Anti-STEAP1 antigen-binding protein

Abstract

The present disclosure provides novel antigen binding proteins that bind to STEAP1 and methods of use.

Description

[Technical field]

[0001] The present disclosure provides novel antigen binding proteins that bind to six transmembrane epithelial antigen 1 of the prostate (STEAP1) and uses thereof.

[0002] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Provisional Patent Application No. 62 / 693,216, filed July 2, 2018, and U.S. Provisional Patent Application No. 62 / 800,259, filed February 1, 2019, the disclosures of which are incorporated by reference in their entireties herein.

[0003] Sequence Listing This application contains a sequence listing in computer readable form (Filename: 52601_Seqlisting.txt; Size: 298,914 bytes; Created: June 27, 2019), which is incorporated by reference in its entirety as a separate part of this disclosure. [Background technology]

[0004] Prostate cancer remains one of the most common cancers among men in the United States. US Cancer Statistics Working Group. United States Cancer Statistics: 1999-2014 Incidence and Mortality Web-based Report. Atlanta (GA): Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute; 2017. Although survival rates for prostate cancer are relatively high compared to other cancer types, existing treatment options come with risks and unwanted side effects. For example, surgery comes with risks of nerve damage and impotence, and radiation therapy can increase the risk of developing bladder or gastrointestinal cancer. Traditional chemotherapy comes with many side effects that limit the quality of life of patients during treatment. Summary of the Invention [Problem to be solved by the invention]

[0005] Antibody-based therapy has been successful in treating various diseases, including cancer and autoimmune / inflammatory disorders. Prostate cancer is considered to be particularly suitable for antibody-based therapy, at least in part, due to the presence of prostate cancer-specific antigens. Despite recent progress in elucidating the biological mechanisms underlying carcinogenesis and potential biomarkers, there is a need for alternative antibody-based treatment options for cancer, including prostate cancer. [Means for solving the problem]

[0006] The present disclosure provides an antigen binding protein that binds to STEAP1 of SEQ ID NO:2, comprising: (a) a heavy chain CDR comprising an amino acid sequence that differs by no more than 3, 2 or 1 amino acid from i) vhCDR1 SEQ ID NO:14, vhCDR2 SEQ ID NO:15, or vhCDR2 SEQ ID NO:21, and vhCDR3 SEQ ID NO:16; or ii) vhCDR1 SEQ ID NO:33, vhCDR2 SEQ ID NO:34, and vhCDR3 SEQ ID NO:35; or (b) a light chain CDR comprising i) vlCDR1 SEQ ID NO:11, vlCDR2 SEQ ID NO:12, and vlCDR3 SEQ ID NO:13; or ii) vlCDR1 SEQ ID NO:30, vlCDR2 SEQ ID NO:31, and vlCDR3 SEQ ID NO:32. or (c) a light chain variable domain comprising an amino acid sequence at least 90% identical to SEQ ID NO: 183 or SEQ ID NO: 186; or (d) a heavy chain variable domain comprising an amino acid sequence at least 90% identical to SEQ ID NO: 182, SEQ ID NO: 184 or SEQ ID NO: 185. In various embodiments, the antigen binding protein comprises a vhCDR1 comprising SEQ ID NO: 14, a vhCDR2 comprising SEQ ID NO: 15 or SEQ ID NO: 21, a vhCDR3 comprising SEQ ID NO: 16, a vlCDR1 comprising SEQ ID NO: 11, a vlCDR2 comprising SEQ ID NO: 12 and a vlCDR3 comprising SEQ ID NO: 13. Alternatively, the antigen binding protein comprises a vhCDR1 comprising SEQ ID NO: 33, a vhCDR2 comprising SEQ ID NO: 34, a vhCDR3 comprising SEQ ID NO: 35, a vlCDR1 comprising SEQ ID NO: 30, a vlCDR2 comprising SEQ ID NO: 31 and a vlCDR3 comprising SEQ ID NO: 32. In various embodiments, the antigen binding protein comprises a variable heavy chain domain comprising SEQ ID NO: 182 or SEQ ID NO: 184 and a variable light chain domain comprising SEQ ID NO: 183; for example, the antigen binding protein comprises a variable heavy chain domain comprising SEQ ID NO: 182 and a variable light chain domain comprising SEQ ID NO: 183 or a variable heavy chain domain comprising SEQ ID NO: 184 and a variable light chain domain comprising SEQ ID NO: 183. Alternatively, the antigen binding protein comprises a variable heavy chain domain comprising SEQ ID NO: 185 and a variable light chain domain comprising SEQ ID NO: 186.The disclosure further provides an antigen binding protein comprising a heavy chain comprising SEQ ID NO:201 and a light chain comprising SEQ ID NO:200; or a heavy chain comprising SEQ ID NO:203 and a light chain comprising SEQ ID NO:200.

[0007] The present disclosure further provides a heterodimeric antibody comprising: 1) a first monomer comprising a first variable heavy chain domain; 2) a first constant heavy chain comprising a first CH1 domain and a first Fc domain; and 3) a first monomer comprising a first heavy chain that binds human CD3 and comprises an scFv variable light chain domain, an scFv linker, and an scFv variable heavy chain domain, the scFv being covalently linked using a domain linker between the C-terminus of the CH1 domain and the N-terminus of the first Fc domain. The heterodimeric antibody further comprises a second monomer comprising a second heavy chain comprising a second constant heavy chain comprising a second variable heavy chain domain and a second Fc domain; and a common light chain comprising a variable light chain domain and a constant light chain domain.The first variable heavy chain domain and the variable light chain domain bind to human STEAP1 and the second variable heavy chain domain and the variable light chain domain bind to human STEAP1, (i) the first variable heavy chain domain and the second variable heavy chain domain comprise heavy chain CDRs comprising amino acid sequences that differ by no more than 3, 2 or 1 amino acid from vhCDR1 SEQ ID NO:14, vhCDR2 SEQ ID NO:15, or vhCDR2 SEQ ID NO:21 and vhCDR3 SEQ ID NO:16, and the variable light chain domain comprises light chain CDRs comprising amino acid sequences that differ by no more than 3, 2 or 1 amino acid from vlCDR1 SEQ ID NO:11, vlCDR2 SEQ ID NO:12, and vlCDR3 SEQ ID NO:13; or (ii) the first variable heavy chain domain and the second variable heavy chain domain comprise vhCDR1 SEQ ID NO:33, vhCDR2 SEQ ID NO:34, and vhCDR3 SEQ ID NO:16. and the variable light chain domain comprises a light chain CDR comprising an amino acid sequence that differs by no more than 3, 2 or 1 amino acid from SEQ ID NO:35; or (iii) the first variable heavy chain domain and the second variable heavy chain domain comprise an amino acid sequence that is at least 90% identical to SEQ ID NO:182 or SEQ ID NO:184, and the variable light chain domain comprises an amino acid sequence that is at least 90% identical to SEQ ID NO:183; or (iv) the first variable heavy chain domain and the second variable heavy chain domain comprise an amino acid sequence that is at least 90% identical to SEQ ID NO:185, and the variable light chain domain comprises an amino acid sequence that is at least 90% identical to SEQ ID NO:186. In various embodiments, the first variable heavy domain and the second variable heavy domain comprise the CDR sequences: vhCDR1 comprising SEQ ID NO: 14, vhCDR2 comprising SEQ ID NO: 15 or SEQ ID NO: 21, and vhCDR3 comprising SEQ ID NO: 16; and the variable light domain comprises the CDR sequences: vlCDR1 comprising SEQ ID NO: 11, vlCDR2 comprising SEQ ID NO: 12, and vlCDR3 comprising SEQ ID NO: 13.In various embodiments, the first variable heavy chain domain and the second variable heavy chain domain comprise CDR sequences: vhCDR1 comprising SEQ ID NO: 33, vhCDR2 comprising SEQ ID NO: 34, and vhCDR3 comprising SEQ ID NO: 35; and the variable light chain domain comprises CDR sequences: vlCDR1 comprising SEQ ID NO: 30, vlCDR2 comprising SEQ ID NO: 31, and vlCDR3 comprising SEQ ID NO: 32. In various embodiments, the first variable heavy chain domain and the second variable heavy chain domain comprise SEQ ID NO: 182 or SEQ ID NO: 184 and the variable light chain domain comprises SEQ ID NO: 183 (e.g., SEQ ID NOs: 182 and 183 or SEQ ID NOs: 184 and 183), or the first variable heavy chain domain and the second variable heavy chain domain comprise SEQ ID NO: 185 and the variable light chain domain comprises SEQ ID NO: 186. The scFv comprises CDRs comprising vhCDR1 comprising SEQ ID NO: 170, vhCDR2 comprising SEQ ID NO: 171, vhCDR3 comprising SEQ ID NO: 172, vlCDR1 comprising SEQ ID NO: 174, vlCDR2 comprising SEQ ID NO: 175 and vlCDR3 comprising SEQ ID NO: 176; or the variable heavy and light chain regions of SEQ ID NO: 169 and SEQ ID NO: 173.

[0008] Also provided is a method of treating cancer, such as prostate cancer, comprising administering to a subject in need thereof an antigen binding protein described herein.

[0009] The use of section headings herein is merely for convenience of reading and comprehension and is not intended to be limiting in itself It should be understood that the entire specification is intended to be considered as a unified disclosure and that all combinations of features described herein are contemplated.

[0010] As used herein, unless otherwise defined, scientific and technical terms used in connection with this application shall have the meanings commonly understood by those of ordinary skill in the art. Furthermore, unless the context requires otherwise, singular terms shall include the plural and plural terms shall include the singular. The terms "comprise", "have", "include" and "contain" shall be construed as open-ended terms unless otherwise stated. When an aspect of the invention is described as "comprising" a feature, the embodiment is also considered to "consist" or "consist essentially of" the feature. The use of any examples or exemplary language (e.g., "such as") provided herein is merely intended to further clarify the disclosure and does not impose limitations on the scope of the disclosure unless otherwise asserted. No language in this specification should be construed as indicating any non-claimed element as essential to the practice of the disclosure. Except as described in the examples or otherwise, all numbers indicating quantities of ingredients or reaction conditions used herein should be understood to be modified in all instances by the term "about" as that term would be interpreted by one of ordinary skill in the art.

[0011] The recitation of ranges of values ​​herein, unless otherwise stated herein, is merely intended to serve as a shorthand method of referring individually to each of the separate values ​​falling within that range, and each of the endpoints, all of which are incorporated herein as if each separate value was individually set forth herein.

[0012] Generally, the terms and techniques of cell and tissue culture, molecular biology, immunology, microbiology, genetics, protein and nucleic acid chemistry, manufacturing, formulation, pharmacology, and medicine described herein are well known and commonly used in the art. Unless otherwise specified, the methods and techniques of the present application are generally performed according to conventional methods well known in the art, and such methods and techniques are described in various general and more specific references cited and discussed throughout this specification. See, for example, Sambrook et al., Molecular Cloning: A Laboratory Manual, 3rd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (2001); Ausubel et al., Current Protocols in Molecular Biology, Greene Publishing Associates (1992); and Harlow and Lane, Antibodies: A Laboratory Manual Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1990), which are incorporated herein by reference. Enzymatic reactions and purification techniques are performed according to manufacturer's specifications, or as commonly accomplished in the art, or as described herein. The terminology used in connection with analytical chemistry, synthetic organic chemistry, and medicinal and pharmaceutical chemistry described herein, and the laboratory procedures and techniques thereof, are well known and commonly used in the art. Standard techniques may be used for chemical synthesis, chemical analysis, pharmaceutical preparation, formulation and delivery, and treatment of patients. Percent identity is calculated using methodologies routinely used in the art, including, for example, the methodologies described in U.S. Patent Application Publication No. 2017 / 0342155, which is incorporated herein by reference in its entirety and in particular with reference to paragraphs

[0075] to

[0083] . [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 shows a bispecific antibody of the present disclosure. [Diagram 2] FIG. 2 shows the sequence of the human CD3 epsilon chain (SEQ ID NO:1). [Diagram 3] Figure 3 shows the sequence of human STEAP1 (SEQ ID NO: 2). The sequences of the extracellular loops are underlined. [Figure 4-1] 4A-4E show useful pairs of heterodimerization mutant sets (including asymmetric and pI mutants). [Figure 4-2] 4A-4E show useful pairs of heterodimerization mutant sets (including asymmetric and pI mutants). [Figure 4-3] 4A-4E show useful pairs of heterodimerization mutant sets (including asymmetric and pI mutants). [Figure 4-4] 4A-4E show useful pairs of heterodimerization mutant sets (including asymmetric and pI mutants). [Figure 4-5] 4A-4E show useful pairs of heterodimerization mutant sets (including asymmetric and pI mutants). [Diagram 5] Figure 5 shows a list of antibody constant regions of isosteric variants and their corresponding substitutions. pI_(-) indicates variants with low pI and pI_(+) indicates variants with high pI. These can be optionally and independently combined with heterodimerization variants of the present disclosure (and other variant types as well as outlined herein). [Figure 6] FIG. 6 shows useful truncation mutants that ablate FcγR binding (sometimes referred to as “knockout” or “KO” mutants). [Figure 7] FIG. 7 describes two embodiments of the present disclosure. [Figure 8-1]8A-8B show useful linkers, including charged scFv linkers and domain linkers, that can be used in the antigen binding proteins and heterodimeric antibody formats provided herein. In various embodiments of the present disclosure, charged linkers are useful, for example, to raise or lower the pI of heterodimeric antibodies that utilize one or more scFvs as building blocks. A single prior art scFv linker with a single charge is referred to as "Whitlow" from Whitlow et al., Protein Engineering 6(8):989-995 (1993). This linker was used to reduce aggregation and increase proteolytic stability of scFvs. [Figure 8-2] 8A-8B show useful linkers, including charged scFv linkers and domain linkers, that can be used in the antigen binding proteins and heterodimeric antibody formats provided herein. In various embodiments of the present disclosure, charged linkers are useful, for example, to raise or lower the pI of heterodimeric antibodies that utilize one or more scFvs as building blocks. A single prior art scFv linker with a single charge is referred to as "Whitlow" from Whitlow et al., Protein Engineering 6(8):989-995 (1993). This linker was used to reduce aggregation and increase proteolytic stability of scFvs. [Figure 9] Figure 9 shows a list of engineered heterodimer-asymmetric Fc variants along with heterodimer yield (determined by HPLC-CIEX) and thermal stability (determined by DSC). Thermal stability not determined is indicated by "nd". Additional information is provided in U.S. Patent No. 9,822,186, which is incorporated herein by reference in its entirety. [Figure 10-1]Figures 10A-10B show stability-optimized humanized anti-CD3 variant scFvs. Substitutions are made to the H1_L1.4 scFv sequence. Amino acid numbering is Kabat numbering. Specific variable light and variable heavy chain regions are described. The listed substitutions may be utilized for variable light and variable heavy chain regions other than those specificities listed. Additional information is provided in WO 2017 / 091656, which is incorporated herein by reference in its entirety. [Figure 10-2] Figures 10A-10B show stability-optimized humanized anti-CD3 variant scFvs. Substitutions are made to the H1_L1.4 scFv sequence. Amino acid numbering is Kabat numbering. Specific variable light and variable heavy chain regions are described. The listed substitutions may be utilized for variable light and variable heavy chain regions other than those specificities listed. Additional information is provided in WO 2017 / 091656, which is incorporated herein by reference in its entirety. [Figure 11] 11A-11B show specific detection of STEAP1 on the surface of C4-2B luc cells by mouse STEAP1 antibody Ab-Am. [Figure 12-1] Figures 12A-12C show specific detection of STEAP1 on C4-2B luc prostate cancer cells using mouse STEAP1 antibody Ab-Am (Figure 12A); Ab-A1 XmAb (●) or Ab-A1 XmAb2+1 (■) (Figure 12B) and antibody Ab-A1 XmAb2+1 (Figure 12C) for 1 hour at 4°C. [Figure 12-2] Figures 12A-12C show specific detection of STEAP1 on C4-2B luc prostate cancer cells using mouse STEAP1 antibody Ab-Am (Figure 12A); Ab-A1 XmAb (●) or Ab-A1 XmAb2+1 (■) (Figure 12B) and antibody Ab-A1 XmAb2+1 (Figure 12C) for 1 hour at 4°C. [Figure 13] FIG. 13 shows specific detection of STEAP1 on C4-2B luc prostate cancer by STEAP1 antibody Ab-Bx (Ab-B1 XmAb). [Figure 14]Figures 14A-14C show that STEAP1 antibody Ab-Ax (Figure 14A), STEAP1 antibody Ab-A1 Xmab2+1 (Figure 14B) and STEAP1 antibody Ab-A2(N67Q) Xmab2+1 (Figure 14C) mediated target cell lysis of the human tumor cell line C4-2B luc by human T cells. [Figure 15] FIG. 15 shows that STEAP1 antibodies Ab-A1 Xmab2+1 and Ab-A2(N67Q)Xmab2+1 mediated dose-dependent target cell lysis of the human tumor cell line C4-2B, but not C4-2B luc STEAP1 KO, by human T cells. [Figure 16] Figures 16A-16B show that mouse STEAP1 antibody Ab-Am detected STEAP1 expressed by the 293T cells tested. Figure 16C shows that the STEAP1 conjugate Ab-A2(N67Q)Xmab2+1 mediated dose-dependent target cell lysis of the human cell line 293T stably transfected with human STEAP1 but not the parent human 293T cell line. [Figure 17-1] Figure 17A shows that Ab-Bx (Ab-B1-XmAb) and Ab-B1 Xmab2+1 mediated target cell lysis of C4-2B luc prostate cancer cells. Figure 17B shows that Xmab2+1 Ab-B1 mutants (i.e., Ab-B1-G37A, Ab-B1-S39A, and Ab-B1-G37A / S39A) mediated target cell lysis of C4-2B luc prostate cancer cells. [Figure 17-2] FIG. 17C shows that Xmab2+1 Ab-B1 mutants (i.e., Ab-B1-G37A, Ab-B1-S39A and Ab-B1-G37A / S39A) did not mediate target cell lysis of C4-2B luc STEAP1 knockout prostate cancer cells. [Figure 18-1]Figures 18A-18I show several formats of antigen binding proteins: "bottle opener" format, mAb-Fv, mAb-scFv, central-scFv, central-Fv, 1-arm central-scFv, 1 scFv-mAb, scFv-mAb and dual-scFv. For all of the scFv domains shown, they can be variable heavy-(optional linker)-variable light from N to C terminus or vice versa. In addition, for the 1-arm scFv-mAb, the scFv can be attached either to the N-terminus of the heavy chain monomer or to the N-terminus of the light chain. [Figure 18-2] Figures 18A-18I show several formats of antigen binding proteins: "bottle opener" format, mAb-Fv, mAb-scFv, central-scFv, central-Fv, 1-arm central-scFv, 1 scFv-mAb, scFv-mAb and dual-scFv. For all of the scFv domains shown, they can be variable heavy-(optional linker)-variable light from N to C terminus or vice versa. In addition, for the 1-arm scFv-mAb, the scFv can be attached either to the N-terminus of the heavy chain monomer or to the N-terminus of the light chain. [Figure 19-1] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-2] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-3] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-4] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-5]Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-6] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-7] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-8] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-9] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-10] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-11] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-12] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-13] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-14]Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-15] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-16] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-17] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-18] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-19] Figure 19 provides the sequences of the CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences of the present disclosure. The underlines in the variable region sequences refer to the CDR sequences. [Figure 19-20] Figure 19 provides the sequences of the disclosed CDRs, variable heavy domains, variable light domains, scFvs, linker sequences and monomer sequences. The underlines in the variable region sequences refer to the CDR sequences. [Figure 20]Figures 20A-20B show the results of the T cell-dependent cytotoxicity assay described in Example 9. Figure 20A is a graph showing the specific cytotoxicity (%) mediated by Ab-A2(N67Q)XmAb2+1 alone (open circles) and in combination with anti-PD-1 antibody (closed circles) in one representative T cell donor (y-axis=Log(pM)). Figure 20B shows the EC50 (pM) of Ab-A2(N67Q)XmAb2+1 alone (left) and in combination with anti-PD-1 antibody from four different T cell donors. [Figure 21] Figures 21A-21C are line graphs showing PD-1 expression (%CD3+) in total T cells (Figure 21A), CD8+ T cells (Figure 21B), and CD4+ T cells (Figure 21C) exposed to various amounts of Ab-A2(N67Q)XmAb2+1. Circles and squares in the graphs represent different donors of T cells. PD-1 expression is increased in T cells exposed to heterodimeric antibodies of the present disclosure. [Figure 22] FIG. 22 is a line graph showing tumor volume (mm3; y-axis) over time (days on study, x-axis). Human SK-N-MC cells (5×106 cells / mouse) were injected subcutaneously into the right dorsal flank of female sublethally irradiated NOD / SCID mice on day 1. On day 8, human CD3+ T cells (2×107 cells / mouse) were injected into the peritoneal cavity of all animals except group 1. Vehicle (groups 1 and 2) or Ab-A2(N67Q)XmAb2+ at dose levels of 1.0, 0.1 or 0.01 mg / kg (groups 3, 4, 5, respectively) were administered by intravenous bolus injection on days 12, 19 and 26 (arrows at the top of the graph). Tumor volumes were measured 3 times / week using electronic calipers. Group mean tumor volume [mm3]+ / -SEM is shown. Asterisks in the figures denote statistically significant differences between vehicle (group 2) and Ab-A2(N67Q)XmAb2+ treatment groups (one-way ANOVA; *=p<0.05; ***=p<0.001). [Diagram 23] FIG. 23. Mean and median SK-N-MC human neuroblastoma tumor volumes in female NOD / SCID mice. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] STEAP1 is a 339 amino acid protein that contains six transmembrane domains resulting in three extracellular and two intracellular loops. The amino acid sequence of human STEAP1 is set forth herein as SEQ ID NO:2. The predicted locations of the extracellular loops are amino acids 92-118 (extracellular loop 1), amino acids 185-217 (extracellular loop 2) and amino acids 279-290 (extracellular loop 3). STEAP1 is differentially expressed in prostate cancer compared to normal tissues, and increased expression has been observed in bone and lymph node prostate cancer metastases compared to primary prostate cancer samples. STEAP1 represents an ideal target for diagnostics and antibody-based therapeutics, such as, for example, bispecific anti-STEAP1 / anti-CD3 T cell recruiting antibodies to induce T cell-dependent cytotoxic activity or redirected lysis of prostate cancer cells. The present disclosure provides antigen binding proteins that bind to STEAP1, as further described herein.

[0015] Antigen-binding proteins An "antigen binding protein" is a protein that includes a portion that binds to a designated target antigen (such as STEAP1). An antigen binding protein includes a scaffold or framework portion that enables the antigen binding portion to adopt a conformation that promotes binding of the antigen binding protein to an antigen. In exemplary aspects, the antigen binding protein is an antibody or immunoglobulin (e.g., a heterodimeric and / or bispecific antibody), or an antigen-binding antibody fragment, or an antibody protein product.

[0016] The term "antibody" refers to an intact antigen-binding immunoglobulin. An "antibody" is a type of antigen-binding protein. The antibody can be an IgA, IgD, IgE, IgG, or IgM antibody, including any one of IgG1, IgG2, IgG3, or IgG4. In various embodiments, an intact antibody comprises two full-length heavy chains and two full-length light chains. An antibody has a variable region and a constant region. In the IgG format, the variable region is generally about 100-110 or more amino acids, contains three complementarity determining regions (CDRs), and is primarily responsible for antigen recognition and varies substantially among other antibodies that bind different antigens. The variable region usually comprises at least three heavy or light chain CDRs (Kabat et al., 1991, Sequences of Proteins of Immunological Interest, Public Health Service NIH, Bethesda, Md.; see also Chothia and Lesk, 1987, J. Mol. Biol. 196:901-917; Chothia et al., 1989, Nature 342:877-883), which are located within framework regions (termed framework regions 1-4, FR1, FR2, FR3 and FR4 by Kabat et al., 1991; see also Chothia and Lesk, 1987, supra). The constant region enables the antibody to recruit cells and molecules of the immune system.

[0017] In various aspects, the antibody is a monoclonal antibody. In certain aspects, the antibody is a human antibody. In certain aspects, the antibody (or other antigen binding protein) is chimeric or humanized. The term "chimeric" refers to an antibody that contains domains from two or more different antibodies. A chimeric antibody can, for example, contain a constant domain from one species and a variable domain from a second species, or more commonly, contain a contiguous amino acid sequence from at least two species. Both "chimeric" and "humanized" often refer to antigen binding proteins that combine regions from two or more species. A chimeric antibody can also contain domains from two or more different antibodies within the same species. In one embodiment, a chimeric antibody is a CDR-grafted antibody.

[0018] The term "humanized" when used in reference to an antigen-binding protein refers to an antigen-binding protein (e.g., an antibody) having at least a CDR region derived from a non-human source that has been engineered to have a structure and immune function more similar to a true human antibody than the original source antibody. For example, humanization can involve grafting CDRs derived from a non-human antibody, such as a mouse antibody, into human framework regions. Generally, in a humanized antibody, the entire antibody except for the CDRs is encoded by a polynucleotide of human origin or is identical to such an antibody except within its CDRs. The CDRs, some or all of which are encoded by a nucleic acid derived from a non-human organism, are grafted into the beta-sheet framework of a human antibody variable region to generate an antibody whose specificity is determined by the grafted CDRs. The production of such antibodies is described, for example, in WO 92 / 11018; Jones, 1986, Nature 321:522-525; and Verhoeyen et al., 1988, Science 239:1534-1536, all of which are incorporated by reference in their entireties. "Backmutation" of selected acceptor framework residues to the corresponding donor residues is often utilized to restore affinity lost in the initial grafted construct (see, e.g., U.S. Pat. Nos. 5,530,101; 5,585,089; 5,693,761; 5,693,762; 6,180,370; 5,859,205; 5,821,337; 6,054,297; and 6,407,213, all of which are incorporated by reference in their entireties). The humanized antibody optimally also will comprise at least a portion of an immunoglobulin constant region, typically that of a human immunoglobulin, and thus typically will comprise a human Fc region.

[0019] Various techniques and methods for producing chimeric antibodies, humanized antibodies, and for reconstituting non-human antibodies are well known in the art. Tsurushita & Vasquez, 2004, Humanization of Monoclonal Antibodies, Molecular Biology of B Cells, 533-545, Elsevier Science (USA) and references cited therein, all of which are incorporated by reference in their entirety; Jones et al., 1986, Nature 321:522-525; Riechmann et al., 1988, Nature 332:323-329; Verhoeyen et al., 1988, Science, 239:1534-1536; Queen et al., 1989, Proc Natl Acad Sci, USA 86:10029-33; He et al., 1998, J. Immunol. 160:1029-1035; Carter et al., 1992, Proc Natl Acad Sci USA See, e.g., Presta et al., 1997, Cancer Res. 57(20):4593-9; Gorman et al., 1991, Proc. Natl. Acad. Sci. USA 88:4181-4185; O'Connor et al., 1998, Protein Eng 11:321-8, U.S. Patent Application Publication No. 20030039649; U.S. Patent Nos. 5,869,619, 5,225,539, 5,821,337, and 5,859,205; Padlan et al., 1995, FASEB J. 9:133-39; and Tamura et al., 2000, J. Immunol. 164:1432-41. Humanization or other methods of reducing the immunogenicity of non-human antibody variable regions may include, for example, the resurfacing method as described in Roguska et al., 1994, Proc. Natl. Acad. Sci. USA 91:969-973, which is incorporated by reference in its entirety. The parent antibody may be affinity matured, which is well understood in the art.Structure-based methods can be utilized for humanization and affinity maturation, for example, as described in US Patent Application Publication No. 20060008883. Selection-based methods can be utilized to humanize and / or affinity mature antibody variable regions, including, but not limited to, those described in Wu et al., 1999, J. Mol. Biol. 294:151-162; Baca et al., 1997, J. Biol. Chem. 272(16):10678-10684; Rosok et al., 1996, J. Biol. Chem. 271(37):22611-22618; Rader et al., 1998, Proc. Natl. Acad. Sci. USA 95:8910-8915; Krauss et al., 2003, Protein Engineering 16(10):753-759, all of which are incorporated by reference in their entirety. Humanization can also include selecting amino acid substitutions that make the non-human sequence more similar to the human sequence. Other humanization methods may involve grafting only portions of the CDRs, including, but not limited to, those described in Tan et al., 2002, J. Immunol. 169:1119-1125; De Pascalis et al., 2002, J. Immunol. 169:3076-3084, all of which are incorporated by reference in their entirety.

[0020] In other embodiments, the antigen-binding protein is an antigen-binding antibody fragment, i.e., a fragment of an antibody lacking a portion or all of the antibody light chain and / or a portion or all of the antibody heavy chain. Antibody fragments can be produced recombinantly or can be made by cleaving an intact antibody using enzymes such as papain and pepsin. Papain cleaves an antibody to yield two Fab fragments and an Fc fragment. Pepsin cleaves an antibody to yield an F(ab')2 fragment and a pFc' fragment. In an exemplary case, the antigen-binding antibody fragment is a Fab fragment or an F(ab')2 fragment. A Fab fragment is a monovalent fragment having a VL, VH, CL and CH1 domain. Fab can refer to this region in isolation or in the context of a full-length antibody, antibody fragment, etc. A F(ab')2 fragment is a bivalent fragment having two Fab fragments linked by a disulfide bridge at the hinge region.

[0021] The structure of antibodies has been exploited to create a wide range of alternative forms, spanning a molecular weight range of at least about 12-150 kDa, and ranging in valency (n) from monomers (n=1) to dimers (n=2), trimers (n=3), tetramers (n=4), and potentially even higher; such alternative forms are referred to herein as "antibody protein products" and are examples of antigen-binding proteins. Antibody protein products include those based on the complete antibody structure and those that mimic antibody fragments that retain full antigen-binding ability, such as scFv and VHH / VH (discussed below). A single-chain antibody (scFv) is an antibody in which the VL and VH domains are joined via a linker (e.g., a synthetic sequence of amino acid residues, usually about 15 amino acids in length) to form a continuous protein chain, which linker is long enough to enable the protein chain to fold on itself and form a monovalent antigen-binding site (see, e.g., Bird et al., 1988, Science 242:423-26 and Huston et al., 1988, Proc. Natl. Acad. Sci. USA 85:5879-83).

[0022] The antigen-binding fragment that retains the complete antigen-binding site is the Fv fragment, which consists entirely of the variable (V) region (the VL and VH domains of a single antibody). To link the V region to the scFv (single-chain fragment variable) fragment for molecular stability, a soluble flexible amino acid peptide linker is often used, or a constant (C) domain is added to the V region to generate a Fab fragment (fragment, antigen-binding). scFv and Fab fragments can be easily produced in host cells, e.g., prokaryotic or eukaryotic host cells. Other antibody protein products include disulfide bond stabilized scFv (ds-scFv), single chain Fab (scFab), single chain antibodies (SCAs), domain antibodies (dAbs) (e.g. peptides containing the VH domain, the VL domain or antigen binding fragments of the VH or VL domain), peptides containing Fd fragments (containing the VH and CH1 domains), complementarity determining region (CDR) fragments and dimeric and multimeric antibody formats such as diabodies, triabodies and tetrabodies or minibodies (miniAbs) including different formats consisting of scFv linked to oligomerization domains. The smallest fragments are the VHH / VH of camelid heavy chain Abs and single domain Abs (sdAbs). Peptibodies or peptide-Fc fusions are yet another antibody protein product. The structure of a peptibody consists of a biologically active peptide grafted onto an Fc domain. Peptibodies are further described in the art. See, for example, Shimamoto et al., mAbs 4(5):586-591 (2012).

[0023] Alternatively, the antibody protein product may comprise an alternative protein scaffold or an artificial scaffold, for example with grafted CDRs or CDR derivatives. Such scaffolds include, but are not limited to, scaffolds derived from antibodies, for example with mutations introduced to stabilize the three-dimensional structure of the antigen-binding protein, and fully synthetic scaffolds, for example with biocompatible polymers. See, for example, Korndorfer et al., 2003, Proteins: Structure, Function, and Bioinformatics, Volume 53, Issue 1:121-129; Roque et al., 2004, Biotechnol. Prog. 20:639-654. In addition, peptide antibody mimetics ("PAMs") may be used as scaffolds, as well as scaffolds based on antibody mimetics that utilize fibronectin components.

[0024] In various embodiments, the antigen binding protein comprises a heavy chain CDR comprising an amino acid sequence that differs by no more than 3, 2 or 1 amino acid from i) vhCDR1 SEQ ID NO: 14, vhCDR2 SEQ ID NO: 15, or vhCDR2 SEQ ID NO: 21 and vhCDR3 SEQ ID NO: 16, or ii) vhCDR1 SEQ ID NO: 33, vhCDR2 SEQ ID NO: 34 and vhCDR3 SEQ ID NO: 35; and / or a light chain CDR comprising an amino acid sequence that differs by no more than 3, 2 or 1 amino acid from i) vlCDR1 SEQ ID NO: 11, vlCDR2 SEQ ID NO: 12 and vlCDR3 SEQ ID NO: 13; or ii) vlCDR1 SEQ ID NO: 30, vlCDR2 SEQ ID NO: 31 and vlCDR3 SEQ ID NO: 32. Each such sequence difference is independently either a deletion, insertion or substitution, although substitutions (e.g., conservative substitutions) are preferred. Examples of conservative substitutions include, but are not limited to, exchanges within the following groups: small aliphatic non-polar or slightly polar residues, Ala, Ser, Thr, Pro, Gly; polar, negatively charged residues and their amides and esters, Asp, Asn, Glu, Gln, cysteic acid and homocysteic acid; polar, positively charged residues, His, Arg, Lys, ornithine (Orn); large aliphatic non-polar residues, Met, Leu, Ile, Val, Cys, norleucine (Nle), homocysteine; and large aromatic residues: Phe, Tyr, Trp, acetylphenylalanine.

