Dll3 single domain antibodies and therapeutic compositions thereof

Abstract

This invention provides binding peptides that specifically bind to DLL3. More specifically, this invention provides fusion proteins that bind to DLL3, comprising multivalent and / or multispecific constructs and chimeric antigen receptors. Also provided are pharmaceutical compositions containing such peptides; nucleic acid molecules encoding such peptides, as well as their carriers and cells; and methods of use and applications of the provided DLL3-binding peptides for treating diseases and symptoms such as cancer.

Description

Technical Field

[0001] This invention generally provides binding peptides that specifically bind to DLL3. More specifically, this invention relates to fusion proteins that bind at least DLL3, including multivalent and / or multispecific constructs and chimeric antigen receptors. This invention also provides nucleic acid molecules encoding such peptides, as well as their vectors and cells; and methods of use and applications of the provided DLL3-binding peptides for treating diseases and symptoms such as cancer. Prior Technology

[0002] Delta-like ligand 3 (DLL3) is an inhibitory Notch pathway ligand that is highly upregulated and aberrantly expressed on the cell surface of tumors and cancer cells, including small cell lung cancer (SCLC) and highly neuroendocrine tumors. The expression of DLL3 in various human cancers (including solid tumors) makes it a desired therapeutic target. There is a need for modified therapeutic molecules and agents that target DLL3. This article provides examples that meet these needs. Summary of the Invention

[0003] This document provides a DLL3-binding peptide construct comprising at least one variable domain (DLL3 VHH domain) that specifically binds to the heavy chain only of DLL3. In some embodiments, the DLL3-binding construct includes one or more other binding domains that bind to targets other than DLL3.

[0004] This article provides a DLL3-binding peptide construct wherein at least one DLL3 VHH domain comprises: a complementarity-determining region 1 (CDR1) containing an amino acid sequence selected from the group consisting of: SEQ ID NO: 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, and 456; and a complementarity-determining region 2 (CDR2) containing an amino acid sequence selected from the group consisting of: SEQ ID NO: SEQ ID NOs: 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 384, 410, and 411; and complementarity-determining region 3 (CDR3), which contains an amino acid sequence selected from the group consisting of: SEQ ID NOs: 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 395, and 412-415, and binds to DLL3.

[0005] This article provides a DLL3-binding polypeptide construct comprising at least one variable domain (DLL3 VHH domain) that specifically binds to the heavy chain only of DLL3. This variable domain comprises: a complementarity-determining region 1 (CDR1) containing an amino acid sequence selected from the group consisting of: SEQ ID NO: 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, and 456; and a complementarity-determining region 2 (CDR2) containing an amino acid sequence selected from the group consisting of: SEQ ID NO: SEQ ID NOs: 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 384, 410, and 411; and complementarity-determining region 3 (CDR3), which contains an amino acid sequence selected from the group consisting of: SEQ ID NOs: 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 395, and 412-415, and binds to DLL3.

[0006] In any of the provided embodiments, at least one DLL3 VHH domain comprises: a complementarity-determining region 1 (CDR1) containing an amino acid sequence selected from the group consisting of: SEQ ID NO: 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, and 456; a complementarity-determining region 2 (CDR2) containing an amino acid sequence selected from the group consisting of: SEQ ID NO: 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, and 353; and a complementarity-determining region 3 (CDR3) containing an amino acid sequence selected from the group consisting of: SEQ ID NO: 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, and 456; and a complementarity-determining region 3 (CDR3) containing an amino acid sequence selected from the group consisting of: SEQ ID NO: 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, and 453; and a complementarity NO:354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366 and 367, and combined with DLL3.

[0007] In any of the embodiments provided, DLL3 is human DLL3. In some embodiments, DLL3 has the sequence shown in SEQ ID NO:86 or its mature form, which lacks a signal sequence. In some embodiments, DLL3 has the sequence shown in SEQ ID NO:87 or its mature form, which lacks a signal sequence.

[0008] In some embodiments, at least one DLL3 VHH field is humanized. In some embodiments, DLL3 VHH is a Camelidae VHH. In some embodiments, DLL3 VHH is a humanized form of a Camelidae VHH.

[0009] In any of the embodiments provided, one or more additional binding domains bind to an activating receptor on an immune cell. In some embodiments, the immune cell is a T cell. In any of the embodiments provided, the activating receptor is CD3 (CD3ε). In some instances, the DLL3-binding polypeptide construct of the embodiments has bispecificity against DLL3 and CD3. In some embodiments, the immune cell is a natural killer (NK) cell.

[0010] In some of the embodiments provided, the activating receptor is CD16 (CD16a). In some instances, the DLL3-binding peptide construct has bispecificity against both DLL3 and CD16a.

[0011] In any of the embodiments provided, one or more additional binding domains bind to the interleukin receptor.

[0012] In any of the provided embodiments, one or more additional binding domains comprise an antibody or an antigen-binding fragment thereof. In some embodiments, the one or more additional binding domains are monovalent. In some embodiments, the antibody or antigen-binding fragment thereof is an Fv, a disulfide-stabilized Fv (dsFv), scFv, Fab, a single-domain antibody (sdAb), a VNAR, or a VHH. In some embodiments, the single-domain antibody (sdAb) is a camelid VHH. In some embodiments, the single-domain antibody (sdAb) is a humanized form of a camelid VHH.

[0013] In any of the provided embodiments, one or more additional binding domains are interferons or truncated fragments or variants thereof, capable of binding to interferon receptors. In some embodiments, the interferon is an interferon, or a truncated fragment or variant of an interferon. In some embodiments, the interferon is a type I interferon or a type II interferon; a truncated fragment or variant of a type I interferon; or a truncated fragment or variant of a type II interferon. In some embodiments, a type I interferon is IFN-α or IFN-β or a truncated fragment or variant thereof; or a type II interferon is IFN-γ or a truncated fragment or variant thereof.

[0014] In any of the provided embodiments, the polypeptide comprises an immunoglobulin Fc region. In some embodiments, the polypeptide comprises an immunoglobulin Fc region linking at least one VHH domain to one or more other binding domains. In some embodiments, the DLL3-binding polypeptide construct is a dimer. In some embodiments, the Fc region is a homodimeric Fc region.

[0015] In any of the provided embodiments, the Fc region comprises an amino acid sequence shown in any of SEQ ID NO: 8, 10, 11, 12, or 13, or an amino acid sequence showing at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 8, 10, 11, 12, or 13. In some embodiments, the Fc region is human IgG1.

[0016] In any of the embodiments provided, the DLL3-binding peptide construct is a dimer. In some embodiments, the Fc region is a homodimeric Fc region.

[0017] In any of the embodiments provided, the Fc region is human IgG1.

[0018] In any of the embodiments provided, the Fc region contains the amino acid sequence shown in SEQ ID NO:8 or an amino acid sequence that is at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the sequence in SEQ ID NO:8.

[0019] In some embodiments, the Fc region is a heterodimeric Fc region. In some embodiments, the Fc region exhibits effector function. In some embodiments, the Fc region comprises a polypeptide containing one or more amino acid modifications that reduce effector function and / or reduce binding to effector molecules selected from Fcγ receptors or C1q. In some embodiments, the one or more amino acid modifications are the absence of one or more of Glu233, Leu234, or Leu235.

[0020] In any of the embodiments provided, the Fc region contains the amino acid sequence shown in SEQ ID NO:9 or an amino acid sequence that is at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the sequence in SEQ ID NO:9.

[0021] In any of the embodiments provided, at least one DLL3 VHH domain comprises a VHH domain sequence shown in any of SEQ ID NO:244-318 and 455 or a VHH domain sequence shown in an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO:244-318 and 455, and is bound to DLL3.

[0022] In any of the embodiments provided, at least one DLL3 VHH domain comprises the VHH domain sequence shown in any of SEQ ID NO: 102, 244-318, 401-409, 416, 455, 476-480-488, and 507-518, or the VHH domain sequence shown in an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 102, 244-318, 401-409, 416, 455, 476-480-488, and 507-518, and is bound to DLL3.

[0023] In any of the embodiments provided, at least one DLL3 VHH domain comprises a VHH domain sequence shown in any of SEQ ID NO: 102, 244-275, 277-300, 302-305, 314, 401, 416, 455, 476-480-488, and 507-518, or a VHH domain sequence shown in an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence consistency with any of SEQ ID NO: 102, 244-275, 277-300, 302-305, 314, 401, 416, 455, 476-480-488, and 507-518, and is bound to DLL3.

[0024] In any of the provided embodiments, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO:244, (ii) a humanized variant of SEQ ID NO:244, or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO:244, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises: CDR1, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO:319, 320, 321, 322, 323, 324, 325, and 326; CDR2, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO:336, 337, and 338; and CDR3, which comprises the amino acid sequence shown in SEQ ID NO:354; and is bound to DLL3. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 319, 336, and 354, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 319, 337, and 354, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 319, 338, and 354, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 338, and 354, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 321, 338, and 354, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 322, 338, and 354, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO:323, 338, and 354, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO:324, 338, and 354, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO:325, 338, and 354, respectively.In some embodiments, at least one DLL3 VHH field comprises CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 326, 338, and 354, respectively. In any of the above embodiments, at least one DLL3 VHH field is bound to DLL3. In some embodiments, at least one DLL3 VHH field comprises an amino acid sequence shown in any of SEQ ID NO: 245-257 or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 245-257, and is bound to DLL3. In some embodiments, at least one DLL3 VHH field comprises an amino acid sequence shown in any of SEQ ID NO: 245-257, and is bound to DLL3.

[0025] In any of the provided embodiments, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO:258, (ii) a humanized variant of SEQ ID NO:258, or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO:258, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises: CDR1, which comprises the amino acid sequence shown in SEQ ID NO:327; CDR2, which comprises the amino acid sequence shown in SEQ ID NO:339; and CDR3, which comprises the amino acid sequence shown in SEQ ID NO:355; and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises CDR1, CDR2, and CDR3 shown in SEQ ID NO:327, 339, and 355, respectively, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 259-263 or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 259-263, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 259-263, and is bound to DLL3.

[0026] In any of the provided embodiments, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO:264, (ii) a humanized variant of SEQ ID NO:264, or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO:264, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises: CDR1, which comprises the amino acid sequence shown in SEQ ID NO:328, 329, or 456; CDR2, which comprises the amino acid sequence shown in SEQ ID NO:340; and CDR3, which comprises the amino acid sequence shown in SEQ ID NO:356; and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises CDR1, CDR2, and CDR3 shown in SEQ ID NO:328, 340, and 356, respectively. In some embodiments, at least one DLL3 VHH field comprises CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 329, 340, and 356, respectively. In some embodiments, at least one DLL3 VHH field comprises CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 456, 340, and 356, respectively. In any of the above embodiments, at least one DLL3 VHH field is bound to DLL3. In some embodiments, at least one DLL3 VHH field comprises an amino acid sequence shown in any of SEQ ID NO: 265-274, 416, or 455, or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 265-274, 416, or 455, and is bound to DLL3. In some embodiments, at least one DLL3 VHH field contains an amino acid sequence shown in any of SEQ ID NO: 265-274, 416, or 455, and is bound to DLL3. In some embodiments, at least one DLL3 VHH field contains an amino acid sequence shown in any of SEQ ID NO: 265-274, 416, 455, or 476-478, or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 265-274, 416, 455, or 476-478, and is bound to DLL3.In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 265-274, 416, 455, or 476-478, and binds to DLL3.

[0027] In any of the provided embodiments, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO:275, (ii) a humanized variant of SEQ ID NO:275, or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO:275, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises: CDR1, which comprises the amino acid sequence shown in SEQ ID NO:320; CDR2, which comprises the amino acid sequence shown in SEQ ID NO:341; and CDR3, which comprises the amino acid sequence shown in SEQ ID NO:357; and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 276-279 or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 276-279, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 276-279, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 277-279 and 479 or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 277-279 and 479, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 277-279 and 479, and is bound to DLL3.

[0028] In any of the provided embodiments, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO:280, (ii) a humanized variant of SEQ ID NO:280, or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO:280, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises: CDR1, which comprises the amino acid sequence shown in SEQ ID NO:330; CDR2, which comprises the amino acid sequence shown in SEQ ID NO:342; and CDR3, which comprises the amino acid sequence shown in SEQ ID NO:358; and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO:281-286 or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO:281-286, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO:281-286, and is bound to DLL3.

[0029] In any of the provided embodiments, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO:287, (ii) a humanized variant of SEQ ID NO:287, or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO:287, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises: CDR1, which comprises the amino acid sequence shown in SEQ ID NO:320; CDR2, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO:345, 346, and 347; and CDR3, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO:359, 360, and 361; and is bound to DLL3. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 345, and 359, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 346, and 359, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 347, and 359, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 345, and 360, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 345, and 361, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 347, and 360, respectively. In some embodiments, any of the provided DLL3 VHH fields are bound to DLL3. In some embodiments, at least one DLL3 VHH field contains an amino acid sequence shown in any of SEQ ID NO: 288-298 or 102, or an amino acid sequence showing at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 288-298 or 102, and is bound to DLL3. In some embodiments, at least one DLL3 VHH field contains an amino acid sequence shown in any of SEQ ID NO: 288-298 or 102, and is bound to DLL3.

[0030] In any of the provided embodiments, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO:299, (ii) a humanized variant of SEQ ID NO:299, or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO:299, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises: CDR1, which comprises the amino acid sequence shown in SEQ ID NO:331; CDR2, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO:348, 349, and 350; and CDR3, which comprises the amino acid sequence shown in SEQ ID NO:356; and is bound to DLL3. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 331, 348, and 356, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 331, 349, and 356, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 331, 350, and 356, respectively. In some embodiments, any of the above-described DLL3 VHH fields are combined with DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 300-305 or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 300-305, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 300-305, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 300, 302-305 and 480 or an amino acid sequence showing at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity with any of SEQ ID NO: 300, 302-305 and 480, and is bound to DLL3.In some embodiments, at least one DLL3 VHH domain comprises the amino acid sequence shown in any of SEQ ID NO: 300, 302-305 and 480, and binds to DLL3.

[0031] In any of the provided embodiments, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO: 507, (ii) a humanized variant of SEQ ID NO: 507, or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO: 507, and is bound to DLL3. In any of the provided embodiments, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO: 306, (ii) a humanized variant of SEQ ID NO: 306, or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO: 306, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises: CDR1, which comprises the amino acid sequence shown in SEQ ID NO: 332; CDR2, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 348, 349, and 350; and CDR3, which comprises the amino acid sequence shown in SEQ ID NO: 362; and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises CDR1, CDR2, and CDR3 shown in SEQ ID NO: 332, 348, and 362, respectively. In some embodiments, at least one DLL3 VHH domain comprises CDR1, CDR2, and CDR3 shown in SEQ ID NO: 332, 349, and 362, respectively. In some embodiments, at least one DLL3 VHH domain comprises CDR1, CDR2, and CDR3 shown in SEQ ID NO: 332, 350, and 362, respectively. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 300-305 or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 307-313, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 307-313, and is bound to DLL3.In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 508-514 or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 508-514, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 508-514, and is bound to DLL3.

[0032] In any of the provided embodiments, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO: 401, (ii) a humanized variant of SEQ ID NO: 401, or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO: 401, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain comprises: CDR1, which comprises the amino acid sequence shown in SEQ ID NO: 320; CDR2, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 384, 410, and 411; and CDR3, which comprises an amino acid sequence selected from the group consisting of SEQ ID NO: 395, 412, 413, 414, and 415; and is bound to DLL3. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 384, and 395, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 410, and 395, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 384, and 412, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 384, and 413, respectively. In some embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 320, 384, and 414, respectively. In some embodiments, at least one DLL3 VHH field comprises CDR1, CDR2, and CDR3 as shown in SEQ ID NO:320, 384, and 415, respectively, and is bound to DLL3. In some embodiments, at least one DLL3 VHH field comprises an amino acid sequence as shown in any of SEQ ID NO:402-409 or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO:402-409, and is bound to DLL3.In some embodiments, at least one DLL3 VHH field contains an amino acid sequence shown in any of SEQ ID NO: 402-409 and binds to DLL3. In some embodiments, at least one DLL3 VHH field contains an amino acid sequence shown in any of SEQ ID NO: 481-488 or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 481-488 and binds to DLL3. In some embodiments, at least one DLL3 VHH field contains an amino acid sequence shown in any of SEQ ID NO: 481-488 and binds to DLL3.

[0033] In any of the provided embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in SEQ ID NO: 333, 351, and 363, respectively. In any of the provided embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in 334, 352, and 364, respectively. In any of the provided embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in 320, 353, and 365, respectively. In any of the provided embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in 334, 339, and 366, respectively. In any of the provided embodiments, at least one DLL3 VHH field includes CDR1, CDR2, and CDR3 as shown in 335, 348, and 367, respectively. In any of the provided embodiments, the DLL3 VHH field is combined with DDL3. In some embodiments, at least one DLL3 VHH field comprises a sequence shown below: (i) SEQ ID NO: 314, 315, 316, 317 or 318; (ii) a humanized variant of SEQ ID NO: 314, 315, 316, 317 or 318; or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity with SEQ ID NO: 314, 315, 316, 317 or 318, and is bound to DLL3. In some embodiments, at least one DLL3 VHH field is defined in SEQ ID NO: 314, 315, 316, 317 or 318 and is bound to DLL3. In any of the provided embodiments, at least one DLL3 VHH domain comprises a sequence shown below: (i) SEQ ID NO: 314, 518, 515, 516 or 517; (ii) a humanized variant of SEQ ID NO: 314, 518, 515, 516 or 517; or (iii) an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity with SEQ ID NO: 314, 518, 515, 516 or 517, and is bound to DLL3. In some embodiments, at least one DLL3 VHH domain is expressed in SEQ ID NO: 314, 518, 515, 516 or 517 and is bound to DLL3.

[0034] In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 244. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 251. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 264. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 268. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 275. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 287. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 299. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 507. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 314. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 518. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 515. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 516. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 517. In some embodiments, the DLL3 VHH field comprises the VHH field sequence shown in SEQ ID NO: 455.

[0035] In any of the embodiments provided, the DLL3 VHH domain may contain other amino acids at its N-terminus and / or C-terminus, such as other amino acids for linking to another amino acid sequence (such as another polypeptide). In any of the embodiments provided, the DLL3 VHH domain may contain flexible linkers, such as glycine linkers or linkers consisting primarily of the amino acids glycine and serine, denoted herein as GS linkers. Such linkers of the present invention may have various lengths, such as 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids. In some embodiments, the linker comprises an amino acid sequence selected from the group consisting of: GGSGGS, i.e. (GGS)2 (SEQ ID NO:1); GGSGGSGGS, i.e. (GGS)3 (SEQ ID NO:2); GGSGGSGGSGGS, i.e. (GGS)4 (SEQ ID NO:3); and GGSGGSGGSGGSGGS, i.e. (GGS)5 (SEQ ID NO:4); Gly-Gly (GG), GGG, GGGG (SEQ ID NO:5); GGGGG (SEQ ID NO:6) and GGGGGG (SEQ ID NO:7). In some embodiments, the linker is (GGGGS)n, where n is 1 to 5 (SEQ ID NO: 123); (GGGGGS)n, where n is 1 to 4 (SEQ ID NO: 124); GGGGS (SEQ ID NO: 125); GGGGGS (SEQ ID NO: 126); GGGGGSGGGGGSGGGGS (SEQ ID NO: 127); GGGGSGGGGSGGGGS (SEQ ID NO: 128); GGSGGGGSGGGGSGGGGS (SEQ ID NO: 129); or PGGGG (SEQ ID NO: 450). In some embodiments, the linker is a GG linker. In some embodiments, the DLL3-binding polypeptide comprises a combination of a GS linker and a glycine linker. In some embodiments, the DLL3 VHH domain may contain other linkers at its C-terminus, such as linkers for linking to another amino acid sequence (such as another polypeptide). In any of the embodiments provided, the DLL3 VHH domain may contain a linker at its N-terminus, such as a linker for linking to another amino acid sequence (such as another polypeptide).

[0036] In any of the embodiments provided, at least one DLL3 VHH domain comprises the sequence shown below: (i) SEQ ID NO:244, (ii) a humanized variant of SEQ ID NO:244, or (iii) an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with SEQ ID NO:244, and is bound to DLL3.

[0037] This document provides a multispecific polypeptide construct comprising: (a) a first component comprising a heterodimeric Fc region containing a first Fc polypeptide and a second Fc polypeptide; and (b) a second component comprising an anti-CD3 antibody or antigen-binding fragment containing a heavy chain variable region (VH) and a light chain variable region (VL), wherein the VH and VL constituting the anti-CD3 antibody or antigen-binding fragment are linked to the corresponding polypeptide of the heterodimeric Fc region, wherein the first and second components are coupled by a linker, wherein the heterodimeric Fc region is located at the N-terminus of the anti-CD3 antibody; and wherein one or both of the first and second components comprise at least one antigen-binding domain containing a VHH domain specifically binding to DLL3 (DLL3 VHH domain). In a particular embodiment, the DLL3 VHH domain may include any DLL3 VHH domain sequence provided, including any as described above or elsewhere herein.

[0038] In some embodiments, the multispecific polypeptide construct comprises at least: (i) a first polypeptide comprising a first Fc polypeptide of a heterodimeric Fc region, a linker, and a VH or VL domain of an anti-CD3 antibody or antigen-binding fragment; and (ii) a second polypeptide comprising a second Fc polypeptide of a heterodimeric Fc region, a linker (which may be the same as the linker present in the first polypeptide), and the other of a VH or VL domain of an anti-CD3 antibody or antigen-binding fragment, wherein one or both of the first and second polypeptides comprises at least one DLL3 VHH domain.

[0039] In some embodiments, compared to the homodimeric Fc region peptide, and as appropriate compared to the Fc peptide shown in SEQ ID NO: 8 or its immunoactive fragment, one or both of the first and second Fc peptides of the heterodimeric Fc region contain at least one modification that induces heterodimerization. In some embodiments, each of the first and second Fc peptides of the heterodimeric Fc region independently contains at least one amino acid modification. In some embodiments, each of the first and second Fc peptides of the heterodimeric Fc region contains a mordant modification or a charge mutation that enhances the electrostatic complementarity of the peptide.

[0040] In some embodiments, the amino acid modification is a mortar and pestle modification. In one embodiment, the first Fc polypeptide of the heterodimeric Fc region comprises a modification selected from Thr366Ser, Leu368Ala, Tyr407Val, and combinations thereof, and the second Fc polypeptide of the heterodimeric Fc region comprises a modification of Thr366Trp. In such embodiments, the first and second Fc polypeptides may further comprise a modification of non-cysteine ​​residues to cysteine ​​residues, wherein the modification of the first Fc polypeptide is present at one of Ser354 and Tyr349, and the modification of the second Fc polypeptide is present at the other of Ser354 and Tyr349.

[0041] In some embodiments, the amino acid modification is a charge mutation that enhances the electrostatic complementarity of the peptide. In some embodiments, the first and / or the second Fc peptide, or each of the first and second Fc peptides, includes a modification located at a complementary position, wherein the modification is an amino acid substitution with an amino acid having a charge opposite to that of the complementary amino acid of the other peptide.

[0042] In any of the provided embodiments, one of the first or second Fc polypeptides in the heterodimeric Fc region further comprises a modification at residue Ile253. In some embodiments, the modification is Ile253Arg. In some embodiments, one of the first or second Fc polypeptides in the heterodimeric Fc region further comprises a modification at residue His435. In some embodiments, the modification is His435Arg.

[0043] In some embodiments, the Fc region of any of the provided polypeptides or constructs contains a polypeptide lacking Lys447.

[0044] In some embodiments, the Fc region of any provided polypeptide or construct includes at least one modification that enhances FcRn binding. In some embodiments, the modification is present at a position selected from the group consisting of: Met252, Ser254, Thr256, Met428, Asn434, and combinations thereof. In some embodiments, the modification is selected from the group consisting of: Met252Y, Ser254T, Thr256E, Met428L, Met428V, Asn434S, and combinations thereof. In some embodiments, the modification is present at positions Met252 and Met428. In some embodiments, the modification is Met252Y and Met428L. In some embodiments, the modification is Met252Y and Met428V.

[0045] In any of the embodiments provided, the first Fc polypeptide of the heterodimer Fc region comprises the amino acid sequence shown in any of SEQ ID NO: 103, 107, 115 or 117, and the second Fc polypeptide of the heterodimer Fc region comprises the amino acid sequence shown in any of SEQ ID NO: 104, 108, 111, 113, 119 or 121.

[0046] In any of the provided embodiments, the Fc region of the provided polypeptide or construct comprises a polypeptide containing at least one amino acid modification that reduces effector function and / or reduces binding to effector molecules selected from Fcγ receptors or C1q. In some embodiments, one or more amino acid modifications are performed by the absence of one or more of Glu233, Leu234, or Leu235.

[0047] In any of the embodiments provided, the first Fc polypeptide of the heterodimer Fc region comprises the amino acid sequence shown in any of SEQ ID NO: 105, 109, 116 or 118, and the second Fc polypeptide of the heterodimer Fc region comprises the amino acid sequence shown in any of SEQ ID NO: 106, 110, 112, 114, 120 or 122.

