGAAAGWFDP

98
EVQLVESGGGLVKPGGSLRLSCAASGFTFSSYSMNWVRQAPGKGLEWVSSISSSSSYTYYADSVKGRF

TISRDNAKNSLYLQMNSLRAEDTAVYYCARGAPQGAAAGWFDPWGQGTLVTVSS

99
ARGAPQGAAAGWFDP

100
GAPQGAAAGWFDP

101
EVQLVESGGGLVKPGGSLRLSCAASGFTFSSYSMNWVRQAPGKGLEWVSSISSSSSYTYYADSVKGRF

TISRDNAKNSLYLQMNSLRAEDTAVYYCARGAPLGAAAGWFDPWGQGTLVTVSS

102
ARGAPLGAAAGWFDP

103
GAPLGAAAGWFDP

104
EVQLVESGGGLVKPGGSLRLSCAASGFTFSSYSMNWVRQAPGKGLEWVSSISSSSSYTYYADSVKGRF

TISRDNAKNSLYLQMNSLRAEDTAVYYCARGAPFGAAAGWFDPWGQGTLVTVSS

105
ARGAPFGAAAGWFDP

106
GAPFGAAAGWFDP

107
EVQLVESGGGLVKPGGSLRLSCAASGFTFSSYSMNWVRQAPGKGLEWVSSISSSSSYTYYADSVKGRF

TISRDNAKNSLYLQMNSLRAEDTAVYYCARGAPVGAAAGWFDPWGQGTLVTVSS

108
ARGAPVGAAAGWFDP

109
GAPVGAAAGWFDP

110
EVQLVESGGGLVKPGGSLRLSCAASGFTFSSYSMNWVRQAPGKGLEWVSSISSSSSYTYYADSVKGRF

TISRDNAKNSLYLQMNSLRAEDTAVYYCARGAPXGAAAGWFDPWGQGTLVTVSS, wherein X is

M, L, I, V, Q, or F

111
ARGAPXGAAAGWFDP wherein X is M, L, I, V, Q, or F

112
GAPXGAAAGWFDP wherein X is M, L, I, V, Q, or F

113
QVQLVESGGGLVKPGGSLRLSCAASGFTFSSYGMHWVRQAPGKGLEWVAFIRYDGSNKYYADSVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCAKDRGLGDGTYFDYWGQGTTVTVSS

114
QSALTQPASVSGSPGQSITISCSGSSSNIGNNAVNWYQQLPGKAPKLLIYYDDLLPSGVSDRFSGSKS

GTSAFLAISGLQSEDEADYYCAAWDDSLNGPVFGGGTKLTVL

115
SYAMS

116
QVHLQESGPGLVKPSETLSLTCTVSDDSISSYYWSWIRQPPGKGLEWIGHISYSGSANYNPSLKSRVT

ISVDTSKNQFSLKLSSVTAADTAVYYCANWDDAFNIWGQGTMVTVSS

117
EIVLTQSPGTLSLSPGERATLSCRASQSVSSSYLAWYQQKPGQAPRLLIYGASSRATGIPDRFSGSGS

GTDFTLTISRLEPEDFAVYYCQQYGSSPWTFGQGTKVEIK

118
DKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAK

TKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPSRD

ELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFS

CSVMHEALHNHYTQKSLSLSPG

119
(GlyGlyGlyGlySer)₄ ((G₄S)₄)

120
(GS)_n wherein n is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,

15, 16, 17, 18, 19, or 20

121
(GGS)_n wherein n is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,

15, 16, 17, 18, 19, or 20

122
(GGGS)_n wherein n is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,

15, 16, 17, 18, 19, or 20

123
(GGSG)_n wherein n is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,

15, 16, 17, 18, 19, or 20

124
(GGSGG)_n wherein n is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,

15, 16, 17, 18, 19, or 20

125
(GGGGS)_n wherein n is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,

15, 16, 17, 18, 19, or 20

126
GSGSGSGSGSGSGSGSGSGS

127
GGSGGSGGSGGSGGSGGSGGSGGSGGSGGS

128
GGGSGGGSGGGSGGGSGGGSGGGSGGGSGGGSGGGSGGGS

129
GGSGGGSGGGSGGGSGGGSGGGSGGGSGGGSGGGSGGGSG

130
GGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGG

131
GGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGS

132
GGGGSGGGGSGGGGS

133
GGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGG

GSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGS

134
GGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGS

GGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGG

135
EVQLQESGPGLVQPSQSLSITCTVSGFSLTNYGLHWVRQSPGKGLEWLGVIWSGGKTDYNTPFKSRLS

ISKDISKNQVFFKMNSLQPNDTAIYFCAKYDYDDSLDYWGQGTSVTVSS

136
DIQMNQSPSSLSASLGDTIAITCHASQNINFWLSWYQQKPGNIPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGQGTKLEIK

137
GFSLTNY

138
WSGGK

139
YDYDDSLDY

140
HASQNINFWLS

141
EASNLHT

142
QQSHSYPLT

143
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKGLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSS

144
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGQGTKLEIK

145
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIK

146
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISKDTSKNQ

FSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSS

147
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKGLEWIGVIWSGGKTDYNPSLKSRVT

ISVDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSS

148

EVQLVESGGGVVQPGGSLRLSCAASGFTENAFGMHWVRQAPGKGLEWVAFISSGSTSIYYANTVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSS

149
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISVDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIK

150
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISVDTSKNQ

FSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSS

151
DIQMTQSPSSLSASVGDRVTITCHASQNINEWLSWYQQKPGKX₁PKLLIYEASNLHTGVPSRFSGSGS

GTX₂FTLTISSLQPEDX₃ATYYCQQSHSYPLTFGX₄GTKLEIK wherein X₁ is A or I, X₂

is D or R, X₃ is F or I, and X₄ is Q or G

152
KSSQSLLWNVNQNNYLV

153
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVTANDTAVYYCAKYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIK

154
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISKDTSKNQ

FSLKLSSVTANDTAVYYCAKYDYDDSLDYWGQGTLVTVSS

155
EVQLQESGAELVRSGASIKLSCAASAFNIKDYFIHWVRQRPDQGLEWIGWIDPENDDTEYAPKFQDKA

TMTADTSSNTAYLQLSSLTSADTAVYYCNALWSRGGYFDYWGQGTTLTVSS

156
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIK

157
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISKDTSKNQ

FSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLVTVSS

158
EVLLTQSPAITAASPGEKVTITCSARSSVSYMSWYQQKPGSSPKIWIYGISKLASGVPARFSGSGSGT

YFSFTINNLEAEDVATYYCQQRSYYPFTFGSGTKLEIK

159
AFNIKDY

160
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQANDTAVYYCARYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIK

161
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISKDTSKNQ

FSLKLSSVQANDTAVYYCARYDYDDSLDYWGQGTLVTVSS

162

EVQLVESGGGVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKGLEWVAFISSGSTSIYYANTVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSS

163
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGQGTKLEIK

164
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIK

165
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISKDTSKNQ

FSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSS

166
WSGGS

167
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGSG

TDFTLTISSLQPEDIATYYCQQSHSYPLTFGQGTKLEIK

168
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGS

GTDFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIK

169
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGSG

TDFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISKDTSKNQ

FSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSS

170
DPENDD

171
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDFATYYCQQSHSYPLTFGQGTKLEIK

172
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDFATYYCQQSHSYPLTFGCGTKLEIK

173
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDFATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISKDTSKNQ

FSLKLSSVQANDTAVYYCAKYDYDDSLDYWGQGTLVTVSS

174
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKGLEWIGVIWSGGKTDYNPSLKSRVT

ISVDTSKNQFSLKLSSVTAADTAVYYCARYDYDDSLDYWGQGTLVTVSS

175
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQSHSYPLTFGQGTKLEIK

176
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISVDTSKNQFSLKLSSVTAADTAVYYCARYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGS

GTDFTLTISSLQPEDFATYYCQQSHSYPLTFGCGTKLEIK

177
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISVDTSKNQ

FSLKLSSVTAADTAVYYCARYDYDDSLDYWGQGTLVTVSS

178
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKGLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLVTVSS

179
QHNHGSFLPYT

180
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIK

181
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISKDTSKNQ

FSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLVTVSS

182
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKGLEWIGVIWSGGKTDYNPSLKSRVT