[0025] In various embodiments, the antigen binding protein comprises the following CDR sequences: a) vhCDR1 comprising SEQ ID NO: 14, vhCDR2 comprising SEQ ID NO: 15 or SEQ ID NO: 21 and vhCDR3 comprising SEQ ID NO: 16, or b) vhCDR1 comprising SEQ ID NO: 33, vhCDR2 comprising SEQ ID NO: 34 and vhCDR3 comprising SEQ ID NO: 35. Alternatively or in addition, the antigen binding protein comprises the following CDR sequences: a) vlCDR1 comprising SEQ ID NO: 11, vlCDR2 comprising SEQ ID NO: 12 and vlCDR3 comprising SEQ ID NO: 13, or b) vlCDR1 comprising SEQ ID NO: 30, vlCDR2 comprising SEQ ID NO: 32 and vlCDR3 comprising SEQ ID NO: 33.

[0026] Thus, in various embodiments, the antigen binding protein comprises a vhCDR1 comprising SEQ ID NO: 14, a vhCDR2 comprising SEQ ID NO: 15 or SEQ ID NO: 21, a vhCDR3 comprising SEQ ID NO: 16, a vlCDR1 comprising SEQ ID NO: 11, a vlCDR2 comprising SEQ ID NO: 12 and a vlCDR3 comprising SEQ ID NO: 13. In alternative embodiments, the antigen binding protein comprises a vhCDR1 comprising SEQ ID NO: 33, a vhCDR2 comprising SEQ ID NO: 34, a vhCDR3 comprising SEQ ID NO: 35, a vlCDR1 comprising SEQ ID NO: 30, a vlCDR2 comprising SEQ ID NO: 31 and a vlCDR3 comprising SEQ ID NO: 32.

[0027] In various embodiments, the antigen binding protein comprises a light chain variable domain comprising an amino acid sequence at least 90% identical (e.g., at least 95% identical or 100% identical) to SEQ ID NO: 183 or SEQ ID NO: 186; and / or a heavy chain variable domain comprising an amino acid sequence at least 90% identical (e.g., at least 95% identical or 100% identical) to SEQ ID NO: 182, SEQ ID NO: 184, or SEQ ID NO: 185. For example, the antigen binding protein may comprise (i) SEQ ID NO: 183 and SEQ ID NO: 182, (ii) SEQ ID NO: 184 and SEQ ID NO: 183, or (iii) SEQ ID NO: 185 and SEQ ID NO: 186. In various aspects, the antigen binding protein comprises a light chain variable region and / or a heavy chain variable region comprising a sequence of amino acids that differs from the foregoing amino acid sequences by only 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 residue, each such sequence difference being, independently, either a deletion, insertion, or substitution (e.g., a conservative substitution). In various embodiments, the sequence differences are located outside of the CDRs (eg, within the framework regions).

[0028] In various embodiments, the antigen binding protein comprises a light chain that comprises an amino acid sequence at least 90% identical (e.g., at least 95% identical or 100% identical) to SEQ ID NO: 17 or SEQ ID NO: 36; and / or a heavy chain that comprises an amino acid sequence at least 90% identical (e.g., at least 95% identical or 100% identical) to SEQ ID NO: 18, SEQ ID NO: 199, or SEQ ID NO: 37. For example, the antigen binding protein can comprise: (i) SEQ ID NO: 17 and SEQ ID NO: 18; (ii) SEQ ID NO: 17 and SEQ ID NO: 199; or (iii) SEQ ID NO: 36 and SEQ ID NO: 37.

[0029] In various embodiments, the antigen binding protein comprises a light chain comprising an amino acid sequence at least 90% identical (e.g., at least 95% identical or 100% identical) to SEQ ID NO:200 or SEQ ID NO:204; and / or a heavy chain comprising an amino acid sequence at least 90% identical (e.g., at least 95% identical or 100% identical) to SEQ ID NO:201, SEQ ID NO:203 or SEQ ID NO:205. For example, the antigen binding protein may comprise: (i) SEQ ID NO:200 and SEQ ID NO:201; (ii) SEQ ID NO:200 or SEQ ID NO:203; (iii) SEQ ID NO:204 and SEQ ID NO:205.

[0030] Competition, epitopes, and binding affinity The antigen-binding protein binds to STEAP1 of SEQ ID NO: 2. Specific binding (i.e., binding to STEAP1 that is measurably different from non-specific interactions) can be measured, for example, by measuring binding of a molecule compared to binding of a control molecule, which is a molecule of similar structure that generally does not have binding activity. For example, specific binding can be measured by competition with a control molecule that is similar to the target.

[0031] The binding affinity of an antigen-binding protein to STEAP1 can be described in terms of dissociation constant (Kd). In exemplary embodiments, the Kd of the antigen-binding proteins provided herein is micromolar, nanomolar, picomolar, or femtomolar. Typically, an antigen-binding protein that specifically binds to an antigen will have a Kd that is 20-fold, 50-fold, 100-fold, 500-fold, 1000-fold, 5,000-fold, 10,000-fold or more for the target antigen or epitope than a control molecule. Specific binding to a particular antigen can also be demonstrated, for example, by an antibody having a K or Ka for an antigen or epitope that is at least 20-fold, 50-fold, 100-fold, 500-fold, 1000-fold, 5,000-fold, 10,000-fold or more for the epitope compared to a control, where K or Ka refers to the binding rate of a particular antibody-antigen interaction. In exemplary embodiments, the KD of the antigen-binding proteins provided herein for STEAP1 is at least 10 -7 M or less, 10 -8 M or less, 10 -9 M or less, 10 -10 M or less, 10 -11 M or less or 10 -12 For example, the KD of the antigen binding protein is optionally about 10 -4 ~10 -6 M, or about 10 -7 ~10 -9 M, or about 10 -10 ~10 -12 M, or about 10 -7 ~10 -12 , or about 10 -9 ~10 -12 , or about 10 -13 ~10 -15 Alternatively (or in addition), the antigen binding protein has a low dissociation rate from STEAP1. In some embodiments, the antigen binding protein has a molecular weight of 1×10 -4 s -1 The following K off In another embodiment, K off is 5 x 10 -5 s -1In various embodiments, the antigen binding proteins differentiate between target cells expressing high levels of STEAP1 and those off-target cells that exhibit less STEAP1. For example, in various embodiments, the antigen binding proteins preferentially bind to cells that contain more than about 100,000 STEAP receptors per cell (e.g., about 200,000 STEAP1 receptors per cell) to about 10,000 STEAP1 receptors per cell. It will be understood that the present disclosure regarding competition, binding affinity and binding specificity related to STEAP1 also applies to binding of multispecific antigen binding proteins to a second or third antigen (e.g., CD3) or a different antibody used in combination with the anti-STEAP1 antigen binding protein. For example, in an exemplary embodiment, the Kd of the antigen binding proteins provided herein for CD3 (or PD-1 as described below) is greater than 10 -7 M or less, 10 -8 M or less, 10 -9 M or less, 10 -10 M or less, 10 -11 M or less or 10 -12 For example, the Kd of the antigen binding protein is optionally about 10 -4 ~10 -6 M, or about 10 -7 ~10 -9 M, or about 10 -10 ~10 -12 M, or about 10 -7 ~10 -12 , or about 10 -9 ~10 -12 , or about 10 -13 ~10 -15 Alternatively (or in addition), the antigen binding protein has a low dissociation rate from CD3. In some embodiments, the antigen binding protein has a molecular weight in the range of 1×10 -4 s -1 The following K off In another embodiment, K off is 5 × 10 for CD3 -5 s -1 The following is the result.

[0032] The present disclosure relates to antigen binding proteins as described herein (e.g., XmAb as described herein). 2+1Further provided are antigen binding proteins (e.g., antibodies) that compete for binding to STEAP1 with any of the antigen binding proteins described herein (Ab-A, Ab-A1, Ab-A2, Ab-B or Ab-B1, including in any of the formats described herein). In other words, the present disclosure provides antigen binding proteins that cross-block the binding of a reference antigen binding protein described herein to STEAP1 or are cross-blocked from binding to STEAP1 by a reference antigen binding protein. By "compete" it is meant that one antigen binding protein prevents, reduces or inhibits the binding of the reference antigen binding protein to STEAP1. Many types of competitive binding assays can be used, such as surface plasmon resonance, solid-phase direct or indirect radioimmunoassays (RIA), solid-phase direct or indirect enzyme immunoassays (EIA), sandwich competition assays (see, e.g., Stahli et al., 1983, Methods in Enzymology 9:242-253), solid-phase direct biotin-avidin EIA (see, e.g., Kirkland et al., 1986, J. Immunol. 137:3614-3619), solid-phase direct label assays, solid-phase direct label sandwich assays (see, e.g., Harlow and Lane, 1988, Antibodies, A Laboratory Manual, Cold Spring Harbor Press); solid-phase direct label RIA using 1-125 label (see, e.g., Morel et al., 1988, Antibodies, A Laboratory Manual, Cold Spring Harbor Press); al., 1988, Molec. Immunol. 25:7-15); solid phase direct biotin-avidin EIA (see, e.g., Cheung, et al., 1990, Virology 176:546-552); and direct labeling RIA (Moldenhauer et al., 1990, Scand. J. Immunol. 32:77-82). Typically, such assays involve the use of purified antigen bound to a solid surface or exposed on cells, an unlabeled test antigen-binding protein, and a labeled reference antigen-binding protein. Competitive inhibition is measured by determining the amount of label bound to the solid surface or cells in the presence of the test antigen-binding protein. Usually, the test antigen-binding protein is present in excess.Antigen binding proteins identified by competitive assays (competing antigen binding proteins) include antigen binding proteins that bind to the same epitope as the reference antigen binding protein, epitopes that overlap with the epitope recognized by the reference antigen binding protein, and epitopes that do not overlap but allow for steric hindrance to occur between the test antigen binding protein and the reference antigen binding protein. Usually, when a competing antigen binding protein is present in excess, it will inhibit the binding of the reference antigen binding protein to a common antigen by at least 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70% or 75%. In some cases, binding is inhibited by at least 80%, 85%, 90%, 95% or 97% or more. In at least one embodiment, the antigen binding protein (e.g., an antibody) is coupled to a reference antigen binding protein (e.g., optionally in a bispecific antibody format, such as a bispecific antibody format described in the Examples (e.g., XmAb), such that binding of the reference antigen binding protein to STEAP1 is reduced by at least 80% or at least 90%. 2+1 ) competes with Ab-A, Ab-A1, Ab-A2, Ab-B, or Ab-B1) described herein.

[0033] The antigen binding protein binds to STEAP1 of SEQ ID NO:2. A competing (or cross-blocking) antigen binding protein may bind to an epitope that overlaps with the epitope recognized by the reference antigen binding protein or an epitope that does not overlap but allows for steric hindrance to occur between the test and reference antigen binding proteins. In various embodiments, the antigen binding protein is Ab-A, Ab-A1, Ab-A2(N67Q), Ab-B, or Ab-B1 described herein or a bispecific or heterodimeric version thereof (e.g., Ab-A1 XmAb 2+1 , Ab-A2(N67Q)XmAb 2+1 or Ab-B1 XmAb 2+1). For example, the antigen binding proteins of the present disclosure optionally bind STEAP1 at a region outside the second extracellular loop. The antigen binding proteins, in at least one embodiment, bind to a region of STEAP1 within amino acids 92-118 (extracellular loop 1) and / or amino acids 279-290 (extracellular loop 3). In various aspects, the present disclosure provides antigen binding proteins that bind to regions of STEAP1 within amino acids 92-118 and amino acids 279-290. Also, optionally, the antigen binding proteins do not bind to STEAP2 (UniProtKB No. Q8NFT2; SEQ ID NO: 177). Optionally, the epitope of the reference antigen binding protein and / or the antigen binding protein being tested can be determined by analyzing the x-ray plasma structure of the antigen binding protein bound to STEAP1 or a portion thereof. In one such embodiment, the epitope is defined as those residues on the extracellular portion of STEAP1 that exhibit at least a 10% reduction in solvent accessibility when an antigen-binding protein (reference or test) is bound to it compared to when it is not bound.

[0034] Methods for Producing Antigen Binding Proteins Suitable methods for producing antigen-binding proteins (e.g., antibodies, antigen-binding antibody fragments, and antibody protein products) are known in the art. For example, standard hybridoma methods for producing antibodies are described, for example, in Harlow and Lane (eds.), Antibodies: A Laboratory Manual, CSH Press (1988) and CA. Janeway et al. (eds.), Immunobiology, 5th Ed., Garland Publishing, New York, NY (2001). EBV-hybridoma methods and bacteriophage vector expression systems are described, for example, in Haskard and Archer, J. Immunol. Methods, 74(2), 361-67 (1984), Roder et al., Methods Enzymol., 121, 140-67 (1986), and Huse et al., Science, 246, 1275-81 (1989). Methods for producing antibodies in non-human animals are described, for example, in U.S. Patent Nos. 5,545,806, 5,569,825, 5,714,352 and 5,814,318; and U.S. Patent Application Publication No. 2002 / 0197266, all of which are incorporated herein by reference. In certain embodiments, recombinant antigen-binding proteins that bind to STEAP1 are provided. In this context, a "recombinant protein" is a protein that is made using recombinant techniques, for example, by expression of a recombinant nucleic acid. Methods and techniques for the production of recombinant proteins are well known in the art.

[0035] Molecular evolution of CDRs in the binding site has also been used to generate antigen binding proteins (e.g., antibodies) with increased affinity, such as antibodies with increased affinity for c-erbB-2 as described by Schier et al., 1996, J. Mol. Biol. 263:551. Such techniques are useful for preparing anti-STEAP1 antigen binding proteins (or other antigen binding proteins described herein).

[0036] Methods for testing antigen-binding proteins for their ability to bind to an antigen, such as STEAP1, are known in the art and include, for example, radioimmunoassays (RIA), ELISA, Western blots, immunoprecipitation, surface plasmon resonance (e.g., BIAcore®), and competitive inhibition assays (see, for example, Janeway et al., infra; U.S. Patent Application Publication No. 2002 / 0197266; ​​and U.S. Patent No. 7,872,106, all of which are incorporated herein by reference in their entirety and specifically with respect to competitive assays). Indeed, assays that test the ability of an antigen-binding protein to compete with a second antigen-binding protein for binding to an antigen or an epitope thereof are known in the art and can be used, for example, to test the ability of an antibody to bind to STEAP1. See, e.g., U.S. Patent Publication No. 2014 / 0178905; Chand et al., Biologicals 46:168-171 (2017); Liu et al., Anal Biochem 525:89-91 (2017); and Goolia et al., J Vet Diagn Invest 29(2):250-253 (2017). Surface plasmon resonance can be used to determine the binding constant of an antigen-binding protein and a second antigen-binding protein, and the two binding constants can be compared.

[0037] Multispecific antigen-binding proteins An ongoing problem in antibody technology is the need for bispecific (and / or multispecific) antibodies that bind two (or more) different antigens simultaneously, typically allowing the different antigens to be in close proximity and resulting in novel functionalities and novel therapies. The present disclosure provides novel multispecific antigen-binding proteins that bind STEAP1 and one or more additional target antigens. In a preferred embodiment, the present disclosure provides novel bispecific antigen-binding proteins (e.g., bispecific antibodies) that include an anti-STEAP1 binding domain as described above and a binding region that binds to a second target antigen (which may be a different STEAP1 epitope, but is generally a different antigen). In various aspects, the second antigen is a cell surface molecule present on effector cells, i.e., leukocytes that express one or more FcRs (e.g., FcγRIII) and perform one or more effector functions attributable to the Fc region of the antibody.

[0038] Examples of effector functions include, but are not limited to, Clq binding and complement-dependent cytotoxicity (CDC), Fc receptor binding, antibody-dependent cellular cytotoxicity (ADCC), phagocytosis, downregulation of cell surface receptors, and B cell activation. Examples of effector cells involved in ADCC include, but are not limited to, cytotoxic T cells, peripheral blood mononuclear cells (PBMCs), natural killer (NK) cells, monocytes, and neutrophils. In various embodiments, the present disclosure provides bispecific antigen-binding proteins (e.g., bispecific antibodies) that bind to both CD3 (e.g., SEQ ID NO: 1) and STEAP1 (SEQ ID NO: 2). In various embodiments, the present disclosure provides bispecific antigen-binding proteins (e.g., bispecific antibodies) that bind to both CD3 and extracellular loops 1 and 3 of STEAP1.

[0039] In various aspects, the bispecific antigen binding protein differentiates between target cells expressing high levels of STEAP1 and those off-target cells that exhibit less STEAP1. In this regard, in some embodiments, the bispecific antigen binding proteins (e.g., heterodimeric antibodies) of the present disclosure can preferentially mediate T cell-dependent killing of tumor cells and exhibit reduced "off-target" effects. For example, in some aspects, bispecific antibodies comprising a STEAP1 antigen binding protein and a CD3 antigen binding region described herein preferentially mediate T cell-dependent killing of cells having a surface density of STEAP1 of more than 10,000 (e.g., the EC90 is at least 10-fold less for cells having a surface density of STEAP1 of more than 10,000 compared to cells having a surface density of STEAP1 of less than 10,000).

[0040] The present disclosure provides bispecific antigen binding proteins comprising novel anti-CD3 sequences comprising CDRs and complete variable light and heavy chain sets. In some embodiments, the CD3 binding domain of the bispecific construct (optionally an scFv as described below) comprises a variable heavy chain domain comprising heavy chain CDRs comprising amino acid sequences that differ by no more than 3, 2 or 1 amino acid from vhCDR1 SEQ ID NO: 170, vhCDR2 SEQ ID NO: 171 and vhCDR3 SEQ ID NO: 172, and a variable light chain domain comprising light chain CDRs comprising amino acid sequences that differ by no more than 3, 2 or 1 amino acid from vlCDR1 SEQ ID NO: 174, vlCDR2 SEQ ID NO: 175 and vlCDR3 SEQ ID NO: 176. For example, the disclosure provides a multispecific (e.g., bispecific) construct comprising a variable heavy chain domain comprising an amino acid sequence at least 90% identical (e.g., at least 95% identical or 100% identical) to SEQ ID NO: 169 and a variable light chain domain comprising an amino acid sequence at least 90% identical (e.g., at least 95% identical or 100% identical) to SEQ ID NO: 173.

[0041] For example, the anti-CD3 portion optionally comprises the CDR sequences vhCDR1 comprising SEQ ID NO: 170, vhCDR2 comprising SEQ ID NO: 171, vhCDR3 comprising SEQ ID NO: 172, vlCDR1 comprising SEQ ID NO: 174, vlCDR2 comprising SEQ ID NO: 175 and vlCDR3 comprising SEQ ID NO: 176. In this regard, the CD3 binding region optionally comprises the variable heavy chain region of SEQ ID NO: 169 and the variable light chain region of SEQ ID NO:173.

[0042] Bispecific antigen-binding proteins may comprise two antigen-binding domains (e.g., each antigen is monovalently bound), such as the STEAP1 and CD3 binding domains described herein, or three (or more) antigen-binding domains (e.g., one antigen is bivalently bound and the other is monovalently bound). Bispecific antibodies include, but are not limited to, traditional bispecific immunoglobulins (e.g., BsIgG), IgGs with additional antigen-binding domains (e.g., in which the amino or carboxy terminus of the light or heavy chain is linked to an additional antigen-binding domain, such as a single domain antibody or paired antibody variable domains (e.g., Fv or scFv)), BsAb fragments (e.g., bispecific single chain antibodies), bispecific fusion proteins (e.g., an antigen-binding domain fused to an effector moiety), and BsAb conjugates. See, for example, Spiess et al., Molecular Immunology 67(2) Part A:97-106 (2015), which describes various bispecific formats and is incorporated herein by reference. Examples of bispecific constructs include, but are not limited to, diabodies, single chain diabodies, tandem scFv and Fab2 bispecifics, and engineered constructs including full-length antibodies.See, e.g., Chames & Baty, 2009, mAbs 1[6]:1-9; and Holliger & Hudson, 2005, Nature Biotechnology 23[9]:1126-1136; Wu et al., 2007, Nature Biotechnology 25

[11] :1290-1297; Michaelson et al., 2009, mAbs 1[2]:128-141; WO 2009032782 and WO 2006020258; Zuo et al., 2000, Protein Engineering 13[5]:361-367; U.S. Patent Publication No. 20020103345; Shen et al., 2006, J Biol Chem 2007, 1126-1136, all of which are expressly incorporated herein. 281

[16] :10706-10714; Lu et al., 2005, J Biol Chem 280

[20] :19665-19672; and Kontermann, 2012 MAbs 4(2):182.

[0043] In various embodiments, the bispecific antigen-binding protein is a bispecific single-chain antibody (BiScFv). The light chain variable region and the heavy chain variable region are linked to each other as a single chain as a first antigen-binding domain, optionally linked via a linker to a second antigen-binding domain of similar structure. If a linker is used, the linker is preferably of sufficient length and sequence to ensure that each of the first and second antigen-binding domains can retain their distinct binding specificities independently of each other. Bispecific single chain molecules are known in the art and are further described in U.S. Pat. No. 7,635,472, WO 99 / 54440; Mack, J. Immunol. (1997), 158, 3965-3970; Mack, PNAS, (1995), 92, 7021-7025; Kufer, Cancer Immunol. Immunother., (1997), 45, 193-197; Loffler, Blood, (2000), 95, 6, 2098-2103; Bruhl, Immunol., (2001), 166, 2420-2426; and Kipriyanov, J. Mol. Biol., (1999), 293, 41-56, all of which are incorporated by reference in their entirety.

[0044] Alternative bispecific antigen-binding formats are described, for example, in US Patent Publication No. 2011 / 0054151, which is incorporated herein by reference. For example, a bispecific antigen-binding protein may include a mAb-Fv format in which an IgG antibody is fused to an Fv fragment at the C-terminus. Alternatively, a mAb-Fab format may be used in which an IgG antibody is fused to a Fab at the C-terminus. The mAb-Fab construct contains CH and CL constant domains C-terminus, whereas the mAb-Fv does not contain them in the C-terminal Fv fusion. See Figure 8 of US Patent Publication No. 2011 / 0054151. Optionally, the N-terminal binding regions of the mAb-Fv and mAb-Fab constructs lack the light chain and CH1 domains (i.e., contain a single domain VHH region). mAb-Fv and mAb-Fab constructs contain three variable regions such that they bind a first antigen bivalently and a second antigen monovalently. Suitable bispecific antigen binding formats also include the Fab-Fv and Fab-Fab constructs described in US 2011 / 0054151. Fab-Fv and Fab-Fab immunoglobulins contain an N-terminal Fab fragment that binds a first antigen and a C-terminal Fv or Fab fragment that binds a second antigen.

[0045] In one aspect, the present disclosure relates to the generation of heterodimeric antibodies that bind antigens together and rely on different amino acid variants in the constant regions on each chain to promote heterodimerization formation and / or allow easier purification of the heterodimer over homodimers. Generally, bispecific antibodies are made by including genes for each heavy and light chain in a host cell. This generally results in the formation of the desired heterodimer (AB) and two homodimers (AA and BB). However, the main obstacle in the formation of bispecific antibodies is the difficulty of purifying the heterodimeric antibody from the homodimeric antibody and / or biasing the formation of the heterodimer over the formation of the homodimer.

[0046] The present disclosure provides heterodimeric antibodies that overcome the problems associated with previous techniques. Moreover, in the context of STEAP1 / CD3 bispecific antigen-binding proteins, the heterodimeric antibodies of the present disclosure allow monovalent binding of CD3. CD3 activation of T cells occurs only when their associated T cell receptor (TCR) binds antigen-loaded MHC on antigen-presenting cells at high affinity cell-cell junctions (Kuhns et al., 2006, Immunity 24:133-139). Non-specific bivalent cross-linking of CD3 using anti-CD3 antibodies induces cytokine storm and toxicity (Perruche et al., 2009, J Immunol 183[2]:953-61; Chatenoud & Bluestone, 2007, Nature Reviews Immunology 7:622-632; expressly incorporated by reference). Thus, for practical clinical use, the preferred mode of CD3 co-engagement for redirected killing of target cells is monovalent binding that results in activation only upon binding to the co-engaged target. Thus, in one embodiment, the heterodimeric antibodies of the present disclosure offer the advantages of monovalent binding to CD3 and bivalent binding to STEAP1 in a format that results in efficient antibody production.

[0047] An exemplary heterodimeric antibody format includes one heavy chain with a single chain Fv (scFv) and a second heavy chain in a "conventional" Fab format, i.e., variable heavy and light chains. In other words, a heterodimeric antibody includes a) a first heavy chain including a single chain Fv region (scFv) that binds a first antigen (optionally CD3); b) a second heavy chain including a second variable Fc domain and a first variable heavy domain; and c) a first light chain including a first variable light domain and a first constant light domain, where the first variable heavy domain and the first variable light domain bind a second antigen (optionally STEAP1). To illustrate, the construct comprises one monomer with scFv region-domain linker-Fc domain and a second monomer with a light chain associated with VH-CH1-hinge-CH2-CH3, optionally with heterodimerization variants including steric and pI variants, Fc and FcRn variants and additional antigen binding domains contained in these regions (with optional linkers). In some embodiments, the linker is a hinge region or a fragment thereof. This structure may be referred to herein as the XmAb format "triple F" format (scFv-FAb-Fc) or "bottle opener" format. The two chains are preferably linked by the use of amino acid variants in the constant region (e.g., Fc domain and / or hinge region) that promote the formation of heterodimeric antibodies as described more fully below. Preferably, the scFv binds to CD3 and optionally includes a positively charged scFv linker. Alternatively, the scFv binds to STEAP1. The "triple-F" format is further described in US Pat. No. 9,822,186, which is incorporated by reference herein in general and specifically for its disclosure of heterodimeric antibody structures.

[0048] In another embodiment, the bispecific antigen binding protein is a heterodimeric antibody comprising a first heavy chain comprising a first variable heavy chain domain, a first constant heavy chain comprising a first CH1 domain, and a first monomer comprising a first Fc domain, together with an scFv comprising an scFv variable light chain domain, an scFv linker, and an scFv variable heavy chain domain. The scFv is covalently linked using a domain linker between the C-terminus of the CH1 domain of the heavy chain constant domain and the N-terminus of the first Fc domain, and the scFv binds to CD3. The heterodimeric antibody further comprises a second monomer comprising a second heavy chain comprising a second variable heavy chain domain and a second constant heavy chain comprising a second Fc domain. The heterodimeric antibody further utilizes a common light chain comprising a variable light chain domain and a constant light chain domain associated with the heavy chains forming two identical Fabs that bind to STEAP1. This format is referred to as "XmAb" for bivalent binding to one target antigen. 2+1 " format. Thus, in one embodiment, the heterodimeric antibodies of the present disclosure provide the advantages of monovalent binding to CD3 and bivalent binding to STEAP1 in a format that results in efficient antibody production.

[0049] As described further below, heterodimeric antibodies may also include variants that generate asymmetric variants, pI variants, truncation variants, additional Fc variants, etc. For example, in various embodiments, the first and said second Fc domains have a set of amino acid substitutions selected from the group consisting of S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L and K370S:S364K / E357Q.

[0050] XmAb of the present disclosure 2+1An illustration of the heterodimeric antibody format is provided in Figure 1. Provision of an scFv domain and two Fab portions results in three antigen-binding domains, with the Fab portions of two monomers binding to STEAP1 and the scFv domain binding to CD3. The scFv domain is inserted between the Fc domain and the CH1-Fv region of one of the monomers.

[0051] Preferably, the heterodimeric antibody is of the IgG class with multiple subclasses, including but not limited to IgG1, IgG2, IgG3 and IgG4, but also IgM, IgD, IgG, IgA and IgE are contemplated. It should be understood that the antibody may also include hybrids of isotypes and / or subclasses. For example, pI engineering of IgG1 / G2 hybrids as shown in US Patent Application Publication No. 2009 / 0163699, which is incorporated by reference, is contemplated as part of this disclosure.

[0052] There are several mechanisms that can be used to generate the heterodimers of the present disclosure. In addition, as will be appreciated by those of skill in the art and described more fully below, these mechanisms can be combined to ensure high heterodimerization.

[0053] One mechanism is generally referred to in the art as "knobs and holes" ("KIH"), which refers to amino acid manipulations that result in steric effects that favor heterodimer formation, although unfavorable heterodimer forms may also be used optionally. This may be referred to as "knobs and holes" as described in US Patent Publication No. 20130205756, Ridgway et al., Protein Engineering 9(7):617 (1996); Atwell et al., J. Mol. Biol. 1997 270:26; and US Patent No. 8,216,805, all of which are incorporated herein by reference in their entirety, particularly with respect to disclosure of heterodimeric antibody production. In addition, these "knobs and holes" mutations can be combined with disulfide bonds to make heterodimerization asymmetric, as described in Merchant et al., Nature Biotech. 16:677 (1998). Examples of mutations include T366S / L368A / Y407V paired with T366W and this mutant with the bridging disulfide, T366S / L368A / Y407V / Y349C paired with T366W / S354C, particularly in combination with other heterodimerization mutants including pi mutants as outlined below.

[0054] A further mechanism useful in the generation of heterodimers may be referred to as "electrostatic steering", as described in Gunasekaran et al., J. Biol. Chem. 285(25):19637 (2010), which is incorporated herein by reference in its entirety. This may be referred to herein as "charge pairing". In this embodiment, electrostatics are used to asymmetrically drive the formation towards heterodimerization. As one skilled in the art will appreciate, these may have an impact on pI and therefore purification, and therefore may also be considered to be pI mutants in some cases. However, because they were generated to force heterodimerization and were not used as a purification tool, they are classified as "steric mutants". These include, but are not limited to, D221E / P228E / L368E paired with D221R / P228R / K409R (i.e., these are the corresponding sets of monomers) and C220E / P228E / 368E paired with C220R / E224R / P228R / K409R. In some embodiments of the framework regions, the mutation at position 220 removes a cysteine ​​that is no longer required for heavy and light chain disulfide formation. "Stereovariants" are optional embodiments of the present disclosure.

[0055] There are several mechanisms that can facilitate the purification of heterodimeric proteins. Some rely on the use of pI mutants, so that each monomer has a different pI, thereby allowing isoelectric purification of AA, AB and BB dimeric proteins. Alternatively, separation can be performed based on size. It is also possible to "make asymmetric" so that heterodimer formation predominates over homodimers, as generally outlined below. Thus, a combination of stereochemical heterodimerization mutants and pI or charge pair mutants can be used in the context of the present disclosure. Additionally, the scFv can include a charged scFv linker (either positive or negative) that further increases the pI for purification purposes. As will be appreciated by those skilled in the art, some triple F formats with the correct charged scFv linker and no additional pI adjustments are useful, but the present invention also provides for the use of asymmetric mutants (and combinations of Fc, FcRn and KO mutants described herein) that also have a charged scFv linker.

[0056] In embodiments utilizing pI as a separation mechanism, amino acid variants can be introduced into one or both of the monomeric polypeptides; i.e., the pI of one of the monomers (referred to herein for simplicity as "monomer A") can be engineered away from monomer B, or both monomers A and B can be altered such that the pI of monomer A is increased and the pI of monomer B is decreased. The pI change of one or both monomers can be made by removing or adding a charged residue (e.g., replacing a neutral amino acid with a positively or negatively charged amino acid residue, e.g., replacing glycine with glutamic acid), changing a charged residue from a positive or negative charge to the opposite charge (e.g., aspartic acid to lysine), or changing a charged residue to a neutral residue (e.g., loss of charge; lysine to serine). Additionally, suitable pI variants for use in making the heterodimeric antibodies herein are those that are isotype importing pI variants from different IgG isotypes such that the pI can be altered without introducing significant immunogenicity; see Figure 29 of US Patent Publication No. 20140288275, which is incorporated by reference in its entirety.

[0057] Thus, embodiments make sufficient changes in pI in at least one of the monomers so that the heterodimer can separate from the homodimer. This can be accomplished by using a "wild type" heavy chain constant region and a mutant region that has been engineered to raise or lower its pI (wtA-+B or wtA--B), or by raising one region and lowering the other (A+-B- or A-B+). Note that for the purposes of this discussion, it does not matter which monomer contains the scFv and which contains the Fab. A schematic diagram relating to the use of pI mutants is described in Figure 34 of U.S. Pat. No. 9,822,186 (incorporated herein by reference in its entirety and in particular for the discussion of heterodimeric antibody mutants and anti-CD3 sequences). pI mutants can be combined with asymmetric mutants in a "plug and play" format in that the effect of the mutants is easily transferred to different antibodies with different Fv regions and is very stable.

[0058] Thus, in general, embodiments of the disclosure include amino acid variants in the constant region of an antibody that are directed to alter the isoelectric point (pI) of at least one, if not both, of the antibody monomers to form a "pI heterodimer" (i.e., a "pI antibody") by incorporating amino acid substitutions ("pI variants" or "pI substitutions") into one or both of the monomers. Separation of the heterodimer from the two homodimers can be achieved if the pI of the two monomers differs by as little as 0.1 pH units, such as 0.2, 0.3, 0.4 and even 0.5 pH differences or more.

[0059] The number of pI variants to be included in each or both monomers to achieve the desired separation will depend in part on the starting pI of the scFv and Fab. That is, to determine which monomer to engineer or which "direction" (e.g., positive or negative) the Fv sequences of the two target antigens are calculated and a decision made from there. As is known in the art, different Fvs will have different starting pIs utilized. Generally, the pIs are engineered to result in a total pI difference of at least about 0.1 log for each monomer, with 0.2-0.5 being preferred.

[0060] Furthermore, in some cases (depending on the format), heterodimers can be separated from homodimers based on size (e.g., molecular weight). For example, as shown in some embodiments in Figures 18A-I, some formats result in homodimers and heterodimers with different sizes (e.g., for a bottle opener, one homodimer is a "dual scFv" format, one homodimer is a standard antibody, and the heterodimer has one Fab and one scFv). In addition, as shown in Figures 18A-I, some antigens may be bound bivalently (e.g., two antigen binding sites for one antigen). As will be appreciated, any combination of Fabs and scFvs can be utilized to achieve the desired results and combinations.