[0048] In any of the embodiments provided, the anti-CD3 antibody or antigen-binding fragment is monovalent. In some embodiments, the anti-CD3 antibody or antigen-binding fragment is an Fv antibody fragment. In some embodiments, the Fv antibody fragment is a disulfide-stabilized anti-CD3 binding Fv fragment (dsFv).

[0049] In some embodiments, the anti-CD3 antibody or antigen-binding fragment is not a single-chain antibody, for example, it is not a single-chain variable fragment (scFv).

[0050] In some embodiments, the anti-CD3 antibody or antigen-binding fragment comprises VH CDR1, which contains the amino acid sequence TYAMN (SEQ ID NO: 29); VH CDR2, which contains the amino acid sequence RIRSKYNNYATYYADSVKD (SEQ ID NO: 30); VH CDR3, which contains the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); VL CDR1, which contains the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO: 32); VL CDR2, which contains the amino acid sequence GTNKRAP (SEQ ID NO: 33); and VL CDR3, which contains the amino acid sequence ALWYSNLWV (SEQ ID NO: 34). In some embodiments, the anti-CD3 antibody or antigen-binding fragment comprises: VH, having an amino acid sequence of any one of SEQ ID NO: 35-65 or a sequence exhibiting at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any one of SEQ ID NO: 35-65; and VL, having an amino acid sequence of any one of SEQ ID NO: 66-84 and 368 or a sequence exhibiting at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any one of SEQ ID NO: 66-84 and 368. In some embodiments, the anti-CD3 antibody or antigen-binding fragment comprises the amino acid sequence of SEQ ID NO: 47 and the amino acid sequence of SEQ ID NO: 75. In some embodiments, the anti-CD3 antibody or antigen-binding fragment comprises the amino acid sequence of SEQ ID NO: 47 and the amino acid sequence of SEQ ID NO: 368.

[0051] In some embodiments, the VL of the anti-CD3 antibody or antigen-binding fragment is linked to the first Fc polypeptide of the heterodimer Fc, and the VH of the anti-CD3 antibody or antigen-binding fragment is linked to the second Fc polypeptide of the heterodimer Fc.

[0052] In some embodiments, at least one DLL3 VHH domain multispecific polypeptide construct is located at the amino terminus relative to the Fc region and / or at the carboxyl terminus relative to the CD3 binding region of the multispecific polypeptide construct.

[0053] In some embodiments, the multispecific peptide construct includes a first DLL3 VHH domain that specifically binds to DLL3 and a second DLL3 VHH domain that specifically binds to DLL3. In certain embodiments, the first and second DLL3 VHH domains may independently include any provided DLL3 VHH domain sequence, including any as described above or elsewhere herein. In some embodiments, the first and second DLL3 VHH domains are identical. In some embodiments, the first and second DLL3 VHH domains are different. In some embodiments, the first and second DLL3 VHH domains bind different or non-overlapping antigenic determinants of DLL3 and / or do not compete for binding to DLL3.

[0054] In some embodiments, the first or second DLL3 VHH domain is located at the amino terminus relative to the Fc region of the multispecific construct, and the other of the first or second DLL3 VHH domains is located at the carboxyl terminus relative to the CD3 binding region of the multispecific construct.

[0055] In any of the embodiments of the provided multispecific polypeptide constructs, the first component comprises, in order from N-terminus to C-terminus, a first DLL3 VHH domain binding DLL3, a first Fc polypeptide of the heterodimeric Fc region, a linker, a VH or VL domain of an anti-CD3 antibody or antigen-binding fragment, and a second DLL3 VHH domain binding DLL3; and the second polypeptide comprises, in order from N-terminus to C-terminus, a second Fc polypeptide of the heterodimeric Fc region, a linker (which may be the same as the linker present in the first component), and the other of the VH or VL domains of an anti-CD3 antibody or antigen-binding fragment.

[0056] In some embodiments, one or both of the first and second components include at least one costimulatory receptor binding region (CRBR) that binds to a costimulatory receptor. In some embodiments, at least one costimulatory receptor binding region (CRBR) is located at the amino terminus relative to the Fc region of the multispecific polypeptide construct and / or at the carboxyl terminus relative to the CD3 binding region of the multispecific polypeptide construct. In some embodiments, the multispecific polypeptide construct includes only one costimulatory receptor binding region (CRBR).

[0057] In any of the embodiments of the provided multispecific peptide constructs, the first component comprises, in order from N-terminus to C-terminus, a first DLL3 VHH domain for binding DLL3; a first Fc peptide of the heterodimeric Fc region, a linker, a VH or VL domain of an anti-CD3 antibody or antigen-binding fragment, and a second DLL3 VHH domain for binding DLL3; and the second component comprises a CRBR and, in order from N-terminus to C-terminus, another of the following: a second Fc peptide of the heterodimeric Fc region, a linker (which may be the same as the linker present in the first component), and a VH or VL domain of an anti-CD3 antibody or antigen-binding fragment, wherein the CRBR is located at the amino terminus relative to the Fc region or at the carboxyl terminus relative to the anti-CD3 antibody or antigen-binding fragment of the second component.

[0058] In some embodiments, at least one co-stimulatory receptor binding region (CRBR) is or comprises an extracellular domain of a native homologous binding complex of the co-stimulatory receptor or a binding fragment thereof, or a variant thereof, exhibiting binding activity to the co-stimulatory receptor. In some embodiments, at least one co-stimulatory receptor binding region (CRBR) is an antibody or an antigen-binding fragment thereof selected from the group consisting of: Fab fragment, F(ab')2 fragment, Fv fragment, scFv, scAb, dAb, single-domain heavy chain antibody, and single-domain light chain antibody. In some embodiments, the antibody or its antigen-binding fragment is Fv, scFv, Fab, a single-domain antibody (VHH domain), VNAR, or VHH. In some embodiments, the antibody or antigen-binding fragment is a VHH domain. In some embodiments, the VHH domain is a human or humanized VHH domain.

[0059] In any of the provided embodiments, at least one co-stimulatory receptor binding region (CRBR) binds to a co-stimulatory receptor selected from the following: 41BB (CD137), OX40 (CD134), CD27, glucocorticoid-induced TNFR-related protein (GITR), CD28, ICOS, CD40, B cell activating factor receptor (BAFF-R), B cell maturation antigen (BCMA), transmembrane activating factor and CAML interactor (TACI), and NKG2D. In some embodiments, at least one co-stimulatory receptor binding region (CRBR) binds to a co-stimulatory receptor selected from the following: 41BB (CD137), OX40 (CD134), and glucocorticoid-induced TNFR-related protein (GITR).

[0060] In some embodiments, at least one co-stimulatory receptor binding region (CRBR) comprises the amino acid sequence shown in SEQ ID NO:210 or a sequence having at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the sequence shown in SEQ ID NO:210, and binds to 4-1BB.

[0061] In some embodiments, at least one co-stimulatory receptor binding region (CRBR) contains the amino acid sequence shown in SEQ ID NO:210 and binds 4-1BB.

[0062] In some embodiments, at least one co-stimulatory receptor binding region (CRBR) contains the amino acid sequence shown in SEQ ID NO:470 and binds 4-1BB.

[0063] In some embodiments, one or both of the first and second components include at least one inhibitory receptor binding region (IRBR) that binds an inhibitory receptor. In some embodiments, at least one inhibitory receptor binding region (IRBR) is located at the amino terminus relative to the Fc region of the multispecific polypeptide construct and / or at the carboxyl terminus relative to the CD3 binding region of the multispecific polypeptide construct. In some embodiments, the multispecific polypeptide construct includes only one inhibitory receptor binding region (IRBR).

[0064] In any of the embodiments of the provided multispecific polypeptide constructs, the first component comprises, in order from N-terminus to C-terminus: a first DLL3 VHH domain binding DLL3, a first Fc polypeptide of the heterodimeric Fc region, a linker, a VH or VL domain of an anti-CD3 antibody or antigen-binding fragment, and a second DLL3 VHH domain binding DLL3; and the second component comprises an IRBR and, in order from N-terminus to C-terminus: a second Fc polypeptide of the heterodimeric Fc region, a linker (which may be the same as the linker present in the first component), and another of the VH or VL domains of an anti-CD3 antibody or antigen-binding fragment, wherein the IRBR is located at the amino terminus relative to the Fc region or at the carboxyl terminus relative to the anti-CD3 antibody or antigen-binding fragment of the second component.

[0065] In some embodiments, at least one IRBR is or comprises the extracellular domain of a native homologous binding complex of an inhibitory receptor or a binding fragment thereof, or a variant thereof, exhibiting binding activity to the inhibitory receptor. In some embodiments, at least one IRBR is an antibody or an antigen-binding fragment thereof selected from the group consisting of: Fab fragment, F(ab')2 fragment, Fv fragment, scFv, scAb, dAb, single-domain heavy chain antibody, and single-domain light chain antibody. In some embodiments, the antibody or its antigen-binding fragment is Fv, scFv, Fab, a single-domain antibody (VHH domain), VNAR, or VHH. In some embodiments, the antibody or antigen-binding fragment is a VHH domain. In some embodiments, the VHH domain is a human or humanized VHH domain. In some embodiments, at least one IRBR binds to an inhibitory receptor selected from: PD-1, CTLA-4, TIGIT, VISTA, and TIM3. In some embodiments, at least one IRBR binds to PD-1.

[0066] In any of the embodiments provided with the multispecific peptide construct, the first component comprises, in order from N-terminus to C-terminus: a first DLL3 VHH domain binding DLL3, a first Fc peptide of the heterodimeric Fc region, a linker, a VH or VL domain of an anti-CD3 antibody or antigen-binding fragment, and a second DLL3 VHH binding DLL3; and the second component comprises, in order from N-terminus to C-terminus: one of an IRBR or a CRBR, a second Fc peptide of the heterodimeric Fc region, a linker (which may be the same as the linker present in the first component), the other of the VH or VL domain of the anti-CD3 antibody or antigen-binding fragment, and the other of the CRBR or IRBR.

[0067] In any of the embodiments provided, the linker is a peptide or polypeptide linker. In some embodiments, the linker has a length of 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids.

[0068] In some embodiments, the linker is a non-disintegrable linker, such as a linker comprising GS, GGS, GGGGS (SEQ ID NO: 125), GGGGGS (SEQ ID NO: 126), and combinations thereof. In some embodiments, the linker is or comprises the sequence GGGGGSGGGGGSGGGGGS (SEQ ID NO: 127).

[0069] In some embodiments, the linker is a cleavable linker, such as a polypeptide that acts as a protease acceptor. In some embodiments, the protease is produced by immune effector cells, tumors, or cells present in the tumor microenvironment. In some embodiments, the protease is produced by immune effector cells, and the immune effector cells are activated T cells, natural killer (NK) cells, or NK T cells. In some embodiments, the protease is an interstitial protease, matrix metalloproteinase (MMP), granzyme B, or a combination thereof. In some embodiments, the cleavable linker comprises the amino acid sequence GGSGGGG. [IEPD] IGGSGGS (SEQ ID NO: 171).

[0070] This article provides an isolated single-domain antibody that binds to DLL3 and contains any DLL3 VHH domain sequence provided herein, including any as described above or elsewhere herein.

[0071] This article provides a single-domain antibody that binds to DLL3, comprising: a complementarity-determining region 1 (CDR1) containing an amino acid sequence selected from the group consisting of; SEQ ID NO: 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335 and 456; and a complementarity-determining region 2 (CDR2) containing an amino acid sequence selected from the group consisting of: SEQ ID NO: SEQ ID NOs: 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 384, 410, and 411; and complementarity-determining region 3 (CDR3), which contains amino acid sequences selected from the group consisting of: SEQ ID NOs: 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 395, and 412-415.

[0072] This article provides polynucleotides encoding any DLL3-binding polypeptide constructs provided herein.

[0073] This document also provides polynucleotides encoding any of the multispecific polypeptide constructs provided herein or their first or second polypeptides. In some embodiments, the polynucleotide includes a first nucleic acid encoding a first polypeptide of the provided multispecific polypeptide construct and a second nucleic acid encoding a second polypeptide of the provided multispecific polypeptide construct, wherein the first and second nucleic acid sequences are separated by an internal ribosome entry site (IRES); or a nucleic acid encoding a cleavage peptide or a peptide that causes ribosome jumping (such as T2A, P2A, E2A, or F2A).

[0074] It also provides polynucleotides encoding any single-domain antibody provided.

[0075] It also provides a carrier for encoding any polynucleotide provided, such as an expression carrier.

[0076] This article provides a cell comprising any one or more of the provided polynucleotides or any one or more of the provided carriers.

[0077] This document provides a method for preparing peptides, comprising introducing one or more provided polynucleotides or any of one or more carriers into cells and culturing the cells under conditions that produce multispecific peptide constructs. This document provides peptides produced by any of the methods provided herein.

[0078] This article provides an engineered immune cell comprising a chimeric antigen receptor comprising an extracellular domain containing a single-domain antibody with any provided DLL3 VHH domain sequence; a transmembrane domain; and an intracellular signal transduction domain.

[0079] This article provides a pharmaceutical composition comprising any one of the provided DLL3-binding peptide, multispecific peptide construct, single-domain antibody, or engineered immune cell.

[0080] This article provides a method for stimulating or inducing an immune response in an individual, the method comprising administering to an individual in need any of the following, or a pharmaceutical composition: a provided DLL3-binding peptide, a multispecific peptide construct, a single-domain antibody, or engineered immune cells.

[0081] This article also provides a method for treating an individual’s disease or condition, the method comprising administering to an individual in need a therapeutically effective amount of any of the DLL3-binding peptides described herein, any of the multispecific peptide constructs described herein, any of the single-domain antibodies described herein, any of the engineered immune cells described herein, or any of the pharmaceutical compositions described herein. Simple Explanation of the Diagram

[0082] [Figure 1] illustrates the ability of various anti-DLL3 single-domain antibodies (sdAbs) to bind to DLL3 on the cell surface. Binding was assessed in the DLL3-positive cell line SHP-77 by flow cytometry. In this study, DLL3 sdAbs 10D9, 10E5, 8E7, 5A7, 5A8, 5H8, 3G3, 6C5, 6F1, 3B4, 3B12, or 6B4 were operatively linked to human Fc cells.

[0083] [Figures 2A-L] illustrate the ability of sdAbs targeting DLL3 and their humanized variants to bind to DLL3 on the cell surface. Binding was assessed by flow cytometry for DLL3-positive cell line SHP-77 or HEK-293 free cells transiently transfected with a vector encoding DLL3. [Figure 2A] shows the binding of 3G3 and its humanized variants to SHP-77. [Figure 2B] shows the binding of 5A7 and its humanized variants to SHP-77. [Figure 2C] shows the binding of 3C5 and its humanized variants to SHP-77. [Figure 2D] shows the binding of 6C5 and its humanized variants to SHP-77. [Figures 2E-G] illustrate the effects of 5A8 and its humanized variants on SHP-77 ( [Figure 2E] and [2F]) and 293FS(FL) transfected by DLL3 ( [Figure 2G]) combination. [Figure 2H] shows the binding of 10D9 and its humanized variants to SHP-77. [Figure 2I] shows the binding of 10E5 and its humanized variants to SHP-77. [Figures 2J and 2K] show the binding of 8E7 and its humanized variants to SHP-77 and DLL3-transfected 293FS cells, respectively. [Figure 2L] illustrates the binding of 6F1 and its humanized variants to SHP-77. In this paper, DLL3 sdAbs are operatively linked to human Fc.

[0084] [Figures 3A-3E] depict a series of schematic diagrams illustrating various DLL3-targeting restricted CD3-binding constructs. The basic components of the DLL3-targeting restricted CD3-binding constructs of this invention restrict CD3 binding. The antigen-binding domain is located at the amino terminus and / or carboxyl terminus. The Fc region (such as the heterodimeric Fc region) is located at the N-terminus of the CD3-binding region. This positioning of the Fc region adjacent to the CD3-binding region hinders CD3 binding.

[0085] [Figures 4A-4B] demonstrate the binding of the DLL3-targeting restricted CD3 binding construct cx4720 to DLL3-positive cells SHP-77. [Figure 4A]), without binding to naïve human T cells ( [Figure 4B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0086] [Figures 5A-5B] demonstrate the binding of the DLL3-targeting restricted CD3 binding construct cx3715 to DLL3-positive cells SHP-77. [Figure 5A]), without binding to naïve human T cells ( [Figure 5B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0087] [Figures 6A-6B] demonstrate the binding of the DLL3-targeting restricted CD3 binding construct cx4422 to DLL3-positive cells SHP-77. [Figure 6A]), without binding to naïve human T cells ( [Figure 6B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0088] [Figures 7A-7B] show the binding of the DLL3-targeting restricted CD3 binding construct cx3708 to DLL3-positive cells SHP-77. [Figure 7A]) and does not bind to naïve human T cells ( [Figure 7B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0089] [Figures 8A-8B] show the binding of the DLL3-targeting restricted CD3 binding construct cx4052 to DLL3-positive cells SHP-77. [Figure 8A]), without binding to naïve human T cells ( [Figure 8B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0090] [Figures 9A-9B] show the binding of the DLL3-targeting restricted CD3 binding construct cx3985 to DLL3-positive cells SHP-77. [Figure 9A]), without binding to naïve human T cells ( [Figure 9B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0091] [Figures 10A-10B] show the binding of the DLL3-targeting restricted CD3 binding construct cx4059 to DLL3-positive cells SHP-77. [Figure 10A]), without binding to naïve human T cells ( [Figure 10B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0092] [Figures 11A-11B] show the binding of the DLL3-targeting restricted CD3 binding construct cx4087 to DLL3-positive cells SHP-77. [Figure 11A]), without binding to naïve human T cells ( [Figure 11B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0093] [Figures 12A-12B] demonstrate the binding of the DLL3-targeting restricted CD3 binding construct cx4088 to DLL3-positive cells SHP-77. [Figure 12A]), without binding to naïve human T cells ( [Figure 12B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0094] [Figures 13A-13B] show the binding of the DLL3-targeting restricted CD3 binding construct cx4895 to DLL3-positive cells SHP-77. [Figure 13A]), without binding to naïve human T cells ( [Figure 13B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0095] [Figures 14A-14B] demonstrate the binding of the DLL3-targeting restricted CD3 binding construct cx3991 to DLL3-positive cells SHP-77. [Figure 14A]) and does not bind to naïve human T cells ( [Figure 14B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0096] [Figures 15A-15B] demonstrate the binding of the DLL3-targeting restricted CD3 binding construct cx3711 to DLL3-positive cells SHP-77. [Figure 15A]), without binding to naïve human T cells ( [Figure 15B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0097] [Figures 16A-16B] show the binding of the DLL3-targeting restricted CD3 binding construct cx4887 to DLL3-positive cells SHP-77. [Figure 16A]), without binding to naïve human T cells ( [Figure 16B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0098] [Figures 17A-17B] show the binding of the DLL3-targeting restricted CD3 binding construct cx4896 to DLL3-positive cells SHP-77. [Figure 17A]), without binding to naïve human T cells ( [Figure 17B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0099] [Figures 18A-18B] demonstrate the binding of the DLL3-targeting restricted CD3 binding construct cx4899 to DLL3-positive cells SHP-77. [Figure 18A]), without binding to naïve human T cells ( [Figure 18B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0100] [Figures 19A-19B] show the binding of the DLL3-targeting restricted CD3 binding construct cx4406 to DLL3-positive cells SHP-77. [Figure 19A]), without binding to naïve human T cells ( [Figure 19B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0101] [Figures 20A-20B] show the binding of the DLL3-targeting restricted CD3 binding construct cx3707 to DLL3-positive cells SHP-77. [Figure 20A]), without binding to naïve human T cells ( [Figure 20B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0102] [Figures 21A-21B] show the binding of the DLL3-targeting restricted CD3 binding construct cx3710 to DLL3-positive cells SHP-77. [Figure 21A]), without binding to naïve human T cells ( [Figure 21B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0103] [Figures 22A-22B] show the binding of the DLL3-targeting restricted CD3 binding construct cx4888 to DLL3-positive cells SHP-77. [Figure 22A]), without binding to naïve human T cells ( [Figure 22B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0104] [Figures 23A-23B] show the binding of the DLL3-targeting restricted CD3 binding construct cx4890 to DLL3-positive cells SHP-77. [Figure 23A]), without binding to naïve human T cells ( [Figure 23B]). Binding was assessed by flow cytometry using a secondary antibody against human IgG APC. Unless otherwise noted, histograms show normalized cell counts of various constructs relative to fluorescence at 200 nM.

[0105] [Figures 24A-24B] depict the ability of DLL3-targeted CD3-binding constructs to induce DLL3-dependent T cell activation. The Jurkat CD3 NFAT-GFP reporter cell line was used to monitor T cell activation. SHP-77 cells ( [Figure 24A]) and HEK-293 free cells ( [Figure 24B]) were used as antigen-positive and antigen-negative cell lines, respectively.

[0106] [Figures 25A-25D] depict the ability of CD3-binding constructs targeting the dual antigenic determinant DLL3 to induce DLL3-dependent T cell activation. Jaccard CD3 NFAT-GFP report somatic cell lines are used to monitor T cell activation. SHP-77 cells (Figure...) [22A] and [22C]) and HEK-293 free cells ( [Figure 22B] and [22D]) were used as antigen-positive and antigen-negative cell lines, respectively.

[0107] [Figures 26A-26B] depict diagrams illustrating the targeting of DLL3-restricted CD3-binding constructs mediated by DLL3-positive SHP-77 cells ( [Figure 26A]) and DLL3-negative HEK-293 free cells ( [Figure 26B]) The ability of antigen-specific T cell cytotoxicity.

[0108] [Figures 27A-27D] depict diagrams illustrating the targeting of DLL3-restricted CD3-binding constructs mediated by DLL3-positive SHP-77 cells. [Figure 27A] and [27C]) and DLL3-negative HEK-293 free cells ( [Figure 27B] and [27D]) The ability of antigen-specific T cell activation.

[0109] [Figure 28] Demonstrates the ability of DLL3-targeted restricted CD3 binding constructs to induce cytokine production from T cells in an antigen-dependent manner. Cytokine production was monitored using an ELISA method.

[0110] [Figure 29A] is a schematic diagram of three restricted CD3 constructs targeting DLL3, consisting of two polypeptides (chain 1 and chain 2). Chain 1 contains a heterodimer Fc "sodium", which is linked via a non-cleavable linker to an anti-CD3 VL domain modified with G100C, which is linked to an sdAb targeting the co-stimulatory receptor. Chain 2 contains an sdAb targeting DLL3, which is linked to a complementary heterodimer Fc "club," which is then linked via a linker as described above to a G44C-modified anti-CD3 VH domain, which is linked to a second DLL3-targeting sdAb (top diagram); or contains an sdAb targeting DLL3, which is linked to a complementary heterodimer Fc "club," which is then linked via a linker as described above to a G44C-modified anti-CD3 VH domain, which is linked to a DLL3-targeting sdAb (middle diagram); or contains an sdAb targeting DLL3, which is linked to a complementary heterodimer Fc "club," which is then linked via a linker as described above to a G44C-modified anti-CD3 VH domain (bottom diagram). The resulting constructs bind to DLL3 in a divalent (top diagram) and monovalent (middle and bottom diagrams) manner. All constructs shown herein contain an sdAb targeting a co-stimulatory receptor.

[0111] [Figure 29B] is a schematic diagram of cx5499, a CD3-restricted construct targeting DLL3, consisting of two polypeptides (chain 1 and chain 2). cx5499 and Figure [29A] (top figure) shows the same cx5352, but lacks the sdAb targeting the co-stimulatory receptor at the C-terminus of chain 1. During co-expression, the CD3 binding domain is correctly assembled via VL:VH binding on the mortar and pestle, respectively. The VH:VL interaction is stabilized by an engineered disulfide bond between the modified residue G44C in the VH domain and G100C in the VL domain.

[0112] [Figures 30A-30B] show SHP-77 cell line, which exhibits DLL3-targeting monovalent (cx5800 and cx5801) and bivalent (cx5352) CD3-restricted constructs. [Figure 30A]), rather than isolated T cells ( [Figure 30B]). The binding was assessed by flow cytometry.

[0113] [Figure 30C] Depicts the ability of representative DLL3-targeting restricted CD3 conjugating constructs to promote CD3 signaling in the presence of DLL3-positive SHP-77 cells. Conjugation of DLL3-positive cells to the bivalent and dual-antigenic construct (cx5352) induced stronger T cell activation than to the monovalent constructs (cx5800 and cx5801). CD3 signaling was assessed using somatic cell lines reported by Jaccard CD3 NFAT luciferase.