ISX₁DTSKNQFSLKLSSVX₂AX₃DTAVYYCAX₄YDYDDSLDYWGQGTLVTVSS; wherein X₁ is

V or K, X₂ is T or Q, X₃ is A or N, and X₄ is R or K

183
LWSRGGYFDY

184
EVQLVESGGGVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSG

GGGSDIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSG

SGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIK

185
DIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSGSGSG

QDYTFTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGG

GVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKGRFTISRDNSKN

TLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSS

186
SARSSVSYMS

187
DGYPTGGAMDY

188
GISKLAS

189
QQRSYYPFT

190
EVQLQESGPELEKPGASVRISCKASGYSFTAYNMNWVKQSNGKSLEWIGNIDPSYGDATYNQKFKGKA

TLTVDKSSSTAYMQLKSLTSEDSAVYYCARDNYYGSGYFDYWGQGTTLTVSS

191
SVLMTQTPLSLPVSLGDRASISCRSSQGIVHINGNTYLEWYLQKPGQSPKLLIYKVSNRFSGVPDRFS

GSGSGTDFTLKISRVEAEDLGVYYCFQGSHVPWTFGGGTKLEIK

192
GFTFNSF

193
EVQLQESGGGLVQPGGSRKLSCAASGFTFNSFGMHWVRQAPEKGLEWVAFISSGSTSIYYANTVKGRF

TISRDNPKNTLFLQMTSLRSEDTAMYYCARDGYPTGGAMDYWGQGTSVTVSS

194
DIKMTQSPSSMYASLGERVTITCKASQDIYNYLSWFQLKPGKSPRPLIYRANILVSGVPSKFSGSGSG

QDYSLTINSLEYEDLGIYYCLQFDAFPFTFGSGTKLEIK

195
GFTFNAF

196
SSGSTS

197
GYSFTAY

198
KASQDIYNYLS

199
RANILVS

200
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVILSGGWTDYNPSLKSRVTISKDTSKNQ

FSLKLSSVQAADTAVYYCAKGDYGDALDYWGQGTLVTVSS

201
LQFDAFPFT

202
DPSYGD

203
DIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSGSGSG

QDYTFTISSLQPEDIATYYCLQFDAFPFTFGSGTKLEIK

204
EVQLVESGGGVVQPGGSLRLSCAASGFTFNAFGMHWVRQAPGKCLEWVAFISSGSTSTYYANTVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSG

GGGSDIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSG

SGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIK

205
DIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSGSGSG

QDYTFTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGG

GVVQPGGSLRLSCAASGFTFNAFGMHWVRQAPGKCLEWVAFISSGSTSTYYANTVKGRFTISRDNSKN

TLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSS

206
GFSLTSY

207
DNYYGSGYFDY

208
EVQLVESGGGVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKCLEWVAFISSGSTSTYYANTVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCARSGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSG

GGGSDIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSG

SGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIK

209
DIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSGSGSG

QDYTFTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGG

GVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKCLEWVAFISSGSTSTYYANTVKGRFTISRDNSKN

TLYLQMNSLRAEDTAVYYCARSGYPTGGAMDYWGQGTSVTVSS

210

EVQLVESGGGVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKGLEWVAFISSGSTSTYYANTVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCARSGYPTGGAMDYWGQGTSVTVSS

211
RSSQGIVHINGNTYLE

212
KVSNRFS

213
SGYPTGGAMDY

214
FQGSHVPWT

215
EVQLVESGGGVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKCLEWVAFISSGSTSTYYANTVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSG

GGGSDIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSG

SGSGQDYTLTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIK

216
DIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSGSGSG

QDYTLTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGG

GVVQPGGSLRLSCAASGFTFNSFGMHWVRQAPGKCLEWVAFISSGSTSTYYANTVKGRFTISRDNSKN

TLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSS

217
EVQLQESGAELVRSGASVKLSCTVSGFNIKDYYMHWVKQRPEQGLEWIGWIDPENGDTENVPKFQGKA

TMTADTSSNTAYLQLRSLTSEDTAVYYCKSYYYDSSSRYVDVWGAGTTVTVSS

218
DIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSGSGSG

QDYTLTISSLQPEDIATYYCLQFDAFPFTFGSGTKLEIK

219
GIVMTQAPLTLSVTIGQPASISCKSSQSLLDSDGKTFLNWFLQRPGQSPKRLISLVSKLDSGVPDRFT

GSGSGTDFTLKLSRVEPEDLGVYYCWQGTHFPYTFGGGTKLEIK

220
GFNIKDY

221
GFSLTSF

222
DPENGD

223
SYFAMDY

224
KSSQSLLDSDGKTFLN

225
LVSKLDS

226
GASIRES

227
WQGTHFPYT

228
EVQLQESGAELMKPGASVKISCRTTGYTFSTYWIEWVKQRPGRGPEWIGELFPGNSDTTLNEKFTGKA

TFTADSSSNTAYMQLSSLTSEDSAVYYCARSGYYGSSLDYWGQGTTLTVSS

229
GIVMTQSPASLSASVGETVTITCRAGENIHSYLAWYQQKQGKSPQLLVYNAKTLAEGVPSRFSGSGSG

TQFSLKINSLQPEDFGSYYCQHHYGTPRTFGGGTKLEIK

230
GYTFSTY

231
FPGNSD

232
SGYYGSSLDY

233
RAGENIHSYLA

234
NAKTLAE

235
QHHYGTPRT

236
WSGGN

237
GIVMTQSPSSLAVTAGEKVTMRCKSSQSLLWNVNQNNYLVWYQQKQGQPPKLLIYGASIRESWVPDRF

TGSGSGTDFTLTISNVHAEDLAVYYCQHNHGSFLPYTFGGGTKLEIK

238
EVQLQESGPGLVQPSQSLSITCTVSGFSLTSFGIHWVRQSPGKGLEWLGVIWSGGNTDSNAAFISRLS

ITKDISKSQVFFKMNSLQVTDTAIYYCARSYFAMDYWGQGTSVTVSS

239
GASIRQS

240
KSSQSLLWNVNQNNYLL

241
THFGMDY

242
QVQLRQSGPGLVQPSQSLSITCTVSGFSLTSYGVHWVRQSPGKGLEWLGVMWSGGSTDYNAAFMSRLS

ISKDNSKSQVFFTMNSLQADDTAIYYCARTHFGMDYWGQGTPVTVSS

243
GIVMTQSPSSLAVTAGEKVTMRCKSSQSLLWSVNQNNYLLWYQQKQGQPPKLLIYGASIRQSWVPDRF

TGSGSGTDFTLSISNVHAEDLAVYYCQHNHGSFLPYTFGGGTKLEIK

244
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQFPGKGLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVTANDTAVYYCAKYDYDDSLDYWGQGTLVTVSS

245
KSSQSLLWSVNQNNYLL

246
XGYPTGGAMDY wherein X is D or S

247
EVQLQESGPGLVQPSQSLSITCTVSGFSLTSFGVHWVRQSPGKGLEWLGVIWSGGSTDSNAAFISRLT

ITKDNSKSQVFFKMNSLQATDTAIYYCARSYFAMDYWGQGTSVSVSS

248
DIVMTQSPSSLAVTAGEKVTMRCKSSQSLLWNVNQNNYLLWYQQKQGQPPKLLIYGASIRESWVPDRF

TGSGSGTDFTLTISNVHVEDLAVYYCQHNHGSFLPYTFGGGTKLEIK

249
EVQLVESGGGVVQPGGSLRLSCAASGFTENX₁FGMHWVRQAPGKGLEWVAFISSGSTSIYYANTVKGR

FTISRDNSKNTLYLQMNSLRAEDTAVYYCARX₂GYPTGGAMDYWGQGTSVTVSS wherein X₁ is

S or A and X₂ is D or S

250
DIX₁MTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSGSGS

GQDYTX₂TISSLQPEDIATYYCLQFDAFPFTFGSGTKLEIK wherein X₁ is Q or K and

X₂ is F or L

251
GFTFNXF wherein X is Sor A

252
EVQLVESGGGVVQPGGSLRLSCAASGFTFNAFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSG

GGGSDIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSG

SGSGQDYTLTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIK

253
DIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSGSGSG

QDYTLTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGG

GVVQPGGSLRLSCAASGFTFNAFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKGRFTISRDNSKN

TLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSS

254
EVQLVESGGGVVQPGGSLRLSCAASGFTENSFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCARSGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSG

GGGSDIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSG

SGSGQDYTLTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIK

255
DIKMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSGSGSG

QDYTLTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGG

GVVQPGGSLRLSCAASGFTENSFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKGRFTISRDNSKN

TLYLQMNSLRAEDTAVYYCARSGYPTGGAMDYWGQGTSVTVSS

256
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKGLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQANDTAVYYCARYDYDDSLDYWGQGTLVTVSS

257
YYYDSSSRYVDV

258
MSEEVTYADLQFQNSSEMEKIPEIGKFGEKAPPAPSHVWRPAALFLTLLCLLLLIGLGVLASMFHVTL

KIEMKKMNKLQNISEELQRNISLQLMSNMNISNKIRNLSTTLQTIATKLCRELYSKEQEHKCKPCPRR

WIWHKDSCYFLSDDVQTWQESKMACAAQNASLLKINNKNALEFIKSQSRSYDYWLGLSPEEDSTRGMR

VDNIINSSAWVIRNAPDLNNMYCGYINRLYVQYYHCTYKKRMICEKMANPVQLGSTYFREA

259
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGSDKTHTCPPCPAPELLGGPSVFLFPPK

PKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWL

NGKEYKCKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVKGFYPSDIAVEWE

SNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG

260
EVQLVESGGGLVKPGGSLRLSCAASGFTFSSYSMNWVRQAPGKGLEWVSSISSSSSYTYYADSVKGRF

TISRDNAKNSLYLQMNSLRAEDTAVYYCARGAPIGAAAGWFDPWGQGTLVTVSSASTKGPSVFPLAPS

SKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYI

CNVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGGPSVFLEPPKPKDTLMISRTPEVTCVVVDVS

HEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKT

ISKAKGQPREPQVCTLPPSRDELTENQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGS

FFLYSWLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG

261
DIQMTQSPSSVSASVGDRVTITCRASQGISSWLAWYQQKPGKAPKLLIYAASSLQSGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQGVSFPRTEGGGTKVEIKRTVAAPSVFIFPPSDEQLKSGTASVVCL1

NNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSP

VTKSFNRGEC

262
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVTISKDTSKNQ

FSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLVTVSSGSDKTHTCPPCPAPELLGGPSVFLFPPK

PKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWL

NGKEYKCKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVKGFYPSDIAVEWE

SNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG

263
EVQLVQSGAEVKKPGASVKVSCKASGYTFTSYYMHWVRQAPGQGLEWMGIINPSGGSTSYAQKFQGRV

TMTRDTSTSTVYMELSSLRSEDTAVYYCARGNYGDEFDYWGQGTLVTVSS

264
DIQMTQSPSSLSASVGDRVTITCRASQSISSYLNWYQQKPGKAPKLLIYAASSLQSGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQSYSTPPTFGQGTKVEIK

265
DIQMTQSPSSLSASVGDRVTITCRASQSVSTSSYNYMHWYQQKPGKPPKLLIKYASNLESGVPSRFSG

SGSGTDFTLTISSLQPEDFATYYCQHSWEIPLTFGQGTKVEIK

266
EVQLVQSGAEVKKPGASVKVSCKASGYSFTDYYMHWVRQAPGQGLEWIGRINPYNGAAFYSQNFKDRV

TLTVDTSTSTAYLELSSLRSEDTAVYYCAIERGADLEGYAMDYWGQGTLVTVSS

267
ATGGACATGAGGGTACCAGCTCAGCTGCTGGGCCTGCTGCTGCTTTGGCTTCCTGGCGCTAGATGCCA

ATTGCAGCTGCAAGAATCCGGACCAGGGCTCGTAAAACCTTCTGAAACCCTCTCCCTCACTTGCACTG

TGTCTGGATTCTCTCTCACGAACTACGGGTTGCATTGGATCCGACAGCCGCCTGGCAAGTGTCTTGAG

TGGATTGGCGTTATCTGGTCAGGAGGGAAGACGGATTATAATCCAAGTCTGAAATCCAGAGTGACAAT

TTCTAAGGACACGTCCAAAAATCAGTTTAGCCTCAAACTTTCCTCAGTTCAGGCTGCTGATACAGCGG

TGTACTATTGCGCAAAGTACGACTATGATGACAGTTTGGACTACTGGGGACAGGGTACACTGGTAACA

GTGAGTTCTGGAGGCGGCGGTAGTGGAGGCGGGGGAAGTGGTGGCGGAGGGAGTGGAGGAGGGGGCAG

TGATATTCAAATGACGCAGAGCCCCAGTTCCCTTTCAGCTAGCGTTGGCGACAGAGTAACGATAACGT

GTCACGCTTCCCAGAACATTAATTTTTGGCTTAGCTGGTATCAACAGAAACCGGGAAAAGCACCCAAG

CTGCTTATCTATGAAGCGTCCAATCTCCACACGGGTGTCCCATCAAGATTTTCTGGATCAGGCAGTGG

CACACGATTCACACTGACCATTAGTTCCTTGCAGCCCGAGGATATCGCAACTTACTACTGTCAGCAAA

GTCACTCTTACCCGTTGACATTCGGATGTGGAACTAAACTGGAGATTAAGGGATCCGATAAGACCCAC

ACCTGTCCTCCATGTCCTGCTCCAGAACTGCTCGGCGGACCTTCCGTGTTCCTGTTTCCTCCAAAGCC

TAAGGACACCCTGATGATCAGCAGGACCCCTGAAGTGACCTGCGTGGTGGTGGATGTGTCTCACGAGG

ACCCAGAAGTGAAGTTCAATTGGTACGTGGACGGCGTGGAAGTGCACAACGCCAAGACCAAGCCTAGA

GAGGAACAGTACAACAGCACCTACAGAGTGGTGTCCGTGCTGACCGTGCTGCACCAGGATTGGCTGAA

CGGCAAAGAGTACAAGTGCAAGGTGTCCAACAAGGCCCTGCCTGCTCCTATCGAGAAAACCATCAGCA

AGGCCAAGGGCCAGCCTCGCGAGCCTAGAGTGTATACCTTGCCTCCATGCCGGGACGAGCTGACCAAG

AATCAGGTGTCCCTGACCTGCCTGGTCAAGGGCTTCTACCCTTCCGATATCGCCGTGGAATGGGAGAG

CAATGGCCAGCCAGAGAACAACTACAAGACCACACCTCCTGTGCTGGTGTCCGACGGCAGCTTTACCC

TGTACAGCAAGCTGACAGTGGACAAGAGCAGATGGCAGCAGGGCAACGTGTTCTCCTGCAGCGTGATG

CACGAGGCCCTGCACAACCACTACACCCAGAAAAGCCTGAGCCTGTCTCCTGGCTAA

268
ATGGGCTGGTCCTGCATCATCCTGTTTCTGGTGGCCACAGCCACAGGCGTGCACTCTGAGGTGCAGCT

GGTTGAATCTGGCGGCGGACTTGTGAAGCCTGGCGGATCTCTGAGACTGAGCTGTGCCGCCAGCGGCT

TCACCTTTAGCAGCTACAGCATGAACTGGGTCCGACAGGCCCCTGGCAAAGGCCTTGAATGGGTGTCC

AGCATCAGCAGCAGCTCCAGCTACATCTACTACGCCGACAGCGTGAAGGGCAGATTCACCATCAGCCG

GGACAACGCCAAGAACAGCCTGTACCTGCAGATGAACTCCCTGAGAGCCGAGGACACCGCCGTGTACT

ATTGTGCTAGAGGCGCCCCTATTGGAGCCGCCGCTGGATGGTTCGATCCTTGGGGACAGGGAACCCTG

GTCACCGTTTCTTCTGCCTCTACAAAGGGCCCCAGCGTTTTCCCACTGGCTCCTAGCAGCAAGAGCAC

AAGCGGAGGAACAGCTGCCCTGGGCTGTCTGGTCAAGGACTACTTTCCTGAGCCTGTGACCGTGTCCT

GGAACAGCGGAGCACTGACTAGCGGCGTGCACACATTTCCAGCCGTGCTGCAAAGCAGCGGCCTGTAC

TCTCTGAGCAGCGTCGTGACAGTGCCTAGCAGCTCTCTGGGCACCCAGACCTACATCTGCAATGTGAA

CCACAAGCCTAGCAACACCAAGGTGGACAAGAAGGTGGAACCCAAGAGCTGCGACAAGACCCACACCT

GTCCTCCATGTCCTGCTCCAGAACTGCTCGGCGGACCTTCCGTGTTCCTGTTTCCTCCAAAGCCTAAG

GACACCCTGATGATCAGCAGAACCCCTGAAGTGACCTGCGTGGTGGTGGATGTGTCTCACGAGGACCC

CGAAGTGAAGTTCAATTGGTACGTGGACGGCGTGGAAGTGCACAACGCCAAGACCAAGCCTAGAGAGG

AACAGTACAACAGCACCTACAGAGTGGTGTCCGTGCTGACAGTGCTGCACCAGGATTGGCTGAACGGC

AAAGAGTACAAGTGCAAGGTGTCCAACAAGGCCCTGCCTGCTCCTATCGAGAAAACCATCAGCAAGGC

CAAGGGCCAGCCTCGCGAACCTCAAGTCTGTACACTGCCTCCTAGCCGGGATGAGCTGACCGAGAATC

AGGTGTCCCTGACCTGTCTCGTGAAGGGCTTCTACCCCTCCGATATCGCCGTGGAATGGGAGAGCAAT

GGCCAGCCAGAGAACAACTACAAGACAACCCCTCCTGTGCTGGACAGCGACGGCTCATTCTTCCTGTA

CAGCTGGCTGACCGTGGACAAGTCCAGATGGCAGCAGGGCAACGTGTTCTCCTGCAGCGTGATGCACG

AGGCCCTGCACAACCACTACACCCAGAAGTCCCTGTCTCTGAGCCCAGGCTAA

269
ATGGACATGAGAGTTCCAGCTCAGCTGCTGGGCCTGCTGCTGCTTTGGCTTCCTGGCGCTAGATGCGA

CATCCAGATGACACAGAGCCCCAGCTCCGTGTCTGCCTCTGTGGGAGACAGAGTGACCATCACCTGTA

GAGCCAGCCAGGGCATCTCTTCTTGGCTGGCCTGGTATCAGCAGAAGCCTGGCAAGGCCCCTAAGCTG