[0061] The use of constant regions of the heavy chains provides a further modular approach to the design and purification of multispecific proteins, including antibodies, when heterodimerization is achieved using pI variants. Thus, in some embodiments, heterodimerization variants (including asymmetric and purified heterodimerization variants) are not included in the variable region, so that each individual antibody must be engineered. In addition, in some embodiments, the potential for immunogenicity resulting from pI variants is significantly reduced by importing pI variants from different IgG isotypes such that the pI is changed without introducing significant immunogenicity. Thus, an additional problem to be solved is the elucidation of low pI constant domains with high human sequence content, e.g., minimizing or avoiding non-human residues at any particular position.

[0062] Potential collateral benefits of pI engineering are also increased serum half-life and increased FcRn binding. That is, lowering the pI of antibody constant domains (including those found in antibodies and Fc fusions) can result in prolonged serum retention in vivo, as described in US Patent Publication No. 20120028304 (incorporated by reference in its entirety). These pI variants for prolonged serum half-life also facilitate pI changes for purification.

[0063] Heterodimeric fusion proteins of the present disclosure can be in a variety of configurations, as generally shown in Figures 18A-I. Some figures show "single-ended" configurations, with one specificity on one "arm" of the molecule and a different specificity on the other "arm." Other figures show "dual-ended" configurations, with at least one specificity on the "top" of the molecule and one or more different specificities on the "bottom" of the molecule. One heterodimeric scaffold useful in the present disclosure is the "triple-F" or "bottle opener" scaffold format shown in Figure 18A and as described above. The "triple-F" format has several distinct advantages. Antibody analogs that rely on two scFv constructs often have stability and aggregation problems that can be alleviated by the constructs described herein with the addition of "conventional" heavy and light chain pairing. Additionally, in contrast to formats that rely on two heavy and two light chains, there are no issues with improper pairing of heavy and light chains (e.g., heavy chain 1 pairing with light chain 2, etc.). Additional useful antigen binding protein formats are described below.

[0064] In various embodiments, the scFv of the heterodimeric antibody comprises the anti-CD3 CDR sequences described herein. For example, in various embodiments, the scFv comprises a vhCDR1 comprising SEQ ID NO: 170, a vhCDR2 comprising SEQ ID NO: 171, a vhCDR3 comprising SEQ ID NO: 172, a vlCDR1 comprising SEQ ID NO: 174, a vlCDR2 comprising SEQ ID NO: 175, and a vlCDR3 comprising SEQ ID NO: 176. For example, the scFv optionally comprises a variable heavy chain region of SEQ ID NO: 169 and a variable light chain region of SEQ ID NO: 173. In various embodiments, the scFv comprises the sequence of SEQ ID NO: 44.

[0065] In some embodiments, the scFv comprises an amino acid sequence set forth in FIG. 19, e.g., the amino acid sequence set forth in SEQ ID NO: 44. The sequences set forth in FIG. 19 provide antigen binding domains of different affinities. In some indications, stronger affinity may be preferred, while in others, lower affinity may be useful. Thus, in some embodiments, the present disclosure provides heterodimeric antibodies comprising anti-CD3 antigen binding domains that are "strong" or "high affinity" binders to CD3 (e.g., one example is the heavy and light chain variable domains shown as H1.30_L1.47 (optionally including an appropriate charged linker)). In other embodiments, the present disclosure provides heterodimeric antibodies comprising anti-CD3 antigen binding domains that are "less" or "lower affinity" binders to CD3.

[0066] Exemplary scFv linkers are well known in the art, generally are 10-25 amino acids in length, and contain glycine and serine. "Charged scFv linker" refers to an scFv linker that utilizes charged amino acids for use in the generation and purification of heterodimeric antibodies comprising at least one scFv. Suitable charged scFv linkers are shown in Figures 8A, 8B, and 19, although others may be used. In general, charged scFv linkers contemplated for use in connection with the present disclosure include the conventionally used (GGGGS) 3~5(SEQ ID NO: 179) sequence. The charged scFv optionally comprises an amino acid sequence selected from: IRPRAIGGSKPRVA (SEQ ID NO: 145), GKGGSGKGGSGKGGS (SEQ ID NO: 146), GGKGSGGKGSGGKGS (SEQ ID NO: 147), GGGKSGGGKSGGGKS (SEQ ID NO: 148), GKGKSGKGKSGKGKS (SEQ ID NO: 149), GGGKSGGKGSGKGGS (SEQ ID NO: 150), GKPGSGKPGSGKPGS (SEQ ID NO: 151), GKPGSGKPGSGKPGSGKPGS (SEQ ID NO: 152), or GKGKSGKGKSGKGKSGKGS (SEQ ID NO: 153). In various embodiments, the scFv comprises the amino acid sequence GKPGSGKPGSGKPGSGKPGS (SEQ ID NO: 152).

[0067] In an exemplary embodiment, the scFv comprises any one of the CDR sequences provided herein (e.g., any one or more of SEQ ID NOs: 4 to 6, 8 to 10, 11 to 17, 21, 23 to 25, 27 to 29, 30 to 35, 170 to 173, and 174 to 176, any one or more of SEQ ID NOs: 3, 7, 22, 26, 41, 42, 45, 46, 49, 50, 53, 54, 57, 58, 61, 62, 65, 66, 69, 70, 73, 74, 77, 78, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 120, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153 any one or more of variable region sequences of SEQ ID NOs: 1, 82, 85, 86, 89, 90, 93, 94, 97, 98, 101, 102, 105, 106, 109, 110, 113, 114, 117, 118, 121, 122, 125, 126, 129, 130, 133, 134, 137, 138, 141, 142, 169, 173, and 182 to 186; any one of scFv linker sequences of SEQ ID NOs: 143 to 168 and / or SEQ ID NOs: 19, 20, 38, 40, 43, 44, 47, 48, 51, 52, 55, 56, 59, 60, 63, 64, 67, 68, 71, 72, 75, 76, 79, 80, 83, 84, 87, 88, 91, 92, 95, 96, 99, 100, 104, 104, 107, 108, 111, 112, 115, 116, 119, 120, 123, 124, 127, 128, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, The scFv may comprise a CDR sequence, a variable region sequence, an scFv linker sequence, or an scFv sequence having at least about 70%, at least about 80%, at least about 85%, at least about 90%, or more than about 90% (e.g., about 91%, about 92%, about 93%, about 94%, about 95%, about 96%, about 97%, about 98%, or about 99%) sequence identity with any one of the scFv sequences set forth in SEQ ID NOs: 4 to 6, 8 to 10, 11 to 17, 21, 23 to 25, 27 to 29, 30 to 35, 170 to 173, and 174 to 176, but including one or two amino acid substitutions.Alternatively, in various embodiments, the scFv can comprise variable region sequences that have been modified with respect to SEQ ID NO: 3, 7, 22, 26, 41, 42, 45, 46, 49, 50, 53, 54, 57, 58, 61, 62, 65, 66, 69, 70, 73, 74, 77, 78, 81, 82, 85, 86, 89, 90, 93, 94, 97, 98, 101, 102, 105, 106, 109, 110, 113, 114, 117, 118, 121, 122, 125, 126, 129, 130, 133, 134, 137, 138, 141, 142, 169, 173 or 182-186, wherein the modifications are outside of the CDR sequences.

[0068] The first and second variable heavy chain domains of the heterodimeric antibody in various aspects comprise the anti-STEAP1 CDR or variable region sequences described herein. For example, in some embodiments, the first and second variable heavy chain domains of the heterodimeric antibody comprise a vhCDR1 comprising SEQ ID NO: 14, a vhCDR2 comprising SEQ ID NO: 15 or SEQ ID NO: 21, and a vhCDR3 comprising SEQ ID NO: 16; and the variable light chain domain comprises a vlCDR1 comprising SEQ ID NO: 11, a vlCDR2 comprising SEQ ID NO: 12, and a vlCDR3 comprising SEQ ID NO: 13. Alternatively, the first and second variable heavy chain domains comprise a vhCDR1 comprising SEQ ID NO: 33, a vhCDR2 comprising SEQ ID NO: 34, and a vhCDR3 comprising SEQ ID NO: 35; and the variable light chain domain comprises a vlCDR1 comprising SEQ ID NO: 30, a vlCDR2 comprising SEQ ID NO: 31, and a vlCDR3 comprising SEQ ID NO: 32. In a preferred embodiment, the first variable heavy chain domain and the second variable heavy chain domain comprise SEQ ID NO: 182 or SEQ ID NO: 184, and the variable light chain domain comprises SEQ ID NO: 183. Alternatively, the first variable heavy chain domain and the second variable heavy chain domain comprise SEQ ID NO: 185, and the variable light chain domain comprises SEQ ID NO: 186.

[0069] In various embodiments of the disclosure, the heterodimeric antibody comprises: a) a first monomer comprising the sequence of SEQ ID NO: 19 or 20, a second monomer comprising the sequence of SEQ ID NO: 18, and a common light chain comprising the sequence of SEQ ID NO: 17; or b) a first monomer comprising the sequence of SEQ ID NO: 38, a second monomer comprising the sequence of SEQ ID NO: 37, and a common light chain comprising the sequence of SEQ ID NO: 36.

[0070] In various embodiments of the disclosure, the heterodimeric antibody comprises a first monomer comprising the sequence of SEQ ID NO:202 or 207, a second monomer comprising the sequence of SEQ ID NO:201 or 203, and a common light chain comprising the sequence of SEQ ID NO:200 (e.g., a first monomer comprising the sequence of SEQ ID NO:202, a second monomer comprising the sequence of SEQ ID NO:201, and a light chain comprising the sequence of SEQ ID NO:200; or a first monomer comprising the sequence of SEQ ID NO:207, a second monomer comprising the sequence of SEQ ID NO:203, and a light chain comprising the sequence of SEQ ID NO:200). Alternatively, the heterodimeric antibody may comprise a first monomer comprising the sequence of SEQ ID NO:206, a second monomer comprising the sequence of SEQ ID NO:205, and a common light chain comprising the sequence of SEQ ID NO:204.

[0071] In an exemplary embodiment, the first and / or second variable heavy chain domains are selected from the group consisting of the sequences provided herein (e.g., any one or more of the CDR sequences of SEQ ID NOs: 4-6, 14-17, 21, 23-25, 33-35, and 170-172, or any one or more of the CDR sequences of SEQ ID NOs: 3, 22, 41, 45, 49, 53, 57, 61, 65, 69, 73, 77, 81, 85, 89, 93, 97, 101, 105, 109, 113, 117, 121, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 18 , 129, 133, 137, 141, 169, 182, 184, and 185). For example, the first and / or second variable heavy chain domain may comprise a CDR sequence or variable region sequence set forth in any one or more of SEQ ID NOs: 4-6, 14-17, 21, 23-25, 33-35, and 170-172, but comprising one or two amino acid substitutions. Alternatively, in various embodiments, the first and / or second variable heavy chain domain may comprise a variable region sequence modified with respect to SEQ ID NO: 3, 22, 41, 45, 49, 53, 57, 61, 65, 69, 73, 77, 81, 85, 89, 93, 97, 101, 105, 109, 113, 117, 121, 125, 129, 133, 137, 141, 169, 182, 184 or 185, wherein the modification is outside of the CDR sequences. Similarly, the variable light chain domain may be any one of the CDR sequences of any one or more of the sequences provided herein (e.g., SEQ ID NOs: 8 to 10, 11 to 13, 27 to 29, 30 to 32, and 174 to 176, or any one or more of the CDR sequences of SEQ ID NOs: 7, 26, 42, 46, 50, 54, 58, 62, 66, 70, 74, 78, 82, 86, 90, 94, 98, 102, 106, 110, 114, 118, 122, 126, 130, 132, 134, 136, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 1 The present invention may comprise a CDR or variable region sequence having at least about 70%, at least about 80%, at least about 85%, at least about 90% or more than about 90% (e.g., about 91%, about 92%, about 93%, about 94%, about 95%, about 96%, about 97%, about 98% or about 99%) sequence identity to any one of the variable region sequences of any one of SEQ ID NO:4, 138, 142, 173, 183 and 186.For example, the variable light chain domain may comprise a CDR sequence as set forth in any one or more of SEQ ID NOs: 8-10, 11-13, 27-29, 30-32, and 174-176, but containing one or two amino acid substitutions. Alternatively, in various embodiments, the variable light chain domain may comprise a variable region sequence that is modified with respect to SEQ ID NOs: 7, 26, 42, 46, 50, 54, 58, 62, 66, 70, 74, 78, 82, 86, 90, 94, 98, 102, 106, 110, 114, 118, 122, 126, 130, 134, 138, 142, 173, 183, or 186, wherein the modification is external to the CDR sequences. Optionally, the first monomer can comprise an amino acid sequence having at least about 70%, at least about 80%, at least about 85%, at least about 90% or more than about 90% (e.g., about 91%, about 92%, about 93%, about 94%, about 95%, about 96%, about 97%, about 98% or about 99%) sequence identity to any of the sequences provided herein (SEQ ID NO: 18, 199 or 37; or SEQ ID NO: 202, 207 or 206); the second monomer can comprise an amino acid sequence having at least about 70%, at least about 80%, at least about 85%, at least about 90% or more than about 90% sequence identity to any of the sequences provided herein (SEQ ID NO: 19, 20, 38, 202, 206 or 207); and / or the common light chain may comprise an amino acid sequence that has at least about 70%, at least about 80%, at least about 85%, at least about 90% or more (e.g., about 91%, about 92%, about 93%, about 94%, about 95%, about 96%, about 97%, about 98% or about 99%) sequence identity to any of the sequences provided herein (SEQ ID NOs: 17, 36, 200 or 204).

[0072] In some embodiments, full-length heterodimeric antibodies are utilized. "Full-length" refers to the structure that constitutes the natural biological form of an antibody, including variable and constant regions that include one or more modifications as outlined herein. The heterodimeric antibodies of the present disclosure can be monoclonal, synthetic, chimeric, and / or humanized. In the context of heterodimeric antibodies, antigen-binding antibody fragments contain at least one constant domain that can be engineered to generate heterodimers, such as pI engineering. Other antibody fragments include those that contain one or more of the CH1, CH2, CH3, hinge, and CL domains of the present invention that have been pI engineered. For example, an Fc fusion is a fusion of the Fc region (CH2 and CH3, optionally with a hinge region) fused to another protein. Several Fc fusions are known in the art and can be improved by the addition of the heterodimerization variants of the present invention. Antibody fusions can be made comprising CH1; CH1, CH2 and CH3; CH2; CH3; CH2 and CH3; CH1 and CH3, any or all of which may be made using any combination of the heterodimerization variants described herein, optionally with a hinge region.

[0073] Antigen binding proteins, including the heterodimeric antibodies of the present disclosure, are generally isolated or recombinant. Nucleic acids, vectors, and host cells encoding all or part of the heterodimeric antibodies described herein are described herein and are contemplated as part of the present disclosure.

[0074] Antibody structure / Fc region modification The present disclosure includes antibodies with modified Fc variants with amino acid modifications relative to the wild-type antibody sequence. The variants are defined according to the amino acid modifications that compose them. Thus, for example, N434S or 434S is an Fc variant with a substitution serine at position 434 relative to the parent Fc polypeptide, where the numbering is according to the EU index. Similarly, M428L / N434S defines an Fc variant with substitutions M428L and N434S relative to the parent Fc polypeptide. The identity of the wild-type amino acids may not be specified, in which case the aforementioned variant is referred to as 428L / 434S. The order in which the substitutions are made is arbitrary, i.e., for example, 428L / 434S is the same Fc variant as M428L / N434S, etc. The modifications may be additions, deletions or substitutions. The substitutions may include naturally occurring amino acids and optionally synthetic amino acids. See, for example, U.S. Pat. No. 6,586,207, all of which are incorporated by reference in their entirety; U.S. Patent Application Publication No. 20040214988; WO 98 / 48032, WO 03 / 073238, WO 05 / 35727 A2, and WO 05 / 74524 A2; JW Hin et al., (2002), Journal of the American Chemical Society 124:9026-9027; JW Hin, & P.G. Schultz, (2002), ChemBioChem 11:1135-1137; JW Hin, et al., (2002), PICAS United States of America 99:11020-11024; and L. Wang, & P. ​​G. Schultz, (2002), Chem. 1-10.

[0075] For all positions discussed in this disclosure regarding antibodies and other antigen-binding proteins, unless otherwise stated, the numbering of amino acid positions is according to the EU index. The EU index or EU index as Kabat or EU numbering scheme refers to the numbering of EU antibodies (Edelman et al., 1969, Proc Natl Acad Sci USA 63:78-85, which is incorporated herein by reference in its entirety). For example, each variable heavy chain region (VH) and variable light chain region (VL) is understood to be composed of three hypervariable regions ("complementarity determining regions", "CDRs") and four FRs arranged from amino terminus to carboxy terminus in the following order: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4. The hypervariable regions generally correspond to approximately amino acid residues 24-34 (LCDR1; "L" refers to light chain), 50-56 (LCDR2), and 89-97 (LCDR3) in the light chain variable region and approximately amino acid residues 31-35B (HCDR1; "H" refers to heavy chain), 50-65 (HCDR2), and 95-102 (HCDR3) in the heavy chain variable region; Kabat et al., SEQUENCES OF PROTEINS OF IMMUNOLOGICAL INTEREST, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md. (1991) and / or those residues that form the hypervariable loops (e.g., residues 26-32 (LCDR1), 50-52 (LCDR2), and 91-96 (LCDR3) in the light chain variable region and 26-32 (HCDR1), 53-55 (HCDR2), and 96-101 (HCDR3) in the heavy chain variable region; Chothia and Lesk (1987) J. Mol. Biol. 196:901-917).

[0076] As understood in the art, the exact numbering and arrangement of the CDRs may vary between different numbering systems. However, it should be understood that the present disclosure of the variable heavy and / or variable light chain sequences includes the disclosure of the associated (unique) CDRs. Thus, the present disclosure of each variable heavy chain region is a disclosure of the vhCDRs (e.g., vhCDR1, vhCDR2 and vhCDR3), and the present disclosure of each variable light chain region is a disclosure of the vlCDRs (e.g., vlCDR1, vlCDR2 and vlCDR3). A useful comparison of CDR numbering is given below, see Lafranc et al, Dev. Comp. Immunol. 27(1):55-77 (2003).

[0077] [Table 1]

[0078] Throughout this specification, the Kabat numbering system is generally used when referring to residues in the variable domain (approximately residues 1-107 for light chain variable regions and residues 1-113 for heavy chain variable regions) (e.g., Kabat et al., supra (1991)).

[0079] The carboxy-terminal portion of each chain defines a constant region that is primarily responsible for effector function. Kabat et al. collected a large number of primary sequences of heavy and light chain variable regions. Based on the degree of sequence conservation, the authors classified and compiled a list of the individual primary sequences into CDRs and frameworks (see SEQUENCES OF IMMUNOLOGICAL INTEREST, 5th edition, NIH publication, No. 91-3242, E.A. Kabat et al., incorporated by reference in its entirety).

[0080] In the IgG subclass of immunoglobulins, there are several immunoglobulin domains in the heavy chain. By "immunoglobulin (Ig) domain" herein is meant a region of an immunoglobulin having a distinct tertiary structure. Of interest are heavy chain domains that include the constant heavy (CH) domain and the hinge domain. In the context of IgG antibodies, IgG isotypes each have three CH regions. Thus, in the context of IgG, the "CH" domains are as follows: "CH1" refers to positions 118-220 according to the EU index as in Kabat; "CH2" refers to positions 237-340 according to the EU index as in Kabat, and "CH3" refers to positions 341-447 according to the EU index as in Kabat. As shown herein and below, the pI variants can be in one or more of the CH and hinge regions discussed below. In various embodiments, the sequences shown herein start at position 118 of the CH1 region; the variable regions are not included except where specified.

[0081] Another type of Ig domain of the heavy chain is the hinge region. By "hinge" or "hinge region" or "antibody hinge region" or "immunoglobulin hinge region" herein is meant a flexible polypeptide comprising the amino acids between the first and second constant domains of an antibody. Structurally, the IgG CH1 domain ends at EU position 220, and the IgG CH2 domain starts at residue EU position 237. Thus, for IgG, the antibody hinge is defined herein to comprise positions 221 (D221 in IgG1) to 236 (G236 in IgG1), where the numbering is according to the EU index as in Kabat. In some embodiments, the lower hinge is included, for example in relation to the Fc region, and "lower hinge" generally refers to positions 226 to 230. As described herein, pI variants can also be made in the hinge region.

[0082] As used herein, "Fc" or "Fc region" or "Fc domain" refers to a polypeptide comprising the constant region of an antibody excluding the first constant region immunoglobulin domain and optionally a portion of the hinge. Thus, Fc refers to the last two constant region immunoglobulin domains of IgA, IgD and IgG, and the last three constant region immunoglobulin domains of IgE and IgM, and the flexible hinge N-terminal to these domains. For IgA and IgM, Fc may include the J chain. For IgG, the Fc domain includes immunoglobulin domains Cγ2 and Cγ3 (Cγ2 and Cγ3) and the lower hinge region between Cγ1 (Cγ1) and Cγ2 (Cγ2). Although the boundaries of the Fc region may vary, the human IgG heavy chain Fc region is usually defined to include residues C226 or P230 to the carboxyl terminus, with numbering according to the EU index as described in Kabat.

[0083] Amino acid variants may be introduced into the antigen binding proteins (e.g., bispecific antibodies) of the present disclosure to add additional functionality. For example, amino acid changes may be made within the Fc region (either in one monomer or both) to facilitate increased ADCC or CDC (e.g., altered binding to Fcγ receptors), to allow or increase the yield of toxicity and drug loading (e.g., for ADC), and to increase binding to FcRn and / or to extend the serum half-life of the resulting molecule. Effector functions that may be modulated by altering the amino acid sequence include, but are not limited to, ADCC, ADCP, and CDC. Any variant outlined herein may be optionally and independently combined with other variants.

[0084] "FcRn" or "neonatal Fc receptor" refers to a protein that binds to an IgG antibody Fc region and is at least partially encoded by the FcRn gene. FcRn can be from any organism, including but not limited to human, mouse, rat, rabbit, and monkey. As known in the art, a functional FcRn protein comprises two polypeptides, often referred to as heavy and light chains. The light chain is beta-2-microglobulin, and the heavy chain is encoded by the FcRn gene. Unless otherwise stated herein, FcRn or FcRn protein refers to the complex of FcRn heavy chain and beta-2-microglobulin. Various FcRn variants are used to increase binding to the FcRn receptor and in some cases to extend serum half-life. Fc variants that confer increased binding to the FcRn receptor and a corresponding increase in serum half-life include, but are not limited to, 434A, 434S, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L, 252Y, 252Y / 254T / 256E, and 259I / 308F / 428L. For clarity, since each heavy chain is different, the FcRn variant (and Fc variant) may be present in one or both monomers.

[0085] Another class of functional variants are "Fc gamma truncation variants" or "Fc knockout (FcKO or KO) variants. In these embodiments, for some therapeutic applications, it is desirable to reduce or eliminate the normal binding of the Fc domain to one or more or all of the Fc gamma receptors (e.g., FcγR1, FcγRIIa, FcγRIIb, FcγRIIIa, etc.) to avoid additional mechanisms of action. By "Fc gamma receptor", "FcγR" or "Fc gamma R" is meant any member of a family of proteins that bind to the IgG antibody Fc region and are encoded by the FcγR gene. In humans, this family is associated with These include, but are not limited to, FcγRI (CD64), which includes isoforms FcγRIa, FcγRIb, and FcγRIc; FcγRII (CD32), which includes isoforms FcγRIIa (including allotypes H131 and R131), FcγRIIb (including allotypes FcγRIIb-1 and FcγRIIb-2), and FcγRIIc; and FcγRIII (CD16), which includes isoforms FcγRIIIa (including allotypes V158 and F158) and FcγRIIIb (including allotypes FcγRIIb-NA1 and FcγRIIb-NA2) (see Jefferies, supra, which is incorporated by reference in its entirety). et al., 2002, Immunol Lett 82:57-65). FcγRs can be from any organism, including, but not limited to, human, mouse, rat, rabbit, and monkey. Mouse FcγRs include, but are not limited to, FcγRI (CD64), FcγRII (CD32), FcγRIII (CD16), and FcγRIII-2 (CD16-2). In many embodiments, it is generally desirable to truncate FcγRIIIa binding to eliminate or significantly reduce ADCC activity. Figure 36 in U.S. Pat. No. 9,822,186 shows the use of Fc knockouts (or truncation mutants) that retain wild-type stability but eliminate all FcγR binding.

[0086] Exemplary truncation mutants include those selected from the group consisting of G236R / L328R, E233P / L234V / L235A / G236del / S239K, E233P / L234V / L235A / G236del / S267K, E233P / L234V / L235A / G236del / S239K / A327G, E233P / L234V / L235A / G236del / S267K / A327G, and E233P / L234V / L235A / G236del. It should be noted that the truncation mutants referred to herein ablate FcγR binding, but generally do not ablate FcRn binding.

[0087] As is known in the art, the Fc domain of human IgG1 has the highest binding to Fγ receptors, and therefore truncation mutants may be used when the constant domain (or Fc domain) in the scaffold of the heterodimeric antibody is IgG1. Alternatively or in addition to truncation mutants in the IgG1 scaffold, a mutation at glycosylation position 297 (typically A or S) can, for example, significantly ablate binding to FcγRIIIa. Human IgG2 and IgG4 naturally have reduced binding to Fcγ receptors, and therefore those scaffolds may be used with or without truncation mutants.

[0088] Deamidation can significantly affect antibody activity and stability. In various embodiments, the heterodimeric antibody comprises one or more substitutions to remove deamidation sites. In this regard, the heterodimeric antibody optionally comprises a substitution at position N67, such as substitution N67Q.

[0089] Heterodimeric heavy chain constant region The present disclosure provides heterodimeric antibodies based on the use of a monomer containing a variant heavy chain constant region as the first domain. A "monomer" in this specification means one half of a heterodimeric protein. It should be noted that conventional antibodies are actually tetramers (two heavy chains and two light chains). For ease of reference, in the context of this disclosure, a pair containing a heavy chain and a light chain is considered a "monomer". The heavy chain region containing the scFv (and in some cases Fab) is considered a monomer. Essentially, each monomer contains enough heavy chain constant region to allow heterodimerization engineering, whether it is the entire constant region, e.g., CH1-hinge-CH2-CH3, Fc region (CH2-CH3) or only the CH3 domain.

[0090] The variant heavy chain constant region may comprise all or a portion of the heavy chain constant region, including the full-length construct, CH1-hinge-CH2-CH3, or portions thereof, including, for example, CH2-CH3 or CH3 alone. In addition, the heavy chain regions of each monomer may be the same backbone (CH1-hinge-CH2-CH3 or CH2-CH3) or different. N- and C-terminal truncations and additions are also included in the definition. For example, some pI variants include the addition of a charged amino acid to the C-terminus of the heavy chain domain.

[0091] In addition to the heterodimerization variants outlined herein (e.g., steric and pI variants), the heavy chain regions may also contain additional amino acid substitutions, including changes to alter FcγR and FcRn binding.

[0092] Heterodimerization mutants include several different types of mutants, including but not limited to steric mutants (including charge mutants) and pI mutants, which can be optionally and independently combined with any other mutants. In these embodiments, it is important to match "monomer A" with "monomer B", i.e., if the heterodimeric protein relies on both steric and pI mutants, they need to be appropriately matched to each monomer, for example, a set of functional steric mutants (one set for monomer A, one set for monomer B) is combined with a set of pI mutants (one set for monomer A, one set for monomer B) such that the mutants for each monomer are designed to achieve the desired function. For examples where steric mutants can also change charge, the appropriate set needs to be matched to the appropriate monomer.

[0093] The heterodimerization variants outlined herein (including but not limited to those shown in the figures) can be optionally and independently combined with any other variants and with any other monomers. What is important for heterodimerization is that there are "sets" of variants, one set for one monomer and one set for the other. Whether these are combined one-to-one (e.g., a list of monomer 1 occurring simultaneously) or switched (pi variants of monomer 1 with steric variants of monomer 2) is irrelevant. However, the "chain state" should be preserved when the combination is made as outlined above, so that heterodimerization is favored. For example, charge variants that raise the pi should be used with increased pi variants and / or scFv linkers with increased pi, etc. In the context of monomers of heterodimeric proteins, "chain state" means that the heterodimerization mutants are incorporated into each monomer to preserve the ability to "match" to form heterodimers, similar to a "matched" double strand of DNA. For example, if some pI mutants are engineered into monomer A (e.g., to make the pI higher), the steric mutants that are "charge pairs" that can be utilized do not interfere with the pI mutants, e.g., the charge mutants that increase the pI are placed in the same "chain" or "monomer" to preserve both functionality. Furthermore, for additional Fc mutants (for FcγR binding, FcRn binding, truncation mutants, etc.), either or both monomers can independently and optionally contain any of the listed mutants. In some cases, both monomers have additional mutants, and in some cases, only one monomer has additional mutants or they can be combined.

[0094] Stereovariants In some embodiments, heterodimer formation is promoted by the addition of steric variants, i.e., by varying the amino acids in each heavy chain, different heavy chains are more likely to associate to form heterodimeric structures than to form homodimers with the same Fc amino acid sequence. Representative suitable steric variants are shown in the figures.

[0095] One mechanism for generating stereomutants is the "knob and hole" mechanism described above. An additional mechanism useful in generating heterodimers may be called "electrostatic steering" as described in Gunasekaran et al., J.Biol.Chem.285(25):19637 (2010), which is incorporated herein by reference in its entirety. This may be referred to herein as "charge pairing". In this embodiment, electrostatics are used to asymmetrically form toward heterodimerization. These may have an effect on pI and therefore purification, and therefore may also be considered to be pI mutants in some cases. However, because they were generated to force heterodimerization and were not used as a purification tool, they are classified as "stereomutants". These include, but are not limited to, mutants that result in greater than 75% heterodimerization, such as D221E / P228E / L368E paired with D221R / P228R / K409R (e.g., these are the "corresponding set of monomers") and C220E / P228E / 368E paired with C220R / E224R / P228R / K409R.

[0096] In some embodiments, asymmetric mutants favor heterodimerization based on both "knobs and holes" and "electrostatic steering" mechanisms, and at the same time. These mutants are provided in "pairs" of "sets", i.e., one set of the pair is incorporated into the first monomer and the other set of the pair is incorporated into the second monomer. It should be noted that these sets do not necessarily behave as "knobs-in-holes" mutants with a one-to-one correspondence between residues in one monomer and residues in the other. That is, these pairs of sets may instead form an interface between the two monomers that favors heterodimer formation and suppresses homodimer formation, allowing the percentage of heterodimers that form spontaneously under biological conditions to exceed 90%, instead of the expected 50% (25% homodimer A / A: 50% heterodimer A / B: 25% homodimer B / B). Exemplary heterodimerization "asymmetric" mutants are shown in FIG. 4. Examples of such asymmetric mutants include pairs of sets of mutations including, but not limited to, S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q; and T366S / L368A / Y407V:T366W (optionally including a bridging disulfide, T366S / L368A / Y407V / Y349C:T366W / S354C).

[0097] Additional monomer A and monomer B variants that can be optionally and independently combined in any amount with other variants, such as the pI variants outlined herein or other stereovariants shown in Figure 37 of US Patent Publication No. 2012 / 0149876, the figures and legends of which are expressly incorporated herein by reference.

[0098] In some embodiments, the steric variants outlined herein may be optionally and independently incorporated into one or both monomers along with any heterodimerization variants, including pI variants (or other variants such as Fc variants, FcRn variants, truncation variants, etc.).

[0099] pI mutants for heterodimers In general, there are two classes of pI mutants: those that increase the pI of a protein (basic changes) and those that decrease the pI of a protein (acidic changes). As described herein, all combinations of these mutants can be implemented: one monomer may be wild-type or a mutant that does not exhibit a significantly different pI from wild-type, and the other may be either more basic or more acidic. Alternatively, each monomer is altered, one more basic and one more acidic. Exemplary combinations of pI mutants are shown in the figures.

[0100] In various embodiments, for example in the format of Figures 18A, E, F, G, H and I, a preferred combination of pI variants has one monomer (negative Fab side) comprising the 208D / 295E / 384D / 418E / 421D variants (N208D / Q295E / N384D / Q418E / N421D for human IgG1) and a second monomer (positive scFv side) comprising a positively charged scFv linker comprising (GKPGS)4. However, as will be appreciated by those skilled in the art, the first monomer comprises a CH1 domain comprising position 208. Thus, in constructs that do not include a CH1 domain (e.g., for antibodies that do not utilize a CH1 domain in one of the domains, in a dual-scFv format or a "one-arm" format such as those shown in Figures 18B, C or D), a preferred negative pI mutant Fc set comprises the 295E / 384D / 418E / 421D mutants (Q295E / N384D / Q418E / N421D for human IgG1).

[0101] Acidic pI change If one monomer comprising a variant heavy chain constant domain is to be made more positive (e.g., to lower the pI), one or more of the following modifications (e.g., substitutions) are suitable in the context of the present disclosure: S119E, K133E, K133Q, T164E, K205E, K205Q, N208D, K210E, K210Q, K274E, K320E, K322E, K326E, K334E, R355E, K392E, deletion of K447, addition of the peptide DEDE at the C-terminus, G137E, N203D, K274Q, R355Q, K392N, and Q419E. These changes are described for IgG1, but all isotypes, not just isotype hybrids, can be modified in this manner. When the heavy chain constant domain is derived from IgG2-4, R133E and R133Q may also be used.

[0102] Basic pI change Where one monomer comprising a variant heavy chain constant domain is to be made more negative (e.g., to increase the pI), one or more of the following exemplary substitutions are suitable in the context of the present disclosure: Q196K, P217R, P228R, N276K and H435R. These changes are described for IgG1, however all isotypes can be modified in this manner, not just isotype hybrids.

[0103] Heterodimeric antibody light chain variants pI variants can also be made in the antibody light chain. Amino acid modifications to lower the pI of the light chain include, but are not limited to, K126E, K126Q, K145E, K145Q, N152D, S156E, K169E, S202E, K207E, and the addition of peptide DEDE at the C-terminus of the light chain. Changes in this classification based on the constant lambda light chain include, but are not limited to, one or more substitutions at R108Q, Q124E, K126Q, N138D, K145T, and Q199E. In addition, it is also possible to increase the pI of the light chain, and this is contemplated in various aspects of the present disclosure.