[0114] [Figures 31A-31E] demonstrate the ability of the representative CD3-restricted binding construct cx5499 targeting DLL3 to induce T cell-mediated cytotoxicity and T cell activation in the presence of DLL3-positive SHP-77 cells. [Figure 31A] shows the ability of a representative CD3-binding construct targeting DLL3 to induce T cell-mediated cytotoxicity in the presence of DLL3-positive SHP-77 cells. [Figures 31B-31E] demonstrate the ability of representative CD3-restricted binding constructs targeting DLL3 to induce T cell activation in the presence of DLL3-positive SHP-77 cells, as assessed by: CD25 expression on CD4+ T cells ( [Figure 31B]), CD69 expression on CD4+ T cells ( [Figure 31C]), CD25 expression on CD8+ T cells ( [Figure 31D]) and CD69 expression on CD8+ T cells ( [Figure 31E]).

[0115] [Figure 32] This demonstrates the ability of the CD3-restricted binding construct cx5352, which targets DLL3 and has a 41BB binding domain, to mediate 41BB signaling, but not the identical construct cx5499, which lacks a 41BB binding domain. Somatic monitoring of DLL3-dependent 41BB signaling was reported using Jacobart 41BB NFkB luciferase co-cultured with parental HEK-293 cells or HEK-293 cells transiently expressing truncated DLL3, and activity is expressed in relative luciferase units (RLUs).

[0116] [Figures 33A-B] depict an exemplary restricted CD3 binding construct cx5352, which targets DLL3, driving the action of the DLL3-positive cell line SHP-77. [Figure 33A]) or DLL3-negative cell line HEK-293FS ( [Figure 33B] The efficacy of T cell-mediated cytotoxicity. Three different T cell donors were used as the effector cell source in this analysis.

[0117] [Figures 34A-C] illustrate the difference in T cell-mediated cytotoxicity driven by cx5352, an exemplary restricted CD3-binding construct targeting DLL3 with a 41BB binding domain, and cx5499, an exemplary restricted CD3-binding construct targeting DLL3 without a 41BB binding domain. In this paper, the following ratios of effectors (T cells) to target cells (SHP-77) (effector:target cell ratio) were compared: 10:1 ( [Figure 34A]), 5:1 ( [Figure 34B]) or 1.25:1 ( [Figure 34C]). Human PBMCs are used as a source of T cells.

[0118] [Figure 35] This figure shows a comparison of IFNγ production in T cells titrated with cx5352, a representative restricted CD3 conjugate targeting DLL3 with a 41BB binding domain, and cx5499, a representative restricted CD3 conjugate targeting DLL3 without a 41BB binding domain, in the presence of the DLL3-positive cell line SHP-77.

[0119] [Figures 36A-B] illustrate exemplary restricted CD3 conjugates targeting TAAs with / without costimulatory receptor-binding regions (CRBRs). The construct has sdAb and Fc of the targeting antigen located at the N-terminus and C-terminus of one chain of the heterodimer, and has a costimulatory receptor-binding region (CRBR) and Fc located at the C-terminus of the opposite chain of the heterodimer, and has VH and VL of CD3-binding Fv located on opposite sides of each other.

[0120] [Figures 37A-D] depict The T-cell reporter body analysis results of the exemplary constructs described in [Figures 36A-B]. [Figure 37A] and [37B] Describes the mean fluorescence intensity (MFI) of GFP reporter cells when TAA-positive cell line A375 or TAA-negative cell line CCRF-CEM are co-cultured with JKC CD3 NFAT-GFP reporter cells. [Figure 37C] and [37D] Describes the relative luminescent units (RLU) of luciferase reporter cells when TAA-positive cell line A375 or TAA-negative cell line CCRF-CEM are co-cultured with Jekkaal CD3 NFAT luciferase reporter cells.

[0121] [Figures 38A-B] depict the results of T-cell reporter somatic cell analysis using the Jockaert reporter somatic cell line, which was engineered to stably express CD16a using an NFAT-driven luciferase reporter gene. [Figure 38A] Depicts the relative luminescent units (RLU) of luciferase reporter cells when CHO cells expressing DLL3 are co-cultured with reporter somatic cells and titrated with a DLL3-targeting construct containing IgG1 Fc. [Figure 38B] Depicts the RLU of luciferase reporter cells after titration with a DLL3-targeting construct containing IgG1 Fc in the absence of DLL3-expressing cells. Implementation [Cross-reference to related applications]

[0122] This application claims priority to the following U.S. provisional applications: 62 / 744,638, filed October 11, 2018, entitled "DLL3 Single-Domain Antibody and Therapeutic Composition Thereof"; 62 / 832,265, filed April 10, 2019, entitled "DLL3 Single-Domain Antibody and Therapeutic Composition Thereof"; and 62 / 877,815, filed July 23, 2019, entitled "DLL3 Single-Domain Antibody and Therapeutic Composition Thereof," the contents of which are incorporated herein by reference in their entirety for all purposes. [The sequence list is incorporated by reference]

[0123] This application, together with the sequence list, is filed electronically. The sequence list is provided as a file named 744952000641SeqList.TXT, created on October 8, 2019, and is 431 kilobytes in size. Information from the electronic sequence list is incorporated herein by reference in its entirety.

[0124] This document provides polypeptides that specifically bind to DLL3, hereinafter also referred to as DLL3-binding polypeptides. In some embodiments, the provided binding polypeptides comprise at least one VHH domain that binds to DLL3. In some embodiments, the DLL3-binding polypeptides provided herein comprise one, two, three, four, five, six, seven, or eight VHH domains that each individually bind to DLL3. In some embodiments, the DLL3-binding polypeptides provided herein comprise one, two, three, or four VHH domains that bind to DLL3. In some embodiments, the DLL3-binding polypeptide is single-specific. In some embodiments, the DLL3-binding polypeptide is multi-specific. For example, the provided DLL3-binding polypeptides may comprise at least one VHH domain that binds to DLL3 and a polypeptide that binds to one or more other binding domains of a target protein other than DLL3, such as one or more other VHH domains.

[0125] In some embodiments, the DLL3-binding polypeptide comprises at least one VHH domain and an Fc domain for binding DLL3. In some embodiments, the DLL3-binding polypeptide provided herein comprises one, two, three, or four VHH domains and Fc domains for binding DLL3. In some embodiments, the Fc domain mediates dimerization of the DLL3-binding polypeptide under physiological conditions to form a dimer that doubles the number of DLL3 binding sites. For example, a DLL3-binding polypeptide comprising three VHH domains and an Fc region for binding DLL3 is in trivalent monomeric form, but under physiological conditions, the Fc region can mediate dimerization, such that the DLL3-binding polypeptide exists as a hexavalent dimer under these conditions.

[0126] DLL3 is a member of the δ-protein ligand family. The δ-protein ligand family acts as Notch ligands characterized by the DSL domain, EGF repeat, and transmembrane domain. DLL3 is highly expressed in the fetal brain, but not in normal adult tissues. In contrast, DLL3 is expressed on the surface of various tumor cells and tumor blood vessels, including (but not limited to) small cell lung cancer (SCLC), large cell neuroendocrine carcinoma (LCNEC), and ovarian cancer. During normal development, unlike Notch-activated δ-protein family members, DLL3 inhibits the activation of both cis- and trans-acting Notch pathways. This inhibition is achieved by interacting with Notch and DLL1, respectively, and redirecting or retaining them in late endosome / lysosomal compartments or the Golgi apparatus, thereby preventing their localization on the cell surface (Chapman et al., 2011, Hum Mol Genet. 20(5): 905-16.; Serth et al., 2015, PLoS One. 10(4): e0123776.). Notably, Notch activation inhibits tumor growth in neuroendocrine tumors (Kunnimalaiyaan and Chen, 2007, Oncologist. 12(5): 535-42). These observations suggest that DLL3 may be associated with the neuroendocrine phenotype by downregulating Notch signaling, thereby leading to neuroendocrine tumors. DLL3 is extremely limited in normal adult tissues, but is widely present in malignant SCLC, LCNEC, melanoma, glioblastoma, and extrapulmonary neuroendocrine carcinoma (NEC) (Saunders et al., 2015, Sci Transl Med. 2015, 7(302): 302ra136; Peng et al., 2016, J. Clin. Oncol., 34, Vol. 15, Supplement 11611-11611). In addition, high levels of DLL3 expression in tumor tissue are associated with the following: poor survival rates in advanced tumor stages and / or glioblastoma, medullary thyroid and neuroendocrine pancreatic cancer (Peng et al., 2016, J. Clin. Oncol., 34, Vol. 15, Supplement 11611-11611), SCLC and LCNEC (Saunders et al., 2015, Sci Transl Med. 2015, 7(302): 302ra136), and some ovarian cancers (Hu et al., 2014, Cancer Res. 74(12): 3282-3293).

[0127] An exemplary sequence of a typical human DLL3 is shown below: (SEQ ID NO:86, signal sequence is underlined)

[0128] An exemplary sequence of atypical human DLL3 is shown below: (SEQ ID NO:87, signal sequence is underlined)

[0129] In some cases, the provided DLL3-binding peptides directly block or inhibit the activity of DLL3, and in some forms, they can be used as therapeutic agents to inhibit or reduce tumor cell growth or survival.

[0130] Multiple DLL3 peptide binding forms are provided. In some instances, the DLL3-binding peptide includes the DLL3 VHH-Fc peptide. In some embodiments, the Fc is an Fc exhibiting immune effector activity, such as one or more effector functions, such as antibody-dependent cytotoxicity (ADCC), antibody-dependent phagocytosis (ADCP), and / or complement-dependent cytotoxicity (CDC).

[0131] In some embodiments, the provided DLL3-binding peptides can be used to stimulate an individual's immune response, and in some embodiments, to treat an individual's disease or condition, such as cancer. In some embodiments, the DLL3-binding peptides provided herein, such as DLL3-Fc, can bind to tumor cells expressing DLL3 and induce an active immune response against tumor cells expressing DLL3. In some cases, the active immune response can cause cancer cell death (e.g., antibody binding to cancer cells induces apoptotic cell death) or inhibit cancer cell growth (e.g., block cell cycle progression). In other cases, the DLL3-binding peptides provided herein, such as DLL3 VHH-Fc, can bind to cancer cells and antibody-dependent cytotoxicity (ADCC) can eliminate the cancer cells bound by the DLL3-binding peptide. In some embodiments, the provided DLL3 VHH-binding peptides can also activate cellular and humoral immune responses and recruit more natural killer cells or increase the production of cytokines (e.g., IL-2, IFN-γ, IL-12, TNF-α, TNF-β, etc.) that further activate the individual's immune system to destroy cancer cells. In another embodiment, DLL3-binding peptides, such as DLL3 VHH-Fc, can bind to cancer cells, and macrophages or other phagocytes can modulate cancer cells, such as via CDC or ADCP processes.

[0132] In other forms, this paper also provides VHH-binding peptides exhibiting multispecific binding. In some cases, the binding peptides include peptides exhibiting dual affinity for DLL3 and T-cell antigens such as CD3. In some forms, such dual-affinity molecules can bind to or activate T cells at the tumor site when binding to DLL3 expressed by the tumor. Specifically, such molecules provided herein are particularly those exhibiting restricted CD3 binding. This paper also provides engineered cells, such as engineered T cells, that express chimeric antigen receptors containing DLL3-binding peptides.

[0133] All publications mentioned in this application, including patent documents, scientific papers, and databases, are incorporated herein by full reference for all purposes, as if each publication were individually incorporated by reference. Where the definitions set forth herein contradict or otherwise differ from those set forth in the patents, applications, publications, and other publications incorporated herein by reference, the definitions set forth herein shall prevail, not those incorporated herein by reference.

[0134] The techniques and procedures described or mentioned herein are generally well understood and commonly used by those familiar with the techniques, such as the widely used methods described in the following literature: Sambrook et al., Molecular Cloning: A Laboratory Manual 3rd Edition (2001), Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NYCURRENT PROTOCOLS IN MOLECULAR BIOLOGY (edited by F.M. Susubel et al., (2003)); Series of Methods in Enzymology (Academic Press, Inc.): PCR 2: A PRACTICAL APPROACH (edited by M.J. MacPherson, B.D. Hames, and G.G. Taylor, (1995)); Harlow and Lane (edited by 1988) ANTIBODIES, A LABORATORY MANUAL, and ANIMAL CELL CULTURE (edited by R.R. Freshney, (1987)); Oligonucleotide Synthesis (edited by M.J. Gait, 1984); Methods in Molecular Cloning. Biology, Humana Press; Cell Biology: A Laboratory Notebook (JECellis editor, 1998) Academic Press; Animal Cell Culture (RI Freshney) editor, 1987); Introduction to Cell and Tissue Culture (JPMather and PERoberts, 1998) Plenum Press; Cell and Tissue Laboratory Culture Procedures (edited by A.Doyle, JBGriffiths and DG Newell, 1993-8) J.Wiley and Sons; Handbook of Experimental Immunology (edited by DMWeir and CC Blackwell); Gene Transfer Vectors for Mammalian Cells (JMMiller and MPCalos (ed., 1987); PCR: The Polymerase Chain Reaction (Mullis et al., ed., 1994); Current Protocols in Immunology (J. E. Coligan et al., ed., 1991); Short Protocols in Molecular Biology (Wiley and Sons, 1999); Immunobiology (CA. Janeway and P. Travers, 1997); Antibodies (P. Finch, 1997); Antibodies: A Practical Approach (D. Catty, ed., IRL Press, 1988-1989); Monoclonal Antibodies: A Practical Approach (P. Shepherd and C. Dean, ed., Oxford University Press, 2000); Using Antibodies: A Laboratory Manual (E. Harlow and D. Lane, Cold Spring Harbor Laboratory Press, 1999); The Antibodies (M. Zanetti and JD. Capra, ed., Harwood Academic Press, 1999). Publishers, 1995); and Cancer: Principles and Practice of Oncology (edited by VT DeVita et al., JBLippincott Company, 1993); and its latest version.

[0135] The chapter titles used in this article are for organizational purposes only and should not be construed as limiting the topics described. [I. Definition]

[0136] Unless otherwise defined, scientific and technical terms used in connection with this invention will have the meanings commonly understood by one of ordinary skill in the art. Furthermore, unless the context otherwise requires or explicitly indicates otherwise, singular terms shall include plural terms and plural terms shall include singular terms. In the event of any inconsistency in definitions among various sources or references, the definitions provided herein shall prevail.

[0137] It should be understood that the embodiments of the invention described herein include those "composed of" and / or "substantially composed of" the embodiments. Unless otherwise indicated, as used herein, the singular forms "a / an" and "the" include a plurality of indicators. The use of the term "or" herein is not intended to imply that alternatives are mutually exclusive.

[0138] In this application, unless explicitly stated or understood by one skilled in the art, the use of "or" means "and / or". In the case of multiple sub-items, the use of "or" refers to more than one of the aforementioned independent or sub-items.

[0139] As used herein, the term "about" refers to a common range of error for individual values ​​that is readily apparent to those skilled in the art. References to "about" a value or parameter herein include (and describe) embodiments of that value or parameter itself. For example, a description relating to "about X" includes a description of "X".

[0140] The terms "nucleic acid molecule," "nucleic acid," and "polynucleotide" are used interchangeably and refer to nucleotide polymers. Such nucleotide polymers may contain natural and / or non-natural nucleotides and include (but are not limited to) DNA, RNA, and PNA. "Nucleic acid sequence" refers to the linear sequence of nucleotides contained in a nucleic acid molecule or polynucleotide.

[0141] As used herein, “isolated polynucleotide” should mean a polynucleotide of genomic, cDNA or synthetic origin or a combination thereof, in which the isolated polynucleotide (1) is not associated with all or part of polynucleotides found in nature, (2) is operatively linked to polynucleotides in nature that are not linked thereto, or (3) is not present in nature as part of a larger sequence.

[0142] The terms "peptide" and "protein" are used interchangeably and refer to polymers of amino acid residues, and are not limited to a minimum length. Such polymers of amino acid residues may contain native or non-native amino acid residues and include, but are not limited to, peptides, oligopeptides, dimers, trimers, and polymers of amino acid residues. Full-length proteins and their fragments are both covered in this definition. The term also includes post-expression modifications of peptides, such as glycosylation, sialylation, acetylation, phosphorylation, and similar modifications. Furthermore, for the purposes of this invention, "peptide" refers to a protein that includes modifications to its native sequence, such as deletions, additions, and substitutions (which are generally conserved in practice), provided that the protein retains the desired activity. Such modifications may be intentional, such as induced by site-directed mutagenesis; or may be accidental, such as by mutations in the host that produces the protein or by errors caused by PCR amplification.

[0143] The term "isolated protein" as used herein means a target protein that (1) is free from at least some of the other proteins that are normally found with it in nature, (2) is substantially free from other proteins of the same origin, such as those of the same species, (3) is expressed by cells of a different species, (4) has been isolated from at least about 50% of the polynucleotides, lipids, carbohydrates or other materials associated with it in nature, (5) is not associated with the protein fraction associated with "isolated protein" in nature (by covalent or non-covalent interaction), (6) is operable to be associated with polypeptides that are not associated with it in nature (by covalent or non-covalent interaction), or (7) is not present in nature. Such isolated proteins may be encoded by genomic DNA, cDNA, mRNA or other RNA, may be of synthetic origin, or any combination thereof. In some embodiments, the isolated protein is substantially pure or substantially free from proteins or polypeptides or other contaminants found in its natural environment that would interfere with its use (therapeutic, diagnostic, preventative, research or otherwise).

[0144] As used herein, "substantially pure" means that the target species is the dominant species present (i.e., more abundant than any other individual species in the composition in moles), and that the substantially purified portion is a composition in which the target species constitutes at least about 50% (in moles) of all macromolecular species present. Generally, a substantially pure composition will contain more than about 80% of all macromolecular species present in the composition, for example, more than about 85%, 90%, 95%, and 99% in some embodiments. In some embodiments, the target species is purified to substantially homogeneity (contaminant species cannot be detected in the composition by conventional detection methods), wherein the composition consists essentially of a single macromolecular species.

[0145] As used herein, "operationally linked" means that the component described in this way is positioned in a relationship that allows it to function in the intended manner. "Operationally linked" to the control sequence of the coding sequence is achieved in a manner that allows the performance of the coding sequence to be compatible with the control sequence.

[0146] The term "specific binding" to an antigen or antigenic determinant is a well-understood term in this art, and the methods used to determine such specific binding are also well-known in this art. A molecule is said to exhibit "specific binding" or "preferential binding" if it reacts or associates with a specific cell or substance more frequently, more rapidly, for a longer duration, and / or with a greater affinity than with alternative cells or substances. A single-domain antibody (sdAb) or VHH-containing peptide is said to "specifically bind" or "preferentially bind" to a target if its binding affinity to the target is greater, easier, and / or longer-lasting than its binding affinity to other substances. For example, an sdAb or VHH-containing peptide that specifically or preferentially binds to a DLL3 antigenic determinant is an sdAb or VHH-containing peptide that binds to this antigenic determinant with greater, easier, and / or longer-lasting affinity than it binds to other DLL3 or non-DLL3 antigenic determinants. By reading this definition, it should also be understood that, for example, an sdAb or VHH-containing polypeptide that specifically or preferentially binds to a first target may or may not specifically or preferentially bind to a second target. Therefore, "specific binding" or "preferential binding" does not necessarily require (although it may include) exclusive binding. Generally speaking, but not necessarily, the mention of binding means preferential binding. "Specificity" refers to the ability of a binding protein to selectively bind to an antigen.

[0147] As used herein, the term "antigenic determinant" refers to the binding site of an antigen-binding molecule (e.g., sdAb or VHH-containing polypeptide) on a target molecule (e.g., an antigen, such as a protein, nucleic acid, carbohydrate, or lipid). Antigenic determinants often comprise a group of chemically active molecules, such as amino acids, polypeptides, or sugar side chains, and possess specific three-dimensional structural features and mass-charge ratio characteristics. Antigenic determinants can be formed from adjacent and / or fused non-adjacent residues (e.g., amino acid, nucleotide, sugar, lipid moieties) of the target molecule. Antigenic determinants formed from adjacent residues (e.g., amino acid, nucleotide, sugar, lipid moieties) are generally retained upon exposure to denaturing solvents, while antigenic determinants formed by tertiary folding are generally lost upon treatment with denaturing solvents. Antigenic determinants may include (but are not limited to) at least 3, at least 5, or 8-10 residues (e.g., amino acids or nucleotides). In some embodiments, the length of the antigenic determinant is less than 20 residues (e.g., amino acids or nucleotides), less than 15 residues, or less than 12 residues. If two antibodies competitively bind to an antigen, they may bind to the same antigenic determinant within that antigen. In some embodiments, antigenic determinants can be identified by a minimum distance from the CDR residues on the antigen-binding molecule. In some embodiments, antigenic determinants can be identified by this distance, and further limited to residues associated with bonds (e.g., hydrogen bonds) between residues of the antigen-binding molecule and antigen residues. Antigenic determinants can also be identified by various scans, such as alanine or arginine scans, which can indicate one or more residues of the antigen-binding molecule that can interact. Unless explicitly stated otherwise, a set of residues as an antigenic determinant does not exclude other residues as part of the antigenic determinant of a particular antigen-binding molecule. In fact, the presence of this set represents a minimal set of antigenic determinants (or a group of species). Therefore, in some embodiments, a set of residues identified as an antigenic determinant represents the minimally relevant antigenic determinant of the antigen, rather than an exclusive list of antigenic determinants on the antigen.

[0148] A "nonlinear antigenic determinant" or "configurational antigenic determinant" comprises a non-adjacent polypeptide, amino acid, and / or sugar within an antigen protein to which an antigen-binding molecule specific to the antigenic determinant binds. In some embodiments, at least one residue will not be adjacent to other indicated residues of the antigenic determinant; however, one or more residues may be adjacent to other residues.

[0149] A "linear antigenic determinant" comprises adjacent polypeptides, amino acids, and / or sugars within an antigen protein to which an antigen-binding molecule specifically binds. It should be noted that in some embodiments, not every residue within the linear antigenic determinant needs to directly bind (or be bonded) to the antigen-binding molecule. In some embodiments, the linear antigenic determinant may be derived from immunization with a peptide that is substantially composed of the linear antigenic determinant sequence, or from a protein structural portion relatively separated from the rest of the protein (so that the antigen-binding molecule can interact at least primarily, only with that sequence portion).

[0150] The terms "antibody" and "antigen-binding molecule" are used interchangeably in the broadest sense and encompass a variety of polypeptides containing an antibody-like antigen-binding domain, including (but not limited to) conventional antibodies (typically containing at least one heavy chain and at least one light chain), single-domain antibodies (sdAbs, which contain only one chain typically resembling a heavy chain), VHH-containing polypeptides (containing at least one heavy chain, and polypeptides containing only an antibody variable domain or VHH), and fragments of any of the foregoing, provided they exhibit the desired antigen-binding activity. In some embodiments, the antibody contains a dimerizing domain. Such dimerizing domains include (but are not limited to) a heavy chain constant domain (containing CH1, a hinge, CH2, and CH3, wherein CH1 typically pairs with the light chain constant domain CL, and the hinge regulates dimerization) and an Fc domain (containing a hinge, CH2, and CH3, wherein the hinge regulates dimerization).

[0151] The term antibody also includes (but is not limited to) chimeric antibodies, humanized antibodies, and antibodies from various species, such as camels (including llamas), sharks, mice, humans, and cynomolgus monkeys.

[0152] The term "variable region" or "variable domain" refers to a structural domain in the heavy or light chain of an antibody that participates in the binding of the antibody to the antigen. The variable domains of the heavy and light chains (VH and VL, respectively) of native antibodies typically have similar structures, with each domain containing four conserved framework regions (FRs) and three core regions (CDRs). (See, for example, Kindt et al., Kuby Immunology, 6th ed., WH Freeman and Co., p. 91 (2007)). A single VH or VL domain can be sufficient to confer antigen-binding specificity, as in single-domain antibodies such as VHH. Furthermore, antibodies binding to specific antigens can be isolated using the VH or VL domains of antibodies binding to antigens to screen libraries of complementary VL or VH domains separately. See, for example, Portolano et al., J. Immunol. 150:880-887 (1993); Clarkson et al., Nature 352:624-628 (1991).

[0153] "Antibody fragment" or "antigen-binding fragment" refers to a molecule, other than a known or complete antibody, that contains at least a variable region of a known or complete antibody that binds to an antigen. Examples of antibody fragments include (but are not limited to) Fv, single-chain Fv (sdFv), Fab, Fab', Fab'-SH, F(ab')2; bifunctional antibodies; linear antibodies; and single-domain antibodies containing only the VH region (VHH).

[0154] As used herein, when referring to binding molecules, "monovalent" means that the binding molecule has a single antigen recognition site that is specific to the target antigen. Examples of monovalent binding molecules include, for example, monovalent antibody fragments, protein-binding molecules with antibody-like binding properties, or MHC molecules. Examples of monovalent antibody fragments include (but are not limited to) Fab fragments, Fv fragments, and single-chain Fv fragments (scFv).

[0155] The terms "single-domain antibody," "sdAb," and "VHH" are used interchangeably herein to refer to an antibody having a single monomeric domain antigen-binding / recognition domain. Such antibodies include camel antibodies or shark antibodies. In some embodiments, a VHH comprises three CDRs and four framework regions, referred to as FR1, CDR1, FR2, CDR2, FR3, CDR3, and FR4. In some embodiments, a VHH may be truncated at the N-terminus or C-terminus such that it contains only a portion of FR1 and / or FR4, or lacks one or both of these framework regions, provided that the VHH substantially maintains antigen binding and specificity.