CTGATCTATGCCGCTAGCTCTCTGCAGTCTGGCGTGCCCTCTAGATTTTCTGGCAGCGGCTCTGGCAC

CGACTTCACCCTGACCATATCTAGCCTGCAGCCTGAGGACTTCGCCACCTACTATTGTCAGCAGGGCG

TGTCCTTTCCACGGACCTTTGGCGGCGGAACAAAGGTGGAAATCAAGCGGACAGTGGCCGCTCCTAGC

GTGTTCATCTTTCCACCTAGCGACGAGCAGCTGAAGTCCGGCACAGCCTCTGTTGTGTGCCTGCTGAA

CAACTTCTACCCCAGAGAAGCCAAGGTGCAGTGGAAGGTGGACAATGCCCTGCAGAGCGGCAACAGCC

AAGAGAGCGTGACAGAGCAGGACAGCAAGGACTCCACCTACAGCCTGAGCAGCACCCTGACACTGAGC

AAGGCCGACTACGAGAAGCACAAAGTGTACGCCTGCGAAGTGACCCACCAGGGCCTTTCTAGCCCTGT

GACCAAGAGCTTCAACCGGGGCGAGTGTTGA

270
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKGLEWIGVIWVGGATDYNPSLKSRVT

ISVDTSKNQFSLKLSSVQAADTAVYYCAKGDYGDTLDYWGQGTLVTVSS

271
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQSHSYPLTFGSGTKLEIK

272
WVGGA

273
GDYGDTLDY

274
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWVGGATDYNPSLKSRVT

ISVDTSKNQFSLKLSSVQAADTAVYYCAKGDYGDTLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGS

GTDFTLTISSLQPEDFATYYCQQSHSYPLTFGCGTKLEIK

275
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQSHSYPLTFGCGTKLEIKGGGGSGGGGSGGGGSGGGGSQLQLQESGP

GLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWVGGATDYNPSLKSRVTISVDTSKNQ

FSLKLSSVQAADTAVYYCAKGDYGDTLDYWGQGTLVTVSS

276
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWVGGATDYNPSLKSRVT

ISVDTSKNQFSLKLSSVQAADTAVYYCAKGDYGDTLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGS

GTDFTLTISSLQPEDFATYYCQQSHSYPLTEGCGTKLEIKGSDKTHTCPPCPAPELLGGPSVFLEPPK

PKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWL

NGKEYKCKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVKGFYPSDIAVEWE

SNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG

277
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVILSGGWTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQAADTAVYYCAKGDYGDALDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGSDKTHTCPPCPAPELLGGPSVFLFPPK

PKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWL

NGKEYKCKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVKGFYPSDIAVEWE

SNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG

278
EVQLVESGGGVVQPGGSLRLSCAASGETFNAFGMHWVRQAPGKCLEWVAFISSGSTSIYYANTVKGRF

TISRDNSKNTLYLQMNSLRAEDTAVYYCARDGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSG

GGGSDIQMTQSPSSLSASVGDRVTITCKASQDIYNYLSWFQQKPGKAPKPLIYRANILVSGVPSRFSG

SGSGQDYTFTISSLQPEDIATYYCLQFDAFPFTFGCGTKLEIKGSDKTHTCPPCPAPELLGGPSVFLF

PPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQ

DWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVKGFYPSDIAV

EWESNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG

279
GDYGDALDY

280
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVILSGGWTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQAADTAVYYCAKGDYGDALDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIK

281
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGG

SDIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKIPKLLIYEASNLHTGVPSRFSGSGS

GTRFTLTISSLQPEDIATYYCQQSHSYPLTFGCGTKLEIKGSDKTHTCPPCPAPELLGGPSVFLFPPK

PKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWL

NGKEYKCKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVKGFYPSDIAVEWE

SNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG

282
DIQMTQSPSTLSASVGDRVTITCRASNSISSWLAWYQQKPGKAPKLLIYEASSTKSGVPSRFSGSGSG

TEFTLTISSLQPDDFATYYCQQYDDLPTFGGGTKVEIK

283
YDYDDALDY

284
YDYDDILDY

285
YDYDDLLDY

286
DIQMTQSPSSVSASVGDRVTITCRASQGISSWLAWYQQKPGKAPKLLIYAASSLQSGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQGVSFPRTFGGGTKVEIKGGGGSGGGGSGGGGSGGGGSEVQLVESGG