[0104] Isotype variants In addition, various embodiments of the present disclosure involve the "import" of pI amino acids at specific positions from one IgG isotype to another, thus reducing or eliminating the possibility of undesired immunogenicity being introduced into the variant. Some of these are shown in FIG. 21 of US Patent Application Publication No. 2014 / 0370013, incorporated herein by reference. Namely, IgG1 is a common isotype for therapeutic antibodies for a variety of reasons, including high effector function. However, the heavy chain constant region of IgG1 has a higher pI than that of IgG2 (8.10 versus 7.31). By introducing IgG2 residues at specific positions into the IgG1 backbone, the pI of the resulting monomer is lowered (or raised) and further exhibits a longer serum half-life. For example, IgG1 has glycine (pI 5.97) at position 137, and IgG2 has glutamic acid (pI 3.22). Importing glutamic acid will affect the pI of the resulting protein. Several amino acid substitutions are generally required to significantly affect the pI of the mutant antibody, however, it should be noted that even changes in the IgG2 molecule can allow for an extension of serum half-life, as discussed below.

[0105] In other embodiments, non-isotypic amino acid changes are made to reduce the overall charge state of the resulting protein (e.g., by changing from a higher pI amino acid to a lower pI amino acid) or to allow for adaptations in structure, such as for stability.

[0106] In addition, by pI engineering both the heavy and light chain constant domains, significant changes can be observed in each monomer of the heterodimer. As discussed herein, having two monomers differing in pI by at least 0.5 can allow for separation by ion exchange chromatography or isoelectric focusing or other methods that are sensitive to isoelectric point.

[0107] Additionally, pI variants can be made that are isosteric, eg, charge variants that are approximately the same size as the parent amino acid, and are contemplated herein.

[0108] Calculating pI The pI of each monomer may depend on the pI of the variant heavy chain constant domain as well as the pI of the entire monomer, including the variant heavy chain constant domain and the fusion partner. Thus, in some embodiments, the change in pI is calculated based on the variant heavy chain constant domain. Alternatively, the pI of each monomer can be compared. Similarly, the pI of the "starting" variable region (e.g., either scFv or Fab) is calculated to inform which monomers will be designed in which orientation.

[0109] pI variants that confer better FcRn binding in vivo pI mutants that lower the pI of the monomer may show the added benefit of improving serum retention in vivo.

[0110] The Fc region is believed to have a longer half-life in vivo because binding to FcRn at pH 6 in endosomes occupies Fc (Ghetie and Ward, 1997 Immunol Today. 18(12):592-598, incorporated herein by reference). The endosomal compartment then recycles Fc to the cell surface. Once the compartment is open to the extracellular space, a higher pH, about 7.4, induces the release of Fc into the blood. The increased affinity of Fc for FcRn at pH 7.4 is believed to prevent the release of Fc into the blood. Thus, Fc mutations that would extend the half-life of Fc in vivo would ideally increase FcRn binding at low pH while still allowing the release of Fc at higher pH. The amino acid histidine changes its charge state in the pH range of 6.0 to 7.4. It is therefore not surprising to find His residues at key positions in the Fc / FcRn complex.

[0111] Recently, it has been suggested that antibodies with variable regions with lower isoelectric points may also have longer serum half-lives (Igawa et al., 2010 PEDS.23(5):385-392, incorporated by reference in its entirety). Constant region variants with reduced pI and extended half-lives offer a more modular approach to improving the pharmacokinetic properties of antibodies.

[0112] pI variants useful in this embodiment and their use for purification optimization are disclosed in the figures.

[0113] Mutant combinations As will be appreciated by those skilled in the art, all of the listed heterodimerization mutants can be optionally and independently combined in any way as long as they retain their "chain state" or "monomer boundary". In addition, all of these mutants can be combined in any of the heterodimerization formats. In the case of pI mutants, an exemplary embodiment is shown in the figures, but other combinations can be made following the basic rules of modifying the pI difference between the two monomers to facilitate purification.

[0114] Antigen-binding protein (e.g., antibody) format One heterodimeric scaffold useful in the context of the present disclosure is the "triple F" or "bottle opener" scaffold format described above and depicted in FIG. 18. In this embodiment, one heavy chain of the antibody contains a single chain Fv ("scFv" as defined below) and the other heavy chain is a "normal" Fab format comprising a variable heavy chain and a light chain. Many of the embodiments outlined herein generally rely on the bottle opener format, which comprises a first monomer comprising an scFv comprising a variable heavy chain domain and a variable light chain domain covalently linked using an scFv linker (often, but not all, charged), where the scFv is typically covalently linked to the N-terminus of the first Fc domain via a domain linker (which may be uncharged or charged, as outlined herein, and may be exogenous or endogenous (e.g., all or part of a natural hinge domain)). The second monomer of the bottle opener format is a heavy chain, and the composition further comprises a light chain.

[0115] In addition, the Fc domain of the bottle opener format generally comprises an asymmetric mutant (e.g., selected from the group consisting of S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W; and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally a truncation mutant, optionally a charged scFv linker, and the heavy chain comprises a pI mutant. In some embodiments, the bottle opener format comprises asymmetric mutants, pI mutants and truncation mutants. Thus, some embodiments comprise a) a first monomer comprising a charged scFv linker, asymmetric mutants S364K / E357Q, truncation mutants E233P / L234V / L235A / G236del / S267K and an Fv ("scFv monomer"); b) a second monomer comprising asymmetric mutants L368D / K370S, pI mutants N208D / Q295E / N384D / Q418E / N421D, truncation mutants E233P / L234V / L235A / G236del / S267K and a variable heavy chain domain that together with the variable light chain domain constitutes an Fv that binds to a second antigen ("Fab monomer"); and c) a bottle opener format comprising a light chain.

[0116] In some embodiments, the bottle opener format comprises asymmetric mutants, pI mutants, truncation mutants and FcRn mutants. Thus, some embodiments include a) a first monomer ("scFv monomer") comprising a charged scFv linker, asymmetric mutants S364K / E357Q, truncation mutants E233P / L234V / L235A / G236del / S267K, FcRn mutant M428L / N434S and an Fv that binds a first antigen; b) a second monomer ("Fab monomer") comprising asymmetric mutants L368D / K370S, pI mutants N208D / Q295E / N384D / Q418E / N421D, truncation mutants E233P / L234V / L235A / G236del / S267K, FcRn mutant M428L / N434S and a variable heavy chain domain which together with the variable light chain domain constitutes an Fv that binds a second antigen; and c) a bottle opener format comprising a light chain.

[0117] Another heterodimeric scaffold useful in the context of the present disclosure is the mAb-Fv format shown in Figure 18H. In this embodiment, the format relies on the use of an "additional" variable heavy chain domain C-terminally attached to one monomer and an "additional" variable light chain domain C-terminally attached to the other monomer, thereby forming a third antigen-binding domain, where the Fab portions of the two monomers bind one antigen and the "additional" scFv domain binds a different antigen.

[0118] In this embodiment, the first monomer comprises a first heavy chain comprising a first variable heavy domain and a first constant heavy domain comprising a first Fc domain, with a first variable light domain covalently linked to the C-terminus of the first Fc domain using a domain linker (vhl-CHl-[domain linker (e.g. hinge)]-CH2-CH3-[optional domain linker]-vl2). The second monomer comprises a second variable heavy domain of a second constant heavy domain comprising a second Fc domain and a third variable light domain covalently linked to the C-terminus of the second Fc domain using a domain linker (vhl-CHl-domain linker (e.g. hinge)-CH2-CH3-[optional domain linker]-vh2). The two C-terminally linked variable domains constitute an scFv. This embodiment further utilizes a common light chain comprising the variable light domain and constant light domain associated with the heavy chains forming the two identical Fabs. For many of the embodiments herein, these constructs include asymmetric mutants, pI mutants, truncation mutants, additional Fc mutants, etc. as desired and described herein.

[0119] Optionally, the Fc domain of the mAb-Fv format comprises an asymmetric mutant (e.g., selected from the group consisting of S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W; and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally a truncation mutant, optionally a charged scFv linker, and the heavy chain comprises a pI mutant. In some embodiments, the mAb-Fv format includes asymmetric mutants, pI mutants and truncation mutants. Thus, some embodiments include a) a first monomer comprising the asymmetric mutant S364K / E357Q, the truncation mutant 233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain that together with the first variable light chain domain of the light chain constitutes an Fv that binds to an antigen, and a second variable heavy chain domain; b) a first monomer comprising the asymmetric mutant L368D / K370S, the pI mutant N208D / Q295E / N384D / Q4 a second monomer comprising 18E / N421D, a truncation mutant E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain which together with the first variable light chain domain constitutes an Fv that binds to a first antigen and a second variable light chain which together with the second variable heavy chain forms an Fv that binds to a second antigen; and c) a bottle opener format comprising a light chain comprising the first variable light chain domain and a constant light chain domain.

[0120] Yet another heterodimeric scaffold useful in the present disclosure is the mAb-scFv format shown in Figure 18I. In this embodiment, the format relies on the use of C-terminal attachment of an scFv to one of the monomers, thereby forming a third antigen-binding domain, where the Fab portions of the two monomers bind one antigen and the "additional" scFv domain binds a different antigen. In this embodiment, the first monomer comprises a first heavy chain (including a variable heavy domain and a constant domain) with a scFv covalently attached to its C-terminus, which comprises an scFv variable light domain, an scFv linker and an scFv variable heavy domain, in either orientation (vhl-CHl-domain linker-CH2-CH3-[optional domain linker]-vh2-scFv linker-vl2 or vhl-CHl-domain linker-CH2-CH3-[optional domain linker]-vl2-scFv linker-vh2). This embodiment further utilizes a common light chain that includes a variable light domain and a constant light domain associated with the heavy chain that form two identical Fabs that bind to one of the target antigens. For many of the embodiments herein, these constructs include asymmetric mutants, pI mutants, truncation mutants, additional Fc mutants, etc. as desired and described herein.

[0121] In addition, the Fc domain of the mAb-scFv format optionally comprises an asymmetric mutant (e.g., selected from the group consisting of S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W; and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally a truncation mutant, optionally a charged scFv linker, and the heavy chain comprises a pI mutant. In some embodiments, the mAb-scFv format includes asymmetric mutants, pI mutants and truncation mutants. Thus, some embodiments include a) a first monomer comprising the asymmetric mutant S364K / E357Q, the truncation mutants E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain that together with the first variable light chain domain of the light chain constitute an Fv that binds to a first antigen, and a second variable heavy chain domain; b) a first monomer comprising the asymmetric mutant L368D / K370S, the pI mutants N208D / Q295E / N3 a second monomer comprising a first variable heavy chain domain which, together with the first variable light chain domain, constitutes an Fv that binds to a first antigen, and a second variable light chain which, together with the second variable heavy chain, forms an Fv that binds to a second antigen; and c) a light chain comprising the first variable light chain domain and a constant light chain domain.

[0122] Yet another heterodimeric scaffold useful in the present disclosure is the central-scFv or "mAb 2+1" format. That format relies on the use of an inserted scFv domain thus forming a third antigen-binding domain, where the Fab portions of the two monomers bind one target and the "additional" scFv domain binds another target. The scFv domain is inserted between the Fc domain and the CH1-Fv region of one of the monomers, thereby resulting in a third antigen-binding domain. In this embodiment, one monomer comprises a first heavy chain comprising a first variable heavy domain, a CH1 domain (and optional linker / hinge) and an Fc domain, together with an scFv comprising an scFv variable light domain, an scFv linker and an scFv variable heavy domain. The scFv is covalently linked using an optional domain linker between the C-terminus of the CH1 domain of the heavy chain constant domain and the N-terminus of the first Fc domain (VH1-CH1-[optional domain linker]-VH2-scFv linker-VL2-[optional domain linker including hinge]-CH2-CH3 or in the opposite orientation with respect to the scFv, VH1-CH1-[optional domain linker]-VL2-scFv linker-VH2-[optional domain linker including hinge]-CH2-CH3). In some embodiments, the first monomer is VH1-CH1-domain linker-VH2-scFv linker-VL2 domain linker-CH2-CH3. The other monomer is a standard Fab side (i.e. VH1-CH1-domain linker (e.g. hinge)-CH2-CH3). This embodiment further utilizes a common light chain that includes a variable light domain and a constant light domain associated with the heavy chain that forms two identical Fabs that bind to the target. For many of the embodiments herein, these constructs include asymmetric mutants, pI mutants, truncation mutants, additional Fc mutants, etc. as desired and described herein.

[0123] In various embodiments, the antigen binding protein comprises a first heavy chain comprising VH1-CH1-[domain linker]-VH2-scFv linker-VL2-[domain linker (optionally including a hinge)]-CH2-CH3; a second heavy chain comprising VH1-CH1-domain linker-CH2-CH3; and a common light chain comprising VL1; VH1 and VL1 bind to STEAP1, and VH2 and VL2 bind to CD3. In this format, VH2 optionally comprises the CDR sequences of SEQ ID NO: 170 (CDR1), SEQ ID NO: 171 (CDR2) and SEQ ID NO: 172 (CDR3), while VL2 comprises the CDR sequences of SEQ ID NO: 174 (CDR1), SEQ ID NO: 175 (CDR2) and SEQ ID NO: 176 (CDR3). VH1 comprises the CDR sequences of SEQ ID NO:14 (CDR1), SEQ ID NO:15 or 21 (CDR2) and SEQ ID NO:16 (CDR3); and VL1 comprises the CDR sequences of SEQ ID NO:11 (CDR1), SEQ ID NO:12 (CDR2) and SEQ ID NO:13 (CDR3). Alternatively, VH1 comprises the CDR sequences of SEQ ID NO:33 (CDR1), SEQ ID NO:34 (CDR2) and SEQ ID NO:35 (CDR3); and VL1 comprises the CDR sequences of SEQ ID NO:30 (CDR1), SEQ ID NO:31 (CDR2) and SEQ ID NO:32 (CDR3). Optionally, the antigen binding protein comprises modifications in a first heavy chain including but not limited to E233P, delL234, L235V, G236A, S267K, r292c, n297g, v302c, E357Q and S364K (EU numbering, lower case letters refer to SEFL2 substitutions further described herein), and the second heavy chain comprises modifications including but not limited to N208D, E233P, delL234, L235V, G236A, S267K, r292c, Q295E, n297g, v302c, L368D, K370S, N384D, Q418E and N421D (EU numbering, lower case letters refer to SEFL2 substitutions further described herein). In connection with this embodiment, a linker for use is optionally GKPGSGKPGSGKPGSGKPGS (SEQ ID NO: 152).

[0124] The Fc domain of the central scFv format optionally comprises an asymmetric mutant (e.g., selected from the group consisting of S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally a truncation mutant, optionally a charged scFv linker, and the heavy chain comprises a pI mutant. In some embodiments, the central scFv format comprises asymmetric mutants, pI mutants and truncation mutants. Thus, some embodiments comprise a) a first monomer comprising the asymmetric mutant S364K / E357Q, the truncation mutants E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain constituting an Fv that binds to a first target together with a first variable light chain domain of the light chain and a second variable heavy chain domain; b) a first monomer comprising the asymmetric mutant L368D / K370S, the pI mutants N208D / Q295E / N3 a second monomer comprising a first variable heavy chain domain which, together with the first variable light chain domain, constitutes an Fv that binds to a first target, and a second variable light chain which, together with the second variable heavy chain, forms an Fv that binds to a second target; and c) a light chain comprising the first variable light chain domain and a constant light chain domain.

[0125] Another heterodimeric scaffold particularly useful in the present disclosure is the mid-Fv format shown in Figure 18G. The format relies on the use of an inserted scFv domain thus forming a third antigen-binding domain, where the Fab portions of the two monomers bind one target and the "additional" scFv domain binds another target. The scFv domain is inserted between the Fc domain and the CH1-Fv region of the monomer, thereby resulting in a third antigen-binding domain, where each monomer contains the components of an scFv (e.g., one monomer contains a variable heavy chain domain and the other contains a variable light chain domain). In this embodiment, one monomer contains a first heavy chain containing a first variable heavy chain domain, a CH1 domain and an Fc domain, and an additional variable light chain domain. The light chain domain is covalently linked using a domain linker between the C-terminus of the CH1 domain of the heavy chain constant domain and the N-terminus of the first Fc domain (vhl-CHI-[optional domain linker]-vl2-hinge-CH2-CH3). The other monomer comprises a first heavy chain comprising a first variable heavy domain, a CH1 domain and an Fc domain, and an additional variable heavy domain (vhl-CHI-[optional domain linker]-vh2-hinge-CH2-CH3). The light chain domain is covalently linked using a domain linker between the C-terminus of the CH1 domain of the heavy chain constant domain and the N-terminus of the first Fc domain. This embodiment further utilizes a common light chain comprising variable light domains and constant light domains associated with the heavy chains that form two identical Fabs that bind to the target. For many of the embodiments herein, these constructs include asymmetric mutants, pI mutants, truncation mutants, additional Fc mutants, etc. as desired and described herein.

[0126] A further heterodimeric scaffold useful in the context of the present disclosure is the one-arm central-scFv format shown in FIG. 18C. In this embodiment, one monomer contains only the Fc domain, while the other monomer uses an inserted scFv domain, thereby forming the second antigen-binding domain. In this format, the Fab portion binds one target and the scFv binds the other target. The scFv domain is inserted between the Fc domain and the CH1-Fv region of one of the monomers. In this embodiment, one monomer contains a first heavy chain containing a first variable heavy domain, a CH1 domain and an Fc domain, together with an scFv containing an scFv variable light domain, an scFv linker and an scFv variable heavy domain. The scFv is covalently linked between the C-terminus of the CH1 domain of the heavy chain constant domain and the N-terminus of the first Fc domain using a domain linker. The second monomer contains an Fc domain. This embodiment further utilizes a light chain that includes a variable light domain and a constant light domain associated with the heavy chain forming a Fab. For many of the embodiments herein, these constructs include asymmetric mutants, pI mutants, truncation mutants, additional Fc mutants, etc. as desired and described herein.

[0127] In addition, the Fc domain of the one-arm central-scFv format optionally comprises an asymmetric mutant (e.g., selected from the group consisting of S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W and T366S / L368A / Y407V Y349C:T366W / S354C), optionally a truncation mutant, optionally a charged scFv linker, and the heavy chain comprises a pI mutant. In some embodiments, the one-arm central scFv format comprises asymmetric mutants, pI mutants and truncation mutants. Thus, some embodiments comprise a) a first monomer comprising the asymmetric mutants S364K / E357Q, the truncation mutants E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain constituting an Fv that binds to a first target together with a first variable light chain domain of the light chain and a second variable heavy chain domain; b) a first monomer comprising the asymmetric mutants L368D / K370S, the pI mutants N208D / Q295E / N384 D / Q418E / N421D, truncation mutants E233P / L234V / L235A / G236del / S267K and a first variable heavy domain constituting an Fv that binds to a first target together with the first variable light domain and a second variable light chain that forms an Fv that binds to a second target together with the second variable heavy chain; and c) a bottle opener format comprising the first variable light domain and a constant light domain. In some embodiments, the one arm central scFv format comprises an asymmetric mutant, a pi mutant, a truncation mutant and an FcRn mutant.Thus, some embodiments include a) a first monomer comprising an asymmetric mutant S364K / E357Q, a truncation mutant E233P / L234V / L235A / G236del / S267K, an FcRn mutant M428L / N434S and a first variable heavy chain domain constituting an Fv that binds to a first target together with a first variable light chain domain of the light chain and a second variable heavy chain domain; b) a first monomer comprising an asymmetric mutant L368D / K370S, a pi mutant N208D / Q295E / N384D / Q41 a second monomer comprising a first variable heavy chain domain which together with the first variable light chain domain constitutes an Fv that binds to a first target as outlined herein and a second variable light chain which together with the second variable heavy chain forms an Fv that binds to a second target; and c) a light chain comprising the first variable light chain domain and a constant light chain domain.

[0128] A further heterodimeric scaffold useful in the present disclosure is the one-arm scFv-mAb format shown in Figure 18D. In this embodiment, one monomer contains only the Fc domain, while the other monomer uses a scFv domain linked at the N-terminus of the heavy chain, generally via the use of a linker: vh-scFv linker-vl-[optional domain linker]-CHl-hinge-CH2-CH3 or (opposite orientation) vl-scFv linker-vh-[optional domain linker]-CHl-hinge-CH2-CH3. In this format, the Fab portion binds one target and the scFv binds the other target. This embodiment further utilizes a light chain that includes a variable light domain and a constant light domain associated with the heavy chain that forms the Fab. For many of the embodiments herein, these constructs include asymmetric mutants, pI mutants, truncation mutants, additional Fc mutants, etc. as desired and described herein.

[0129] The Fc domain of the one-arm scFv-mAb comprises an asymmetric mutant (e.g., selected from the group consisting of S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W; and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally a truncation mutant, optionally a charged scFv linker, and the heavy chain comprises a pi mutant. In some embodiments, the one-arm scFv-mAb format comprises asymmetric mutants, pi mutants and truncation mutants. Thus, some embodiments comprise a) a first monomer comprising the asymmetric mutant S364K / E357Q, the truncation mutants E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain constituting an Fv that binds to a first target together with a first variable light chain domain of the light chain and a second variable heavy chain domain; b) a first monomer comprising the asymmetric mutant L368D / K370S, the pi mutants N208D / Q295E / N384D / Q418 ... 421D, truncation mutants E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain which together with the first variable light chain domain constitutes an Fv that binds to a first target as outlined herein and a second variable light chain which together with the second variable heavy chain forms an Fv that binds to a second target; and c) a bottle opener format comprising a light chain comprising the first variable light chain domain and a constant light chain domain. In some embodiments, the one-arm scFv-mAb format comprises asymmetric mutants, pI mutants, truncation mutants and FcRn mutants.Thus, some embodiments include a) a first monomer comprising an asymmetric mutant S364K / E357Q, a truncation mutant E233P / L234V / L235A / G236del / S267K, an FcRn mutant M428L / N434S and a first variable heavy chain domain constituting an Fv that binds to a first target together with a first variable light chain domain of the light chain and a second variable heavy chain domain; b) a first monomer comprising an asymmetric mutant L368D / K370S, a pi mutant N208D / Q295E / N384D / Q418E / N 421D, truncation mutants E233P / L234V / L235A / G236del / S267K, FcRn mutant M428L / N434S and a first variable heavy chain domain which together with the first variable light chain domain constitutes an Fv that binds to a first target as outlined herein and a second variable light chain which together with the second variable heavy chain forms an Fv that binds to a second target; and c) a bottle opener format comprising a light chain comprising the first variable light chain domain and a constant light chain domain.

[0130] Another heterodimeric scaffold useful in the present disclosure is the mAb-scFv format shown in Figure 18E. In this embodiment, the format relies on the use of N-terminal attachment of an scFv to one of the monomers, thereby forming a third antigen-binding domain, where the Fab portions of the two monomers bind one target and the "additional" scFv domain binds a different target. In this embodiment, the first monomer comprises a first heavy chain (comprising a variable heavy domain and a constant domain) with a covalently attached scFv at its N-terminus comprising a scFv variable light domain, a scFv linker and a scFv variable heavy domain in either orientation ((vhl-scFv linker-vll-[optional domain linker]-vh2-CHl-hinge-CH2-CH3) or (with the scFv in the opposite orientation) (vll-scFv linker-vhl-[optional domain linker]-vh2-CHl-hinge-CH2-CH3)). This embodiment further utilizes a common light chain that includes a variable light domain and a constant light domain associated with the heavy chain that forms two identical Fabs that bind to one of the target antigens. For many of the embodiments herein, these constructs include asymmetric mutants, pi mutants, truncation mutants, additional Fc mutants, etc. as desired and described herein.

[0131] The Fc domain of the scFv-mAb format optionally comprises an asymmetric mutant (e.g., selected from the group consisting of S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W; and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally a truncation mutant, optionally a charged scFv linker, and the heavy chain comprises a pI mutant. In some embodiments, the mAb-scFv format includes asymmetric mutants, pI mutants and truncation mutants. Thus, some embodiments include a) a first monomer comprising an asymmetric mutant S364K / E357Q, a truncation mutant E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain constituting an Fv that binds to a first target together with a first variable light chain domain of the light chain and a second variable heavy chain domain; b) an asymmetric mutant L368D / K370S, a pI mutant N208D / Q295E / N384D / Q418E / N 421D, truncation mutants E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain which together with the first variable light chain domain constitutes an Fv that binds to a first target as outlined herein and a second variable light chain which together with the second variable heavy chain forms an Fv that binds to a second target; and c) a bottle opener format comprising a light chain comprising the first variable light chain domain and a constant light chain domain. In some embodiments, the mAb-scFv format comprises asymmetric mutants, pI mutants, truncation mutants and FcRn mutants.Thus, some embodiments include a) a first monomer comprising an asymmetric mutant S364K / E357Q, a truncation mutant E233P / L234V / L235A / G236del / S267K, an FcRn mutant M428L / N434S and a first variable heavy chain domain constituting an Fv that binds to a first target together with a first variable light chain domain of the light chain and a second variable heavy chain domain; b) a first monomer comprising an asymmetric mutant L368D / K370S, a pI mutant N208D / Q295E / N384D / a second monomer comprising a first variable heavy chain domain which together with the first variable light chain domain constitutes an Fv that binds to a first target and a second variable light chain which together with the second variable heavy chain forms an Fv that binds to a second target; and c) a bottle opener format comprising a light chain comprising the first variable light chain domain and a constant light chain domain.

[0132] The disclosure also provides dual-scFv formats, such as that shown in Figure 18B. In this embodiment, the heterodimeric antigen binding protein is composed of two scFv-Fc monomers, both in either the (vh-scFv linker-vl-[optional domain linker]-CH2-CH3) format or the (vl-scFv linker-vh-[optional domain linker]-CH2-CH3) format, or with one monomer in one orientation and the other monomer in the other orientation. The Fc domain of the dual-scFv format optionally comprises an asymmetric mutant (e.g., selected from the group consisting of S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411T / E360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W; and T366S / L368A / Y407V / Y349C:T366W / S354C), optionally a truncation mutant, optionally a charged scFv linker, and the heavy chain comprises a pI mutant.

[0133] In some embodiments, the dual scFv format comprises asymmetric mutants, pI mutants and truncation mutants. Thus, some embodiments comprise a) a first monomer comprising an asymmetric mutant S364K / E357Q, a truncation mutant E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain that together with a first variable light chain domain of the light chain constitute an Fv that binds to a first target, and a second variable heavy chain domain; b) an asymmetric mutant L368D / K370S, a pI mutant N208D / Q295E / N384D / Q41 and c) a light chain comprising a first variable light chain domain and a constant light chain domain. In some embodiments, the dual scFv formats include asymmetric mutants, pI mutants, truncation mutants E233P / L234V / L235A / G236del / S267K and a first variable heavy chain domain which together with the first variable light chain domain constitutes an Fv that binds to a first target as outlined herein and a second variable light chain which together with the second variable heavy chain forms an Fv that binds to a second target; and c) a light chain comprising a first variable light chain domain and a constant light chain domain. In some embodiments, the dual scFv formats include asymmetric mutants, pI mutants, truncation mutants and FcRn mutants. Thus, some embodiments include a) a first monomer comprising an asymmetric mutant S364K / E357Q, a truncation mutant E233P / L234V / L235A / G236del / S267K, an FcRn mutant M428L / N434S and a first variable heavy chain domain constituting an Fv that binds to a first target together with a first variable light chain domain of the light chain and a second variable heavy chain domain; b) a first monomer comprising an asymmetric mutant L368D / K370S, a pi mutant N208D / Q295E / N384D / a second monomer comprising a first variable heavy chain domain which together with the first variable light chain domain constitutes an Fv that binds to a first target and a second variable light chain which together with the second variable heavy chain forms an Fv that binds to a second target; and c) a bottle opener format comprising a light chain comprising the first variable light chain domain and a constant light chain domain.

[0134] Further description of antibody formats is provided in WO 2017 / 218707, which is incorporated herein by reference.

[0135] antibody binding The bispecific antigen binding proteins (e.g., heterodimeric antibodies) of the present disclosure, in various embodiments, bind to CD3 and STEAP1. The different binding regions are independently -4 M or less, 10 -5 M or less, 10 -6 M or less, 10 -7 M or less, 10 -8 M or less, 10 -9 M or less, 10 -10 M or less, 10 -11 M or less or 10 -12 The binding affinities disclosed herein are for their corresponding antigens with a KD of less than or equal to M, where KD refers to the dissociation rate of a specific antibody-antigen interaction. Binding affinities are described further above. The STEAP1 binding region does not need to bind to STEAP1 with the same affinity as, for example, the CD3 binding region binds to CD3. The binding affinities disclosed in relation to bispecific antigen-binding proteins also apply to any of the monospecific constructs described herein, including constructs that bind to PD-1.

[0136] Additional antibody modifications In addition to the modifications outlined above, other modifications can be made. For example, the molecule can be stabilized by incorporating disulfide bridges that link the VH and VL domains (Reiter et al., 1996, Nature Biotech. 14:1239-1245, incorporated by reference in its entirety). In addition, there are various covalent modifications of antibodies that can be made as outlined below.

[0137] Covalent modifications of antibodies are included within the scope of the present disclosure, which are generally, although not always, performed post-translationally. For example, some types of covalent modifications of antibodies are introduced into the molecule by reacting specific amino acid residues of the antibody with organic derivatizing agents that can react with selected side chains or N- or C-terminal residues.

[0138] Cysteinyl residues most commonly are reacted with α-haloacetates (and corresponding amines), such as chloroacetic acid or chloroacetamide to give carboxymethyl or carboxyamidomethyl derivatives. Cysteinyl residues can also be derivatized by reaction with bromotrifluoroacetone, α-bromo-β-(5-imidozoyl)propionic acid, chloroacetyl phosphate, N-alkylmaleimides, 3-nitro-2-pyridyl disulfide, methyl 2-pyridyl disulfide, p-chloromercuribenzoate, 2-chloromercuri-4-nitrophenol, or chloro-7-nitrobenzo-2-oxa-1,3-diazole, and the like.

[0139] In addition, modification with cysteine ​​is particularly useful in antibody-drug conjugate (ADC) applications, which are described further below. In some embodiments, the constant region of an antibody can be engineered to contain one or more cysteines that are specifically "thiol-reactive" to allow for more specific and controlled placement of drug moieties. See, for example, U.S. Patent No. 7,521,541, which is incorporated herein by reference in its entirety.

[0140] Histidyl residues are derivatized by reaction with diethylpyrocarbonate at pH 5.5-7.0 because this agent is relatively specific for the histidyl side chain. Para-bromophenacyl bromide is also useful; the reaction is preferably performed in 0.1 M sodium cacodylate at pH 6.0.

[0141] Lysinyl and amino terminal residues react with succinic or other carboxylic acid anhydrides. Derivatization with these agents reverses the charge of lysinyl residues. Other suitable agents for derivatizing alpha-amino-containing residues include imidoesters such as methyl picolinimidate; pyridoxal phosphate; pyridoxal; chloroborohydrides; trinitrobenzenesulfonic acid; O-methylisourea; 2,4-pentanedione; and transaminase-catalyzed reactions with glyoxylate.

[0142] Arginyl residues are modified by reaction with one or more conventional reagents, among them phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin. Due to the high pKa of the guanidine functional group, derivatization of arginine residues requires that the reaction be carried out under alkaline conditions. Furthermore, these reagents can react with lysine groups and the arginine epsilon-amino group.

[0143] The specific modification of tyrosyl residues may be undertaken for the introduction of spectral labels into tyrosyl residues by reaction with aromatic diazonium compounds or tetranitromethane. Most commonly, N-acetylimidizole and tetranitromethane are used to form O-acetyl tyrosyl species and 3-nitro derivatives, respectively. The chloramine T method described above, in which tyrosyl residues are iodinated with 125I or 131I to prepare labeled proteins for use in radioimmunoassay, is preferred.

[0144] Carboxyl side groups (aspartyl or glutamyl) are selectively modified by reaction with carbodiimides (R'-N=C=N--R'), where R and R' are optionally different alkyl groups, such as 1-cyclohexyl-3-(2-morpholinyl-4-ethyl)carbodiimide or 1-ethyl-3-(4-azonia-4,4-dimethylpentyl)carbodiimide. Furthermore, aspartyl and glutamyl residues are converted to asparaginyl and glutaminyl residues by reaction with ammonium ions.

[0145] Derivatization with bifunctional agents is useful for crosslinking antibodies to water-insoluble support matrices or surfaces for use in a variety of methods. Commonly used crosslinking agents include homobifunctional imidoesters including 1,1-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters such as esters with 4-azidosalicylic acid, disuccinimidyl esters such as 3,3'-dithiobis(succinimidylpropionate), and bifunctional maleimides such as bis-N-maleimido-1,8-octane. Derivatization agents such as methyl-3-[(p-azidophenyl)dithio]propioimidate provide photoactivatable intermediates that can form crosslinks in the presence of light. Instead, reactive water-insoluble matrices are utilized for protein immobilization, such as cyanogen-bromide activated carbohydrates and reactive materials described in U.S. Pat. Nos. 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537; and 4,330,440, all of which are incorporated by reference in their entirety.

[0146] Glutaminyl and asparaginyl residues are frequently deamidated to the corresponding glutamyl and aspartyl residues, respectively. Alternatively, these residues are deamidated under mildly acidic conditions. Either form of these residues is included within the scope of the disclosure.

[0147] Other modifications include hydroxylation of proline and lysine, phosphorylation of the hydroxyl group of seryl or threonyl residues, methylation of the α-amino groups of lysine, arginine and histidine side chains (TECreighton, Proteins: Structure and Molecular Properties, W.H. Freeman & Co., San Francisco, pp. 79-86

[1983] (incorporated herein in its entirety)), acetylation of the N-terminal amine and amidation of any C-terminal carboxyl group.

[0148] In addition, as will be appreciated by one of skill in the art, labels (including fluorescent, enzymatic, magnetic, radioactive, etc.) can be added to any of the antigen binding proteins described herein (and other compositions of the disclosure).