[0156] The term "VHH-containing polypeptide" refers to a polypeptide containing at least one VHH domain. In some embodiments, a VHH polypeptide contains two, three, four, or more VHH domains, wherein the VHH domains may be the same or different. In some embodiments, a VHH-containing polypeptide contains an Fc domain. In some such embodiments, the VHH polypeptide may form a dimer. Non-limiting structures of VHH-containing polypeptides include VHH1-Fc, VHH1-VHH2-Fc, and VHH1-VHH2-VHH3-Fc, wherein VHH1, VHH2, and VHH3 may be the same or different. In some embodiments of these structures, a VHH may be linked to another VHH via a linker, or a VHH may be linked to an Fc via a linker. In some such embodiments, the linker contains 1-20 amino acids, preferably 1-20 amino acids that primarily contain glycine and, where appropriate, serine. In some embodiments, when the VHH-containing polypeptide contains an Fc, it forms a dimer. Therefore, if the structure VHH1-VHH2-Fc forms a dimer, it is considered to be tetravalent (i.e., the dimer has four VHH domains). Similarly, if the structure VHH1-VHH2-VHH3-Fc forms a dimer, it is considered to be hexavalent (i.e., the dimer has six VHH domains).

[0157] As used herein, a DLL3-binding polypeptide is a polypeptide or protein that specifically binds to DLL3. Typically, a DLL3-binding polypeptide in this context is a VHH-containing polypeptide containing at least one VHH domain that binds to DLL3. DLL3-binding polypeptides include conjugates, including fusion proteins. DLL3-binding polypeptides include fusion proteins, including fusion proteins containing an Fc domain. In some embodiments, a DLL3-binding polypeptide contains two, three, four, or more VHH domains, each specifically binding to a VHH domain of DLL3, wherein the VHH domains may be identical or different. In some embodiments, a DLL3-binding polypeptide is multivalent. In some embodiments, a DLL3-binding polypeptide is multispecific. In some cases, a DLL3-binding polypeptide may contain one or more domains that bind to one or more other domains of other antigens besides DLL3.

[0158] The term "monoclonal antibody" refers to an antibody (including sdAbs or VHH-containing peptides) that is substantially homogeneous, meaning that the individual antibodies constituting this population are identical except for the possible small amounts of naturally occurring mutations. Monoclonal antibodies are highly specific for a single antigenic site. Furthermore, in contrast to polyclonal antibody formulations, which typically consist of different antibodies targeting different determinants (antigenic determinants), each monoclonal antibody system targets a single determinant on the antigen. Therefore, a sample of a monoclonal antibody can bind to the same antigenic determinant. The modifier "monoclonal" indicates that the antibody is derived from a substantially homogeneous population of antibodies and should not be construed as requiring any particular method to produce the antibody. For example, monoclonal antibodies can be manufactured using the fusion tumor method first described in Kohler and Milstein, 1975, Nature 256:495, or using recombinant DNA methods such as those described in U.S. Patent No. 4,816,567. For example, monoclonal antibodies can also be isolated from phage libraries generated using the technique described in McCafferty et al., 1990, Nature 348:552-554.

[0159] The term "CDR" stands for Complementary Determination Region, as defined by a person skilled in this art using at least one identification method. The exact amino acid sequence boundaries of a given CDR or FR can be readily determined using any of a number of well-known schemes, including those described in the following literature: Kabat et al. (1991), "Sequences of Proteins of Immunological Interest", 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD ("Kabat" numbering scheme); Al-Lazikani et al. (1997) JMB 273, 927-948 ("Chothia" numbering scheme); MacCallum et al., J. Mol. Biol. 262: 732-745 (1996), "Antibody-antigen interactions: Contact analysis and binding site topography", J. Mol. Biol. 262, 732-745 ("Contact" numbering scheme); Lefranc MP et al., "IMGT unique numbering for immunoglobulin and T cell receptor variable domains and Ig superfamily V-like domains", Dev Comp. Immunol, January 2003; 27(1):55-77 (“IMGT” numbering scheme); Honegger A and Plückthun A, “Yet another numbering scheme for immunoglobulin variable domains: an automatic modeling and analysis tool”, J Mol Biol, June 8, 2001; 309(3):657-70 (“Aho” numbering scheme); and Martin et al., “Modeling antibody hypervariable loops: a combined algorithm”, PNAS, 1989, 86(23):9268-9272 (“AbM” numbering scheme).

[0160] The boundaries of a given CDR or FR can vary depending on the identification scheme used. For example, the Kabat scheme is based on structural alignment, while the Chothia scheme is based on structural information. Both Kabat and Chothia schemes number antibodies based on the length of the most commonly used antibody region sequence, where some antibodies exhibit insertions and deletions indicated by insert letters (e.g., "30a"). The two schemes result in different numbers for certain insertions and deletions ("insertions and deletions") located in different positions. The contact scheme is based on the analysis of complex crystal structures and is similar to the Chothia numbering scheme in several ways. The AbM scheme is a compromise between the Kabat and Chothia definitions, based on the scheme used by Oxford Molecular's AbM antibody modeling software.

[0161] In some embodiments, CDRs may be defined according to any of the Chothia numbering scheme, the Kabat numbering scheme, a combination of Kabat and Chothia, the AbM definition, and / or the contact definition. The VHH contains three CDRs, referred to as CDR1, CDR2, and CDR3. Table 1 below lists illustrative positional boundaries of CDR-H1, CDR-H2, and CDR-H3 as identified by the Kabat, Chothia, AbM, and contact schemes, respectively. For CDR-H1, residue numbering is listed using both the Kabat and Chothia numbering schemes. FRs are located between CDRs; for example, FR-H1 precedes CDR-H1, FR-H2 is between CDR-H1 and CDR-H2, FR-H3 is between CDR-H2 and CDR-H3, and so on. It should be noted that because the Kabat numbering scheme shown makes the insertion point at H35A and H35B, the end of the Chothia CDR-H1 ring varies between H32 and H34 when numbered using the Kabat numbering convention shown, depending on the ring length.

[0162]

[0163] Therefore, unless otherwise stated, the “CDR” or “complementarity-determining region” of a given antibody or its regions (such as its variable regions) or individually designated CDRs (e.g., CDR-H1, CDR-H2, CDR-H3) should be understood to encompass the complementarity-determining region (or specific complementarity-determining region) as defined by any of the foregoing schemes. For example, when stating that a particular CDR (e.g., CDR-H3) contains the amino acid sequence of the corresponding CDR in a given VHH amino acid sequence, it should be understood that such CDR has the sequence within VHH as defined by any of the foregoing schemes for the corresponding CDR (e.g., CDR-H3). In some embodiments, a specific CDR sequence is specified. Various numbering schemes are used to describe exemplary CDR sequences of the provided antibodies (see, for example, Table 1), but it should be understood that the provided antibodies may include CDRs as described according to any of the other numbering schemes described above or other numbering schemes known to a skilled person.

[0164] As used herein, "conjugate," "conjugate," or any grammatical variation thereof refers to two or more compounds joined or linked together by any joining or linking method known in the art, resulting in the formation of another compound. It can also refer to compounds produced by joining or linking two or more compounds together. For example, illustrative conjugates are those directly or indirectly linked to the VHH domain of one or more chemical moieties or polypeptides. Such conjugates include fusion proteins, conjugates produced by chemical conjugates, and conjugates produced by any other method.

[0165] Immunoglobulin Fc fusions (“Fc-fusions”), such as VHH-Fc, are molecules containing one or more VHH domains operably linked to the Fc region of an immunoglobulin, which may be indirectly or directly linked to one or more VHH domains. Various linkers known in the art and used as appropriate to link Fc to a fusion complex to produce an Fc-fusion. In some such embodiments, the linker contains 1-20 amino acids, preferably 1-20 amino acids primarily containing glycine and, where appropriate, serine. Fc-fusions of the same species may dimerize to form a homodimer of Fc-fusions, or different species may be used to form heterodimers of Fc-fusions. In some embodiments, the Fc is a mammalian Fc, such as a human Fc.

[0166] As used herein, the term "heavy chain constant region" refers to a region containing at least three heavy chain constant domains: CH1, hinge, CH2, and CH3. Of course, unless otherwise specified, any loss or variation within these domains that does not alter function is covered by the term "heavy chain constant region." Non-limiting illustrative heavy chain constant regions include γ, δ, and α. Non-limiting illustrative heavy chain constant regions also include ε and μ. Each heavy chain constant region corresponds to an antibody isotype. For example, an anti-systemic IgG antibody containing the γ constant region, an anti-systemic IgD antibody containing the δ constant region, and an anti-systemic IgA antibody containing the α constant region. Additionally, an anti-systemic IgM antibody containing the μ constant region and an anti-systemic IgE antibody containing the ε constant region. Some isotypes can be further subdivided into subclasses. For example, IgG antibodies include (but are not limited to) IgG1 (containing the γ1 constant region), IgG2 (containing the γ2 constant region), IgG3 (containing the γ3 constant region) and IgG4 (containing the γ4 constant region) antibodies; IgA antibodies include (but are not limited to) IgA1 (containing the α1 constant region) and IgA2 (containing the α2 constant region) antibodies; and IgM antibodies include (but are not limited to) IgM1 and IgM2.

[0167] As used herein, "Fc region" refers to a portion of the heavy chain constant region containing CH2 and CH3. In some embodiments, the Fc region includes a hinge, CH2, and CH3. In various embodiments, when the Fc region includes a hinge, the hinge mediates dimerization between two Fc-containing polypeptides. The Fc region can be any antibody heavy chain constant region isotype discussed herein. In some embodiments, the Fc region is IgG1, IgG2, IgG3, or IgG4.

[0168] "Functional Fc fragments" possess the "effective functions" of the native Fc region. Exemplary "effective functions" include Fc receptor binding; Clq binding and complement-dependent cytotoxicity (CDC); Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; downregulation of cell surface receptors (e.g., B cell receptors); and B cell activation. These effector functions generally require a combination of the Fc region and a binding domain (e.g., antibody variable domain) and can be assessed using various analyses.

[0169] The "native sequence Fc region" contains amino acid sequences identical to those found in naturally occurring Fc regions. The native sequence human Fc regions include the native sequence human IgG1 Fc region (non-A and A variants); the native sequence human IgG2 Fc region; the native sequence human IgG3 Fc region; and the native sequence human IgG4 Fc region, as well as their naturally occurring variants.

[0170] The "mutated Fc region" comprises an amino acid sequence that differs from the amino acid sequence of the native Fc region by at least one amino acid modification. In some embodiments, the "mutated Fc region" comprises an amino acid sequence that differs from the amino acid sequence of the native Fc region by at least one amino acid modification, but retains at least one effector function of the native Fc region. In some embodiments, the mutated Fc region has at least one amino acid substitution compared to the native Fc region or compared to the Fc region of the parent peptide, for example, about one to about ten amino acid substitutions in the native Fc region or in the Fc region of the parent peptide, and preferably about one to about five amino acid substitutions. In some embodiments, the mutated Fc region herein has at least about 80% sequence identity, at least about 90% sequence identity, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99% sequence identity with the native Fc region and / or with the Fc region of the parent peptide.

[0171] Generally, the residues in the immunoglobulin heavy chain or its portions, such as the Fc region, are numbered according to the EU index in Kabat et al., Sequences of Proteins of Immunological Interest, 5th Edition, Public Health Service, National Institutes of Health, Bethesda, Md. (1991). "The EU index in Kabat's work" refers to the residue numbers of human IgG1 EU antibodies.

[0172] "Fc receptor" or "FcR" describes a receptor that binds to the Fc region of an antibody. In some embodiments, FcγR is a native human FcR. In some embodiments, FcR is an FcR (γ receptor) that binds to IgG antibodies and includes receptors of the FcγRI, FcγRII, and FcγRIII subclasses, including variants of their paired genes and alternative splice forms. FcγRII receptors include FcγRIIA ("activating receptor") and FcγRIIB ("inhibiting receptor"), which have similar amino acid sequences and differ primarily in their cytoplasmic domains. The activating receptor FcγRIIA contains an activating motif (ITAM) based on the immunoreceptor tyrosine in its cytoplasmic domain. The inhibiting receptor FcγRIIB contains an inhibiting motif (ITIM) based on the immunoreceptor tyrosine in its cytoplasmic domain. (See, for example, Daeron, Annu. Rev. Immunol. 15:203-234 (1997)). FcRs have been described, for example, in Ravetch and Kinet, Annu. Rev. Immunol 9:457-92 (1991); Capel et al., Immunomethods 4:25-34 (1994); and de Haas et al., J. Lab. Clin. Med. 126:330-41 (1995). Other FcRs, including those to be identified in the future, are covered by the term “FcR” in this document. For example, the terms “Fc receptor” or “FcR” also include the neonatal receptor FcRn, which is responsible for transferring maternal IgG to the fetus (Guyer et al., J. Immunol. 117:587 (1976) and Kim et al., J. Immunol. 24:249 (1994)) and regulating immunoglobulin homeostasis. Methods for measuring the binding of FcRn are known (see, for example, Ghetie and Ward, Immunol. Today 18(12):592-598(1997); Ghetie et al., Nature Biotechnology, 15(7):637-640(1997); Hinton et al., J. Biol. Chem. 279(8):6213-6216(2004); WO 2004 / 92219(Hinton et al.).

[0173] As discussed herein, the term "receptor human framework" as used herein refers to a framework comprising the amino acid sequence of a heavy chain variable domain (VH) framework derived from the human immunoglobulin framework or the human common framework. Receptor human frameworks derived from the human immunoglobulin framework or the human common framework may contain the same amino acid sequence, or may contain amino acid sequence alterations. In some embodiments, in all human frameworks within a single antigen-binding domain such as VHH, the number of amino acid alterations is less than 10, or less than 9, or less than 8, or less than 7, or less than 6, or less than 5, or less than 4, or less than 3.

[0174] As used herein, "chimeric antigen receptor" or "CAR" refers to an engineered receptor that specifically introduces an antigen into the engineered cell (e.g., T cells, such as primitive T cells, central memory T cells, effector memory T cells, or combinations thereof) via an antigen-binding domain, thus combining the antigen-binding properties of the antigen-binding domain with the T cell activities (e.g., lysis and self-renewal) of the T cell. CARs typically include an extracellular antigen-binding domain (extracellular domain), a transmembrane domain, and an intracellular signaling domain. The intracellular signaling domain generally contains at least one ITAM signaling domain, for example derived from CD3ζ, and, depending on the case, at least one co-stimulatory signaling domain, for example derived from CD28 or 4-1BB. In the CARs presented herein, the VHH domain forms the antigen-binding domain and, when expressed in the cell, is located extracellularly.

[0175] "Affinity" refers to the strength of the sum of non-covalent interactions between a single binding site of a molecule (e.g., an antibody or a VHH-containing polypeptide) and its bound complex (e.g., an antigen). The affinity or apparent affinity of molecule X for its complex Y can generally be expressed by the dissociation constant (KD) or KD-apparent, respectively. Affinity can be measured by commonly used methods known in this technology (such as ELISA KD, KinExA, flow cytometry, and / or surface plasma resonance devices), including those described herein. These methods include (but are not limited to) methods involving BIAcore®, Octet®, or flow cytometry.

[0176] As used herein, the term "KD" refers to the equilibrium dissociation constant of antigen-binding molecules / antigen interactions. When the term "KD" is used herein, it includes both KD and KD-apparent.

[0177] In some embodiments, the KD of antigen-binding molecules is measured using flow cytometry, employing cell lines expressing the antigen and fitting the average fluorescence measured at each antibody concentration to a nonlinear unit-point equation (Prism Software graphpad). In some such embodiments, KD is KD-apparent.

[0178] The term "bioactivity" refers to any one or more biological properties of a molecule (whether naturally occurring in vivo or provided or achieved through recombinant means). Biological properties include (but are not limited to) binding ligands, inducing or increasing cell proliferation (such as T cell proliferation), and inducing or increasing the expression of cytokines.

[0179] "Affinity maturation" refers to VHH-containing peptides that have undergone such modifications compared to parental VHH-containing peptides without one or more modifications. These modifications improve the affinity of VHH-containing peptides for antigens.

[0180] As used herein, "humanized VHH" refers to a VHH in which one or more of its structural regions have been substantially replaced by human structural regions. In some cases, certain structural region (FR) residues of human immunoglobulins are replaced with corresponding non-human residues. Furthermore, humanized VHHs may include residues not found in the original VHH or the human structural sequence, but included to further improve and optimize the efficacy of the VHH or VHH-containing peptides. In some embodiments, the humanized VHH-containing peptide includes a human Fc region. It should be understood that humanized sequences can be identified by their primary sequence and do not necessarily represent the process of antibody production.

[0181] As used herein, "substantially similar" or "substantially identical" means that the similarity between two or more values ​​is high enough that, in the context of the biometrics measured by those values, a person skilled in the art would consider the difference between the two or more values ​​to be minimal or non-biologically and / or statistically significant. In some embodiments, the difference between two or more substantially similar values ​​does not exceed about any one of the following: 5%, 10%, 15%, 20%, 25%, or 50%.

[0182] A peptide "variant" is defined as a bioactive peptide that, after sequence alignment and, where necessary, the introduction of gaps to achieve the maximum percentage of sequence identity, and without considering any conserved substitutions as part of the sequence identity, has at least about 80% amino acid sequence identity with the native sequence peptide. Such variants include, for example, peptides with the addition or deletion of one or more amino acid residues at the N-terminus or C-terminus of the peptide. In some embodiments, the variant will have at least about 80% amino acid sequence identity. In some embodiments, the variant will have at least about 90% amino acid sequence identity. In some embodiments, the variant will have at least about 95% amino acid sequence identity with the native sequence peptide.

[0183] As used herein, the "percentage of amino acid sequence identity (%)" and "homology" for peptide, polypeptide, or antibody sequences are defined as the percentage of amino acid residues in a candidate sequence that are identical to amino acid residues in a specific peptide or polypeptide sequence, after alignment of sequences and, where necessary, the introduction of gaps to achieve the maximum percentage of sequence identity, without considering any conserved substitutions as part of the sequence identity. Alignment for determining the percentage of amino acid sequence identity can be performed in various ways within this technique, such as using publicly available computer software, such as BLAST, BLAST-2, ALIGN, or MEGALIGN™ (DNASTAR) software. Those skilled in this technique can determine the appropriate parameters used to measure alignment, including any algorithms required to achieve maximum alignment across the full length of the compared sequences.

[0184] Amino acid substitution may include (but is not limited to) replacing one amino acid in a polypeptide with another. Exemplary substitutions are shown in Table 2. Amino acid substitutions can be introduced into antibodies of interest and products screened for desired activities, such as retention / modification of antigen binding, reduction of immunogenicity, or improvement of ADCC or CDC.

[0185]

[0186] Amino acids can be grouped according to their common side chain characteristics: (1) Hydrophobic: Leucine, Met, Ala, Val, Leu, Ile; (2) Neutral hydrophilic: Cys, Ser, Thr, Asn, Gln; (3) Acidic: Asp, Glu; (4) Basic: His, Lys, Arg; (5) Residues that affect chain orientation: Gly, Pro; (6) Aromatic: Trp, Tyr, Phe.

[0187] Non-conservative substitution would require replacing members of one of these categories with members of another category.

[0188] The term "vector" is used to describe a polynucleotide that can be engineered to contain selected polynucleotides capable of replicating in a host cell. A vector may include one or more of the following elements: an origin of replication, one or more regulatory sequences (such as promoters and / or enhancers) that regulate the expression of the polypeptide of interest, and / or one or more selectable marker genes (such as antibiotic resistance genes and genes that can be used in colorimetric analysis, such as β-galactose). The term "expression vector" refers to a vector used to express the polypeptide of interest in a host cell.

[0189] "Host cell" refers to a cell that can be, or has been, a carrier or an acceptor of isolated polynucleotides. Host cells can be prokaryotic or eukaryotic. Exemplary eukaryotic cells include mammalian cells, such as primate or non-primate cells; fungal cells, such as yeast; plant cells; and insect cells. Non-limiting exemplary mammalian cells include (but are not limited to) NSO cells, PER.C6® cells (Crucell), and 293 and CHO cells, and their derivatives, such as 293-6E, CHO-DG44, CHO-K1, CHO-S, and CHO-DS cells. Host cells include the offspring of a single host cell, and the offspring may not necessarily be completely identical to the original parent cell (in terms of morphology or genomic DNA complementarity) due to natural, accidental, or intentional mutations. Host cells include cells transfected in vivo with the polynucleotides provided herein.

[0190] As used herein, the term "separated" refers to a molecule that has been separated from at least some of the components normally found in nature with or produced therefrom. For example, a polypeptide is called "separated" when it is separated from at least some of the components of the cell that produced it. In the case of a polypeptide being secreted by a cell after expression, the physical separation of the supernatant containing the polypeptide from the cell that produced it is considered "separated" polypeptide. Similarly, a polynucleotide is called "separated" when it is not part of a larger polynucleotide (such as genomic DNA or mitochondrial DNA in the case of DNA polynucleotides) that is normally found in nature, or when it is separated from at least some of the components of the cell that produced it (e.g., in the case of RNA polynucleotides). Therefore, DNA polynucleotides contained in carriers within host cells can be called "separated".

[0191] The terms "individual" and "subject" are used interchangeably herein and refer to an animal, such as a mammal. The term "patient" includes human and veterinary individuals. In some embodiments, methods of treating mammals are provided, including (but not limited to) humans, rodents, apes, felines, canines, equines, bovines, suidae, sheep, goats, laboratory mammals, livestock mammals, locusts mammals, and pet mammals. An individual may be male or female and may be of any age, including infants, young children, adolescents, adults, and elderly individuals. In some instances, "individual" or "subject" refers to an individual who requires treatment for a disease or condition. In some embodiments, an individual receiving treatment may be a patient, indicating that the individual has been identified as having a condition associated with the treatment or is at great risk of developing the condition. In certain embodiments, the individual is human, such as a human patient.

[0192] As used in this article, "disease" or "symptom" refers to a condition that requires and / or expects treatment.

[0193] Unless otherwise stated, the terms "tumor cell," "cancer cell," "cancer," "tumor," and / or "necrophyte" are used interchangeably herein and refer to cells exhibiting uncontrolled growth and / or abnormally increased cell survival and / or inhibited apoptosis, interfering with the normal function of organs and systems of the body. This definition includes benign and malignant cancers, polyps, hyperplasia, and latent tumors or micrometastases.

[0194] The terms "cancer" and "tumor" encompass solid cancers and hematologic / lymphomas, and also include malignant, potentially malignant, and benign growths such as dysplasia. This definition also includes cells with abnormal proliferation that the immune system cannot prevent (e.g., immune evasion and immune escape mechanisms) (e.g., virus-infected cells). Exemplary cancers include (but are not limited to): basal cell carcinoma, biliary tract cancer; bladder cancer; bone cancer; brain and central nervous system cancers; breast cancer; peritoneal cancer; cervical cancer; choriocarcinoma; colon and rectal cancer; connective tissue cancer; digestive system cancers; endometrial cancer; esophageal cancer; eye cancer; head and neck cancer; gastric cancer (including gastrointestinal cancer); glioblastoma; liver cancer; hepatoma; intraepithelial neoplasia; kidney cancer or renal cell carcinoma; laryngeal cancer; leukemia; liver cancer; lung cancer (e.g., small cell lung cancer, non-small cell lung cancer). Cancers include: lung cancer, lung adenocarcinoma, and squamous cell carcinoma; melanoma; myeloma; neuroblastoma; oral cancer (lip cancer, tongue cancer, oral cancer, and pharyngeal cancer); ovarian cancer; pancreatic cancer; prostate cancer; retinoblastoma; rhabdomyosarcoma; rectal cancer; respiratory system cancers; salivary gland cancer; sarcoma; skin cancer; squamous cell carcinoma; gastric cancer; testicular cancer; thyroid cancer; uterine or endometrial cancer; urinary system cancers; vulvar cancer; lymphoma, including Hodgkin's lymphoma. Lymphoma and non-Hodgkin's lymphoma, as well as B-cell lymphomas (including low-grade / follicular non-Hodgkin's lymphoma (NHL); small lymphocytic (SL) NHL; intermediate / follicular NHL; intermediate diffuse NHL; advanced immunoblastic NHL; advanced lymphoblastic NHL; advanced small nucleated cell NHL; mass lesion NHL; mantle cell lymphoma; AIDS-related lymphoma; and Waldenstrom's macroglobulinemia; chronic lymphocytic leukemia (CLL); acute lymphoblastic leukemia (ALL); hairy cell leukemia; chronic myeloblastic leukemia; and other carcinomas and sarcomas; and post-transplant lymphoproliferative disorders (PTLD), as well as abnormal angiogenesis associated with mammary spot disease, edema (such as that associated with brain tumors) and Meigs' syndrome.