GLVKPGGSLRLSCAASGETFSSYSMNWVRQAPGKGLEWVSSISSSSSYIYYADSVKGRFTISRDNAKN

SLYLQMNSLRAEDTAVYYCARGAPIGAAAGWFDPWGQGTLVTVSSGSDKTHTCPPCPAPELLGGPSVF

LFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVL

HQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPRVYTLPPCRDELTKNQVSLTCLVKGFYPSDI

AVEWESNGQPENNYKTTPPVLVSDGSFTLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSP

G

287
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKGLEWIGVIWSGGKTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQAADTAVYYCAKYDYDDSLDYWGQGTLVTVSSASTKGPSVFPLAPSSKSTS

GGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNH

KPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPE

VKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAK

GQPREPQVCTLPPSRDELTENQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYS

WLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPG

288
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGGGTKLEIKRTVAAPSVFIFPPSDEQLKSGTASVVCL1

NNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSP

VTKSFNRGEC

289
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGGGTKLEIK

290
YDYDDVLDY

291
YDYDDTLDY

292
YDYDESLDY

293
DYDDSLDY

294
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKGLEWIGVILSGGWTDYNPSLKSRVT

ISKDTSKNQFSLKLSSVQAADTAVYYCAKGDYGDALDYWGQGTLVTVSS

295
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TRFTLTISSLQPEDIATYYCQQSHSYPLTFGSGTKLEIK

296
LSGGW

297
GFTFSSY

298
SSSSSY

299
LSCAASGFTFSSYSMNWVR

300
NSLYLQMNSLR

301
DTLMISR

302
WQQGNVFSCSVMHEALHNHYTQK

303
DIQMTQSPSSVSASVGDR

304
YDYDDSLDYWGQGTLVTVSSGGGGSGGGGSGGGGSGGGGSDIQMTQSPSSLSASVGDRVTITCHASQN

INFWLSWYQQKPGKAPK

305
XDYDDSLDY, wherein X = any amino acid

306
YXYDDSLDY, wherein X = any amino acid

307
YDXDDSLDY, wherein X = any amino acid

308
YDYXDSLDY, wherein X = any amino acid

309
YDYDXSLDY, wherein X = any amino acid

310
YDYDDXLDY, wherein X = any amino acid

311
YDYDDSXDY, wherein X = any amino acid

312
YDYDDSLXY, wherein X = any amino acid

313
YDYDDSLDX, wherein X = any amino acid

314
XXYDDSLDY, wherein X = any amino acid

315
XDXDDSLDY, wherein X = any amino acid

316
XDYXDSLDY, wherein X = any amino acid

317
XDYDXSLDY, wherein X = any amino acid

318
XDYDDXLDY, wherein X = any amino acid

319
XDYDDSXDY, wherein X = any amino acid

320
XDYDDSLXY, wherein X = any amino acid

321
XDYDDSLDX, wherein X = any amino acid

322

G
DYGDSLDY

323

XSGGK, wherein X = any amino acid

324
WXGGK, wherein X = any amino acid

325
WSXGK, wherein X = any amino acid

326
WSGXK, wherein X = any amino acid

327
WSGGX, wherein X = any amino acid

328

XXGGK, wherein X = any amino acid

329

XSXGK, wherein X = any amino acid

330

XSGXK, wherein X = any amino acid

331

XSGGX, wherein X = any amino acid

332
WXXGK, wherein X = any amino acid

333
WXGXK, wherein X = any amino acid

334
WXGGX, wherein X = any amino acid

335
QLQLQESGPGLVKPSETLSLTCTVSGFSLTNYGLHWIRQPPGKCLEWIGVIWVGGATDYNPSLKSRVT

ISVDTSKNQFSLKLSSVQAADTAVYYCAKGDYGDTLDYWGQGTLVTVSS

336
DIQMTQSPSSLSASVGDRVTITCHASQNINFWLSWYQQKPGKAPKLLIYEASNLHTGVPSRFSGSGSG

TDFTLTISSLQPEDFATYYCQQSHSYPLTFGCGTKLEIK

337
RASQGISSWA

338
SSSSY

339
GAPMIGAAAGWFDP

340
LSCAASGFTFNAFGMHWVR

341
NTLYLQMNSLR

342
DGYPTGGAMDYWGQGTSVTVSSGGGGSGGGGSGGGGSGGGGSDIQMTQSPSSLSASVGDR

PROTEINS BINDING NKG2D, CD16 AND CLEC12A

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

International Classifications

Abstract

Description

Claims

Parent Case Info

Provisional Applications (1)