[0149] Glycosylation Another type of covalent modification is an alteration in glycosylation. In another embodiment, the antibodies (or other types of antigen binding proteins) disclosed herein may be modified to include one or more engineered glycoforms. As used herein, "engineered glycoform" refers to a carbohydrate composition covalently attached to an antibody, said carbohydrate composition being chemically distinct from that of the parent antibody. Engineered glycoforms may be useful for a variety of purposes, including but not limited to enhancing or reducing effector function. A preferred form of engineered glycoform is afucosylation, which has been shown to correlate with increased ADCC function, presumably via tighter binding to the FcγRIIIa receptor. In this context, "afucosylation" means that the majority of antibodies produced in a host cell are substantially devoid of fucose, e.g., 90%, 95% or 98% of the antibodies produced have no recognizable fucose as a component of the antibody's carbohydrate chain (generally attached at N297 in the Fc region). A defined functionally afucosylated antibody generally exhibits an affinity for the FcγRIIIa receptor of at least 50% or more.

[0150] Optionally, the heterodimeric antibody comprises a sequence modification to remove another glycosylation site, e.g., at one or more of positions 292, 297, or 302. One non-limiting example comprises the introduction of one or more suitable effector function (SEFL2) mutations (e.g., in the IgG1 backbone), as further described, e.g., in U.S. Pat. No. 9,546,203, which is incorporated herein by reference in its entirety and in particular with respect to its description of SEFL2 mutations. This modification may be used in addition to any other modification disclosed herein, e.g., the N67Q modification to reduce deamidation.

[0151] Engineered glycoforms can be generated by a variety of methods known in the art, see, for example, Umana et al., 1999, Nat Biotechnol 17:176-180; Davies et al., 2001, Biotechnol Bioeng 74:288-294; Shields et al., 2002, J Biol Chem 277:26733-26740; Shinkawa et al., 2003, J Biol Chem 277:26733-26740, all of which are incorporated by reference in their entirety. 278:3466-3473; U.S. Patent No. 6,602,684; U.S. Patent Application Publication Nos. 2003 / 0157108 and 2003;0003097; and WO 00 / 61739 A1 and WO 01 / 29246 A1, WO 02 / 31140 A1, WO 02 / 30954 A1, as well as Potelligent® technology (Biowa, Inc., Princeton, NJ) and GlycoMAb® glycosylation engineering technology (Glycart Biotechnology AG, Zuerich, Switzerland). Many of these techniques are based on controlling the level of fucosylation and / or bisecting oligosaccharides covalently attached to the Fc region, for example, by expressing IgG in different organisms or engineered or otherwise cell lines (e.g., Lec-13 CHO cells or rat hybridoma YB2 / 0 cells), by modulating enzymes involved in the glycosylation pathway (e.g., FUT8 [α1,6-fucosyltransferase] and / or β1-4-N-acetylglucosaminyltransferase III [GnTIII]), or by modifying the carbohydrates after the IgG is expressed. For example, "glycoengineered antibody technology" works by adding modified sugars that inhibit fucosylation during production; see, for example, US Patent Publication No. 20090317869, which is incorporated by reference in its entirety. Engineered glycoforms usually refer to different carbohydrates or oligosaccharides; thus, antibodies may contain engineered glycoforms.

[0152] Alternatively, engineered glycoforms can refer to IgG variants that contain different carbohydrates or oligosaccharides. As is known in the art, glycosylation patterns can depend on both the sequence of the protein (e.g., the presence or absence of specific glycosylated amino acid residues, discussed below) or the host cell or organism that produces the protein. Specific expression systems are discussed below.

[0153] Glycosylation of polypeptides is usually either N-linked or O-linked. N-linked refers to the attachment of a sugar chain to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine, where X is any amino acid except proline, are the recognition sequences for enzymatic attachment of a sugar chain to the asparagine side chain. Thus, the presence of either of these tripeptide sequences in a polypeptide creates a potential glycosylation site. O-linked glycosylation most commonly refers to the addition of one of the sugars N-acetylgalactosamine, galactose, or xylose to serine or threonine, although the hydroxyamino acids 5-hydroxyproline or 5-hydroxylysine may also be used.

[0154] Addition of glycosylation sites to the antibody is conveniently accomplished by altering the amino acid sequence to include one or more of the above-mentioned tripeptide sequences (for N-linked glycosylation sites). Alterations may also be made by the addition to, or substitution by, one or more serine or threonine residues from the starting sequence (for O-linked glycosylation sites). For ease of use, the antibody amino acid sequence is preferably altered by mutating the DNA encoding the target polypeptide at preselected bases such that the change at the DNA level specifically generates a codon that will be translated into the desired amino acid.

[0155] Another means of increasing the number of carbohydrate chains on an antigen-binding protein (e.g., an antibody) is by chemical or enzymatic coupling of glycosides to the protein. These procedures are advantageous in that they do not require production of the protein in a host cell that has glycosylation capabilities for N-linked and O-linked glycosylation. Depending on the linkage mode used, sugars can be added to (a) arginines and histidines, (b) free carboxyl groups, (c) free sulfhydryl groups such as those of cysteine, (d) free hydroxyl groups such as those of serine, threonine or hydroxyproline, (e) aromatic residues such as those of phenylalanine, tyrosine or tryptophan, or (f) the amide group of glutamine. These methods are described in WO 87 / 05330 and Aplin and Wriston, 1981, CRC Crit. Rev. Biochem., pp. 259-306, both of which are incorporated by reference in their entirety.

[0156] Removal of carbohydrate moieties present on the starting antibody (e.g., post-translationally) can be accomplished chemically or enzymatically. Chemical deglycosylation requires exposure of the protein to the compound trifluoromethanesulfonic acid, or an equivalent compound. This treatment cleaves most or all sugars except the linking sugar (N-acetylglucosamine or N-acetylgalactosamine), while leaving the polypeptide intact. Chemical deglycosylation is described by Hakimuddin et al., 1987, Arch. Biochem. Biophys. 259:52 and Edge et al., 1981, Anal. Biochem. 118:131, both of which are incorporated by reference in their entireties. Enzymatic cleavage of carbohydrate moieties on polypeptides can be accomplished by the use of a variety of endo- and exoglycosidases as described by Thotakura et al., 1987, Meth. Enzymol. 138:350, which are incorporated by reference in their entireties. Glycosylation at potential glycosylation sites can be prevented by the use of the compound tunicamycin as described by Duskin et al., 1982, J. Biol. Chem. 257:3105, which is incorporated by reference in its entirety. Tunicamycin blocks the formation of protein-N-glycosidic bonds.

[0157] Another type of covalent modification of antibodies involves linking the antibody to various non-proteinaceous polymers, including, but not limited to, various polyols such as polyethylene glycol, polypropylene glycol, or polyoxyalkylenes, for example, in the manner described in the 2005-2006 PEG catalog from Nektar Therapeutics (available on the Nektar website) or in U.S. Pat. Nos. 4,640,835; 4,496,689; 4,301,144; 4,670,417; 4,791,192; or 4,179,337, all of which are incorporated by reference in their entirety. In addition, as is known in the art, amino acid substitutions can be made at various positions within the antibody to facilitate the addition of polymers such as PEG. See, for example, U.S. Patent Application Publication No. 2005 / 0114037A1, which is incorporated by reference in its entirety.

[0158] Additional Fc variants for added functionality In addition to the pI amino acid variants and other variants described above, there are a number of useful Fc amino acid modifications that can be made for a variety of reasons, including, but not limited to, modifying binding to one or more FcγR receptors, modifying binding to the FcRn receptor, etc. The following modifications can be utilized in addition to or instead of any of the above modifications.

[0159] FcγR variants There are several useful Fc substitutions that can be made to alter binding to one or more of the FcγR receptors. Substitutions that result in increased and decreased binding can be useful. For example, increased binding to FcγRIIIa generally results in increased ADCC (antibody-dependent cellular cytotoxicity; a cellular reaction in which non-specific cytotoxic cells expressing FcγR recognize bound antibodies on target cells and subsequently cause lysis of the target cells). Similarly, decreased binding to FcγRIIb (an inhibitory receptor) can be beneficial in some situations as well. Amino acid substitutions that are useful in this disclosure include those listed in US Patent Publication Nos. 2006 / 0024298 (particularly FIG. 41), 2006 / 0121032, 2006 / 0235208, and 2007 / 0148170, all of which are expressly incorporated herein by reference in their entirety and specifically with respect to the variants disclosed. Particular mutants that are useful include, but are not limited to, 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D, 332E / 330L and 299T.

[0160] In addition, there are additional Fc substitutions useful in increasing binding to the FcRn receptor and increasing serum half-life, as specifically disclosed in U.S. Patent Application Publication No. 2009 / 0163699, which is incorporated herein by reference in its entirety, including, but not limited to, 434S, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L and 259I / 308F / 428L.

[0161] Fc truncation mutants Additional variants useful in the context of the present disclosure are those that truncate (e.g., reduce or eliminate) binding to Fcγ receptors. This may be desirable to reduce a potential mechanism of action of a heterodimeric antibody (e.g., reduce ADCC activity). Some suitable Fc truncation variants are shown in Figure 6, which may optionally and independently be included in combination with or excluded from any other heterodimerization variants, including pI and conformational variants.

[0162] In some embodiments, a first monomer ("negative side") containing the pI mutants N208D / Q295E / N384D / Q418E / N421D, the asymmetric mutants 368D / 370S and the truncation mutants E233P / L234V / L235A / G236del / S267K paired with a positive side that does not contain a pI mutant and contains the asymmetric mutants S364K / E357Q and the truncation mutants E233P / L234V / L235A / G236del / S267K (both monomers optionally containing FcRn mutants 428L / 434S), is particularly useful, where the positive side is a monomer that contains an scFv and contains a charged scFv linker. The second embodiment comprises a nucleotide sequence comprising I199T / N203D / K274Q / K276K, pI mutants Q196K / I199T / P271R / P228R / N276K, asymmetric mutants S364K / E357Q and truncation mutants E233P / L234V / L235A / G236del / S267K paired with a positive side (both monomers optionally containing FcRn mutants 428L / 434S). A first negative monomer is utilized that contains R355Q / Q419E / K447del, asymmetric mutants S364K / E357Q and truncation mutants E233P / L234V / L235A / G236del / S267K (both monomers optionally containing FcRn mutants 428L / 434S), and the positive side is a monomer that contains an scFv and contains a charged scFv linker. The third embodiment is a positive monomer with no pI mutations, I199T / N203D / K274Q / R355Q / N384S / K392N / V paired with an asymmetric mutation S364K / E357Q and a truncation mutation E233P / L234V / L235A / G236del / S267K (both monomers optionally containing FcRn mutations 428L / 434S). A first negative monomer is utilized that contains 397M / Q419E / K447del, asymmetric mutants S364K / E357Q and truncation mutants E233P / L234V / L235A / G236del / S267K (both monomers optionally containing FcRn mutants 428L / 434S), and the positive side is a monomer that contains an scFv and contains a charged scFv linker.A fourth embodiment utilizes a first monomer ("negative side") containing pI mutants N208D / Q295E / N384D / Q418E / N421D, asymmetric mutants 368D / 370S and truncation mutants E233P / L234V / L235A / G236del / S239 paired with a positive side that does not contain pI mutants and contains asymmetric mutants S364K / E357Q and truncation mutants E233P / L234V / L235A / G236del / S239K (both monomers optionally containing FcRn mutants 428L / 434S). A fifth embodiment utilizes a first negative side monomer (optionally containing FcRn mutant 428L / 434S) comprising I199T / N203D / K274Q / R355Q / Q419E / K447del, asymmetric mutants S364K / E357Q and truncation mutants E233P / L234V / L235A / G236del / S239K paired with a positive side (optionally containing FcRn mutant 428L / 434S) comprising pI mutants Q196K / I199T / P271R / P228R / N276K, asymmetric mutants S364K / E357Q and truncation mutants E233P / L234V / L235A / G236del / S239K. A sixth embodiment is a combination of I199T / N203D / K274Q / R355Q / N384S / K392N / V397M / Q419E / K44K paired with the asymmetric mutants S364K / E357Q and truncation mutants E233P / L234V / L235A / G236del / S239K of the positive monomer (both monomers optionally containing the FcRn mutants 428L / 434S). The first negative monomer (optionally containing FcRn mutants 428L / 434S) is utilized, which contains 7del, asymmetric mutants S364K / E357Q and truncation mutants E233P / L234V / L235A / G236del / S267K, and the positive side is an scFv monomer and contains a charged scFv linker (particularly when the scFv is anti-CD3).A seventh embodiment utilizes a first monomer ("negative side") containing pI mutants N208D / Q295E / N384D / Q418E / N421D, asymmetric mutants 368D / 370S and truncation mutants S239K / S267K paired with a positive side ("negative side") that does not contain pI mutants and contains asymmetric mutants S364K / E357Q and truncation mutants S239K / S267K (both monomers optionally containing FcRn mutants 428L / 434S), where the positive side is an scFv monomer and contains a charged scFv linker. An eighth embodiment utilizes a first negative side monomer (optionally containing FcRn mutant 428L / 434S) comprising I199T / N203D / K274Q / R355Q / Q419E / K447del, asymmetric mutants S364K / E357Q and truncation mutants S239K / S267K paired with a positive side (optionally containing FcRn mutant 428L / 434S) comprising pI mutants Q196K / I199T / P271R / P228R / N276K, asymmetric mutants S364K / E357Q and truncation mutants S239K / S267K, where the positive side is an scFv monomer and contains a charged scFv linker. A ninth embodiment utilizes I199T / N203D / K274Q / R355Q / N384S / K392N / V397M / Q419E / K447del, a first negative monomer comprising asymmetric mutants S364K / E357Q and truncation mutants S239K / S267K (optionally both monomers containing FcRn mutants 428L / 434S) paired with a positive monomer (optionally both monomers containing FcRn mutants 428L / 434S) that does not have a pI mutant and has asymmetric mutants S364K / E357Q and truncation mutants S239K / S267K, where the positive side is an scFv monomer and contains a charged scFv linker.A tenth embodiment utilizes a first monomer ("negative side") containing pI mutants N208D / Q295E / N384D / Q418E / N421D, asymmetric mutants 368D / 370S and truncation mutants S267K / P329K paired with a positive side ("negative side") that does not contain pI mutants and contains asymmetric mutants S364 / E357Q and truncation mutants S267K / P329K (both monomers optionally containing FcRn mutants 428L / 434S), where the positive side is an scFv monomer and contains a charged scFv linker. An eleventh embodiment utilizes a first negative side monomer (optionally containing FcRn mutant 428L / 434S) comprising I199T / N203D / K274Q / R355Q / Q419E / K447del, asymmetric mutants S364K / E357Q and truncation mutants S267K / P329K paired with a positive side (optionally containing FcRn mutant 428L / 434S) comprising pI mutants Q196K / I199T / P271R / P228R / N276K, asymmetric mutants S364K / E357Q and truncation mutants S267K / P329K, where the positive side is an scFv monomer and contains a charged scFv linker. A twelfth embodiment utilizes a first negative monomer comprising I199T / N203D / K274Q / R355Q / N384S / K392N / V397M / Q419E / K447del, asymmetric mutants S364K / E357Q and truncation mutants S267K / P329K (optionally both monomers containing FcRn mutants 428L / 434S) paired with a positive monomer (optionally both monomers containing FcRn mutants 428L / 434S) that does not have a pI mutant and has asymmetric mutants S364K / E357Q and truncation mutants S267K / P329K, where the positive side is a scFv monomer and contains a charged scFv linker.

[0163] In various embodiments, a first monomer (i.e., a "Fab-scFv-Fc" heavy chain) comprises a first heavy chain comprising a first variable heavy chain domain, a first constant heavy chain comprising a first CH1 domain and a first Fc domain, an scFv that binds human CD3 and comprises an scFv variable light chain domain, an scFv linker and an scFv variable heavy chain domain heavy chain, the first monomer comprises a deletion in the upper hinge and CH2 and CH3 substitutions are introduced. The substitutions include, for example, one or more (e.g., all) of E233P, delL234, L235V, G236A, S267K, r292c, n297g, v302c, E357Q and S364K (EU numbering, lower case letters refer to SEFL2 substitutions). A second monomer comprising a second heavy chain comprising a second constant heavy chain comprising a second variable heavy domain and a second Fc domain optionally comprises one or more (e.g., all) of the following mutations: N208D, E233P, delL234, L235V, G236A, S267K, r292c, Q295E, n297g, v302c, L368D, K370S, N384D, Q418E and N421D (EU numbering, lower case letters refer to SEFL2 substitutions).

[0164] Linker A "linker" herein is referred to as a "linker sequence" or "spacer" or grammatically correct equivalents. Homo- or heterobifunctional linkers are well known (see 1994 Pierce Chemical Company catalog, technical section on cross-linkers, pages 155-200, incorporated by reference in its entirety). (Note the distinction between "linkers" in general, "scFv linkers" and "charged scFv linkers".) Several strategies can be used to covalently link molecules together. These include, but are not limited to, polypeptide bonds between the N- and C-termini of proteins or protein domains, linkages via disulfide bonds and linkages via chemical cross-linking reagents. In one aspect of this embodiment, the linker is a peptide bond generated by recombinant technology or peptide synthesis. The linker peptide may primarily comprise the following amino acid residues: Gly, Ser, Ala or Thr. The linker peptide should be of sufficient length to link the two molecules such that they assume the correct conformation relative to one another and thus retain the desired activity. In one embodiment, the linker is about 1-50 amino acids long, preferably about 1-30 amino acids long. In one embodiment, linkers of 1-20 amino acids long may be used. Useful linkers include glycine-serine polymers, such as (GS)n, (GSGGS)n (SEQ ID NO: 178), (GGGGS)n (SEQ ID NO: 179), and (GGGS)n (SEQ ID NO: 180), where n is an integer of at least 1; glycine-alanine polymers; alanine-serine polymers; and other flexible linkers. Alternatively, a variety of non-proteinaceous polymers may be useful as linkers, including, but not limited to, polyethylene glycol (PEG), polypropylene glycol, polyoxyalkylenes, or copolymers of polyethylene glycol and polypropylene glycol.

[0165] Other linker sequences may include any sequence of any length of the CL / CH1 domain, but not all residues of the CL / CH1 domain; for example, the first 5-12 amino acid residues of the CL / CH1 domain. Linkers may be derived from immunoglobulin light chains, such as Cκ or Cλ. Linkers may be derived from immunoglobulin heavy chains of any isotype, including Cγ1, Cγ2, Cγ3, Cγ4, Cα1, Cα2, Cδ, Cε, and Cμ. Linker sequences may also be derived from other proteins, such as Ig-like proteins (e.g., TCR, FcR, KIR), sequences from hinge regions, and other naturally occurring sequences from other proteins.

[0166] In some embodiments, the linker is a "domain linker" used to link together any two domains outlined herein. For example, in Figure 18F, the domain linker that connects the C-terminus of the CH1 domain of the Fab to the N-terminus of the scFv can be present with another optional domain linker that connects the C-terminus of the scFv to the CH2 domain (although in many embodiments a hinge is used as this domain linker). In some embodiments, the linker is a hinge region and fragments thereof.

[0167] Antibody-drug conjugates The antigen binding proteins (e.g., antibodies or heterodimeric antibodies) of the present disclosure are optionally conjugated with a drug to form an antibody-drug conjugate (ADC). In general, ADCs are used in a variety of contexts, including cancer research applications, where the use of antibody-drug conjugates for localized delivery of cytotoxic or cytostatic agents allows for targeted delivery of the drug moiety to tumors, which may allow for greater efficacy, lower toxicity, and the like. Overviews of this technology are provided in Ducry et al., Bioconjugate Chem., 21:5-13 (2010); Carter et al., Cancer J. 14(3):154 (2008); and Senter, Current Opin. Chem. Biol. 13:235-244 (2009), all of which are incorporated herein by reference in their entirety.

[0168] Generally, conjugation is performed by covalent attachment to the antibody as described further below, and generally relies on a linker, often a peptide bond, which can be designed to be susceptible to cleavage by proteases or otherwise at the target site, as described herein. In addition, linker-drug unit (LU-D) attachment can be achieved by attachment to a cysteine ​​in the antibody. The number of drug moieties per antibody can vary depending on the conditions of the reaction, and can range from 1:1 to 10:1 drug:antibody. As will be understood by those skilled in the art, the actual number is an average value.

[0169] The drug of the ADC may be selected from any of a number of agents, including, but not limited to, a cytotoxic agent such as a chemotherapeutic agent, a growth inhibitory agent, a toxin (e.g., an enzymatically active toxin of bacterial, fungal, plant or animal origin or a fragment thereof), or a radioisotope (i.e., a radioconjugate). The present disclosure further provides methods of using the ADC.

[0170] In the context of the present disclosure, drugs for use include cytotoxic drugs, particularly those used for cancer therapy.Such drugs generally include DNA damaging drugs, antimetabolites, natural products and their analogs.Exemplary classes of cytotoxic drugs include enzyme inhibitors such as dihydrofolate reductase inhibitors and thymidylate synthase inhibitors, DNA interfering agents, DNA cleaving agents, topoisomerase inhibitors, the anthracycline family of drugs, vinca drugs, mitomycins, bleomycins, cytotoxic nucleosides, the pteridine family of drugs, diynenes, podophyllotoxins, dolastatins, maytansinoids, differentiation inducers and taxol.

[0171] Members of these classes include, for example, methotrexate, methopterin, dichloromethotrexate, 5-fluorouracil, 6-mercaptopurine, cytosine arabinoside, melphalan, leurosine, leurosideine, actinomycin, daunorubicin, doxorubicin, mitomycin C, mitomycin A, caminomycin, aminopterin, tallysomycin, podophyllotoxin and podophyllotoxin derivatives such as etoposide or etoposide phosphate, vinblastine, and the like. Examples of compounds that may be used in the treatment of rheumatoid arthritis include, but are not limited to, vincristine, vindesine, taxanes including taxol, taxotere retinoic acid, butyric acid, N8-acetylspermidine, camptothecins, calicheamicin, esperamicin, ene-diynes, duocarmycin A, duocarmycin SA, calicheamicin, camptothecins, maytansinoids (including DM1), monomethylauristatin E (MMAE), monomethylauristatin F (MMAF) and maytansinoids (DM4) and their analogs.

[0172] Toxins may be used as antibody-toxin conjugates, including bacterial toxins such as diphtheria toxin, plant toxins such as ricin, small molecule toxins such as geldanamycin (Mandler et al (2000) J. Nat. Cancer Inst. 92(19):1573-1581; Mandler et al (2000) Bioorganic & Med. Chem. Letters 10:1025-1028; Mandler et al (2002) Bioconjugate Chem. 13:786-791), maytansinoids (EP 1391213; Liu et al., (1996) Proc. Natl. Acad. Sci. USA 93:8618-8623) and calicheamicin (Lode et al (1998) Cancer Res. 58:2928; Hinman et al (1993) Cancer Res. 53:3336-3342. Toxins may exert their cytotoxic and cytostatic effects by mechanisms including tubulin binding, DNA binding, or topoisomerase inhibition.

[0173] Conjugates of antibodies (or other antigen binding proteins) and one or more small molecule toxins such as maytansinoids, dolastatins, auristatins, trichothecins, calicheamicins, and CC1065, and derivatives of these toxins that have toxic activity are contemplated.

[0174] Maytansine compounds suitable for use as maytansinoid drug moieties are well known in the art and can be isolated from natural sources according to known methods produced using genetic engineering techniques (see Yu et al (2002) PNAS 99:7968-7973), or synthetically prepared maytansinol and maytansinol analogues according to known methods. As described below, the drugs can be modified by the incorporation of functionally active groups, such as thiol or amine groups for conjugation to antibodies.

[0175] Exemplary maytansinoid drug moieties include those with modified aromatic rings, such as C-19-dechloro (U.S. Pat. No. 4,256,746) (prepared by lithium aluminum hydride reduction of ansamitocin P2); C-20-hydroxy (or C-20-demethyl) + / - C-19-dechloro (U.S. Pat. Nos. 4,361,650 and 4,307,016) (prepared by demethylation using Streptomyces or Actinomyces or dechlorination using LAH); and C-20-demethoxy, C-20-acyloxy (--OCOR), + / -dechloro (U.S. Pat. No. 4,294,757) (prepared by acylation using acyl chlorides), as well as those with modifications at other positions.

[0176] Exemplary maytansinoid drug moieties include C-9-SH (U.S. Pat. No. 4,424,219) (prepared by reduction of maytansinol with H2S or P2S5); C-14-alkoxymethyl (demethoxy / CH2OR) (U.S. Pat. No. 4,331,598); C-14-hydroxymethyl or acyloxymethyl (CH2OH or CHOAc) (U.S. Pat. No. 4,450,254) (prepared from Nocardia); C-15-hydroxy / acyloxy (U.S. Pat. No. 4,364,866) (prepared by conversion of maytansinol by Streptomyces); C-15-methoxy (U.S. Pat. Nos. 4,313,946 and 4,315,929) (prepared from Trewia nudolflora nudlflora); C-18-N-demethyl (U.S. Pat. Nos. 4,362,663 and 4,322,348) (prepared by demethylation of maytansinol with Streptomyces); and 4,5-deoxy (U.S. Pat. No. 4,371,533) (prepared by titanium trichloride / LAH reduction of maytansinol).

[0177] Particularly useful are DM1 (disclosed in U.S. Patent No. 5,208,020, which is incorporated by reference) and DM4 (disclosed in U.S. Patent No. 7,276,497, which is incorporated by reference). See also U.S. Patent Nos. 5,416,064; 6,441,163; 7,303,749; 7,368,565; and 7,601,354; WO 01 / 24763, WO 02 / 098883, WO 02 / 16368, and WO 04 / 1033272; and U.S. Patent Application No. 12 / 631,508, all of which are expressly incorporated by reference in their entirety.

[0178] ADCs containing maytansinoids, methods of making them and their therapeutic uses are disclosed, for example, in U.S. Patent Nos. 5,208,020; 5,416,064; 6,441,163 and European Patent No. 0425235B1, the disclosures of which are expressly incorporated herein by reference. Liu et al., Proc. Natl. Acad. Sci. USA 93:8618-8623 (1996) described ADCs containing DM1 engineered with a maytansinoid linked to the monoclonal antibody C242, which is directed against human colorectal cancer.

[0179] Chari et al., Cancer Research 52:127-131 (1992) described ADCs in which maytansinoids were conjugated via a disulfide linker to the murine antibody A7, which binds to an antigen on a human colon cancer cell line, or to another murine monoclonal antibody, TA.1, which binds to the HER-2 / neu oncogene. The drug conjugates achieved a degree of cytotoxicity similar to that of the free maytansinoid drug, which could be increased by increasing the number of maytansinoid molecules per antibody molecule.

[0180] In some embodiments, the ADC comprises a dolastatin or dolastatin peptidic analog or derivative or an auristatin (U.S. Pat. Nos. 5,635,483 and 5,780,588). Dolastatins and auristatins have been shown to interfere with microtubule dynamics, GTP hydrolysis, and nuclear and cell division (Woyke et al (2001) Antimicrob. Agents and Chemother. 45(12):3580-3584), and have anticancer (U.S. Pat. No. 5,663,149) and antifungal activity (Pettit et al (1998) Antimicrob. Agents Chemother. 42:2961-2965). The dolastatin or auristatin drug moiety may be attached to the antibody through the N (amino) terminus or the C (carboxyl) terminus of the peptidic drug moiety (WO 02 / 088172). In various embodiments, the heterodimeric antibody is part of a therapeutic regimen that also includes administration of eribulin.

[0181] Exemplary auristatin embodiments include monomethyl auristatin drug moieties DE and DF linked to the N-terminus as disclosed in Senter et al, Proceedings of the American Association for Cancer Research, Volume 45, Abstract Number 623, filed March 28, 2004, and described in U.S. Patent Application Publication No. 2005 / 0238648, the disclosure of which is expressly incorporated by reference in its entirety. An exemplary auristatin embodiment is MMAE (see U.S. Patent Application Publication No. 6,884,869, which is expressly incorporated by reference in its entirety). Another exemplary auristatin embodiment is MMAF (see U.S. Patent Application Publication No. 2005 / 0238649 and U.S. Patent Nos. 5,767,237 and 6,124,431, which are expressly incorporated by reference in their entirety).

[0182] Typically, peptide-based drug moieties can be prepared by forming a peptide bond between two or more amino acids and / or peptide fragments. Such peptide bonds can be made, for example, according to solution phase synthesis methods well known in the field of peptide chemistry (see E. Schroder and K. Lubke, "The Peptides", volume 1, pp76-136, 1965, Academic Press). Auristatin / dolastatin drug moieties may be prepared according to the methods of U.S. Pat. Nos. 5,635,483 and 5,780,588; Pettit et al (1989) J. Am. Chem. Soc. 111:5463-5465; Pettit et al (1998) Anti-Cancer Drug Design 13:243-277; Pettit, GR, et al. Synthesis, 1996, 719-725; Pettit et al (1996) J. Chem. Soc. Perkin Trans. 1 5:859-863; and Doronina (2003) Nat Biotechnol 21(7):778-784.

[0183] In other embodiments, the ADC comprises one or more calicheamicin molecules. For example, Mylotarg is the first commercially available ADC drug and utilizes calicheamicin gamma 1 as the payload (see U.S. Patent No. 4,970,198, which is incorporated by reference in its entirety). Additional calicheamicin derivatives are described in U.S. Patent Nos. 5,264,586, 5,384,412, 5,550,246, 5,739,116, 5,773,001, 5,767,285, and 5,877,296, all of which are expressly incorporated by reference. The calicheamicin family of antibiotics can generate double-stranded DNA breaks at sub-picomolar concentrations. For the preparation of conjugates of the calicheamicin family, see U.S. Pat. Nos. 5,712,374, 5,714,586, 5,739,116, 5,767,285, 5,770,701, 5,770,710, 5,773,001, and 5,877,296 (all to American Cyanamid Company). Structural analogs of calicheamicin that may be used include, but are not limited to, gamma 1I, alpha 2I, alpha 2I, N-acetyl-gamma 1I, PSAG, and theta 11 (Hinman et al., Cancer Research 53:3336-3342 (1993); Lode et al., Cancer Research 58:2925-2928 (1998) and the aforementioned U.S. patents to American Cyanamid). Another antitumor drug to which the antibody may be conjugated is QFA, an antifolate. Both calicheamicin and QFA have intracellular sites of action and do not readily cross the plasma membrane. Thus, cellular uptake of these agents via antibody-mediated internalization greatly enhances their cytotoxic effects.

[0184] CC-1065 (see U.S. Pat. No. 4,169,888, incorporated by reference) and the duocarmycins are members of a family of antitumor antibiotics utilized in ADCs. These antibiotics appear to function by sequence-selectively alkylating DNA at the N3 of adenine in the minor groove, initiating a cascade of events that lead to apoptosis. Important members of the duocarmycin family include duocarmycin A (U.S. Pat. No. 4,923,990, incorporated by reference) and duocarmycin SA (U.S. Pat. No. 5,101,038, incorporated by reference), as well as U.S. Pat. Nos. 7,517,903; 7,691,962; 5,101,038; 5,641,780; 5,187,186; 5,070,092; 5,070,092; 5,641,780, all of which are expressly incorporated by reference. Nos. 5,101,038; 5,084,468; 5,475,092; 5,585,499; 5,703,080; 6,989,452; 7,087,600; 7,129,261; 7,498,302; 7,507,420; and 5,846,545; as well as numerous analogues as described in WO 2007 / 089149 and WO 2009 / 017394 A1.

[0185] Other cytotoxic agents Other anti-tumor drugs that can be conjugated to the antigen binding proteins include BCNU, streptozoicin, vincristine and 5-fluorouracil, the family of drugs known collectively as LL-E33288 conjugates described in U.S. Pat. Nos. 5,053,394 and 5,770,710, and esperamicin (U.S. Pat. No. 5,877,296).

[0186] Enzymatically active toxins and fragments thereof that may be used include diphtheria A chain, nonbinding active fragments of diphtheria toxin, exotoxin A chain (from Pseudomonas aeruginosa), ricin A chain, abrin A chain, modeccin A chain, alpha-sarcin, Aleurites fordii proteins, dianthin proteins, Phytolaca americana proteins (PAPI, PAPII and PAP-S), Momordica charantia inhibitor, curcin, crotin, Sapaonaria officinalis inhibitor, gelonin, mitogenin, restrictocin, phenomycin, enomycin and the trichothecenes. See, e.g., WO 93 / 21232.

[0187] The present disclosure further contemplates ADCs formed between an antigen binding protein and a compound with nucleolytic activity (e.g., a ribonuclease or deoxyribonuclease; a DNA endonuclease such as DNase).

[0188] For selective destruction of tumors, antigen-binding proteins (e.g., antibodies or heterodimeric antibodies) may contain highly radioactive atoms. A variety of radioisotopes are available for the production of radioconjugates. Examples include At211, I131, I125, Y90, Re186, Re188, Sm153, Bi212, P32, Pb212 and Lu radioisotopes.

[0189] Radioactive or other labels can be incorporated in the conjugate in known manner. For example, peptides can be biosynthesized or synthesized by chemical amino acid synthesis using suitable amino acid precursors, for example, containing fluorine-19 instead of hydrogen. Labels such as Tc99m or I123, Re186, Re188 and In111 can be attached via cysteine ​​residues in the peptide. Yttrium-90 can be attached via lysine residues. The iodogen method (Fraker et al (1978) Biochem. Biophys. Res. Commun. 80:49-57) can be used to incorporate iodine-123. "Monoclonal Antibodies in Immunoscintigraphy" (Chatal, CRC Press 1989) describes other methods in detail.

[0190] In some cases, separation, purification and characterization of homogeneous ADCs where p is a constant value from ADCs with other drug loadings can be achieved by means such as reverse-phase HPLC or electrophoresis. In exemplary embodiments, p is 2, 3, 4, 5, 6, 7 or 8, or fractions thereof.

[0191] It will be appreciated that chemical modifications can also be made to a desired compound to make reactions of that compound more favorable for the purpose of making the conjugates of the invention. For example, functional groups such as amines, hydroxyls, or sulfhydryls can be added to a drug at positions that have minimal or acceptable effect on the activity or other properties of the drug.