[0195] As used herein, the term "non-tumor cell" refers to normal cells or tissues. Exemplary non-tumor cells include (but are not limited to): T cells, B cells, natural killer (NK) cells, natural killer T (NKT) cells, dendritic cells, monocytes, macrophages, epithelial cells, fibroblasts, hepatocytes, interstitial kidney cells, fibroblast-like synovial cells, osteoblasts, and cells located in the breast, skeletal muscle, pancreas, stomach, ovary, small intestine, placenta, uterus, testes, kidney, lung, heart, brain, liver, prostate, colon, lymphatic organs, bone, and bone-derived mesenchymal stem cells. As used herein, the term "peripheral cells or tissues" refers to non-tumor cells that are not located near tumor cells and / or within the tumor microenvironment.

[0196] As used herein, the term "cells or tissues within the tumor microenvironment" refers to cells, molecules, extracellular matrix, and / or blood vessels surrounding and / or feeding tumor cells. Exemplary cells or tissues within the tumor microenvironment include (but are not limited to): tumor vascular structures; tumor-infiltrating lymphocytes; fibroblastic reticular cells; endothelial progenitor cells (EPCs); cancer-associated fibroblasts; outer layer cells; other stromal cells; components of the extracellular matrix (ECM); dendritic cells; antigen-presenting cells; T cells; regulatory T cells (Treg cells); macrophages; neutrophils; bone marrow-derived suppressor cells (MDSCs); and other immune cells located near the tumor. As described below, methods well-known in this art for identifying tumor cells and / or cells / tissues within the tumor microenvironment are described in detail below.

[0197] In some embodiments, “increase” or “decrease” refers to a statistically significant increase or decrease, respectively. As will be apparent to those skilled in the art, “modulation” may also include, compared to conditions without the assay reagent, changes in the affinity, specificity, and / or selectivity of the target or antigen for its ligands, binding complexes, complexes used to associate into homopolymeric or heteropolymeric forms, or receptors (which may be increases or decreases); changes in the sensitivity of the target or antigen to one or more conditions (such as pH, ionic strength, presence of cofactors, etc.) in the medium or surrounding environment in which the target or antigen is present (which may be increases or decreases); and / or cell proliferation or cytokine production. Depending on the target involved, this may be determined by any suitable method and / or using any suitable analysis known herein or described herein.

[0198] As used in this article, "immune response" means a cellular and / or humoral immune response that is sufficient to suppress or prevent the onset of disease or alleviate disease symptoms (such as cancer or metastasis). "Immune response" can include both innate and adaptive immune system states.

[0199] As used herein, the terms "treating," "treatment," or "therapy" for disease, symptom, or condition refer to a method of achieving a beneficial or desired clinical outcome. As used herein, "treatment" encompasses any administration or application of a therapeutic agent for a disease in mammals, including humans. For the purposes of this invention, beneficial or desired clinical outcomes include, but are not limited to, any one or more of the following: relief of one or more symptoms, reduction of disease severity, prevention or delay of disease spread (e.g., metastasis, such as to the lungs or lymph nodes), prevention or delay of disease recurrence, delay or slow disease progression, improvement of disease symptoms, inhibition of disease or disease progression, inhibition or slowing of disease or its progression, arrest of its development, and remission (whether partial or complete). "Treatment" also encompasses the reduction of the pathological consequences of proliferative diseases. The methods provided herein cover any one or more of these therapeutic approaches. According to the foregoing, the term "treatment" does not require the complete removal of all aspects of a symptom.

[0200] In the context of cancer, the terms “treatment” or “inhibit,” “inhibiting,” or “inhibition” used herein to refer to at least one of the following: a statistically significant reduction in the rate of tumor growth; cessation of tumor growth; or a reduction in tumor size, mass, metabolic activity, or volume as measured by indicators such as (but not limited to) the Response Evaluation Criteria for Solid Tumors (RECIST); or a statistically significant increase in progression-free survival (PFS) or overall survival (OS).

[0201] "Improvement" means that one or more symptoms are reduced or improved compared to when no treatment was given. "Improvement" also includes a reduction or decrease in the duration of symptoms.

[0202] "Preventing," "prophylaxis," or "prevention" of a disease or condition means administering a pharmaceutical composition, alone or in combination with another compound, to prevent the occurrence or onset of a disease or condition or some or all of its symptoms, or to reduce the likelihood of the onset of a disease or condition.

[0203] The terms "inhibition" or "inhibit" refer to a reduction or cessation of any phenotypic trait, or a decrease or cessation of the occurrence, extent, or likelihood of that trait. "Reduction" or "inhibit" means a reduction, decrease, or cessation of activity, function, and / or quantity relative to a reference. In some embodiments, "reduction" or "inhibit" means a reduction of 10% or more in the total population. In some embodiments, "reduction" or "inhibit" means a reduction of 50% or more in the total population. In some embodiments, "reduction" or "inhibit" means a reduction of 75%, 85%, 90%, 95%, or more in the total population. In some embodiments, the quantities described above are inhibited or reduced relative to a control over the same period of time.

[0204] As used herein, "delaying disease progression" means slowing, hindering, mitigating, blocking, stabilizing, containing, and / or delaying the development of a disease (such as cancer). This delay can vary in length, depending on the disease being treated and / or the individual's medical history. As will be apparent to those skilled in this technique, sufficient or significant delay can effectively encompass prevention, as the individual does not develop the disease. For example, it can delay the development of advanced cancers, such as metastases.

[0205] As used herein, “prevention” includes providing preventative measures against the onset or recurrence of a disease in individuals who may be susceptible to the disease but have not yet been diagnosed with it. Unless otherwise specified, the terms “reduction,” “suppression,” or “prevention” do not imply or require constant, complete prevention, but only for the period in which they are measured.

[0206] The term "anticancer agent" is used in its broadest sense herein to refer to an agent used to treat one or more cancers. Examples of such agents include (but are not limited to) chemotherapeutic agents, anticancer biologics (such as interleukins, receptor extracellular domain-Fc fusions, and antibodies), radiation therapy, CAR-T therapy, therapeutic oligonucleotides (such as antisense oligonucleotides and siRNA), and oncolytic viruses.

[0207] The term "biological sample" means a quantity of material derived from living organisms or previously living organisms. Such material includes (but is not limited to) blood (e.g., whole blood), plasma, serum, urine, amniotic fluid, synovial fluid, endothelial cells, white blood cells, monocytes, other cells, organs, tissues, bone marrow, lymph nodes, and spleen.

[0208] The terms "control" or "reference" refer to a composition known to be free of the analyte ("negative control") or a composition known to contain the analyte ("positive control"). A positive control may contain a known concentration of the analyte.

[0209] The term "effective dose" or "therapeutic effective dose" refers to the amount and / or concentration at which a composition containing an active ingredient (e.g., sdAb or VHH-containing peptide), when administered to a patient alone (i.e., as a monotherapy) or in combination with other therapeutic agents, statistically significantly reduces disease progression, for example, by improving or eliminating symptoms and / or the cause of disease. An effective dose can be an amount that reduces, diminishes, or alleviates at least one symptom or biological response or effect associated with the disease or condition, prevents disease or condition progression, or improves the patient's physiological function. Therapeutic effective doses of compositions containing active agents can vary depending on factors such as disease symptom, individual age, sex, weight, and the ability of the active agent to elicit the desired response in the individual. A therapeutic effective dose is also an amount in which the beneficial therapeutic effect outweighs any toxic or harmful effects of the active agent. A therapeutic effective dose can be delivered in one or more administrations. A therapeutic effective dose refers to an amount at which the desired therapeutic and / or preventative outcome can be effectively achieved at the necessary dose and time.

[0210] As used herein, a composition means any mixture of two or more products, substances, or compounds (including cells). It may be a solution, suspension, liquid, powder, paste, aqueous solution, non-aqueous solution, or any combination thereof.

[0211] The terms "pharmaceutical formulation" and "pharmaceutical composition" refer to preparations in a form that allows for the bioactivity of the active ingredient and contains no other components that would cause unacceptable toxicity to the individual to whom the formulation will be administered. Therefore, they are compositions suitable for pharmaceutical use in mammalian individuals, often humans. Pharmaceutical compositions typically contain an effective amount of an active agent (e.g., an sdAb or a VHH-containing peptide) and a carrier, excipient, or diluent. The carrier, excipient, or diluent is typically a pharmaceutically acceptable carrier, excipient, or diluent. Such formulations may be sterile.

[0212] "Pharmaceutically acceptable carriers" refer to non-toxic solid, semi-solid, or liquid fillers, diluents, encapsulating materials, formulation aids, or carriers known in this art, used together with therapeutic agents to form a "pharmaceutical composition" administered to an individual. Pharmaceutically acceptable carriers are non-toxic to the recipient at the dosage and concentration used and are compatible with other components of the formulation. Pharmaceutically acceptable carriers are suitable for the formulation used.

[0213] Administering in combination with one or more other therapeutic agents includes simultaneous (parallel) administration and sequential administration in any order.

[0214] The term "parallel" is used herein to refer to the administration of two or more therapeutic agents, wherein at least part of the administration overlaps in time, or the administration of one therapeutic agent falls within a relatively short period of time relative to the administration of another therapeutic agent, or the therapeutic effects of the two agents overlap for at least a period of time.

[0215] The term "in sequence" is used in this document to refer to the administration of two or more therapeutic agents in a non-overlapping manner in time, or where the therapeutic effects of such agents do not overlap.

[0216] As used in this article, "combination" means administering one treatment modality in addition to another. Therefore, "combination" means administering one treatment modality before, during, or after administering one treatment modality to an individual.

[0217] The term "instructions for use" refers to the instructions typically included in the commercial packaging of therapeutic products, which contain information about indications, usage, dosage, administration, combination therapy, contraindications and / or warnings related to the use of such therapeutic products.

[0218] "Article" means any manufactured article (e.g., packaging or container) or kit containing at least one reagent, such as a medicine for treating a disease or condition (e.g., cancer), or a probe for specifically detecting the biomarkers described herein. In some embodiments, the manufactured article or kit is marketed, distributed, or sold in unit form for performing the methods described herein.

[0219] The terms "label" and "detectable label" refer to, for example, a portion attached to an antibody or antigen to make the reaction (e.g., binding) between members of a specific binding pair detectable. The labeled member of a specific binding pair is called a "detectably labeled" protein. Therefore, the term "labeled binding protein" refers to a protein incorporating a label for identification of binding proteins. In some embodiments, the label can be a detectable label that produces a signal detectable by visual or instrumental means, such as a radiolabeled amino acid or a polypeptide attached to a biotinylated portion detectable by labeled avidin (e.g., streptomycin containing a fluorescent label or enzyme activity detectable by optical or colorimetric methods). Examples of labels used for peptides include (but are not limited to) the following: radioisotopes or radionuclides (e.g., 3H, 14C, 35S, 90Y, 99Tc, 111In, 125I, 131I, 177Lu, 166Ho, or 153Sm); chromogens, fluorescent labels (e.g., FITC, rhodamine, lanthanide phosphor), enzyme labels (e.g., horseradish peroxidase, luciferase, alkaline phosphatase); chemiluminescent labels; biotinyl groups; predetermined peptide antigenic determinants recognized by secondary reporter antibodies (e.g., leucine zipper pairs, binding sites of secondary antibodies, metal-binding domains, antigenic determinant tags); and magnetizing agents, such as thiocyanate chelates. Representative examples of labels commonly used in immunoassays include light-producing portions, such as acrylamide compounds, and fluorescent portions, such as luciferin. In this respect, a part may be undetectably marked on its own, but may become detectable after reacting with another part. [II. Combining the VHH field of DLL3]

[0220] This document provides a DLL3-binding polypeptide, which is a VHH-containing polypeptide containing at least one VHH domain specifically binding to DLL3. In some embodiments, the VHH domain binds human DLL3. In any of the embodiments provided, the VHH domain binds DLL3 having the sequence shown in SEQ ID NO: 86 or a mature form thereof, which lacks a signal sequence. In any of the embodiments provided, the VHH domain binds DLL3 having the sequence shown in SEQ ID NO: 87 or a mature form thereof, which lacks a signal sequence.

[0221] In some embodiments, the VHH-containing polypeptide has multiple copies of the VHH domain provided herein. In such embodiments, the VHH-containing polypeptide may have multiple copies of the same VHH domain. In some embodiments, the VHH-containing polypeptide may have multiple copies of the VHH domain, which, although different, recognize the same antigenic determinant on DLL3. The VHH-containing polypeptide can take many forms, including any of the forms described in Section III below.

[0222] The VHH domain is an antibody fragment, which is a single monomeric antibody variable domain capable of selectively binding to a specific antigen. With a molecular weight of only 12-15 kDa, the VHH domain (also known as a single-domain antibody) is much smaller than common antibodies (150-160 kDa) composed of two protein heavy chains and two light chains, and even smaller than Fab fragments (about 50 kDa, one light chain and half a heavy chain) and single-chain variable fragments (about 25 kDa, two variable domains: one from the light chain and one from the heavy chain).

[0223] A single-domain antibody is an antibody whose complementarity-determining region is a portion of a single-domain polypeptide. Examples include (but are not limited to) heavy-chain antibodies, naturally occurring antibodies lacking a light chain, single-domain antibodies derived from known four-chain antibodies, engineered antibodies, and single-domain backbones other than those derived from the antibody backbone. Single-domain antibodies can be derived from any species, including but not limited to mice, humans, camels, llamas, alpacas, llamas, chestnut llamas, sharks, goats, rabbits, and / or cattle. In some embodiments, the single-domain antibody used herein is a naturally occurring single-domain antibody, referred to as a light-chain-deficient heavy-chain antibody. For clarity, the variable domain of this naturally occurring light-chain-deficient heavy-chain molecule is referred to herein as VHH to distinguish it from the known VH of four-chain immunoglobulins. Such VHH molecules can be derived from antibodies produced in camel species, such as camels, llamas, dromedaries, alpacas, llamas, and chestnut llamas. Other species besides camels can produce naturally occurring light-chain-deficient heavy-chain antibodies; such VHHs are within the scope of this invention.

[0224] Methods for screening VHH domains with desired specificity for DLL3, including VHH-binding peptides, include (but are not limited to) enzyme-linked immunosorbent assay (ELISA), enzyme assays, flow cytometry, and other immune-mediated techniques known in this field.

[0225] The VHH domains provided in this document are, in particular, DLL3 VHH (derived from the llama) and humanized sequences, such as any of those described below.

[0226] In some embodiments, the VHH domain of DLL3 can be humanized. Humanized antibodies (such as VHH-containing peptides) are suitable as therapeutic molecules because they reduce or eliminate human immune responses to non-human antibodies, which can induce immune responses to antibody therapeutics and reduce the efficacy of the therapeutics. Generally, humanized antibodies comprise one or more variable domains, wherein the CDR (or a portion thereof) is derived from a non-human antibody, and the FR (or a portion thereof) is derived from a human antibody sequence. Humanized antibodies may also include at least a portion of a human constant region. In some embodiments, some FR residues in a humanized antibody are substituted with corresponding residues from a non-human antibody (e.g., an antibody from which the CDR residues are derived), thereby, for example, restoring or improving antibody specificity or affinity.

[0227] Humanized antibodies and their manufacturing methods are reviewed in, for example, Almagro and Fransson, (2008) Front. Biosci. 13:1619-1633, and further described in, for example, Riechmann et al., (1988) Nature 332:323-329; Queen et al., (1989) Proc. Natl Acad. Sci USA 86:10029-10033; US Patent Nos. 5,821,337, 7,527,791, 6,982,321 and 7,087,409; Kashmiri et al., (2005) Methods 36:25-34; Padlan, (1991) Mol. Immunol. 28:489-498 (describing "surface remodeling"); Dall'Acqua et al., (2005) Methods 36:43-60 (describes "FR reorganization"); and Osbourn et al., (2005) Methods 36:61-68 and Klimka et al., (2000) Br.J. Cancer, 83:252-260 (describes the "guided selection" method for FR reorganization).

[0228] Human scaffold regions that can be used for humanization include (but are not limited to): scaffold regions selected using the "best fit" method (see, for example, Sims et al. (1993) J. Immunol. 151:2296); and scaffold regions derived from the common sequences of human antibodies with specific subsets of heavy chain variable regions (see, for example, Carter et al. (1992) Proc. Natl. Acad. Sci. USA, 89:4285; and Presta et al. (1993) J. Immunol, 151:262). 3) Human maturation (somatic mutation) scaffold regions or human germline scaffold regions (see, for example, Almagro and Fransson, (2008) Front. Biosci. 13: 1619-1633); and scaffold regions derived from screened FR libraries (see, for example, Baca et al., (1997) J. Biol. Chem. 272: 10678-10684 and Rosok et al., (1996) J. Biol. Chem. 271: 22611-22618). Humanized VHHs are typically prepared by substituting human FR regions with FR regions. In some embodiments, certain FR residues of human FRs are substituted to modify one or more properties of the humanized VHH. VHH domains with such substituted residues are still referred to herein as "humanized".

[0229] This document provides a VHH domain incorporating DLL3, wherein the VHH domain comprises a VHH amino acid sequence selected from any of SEQ ID NO: 102, 244-318, 401-409, 416, 455 and 476-480-488, 507-518, or CDR1, CDR2, and CDR3 contained in an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO: 102, 244-318, 401-409, 416, 455 and 476-480-488, 507-518. In some embodiments, the DLL3 VHH domain provided herein contains CDR1 shown in any of SEQ ID NO:319-335 or 456, CDR2 shown in any of SEQ ID NO:336-353, 384, 410 and 411, and CDR3 shown in any of SEQ ID NO:354-367, 395 and 412-415. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in any of SEQ ID NO: 102, 244-318, 401-409, 416, 455 and 476-480-488, 507-518, or an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identical to the VHH region amino acid selected from any of SEQ ID NO: 102, 244-318, 401-409, 416, 455 and 476-480-488, 507-518. In some embodiments, the DLL3 VHH domain has an amino acid sequence as shown in any one of SEQ ID NO: 102, 244-318, 401-409, 416, 455 and 476-480-488, 507-518.

[0230] This document provides a VHH domain incorporating DLL3, wherein the VHH domain comprises a VHH amino acid sequence selected from any of SEQ ID NO: 244-318 and 455, or CDR1, CDR2, and CDR3 contained in an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO: 244-318 and 455. In some embodiments, the DLL3 VHH domain provided herein comprises CDR1 shown in any of SEQ ID NO: 319-335, CDR2 shown in any of SEQ ID NO: 336-353, and CDR3 shown in any of SEQ ID NO: 354-367. The provided DLL3 VHH domain has an amino acid sequence shown in any of SEQ ID NO: 1-114 or an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identical to the VHH region amino acid selected from any of SEQ ID NO: 244-318. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in any of SEQ ID NO: 244-318 and 455.

[0231] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2, and CDR3 contained in the VHH domain of an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected VHH region amino acid shown in SEQ ID NO:244. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in SEQ ID NO:244 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected amino acid shown in SEQ ID NO:244. In some embodiments, the DLL3 VHH domain is a humanized variant of the amino acid sequence shown in SEQ ID NO:244.

[0232] In some embodiments, the DLL3 VHH domain provided herein contains CDR1 as shown in any one of SEQ ID NO: 319, 320, 321, 322, 323, 324, 325, 326; CDR2 as shown in any one of SEQ ID NO: 336, 337, 338; and CDR3 as shown in SEQ ID NO: 354.

[0233] In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 319, 336, and 354, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 319, 337, and 354, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 319, 338, and 354, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 338, and 354, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 321, 338, and 354, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 322, 338, and 354, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 323, 338, and 354, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 324, 338, and 354, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 325, 338, and 354, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 326, 338, and 354, respectively.

[0234] In some states, the VHH domain of DLL3 includes a VHH amino acid sequence selected from any of SEQ ID NO:245-257 or CDR1, CDR2, and CDR3 contained in an amino acid sequence that has at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:245-257.

[0235] In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO: 245-257 or an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identical to the VHH region amino acids selected from any of SEQ ID NO: 245-257. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO: 245-257.

[0236] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2, and CDR3 contained in the VHH domain of an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected VHH region amino acid shown in SEQ ID NO:258. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in SEQ ID NO:258 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected amino acid shown in SEQ ID NO:258. In some embodiments, the DLL3 VHH domain is a humanized variant of the amino acid sequence shown in SEQ ID NO:258.

[0237] In some embodiments, the DLL3 VHH domain provided herein contains CDR1 shown in SEQ ID NO:327, CDR2 shown in SEQ ID NO:339, and CDR3 shown in SEQ ID NO:355.

[0238] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2 and CDR3 as shown in any of SEQ ID NO: 327, 339 and 355, respectively.

[0239] In some states, the VHH domain of DLL3 includes a VHH amino acid sequence selected from any of SEQ ID NO:259-263 or CDR1, CDR2, and CDR3 contained in an amino acid sequence that has at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:259-263.

[0240] In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO: 259-263 or an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identical to the VHH region amino acids selected from any of SEQ ID NO: 259-263. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO: 259-263.

[0241] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2, and CDR3 contained in the VHH domain of an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected VHH region amino acid shown in SEQ ID NO:264. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in SEQ ID NO:264 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected amino acid shown in SEQ ID NO:264. In some embodiments, the DLL3 VHH domain is a humanized variant of the amino acid sequence shown in SEQ ID NO:264.

[0242] In some embodiments, the DLL3 VHH domain provided herein contains CDR1 shown in SEQ ID NO:328, 329 or 456, CDR2 shown in SEQ ID NO:340 and CDR3 shown in SEQ ID NO:356.

[0243] In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 328, 340, and 356, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 329, 340, and 356, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 456, 340, and 356, respectively.

[0244] In some states, the VHH domain of DLL3 includes a VHH amino acid sequence selected from any of SEQ ID NO:265-274, 416, 455 or 476-478, or CDR1, CDR2 and CDR3 contained in an amino acid sequence that has at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:265-274, 416, 455 or 476-478.

[0245] In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO: 265-274, 416, 455, or 476-478, or an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identical to the VHH region amino acid selected from any of SEQ ID NO: 265-274, 416, 455, or 476-478. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO: 265-274, 416, 455, or 476-478.

[0246] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2, and CDR3 contained in the VHH domain of an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected VHH region amino acid shown in SEQ ID NO: 275. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in SEQ ID NO: 275 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected amino acid shown in SEQ ID NO: 105. In some embodiments, the DLL3 VHH domain is a humanized variant of the amino acid sequence shown in SEQ ID NO: 275.

[0247] In some embodiments, the DLL3 VHH domain provided herein contains CDR1 shown in SEQ ID NO:320, CDR2 shown in SEQ ID NO:341, and CDR3 shown in SEQ ID NO:357.

[0248] In some states, the VHH domain of DLL3 includes a VHH amino acid sequence selected from any of SEQ ID NO:276-279 or 479, or CDR1, CDR2, and CDR3 contained in an amino acid sequence that has at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:276-279 or 479.

[0249] In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO: 276-279 or 479, or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acids selected from any of SEQ ID NO: 276-279 or 479. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO: 276-279 or 479.

[0250] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2, and CDR3 contained in the VHH domain of an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected VHH region amino acid shown in SEQ ID NO:280. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in SEQ ID NO:280 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected amino acid shown in SEQ ID NO:280. In some embodiments, the DLL3 VHH domain is a humanized variant of the amino acid sequence shown in SEQ ID NO:280.

[0251] In some embodiments, the DLL3 VHH domain provided herein contains CDR1 shown in SEQ ID NO:330, CDR2 shown in SEQ ID NO:342, and CDR3 shown in SEQ ID NO:358.

[0252] In some states, the VHH domain of DLL3 includes a VHH amino acid sequence selected from any of SEQ ID NO:281-286 or CDR1, CDR2, and CDR3 contained in an amino acid sequence that has at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:281-286.

[0253] In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO: 281-286 or an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identical to the VHH region amino acids selected from any of SEQ ID NO: 281-286. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO: 281-286.

[0254] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2, and CDR3 contained in the VHH domain of an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected VHH region amino acid shown in SEQ ID NO:287. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in SEQ ID NO:287 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected amino acid shown in SEQ ID NO:287. In some embodiments, the DLL3 VHH domain is a humanized variant of the amino acid sequence shown in SEQ ID NO:287.

[0255] In some embodiments, the DLL3 VHH domain provided herein contains CDR1 as shown in SEQ ID NO:320; CDR2 as shown in any one of SEQ ID NO:345, 346, 347; and CDR3 as shown in any one of SEQ ID NO:359, 360, 361.

[0256] In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 345, and 359, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 346, and 359, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 347, and 359, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 345, and 360, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 345, and 361, respectively. In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2 and CDR3 as shown in any of SEQ ID NO: 320, 347 and 360, respectively.

[0257] In some states, the VHH domain of DLL3 includes a VHH amino acid sequence selected from any of SEQ ID NO:288-298 or 102, or CDR1, CDR2, and CDR3 contained in an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:288-298 or 102.

[0258] In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO: 288-298 or 102, or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acids selected from any of SEQ ID NO: 288-298 or 102. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO: 288-298 or 102.

[0259] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2, and CDR3 contained in the VHH domain of an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected VHH region amino acid shown in SEQ ID NO: 299. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in SEQ ID NO: 299 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected amino acid shown in SEQ ID NO: 299. In some embodiments, the DLL3 VHH domain is a humanized variant of the amino acid sequence shown in SEQ ID NO: 299.