[0192] Linker Units Typically, antigen binding protein-drug conjugates include a linker unit between the drug unit and the antigen binding protein unit. In some embodiments, the linker is cleavable under intracellular or extracellular conditions, such that cleavage of the linker releases the drug unit from the antigen binding protein in the appropriate environment. For example, solid tumors that secrete certain proteases may serve as targets for a cleavable linker; in other embodiments, it is the intracellular proteases that are utilized. In still other embodiments, the linker unit is not cleavable and the drug is released, for example, by antibody degradation in the lysosome.

[0193] In some embodiments, the linker is cleavable by a cleaving agent present in the intracellular environment (e.g., within a lysosome or endosome or a vesicular intracellular structure). The linker can be, for example, a peptidyl linker that is cleaved by an intracellular peptidase or protease enzyme, including, but not limited to, a lysosomal or endosomal protease. In some embodiments, the peptidyl linker is at least 2 amino acids long or at least 3 or more amino acids long.

[0194] Cleavage agents may include, but are not limited to, cathepsin B and D and plasmin, all of which are known to hydrolyze dipeptide drug derivatives resulting in the release of active drug inside target cells (see, e.g., Dubowchik and Walker, 1999, Pharm. Therapeutics 83:67-123). Peptidyl linkers cleavable by enzymes present in CD38-expressing cells. For example, peptidyl linkers cleavable by cathepsin B, a thiol-dependent protease highly expressed in cancerous tissues, may be used (e.g., Phe-Leu or Gly-Phe-Leu-Gly linkers (SEQ ID NO: 181)). Other examples of such linkers are described in U.S. Patent No. 6,214,345, which is incorporated by reference in its entirety.

[0195] In some embodiments, the peptidyl linker cleavable by an intracellular protease is a Val-Cit linker or a Phe-Lys linker (see, for example, U.S. Pat. No. 6,214,345, which describes the synthesis of doxorubicin with a val-cit linker).

[0196] In other embodiments, the cleavable linker is pH-sensitive, i.e., sensitive to hydrolysis at a certain pH value. Typically, pH-sensitive linkers can be hydrolyzed under acidic conditions. For example, acid-labile linkers (e.g., hydrazones, semicarbazones, thiosemicarbazones, cis-aconitic amides, orthoesters, acetals, ketals, etc.) that are hydrolyzable in lysosomes can be used. (See, for example, U.S. Pat. Nos. 5,122,368; 5,824,805; and 5,622,929; Dubowchik and Walker, 1999, Pharm. Therapeutics 83:67-123; Neville et al., 1989, Biol. Chem. 264:14653-14661). Such linkers are relatively stable under neutral pH conditions, such as those in blood, but are unstable below pH 5.5 or 5.0, which is the approximate pH of lysosomes. In certain embodiments, the hydrolyzable linker is a thioether linker, such as a thioether attached to a therapeutic agent via an acylhydrazone bond (see, eg, US Pat. No. 5,622,929).

[0197] In yet other embodiments, the linker is cleavable under reducing conditions (e.g., a disulfide linker). A variety of disulfide linkers are known in the art, including, for example, SATA (N-succinimidyl-5-acetylthioacetate), SPDP (N-succinimidyl-3-(2-pyridyldithio)propionate), SPDB (N-succinimidyl-3-(2-pyridyldithio)butyrate) and SMPT (N-succinimidyl-oxycarbonyl-alpha-methyl-alpha-(2-pyridyl-dithio)toluene)-and those that can be formed using SPDB and SMPT. See, e.g., Thorpe et al., 1987, Cancer Res. 47:5924-5931; Wawrzynczak et al., In Immunoconjugates: Antibody Conjugates in Radioimagery and Therapy of Cancer (C. W. Vogel ed., Oxford U. Press, 1987). See also U.S. Patent No. 4,880,935.

[0198] In other embodiments, the linker is a malonate linker (Johnson et al., 1995, Anticancer Res. 15:1387-93), a maleimidobenzoyl linker (Lau et al., 1995, Bioorg-Med-Chem. 3(10):1299-1304), or a 3'-N-amide analog (Lau et al., 1995, Bioorg-Med-Chem. 3(10):1305-12).

[0199] In yet other embodiments, the linker unit is not cleavable and the drug is released by antibody degradation, see, e.g., U.S. Patent Application Publication No. 2005 / 0238649, which is incorporated by reference herein in its entirety.

[0200] In many embodiments, the linker is self-immolative. As used herein, the term "self-immolative spacer" refers to a bifunctional chemical moiety that can covalently link two spaced apart chemical moieties together into a stable three-part molecule. It will spontaneously separate from the second chemical moiety when its bond to the first moiety is broken. See, for example, WO 2007059404 A2, WO 06110476 A2, WO 05112919 A2, WO 2010 / 062171, WO 09 / 017394, WO 07 / 089149, WO 07 / 018431, WO 04 / 043493 and WO 02 / 083180 concerning drug-cleavable substrate conjugates, where the drug and cleavable substrate are optionally attached via a self-immolative linker, and all are expressly incorporated by reference.

[0201] In many cases, the linker is substantially insensitive to the extracellular environment, i.e., about 20%, 15%, 10%, 5%, 3%, or less than about 1% of the linkers in a sample of the antigen binding protein-drug conjugate are cleaved when the antigen binding protein-drug conjugate is present in an extracellular environment (e.g., in plasma). Whether a linker is substantially insensitive to the extracellular environment can be determined, for example, by incubating the antigen binding protein-drug conjugate compound with plasma for a predetermined period of time (e.g., 2, 4, 8, 16, or 24 hours) and subsequently quantifying the amount of free drug present in the plasma.

[0202] In other, non-mutually exclusive, embodiments, the linker promotes cellular internalization. In certain embodiments, the linker promotes cellular internalization when conjugated to a therapeutic agent (i.e., in the context of the linker-therapeutic agent portion of an ADC as described herein). In yet other embodiments, the linker promotes cellular internalization when conjugated to both an auristatin compound and an antigen binding protein of the disclosure.

[0203] Various exemplary linkers that can be used with the present compositions and methods are described in International Publication No. WO 2004-010957 and U.S. Patent Application Publication Nos. 2006 / 0074008, 20050238649, and 2006 / 0024317, each of which is incorporated by reference in its entirety.

[0204] It will be understood that the above therapeutic agents may be administered separately, i.e., may not be conjugated to an antigen binding protein in various embodiments.

[0205] Drug burden Drug loading is represented by p, which is the average number of drug moieties per antigen-binding protein in the molecule. Drug loading ("p") can be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or more moieties (D) per antigen-binding protein, although often the average number is a fraction or decimal. In general, drug loadings of 1-4 are often useful, and 1-2 are also useful. The ADCs of the present disclosure include a collection of antigen-binding proteins conjugated with a range of 1-20 drug moieties. The average number of drug moieties per antigen-binding protein in preparations of ADCs from conjugation reactions can be characterized by conventional means, such as mass spectrometry and ELISA assays.

[0206] The quantitative distribution of the ADC in terms of p may also be determined. In some cases, separation, purification, and characterization of homogeneous ADCs with a constant p from ADCs with other drug loads may be achieved by means such as electrophoresis.

[0207] For some ADCs, p may be limited by the number of binding sites on the antigen-binding protein. For example, if the linkage is a cysteine ​​thiol, as in the exemplary embodiment above, the antigen-binding protein may have only one or more cysteine ​​thiol groups, or may have only one or more sufficiently reactive thiol groups to which a linker may be attached. In certain embodiments, higher drug loading, e.g., p>5, may cause aggregation, insolubility, toxicity, or loss of cell permeability of certain antibody-drug conjugates. In certain embodiments, the drug loading for the ADCs of the present disclosure ranges from 1 to about 8; from about 2 to about 6; from about 3 to about 5; from about 3 to about 4; from about 3.1 to about 3.9; from about 3.2 to about 3.8; from about 3.2 to about 3.7; from about 3.2 to about 3.6; from about 3.3 to about 3.8; or from about 3.3 to about 3.7. Indeed, for certain ADCs, the optimal ratio of drug moieties per antigen-binding protein may be less than 8 and may be from about 2 to about 5. See US Patent Application Publication No. 2005 / 0238649A1, which is incorporated herein by reference in its entirety.

[0208] In certain embodiments, less than the theoretical maximum of the drug moiety is conjugated to the antigen-binding protein during the conjugation reaction. The antigen-binding protein may contain lysine residues that do not react with, for example, the drug-linker intermediate or linker reagent as described below. In general, antibodies do not contain many free and reactive cysteine ​​thiol groups that can be linked to drug moieties; in fact, most cysteine ​​thiol residues in antibodies exist as disulfide bridges. In certain embodiments, the antibody may be reacted under partial or total reducing conditions with a reducing agent such as dithiothreitol (DTT) or tricarbonylethylphosphine (TCEP) to generate reactive cysteine ​​thiol groups. In certain embodiments, the antibody is subjected to denaturing conditions to expose reactive nucleophilic groups such as lysine or cysteine.

[0209] The loading (drug / antigen binding protein ratio) of the ADC can be controlled in a variety of ways, for example, by (i) limiting the molar excess of drug-linker intermediate or linker reagent relative to antibody, (ii) limiting the conjugation reaction time or temperature, (iii) reactive conditions for partial or limited cysteine ​​thiol modification, (iv) engineering the amino acid sequence of the antigen binding protein by incorporation techniques such that the number and position of cysteine ​​residues are modified to control the number and / or position of linker-drug linkages (such as, for example, ThioMab or ThioFab prepared as disclosed in WO 2006 / 034488, which is incorporated herein by reference in its entirety).

[0210] It should be understood that when two or more nucleophilic groups react with a drug-linker intermediate or linker reagent followed by a drug moiety reagent, the resulting product is a mixture of ADC compounds with a distribution of one or more drug moieties attached to antigen binding proteins. The average number of drugs per antigen binding protein can be calculated from the mixture by a double ELISA antibody assay specific for the antigen binding protein and specific for the drug. Individual ADC molecules can be identified in the mixture by mass spectrometry and separated by HPLC, for example, hydrophobic interaction chromatography.

[0211] In some embodiments, homogenous ADCs having a single loading value can be isolated from the conjugation mixture by electrophoresis or chromatography.

[0212] composition Formulations for use according to the present disclosure are prepared for storage by mixing the antigen binding protein having the desired purity in the form of a lyophilized formulation or aqueous solution with optional pharma- ceutically acceptable carriers, excipients or stabilizers (Remington's Pharmaceutical Sciences 16th edition, Osol, A. Ed.

[1980] ). Compositions of the present disclosure are preferably sterile. Acceptable carriers, excipients or stabilizers are non-toxic to recipients at the dosages and concentrations employed and include buffers such as phosphates, citrates and other organic acids; antioxidants including ascorbic acid and methionine; preservatives (such as octadecyldimethylbenzylammonium chloride; hexamethonium chloride; benzalkonium chloride; benzethonium chloride; phenol, butyl or benzyl alcohol; alkyl parabens such as methyl or propyl paraben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight (less than about 10 residues) polypeptides; serum albumin, These include proteins such as gelatin or immunoglobulins; hydrophilic polymers (such as polyvinylpyrrolidone); amino acids such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates including glucose, mannose, or dextrins; chelating agents such as EDTA; sugars such as sucrose, mannitol, trehalose, or sorbitol; salt-forming counterions such as sodium; metal complexes (e.g., Zn-protein complexes); and / or non-ionic surfactants such as TWEEN®, PLURONICS®, or polyethylene glycol (PEG).

[0213] The formulation may also contain two or more active compounds as necessary for the particular indication being treated, preferably with complementary activities that do not adversely affect each other. For example, it may be desirable to provide antigen binding proteins with other specificities. Alternatively or in addition, the composition may include a cytotoxic agent, such as any of the drugs mentioned herein, a cytokine, a growth inhibitor and / or a small molecule antagonist. Such molecules are suitably present in combination in an amount effective for the intended purpose.

[0214] The active ingredient may also be incorporated into microcapsules (e.g., hydroxymethylcellulose or gelatin microcapsules and poly(methyl methacrylate) microcapsules, respectively), colloidal drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles and nanocapsules) or macroemulsions prepared, for example, by coacervation techniques or by interfacial polymerization. Such techniques are disclosed in Remington's Pharmaceutical Sciences 16th edition, Osol, A. Ed. (1980).

[0215] Sustained release formulations can be prepared. Suitable examples of sustained release formulations include semipermeable matrices of solid hydrophobic polymers containing antigen-binding proteins, which matrices are in the form of shaped articles, such as films or microcapsules. Examples of sustained release matrices include polyesters, hydrogels (e.g., poly(2-hydroxyethyl-methacrylate) or poly(vinyl alcohol)), polylactic acid (U.S. Pat. No. 3,773,919), copolymers of L-glutamic acid and gamma-ethyl-L-glutamate, non-degradable ethylene-vinyl acetate copolymers, degradable lactic acid-glycolic acid copolymers such as LUPRON DEPOT™ (injectable microspheres composed of lactic acid-glycolic acid copolymer and leuprolide acetate), and poly-D-(−)-3-hydroxybutyric acid. Polymers such as ethylene-vinyl acetate and lactic acid-glycolic acid can release molecules for more than 100 days, while certain hydrogels release proteins for shorter periods of time.

[0216] If encapsulated antibodies remain in the body for a long time, they may denature or aggregate as a result of exposure to moisture at 37°C, resulting in loss of biological activity and possible changes in immunogenicity. Rational strategies for stabilization can be devised depending on the mechanism involved. For example, if the aggregation mechanism is found to be the formation of intermolecular SS bonds by thio-disulfide exchange, stabilization can be achieved by modification of sulfhydryl residues, lyophilization of acidic solutions, control of moisture content, use of appropriate additives, and development of specific polymer matrix compositions.

[0217] Mode of administration The antigen binding protein and optionally a combination therapy such as a chemotherapeutic agent or another antibody therapy (e.g., an anti-PD-1 antibody) are administered to the subject according to a clinically accepted method, such as intravenous, intramuscular, intraperitoneal, subcutaneous, intraarticular, intralesional, intrasynovial, intrathecal, oral, topical, intratumoral, afferent lymphatic, or inhalation routes. Intravenous or subcutaneous administration of the antigen binding protein is preferred. Bolus injection and continuous infusion are contemplated, as is local administration, for example, at the site of disease or injury. The use of the antigen binding protein (optionally with another therapeutic agent) in ex vivo procedures is also contemplated. For example, the patient's blood or other bodily fluids may be contacted with the antigen binding protein ex vivo and may optionally be administered. The antigen binding protein may be bound to a suitable insoluble matrix or solid support material.

[0218] How to use The present disclosure provides a method of treating a subject in need thereof, comprising administering to the subject an antigen binding protein (e.g., an antibody or heterodimeric antibody) described herein. In various embodiments, the present disclosure provides a method of treating a cancer (such as prostate cancer or Ewing's sarcoma) in a subject in need thereof, comprising administering to the subject an antigen binding protein (e.g., an antibody or heterodimeric antibody) described herein. The present disclosure further provides the use of an antigen binding protein (e.g., an antibody or heterodimeric antibody) of the present disclosure for treating a cancer (e.g., prostate cancer or Ewing's sarcoma) in a subject in need thereof. The cancer is preferably a cancer with increased expression of STEAP1 (e.g., greater than 10,000 STEAP1 / cell). Examples of cancer include, but are not limited to, cancer of the prostate, breast, pancreas, bladder, gastrointestinal tract, testis, ovary, cervix, as well as sarcoma (Ewing's sarcoma) and melanoma.

[0219] The method of treating a subject described herein is intended to bring about an improvement in disease or condition and / or an improvement in symptoms associated with the disease or condition.For example, a therapeutic response may refer to one or more of the following improvements in disease: (1) a decrease in the number of regenerative cells; (2) an increase in regenerative cell death; (3) an inhibition of regenerative cell survival; (5) an inhibition (i.e., some slowing, preferably stopping) of tumor growth or the appearance of new lesions; (6) an increase in patient survival rate; and / or (7) some relief of one or more symptoms associated with disease or condition (e.g., in connection with prostate cancer, frequent urination, nocturia, hematuria, dysuria or bone pain; in connection with Ewing's sarcoma, pain, swelling or tenderness in the affected area).

[0220] The therapeutic response in any given disease or condition can be determined by standardized response criteria specific to that disease or condition.Tumor response can be evaluated using screening methods such as magnetic resonance imaging (MRI) scan, X-ray, computed tomography (CT) scan, positron emission tomography (PET) scan, bone scan, ultrasound, tumor biopsy collection, tumor cell count in circulation and / or tumor antigen (e.g., prostate specific antigen (PSA) and / or alpha fetoprotein (AFP)).In addition to these therapeutic responses, the subject receiving therapy can experience the beneficial effect of improvement in symptoms related to the disease.

[0221] The subject is a mammal, preferably a human, optionally a male human. In relation to cancer, the subject may be diagnosed with any stage of the disease (i.e., stage I, stage II, stage III or stage IV prostate cancer) or may be diagnosed at risk of developing a cancer that has not yet been clinically confirmed.

[0222] With respect to prostate cancer, the subject may experience a reduction in symptoms associated with prostate cancer (such as those described herein), a reduction in tumor size, a reduction in the level of prostate cancer markers, a reduction in the rate of new lesion appearance, etc. In various embodiments, the method of the present disclosure further comprises monitoring the treatment in the subject. Improvements in the health of the subject are contemplated (e.g., no clinically detectable disease, any reduction (such as at least about 50% reduction) of measurable tumor burden (i.e. the number of malignant cells present in the subject or the measured volume of tumor mass) in the absence of new lesions, reduction in pain, improvement in urination).

[0223] Treatment according to the present disclosure includes a "therapeutically effective amount" of the pharmaceutical used. A "therapeutically effective amount" refers to an amount effective at dosages and for periods of time necessary to achieve the desired therapeutic result. A therapeutically effective amount may vary depending on factors such as the disease state, age, sex, and weight of the individual, and the ability of the pharmaceutical to elicit a desired response in the individual. A therapeutically effective amount is an amount in which any toxic or adverse effects are outweighed by the therapeutically beneficial effects. A "therapeutically effective amount" for tumor therapy may also be measured by its ability to stabilize disease progression. An exemplary non-limiting range for a therapeutically effective amount of an antigen binding protein of the present disclosure is about 0.1-100 mg / kg. Parenteral compositions may be formulated in dosage unit forms for ease of administration and uniformity of dosage. As used herein, dosage unit forms refer to physically discrete units adapted as unitary dosages for the subjects to be treated; each unit contains a predetermined amount of a therapeutic agent calculated to produce the desired biological effect in association with the required pharmaceutical carrier.

[0224] In some embodiments, the antigen binding proteins (e.g., monospecific or heterodimeric antibodies) are used in combination with one or more additional therapeutic agents, such as chemotherapeutic or immunotherapeutic agents. The additional therapeutic agents may be administered sequentially (within minutes, hours, days, or weeks of each other) or in parallel. They may also be administered to the patient in a single premixed composition.

[0225] Non-limiting examples of DNA damaging chemotherapeutic agents include topoisomerase I inhibitors (e.g., irinotecan, topotecan, camptothecin and its analogs or metabolites, and doxorubicin); topoisomerase II inhibitors (e.g., etoposide, teniposide, and daunorubicin); alkylating agents (e.g., melphalan, chlorambucil, busulfan, thiotepa, ifosfamide, carmustine, lomustine, semustine, streptozocin, decarbazone, DNA interferents (e.g., cisplatin, oxaliplatin, epirubicin, and carboplatin); DNA interferents and free radical generators such as bleomycin; and nucleoside mimetics (e.g., 5-fluorouracil, capecitibin, gemcitabine, fludarabine, cytarabine, mercaptopurine, thioguanine, pentostatin, and hydroxyurea).

[0226] Chemotherapeutic agents that interfere with cell replication include, but are not limited to, paclitaxel, docetaxel, and related analogs; cabazitaxel; vincristine, vinblastine, and related analogs; thalidomide, lenalidomide, and related analogs (e.g., CC-5013 and CC-4047); protein tyrosine kinase inhibitors (e.g., imatinib mesylate and gefitinib); proteasome inhibitors (e.g., bortezomib); NF-κB inhibitors, including inhibitors of IκB kinase; antibodies that bind to proteins overexpressed in cancer, thereby downregulating cell replication (e.g., trastuzumab, rituximab, cetuximab, and bevacizumab); and other inhibitors of proteins or enzymes known to be upregulated, overexpressed, or activated in cancer, whose inhibition downregulates cell replication.

[0227] In some embodiments, antigen binding proteins (e.g., monospecific or heterodimeric antibodies) of the present disclosure may be used prior to, concomitantly with, or following treatment with docetaxol. In various aspects, the antigen binding proteins (e.g., monospecific or heterodimeric antibodies) are administered as part of a treatment regimen that includes surgery and / or radiation (e.g., external beam or brachytherapy).

[0228] In various embodiments, the antigen binding protein (e.g., a monospecific antibody or a heterodimeric antibody) is provided as part of a treatment regimen that also includes administration of hormone therapy (e.g., androgen deprivation therapy, such as agents that block the release or production of luteinizing hormone releasing hormone (e.g., leuprolide, goserelin, triptorelin, or degarelix), antiandrogens (e.g., bicalutamide, flutamide, or nilutamide), ketoconazole, abiraterone acetate, enzalutamide).

[0229] In various aspects, the antigen binding protein (e.g., a monospecific antibody or a heterodimeric antibody) is provided as part of a treatment regimen that also includes administration of another immunotherapy (e.g., sipuleucel-T, bevacizumab, atezolizumab, avelumab, ipilimumab, tremelimumab, AM-224, MDX-1105, eftilagimod alpha (IMP321) or enoblituzumab (MGA271)). In this regard, the methods optionally include administration of another antigen binding protein that targets a different antigen, such as a cancer-associated antigen or an antigen associated with an immune response. For example, in various embodiments, an anti-STEAP1 antigen binding protein is administered to the subject along with a PD-1 targeting antigen binding protein (e.g., an antibody) that reduces, blocks, inhibits, suppresses or interferes with signaling resulting from the interaction of PD-1 with one or more of its binding partners, such as PD-L1 or PD-L2. In certain aspects, the PD-1 antigen binding protein inhibits the binding of PD-1 to PD-L1 and / or PD-L2. In one embodiment, the PD-1 antigen binding protein reduces negative costimulatory signals mediated by or through cell surface proteins expressed on T lymphocytes that mediate signaling through PD-1, such that dysfunctional T cells are less dysfunctional (e.g., to enhance effector responses to antigen recognition). Examples of anti-PD-1 antibodies include nivolumab (BMS-936558), pembrolizumab (MK-3475), BMS 936558, BMS-936559, TSR-042 (Tesaro), ePDR001 (Novartis), and pidilizumab (CT-011). Although the present disclosure refers to PD-1 antigen binding proteins, the disclosure also contemplates the use of other PD-1 binding antagonists that reduce, block, inhibit, suppress, or interfere with signaling resulting from the interaction of PD-1 with one or more of its binding partners, such as PD-L1 or PD-L2.

[0230] The disclosure provided herein with respect to anti-STEAP1 antigen binding proteins also applies to anti-PD-1 antigen binding proteins. For example, in various cases, the anti-PD-1 antigen binding protein is an antibody, such as a monoclonal IgG. The anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product is monovalent or bivalent. In exemplary embodiments, the anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product binds to human PD-1 having the amino acid sequence of SEQ ID NO: 187. In exemplary embodiments, the anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product binds to cynomolgus PD-1 having the amino acid sequence of SEQ ID NO: 188. In exemplary cases, the anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product binds to both human PD-1 and cynomolgus PD-1.

[0231] In exemplary embodiments, the binding strength of an anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product to PD-1 may be described in terms of KD. In exemplary aspects, the KD of an anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product provided herein is about 10 -1 M, about 10 -2 M, about 10 -3 M, about 10 -4 M, about 10 -5 M, about 10 -6 M, about 10 -7 M, about 10 -8 M, about 10 -9 M or less. In exemplary embodiments, the KD of the anti-PD-1 antibodies, antigen-binding antibody fragments thereof, or anti-PD-1 antibody protein products provided herein is micromolar, nanomolar, picomolar, or femtomolar. In exemplary embodiments, the KD of the anti-PD-1 antibodies, antigen-binding antibody fragments thereof, or anti-PD-1 antibody protein products provided herein is about 10 -4 ~10 -6 M or 10 -7 ~10 -9 M or 10 -10 ~10 -12 M or 10 -13 ~10-15 In exemplary embodiments, the anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product has a high affinity for human PD-1, cynomolgus PD-1, or both. In exemplary embodiments, the anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product has a KD for human PD-1 of less than 100 pM, optionally between about 1 pM and about 50 pM. In exemplary embodiments, the anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product has a KD for human PD-1 within about 1 pM to about 20 pM, or less than about 10 pM. In exemplary embodiments, the anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product has a KD for cynomolgus PD-1 of less than 100 pM, optionally between about 1 pM and about 75 pM. In exemplary aspects, the anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product has a KD for cynomolgus PD-1 that is within about 1 pM to about 20 pM, or less than about 10 pM.

[0232] In exemplary embodiments, the anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product inhibits at least 50% of the binding interaction between PD-1 and PD-L1 or PD-L2. In exemplary embodiments, the anti-PD-1 antibody, antigen-binding antibody fragment thereof, or anti-PD-1 antibody protein product exhibits at least about 50%, at least about 60%, or at least about 70% inhibition of the binding interaction between PD-1 and PD-L1 or PD-L2.

[0233] In exemplary cases, the anti-PD-1 antibody, its antigen-binding antibody fragment, or anti-PD-1 antibody protein product inhibits PD-1-mediated production of IL-2 by T cells in a mixed lymphocyte reaction (MLR). In exemplary embodiments, the IC50 of the anti-PD-1 antibody, its antigen-binding antibody fragment, or anti-PD-1 antibody protein product in an MLR is within about 0.1 nM to about 5 nM. In exemplary embodiments, the IC50 of the anti-PD-1 antibody, its antigen-binding antibody fragment, or anti-PD-1 antibody protein product in an MLR is less than 2 nM or less than 1 nM. In exemplary embodiments, the IC50 of the anti-PD-1 antibody, its antigen-binding antibody fragment, or anti-PD-1 antibody protein product in an MLR is about 0.5 nM to about 2 nM.

[0234] Methods for testing antibodies for the ability to bind to PD-1 are known in the art and include any suitable antibody-antigen binding assay, such as, for example, radioimmunoassay (RIA), ELISA, Western blot, immunoprecipitation, SPR, and competitive inhibition assays (see, e.g., Janeway et al., infra and U.S. Patent Application Publication No. 2002 / 0197266 and the section above regarding competitive assays). Other binding assays can be used to test the ability of an antibody to bind to PD-1, such as competitive binding or competition assays that test the ability of an antibody to compete with a second antibody for binding to an antigen or an epitope thereof. See, e.g., US Patent Publication No. 2014 / 0178905; Chand et al., Biologicals 46:168-171 (2017); Liu et al., Anal Biochem 525:89-91 (2017); and Goolia et al., J Vet Diagn Invest 29(2):250-253 (2017). Other methods of comparing two antibodies are also known in the art, including, for example, surface plasmon resonance (SPR). SPR can be used to determine the binding constant of an antibody and a second antibody, and the two binding constants can be compared. The present disclosure contemplates the use of anti-PD-1 antigen binding proteins in connection with the disclosed methods that compete with or cross-block the binding of any of the anti-PD-1 antibodies described herein to the PD-1 protein.

[0235] A representative method for characterizing human and cynomolgus PD-1 binding affinity is as follows: Antibodies are incubated in wells containing three-fold dilutions of soluble recombinant receptor human PD-1(1-170)-FLAG-His or cynomolgus PD-1(1-167)-FLAG-His. In both cases, a maximum PD-1 concentration of 30 nM may be selected. 300 seconds of association and 500 seconds of dissociation may be used as these parameters usually provide sufficient curvature for an accurate kinetic fit. Human / cynomolgus PD-1 binding affinity may be quantified on ForteBio Octet HTX and RED384 instruments. Standard Octet sample buffer may be used for sample dilution and binding baseline, binding and dissociation steps (e.g., 10 mM Tris, pH 7.5, 150 mM NaCl, 1 mM CaCl2, 0.10 mg.ml BSA, 0.13% (v / v) Triton X-100). ForteBio raw data can be processed using standard instrument data analysis software (v9 and v10) in the following manner: (a) two reference curves with immobilized target but no interaction (i.e. buffer only) are averaged and subtracted from the remaining sample curve in the same column; (b) the association and dissociation curves are isolated and aligned to the Y-axis; (c) the association and dissociation steps are aligned; (d) Savitzky-Golay filtering is performed to reduce signal noise, and (e) the set of association and dissociation curves obtained for each sample-target interaction is globally fitted with a single 1:1 binding model to determine measured association and dissociation rate constants, k and k; the equilibrium dissociation constant, K, is calculated as the ratio of the dissociation rate constant to the association rate constant (= k / ka).

[0236] In an exemplary embodiment, the anti-PD-1 antibody (or antigen-binding antibody fragment or antibody protein product thereof) comprises a heavy chain (HC) complementarity determining region 1 (vhCDR1) amino acid sequence set forth in SEQ ID NO:189, a HC CDR2 (vhCDR2) amino acid sequence set forth in SEQ ID NO:190, a HC CDR3 (vhCDR3) amino acid sequence set forth in SEQ ID NO:191, a light chain (LC) CDR1 (vlCDR1) amino acid sequence set forth in SEQ ID NO:192, a LC CDR2 (vlCDR2) amino acid sequence set forth in SEQ ID NO:193, and a LC CDR3 (vlCDR3) amino acid sequence set forth in SEQ ID NO:194. In exemplary embodiments, the anti-PD-1 antibody (or antigen-binding antibody fragment or antibody protein product thereof) comprises a heavy chain variable region (vh) comprising an amino acid sequence at least 90% identical (e.g., 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical) to the amino acid sequence of SEQ ID NO:195 and / or a light chain variable region (vl) comprising an amino acid sequence at least 90% identical (e.g., 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical) to the amino acid sequence of SEQ ID NO:196. In exemplary embodiments, the anti-PD-1 antibody (or antigen-binding antibody fragment or antibody protein product thereof) comprises a heavy chain comprising an amino acid sequence at least 90% identical (e.g., 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical) to the amino acid sequence of SEQ ID NO:197 and / or a light chain comprising an amino acid sequence at least 90% identical (e.g., 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% identical) to the amino acid sequence of SEQ ID NO:198.

[0237] In exemplary embodiments, the anti-STEAP1 constructs described herein are part of a therapeutic regimen that includes administration of a cytokine, lymphokine, growth factor, or other hematopoietic factor effective to inhibit tumor metastasis and / or have an antiproliferative effect on at least one cell population, including, but not limited to, M-CSF, GM-CSF, TNF, IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-11, IL-12, IL-13, IL-14, IL-15, IL-16, IL-17, IL-18, IFN, TNFα, TNF1, TNF2, G-CSF, Meg-CSF, GM-CSF, thrombopoietin, stem cell factor, and erythropoietin.Additional growth factors include, for example, angiogenin, bone morphogenetic protein-1, bone morphogenetic protein-2, bone morphogenetic protein-3, bone morphogenetic protein-4, bone morphogenetic protein-5, bone morphogenetic protein-6, bone morphogenetic protein-7, bone morphogenetic protein-8, bone morphogenetic protein-9, bone morphogenetic protein-10, bone morphogenetic protein-11, bone morphogenetic protein-12, bone morphogenetic protein-13, bone morphogenetic protein-14, bone morphogenetic protein-15, bone morphogenetic protein receptor IA, bone morphogenetic protein receptor IB, brain-derived neurotrophic factor, ciliary neurotrophic factor, ciliary neurotrophic factor receptor α, cytokine-induced neutrophil chemotactic factor 1, cytokine-induced neutrophil chemotactic factor 2α, cytokine-induced neutrophil chemotactic factor 2β, beta endothelial growth factor, endothelin 1, epithelial-derived neutrophil attractant, glial cell line-derived neurotrophic factor receptor α1, glial cell line-derived neurotrophic factor receptor α2, growth-associated protein, growth-associated protein α, growth-associated protein β, growth-associated protein γ, heparin-binding epidermal growth factor, Hepatocyte growth factor, hepatocyte growth factor receptor, insulin-like growth factor I, insulin-like growth factor receptor, insulin-like growth factor II, insulin-like growth factor binding protein, keratinocyte growth factor, leukemia inhibitory factor, leukemia inhibitory factor receptor alpha, nerve growth factor, nerve growth factor receptor, neurotrophin-3, neurotrophin-4, pre-B cell growth stimulatory factor, stem cell factor, stem cell factor receptor, transforming growth factor alpha, transforming growth factor beta, transforming growth factor beta 1, transforming growth factor beta 1.2, transforming growth factor beta 2, transforming growth factor beta 3, transforming growth factor beta 5, latent transforming growth factor beta 1, transforming growth factor beta binding protein I, transforming growth factor beta binding protein II, transforming growth factor beta binding protein III, tumor necrosis factor receptor type I, tumor necrosis factor receptor type II, urokinase-type plasminogen activator receptor, and chimeric proteins and biologically or immunologically active fragments thereof.In an exemplary embodiment, the anti-STEAP1 construct is administered as part of a therapeutic regimen that includes administration of an antibody specific for any one of the aforementioned cytokines, lymphokines, growth factors, or other hematopoietic factors.

[0238] The present disclosure contemplates the use of an anti-STEAP1 antigen binding protein or heterodimeric antibody in the preparation of a medicament for treating cancer in a subject in need thereof. Optionally, the medicament is for administering an effective amount of the anti-STEAP1 antigen binding protein or heterodimeric antibody in combination with an effective amount of an anti-PD-1 antigen binding protein (e.g., any of the anti-PD-1 antigen binding proteins described herein).

[0239] The present disclosure further contemplates an anti-STEAP1 antigen binding protein or heterodimeric antibody described herein for use in treating cancer in a subject in need thereof (i.e., in a method of treating a cancer, such as prostate cancer or Ewing's sarcoma, in a subject in need thereof). Optionally, the anti-STEAP1 antigen binding protein or heterodimeric antibody is administered together with an anti-PD1 antigen binding protein. "Administered together with" means that the anti-STEAP1 antigen binding protein or heterodimeric antibody is part of a treatment regimen that includes administration of the anti-PD1 antigen binding protein. Indeed, the anti-STEAP1 antigen binding protein (e.g., a heterodimeric antibody) may be used prior to, simultaneously with, or after treatment with the anti-PD1 antigen binding protein. Administration of the anti-STEAP1 antigen binding protein or heterodimeric antibody and the anti-PD1 antigen binding protein need not be simultaneous, although the present disclosure contemplates embodiments in which the components are included in the same pharmaceutical composition and administered together. The disclosure also provides methods of treatment in which the anti-STEAP1 antigen binding protein or heterodimeric antibody and the anti-PD1 antigen binding protein are present in separate pharmaceutical compositions that are administered in parallel or close in time. The anti-STEAP1 antigen binding protein or heterodimeric antibody and the anti-PD1 antigen binding protein may be administered sequentially (e.g., within minutes, hours, days or weeks of each other) in any order. Modes of administration are described above.

[0240] The anti-STEAP1 antigen binding proteins described herein can also be used, for example, in assays to detect the presence of STEAP1, either in vitro or in vivo. The antigen binding proteins can also be used to purify STEAP1, for example, by immunoaffinity chromatography.

[0241] Nucleic acids, vectors and host cells The present disclosure further provides nucleic acid compositions encoding the antigen binding proteins (e.g., monospecific or heterodimeric antibodies) described herein. Nucleic acids encoding components of the antigen binding proteins of the present disclosure may be incorporated into expression vectors, as known in the art and depending on the host cell used to produce the antigen binding protein. Examples of expression vectors include, but are not limited to, plasmids, viral vectors, non-episomal mammalian vectors, and other expression vectors. In general, the nucleic acid sequence encoding the desired polypeptide is operably linked to any number of regulatory elements (promoter, origin of replication, selection marker, ribosome binding site, inducer, etc.). Expression vectors may be extrachromosomal or integrative vectors.

[0242] The nucleic acids and / or expression vectors of the disclosure are optionally introduced into any number of different types of host cells as are well known in the art, including mammalian, bacterial, yeast, insect and / or fungal cells, with mammalian cells (e.g., CHO cells) being useful in many embodiments. In another aspect, the disclosure provides such host cells into which an expression vector encoding an antigen binding protein has been introduced. Examples of suitable mammalian host cell lines include monkey kidney cell line COS-7 (ATCC CRL 1651) (Gluzman et al., 1981, Cell 23:175), L cells, human embryonic kidney 293 cells or derivatives thereof (e.g., HEK293T, HEK293-EBNA), C127 cells, mouse embryonic fibroblast (3T3 cells) (ATCC CCL 163), Chinese hamster ovary (CHO) cells and derivatives thereof (e.g., CHO-K1, CHO pro-3), mouse melanoma cells (e.g., NS0, GS-NS0, Sp2 / 0), human cervical carcinoma cells (HeLa cells), baby hamster kidney (BHK) cells (ATCC CRL 1651), and the like. 10) Cell lines, human osteosarcoma epithelial cells U2-OS, adenocarcinoma human alveolar basal epithelial cells (A549), human fibrosarcoma cells (HT1080), mouse brain tumor cells (CAD), embryonal carcinoma cells (P19), mouse neuroblastoma cells (N2a), human breast cancer cells (MCF-7), retinoblastoma cells (Y79), human retinoblastoma cells (SO-Rb50), human hepatoma cells (Hep G2), mouse B myeloma cells (J558L) and African green monkey kidney cells (e.g., COS cells, VERO cells and their derivatives (CVI / EBNA cell line (ATCC CCL 11111) derived from the African green monkey kidney cell line CVI as described by McMahan et al., 1991, EMBO J. 10:2821). 70). The transformed cells can be cultured under conditions that promote expression of the antigen-binding protein, and the protein can be recovered by conventional purification procedures. One such purification procedure includes, for example, the use of affinity chromatography over a matrix to which all or a portion of STEAP1 (e.g., one or more extracellular loops) is bound.Antigen binding proteins contemplated for use herein include substantially homogeneous recombinant antigen binding proteins that are substantially free of contaminating endogenous material.

[0243] With respect to heterodimeric antibodies, in various embodiments, compositions are provided that include a nucleic acid encoding a first monomer, a nucleic acid encoding a second monomer, and a nucleic acid encoding a common light chain. The disclosure also provides nucleic acid constructs that encode portions of the monomers and the common light chain, such as an anti-STEAP1 Fab or antibody fragment comprising the six CDRs disclosed herein that bind to STEAP1, an anti-CD3 scFv, a variable light chain domain and / or a variable heavy chain domain that binds to STEAP1 and / or CD3, etc.

[0244] In some embodiments, the nucleic acid encoding each monomer and optionally the nucleic acid encoding the common light chain are contained in a single expression vector, typically under different or the same promoter control. In various embodiments, each of these two or three nucleic acids is contained on a different expression vector. As described in US Patent Publication No. 2016 / 0215063 (incorporated herein by reference in its entirety and in particular for the discussion of recombinant antibody production), different vector ratios can be used to drive heterodimer formation. Surprisingly, when an antibody construct comprises a first monomer:second monomer:light chain in a ratio of 1:1:2, no ratio necessarily produces the best results. See Figure 65 of US Patent Publication No. 2016 / 0215063, incorporated herein by reference. In various aspects, the disclosure provides a nucleic acid composition comprising: a) a first expression vector comprising a first nucleic acid encoding a first monomer; b) a second expression vector comprising a second nucleic acid encoding a second monomer; and c) a third expression vector comprising a third nucleic acid encoding a common light chain. In alternative embodiments, the third nucleic acid encoding the common light chain is present on the same expression vector as the first or second nucleic acid.

[0245] Heterodimeric antibodies are optionally produced by culturing host cells containing an expression vector. Once produced, an antibody purification step is usually performed, which includes an ion exchange chromatography step. As discussed herein, the two monomers have a pI that differs by at least 0.5, which may allow for separation by ion exchange chromatography or isoelectric focusing or other methods that are sensitive to isoelectric point. That is, it includes pI substitutions that change the isoelectric point (pI) of each monomer, so that each monomer has a different pI, and the heterodimer also has a different pI, thereby facilitating isoelectric purification of the heterodimer (e.g., anion exchange column, cation exchange column). These substitutions also facilitate the identification and monitoring of contaminating mAb homodimers after purification (e.g., IEF gel, cIEF and analytical IEX columns).

[0246] kit In some embodiments, the antigen binding proteins of the present disclosure are provided in a kit. In exemplary aspects, the kit includes the antigen binding protein as a unit dose (i.e., a discrete amount dispersed in a suitable carrier). In exemplary aspects, the kit includes several unit doses, for example, optionally a weekly or monthly supply of unit doses, each unit dose being individually packaged or otherwise separated from the other unit doses. In some embodiments, the components of the kit / unit dose are packaged with instructions for administration to a patient. In some embodiments, the kit includes one or more devices for administration to a patient, for example, a needle and a delivery device (such as a syringe). In some aspects, the antigen binding protein is prepackaged in a ready-to-use form, for example, a syringe, an intravenous bag, etc., although it is also contemplated that the antigen binding protein may be provided in a lyophilized form that requires reconstitution. In some aspects, the kit further includes other therapeutic or diagnostic agents, including any of those described herein, or a pharma- ceutically acceptable carrier (e.g., a solvent, buffer, diluent, etc.).

[0247] All references cited in this specification, including publications, patent applications, and patents, are herein incorporated by reference to the same extent as if each reference was individually and specifically indicated to be incorporated by reference and was set forth in its entirety herein.

[0248] The following examples are offered merely to illustrate the present disclosure and are not intended to limit its scope in any way. EXAMPLES

[0249] Example 1 This example describes the detection of STEAP1 on the surface of prostate cancer cells using a monospecific antibody as described herein.

[0250] Prostate cancer cells engineered to lose STEAP1 expression using a CRISPR construct directed against STEAP1 (C4-2B luc cells (FIG. 11A) or C4-2B luc STEAP1 KO Cells (FIG. 11B) were incubated with isotype control antibody or anti-STEAP1 mouse monoclonal antibody (Ab-Am) at a concentration of 10 μg / mL for 1 h at 4° C. Cell-bound Ab-Am was detected by flow cytometry after incubation with FITC-conjugated (FIG. 11A) or APC-conjugated anti-mouse IgG secondary antibody (FIG. 11B). FITC or APC fluorescence, identifying STEAP1-dependent signals, was plotted in histograms (filled grey histograms) and compared to isotype control (white histograms). Results are shown in FIGS. 11A and 11B.

[0251] Example 2 This example describes the characterization of anti-STEAP1 antibodies.

[0252] A panel of 22 anti-human STEAP1 mouse monoclonal antibodies was generated. Antibody A demonstrated improved flow cytometry binding properties (parental mAb binding LnCAP(+) / DU145(-) FACS shift (fold)) compared to others tested: Ab-A (60.1), Ab-B (3.6), Ab-C (2.4) and Ab-D (4.3).

[0253] To determine the regions of STEAP1 recognized by the antibodies of the present disclosure, chimeric constructs were made in which each of the three extracellular loops of STEAP1 was replaced with the corresponding regions of STEAP2 and expressed in 293 cells. Ab-A binds to STEAP1 but not STEAP2. Replacing extracellular loops 1 and 3 of STEAP1 with the corresponding loops from STEAP2 abrogated binding, whereas Ab-A binding to STEAP1 was abolished when extracellular loop 2 was replaced with its STEAP2 counterpart. Was not Ab-A appears to bind to STEAP1 outside of extracellular loop 2.

[0254] Heterodimeric antibodies comprising the STEAP-1-binding and anti-CD3-binding arms of Ab-A1, Ab-A2(N67Q) and Ab-B1 were prepared in the "XmAb" format as described, for example, in US Patent Application Publication No. 2016 / 0215063. These heterodimeric antibodies exhibited TDCC activity (pM): Ab-A1x (273.8), Ab-A2x (387.9) and Ab-B1x (128.7).

[0255] Example 3 This example illustrates the use of the heterodimer anti-STEAP1 / anti-CD3 (Xmab 2+1 ) to C4-2B cells (characterized by EC50) is compared with an anti-STEAP1 / anti-CD3 bispecific antibody (XmAb) with a different backbone.

[0256] Three anti-STEAP1 humanized antibodies in the "XmAb" format (Ab-A1, Ab-A2(N67Q) and Ab-B1) were made as described, for example, in US Patent Application Publication No. 2016 / 0215063 (incorporated herein by reference, particularly with respect to the discussion of the "bottle opener" format). The XmAb format involves a first heavy chain comprising an Fc domain bound to an anti-CD3 scFv; a second heavy chain comprising an Fc domain and a first variable heavy chain domain; and a light chain comprising a variable light chain domain and a constant light chain domain. The variable heavy chain domain and the variable light chain domain bind to STEAP1. The two anti-STEAP1 humanized antibodies are heterodimeric XmAbs. 2+1 Format (Ab-A1 XmAb 2+1 and Ab-B1 XmAb 2+1 ) Ab-A1 XmAb 2+1 and Ab-B1 XmAb 2+1 The CDR sequences are represented by SEQ ID NOs: 11 to 16 (Ab-A1 XmAb 2+1 ) and SEQ ID NOs: 30 to 35 (Ab-B1 XmAb 2+1 ) As used herein, Ab-A2(N67Q)XmAb 2+1 Ab-A1 XmAb with the N67Q modification 2+1 A mutant of Ab-A2(N67Q)XmAb was also generated. 2+1 The CDR sequences are set forth in SEQ ID NOs: 11 to 13, 14, 16 and 21. Antibody designations Ab-A2 and Ab-A2(N67Q)XmAb 2+1 are used interchangeably herein. The ability of the heterodimeric bispecific antibodies to bind to STEAP1 expressed on the surface of C4-2B prostate cancer cells was evaluated together with three mouse anti-STEAP1 antibodies in the non-humanized XmAb format (Ab-Mx1, Ab-Mx2 and Ab-Mx3).

[0257] C4-2B-Luc cells were incubated with increasing concentrations of Ab-A1 XmAb up to 5 μM. 2+1The cells were incubated with Ab-A1 Xmab, Ab-B1 Xmab, Ab-Mx1, Ab-Mx2 and Ab-Mx3 for 1 h at 4° C. Cell-bound antibodies were detected by flow cytometry after incubation with APC-conjugated anti-human IgG secondary antibody and the mean fluorescence intensity (MFI) of the APC channel at increasing concentrations of the corresponding antibodies tested, as shown in Table 3. 2+1 showed cell binding that was 65-fold lower than the binding EC50 of the same conjugate in XmAb format (i.e., Ab-A1 Xmab), indicating a very potent binding activity that exceeds that of the corresponding XmAb. 2+1 The format significantly enhanced binding of the Ab-A conjugate to STEAP1 expressed on prostate cancer cells.

[0258] [Table 2]

[0259] The experiments were performed using different formats (Ab-A (conventional, monospecific, non-humanized), Ab-A1 XmAb format, Ab-A1 Xmab 2+1 Format and Ab-A2(N67Q)Xmab 2+1 The study was repeated in Ab-A (with N67Q modification) as well as Ab-B XmAb formats. C4-2B-Luc cells were incubated with increasing concentrations of anti-STEAPI XmAb or XmAb up to 5 μM. 2+1 The cells were incubated with the molecules for 1 hour at 4° C. Cell-bound XmAb was detected by flow cytometry after incubation with APC-conjugated anti-human IgG secondary antibody, and the mean fluorescence intensity (MFI) of the APC channel at increasing concentrations of the corresponding anti-STEAPI XmAb molecules was shown. The results are shown in Figures 12A-12C and 13 and Table 4 below.

[0260] [Table 3]

[0261] All antibodies bound to STEAP1 regardless of the heterodimeric format. 2+1 The format demonstrated improved binding to STEAP1 compared to other antibody formats.

[0262] TDCC activity was also assessed using a method similar to that described above. All tested antibodies were XmAb 2+1 The following antibodies showed TDCC activity, demonstrating superior activity to the Xmab antibody: Ab-A XmAb (EC50=274pM, EC90=438pM), Ab-A1 Xmab (EC50=388pM, EC90=722pM), Ab-B1 Xmab (EC50=129pM, EC90=265pM), Ab-A1 Xmab 2+1 (EC50=6pM, EC90=11pM) and Ab-B1 Xmab 2+1 (EC50=19pM, EC90=43pM).

[0263] Example 5 This example shows anti-STEAP1 XmAb and XmAb 2+1 To characterize lysis of the human tumor cell line C4-2B luc by human T cells mediated by .

[0264] C4-2B luc prostate cancer cells were co-cultured with human pan-T cells at an E:T cell ratio of 10:1 and increasing concentrations of (Figure 14A) Ab-A1 XmAb, (Figure 14B) Ab-A1 XmAb 2+1 or (Figure 14C) Ab-A2(N67Q)XmAb 2+1 The Ab-A1 Xmab, Ab-A1 XmAb, and Ab-A1 XmAb were co-cultured for 48 hours in the same format (with N67Q substitution). Target cell lysis was monitored by measuring luciferase activity, and specific cytotoxicity was plotted at each concentration compared to the control condition without XmAb. As shown in Figures 14A-14C, Ab-A1 Xmab, Ab-A1 XmAb 2+1 and Ab-A2(N67Q)XmAb 2+1 was successful in mediating target cell lysis.

[0265] Example 6 This example demonstrates the ability of the heterodimeric antibodies of the present disclosure to discriminate between STEAP1 expressing cells and cells that do not express STEAP.

[0266] STEAP1-positive C4-2B luc prostate cancer cells (●) and STEAP1-negative C4-2B luc STEAP1 KO Cells (■) were incubated with human pan-T cells at an E:T cell ratio of 10:1 and increasing concentrations of Ab-A1 XmAb 2+1 Lysis of target cells co-cultured with Ab-A1 for 48 hours was monitored by measuring luciferase activity, and specific cytotoxicity was plotted at each concentration relative to the control condition (lacking XmAb). The results are shown in Figure 15 and Table 5 below. Ab-A1 XmAb 2+1 mediated target cell lysis in a dose-dependent manner in the human tumor cell line C4-2B luc, but not in C4-2B luc cells engineered to knock out STEAP1 expression.

[0267] [Table 4]

[0268] Furthermore, 293T cells stably transfected with human STEAP1 (FIG. 16A) or parental 293T cells (FIG. 16B) were incubated with isotype control antibody or anti-STEAP1 Ab-A mouse monoclonal antibody (Ab-Am; not in bispecific format) at a concentration of 10 μg / mL for 1 h at 4° C. Cell-bound Ab-Am was detected by flow cytometry after incubation with FITC-conjugated anti-mouse IgG secondary antibody (FIG. 16A). FITC fluorescence, identifying STEAP1-dependent signals, was plotted in histograms (filled grey histograms) and compared to isotype control (white histograms). As shown in FIG. 16A and FIG. 16B, Ab-Am detected STEAP-1 expressed in both cell populations tested.

[0269] FIG. 16C shows STEAP1 stable 293T cells (●) and STEAP1 negative parental 293T cells (■) were co-cultured with human pan T cells at an E:T cell ratio of 10:1 and increasing concentrations of Ab-A2(N67Q)XmAb. 2+1 16A-16C and Table 6 below. Anti-STEAP1 / anti-CD3 heterodimeric antibodies selectively mediated cytolysis of STEAP1-expressing cells.

[0270] [Table 5]

[0271] STEAP1-positive C4-2B luc prostate cancer cells were also cultured with human pan-T cells at an E:T cell ratio of 10:1 and increasing concentrations of Ab-B1 XmAb (●) or Ab-B1 XmAb (●). 2+1 As shown in Figure 17A, Ab-B1 XmAb and Ab-B1 XmAb were co-cultured for 48 hours. 2+1 mediated targeted cell lysis of C4-2B luc prostate cancer cells.

[0272] C4-2B luc prostate cancer cells were cocultured with human pan-T cells at an E:T cell ratio of 10:1 and increasing concentrations of XmAb. 2+1 Ab-B1-G37A (XmAb with G37A substitution 2+1 )(■), XmAb 2+1 Ab-B1-S39A (XmAb with S39A substitution 2+1 Format)(▲) or XmAb 2+1 Ab-B1-G37A / S39A (XmAb with both G37A and S39A substitutions) 2+1 As shown in FIG. 17B, Ab-B1 mutants (i.e., Ab-B1-G37A, Ab-B1-S39A, and Ab-B1-G37A / S39A) mediated target cell lysis of C4-2B luc prostate cancer cells. See Table 7 below. Similarly, STEAP1-negative C4-2B lucSTEAP1 KO Cancer cells were cocultured with human pan-T cells at a 10:1 E:T cell ratio and increasing concentrations of Ab-B1 XmAb. 2+1 Lysis of target cells co-cultured for 48 hours with Ab-B1-G37A (■), Ab-B1-S39A (▲) or Ab-B1-G37A / S39A (▼) was monitored by measuring luciferase activity and specific cytotoxicity was plotted at each concentration compared to the control condition without XmAb. Results are shown in FIG. 17C and Table 7 below. Anti-STEAP1 / anti-CD3 heterodimeric antibodies selectively mediated cytolysis of STEAP1 expressing cells, and the XmAb of the present disclosure selectively mediated cytolysis of STEAP1 expressing cells. 2+1 The format outperformed the other formats.

[0273] [Table 6]

[0274] Example 7 This example illustrates the use of a heterodimeric antibody of the present disclosure against human and cynomolgus monkey CD3ε, Ab-A2(N67Q)XmAb. 2+1 Characterize the equilibrium binding constant (KD) of

[0275] Ab-A2(N67Q)XmAb against recombinant human or cynomolgus CD3ε 2+1 The affinity of Ab-A2(N67Q)XmAb was measured using surface plasmon resonance (SPR-Pioneer FE). Recombinant human CD3ε-Fc and cynomolgus monkey CD3ε-Fc were immobilized on a CM5 chip surface using standard amine coupling procedures at approximately 60 RU. 2+1 was injected at concentrations of 100, 33.3, 11.1 and 3.7 nM. Ab against the ligand -A2(N67Q)XmAb 2+1 The association and dissociation rates of the interactions were recorded for 120 and 300 seconds, respectively, as listed in Table 8 below. The equilibrium dissociation constants (K D ) values ​​are the dissociation rate constant and the association rate constant (k off / k on ) was derived as a ratio of

[0276] [Table 7]

[0277] Example 8 This example illustrates the heterodimeric antibody (Ab-A2(N67Q)XmAb 2+1 ) mediates lysis of target cells displaying a range of STEAP1 surface densities.

[0278] STEAP-1 density on the surface of various target cell lines (SNU-5, C4-2B, Sk-N-MC, LOX-IMVI, VCaP, IM-95, TYKNU, 22RV-1, HBSCM, HUCCT1, PC3, HCT116 and NCIH1869) was assessed. See Table 9 below, which identifies STEAP1 density (number of STEAP I antibody binding sites per cell) in column 3 as measured using the Dako Qifikit method.

[0279] [Table 8]

[0280] In separate studies, STEAP-1 density was assessed on the surface of OE33, EBC1 and A673 cell lines. See Table 10 below, which identifies STEAP1 density (number of STEAP I antibody binding sites per cell) in column 3 as measured using the Dako Qifikit method.

[0281] [Table 9]

[0282] T cells from human donors were treated with increasing concentrations of Ab-A2(N67Q)XmAb 2+1The molecules were incubated with the target cell lines for 48 hours at 37°C. After 48 hours, target cell viability was measured using steady glo (B) or cell titer glo, which measures cell viability. Ab-A2(N67Q)XmAb 2+1 killed all cell lines with various EC90s.

[0283] Ab-A2(N67Q)XmAb 2+1 Ab-A2(N67Q)XmAb can kill cancer cell lines with STEAP1 densities ranging from approximately 200,000 STEAP1 receptors per cell (SNU5 cell line) to approximately 10,000 STEAP1 receptors per cell (LOX-IMV cell line). 2+1 The efficacy of Ab-A2(N67Q)XmAb is reduced when STEAP1 receptor density falls below 10,000 per cell. 2+1 preferentially mediates T cell-dependent killing of cells having a surface density of STEAP1 greater than 10,000 (e.g., the EC90 is at least 10-fold less for cells having a surface density of STEAP1 greater than 10,000 compared to cells having a surface density of STEAP1 less than 10,000).

[0284] Specific killing of cancer cells was achieved by Ab-A2-N67G XmAb 2+1 Compared to Xmab 2+1 Other antibodies in the format (Ab-B-G52A XmAb 2+1 and mouse antibody Ab-Cm XmAb 2+1 ) was evaluated. Ab-A2-N67G XmAb 2+1 demonstrated differential killing between high and low STEAP1 expressing cells, whereas Ab-B-G52A XmAb 2+1 did not discriminate between high and low STEAP1 expressing cells, but instead killed any STEAP1 expressing cells. See Table 11. Ab-A2-N67G XmAb 2+1 leaves normal cells that express STEAP1 at low levels (i.e., less than 10,000 / cell), such as HSMBCs (primary human smooth muscle bronchial cells).

[0285] [Table 10]

[0286] Example 9 This example demonstrates that T cell-dependent cytotoxicity is enhanced using a combination of an anti-CD3 / anti-STEAP1 heterodimeric antibody described herein together with an anti-PD-1 antibody.

[0287] Generation of PD-L1 overexpressing cell line: GP2-293 cells were cultured in DMEM medium supplemented with 10% fetal bovine serum, 1% Pen / Strep, 1% HEPES and 1% GlutaMAX. Cells were plated at 75% confluency in 10cm dishes and incubated overnight at 37℃, 5% CO2. The next morning, cells were transfected. Tube A added 45μL Lipofectamine 3000 and 500μL OptiMEM medium. Tube B added 15μg MSCV_GFP_PD-L1 plasmid, 1.8μg VSV-g plasmid, 30μL P3000 reagent and 500μL OptiMEM medium. Tubes A and B were mixed and incubated at room temperature for 10 minutes. The mixture was added dropwise to the dish of GP2-293 cells, which was incubated overnight at 37℃, 5% CO2. The following morning, the medium was removed and replaced with 10 mL of fresh culture medium. That afternoon, the target cells were plated in 6-well plates at 75% confluency and incubated overnight at 37° C., 5% CO2. The following morning, the viral supernatant was collected from the GP2-293 cells and centrifuged (5 min, 1200 rpm). The supernatant was collected in a new tube and polybrene was added at 1:1000. The medium was removed from the plate containing the target cells and 2 mL of viral supernatant was added. For the supernatant cells, 1E6 cells were centrifuged at 1500 rpm for 5 min, resuspended in 500 μL RPMI supplemented with 10% fetal bovine serum and 1% pen / strep, plated in a 6-well plate, to which 2 mL of viral supernatant was added. The plate containing the target cells and viral supernatant was centrifuged at 1200×g for 1.5 h at 32° C., followed by incubation at 37° C., 5% CO2. Culture medium was added after 5 h. After 4 days, cells were analyzed for GFP and PD-L1 expression by flow cytometry on a FACSympony. PD-L1 was detected using a PE-conjugated antibody, clone 29E.2A3. Cells <70% positive for PD-L1 expression were sorted on a BD Melody sorter to select for cells expressing high levels of PD-L1.

[0288] T-cell-dependent cytotoxicity (TDCC) assay: Ab-A2(N67Q)XmAb 2+1 were diluted in cell culture medium (RPMI, 10% heat-inactivated fetal bovine serum, 1X GlutaMAX, 1X Pen / Strep), serially diluted (1:3, 22 total) and transferred to black clear-bottom 384-well plates using a Bravo liquid handling robot. Human pan T cells (n=4) pre-activated with CD3 / CD28 Dynabeads (1:1, 48 hours) were separated from the beads using a magnet and diluted in cell culture medium. (An aliquot of activated T cells from each donor was assessed for PD-1 expression by flow cytometry. Cells were stained as above and data were collected on a FACSymphony flow cytometer and analyzed using FlowJo v10.1.) Activated T cells (2500 cells / 20 μL; 4 rows / donor) followed by target cells overexpressing PD-L1 were plated into 384-well assay plates (2500 cells / 20 μL; full plate) for a final effector to target cell (E:T) ratio of 1:1. An anti-PD-1 antibody of the present disclosure comprising CDR sequences of SEQ ID NOs: 189-194 (10 μg / mL final at 5 μl) was added to two rows of each T cell donor. Plates were covered with MicroClime lids and incubated at 37° C., 5% CO2 for 24 hours. For assays with luciferase expressing target cells, 30 μL of Steady-Glo, Bright-Glo or One-Glo reagent (Promega) was added. Plates with adherent target cells not expressing luciferase were washed with PBS using an EL406 plate washer to remove T cells and 25 μL of Cell Titer Glo reagent was added. Plates were incubated with the reagent for 10 minutes in the dark at room temperature. Luminescence was detected using a BioTek Neo plate reader. Specific cytotoxicity was measured using Ab-A2 XmAb 2+1 Dose curves were plotted and EC50 values ​​were calculated by 4-parameter variable slope curve fitting using GraphPad Prism software.

[0289] The results of the TDCC assay are shown in Figures 20A and 20B. Ab-A2(N67Q)XmAb 2+1 and the anti-PD-1 antibody combination was Ab-A2(N67Q)XmAb 2+1 It demonstrated enhanced cytotoxicity and a reduced EC50 compared to alone.

[0290] Example 10 This example demonstrates the use of heterodimeric antibodies of the disclosure (e.g., Ab-A2(N67Q)XmAb) in reducing Ewing sarcoma tumor volume in vivo. 2+1 ) capabilities.

[0291] Sublethally irradiated NOD / SCID female immunocompromised mice were given 5 × 10 6 STEAP1-expressing SK-N-MC tumor cells were transplanted at 2 × 10 7 CD3+ human T cells were injected intraperitoneally. Ab-A2(N67Q)XmAb 2+1 Or vehicle control was administered by intravenous (IV) bolus injection at 0.01, 0.1 or 1 mg / kg on days 12, 19 and 26. Tumor volume data over time are presented graphically (Figure 22).

[0292] Ab-A2(N67Q)XmAb 2+1 induced early tumor regression, with relative tumor volumes (RTVs) of <1 in all treatment groups between days 15 and 22, whereas the RTVs in vehicle-treated animals increased persistently until the end of the study. 2+1 Tumors from animals receiving (0.01 mg / kg) began to regrow after 22 days, whereas tumors from animals receiving the higher Ab-A2(N67Q)XmAb 2+1 The mean RTV for mice treated with doses (0.1 and 1 mg / kg) was <1 by days 28 and 25, respectively (Table 12).

[0293] [Table 11]

[0294] When compared to vehicle-treated control Group 2, p values ​​<0.001 were observed for all Ab-A2(N67Q)XmAbs from days 15 to 22. 2+1 Dose levels achieved, after day 22, p values ​​<0.001 for 0.1 and 1 mg / kg Ab-A2(N67Q)XmAb 2+1 At day 28, tumors in vehicle-treated mice (group 2) were on average 5.29-fold larger compared to their starting volume before treatment began, whereas Ab-A2(N67Q)XmAb 2+1 The group mean RTV in the treatment groups was 0.61 (group 3), 1.13 (group 4) and 4.24 (group 5) (Table 1). At the end of the survival period on day 28, the highest Ab-A2(N67Q)XmAb 2+1 9 / 10 animals in the dose group (Group 3) were considered tumor free and had a tumor growth inhibition (TGI) of 97% (Table 13).