[0260] In some embodiments, the DLL3 VHH domain provided herein contains CDR1 as shown in SEQ ID NO:331; CDR2 as shown in any one of SEQ ID NO:348, 349, 350; and CDR3 as shown in SEQ ID NO:356.

[0261] In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 331, 348, and 356, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 331, 349, and 356, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 331, 350, and 356, respectively.

[0262] In some states, the VHH domain of DLL3 includes a VHH amino acid sequence selected from any of SEQ ID NO:300-305 or 480, or CDR1, CDR2, and CDR3 contained in an amino acid sequence that has at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:300-305 or 480.

[0263] In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO: 300-305 or 480, or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acids selected from any of SEQ ID NO: 300-305 or 480. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO: 300-305 or 480.

[0264] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2, and CDR3 contained in the VHH domain of an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected VHH region amino acid shown in SEQ ID NO: 306. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in SEQ ID NO: 306 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected amino acid shown in SEQ ID NO: 306. In some embodiments, the DLL3 VHH domain is a humanized variant of the amino acid sequence shown in SEQ ID NO: 306.

[0265] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2, and CDR3 contained in the VHH domain of an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected VHH region amino acid shown in SEQ ID NO: 507. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in SEQ ID NO: 507 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected amino acid shown in SEQ ID NO: 507. In some embodiments, the DLL3 VHH domain is a humanized variant of the amino acid sequence shown in SEQ ID NO: 507.

[0266] In some embodiments, the DLL3 VHH domain provided herein contains CDR1 as shown in SEQ ID NO:332; CDR2 as shown in any one of SEQ ID NO:348, 349, 350; and CDR3 as shown in SEQ ID NO:362.

[0267] In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 332, 348, and 362, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 332, 349, and 362, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 332, 350, and 362, respectively.

[0268] In some states, the VHH domain of DLL3 includes a VHH amino acid sequence selected from any of SEQ ID NO:307-313 or CDR1, CDR2, and CDR3 contained in an amino acid sequence that has at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:307-313.

[0269] In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO: 307-313 or an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identical to the VHH region amino acids selected from any of SEQ ID NO: 307-313. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO: 307-313.

[0270] In some states, the VHH domain of DLL3 includes a VHH amino acid sequence selected from any of SEQ ID NO:508-514 or CDR1, CDR2, and CDR3 contained in an amino acid sequence that has at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:508-514.

[0271] In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO: 508-514 or an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identical to the VHH region amino acids selected from any of SEQ ID NO: 508-514. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO: 508-514.

[0272] In some embodiments, the DLL3 VHH domain provided herein contains CDR1, CDR2, and CDR3 contained in the VHH domain of an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected VHH region amino acid shown in SEQ ID NO: 401. In some embodiments, the DLL3 VHH domain has an amino acid sequence shown in SEQ ID NO: 401 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the selected amino acid shown in SEQ ID NO: 401. In some embodiments, the DLL3 VHH domain is a humanized variant of the amino acid sequence shown in SEQ ID NO: 401.

[0273] In some embodiments, the DLL3 VHH domain provided herein contains CDR1 as shown in SEQ ID NO:320; CDR2 as shown in any one of SEQ ID NO:384, 410, 411; and CDR3 as shown in any one of SEQ ID NO:395, 412, 413, 414, 415.

[0274] In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 384, and 395, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 410, and 395, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 411, and 395, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 384, and 412, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 384, and 413, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 384, and 414, respectively. In some embodiments, the DLL3 VHH field provided herein contains CDR1, CDR2, and CDR3 as shown in any one of SEQ ID NO: 320, 384, and 415, respectively.

[0275] In some states, the VHH domain binding to DLL3 includes CDR1, CDR2, and CDR3 contained in an amino acid sequence selected from any of SEQ ID NO:402-409 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:402-409. In some states, the VHH domain binding to DLL3 includes CDR1, CDR2, and CDR3 contained in an amino acid sequence selected from any of SEQ ID NO:481-488 or an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with the VHH region amino acid selected from any of SEQ ID NO:481-488.

[0276] In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO:402-409 or an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identical to the VHH region amino acids selected from any of SEQ ID NO:402-409. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO:402-409. In some cases, the provided DLL3 VHH domain is a humanized variant having an amino acid sequence shown in any of SEQ ID NO:481-488 or an amino acid sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identical to the VHH region amino acids selected from any of SEQ ID NO:481-488. In some embodiments, the DLL3 humanized VHH domain has an amino acid sequence shown in any of SEQ ID NO:481-488. [III. Fusion proteins and conjugates containing DLL3-binding peptides]

[0277] This document provides fusion proteins and conjugates containing DLL3-binding peptides, wherein the DLL3-binding peptides contain at least one DLL3-specific VHH domain directly or indirectly linked to one or more other domains or portions thereof. In some embodiments, the fusion proteins or conjugates of the present invention consist of a single peptide. In other embodiments, the fusion proteins or conjugates of the present invention consist of more than one peptide. In some embodiments, the DLL3-binding peptides of the present invention also have at least one DLL3-specific VHH domain. In some forms, the DLL3-binding peptides are multivalent. In some embodiments, the DLL3-binding peptides include two or more replicas that specifically bind to the VHH domain of DLL3, such as three or more, four or more, five or more, or six or more replicas that specifically bind to the VHH domain of DLL3. In some forms, the DLL3-binding peptides are multispecific. For example, in some cases, one or more other domains may be binding to one or more other antigens or proteins.

[0278] In some embodiments, the DLL3-binding polypeptide of the present invention comprises two or more polypeptide sequences operatively linked via amino acid linkers. In some embodiments, such linkers are primarily composed of the amino acids glycine and serine, referred to herein as GS linkers. The GS linkers of the fusion protein of the present invention may have various lengths, for example, lengths of 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids. In some embodiments, the GS linker comprises an amino acid sequence selected from the group consisting of: GGSGGS, i.e., (GGS)2 (SEQ ID NO:1); GGSGGSGGS, i.e., (GGS)3 (SEQ ID NO:2); GGSGGSGGSGGS, i.e., (GGS)4 (SEQ ID NO:3); and GGSGGSGGSGGSGGS, i.e., (GGS)5 (SEQ ID NO:4). In some embodiments, the linker is a flexible linker containing glycine residues, such as (by means of non-limiting examples) GG, GGG, GGGG (SEQ ID NO:5), GGGGG (SEQ ID NO:6), and GGGGGG (SEQ ID NO:7). In some embodiments, the DDL3-binding polypeptide comprises a combination of a GS linker and a glycine linker. In some embodiments, the linker is (GGGGS)n, where n is 1 to 5 (SEQ ID NO: 123); (GGGGGS)n, where n is 1 to 4 (SEQ ID NO: 124); GGGGS (SEQ ID NO: 125); GGGGGS (SEQ ID NO: 126); GGGGGSGGGGGSGGGGS (SEQ ID NO: 127); GGGGSGGGGSGGGGS (SEQ ID NO: 128); GGSGGGGSGGGGSGGGGS (SEQ ID NO: 129); or PGGGG (SEQ ID NO: 450). In some embodiments, the linker is a GS linker. In some embodiments, the DLL3-binding polypeptide comprises a combination of a GS linker and a glycine linker.

[0279] [A.Fc fusion]

[0280] This article provides a DLL3-binding polypeptide, which is a fusion protein containing at least one VHH domain and Fc domain of the DLL3-binding polypeptide provided herein. In some embodiments, the DLL3-binding polypeptide provided herein comprises one, two, three, or four VHH domains and Fc domains of the DLL3-binding polypeptide.

[0281] In some embodiments, the incorporation of the immunoglobulin Fc region in the fusion protein may be composed of two polypeptides that together form a dimer. In some embodiments, the Fc domain mediates the dimerization of the DLL3-binding polypeptide under physiological conditions (such as when expressed by cells) to form a dimer that doubles the number of DLL3 binding sites. For example, a DLL3-binding polypeptide comprising three DLL3-binding VHH domains and an Fc region is in trivalent monomeric form, but the Fc region can mediate dimerization such that the DLL3-binding polypeptide exists as a hexavalent dimer under these conditions. In some embodiments, the DLL3 VHH domain is fused to the IgG Fc region, and in these embodiments, the fusion protein is a divalent molecule with two DLL3 VHH domains per molecule. In some embodiments, two DLL3-binding domains (2×) are fused to the IgG Fc region, and in these embodiments, the fusion protein is a tetravalent molecule with four DLL3 VHH domains per molecule. In some embodiments, three DLL3 binding domains (3×) are fused to the IgG Fc region, and in these embodiments, the fusion protein is a hexavalent molecule with six DLL3 VHH domains per molecule.

[0282] In some embodiments, the multivalent DLL3-binding polypeptide is bivalent. In some embodiments, the bivalent DLL3-binding polypeptide of the present invention comprises two copies of a DLL3-binding polypeptide having the following structure: (DLL3 VHH)-linker-Fc. In some embodiments, the multivalent DLL3-binding polypeptide is tetravalent. In some embodiments, the tetravalent DLL3-binding polypeptide of the present invention comprises two copies of a DLL3 polypeptide having the following structure: (DLL3 VHH)-linker-(DLL3 VHH)-linker-Fc. In some embodiments, the multivalent DLL3-binding polypeptide is hexavalent. In some embodiments, the hexavalent DLL3-binding polypeptide of the present invention comprises two copies of a DLL3 polypeptide having the following structure: (DLL3 VHH)-linker-(DLL3 VHH)-linker-(DLL3 VHH)-linker-Fc.

[0283] In some cases, the CH3 domain of the Fc region can be used as a homodimerizing domain, so that the resulting fusion protein is formed from two identical polypeptides. In other cases, the CH3 dimer intercalation region of the Fc region can be mutated to enable heterodimerization. For example, the heterodimerizing domain can be incorporated into the fusion protein to make the construct an asymmetric fusion protein.

[0284] In any of the provided embodiments, the DLL3 VHH field can be any of the VHH fields described above. In some embodiments, the DLL3 VHH field is a humanized VHH field incorporating DLL3.

[0285] In various embodiments, the Fc domain included in the DLL3-binding polypeptide is a human Fc domain or is derived from a human Fc domain. In some embodiments, the fusion protein contains an immunoglobulin Fc region. In some embodiments, the immunoglobulin Fc region is an IgG isotype selected from the group consisting of IgG1 isotype, IgG2 isotype, IgG3 isotype, and IgG4 subtype.

[0286] In some embodiments, the immunoglobulin Fc region or its immunoactive fragment is an IgG isotype. For example, the immunoglobulin Fc region of the fusion protein is a human IgG1 isotype, which has the following amino acid sequence: (SEQ ID NO:8)

[0287] In some embodiments, the immunoglobulin Fc region or its immunoactive fragment comprises a human IgG1 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO:8.

[0288] In some embodiments, where the fusion protein of the present invention comprises an Fc polypeptide, the Fc polypeptide is mutated or modified. In some cases, the mutation comprises one or more amino acid substitutions that reduce the effector function of the Fc polypeptide. Various examples of Fc polypeptide mutations that alter, such as those reducing effector function, are known, including any of those described below. In some embodiments, references to amino acid substitutions in the Fc region are made by way of the Kabat EU number (also known as the Kabat number) unless described with respect to a specific SEQ ID NO. EU numbers are known and based on the latest IMGT Scientific Chart (IMGT®, the international ImMunoGeneTics information system®, http: / / www.imgt.org / IMGTScientificChart / Numbering / Hu_IGHGnber.html (established: May 17, 2001, last updated: January 10, 2013) and the EU index reported in Kabat, EA et al., Sequences of Proteins of Immunological Interest, 5th edition, US Department of Health and Human Services, NIH Publication No. 91-3242 (1991).

[0289] In some embodiments, Fc regions exhibiting reduced effector functions may be ideal candidates for applications requiring binding to DLL3 or CD3 but where certain effector functions (such as CDC and ADCC) are unnecessary or harmful. In vitro and / or in vivo cytotoxicity assays can be performed to confirm the reduction / elimination of CDC and / or ADCC activity. For example, Fc receptor (FcR) binding assays can be performed to ensure that multispecific polypeptide constructs and / or their cleavage components lack FcγR binding (and therefore may lack ADCC activity), but retain FcRn binding capacity. Primary NK cells, used to mediate ADCC, express only FcγRIII, while monocytes express FcγRI, FcγRII, and FcγRIII. Non-limiting examples of in vitro analyses for evaluating the ADCC activity of molecules of interest are described below: U.S. Patent No. 5,500,362 (see, for example, Hellstrom, I. et al., Proc. Nat'l Acad. Sci. USA 83:7059-7063 (1986)) and Hellstrom, I. et al., Proc. Nat'l Acad. Sci. USA 82:1499-1502 (1985); U.S. Patent No. 5,821,337 (see Bruggemann, M. et al., J. Exp. Med. 166:1351-1361 (1987)). Alternatively, non-radioactive analytical methods can be used (see, for example, the ACTITM non-radioactive cytotoxicity assay for flow cytometry (Cell Technology, Inc. Mountain View, Calif.); and the CytoTox 96TM non-radioactive cytotoxicity assay (Promega, Madison, Wis.)). Effector cells suitable for such analyses include peripheral blood mononuclear cells (PBMCs) and natural killer (NK) cells. Alternatively, the ADCC activity of the molecule of interest can be assessed in vivo in animal models, as disclosed in Clynes et al., Proc. Nat'l Acad. Sci. USA 95:652-656 (1998). C1q binding assays can also be performed to confirm that multispecific polypeptide constructs or their cleavage fractions cannot bind C1q and therefore lack CDC activity. See, for example, C1q and C3c binding ELISAs in WO 2006 / 029879 and WO 2005 / 100402. To assess complement activation, CDC analysis can be performed (see, for example, Gazzano-Santoro et al., J. Immunol. Methods 202:163 (1996); Cragg, MS et al., Blood 101:1045-1052 (2003); and Cragg, MS and MJ Glennie, Blood 103:2738-2743 (2004)).FcRn binding and in vivo clearance / half-life can also be determined using methods known in this technique (see, for example, Petkova, SB et al., Int'l. Immunol. 18(12):1759-1769(2006)).

[0290] In some embodiments, the human IgG Fc region is modified to alter antibody-dependent cytotoxicity (ADCC) and / or complement-dependent cytotoxicity (CDC), for example, through amino acid modifications as described below: Natsume et al., 2008 Cancer Res, 68(10):3863-72; Idusogie et al., 2001 J Immunol, 166(4):2571-5; Moore et al., 2010 mAbs, 2(2):181-189; Lazar et al., 2006 PNAS, 103(11):4005-4010; Shields et al., 2001 JBC, 276(9):6591-6604; Stavenhagen et al., 2007 Cancer Res, 67(18):8882-8890; Stavenhagen et al., 2008 Advan.Enzyme The following are reviews in Regul., 48:152-164; Alegre et al., 1992 J Immunol, 148:3461-3468; Kaneko and Niwa, 2011 Biodrugs, 25(1):1-11.

[0291] Examples of mutations that enhance ADCC include modifications at Ser239 and Ile332, such as Ser239Asp and Ile332Glu (S239D, I332E). Examples of mutations that enhance CDC include modifications at Lys326 and Glu333. In some embodiments, the Fc region is modified at one or both of these locations, for example, using the Kabat numbering system as Lys326Ala and / or Glu333Ala (K326A and E333A).

[0292] In some embodiments, the Fc region of the fusion protein is altered at one or more of the following locations to reduce Fc receptor binding: Leu 234 (L234), Leu235 (L235), Asp265 (D265), Asp270 (D270), Ser298 (S298), Asn297 (N297), Asn325 (N325), Ala327 (A327), or Pro329 (P329). For example, Leu 234Ala (L234A), Leu235Ala (L235A), Leu235Glu (L235E), Asp265Asn (D265N), Asp265Ala (D265A), Asp270Asn (D270N), Ser298Asn (S298N), Asn297Ala (N297A), Pro329Ala (P329A) or Pro239Gly (P329G), Asn325Glu (N325E) or Ala327Ser (A327S). In a preferred embodiment, modifications within the Fc region reduce binding to the Fc receptor-γ receptor while minimizing the impact on binding to the neonatal Fc receptor (FcRn).

[0293] In some embodiments, the human IgG1 Fc region is modified at amino acid Asn297 (Kabat designation) to prevent glycosylation of the fusion protein, for example, Asn297Ala (N297A) or Asn297Asp (N297D). In some embodiments, the Fc region of the fusion protein is modified at amino acid Leu235 (Kabat designation) to alter Fc receptor interaction, for example, Leu235Glu (L235E) or Leu235Ala (L235A). In some embodiments, the Fc region of the fusion protein is modified at amino acid Leu234 (Kabat designation) to alter Fc receptor interaction, for example, Leu234Ala (L234A). In some embodiments, the Fc region of the fusion protein is modified at amino acid Leu234 (Kabat designation) to alter Fc receptor interaction, for example, Leu235Glu (L235E). In some embodiments, the Fc region of the fusion protein is altered at amino acids 234 and 235, for example, Leu234Ala and Leu235Ala (L234A / L235A) or Leu234Val and Leu235Ala (L234V / L235A). In some embodiments, the Fc region of the fusion protein is altered at amino acids 234, 235, and 297, for example, Leu234Ala, Leu235Ala, and Asn297Ala (L234A / L235A / N297A). In some embodiments, the Fc region of the fusion protein is altered at amino acids 234, 235, and 329, for example, Leu234Ala, Leu235Ala, and Pro239Ala (L234A / L235A / P329A). In some embodiments, the Fc region of the fusion protein is modified at amino acid Asp265 (Kabat number) to alter Fc receptor interaction, for example, Asp265Ala (D265A). In some embodiments, the Fc region of the fusion protein is modified at amino acid Pro329 (Kabat number) to alter Fc receptor interaction, for example, Pro329Ala (P329A) or Pro329Gly (P329G). In some embodiments, the Fc region of the fusion protein is altered at amino acids 265 and 329, for example, Asp265Ala and Pro329Ala (D265A / P329A) or Asp265Ala and Pro329Gly (D265A / P329G). In some embodiments, the Fc region of the fusion protein is altered at amino acids 234, 235, and 265, for example, Leu234Ala, Leu235Ala, Asp265Ala (L234A / L235A / D265A).In some embodiments, the Fc region of the fusion protein is altered at amino acids 234, 235, and 329, for example, Leu234Ala, Leu235Ala, Pro329Gly (L234A / L235A / P329G). In some embodiments, the Fc region of the fusion protein is altered at amino acids 234, 235, 265, and 329, for example, Leu234Ala, Leu235Ala, Asp265Ala, Pro329Gly (L234A / L235A / D265A / P329G). In some embodiments, the Fc region of the fusion protein is altered at Gly235 to reduce Fc receptor binding. For example, Gly235 is absent from the fusion protein. In some embodiments, the human IgG1 Fc region is modified at amino acid Gly236 to enhance interaction with CD32A, for example, Gly236Ala (G236A). In some embodiments, the human IgG1 Fc region lacks Lys447 (Kabat et al., 1991, Sequences of Proteins of Immunological Interest, EU Index).

[0294] In some embodiments, the Fc region of the fusion protein lacks one or more of the following amino acids to reduce Fc receptor binding: Glu233 (E233), Leu234 (L234), or Leu235 (L235). In other embodiments, the Fc region of the fusion protein lacks one or more of the following amino acids: Glu233 (E233), Leu234 (L234), or Leu235 (L235), and is modified at one or more of Asp265 (D265), Asn297 (N297), or Pro329 (P329) to reduce Fc receptor binding. For example, the Fc region of a DLL3-binding peptide is derived from the human Fc domain and contains the deletion of three amino acids E233, L234, and L235 in the lower hinge corresponding to IgG1. In some forms, such Fc peptides do not engorge FcγRs and are therefore referred to as "effect-silencing" or "effect-free." For example, the deletion of these three amino acids in the Fc region reduces the binding of complement protein C1q. In some embodiments, peptides with the Fc region lacking these three amino acids retain binding to FcRn, thus exhibiting an extended half-life and transcellular transport, which is associated with FcRn-mediated recycling. Such modified Fc regions are referred to as "Fc xELL" or "Fc deletion" and have the following amino acid sequence: (SEQ ID NO:9)

[0295] In some embodiments, the immunoglobulin Fc region or its immunoactive fragment comprises a human IgG1 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO:9.

[0296] In some embodiments, the human IgG Fc region is modified to enhance FcRn binding. Examples of Fc mutations that enhance binding to FcRn are Met252Tyr, Ser254Thr, Thr256Glu (M252Y, S254T, T256E, respectively) (Kabat No., Dall' Acqua et al. 2006, J. Biol Chem Vol. 281(33) 23514-23524), Met428Leu and Asn434Ser (M428L, N434S) (Zalevsky et al. 2010 Nature Biotech, Vol. 28(2) 157-159) or Met252Ile, Thr256Asp, Met428Leu (M252I, T256D, M428L, respectively) (Kabat et al. 1991 Sequences of Proteins of Immunological Interest EU Index).

[0297] In some embodiments, the Fc domain of the DLL3-binding peptide is derived from the human Fc domain and includes the mutants M252Y and M428V, referred to herein as "Fc-YV". In some embodiments, the mutated or modified Fc peptide includes the following mutations: M252Y and M428L, using the Kabat numbering system. In some embodiments, these mutations enhance binding to FcRn at an acidic pH (close to 6.5) in the endosome and lose detectable binding at a neutral pH (approximately 7.2), thereby enhancing FcRn-mediated recycling and prolonging the half-life.

[0298] In specific embodiments of the multispecific peptide constructs provided herein, the Fc domain of the DLL3-binding peptide is derived from the human Fc domain and contains mutations that induce heterodimerization. In some embodiments, such mutations include mutations referred to as "grooves" and "mortars". For example, an amino acid modification at Thr366 within the CH3 domain preferentially pairs with the second CH3 domain of the amino acids modified to smaller amino acids, such as Ser, Ala, and Val (T366S / L368A / Y407V), at positions Thr366, Leu368, and Tyr407, respectively, upon substitution with a larger amino acid, such as Try (T366W). In some embodiments, the "grooves" Fc domain contains the mutated T366W. In some embodiments, the "mortars" Fc domain contains the mutated T366S, L368A, and Y407V. The CH3-modified heterodimer can be further stabilized by introducing disulfide bonds, for example by changing Ser354 to Cys (S354C) and Y349 to Cys (Y349C) in the relative CH3 domain (reviewed in Carter, 2001 Journal of Immunological Methods, 248:7-15). In some embodiments, the Fc domain used for heterodimerization includes other mutations, such as the mutation S354C on the first member of the heterodimer Fc pair, which forms an asymmetric disulfide bond with the corresponding mutation Y349C on the second member of the heterodimer Fc pair. In some embodiments, one member of the heterodimer Fc pair includes a modification of H435R or H435K to prevent protein A binding while maintaining FcRn binding. In some embodiments, one member of the heterodimer Fc pair includes a modification of H435R or H435K, while the second member of the heterodimer Fc pair is not modified at H435. In various embodiments, the mortar Fc domain contains a modification of H435R or H435K (in some cases referred to as "mortar-R" when modified with H435R), while the mortar Fc domain does not contain such a modification. In some cases, the mortar-R mutation modifies the purification of the heterodimer relative to the possible presence of a homodimeric mortar Fc domain.

[0299] In some embodiments, the Fc region of human IgG is modified to prevent dimerization. In these embodiments, the fusion protein of the present invention is a monomer. For example, the residue Thr366 is modified with charged residues, such as Thr366Lys, Thr366Arg, Thr366Asp, or Thr366Glu (T366K, T366R, T366D, or T366E, respectively), to prevent CH3-CH3 dimerization.

[0300] In some embodiments, the immunoglobulin Fc region or immunoactive fragment of the fusion protein is a human IgG2 isotype, which has the following amino acid sequence: (SEQ ID NO:10)

[0301] In some embodiments, the fusion or its immunoactive fragment comprises a human IgG2 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO: 10.

[0302] In some embodiments, the human IgG2 Fc region is modified at amino acid Asn297, for example to prevent antibody glycosylation, such as Asn297Ala (N297A) or Asn297Asp (N297D). In some embodiments, the human IgG2 Fc region lacks Lys447 (Kabat et al., 1991, Sequences of Proteins of Immunological Interest, EU index).

[0303] In some embodiments, the immunoglobulin Fc region or immunoactive fragment of the fusion protein is a human IgG3 isotype, which has the following amino acid sequence: (SEQ ID NO:11)

[0304] In some embodiments, the antibody or its immunologically active fragment comprises a human IgG3 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO: 11.

[0305] In some embodiments, the human IgG3 Fc region is modified at amino acid Asn297 (Kabat number) to prevent antibody glycosylation, for example, Asn297Ala (N297A) or Asn297Asp (N297D). In some embodiments, the human IgG3 Fc region is modified at amino acid 435 to prolong the half-life, for example, Arg435His (R435H). In some embodiments, the human IgG3 Fc region lacks Lys447 (Kabat et al., 1991, Sequences of Proteins of Immunological Interest, EU index).

[0306] In some embodiments, the immunoglobulin Fc region or immunoactive fragment of the fusion protein is a human IgG4 isotype, which has the following amino acid sequence: (SEQ ID NO:12)

[0307] In some embodiments, the antibody or its immunologically active fragment comprises a human IgG4 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO: 12.

[0308] In some embodiments, the immunoglobulin Fc region or immunoactive fragment of the fusion protein is a human IgG4 isotype, which has the following amino acid sequence: (SEQ ID NO:13)

[0309] In some embodiments, the antibody or its immunologically active fragment comprises a human IgG4 polypeptide sequence that is at least 50%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence of SEQ ID NO: 13.

[0310] In some embodiments, the human IgG4 Fc region is modified at amino acid 235 to alter Fc receptor interaction, for example, Leu235Glu (L235E). In some embodiments, the human IgG4 Fc region is modified at amino acid Asn297 (Kabat number) to prevent antibody glycosylation, for example, Asn297Ala (N297A) or Asn297Asp (N297D). In some embodiments, the human IgG4 Fc region lacks Lys447 (Kabat et al., 1991, Sequences of Proteins of Immunological Interest, EU index).

[0311] In some embodiments, the fusion protein contains a polypeptide derived from the hinge region of an immunoglobulin. The hinge region may be selected from any human IgG subclass. For example, the fusion protein may contain a modified IgG1 hinge having the sequence EPKSSDKTHTCPPC (SEQ ID NO: 14), wherein the Cys220 region, which forms a disulfide bond with the C-terminal cysteine ​​of the light chain, is mutated to a serine, such as Cys220Ser (C220S). In other embodiments, the fusion protein contains a truncated hinge having the sequence DKTHTCPPC (SEQ ID NO: 15).

[0312] In some embodiments, the fusion protein has a modified hinge derived from IgG4, which is modified to prevent or reduce chain exchange, such as Ser228Pro (S228P), having the sequence ESKYGPPCPPC (SEQ ID NO: 16). In some embodiments, the fusion protein contains a linker polypeptide. In other embodiments, the fusion protein contains both a linker and a hinge polypeptide.

[0313] In some embodiments, the Fc region lacks or has reduced fucose attached to the N-linked glycan chain at N297. Numerous methods exist to prevent fucoidylation, including (but not limited to) production in FUT8-deficient cell lines; the addition of inhibitors to mammalian cell culture media, such as chestnut succinate; and metabolic engineering of the cell lines used for production.

[0314] In some embodiments, the Fc region is engineered to eliminate recognition by pre-existing antibodies found in humans. In some embodiments, the VHH-containing peptide of the present invention is modified by a mutation at position Leu11, such as Leu11Glu (L11E) or Leu11Lys (L11K). In other embodiments, the single-domain antibody of the present invention is modified by alterations in the carboxyl-terminal region, for example, by having the terminal sequence GQGTLVTVKPGG (SEQ ID NO: 17) or GQGTLVTVEPGG (SEQ ID NO: 18) or modifications thereof. In some embodiments, the VHH-containing peptide of the present invention is modified by a mutation at position 11 and by alterations in the carboxyl-terminal region.

[0315] In some embodiments, one or more polypeptides of the fusion protein of the present invention are operatively linked via amino acid linkers. In some embodiments, such linkers are primarily composed of the amino acids glycine and serine, referred to herein as GS linkers. The GS linkers of the fusion protein of the present invention may have various lengths, for example, lengths of 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids.

[0316] In some embodiments, the GS-linker comprises an amino acid sequence selected from the group consisting of: GGSGGS, i.e., (GGS)2 (SEQ ID NO:1); GGSGGSGGS, i.e., (GGS)3 (SEQ ID NO:2); GGSGGSGGSGGS, i.e., (GGS)4 (SEQ ID NO:3); and GGSGGSGGSGGSGGS, i.e., (GGS)5 (SEQ ID NO:4). In some embodiments, the linker is a flexible linker, which is a glycine linker containing glycine residues, such as (by means of non-limiting examples) GG, GGG, GGGG (SEQ ID NO:5), GGGGG (SEQ ID NO:6) and GGGGGG (SEQ ID NO:7). In some embodiments, the linker is (GGGGS)n, where n is 1 to 5 (SEQ ID NO: 123); (GGGGGS)n, where n is 1 to 4 (SEQ ID NO: 124); GGGGS (SEQ ID NO: 125); GGGGGS (SEQ ID NO: 126); GGGGGSGGGGGSGGGGS (SEQ ID NO: 127); GGGGSGGGGSGGGGS (SEQ ID NO: 128); GGSGGGGSGGGGSGGGGS (SEQ ID NO: 129); or PGGGG (SEQ ID NO: 450). In some embodiments, the fusion protein may include a combination of a GS linker and a glycine linker.

[0317] B. [Conjugate]

[0318] This document provides conjugates containing at least one VHH domain of DLL3 specifically bound as described herein, and one or more other portions. The other portions may be therapeutic agents, such as cytotoxic agents, or may be detection agents. In some embodiments, the portions may be targeting moieties, small molecule drugs (non-peptide drugs less than 500 Daltons), toxins, cell growth inhibitors, cytotoxic agents, immunosuppressants, radioactive reagents suitable for diagnostic purposes, radioactive metal ions for therapeutic purposes, prodrug-activating enzymes, reagents that increase biological half-life, or diagnostic or detection reagents.

[0319] In some embodiments, the conjugate is an antibody-drug conjugate (ADC, also known as an immune conjugate) containing one or more DLL3 VHH domains provided herein that bind to a therapeutic agent, exhibiting cytotoxicity, inhibiting cell growth, or providing some therapeutic benefit. In some embodiments, the cytotoxic agent is a chemotherapeutic agent, a drug, a growth inhibitor, a toxin (e.g., an enzyme-active toxin of bacterial, fungal, plant, or animal origin or fragments thereof), or a radioactive isotope (i.e., a radioactive conjugate). In some embodiments, the antibody-drug conjugates provided by the present invention allow for partial targeted delivery of a drug to a tumor. In some cases, this can induce targeted killing of tumor cells.

[0320] In some embodiments, a DLL3-binding conjugate is provided, comprising at least one DLL3 VHH domain provided herein bound to a therapeutic agent. In some embodiments, the therapeutic agent includes, for example, daunomycin, doxorubicin, methotrexate, and vindesine (Rowland et al., Cancer Immunol. Immunother. 21:183-187, 1986). In some embodiments, the therapeutic agent has intracellular activity. In some embodiments, the DLL3-binding conjugate is internalized, and the therapeutic agent is a cytotoxic agent that blocks cellular protein synthesis, causing cell death. In some embodiments, the therapeutic agent is a cytotoxic agent comprising a polypeptide having ribosomal inactivation activity, including, for example, leucopicrin, bouganin, saporin, ricin, ricin A chain, bryodin, diphtheria toxin, restrictocin, Pseudomonas aeruginosa exotoxin A, and variants thereof. In some embodiments, when the therapeutic agent is a cytotoxic agent comprising a polypeptide having ribosomal inactivation activity, the DLL3-binding conjugate must be internalized after binding to the target cell to make the protein cytotoxic to the cell.

[0321] In some embodiments, a combination of DLL3 is provided, comprising at least one DLL3 VHH domain provided herein combined with a toxin. In some embodiments, toxins include, for example, bacterial toxins (such as diphtheria toxin), plant toxins (such as ricin), small molecule toxins (such as geldanamycin) (Mandler et al., J. Nat. Cancer Inst. 92(19):1573-1581(2000); Mandler et al., Bioorganic & Med. Chem. Letters 10:1025-1028(2000); Mandler et al., Bioconjugate Chem. 13:786-791(2002)), maytansinoids (EP 1391213; Liu et al., Proc. Natl. Acad. Sci. USA 93:8618-8623(1996)), and calicheamicin (Lode et al., Cancer Res. 58:2928(1998); Hinman et al., Cancer Res. 53:3336-3342 (1993)). Toxins can exert their cytotoxic and cell growth-inhibiting effects through mechanisms including tubulin binding, DNA binding, or topoisomerase inhibition.

[0322] In some embodiments, a DLL3-binding conjugate is provided, comprising at least one DLL3 VHH domain provided herein bound to a marker that can indirectly or directly generate a detectable signal. Such IgSF conjugates can be used for research or diagnostic applications, such as in vivo cancer detection. The marker preferably generates a detectable signal directly or indirectly. For example, the marker can be a radiation-impermeable or radioactive isotope, such as 3H, 14C, 32P, 35S, 123I, 125I, 131I; a fluorescent (luciferase) or chemiluminescent (chromophore) compound, such as fluorescein, rhodamine, or luciferin; an enzyme, such as alkaline phosphatase, β-galactose, or horseradish peroxidase; a contrast agent; or a metal ion. In some embodiments, the label is a radioactive atom used for scintillation studies, such as 99Tc or 123I, or a spin label used for nuclear magnetic resonance (NMR) imaging (also known as magnetic resonance imaging, MRI), such as zirconium-89, iodine-123, iodine-131, indium-111, fluorine-19, carbon-13, nitrogen-15, oxygen-17, thiocyanate, manganese, or iron. Zirconium-89 can chelate with a variety of metal chelators and bind to antibodies, for example, for PET imaging (WO 2011 / 056983).

[0323] Combinations with DLL3 can be prepared using any method known in the art. See, for example, WO 2009 / 067800, WO 2011 / 133886 and U.S. Patent Application Publication No. 2014322129, which are incorporated herein by reference in their entirety.

[0324] In some embodiments, attachment may be covalent or non-covalent, for example via a biotin-streptolysin non-covalent interaction. In some embodiments, one, two, three, four, five, or more portions, which may be identical or different, are bound, linked, or fused to the DLL3 VHH domain to form a DLL3-binding conjugate. In some embodiments, such portions are attached to the VHH domain using a variety of molecular biological or chemical binding and linking methods known in the art and described below. In some embodiments, linkers, such as peptide linkers, cleavable linkers, non-cleavable linkers, or linkers that facilitate binding reactions, may be used to link or bind effector moieties to variant peptides or immunomodulatory proteins.

[0325] In some embodiments, the DLL3 VHH domain is bound to one or more portions via a linker (L), for example, about 1 to about 20 drug portions per VHH. In some embodiments, the DLL3-binding conjugate comprises the following components: (VHH domain), (L)q, and (portion)m, wherein the VHH domain is any of the VHH domains as described that are capable of specifically binding DLL3; L is a linker for linking a protein or polypeptide to a portion; m is at least 1; q is 0 or more; and the resulting DLL3-binding conjugate is bound to DLL3. In a particular embodiment, m is 1 to 4 and q is 0 to 8.

[0326] Linkers can consist of one or more linker components. For covalent attachment of antibody and drug moieties, linkers typically have two reactive functional groups, i.e., divalent in a reactive sense. Divalent linker reagents are known for attaching two or more functional or biologically active moieties, such as peptides, nucleic acids, drugs, toxins, antibodies, haptens, and reporter groups, and methods for obtaining the conjugates have been described (Hermanson, GT (1996) Bioconjugate Techniques; Academic Press: New York, pp. 234-242).

[0327] Exemplary linker components include 6-cis-butenylidene-iminohexyl (“MC”), cis-butenylidene-iminopropyl (“MP”), valine-citrulline (“val-cit”), alanine-phenylalanine (“ala-phe”), p-aminobenzoxycarbonyl (“PAB”), N-succinylimino-4-(2-pyridylthio)valerate (“SPP”), N-succinylimino-4-(N-cis-butenylidene-iminomethyl)cyclohexane-1-carboxylate (“SMCC”), and N-succinylimino-(4-iodoacetyl)aminobenzoate (“SIAB”).

[0328] In some embodiments, the linker may comprise amino acid residues. Exemplary amino acid linker components include dipeptides, tripeptides, tetrapeptides, or pentapeptides. Exemplary dipeptides include valine-citrulline (VC or val-cit) and alanine-phenylalanine (AF or ala-phe). Exemplary tripeptides include glycine-valine-citrulline (Gly-val-cit) and glycine-glycine-gly (Gly-gly-gly). The amino acid residues comprising the amino acid linker component include naturally occurring residues as well as minor amino acids and non-naturally occurring amino acid analogs, such as citrulline. The amino acid linker component may be designed and optimized in terms of its selectivity for enzymatic cleavage by specific enzymes, such as tumor-associated proteases, cathepsins B, C, and D, and cytoplasmic proteases.

[0329] VHH domain conjugates to cytotoxic agents can be prepared using a variety of bifunctional protein coupling agents, such as N-succinimino-3-(2-pyridyldithiol)propionate (SPDP), iminothiacyclopentane (IT), bifunctional derivatives of acetilimates (such as dimethyl hexadiimide HCl), active esters (such as disuccinimino adduct), aldehydes (such as glutaraldehyde), diazid compounds (such as bis(p-azidobenzoyl)hexamethylenediamine), diazido derivatives (such as bis-(p-diazobenzoyl)-ethylenediamine), diisocyanates (such as toluene 2,6-diisocyanate), and bifunctional fluorinated compounds (such as 1,5-difluoro-2,4-dinitrobenzene).

[0330] Antibody-drug conjugates can be prepared by a variety of methods known to those skilled in the art, such as organic chemical reactions, conditions, and reagents. In one embodiment, the method includes: (1) reacting a nucleophilic group of the VHH domain with a divalent linker reagent to form VHH-L via a covalent bond, followed by a reaction with the drug moiety D; and (2) reacting a nucleophilic group of the drug moiety with a divalent linker reagent to form DL via a covalent bond, followed by a reaction with a nucleophilic group of the VHH domain.

[0331] Nucleophilic groups on antibodies including the VHH domain include (but are not limited to): (i) N-terminal amino groups; (ii) side-chain amino groups, such as lysine; (iii) side-chain thiols, such as cysteine; and (iv) glycosylated hydroxyl or amino groups of the antibody. The amines, thiols, and hydroxyl groups are nucleophilic and capable of reacting with electrophilic groups on the linker moiety and linker reagent to form covalent bonds. These electrophilic groups include: (i) active esters, such as NHS esters, HOBt esters, haloformates, and acid halides; (ii) alkyl and benzyl halides, such as haloacetylamines; and (iii) aldehydes, ketones, carboxyl groups, and maleic anhydride groups. Other nucleophilic groups can be introduced into the antibody via the reaction of lysine with 2-iminothiocyclopentane (Traut's reagent), causing the amine to convert to a thiol. Reactive thiol groups can be introduced into antibodies (or fragments thereof) by introducing one, two, three, four or more cysteine ​​residues (e.g., to prepare mutant antibodies containing one or more non-natural cysteine ​​amino acid residues).

[0332] Conjugates such as antibody-drug conjugates can also be generated by modifying antibodies, such as VHH domains, to introduce electrophilic moieties that can react with nucleophilic substituents on linker reagents or drugs. The sugar in a glycosylated antibody can be oxidized, for example, with a periodate oxidizing agent to form an aldehyde or ketone group, which can react with the amino group of the linker reagent or drug moiety. The resulting imine Schiff base group can form a stable bond, or can be reduced, for example, with a borohydride reagent to form a stable amine bond. In one embodiment, the carbohydrate moiety of a glycosylated antibody reacts with galactose oxidase or sodium periodate to produce a carbonyl group (aldehyde and ketone) in the protein, which can react with a suitable group on the drug (Hermanson, Bioconjugate Techniques). In another embodiment, a protein containing an N-terminal serine or threonine residue can react with sodium periodate to produce an aldehyde replacing the first amino acid. Such aldehydes can react with the drug moiety or a linker nucleophilic reagent.

[0333] Similarly, nucleophilic groups on the pharmaceutical moiety include (but are not limited to): amines, thiols, hydroxyl groups, acehydrazides, oximes, hydrazines, thiohexacarbazides, hydrazine carboxylates, and arylacehydrazides that can react with electrophilic groups on the linker moiety and the linker reagent to form covalent bonds. These electrophilic groups include: (i) active esters, such as NHS esters, HOBt esters, haloformates, and acid halides; (ii) alkyl and benzyl halides, such as haloacetamides; and (iii) aldehydes, ketones, carboxyl groups, and maleic diacetylimine groups.

[0334] Alternatively, fusion proteins containing the VHH domain and cytotoxic agents can be prepared, for example, through recombinant technology or peptide synthesis. The length of the DNA may include separate regions encoding the two parts of the conjugate, which are adjacent to each other or separated by regions encoding linking peptides that do not disrupt the desired properties of the conjugate.

[0335] [C. Multispecific form]

[0336] This article provides a multispecific DLL3-binding polypeptide containing at least one VHH domain for binding DLL3 and one or more other binding domains. Typically, one or more of these other domains bind to a second antigen or protein other than DLL3. In some embodiments, the one or more other domains are antibody or antigen-binding fragments specific to the second antigen or protein. In some embodiments, the other domains are VHH domains.

[0337] In some embodiments, the multispecific DLL3-binding polypeptide comprises at least one VHH domain that binds DLL3 and at least one other binding domain that binds a second antigen or protein. In some embodiments, the second antigen is a tumor-associated antigen (TAA) or a tumor microenvironment-associated antigen (TMEAA). In some embodiments, the second antigen is an immunomodulatory antigen, wherein the antigen is associated with enhancing or inhibiting signal transduction pathways in immune cells.

[0338] In some cases, multispecific DLL3-binding peptides may further contain an Fc domain, such as any of those described above. In some embodiments, the multispecific DLL3-binding peptides provided herein comprise at least one VHH domain for binding DLL3, at least one other binding domain for binding a second antigen or protein, and an Fc domain. In some embodiments, the Fc domain mediates dimerization of the multispecific DLL3-binding peptide under physiological conditions to form a dimer that doubles the number of binding sites for DLL3 and other antigens or proteins.

[0339] The following describes a non-restrictive, illustrative, multispecific DLL3-binding peptide.

[0340] [ 1. Bispecific T cell conjugation molecules ]

[0341] In some embodiments, the DLL3-binding polypeptide is a bispecific construct that is or includes at least one DLL3 VHH domain as described herein and at least one other binding molecule capable of binding to a surface molecule expressed on T cells. In some embodiments, the surface molecule is a T cell activation component, such as a component of the T cell receptor complex. In certain states, the surface molecule is an activated T cell antigen expressed on T cells and capable of inducing T cell activation upon interaction with the antigen-binding molecule. For example, in some states, the interaction between the antigen-binding molecule and the activated T cell antigen can induce T cell activation by triggering a signaling cascade of the T cell receptor complex. Assays suitable for measuring T cell activation are known and include any assay that measures or assesses proliferation, differentiation, intercytokine secretion, cytotoxic activity, and / or the expression of one or more activation markers. In some embodiments, the simultaneous or near-simultaneous binding of such a DLL3-binding polypeptide to its two targets, DLL3 expressed on target cells and T cell molecules expressed on T cells, such as activated T cell antigens, can cause a transient interaction between the target cells and T cells, thereby inducing T cell activation, such as cytotoxic activity, and subsequently causing lysis of the target cells.

[0342] In some embodiments, the T-cell surface molecule, such as the activated T-cell antigen, is CD3 or CD2. Specifically, the provided bispecific DLL3-binding polypeptide is capable of specifically binding to activated T-cell antigens expressed on human T cells, such as human CD3 or human CD2. In certain embodiments, other domains specifically targeting activated T-cell antigens (e.g., CD3 or CD2) are antibody or antigen-binding fragments. In some embodiments, the DLL3-binding polypeptide may be a bispecific antibody-T-cell conjugating molecule containing at least one DLL3 VHH domain specifically binding to DLL3 and other binding molecules, which are antibody or antigen-binding fragments specific to activated T-cell components (e.g., T-cell surface molecules, such as CD3 or CD2).

[0343] Bispecific antibody T-cell conjugating molecules, particularly bispecific T-cell conjugating molecules (BiTEs), contain tandem scFv molecules fused by flexible linkers (see, for example, Nagorsen and Bauerle, Exp Cell Res 317, 1255-1260 (2011), tandem scFv molecules fused together by, for example, flexible linkers, and further contain an Fc domain composed of first and second units capable of stable association (WO2013026837); bifunctional antibodies and their derivatives, including tandem bifunctional antibodies (Holliger et al., Prot Eng 9, 299-305 (1996); Kipriyanov et al., J Mol Biol 293, 41-66 (1999)); dual affinity retargeting (DART) molecules, which may include bifunctional antibody forms with C-terminal disulfide bridges; or triomab, which includes fully hybridized mouse / rat IgG molecules (Seimetz et al., Cancer Treat Rev 36, 458-). 467 (2010)). Any of the DLL3 VHH domains provided herein can be used to generate similar forms of more than one molecule.

[0344] In some embodiments, other binding domains specifically targeting activated T-cell antigens are antigen-binding fragments selected from Fab fragments, F(ab')2 fragments, Fv fragments, scFv, disulfide-stabilized Fv fragments (dsFv), scAb, dAb, single-domain heavy chain antibodies (VHH), or single-domain light chain antibodies. In some embodiments, the other binding domains are monovalent in relation to binding activated T-cell antigens (such as CD2 or CD3).

[0345] In some embodiments, other binding domains are capable of binding to CD3 or the CD3 complex. The CD3 complex is a complex of at least five membrane-bound polypeptides in a mature T lymphocyte that are non-covalently associated with each other and with the T cell receptor. The CD3 complex includes γ, δ, ε, ζ, and η chains (also referred to as subunits). In some embodiments, other binding molecules are antibody or antigen-binding fragments capable of specifically binding to CD3 or the CD3 complex, also referred to as CD3-binding domains. In some embodiments, CD3-binding domains capable of binding to CD3 or the CD3 complex include one or more replicas of anti-CD3 Fab fragments, anti-CD3 F(ab')2 fragments, anti-CD3 Fv fragments, anti-CD3 scFv, anti-CD3 dsFv, anti-CD3 scAb, anti-CD3 dAb, anti-CD3 single-domain heavy chain antibody (VHH), and anti-CD3 single-domain light chain antibody. In some embodiments, the anti-CD3 binding domain is monovalent with respect to binding CD3.

[0346] In some cases, the CD3-binding domain recognizes the CD3ε chain. In some embodiments, the anti-CD3ε binding domain includes one or more replicas of an anti-CD3ε Fab fragment, an anti-CD3ε F(ab')2 fragment, an anti-CD3ε Fv fragment, an anti-CD3ε scFv, an anti-CD3ε dsFv, an anti-CD3ε scAb, an anti-CD3ε dAb, an anti-CD3ε single-domain heavy chain antibody (VHH), and an anti-CD3ε single-domain light chain antibody. In some embodiments, the anti-CD3ε binding domain is monovalent with respect to binding CD3ε.

[0347] Exemplary monoclonal antibodies against CD3 or CD3 complexes include (but are not limited to) OKT3, SP34, UCHT1, or 64.1 or their antigen-binding fragments (see, for example, June et al., J. Immunol. 136:3945-3952 (1986); Yang et al., J. Immunol. 137:1097-1100 (1986); and Hayward et al., Immunol. 64:87-92 (1988)). In some morphologies, CD3 clustering on T cells, for example by immobilizing or localizing anti-CD3 antibodies to cells or tethers, induces T cell activation, similar to T cell receptor binding, but without the typical specificity of its pure lineage. In one embodiment, the CD3-binding domain monovalently and specifically binds to the CD3 antigen and is derived from OKT3 (ORTHOCLONE-OKT3™ (muromonab-CD3)); humanized OKT3 (US Patent No. 7,635,475 and International Application No. WO2005040220); SP34 (Pessano et al. The EMBO Journal. 4:337-344, 1985); humanized variants of SP34 (WO2015001085); Teplizumab™ (MGA031, Eli Lilly); the anti-CD3 binding molecule described in US2011 / 0275787; UCHT1 (Pollard et al. 1987 J Histochem). Cytochem. 35(11):1329-38;WO2000041474);NI0401(WO2007 / 033230); visilizumab (US Patent No. 5,834,597); BC-3 (Anasetti et al., Transplantation 54:844 (1992); H2C (described in PCT Publication No. WO2008 / 119567); V9 (described in Rodrigues et al., Int J Cancer Supplement 7, 45-50 (1992) and U.S. Patent No. 6,054,297). Other anti-CD3 antibodies may also be used in the constructs provided herein, including any of the following: International Publication PCT Applications Nos. WO199404679, WO2008119567, WO2015095392, WO2016204966, and WO2019133761; Publication Patent Applications Nos. US20170369563, US20180194842, and US20180355038; U.S. Patents Nos. 7,728,114, 7,381,803, and 7,994,289.