[0295] [Table 12]

[0296] Thus, in a clinically relevant xenograft model, Ab-A2(N67Q)XmAb 2+1 demonstrated convincing antitumor activity. Without being limited thereto, the present invention includes the following aspects. [Aspect 1] An antigen-binding protein that binds to STEAP1 of SEQ ID NO:2, (a) a heavy chain CDR comprising an amino acid sequence that differs by no more than 3, 2 or 1 amino acid from i) vhCDR1 SEQ ID NO:14, vhCDR2 SEQ ID NO:15, or vhCDR2 SEQ ID NO:21 and vhCDR3 SEQ ID NO:16, or ii) vhCDR1 SEQ ID NO:33, vhCDR2 SEQ ID NO:34, and vhCDR3 SEQ ID NO:35; or (b) a light chain CDR comprising an amino acid sequence that differs by no more than 3, 2 or 1 amino acid from: i) vlCDR1 SEQ ID NO:11, vlCDR2 SEQ ID NO:12, and vlCDR3 SEQ ID NO:13; or ii) vlCDR1 SEQ ID NO:30, vlCDR2 SEQ ID NO:31, and vlCDR3 SEQ ID NO:32; or (c) a light chain variable domain comprising an amino acid sequence at least 90% identical to SEQ ID NO: 183 or SEQ ID NO: 186; or (d) a heavy chain variable domain comprising an amino acid sequence at least 90% identical to SEQ ID NO: 182, SEQ ID NO: 184, or SEQ ID NO: 185. 2. An antigen-binding protein comprising: [Aspect 2] CDR sequences a) a vhCDR1 comprising SEQ ID NO: 14, a vhCDR2 comprising SEQ ID NO: 15 or SEQ ID NO: 21 and a vhCDR3 comprising SEQ ID NO: 16; or b) vhCDR1 comprising SEQ ID NO: 33, vhCDR2 comprising SEQ ID NO: 34 and vhCDR3 comprising SEQ ID NO: 35 2. The antigen-binding protein of embodiment 1, comprising: [Aspect 3] a) vlCDR1 comprising SEQ ID NO: 11, vlCDR2 comprising SEQ ID NO: 12 and vlCDR3 comprising SEQ ID NO: 13; or b) vlCDR1 comprising SEQ ID NO: 30, vlCDR2 comprising SEQ ID NO: 31 and vlCDR3 comprising SEQ ID NO: 32 3. The antigen-binding protein of embodiment 1 or 2, comprising a CDR sequence selected from: [Aspect 4] vhCDR1 comprising SEQ ID NO: 14, vhCDR2 comprising SEQ ID NO: 15 or SEQ ID NO: 21, vhCDR3 comprising SEQ ID NO: 16, vlCDR1 comprising SEQ ID NO: 11, vlCDR2 comprising SEQ ID NO: 12, and vlCDR3 comprising SEQ ID NO: 13 4. The antigen-binding protein according to any one of aspects 1 to 3, comprising: [Aspect 5] 4. The antigen-binding protein according to any one of Aspects 1 to 3, comprising a vhCDR1 comprising SEQ ID NO: 33, a vhCDR2 comprising SEQ ID NO: 34, a vhCDR3 comprising SEQ ID NO: 35, a vlCDR1 comprising SEQ ID NO: 30, a vlCDR2 comprising SEQ ID NO: 31 and a vlCDR3 comprising SEQ ID NO: 32. [Aspect 6] 4. The antigen-binding protein according to any one of aspects 1 to 3, comprising a variable heavy domain comprising SEQ ID NO: 182, SEQ ID NO: 184 or SEQ ID NO: 185. [Aspect 7] 4. The antigen-binding protein according to any one of aspects 1 to 3, comprising a variable light chain domain comprising SEQ ID NO: 183 or SEQ ID NO: 186. [Aspect 8] 4. The antigen-binding protein according to any one of aspects 1 to 3, comprising a variable heavy chain domain comprising SEQ ID NO: 182 or SEQ ID NO: 184 and a variable light chain domain comprising SEQ ID NO: 183. [Aspect 9] 4. The antigen-binding protein according to any one of aspects 1 to 3, comprising a variable heavy chain domain comprising SEQ ID NO: 185 and a variable light chain domain comprising SEQ ID NO: 186. [Aspect 10] 10. An antigen-binding protein that cross-blocks the binding to STEAP1 of a reference antigen-binding protein according to any one of aspects 1 to 9, or is cross-blocked from binding to STEAP1 by a reference antigen-binding protein according to any one of aspects 1 to 9. [Aspect 11] 11. The antigen-binding protein of aspect 10, wherein said cross-blocking is detected by surface plasmon resonance or ELISA. [Aspect 12] 11. The antigen binding protein of embodiment 10, which does not bind to extracellular loop 2 of STEAP1. [Aspect 13] 11. An antigen-binding protein according to embodiment 10, which preferentially mediates T cell-dependent killing of cells having a surface density of STEAP1 of greater than 10,000. [Aspect 14] An antigen-binding protein that binds to regions of STEAP1 within amino acids 92-118 and amino acids 279-290. [Aspect 15] 15. The antigen-binding protein according to any one of aspects 1 to 14, which is an antibody. [Aspect 16] 16. The antigen-binding protein according to embodiment 15, which is a monoclonal antibody, a chimeric antibody or a humanized antibody. [Aspect 17] 15. The antigen-binding protein according to any one of aspects 1 to 14, which is an antigen-binding antibody fragment. [Aspect 18] 15. The antigen-binding protein according to any one of aspects 1 to 14, comprising a single chain antibody, a diabody, a triabody, a tetrabody or a domain antibody. [Aspect 19] 19. A pharmaceutical composition comprising the antigen-binding protein according to any one of aspects 1 to 18 and a physiologically acceptable excipient. [Aspect 20] The pharmaceutical composition of aspect 19, further comprising an anti-PD-1 antigen binding protein comprising a vhCDR1 comprising the amino acid sequence set forth in SEQ ID NO: 189, a vhCDR2 comprising the amino acid sequence set forth in SEQ ID NO: 190, a vhCDR3 comprising the amino acid sequence set forth in SEQ ID NO: 191, a vlCDR1 comprising the amino acid sequence set forth in SEQ ID NO: 192, a vlCDR2 comprising the amino acid sequence set forth in SEQ ID NO: 193, and a vlCDR3 comprising the amino acid sequence set forth in SEQ ID NO: 194. [Aspect 21] A method of treating cancer comprising administering to a subject in need thereof an antigen-binding protein according to any one of aspects 1 to 18. [Aspect 22] 22. The method of embodiment 21, further comprising administering to the subject an anti-PD-1 antigen binding protein. [Aspect 23] 20. Use of an antigen-binding protein according to any one of aspects 1 to 18 in the preparation of a medicament for treating cancer in a subject in need thereof. [Aspect 24] 24. The use of aspect 23, wherein the medicament is for administering an effective amount of said anti-STEAP1 antigen binding protein in combination with an effective amount of an anti-PD-1 antigen binding protein. [Aspect 25] 20. An antigen-binding protein according to any one of aspects 1 to 18 for use in treating cancer in a subject in need thereof. [Aspect 26] 26. The antigen binding protein for use according to embodiment 25, administered in combination with an anti-PD1 antigen binding protein. [Aspect 27] 27. The method, use or antigen binding protein for use according to any one of aspects 22, 24 or 26, wherein the anti-PD1 antigen binding protein comprises a vhCDR1 comprising the amino acid sequence set forth in SEQ ID NO: 189, a vhCDR2 comprising the amino acid sequence set forth in SEQ ID NO: 190, a vhCDR3 comprising the amino acid sequence set forth in SEQ ID NO: 191, a vlCDR1 comprising the amino acid sequence set forth in SEQ ID NO: 192, a vlCDR2 comprising the amino acid sequence set forth in SEQ ID NO: 193 and a vlCDR3 comprising the amino acid sequence set forth in SEQ ID NO: 194. [Aspect 28] 28. The method, use or antigen binding protein for use according to aspect 27, wherein said anti-PD1 antigen binding protein comprises a heavy chain variable domain comprising an amino acid sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO: 195, and a light chain variable domain comprising an amino acid sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO: 196. [Aspect 29] 29. The method, use or antigen binding protein for use according to embodiment 28, wherein said anti-PD1 antigen binding protein comprises a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 195 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 196. [Aspect 30] 30. The method, use or antigen-binding protein for use according to any one of aspects 27 to 29, wherein the anti-PD-1 antigen-binding protein is an antigen-binding antibody fragment. [Aspect 31] 30. The method, use or antigen-binding protein for use according to any one of aspects 27 to 29, wherein the anti-PD-1 antigen-binding protein is an antibody. [Aspect 32] 30. The method, use or antigen-binding protein for use according to any one of aspects 27 to 29, wherein the anti-PD-1 antigen-binding protein is a monoclonal antibody, a chimeric antibody or a humanized antibody. [Aspect 33] 32. The method, use or antigen binding protein for use according to any one of Aspects 27 to 31, wherein the anti-PD1 antigen binding protein comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 197 and a light chain comprising the amino acid sequence of SEQ ID NO: 198. [Aspect 34] 34. The method, use or antigen-binding protein for use according to any one of aspects 21 to 33, wherein the cancer is prostate cancer. [Aspect 35] 34. The method, use or antigen binding protein for use according to any one of aspects 21 to 33, wherein the cancer is Ewing's sarcoma. [Aspect 36] A polynucleotide comprising a nucleic acid sequence encoding the light chain variable domain and / or the heavy chain variable domain of the antigen-binding protein according to any one of aspects 1 to 14. [Aspect 37] An expression vector comprising the polynucleotide of embodiment 36. [Aspect 38] A composition comprising a polynucleotide comprising a nucleic acid sequence encoding the light chain variable domain of the antigen-binding protein according to any one of aspects 1 to 14, and a polynucleotide comprising a nucleic acid sequence encoding the heavy chain variable domain of the antigen-binding protein according to any one of aspects 1 to 14. [Aspect 39] 39. A method of making an antigen binding protein, the method comprising contacting a host cell with the polynucleotide of aspect 36 or the composition of aspect 38 under conditions allowing expression of said light chain variable domain and said heavy chain variable domain. [Aspect 40] A bispecific antigen-binding protein comprising the antigen-binding protein according to any one of aspects 1 to 14. [Aspect 41] 41. The bispecific antigen-binding protein according to embodiment 40, which binds to STEAP1 and CD3. [Aspect 42] 42. The bispecific antigen-binding protein according to aspect 41, comprising a CD3-binding domain comprising the CDR sequences of SEQ ID NOs: 170 to 172 and 174 to 176. [Aspect 43] A heterodimeric antibody, a) a first monomer, 1) a first variable heavy domain; 2) a first constant heavy chain comprising a first CH1 domain and a first Fc domain; 3) an scFv that binds to human CD3 and comprises an scFv variable light domain, an scFv linker, and an scFv variable heavy domain, the scFv being covalently linked using a domain linker between the C-terminus of the CH1 domain and the N-terminus of the first Fc domain; a first monomer comprising a first heavy chain comprising: b) a second monomer comprising a second heavy chain comprising a second variable heavy chain domain and a second constant heavy chain comprising a second Fc domain; and c) a common light chain comprising a variable light domain and a constant light domain Includes; the first variable heavy chain domain and the variable light chain domain bind to human STEAP1, and the second variable heavy chain domain and the variable light chain domain bind to human STEAP1; (i) the first variable heavy chain domain and the second variable heavy chain domain comprise heavy chain CDRs that comprise amino acid sequences that differ by no more than 3, 2 or 1 amino acid from (a) vhCDR1 SEQ ID NO:14, (b) vhCDR2 SEQ ID NO:15 or SEQ ID NO:21, and (3) vhCDR3 SEQ ID NO:16, and the variable light chain domain comprises light chain CDRs that comprise amino acid sequences that differ by no more than 3, 2 or 1 amino acid from vlCDR1 SEQ ID NO:11, vlCDR2 SEQ ID NO:12, and vlCDR3 SEQ ID NO:13; or (ii) the first variable heavy chain domain and the second variable heavy chain domain comprise heavy chain CDRs that comprise amino acid sequences that differ by no more than 3, 2 or 1 amino acid from vhCDR1 SEQ ID NO:33, vhCDR2 SEQ ID NO:34 and vhCDR3 SEQ ID NO:35, and the variable light chain domain comprises light chain CDRs that comprise amino acid sequences that differ by no more than 3, 2 or 1 amino acid from vlCDR1 SEQ ID NO:30, vlCDR2 SEQ ID NO:31 and vlCDR3 SEQ ID NO:32; or (iii) the first variable heavy domain and the second variable heavy domain comprise an amino acid sequence at least 90% identical to SEQ ID NO: 182 or 184, and the variable light domain comprises an amino acid sequence at least 90% identical to SEQ ID NO: 183; or (iv) a heterodimeric antibody, wherein the first variable heavy domain and the second variable heavy domain comprise an amino acid sequence at least 90% identical to SEQ ID NO: 185, and the variable light domain comprises an amino acid sequence at least 90% identical to SEQ ID NO: 186. [Aspect 44] 44. The heterodimeric antibody of embodiment 43, wherein the first monomer comprises the amino acid substitutions E233P, L235V, G236A, S267K, R292C, N297G, V302C, E357Q, and S364K; the second monomer comprises the amino acid substitutions N208D, E233P, L235V, G236A, S267K, R292C, Q295E, N297G, V302C, L368D, K370S, N384D, Q418E, and N421D; and both monomers comprise a deletion at position 234. [Aspect 45] The scFv comprises (i) a variable heavy chain domain comprising a heavy chain CDR comprising an amino acid sequence that differs by no more than 3, 2 or 1 amino acid from vhCDR1 SEQ ID NO: 170, vhCDR2 SEQ ID NO: 171, and vhCDR3 SEQ ID NO: 172, and a variable light chain domain comprising a light chain CDR comprising an amino acid sequence that differs by no more than 3, 2 or 1 amino acid from vlCDR1 SEQ ID NO: 174, vlCDR2 SEQ ID NO: 175, and vlCDR3 SEQ ID NO: 176; or (ii) a variable heavy chain domain comprising an amino acid sequence at least 90% identical to SEQ ID NO: 169 and a variable light chain domain comprising an amino acid sequence at least 90% identical to SEQ ID NO: 173 45. The heterodimeric antibody of embodiment 43 or 44, comprising: [Aspect 46] 45. The heterodimeric antibody of aspect 43 or 44, wherein the scFv comprises CDRs comprising vhCDR1 comprising SEQ ID NO: 170, vhCDR2 comprising SEQ ID NO: 171, vhCDR3 comprising SEQ ID NO: 172, vlCDR1 comprising SEQ ID NO: 174, vlCDR2 comprising SEQ ID NO: 175 and vlCDR3 comprising SEQ ID NO: 176. [Aspect 47] The heterodimeric antibody according to any one of aspects 43 to 45, wherein the scFv comprises the variable heavy chain region and the variable light chain region of SEQ ID NO:169 and SEQ ID NO:173. [Aspect 48] The heterodimeric antibody according to any one of aspects 43 to 47, wherein the scFv has a charged scFv linker. [Aspect 49] The heterodimeric antibody according to aspect 48, wherein the charged scFv linker has 3 to 8 positive charges and is selected from the group consisting of SEQ ID NOs: 143 to 153. [Aspect 50] 49. The heterodimeric antibody of embodiment 48, wherein the scFv linker comprises SEQ ID NO: 152. [Aspect 51] The heterodimeric antibody according to any one of aspects 43 to 47, wherein the scFv comprises the sequence of SEQ ID NO:44. [Aspect 52] 52. The heterodimeric antibody according to any one of Aspects 43 to 51, wherein the first variable heavy chain domain and the second variable heavy chain domain comprise a vhCDR1 comprising CDR sequence SEQ ID NO: 14, a vhCDR2 comprising SEQ ID NO: 15 or SEQ ID NO: 21, and a vhCDR3 comprising SEQ ID NO: 16, and the variable light chain domain comprises a vlCDR1 comprising CDR sequence SEQ ID NO: 11, a vlCDR2 comprising SEQ ID NO: 12, and a vlCDR3 comprising SEQ ID NO: 13. [Aspect 53] The heterodimeric antibody according to any one of Aspects 43 to 51, wherein the first variable heavy chain domain and the second variable heavy chain domain comprise CDR sequences vhCDR1 comprising SEQ ID NO: 33, vhCDR2 comprising SEQ ID NO: 34, and vhCDR3 comprising SEQ ID NO: 35, and the variable light chain domain comprises CDR sequences vlCDR1 comprising SEQ ID NO: 30, vlCDR2 comprising SEQ ID NO: 31, and vlCDR3 comprising SEQ ID NO: 32. [Aspect 54] 52. The heterodimeric antibody according to any one of aspects 43 to 51, wherein the first variable heavy chain domain and the second variable heavy chain domain comprise SEQ ID NO: 182 or SEQ ID NO: 184, and the variable light chain domain comprises SEQ ID NO: 183. [Aspect 55] 52. The heterodimeric antibody according to any one of aspects 43 to 51, wherein the first variable heavy domain and the second variable heavy domain comprise SEQ ID NO: 185, and the variable light domain comprises SEQ ID NO: 186. [Aspect 56] 54. The heterodimeric antibody according to embodiment 52 or 53, comprising the amino acid substitution N67Q and / or a substitution at one or more of positions 292, 297 or 302. [Aspect 57] a) the first monomer comprises the sequence of SEQ ID NO: 19 or SEQ ID NO: 20, the second monomer comprises the sequence of SEQ ID NO: 18 or 199, and the common light chain comprises the sequence of SEQ ID NO: 17; or b) The heterodimeric antibody according to embodiment 43, wherein the first monomer comprises the sequence of SEQ ID NO: 38, the second monomer comprises the sequence of SEQ ID NO: 37, and the common light chain comprises the sequence of SEQ ID NO: 36. [Aspect 58] a) the first monomer comprises the sequence of SEQ ID NO:202, the second monomer comprises the sequence of SEQ ID NO:201, and the common light chain comprises the sequence of SEQ ID NO:200; b) the first monomer comprises the sequence of SEQ ID NO: 207, the second monomer comprises the sequence of SEQ ID NO: 203, and the common light chain comprises the sequence of SEQ ID NO: 200; or c) The heterodimeric antibody according to embodiment 43, wherein the first monomer comprises the sequence of SEQ ID NO: 206, the second monomer comprises the sequence of SEQ ID NO: 205, and the common light chain comprises the sequence of SEQ ID NO: 204. [Aspect 59] 1. A nucleic acid composition comprising: a) a first nucleic acid encoding a first monomer according to any one of embodiments 43 to 58; b) a second nucleic acid encoding a second monomer according to any one of aspects 43 to 58; and c) a third nucleic acid encoding a common light chain according to any one of embodiments 43 to 58. A nucleic acid composition comprising: [Aspect 60] 1. A nucleic acid composition comprising: a) a first expression vector comprising a first nucleic acid encoding a first monomer according to any one of aspects 43 to 58; b) a second expression vector comprising a second nucleic acid encoding a second monomer according to any one of aspects 43 to 58; and c) a third expression vector comprising a third nucleic acid encoding the common light chain according to any one of aspects 43 to 58. A nucleic acid composition comprising: [Aspect 61] 61. A host cell comprising a nucleic acid composition according to embodiment 59 or 60. [Aspect 62] A pharmaceutical composition comprising the heterodimeric antibody according to any one of aspects 43 to 58. [Aspect 63] A method of treating a subject in need thereof, comprising administering to the subject a heterodimeric antibody according to any one of aspects 43 to 58. [Aspect 64] A method of treating cancer in a subject in need thereof, comprising administering to the subject a heterodimeric antibody according to any one of aspects 43 to 58. [Aspect 65] 65. The method of embodiment 63 or 64, further comprising administering to the subject an anti-PD-1 antigen binding protein. [Aspect 66] 59. Use of the heterodimeric antibody according to any one of aspects 43 to 58 in the preparation of a medicament for treating cancer in a subject in need thereof. [Aspect 67] 67. The use of aspect 66, wherein the medicament is for administering an effective amount of said heterodimeric antibody in combination with an effective amount of an anti-PD-1 antigen binding protein. [Aspect 68] 59. The heterodimeric antibody according to any one of aspects 43 to 58, for use in treating cancer in a subject in need thereof. [Aspect 69] 69. The heterodimeric antibody for use according to embodiment 68, administered in combination with an anti-PD1 antigen binding protein. [Aspect 70] 70. The method, use or antigen binding protein for use according to any one of aspects 65, 67 or 69, wherein the anti-PD1 antigen binding protein comprises a vhCDR1 comprising the amino acid sequence set forth in SEQ ID NO: 189, a vhCDR2 comprising the amino acid sequence set forth in SEQ ID NO: 190, a vhCDR3 comprising the amino acid sequence set forth in SEQ ID NO: 191, a vlCDR1 comprising the amino acid sequence set forth in SEQ ID NO: 192, a vlCDR2 comprising the amino acid sequence set forth in SEQ ID NO: 193 and a vlCDR3 comprising the amino acid sequence set forth in SEQ ID NO: 194. [Aspect 71] 71. The method, use or antigen binding protein for use according to aspect 70, wherein said anti-PD1 antigen binding protein comprises a heavy chain variable domain comprising an amino acid sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO: 195, and a light chain variable domain comprising an amino acid sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO: 196. [Aspect 72] 72. The method, use or antigen binding protein for use according to embodiment 71, wherein said anti-PD1 antigen binding protein comprises a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO: 195 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO: 196. [Aspect 73] 73. The method, use or heterodimeric antibody for use according to any one of aspects 70 to 72, wherein the anti-PD-1 antigen-binding protein is an antigen-binding antibody fragment. [Aspect 74] 73. The method, use or heterodimeric antibody for use according to any one of aspects 70 to 72, wherein the anti-PD-1 antigen-binding protein is an antibody. [Aspect 75] 73. The method, use or heterodimeric antibody for use according to any one of aspects 70 to 72, wherein the anti-PD-1 antigen-binding protein is a monoclonal antibody, a chimeric antibody or a humanized antibody. [Aspect 76] 76. The method, use or heterodimeric antibody for use according to any one of aspects 70 to 75, wherein the anti-PD1 antigen-binding protein comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 197 and a light chain comprising the amino acid sequence of SEQ ID NO: 198. [Aspect 77] 76. The method, use or heterodimeric antibody for use according to any one of aspects 70 to 75, wherein the anti-PD1 antigen-binding protein comprises a heavy chain comprising the amino acid sequence of SEQ ID NO: 208 and a light chain comprising the amino acid sequence of SEQ ID NO: 209. [Aspect 78] 78. The method, use or heterodimeric antibody for use according to any one of aspects 64 to 77, wherein the cancer is prostate cancer. [Aspect 79] 78. The method, use or heterodimeric antibody for use according to any one of aspects 64 to 77, wherein the cancer is Ewing's sarcoma.

Claims

1. An antigen binding protein that binds to STEAP1 of SEQ ID NO:2, a heavy chain variable domain comprising the amino acid sequences of vhCDR1 SEQ ID NO:14, vhCDR2 SEQ ID NO:15 or vhCDR2 SEQ ID NO:21, and vhCDR3 SEQ ID NO:16; and A light chain variable domain comprising the amino acid sequences of vlCDR1 SEQ ID NO:11, vlCDR2 SEQ ID NO:12, and vlCDR3 SEQ ID NO:

13. The antigen-binding protein comprising:

2. A light chain variable domain comprising an amino acid sequence at least 90% identical to SEQ ID NO: 183; and A heavy chain variable domain comprising an amino acid sequence at least 90% identical to SEQ ID NO: 182 or SEQ ID NO:

184.

2. The antigen binding protein of claim 1 , comprising:

3. vhCDR1 comprising SEQ ID NO: 14, vhCDR2 comprising SEQ ID NO: 15, vhCDR3 comprising SEQ ID NO: 16, vlCDR1 comprising SEQ ID NO: 11, vlCDR2 comprising SEQ ID NO: 12, and vlCDR3 comprising SEQ ID NO: 13 The antigen-binding protein of any one of claims 1 to 2, comprising:

4. vhCDR1 comprising SEQ ID NO: 14, vhCDR2 comprising SEQ ID NO:21, vhCDR3 comprising SEQ ID NO: 16, vlCDR1 comprising SEQ ID NO: 11, vlCDR2 comprising SEQ ID NO: 12, and vlCDR3 comprising SEQ ID NO: 13 The antigen-binding protein of any one of claims 1 to 2, comprising:

5. 3. An antigen binding protein according to any one of claims 1 to 2, comprising a variable heavy chain domain comprising SEQ ID NO:182 and a variable light chain domain comprising SEQ ID NO:

183.

6. 3. An antigen binding protein according to any one of claims 1 to 2, comprising a variable heavy chain domain comprising SEQ ID NO:184 and a variable light chain domain comprising SEQ ID NO:

183.

7. The antigen-binding protein of any one of claims 1 to 6, which is an antibody.

8. 8. The antigen-binding protein of claim 7, which is a monoclonal antibody, a chimeric antibody or a humanized antibody.

9. 7. The antigen-binding protein of any one of claims 1 to 6, which is an antigen-binding antibody fragment.

10. 7. The antigen binding protein of any one of claims 1 to 6, comprising a single chain antibody, a diabody, a triabody, or a tetrabody.

11. A pharmaceutical composition comprising the antigen-binding protein of any one of claims 1 to 10 and a physiologically acceptable excipient.

12. 12. The pharmaceutical composition of claim 11, further comprising an anti-PD-1 antigen binding protein comprising a vhCDR1 comprising the amino acid sequence set forth in SEQ ID NO: 189, a vhCDR2 comprising the amino acid sequence set forth in SEQ ID NO: 190, a vhCDR3 comprising the amino acid sequence set forth in SEQ ID NO: 191, a vlCDR1 comprising the amino acid sequence set forth in SEQ ID NO: 192, a vlCDR2 comprising the amino acid sequence set forth in SEQ ID NO: 193, and a vlCDR3 comprising the amino acid sequence set forth in SEQ ID NO:

194.

13. 12. The pharmaceutical composition of claim 11 for use in treating cancer in a subject in need thereof.

14. The pharmaceutical composition of claim 13, administered in combination with an anti-PD-1 antigen binding protein.

15. A polynucleotide comprising a nucleic acid sequence encoding the light chain variable domain or the heavy chain variable domain of the antigen binding protein of any one of claims 1 to 6.

16. An expression vector comprising the polynucleotide described in claim 15.

17. 7. A composition comprising a polynucleotide comprising a nucleic acid sequence encoding the light chain variable domain of an antigen binding protein according to any one of claims 1 to 6 and a polynucleotide comprising a nucleic acid sequence encoding the heavy chain variable domain of an antigen binding protein according to any one of claims 1 to 6.

18. A method for producing an antigen binding protein, comprising: (i) a polynucleotide encoding the light chain variable domain according to any one of claims 1 to 6 and a polynucleotide encoding the heavy chain variable domain according to any one of claims 1 to 6; or (ii) The composition of claim 17 under conditions that allow expression of said light chain variable domain and said heavy chain variable domain.

19. A bispecific antigen-binding protein comprising the antigen-binding protein of any one of claims 1 to 6.

20. 20. The bispecific antigen-binding protein of claim 19, which binds to STEAP1 and CD3.

21. 21. The bispecific antigen binding protein of claim 20, comprising a CD3 binding domain comprising the CDR sequences of vhCDR1 SEQ ID NO: 170, vhCDR2 SEQ ID NO: 171, vhCDR3 SEQ ID NO: 172, vlCDR1 SEQ ID NO: 174, vlCDR2 SEQ ID NO: 175 and vlCDR3 SEQ ID NO:

176.

22. A heterodimeric antibody, a) a first monomer, 1) a first variable heavy domain; 2) a first constant heavy chain comprising a first CH1 domain and a first Fc domain; 3) an scFv that binds to human CD3 and comprises an scFv variable light domain, an scFv linker, and an scFv variable heavy domain, the scFv being covalently linked using a domain linker between the C-terminus of the CH1 domain and the N-terminus of the first Fc domain; a first monomer comprising a first heavy chain comprising: b) a second monomer comprising a second heavy chain comprising a second variable heavy chain domain and a second constant heavy chain comprising a second Fc domain; and c) a common light chain comprising a variable light domain and a constant light domain Including; the first variable heavy chain domain and the variable light chain domain bind to human STEAP1, and the second variable heavy chain domain and the variable light chain domain bind to human STEAP1; The first variable heavy domain and the second variable heavy domain comprise the amino acid sequences of vhCDR1 SEQ ID NO: 14, vhCDR2 SEQ ID NO: 15 or SEQ ID NO: 21, and vhCDR3 SEQ ID NO: 16, and the variable light domain comprises the amino acid sequences of vlCDR1 SEQ ID NO: 11, vlCDR2 SEQ ID NO: 12, and vlCDR3 SEQ ID NO: 13; The heterodimeric antibody.

23. The first variable heavy domain and the second variable heavy domain comprise an amino acid sequence at least 90% identical to SEQ ID NO: 182 or 184, and the variable light domain comprises an amino acid sequence at least 90% identical to SEQ ID NO:

183. The heterodimeric antibody of claim 22.

24. 23. The heterodimeric antibody of claim 22, wherein the first variable heavy domain and the second variable heavy domain comprise CDR sequences vhCDR1 comprising SEQ ID NO: 14, vhCDR2 comprising SEQ ID NO: 15, vhCDR3 comprising SEQ ID NO: 16, and the variable light domain comprises CDR sequences vlCDR1 comprising SEQ ID NO: 11, vlCDR2 comprising SEQ ID NO: 12, and vlCDR3 comprising SEQ ID NO:

13.

25. 23. The heterodimeric antibody of claim 22, wherein the first variable heavy domain and the second variable heavy domain comprise CDR sequences vhCDR1 comprising SEQ ID NO: 14, vhCDR2 comprising SEQ ID NO: 21, vhCDR3 comprising SEQ ID NO: 16, and the variable light domain comprises CDR sequences vlCDR1 comprising SEQ ID NO: 11, vlCDR2 comprising SEQ ID NO: 12, and vlCDR3 comprising SEQ ID NO:

13.

26. 23. The heterodimeric antibody of claim 22, wherein the first variable heavy domain and the second variable heavy domain comprise SEQ ID NO: 182, and the variable light domain comprises SEQ ID NO:

183.

27. 23. The heterodimeric antibody of claim 22, wherein the first variable heavy domain and the second variable heavy domain comprise SEQ ID NO: 184, and the variable light domain comprises SEQ ID NO:

183.

28. A heterodimeric antibody according to any one of claims 22 to 27, wherein, in the EU numbering of IgG1, the first monomer comprises the amino acid substitutions E233P, L235V, G236A, S267K, R292C, N297G, V302C, E357Q and S364K; the second monomer comprises the amino acid substitutions N208D, E233P, L235V, G236A, S267K, R292C, Q295E, N297G, V302C, L368D, K370S, N384D, Q418E and N421D; and both monomers comprise a deletion at position 234.

29. The scFv is a variable heavy chain domain comprising the amino acid sequences of vhCDR1 SEQ ID NO: 170, vhCDR2 SEQ ID NO: 171 and vhCDR3 SEQ ID NO: 172, and a variable light chain domain comprising the amino acid sequences of vlCDR1 SEQ ID NO: 174, vlCDR2 SEQ ID NO: 175 and vlCDR3 SEQ ID NO: 176; The heterodimeric antibody of any one of claims 22 to 28, comprising:

30. A variable heavy chain domain comprising an amino acid sequence at least 90% identical to SEQ ID NO: 169, and a variable light chain domain comprising an amino acid sequence at least 90% identical to SEQ ID NO:

173. The heterodimeric antibody of claim 29 , comprising:

31. The heterodimeric antibody of claim 30, wherein the scFv comprises the variable heavy and light chain regions of SEQ ID NO: 169 and SEQ ID NO:

173.

32. The heterodimeric antibody of any one of claims 22 to 31, wherein the scFv has a charged scFv linker.

33. The heterodimeric antibody of claim 32, wherein the charged scFv linker has 3 to 8 positive charges and is selected from the group consisting of SEQ ID NOs: 144 to 153.

34. The heterodimeric antibody of claim 32, wherein the scFv linker comprises sequence number 152.

35. The heterodimeric antibody of any one of claims 22 to 31, wherein the scFv linker comprises SEQ ID NO:

143.

36. The heterodimeric antibody of any one of claims 22 to 31, wherein the scFv comprises the sequence of SEQ ID NO:

44.

37. The heterodimeric antibody of claim 24, comprising the amino acid substitution N67Q and / or substitutions at one or more of positions 292, 297 or 302 in the EU numbering of IgG1.

38. A heterodimeric antibody comprising a first monomer having the sequence of SEQ ID NO:19, a second monomer having the sequence of SEQ ID NO:18, and a common light chain having the sequence of SEQ ID NO:

17.

39. A heterodimeric antibody comprising a first monomer comprising the sequence of SEQ ID NO:202, a second monomer comprising the sequence of SEQ ID NO:201, and a common light chain comprising the sequence of SEQ ID NO:

200.

40. A heterodimeric antibody comprising a first monomer comprising the sequence of SEQ ID NO:207, a second monomer comprising the sequence of SEQ ID NO:203, and a common light chain comprising the sequence of SEQ ID NO:

200.

41. 1. A nucleic acid composition comprising: a) a first nucleic acid encoding a first monomer according to any one of claims 22 to 40; b) a second nucleic acid encoding a second monomer according to any one of claims 22 to 40, and c) a third nucleic acid encoding a common light chain according to any one of claims 22 to 40 A nucleic acid composition comprising:

42. 1. A nucleic acid composition comprising: a) a first expression vector comprising a first nucleic acid encoding a first monomer according to any one of claims 22 to 40; b) a second expression vector comprising a second nucleic acid encoding a second monomer according to any one of claims 22 to 40; and c) a third expression vector comprising a third nucleic acid encoding a common light chain according to any one of claims 22 to 40. A nucleic acid composition comprising:

43. A host cell comprising the nucleic acid composition of claim 41 or 42.

44. A pharmaceutical composition comprising a heterodimeric antibody described in any one of claims 22 to 40.

45. 45. The pharmaceutical composition of claim 44 for use in treating cancer in a subject in need thereof.

46. 46. ​​The pharmaceutical composition of claim 45, administered in combination with an anti-PD-1 antigen binding protein.

47. 47. The pharmaceutical composition of claim 46, wherein the anti-PD-1 antigen binding protein comprises a vhCDR1 comprising the amino acid sequence set forth in SEQ ID NO: 189, a vhCDR2 comprising the amino acid sequence set forth in SEQ ID NO: 190, a vhCDR3 comprising the amino acid sequence set forth in SEQ ID NO: 191, a vlCDR1 comprising the amino acid sequence set forth in SEQ ID NO: 192, a vlCDR2 comprising the amino acid sequence set forth in SEQ ID NO: 193, and a vlCDR3 comprising the amino acid sequence set forth in SEQ ID NO:

194.

48. 48. The pharmaceutical composition of claim 47, wherein the anti-PD-1 antigen binding protein comprises a heavy chain variable domain comprising an amino acid sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO:195 and a light chain variable domain comprising an amino acid sequence that is at least 90% identical to the amino acid sequence of SEQ ID NO:

196.

49. 49. The pharmaceutical composition of claim 48, wherein the anti-PD-1 antigen binding protein comprises a heavy chain variable domain comprising the amino acid sequence of SEQ ID NO:195 and a light chain variable domain comprising the amino acid sequence of SEQ ID NO:

196.

50. 50. The pharmaceutical composition of any one of claims 46 to 49, wherein the anti-PD-1 antigen-binding protein is an antigen-binding antibody fragment.

51. 50. The pharmaceutical composition of any one of claims 46 to 49, wherein the anti-PD-1 antigen binding protein is an antibody.

52. The pharmaceutical composition of any one of claims 46 to 49, wherein the anti-PD-1 antigen binding protein is a monoclonal antibody, a chimeric antibody, or a humanized antibody.

53. 53. The pharmaceutical composition of any one of claims 46-49, 51 and 52, wherein the anti-PD-1 antigen binding protein comprises a heavy chain comprising the amino acid sequence of SEQ ID NO:197 and a light chain comprising the amino acid sequence of SEQ ID NO:

198.

54. 53. The pharmaceutical composition of any one of claims 46-49, 51 and 52, wherein the anti-PD-1 antigen binding protein comprises a heavy chain comprising the amino acid sequence of SEQ ID NO:208 and a light chain comprising the amino acid sequence of SEQ ID NO:

209.

55. 47. The pharmaceutical composition of claim 14 or claim 46, wherein the anti-PD-1 antigen binding protein is pembrolizumab.

56. 56. The pharmaceutical composition of any one of claims 14 or 46-55, wherein the anti-PD-1 antigen binding protein is administered prior to the antigen binding protein that binds STEAP1 or the heterodimeric antibody.

57. 56. The pharmaceutical composition of any one of claims 14 or 46-55, wherein the anti-PD-1 antigen binding protein is administered simultaneously with the antigen binding protein that binds STEAP1 or the heterodimeric antibody.

58. 56. The pharmaceutical composition of any one of claims 14 or 46-55, wherein the anti-PD-1 antigen binding protein is administered after the antigen binding protein that binds STEAP1 or the heterodimeric antibody.

59. 46. ​​The pharmaceutical composition of claim 13 or claim 45, for administering an effective amount of an antigen binding protein or heterodimeric antibody that binds to STEAP1 together with an effective amount of abiraterone acetate or enzalutemide.

60. 60. The pharmaceutical composition of any one of claims 13, 14 or 45-59, wherein the antigen binding protein that binds STEAP1 or the heterodimeric antibody is administered subcutaneously.

61. The pharmaceutical composition for use according to any one of claims 45 to 60, wherein the cancer is prostate cancer.

62. The pharmaceutical composition for use according to any one of claims 45 to 60, wherein the cancer is Ewing's sarcoma.

63. 10. The antigen-binding protein of any one of claims 1 to 9, which is linked to a detectable label.

64. 64. The antigen binding protein of claim 63, wherein the label is radioactive.

65. 65. The antigen-binding protein of any one of claims 1 to 6, 9, 63 and 64, which is a Fab fragment.

66. F(ab′) ２ 65. The antigen-binding protein of any one of claims 1 to 6, 9, 63 and 64, which is a fragment.

67. 67. An antigen-binding protein according to any one of claims 63 to 66 for use in detecting STEAP1 in a subject.

68. 68. The antigen binding protein of claim 67, wherein the antigen binding protein is linked to a detectable label, and the label is detected using positron emission tomography.

Images