[0348] In some embodiments, the CD3 binding domain contains the variable heavy chain (VH) shown in SEQ ID NO: 19 and / or the variable light chain shown in SEQ ID NO: 20, or VH and / or VL sequences having at least 60%, 70%, 80%, 90%, 95%, 97%, 98%, or 99% identity with these sequences, and specifically binds to CD3. In some embodiments, the CD3 binding domain contains CDRH1, CDRH2, and CDRH3 of the variable heavy chain (VH) shown in SEQ ID NO: 19 and CDRL1, CDRL2, and CDRL3 of the variable light chain shown in SEQ ID NO: 20. In some cases, the CD3 binding region contains a humanized form of the VH sequence shown in SEQ ID NO: 19 and a humanized form of the VL sequence shown in SEQ ID NO: 20. In some embodiments, the CD3 binding region may contain a humanized OKT3-derived VH domain sequence as shown in any of SEQ ID NOs 21, 22, and 23 and / or a VL domain sequence as shown in any of SEQ ID NOs 24, 25, and 26, or a VH and / or VL sequence having at least 60%, 70%, 80%, 90%, 95%, 97%, 98%, or 99% identity with such sequences, and specifically binds to CD3. In some embodiments, the CD3 binding domain is Fab, scFv, Fv, or dsFv, containing any combination of the above VH and VL sequences, particularly any combination of the VH sequence as shown in any of SEQ ID NOs 21, 22, and 23 and the VL sequence as shown in any of SEQ ID NOs 24, 25, and 26.

[0349] In some embodiments, the anti-CD3ε binding domain includes a VH CDR1 sequence comprising at least the amino acid sequence TYAMN (SEQ ID NO: 29); a VH CDR2 sequence comprising at least the amino acid sequence RIRSKYNNYATYYADSVKD (SEQ ID NO: 30); a VH CDR3 sequence comprising at least the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); a VL CDR1 sequence comprising at least the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO: 32); a VL CDR2 sequence comprising at least the amino acid sequence GTNKRAP (SEQ ID NO: 33); and a VL CDR3 sequence comprising at least the amino acid sequence ALWYSNLWV (SEQ ID NO: 34). In some embodiments, the CD3 binding domain is Fab, scFv, Fv, or dsFv, which contains at least a VH CDR1 sequence including the amino acid sequence TYAMN (SEQ ID NO: 29); at least a VH CDR2 sequence including the amino acid sequence RIRSKYNNYATYYADSVKD (SEQ ID NO: 30); at least a VH CDR3 sequence including the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); at least a VL CDR1 sequence including the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO: 32); at least a VL CDR2 sequence including the amino acid sequence GTNKRAP (SEQ ID NO: 33); and at least a VL CDR3 sequence including the amino acid sequence ALWYSNLWV (SEQ ID NO: 34).

[0350] In some embodiments, the CD3 binding domain contains the variable heavy chain (VH) shown in SEQ ID NO:27 and / or the variable light chain shown in SEQ ID NO:28, or a VH and / or VL sequence having at least 60%, 70%, 80%, 90%, 95%, 97%, 98%, or 99% identity with such sequences, and specifically binds to CD3. In some embodiments, the CD3 binding domain contains CDRH1, CDRH2, and CDRH3 of the variable heavy chain (VH) shown in SEQ ID NO:27 and CDRL1, CDRL2, and CDRL3 of the variable light chain shown in SEQ ID NO:28. In some embodiments, the CD3 binding domain contains CDRH1, CDRH2, and CDRH3 shown in SEQ ID NO:29, 30, and 31, respectively, and CDRL1, CDRL2, and CDRL3 of the variable light chain shown in SEQ ID NO:32, 33, and 34, respectively. In some cases, the CD3 binding region comprises a humanized form of the VH sequence shown in SEQ ID NO:27 and a humanized form of the VL sequence shown in SEQ ID NO:28. In some embodiments, the CD3 binding region may contain a humanized VH domain sequence shown in any of SEQ ID NO:35-65, 453, 454 or 460 and / or a VL domain sequence shown in any of SEQ ID NO:66-84, 368, 451 or 452, or a VH and / or VL sequence having at least 60%, 70%, 80%, 90%, 95%, 97%, 98%, or 99% similarity to such sequences, and specifically binds to CD3. In some embodiments, its anti-CD3ε binding domain comprises a variable heavy chain (VH) comprising an amino acid sequence of SEQ ID NO:47 and a variable light chain (VL) comprising an amino acid sequence of SEQ ID NO:75.

[0351] In some embodiments, the CD3-binding domain is Fab, scFv, Fv, or dsFv, containing any combination of the above VH and VL sequences, particularly any combination of the VH sequence shown in any of SEQ ID NO: 35-65, 453, 454, or 460 and the VL sequence shown in any of SEQ ID NO: 66-84, 368, 451, or 452. In some embodiments, the anti-CD3-binding domain is Fab, scFv, Fv, or dsFv, containing a variable heavy chain (VH) comprising the amino acid sequence of SEQ ID NO: 47 and a variable light chain (VL) comprising the amino acid sequence of SEQ ID NO: 75.

[0352] In some embodiments, the CD3 binding domain contains a variable heavy chain (VH) as shown in any of SEQ ID NO: 519, 520, 523 or 524. In some embodiments, the CD3 binding domain contains a variable light chain (VL) as shown in any of SEQ ID NO: 521, 522, 525 or 526.

[0353] The provided bispecific construct can be formatted into any of a number of forms containing at least one DLL3 VHH domain and at least one other domain (such as a CD3 binding domain) specifically targeting activated T cell antigens.

[0354] In one embodiment, the bispecific construct is a bispecific single-domain antibody-linking Fab (S-Fab) containing at least one of the DLL3 VHH domains directly or indirectly linked to a Fab antigen-binding fragment (such as anti-CD3 Fab) specifically targeting a T-cell activating antigen (e.g., CD3). The Fab targeting a T-cell activating antigen, such as anti-CD3 Fab, may contain either the VH or VL sequence. In some embodiments, the DLL3 VHH domain is linked to the C-terminus of the VH or VL chain of the anti-CD3 Fab. In some embodiments, the S-Fab may be further modified, such as by binding to polyethylene glycol (PEG), N-(2-hydroxypropyl)methacrylamide (HPMA) copolymer, proteins (such as albumin), polyglutamic acid, or by extending its plasma half-life (Pan et al., (2018) International Journal of Nanomedicine, 2018: 3189-3201).

[0355] In another embodiment, the bispecific construct is an scFv-single-domain antibody, wherein the construct contains at least one of the DLL3 VHHs directly or indirectly linked to the scFv, which contains VH and VLs specifically targeting an antigen-binding domain of a T-cell activating antigen (e.g., CD3). scFvs targeting T-cell activating antigens, such as anti-CD3 scFvs, may contain either of the VH and VL sequences. In some embodiments, the VHH domain and the scFv are linked by a linker, such as a peptide linker. In some embodiments, the peptide linker may be a peptide linker as described herein. In some embodiments, the VHH domain and the scFv are each linked to an Fc region, such as the N-terminus of an Fc region, via a hinge region or a linker (e.g., a peptide linker), as appropriate. The Fc region may be any of those described herein, such as the human Fc region or a variant thereof, such as the human IgG1 Fc region or a variant thereof. In certain embodiments, the Fc region is formed from a variant Fc domain (e.g., a variant human IgG1 domain) that is mutated or modified to promote heterodimerization, wherein different polypeptides can dimerize to produce heterodimers.

[0356] In another embodiment, the CD3 binding domain is a single-domain antibody, such as a VHH domain that specifically binds to CD3. Single-domain antibodies (including VHH domains that bind to CD3) are known, see, for example, published U.S. Patent Application No. US20160280795. In some embodiments, the CD3 binding domain is the anti-CD3 VHH shown in SEQ ID NO:85; or exhibits a sequence that is at least 60%, 70%, 80%, 90%, 95%, 97%, 98%, or 99% identical to SEQ ID NO:85 and specifically binds to CD3. In such embodiments, the bispecific constructs provided herein may include at least one DLL3 VHH domain and at least one CD3 VHH domain. To format the construct, in some cases, each VHH domain is linked to an Fc region, such as the N-terminus of an Fc region, via a hinge region or a linker (e.g., a peptide linker), as appropriate. The Fc region may be any of those described herein, such as the human Fc region or a variant thereof, such as the human IgG1 Fc region or a variant thereof. In certain embodiments, the Fc region is formed from a variant Fc domain (e.g., a variant human IgG1 domain) that is mutated or modified to promote heterodimerization, wherein different polypeptides can dimerize to produce heterodimers.

[0357] In the above embodiments, exemplary modifications to the Fc region to promote heterodimerization are known, including any of those described below, such as in Table 3. In some embodiments, one Fc polypeptide in the heterodimer Fc contains the amino acid sequence described in any of SEQ ID NO: 103, 107, 115, 440 or 446, and another Fc polypeptide in the heterodimer Fc contains the amino acid sequence described in any of SEQ ID NO: 104, 108, 111, 113, 119, 121, 441, 444, 448. In some embodiments, one Fc polypeptide of the heterodimer Fc contains the amino acid sequence described in any one of SEQ ID NO: 105, 109, 116, 118, 442 or 447, and the other Fc polypeptide of the heterodimer Fc contains the amino acid sequence described in any one of SEQ ID NO: 106, 110, 112, 114, 120, 122, 443, 445 or 449.

[0358] [ 2. Restricted CD3 multispecific constructs ]

[0359] In some embodiments, the DLL3-binding polypeptide is a multispecific polypeptide construct that is a restricted T-cell conjugation fusion protein. In certain embodiments, the restricted multispecific construct provided herein binds an activated T-cell antigen such as CD3 and DLL3. The restricted multispecific polypeptide construct provided herein includes at least: a first component comprising an immunoglobulin Fc region; a second component comprising at least one or more replicas of a CD3-binding domain (referred herein to be an anti-CD3-binding domain or a CD3-binding domain, which are terms used interchangeably herein); and a linker, such as a polypeptide linker, connecting the first and second components. In the provided multispecific polypeptide construct, one or both of the first and second components contain at least one DLL3 VHH domain that engages upon binding to an antigen, such that the restricted CD3-binding region is substantially capable of binding CD3. [Figures 3A to 3E] depict exemplary forms of restricted multispecific structures.

[0360] In some embodiments, the restricted multispecific polypeptide constructs provided herein exist in two states in terms of their ability to bind CD3 and subsequently activate T cells: (1) an “inactive” state when any or all antigen-binding domains do not bind to DLL3, thereby restricting CD3 binding and avoiding or reducing T cell interaction; and (2) an “active” state when any or all antigen-binding domains bind antigens, thereby enabling the CD3-binding region to bind CD3 and allowing T cell interaction.

[0361] In some embodiments, the Fc region is linked to the CD3-binding domain via a linker. In some embodiments, the Fc region is linked to the CD3-binding domain via a non-cleavable linker. In some embodiments, the Fc region is linked to the CD3-binding domain via a cleavable linker or other unstable linker. In some embodiments, the cleavable linker is a linker that can be specifically cleaved in the presence of a protease. In some states, CD3 binding is enhanced after cleavage of the cleavable linker. In some such states, the "active" state can be further amplified by several mechanisms, including cleavage of the linker that binds the CD3-binding domain to the Fc region. In some embodiments, the cleavable linker is a linker containing a protease receptor recognition site. In some embodiments, where the Fc region and the CD3-binding domain are linked by a cleavable linker, cleavage within the linker can enhance CD3 binding.

[0362] Furthermore, in a configuration where the Fc region and CD3 binding region are operatively linked by a cleavable linker, linker cleavage between the Fc region and CD3 binding region can separate the restricted multispecific polypeptide construct into a first component and a second component. Depending on the composition of the restricted multispecific polypeptide construct, the first and second components may have different functions. In some embodiments, the Fc region is a region exhibiting one or more effector functions, such as ADCC, CDC, or ADCP. In such examples, the restricted multispecific polypeptide construct of the present invention can be used to generate a self-amplifying system. For example, in some configurations, a protease-cleavable linker is incorporated between the Fc and CD3 binding domain components to amplify T cell activation capacity by fully exposing the CD3 binding domain. Depending on the specific linker included, the amplification step can be mediated by tumor-associated proteases or by granzymes released after antigen-dependent T cell activation. If a tumor protease-cleavable linker is included, the amplification is mediated by the tumor or the tumor microenvironment. If the granzyme B-mediated linker is included, amplification can be self-mediated by T cells following antigen-dependent activation. Furthermore, if the construct includes an effector Fc, amplification can be mediated by granzymes released from NK cells that emerge via the ADCC mechanism.

[0363] The provided restricted multispecific peptide construct includes a configuration in which a first component containing an Fc region is located at the N-terminus of a second component containing a CD3-binding region. In such embodiments, the first and second components are linked via a linker at the C-terminus of the Fc region. In some embodiments, at least one DLL3 VHH domain is located at the amino-terminal (N-terminal) region of the multispecific peptide construct. In some embodiments, at least one DLL3 VHH domain is located at the carboxyl-terminal (C-terminal) region of the multispecific peptide construct. In some embodiments, the restricted multispecific peptide construct contains at least two DLL3 VHH domains located at the N-terminal and C-terminal regions of the multispecific peptide construct.

[0364] In some embodiments, the restricted multispecific polypeptide construct is a dimer, wherein the dimer is formed by covalent or non-covalent interactions between two polypeptide chains. In some embodiments, the two polypeptide chains are covalently bonded to each other, for example, by interchain disulfide bonds. In some embodiments, the Fc region mediates dimerization via interchain disulfide bonds. In certain embodiments, the restricted multispecific polypeptide construct contains a heterodimeric Fc region, wherein in some cases, the polypeptide chains of the multispecific polypeptide construct are different (heterodimeric). In a specific example of a heterodimeric multispecific polypeptide construct, the CD3-binding region is a double-stranded polypeptide containing VH and VL chains, such as an Fv antibody fragment containing VH and VL. In some embodiments, the Fv antibody fragment comprises a disulfide-stabilized anti-CD3-binding Fv fragment (dsFv).

[0365] In some embodiments, the restricted multispecific peptide construct is formed from or includes two peptides, comprising a first peptide comprising a first Fc peptide of a heterodimeric Fc region, a linker (e.g., a cleavable or non-cleavable linker), and a VH domain of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv); and a second peptide comprising a second Fc peptide of a heterodimeric Fc region, a linker (e.g., a cleavable or non-cleavable linker), and a VL domain of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv). In some embodiments, the first peptide contains one or two VHH domains bound to DLL3. In some embodiments, the second peptide contains one or two VHH domains bound to DLL3. In some embodiments, the restricted multispecific peptide construct contains at least two DLL3 VHH domains. In some cases, at least one DLL3 VHH domain is located at the N-terminus of the Fc peptide, and at least one DLL3 VHH domain is located at the C-terminus of the chain of the CD3 binding region.

[0366] In some embodiments, the first polypeptide or the second polypeptide, or both the first polypeptide and the second polypeptide, further include a co-stimulatory receptor binding region (CRBR) that binds to a co-stimulatory receptor. In some embodiments, the CRBR of the first and / or second polypeptide may be located at the C-terminus of the N-terminus of the Fc polypeptide and / or the C-terminus of the CD3 binding region.

[0367] In some embodiments, the restricted multispecific polypeptide construct contains at least two VHH domains that bind DLL3 and at least one costimulatory receptor binding region (CRBR) that binds a costimulatory receptor. In some embodiments, the restricted multispecific polypeptide construct contains: (1) a first polypeptide comprising, in N-terminal to C-terminal order: a first DLL3 VHH domain, a first Fc polypeptide of a heterodimeric Fc region, a linker (e.g., a cleavable linker), a chain of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv or dsFv) (e.g., VH or VL), and a second DLL3 VHH domain; and (2) a second polypeptide comprising, in N-terminal to C-terminal order: a second Fc polypeptide of a heterodimeric Fc region, the same linker (e.g., the same cleavable linker), another chain of an anti-CD3 antibody or antigen-binding fragment (another chain of VH or VL), and a costimulatory receptor binding region (CRBR) that binds a costimulatory receptor.

[0368] In some embodiments, the first polypeptide or the second polypeptide, or both the first polypeptide and the second polypeptide, further include an inhibitory receptor binding region (IRBR) that binds to an inhibitory receptor. In some embodiments, the IRBR of the first polypeptide and / or the second polypeptide may be located at the C-terminus of the chain of the N-terminus and / or CD3 binding region of the Fc polypeptide.

[0369] In some embodiments, the restricted multispecific polypeptide construct contains at least two VHH domains that bind DLL3 and at least one inhibitory receptor binding region (IRBR) that binds an inhibitory receptor. In some embodiments, the restricted multispecific polypeptide construct contains: (1) a first polypeptide comprising, in N-terminal to C-terminal order: a first DLL3 VHH domain, a first Fc polypeptide of a heterodimeric Fc region, a linker (e.g., a cleavable or non-cleavable linker), a chain of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv or dsFv) (e.g., VH or VL), and a second DLL3 VHH domain; and (2) a second polypeptide comprising, in N-terminal to C-terminal order: a second Fc polypeptide of a heterodimeric Fc region, the same linker (e.g., the same cleavable linker), another chain of an anti-CD3 antibody or antigen-binding fragment (another chain of VH or VL), and an inhibitory receptor binding region (IRBR) that binds an inhibitory receptor.

[0370] In some embodiments, at least one of the first polypeptide or the second polypeptide further includes a co-stimulatory receptor binding region (CRBR) that binds to a co-stimulatory receptor, and at least one of the first polypeptide or the second polypeptide further includes an inhibitory receptor binding region (IRBR) that binds to an inhibitory receptor. In some embodiments, the CRBR of the first and / or second polypeptide may be located at the C-terminus of the chain of the N-terminus and / or the CD3 binding region of the Fc polypeptide. In some embodiments, the IRBR of the first polypeptide and / or the second polypeptide may be located at the C-terminus of the chain of the N-terminus and / or the CD3 binding region of the Fc polypeptide.

[0371] In some embodiments, the restricted multispecific polypeptide construct comprises at least two VHH domains that bind DLL3, a costimulatory receptor binding region (CRBR) that binds a costimulatory receptor, and an inhibitory receptor binding region (IRBR) that binds an inhibitory receptor. In some embodiments, the restricted multispecific polypeptide construct comprises: (1) a first polypeptide comprising, in N-terminal to C-terminal order: a first DLL3 VHH domain, a first Fc polypeptide of a heterodimeric Fc region, a linker (e.g., a cleavable or non-cleavable linker), a chain of an anti-CD3 antibody or antigen-binding fragment (e.g., Fv or dsFv) (e.g., VH or VL), and a second DLL3 VHH domain; and (2) a second polypeptide comprising, in N-terminal to C-terminal order: one of an IRBR or a CRBR, a second Fc polypeptide of a heterodimeric Fc region, the same linker (e.g., the same cleavable or non-cleavable linker), another chain of an anti-CD3 antibody or antigen-binding fragment (another chain of VH or VL), and the other of an IRBR or a CRBR.

[0372] The components of the multispecific polypeptide construct of the present invention are described in more detail below.

[0373] [a.DLL3 VHH antigen-binding domain]

[0374] The restricted multispecific polypeptide constructs of the present invention include at least one DLL3 VHH domain from any of those provided herein. In some embodiments, at least one DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 244-318. In some embodiments, the DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 244-318, 401-409, 416, or 455. In some embodiments, the DLL3 VHH domain comprises an amino acid sequence shown in any of SEQ ID NO: 102, 244-318, 401-409, 416, 455, 476-480-488, and 507-518.

[0375] In certain embodiments, the restricted multispecific polypeptide construct contains at least two DLL3 domains. In some cases, at least one DLL3 VHH domain is located at the amino terminus of the Fc polypeptide relative to the heterodimer Fc, and at least one DLL3 VHH domain is located at the carboxyl terminus of the VH or VL chain relative to the CD3 binding region.

[0376] In a state of a restricted multispecific polypeptide construct containing at least two or more DLL3 VHH domains, each DLL3 VHH domain can bind to the same or overlapping antigenic determinants on DLL3.

[0377] In a state of a restricted multispecific polypeptide construct containing at least two or more DLL3 VHH domains, each of the DLL3 VHH domains can bind to different or non-overlapping antigenic determinants on DLL3.

[0378] In some embodiments, the first and second DLL3 VHH domains bind to different or non-overlapping antigenic determinants of DLL3, and / or do not compete for binding to DLL3.

[0379] In some cases, the VHH domain of the first sdAb contains an amino acid sequence shown in any of 264, 287, 299, 306, or 318, a humanized variant thereof, or an amino acid sequence showing at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of 264, 287, 299, 306, or 318, and binds to DLL3; and the second sdAb The VHH domain contains an amino acid sequence represented by any of 244, 258, 275, 280, 314, 316, or 317, a humanized variant thereof, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of 244, 258, 275, 280, 314, 316, or 317, and binds to DLL3.

[0380] In some cases, the first VHH domain contains an amino acid sequence, a humanized variant thereof, or an amino acid sequence showing at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity with any of SEQ ID NO: 251, 264, 267, 268, 287, 299, 306, 314, 318, 455, 507, or 517, and binds to DLL3; and the second VHH domain contains SEQ ID NO: 251, 264, 267, 268, 287, 299, 306, 314, 318, 455, 507, or 517. The amino acid sequence shown in any of SEQ ID NO: 244, 251, 258, 267, 275, 280, 314, 315, 316, 317, 318, 455, 515, 516, 517, 518, or a humanized variant thereof, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% consistency with any of SEQ ID NO: 244, 251, 258, 267, 275, 280, 314, 315, 316, 317, 318, 455, 515, 516, 517, 518, and binds to DLL3.

[0381] In some cases, the first sdAb VHH domain contains the amino acid sequence shown in SEQ ID NO:264 or its humanized variant shown in any of SEQ ID NO:265-455, the amino acid sequence shown in SEQ ID NO:287 or its humanized variant shown in any of SEQ ID NO:288-298, the amino acid sequence shown in SEQ ID NO:299 or its humanized variant shown in any of SEQ ID NO:300-305, the amino acid sequence shown in SEQ ID NO:306 or its humanized variant shown in any of SEQ ID NO:307-313 or SEQ ID NO:318; and the second sdAb VHH domain contains the amino acid sequence shown in SEQ ID NO:244 or its humanized variant shown in any of SEQ ID NO:245-257, the amino acid sequence shown in SEQ ID NO:258 or SEQ ID NO:264-265-255, or its humanized variant shown in any of SEQ ID NO:265-265-455. The humanized variant shown in any of NO:259-263, the amino acid sequence shown in SEQ ID NO:275, or the humanized variant shown in any of SEQ ID NO:276-279, the amino acid sequence shown in SEQ ID NO:280, or the humanized variant shown in any of SEQ ID NO:281-286, SEQ ID NO:314, SEQ ID NO:316, or SEQ ID NO:317.

[0382] In some cases, the first VHH domain contains the amino acid sequence shown in SEQ ID NO:264 or its humanized variant shown in any of SEQ ID NO:265-274, 416, 455 or 476-478; the amino acid sequence shown in SEQ ID NO:287 or its humanized variant shown in any of SEQ ID NO:102, 288-298; the amino acid sequence shown in SEQ ID NO:299 or its humanized variant shown in any of SEQ ID NO:300-305 or 480; the amino acid sequence shown in SEQ ID NO:306 or its humanized variant shown in any of SEQ ID NO:307-313; the amino acid sequence shown in SEQ ID NO:507 or its humanized variant shown in any of SEQ ID NO:508-514, SEQ ID NO:318 or SEQ ID NO 517; and the second VHH domain contains SEQ ID NO:264 or SEQ ID NO:265-274, 416, 455 or 476-478; the humanized variant shown in any of SEQ ID NO:287 or SEQ ID NO:102, 288-298; the humanized variant shown in any of SEQ ID NO:289 or SEQ ID NO:299; the humanized variant shown in any of SEQ ID NO:300-305 or 480; the humanized variant shown in any of SEQ ID NO:306 or SEQ ID NO:307-313; the humanized variant shown in any of SEQ ID NO:507 or SEQ ID NO:508-514, SEQ ID NO:318 or SEQ ID NO 517; and the second VHH domain contains SEQ ID NO:264 or SEQ ID NO:289 or SEQ ID NO:299. The amino acid sequence shown in NO:244 or any of the humanized variants shown in SEQ ID NO:245-257; the amino acid sequence shown in SEQ ID NO:258 or any of the humanized variants shown in SEQ ID NO:259-263; the amino acid sequence shown in SEQ ID NO:275 or any of the humanized variants shown in SEQ ID NO:276-279 or 479; the amino acid sequence shown in SEQ ID NO:280 or any of the humanized variants shown in SEQ ID NO:281-286, SEQ ID NO:314, SEQ ID NO:316, SEQ ID NO:515, SEQ ID NO:516 or SEQ ID NO:317.

[0383] In some embodiments, the first sdAbVHH domain and the second sdAbVHH domain comprise an amino acid sequence selected from the following: SEQ ID NO:244 and SEQ ID NO:264; SEQ ID NO:314 and SEQ ID NO:318; SEQ ID NO:244 and SEQ ID NO:306; SEQ ID NO:314 and SEQ ID NO:306; SEQ ID NO:314 and SEQ ID NO:299; SEQ ID NO:251 and SEQ ID NO:268; SEQ ID NO:251 and SEQ ID NO:267; SEQ ID NO:275 and SEQ ID NO:318; SEQ ID NO:314 and SEQ ID NO:287; SEQ ID NO:314 and SEQ ID NO:264; SEQ ID NO:316 and SEQ ID NO:318; SEQ ID NO:317 and SEQ ID NO:318; or SEQ ID NO:244 and SEQ ID NO:318. In some embodiments, the first sdAbVHH domain and the second sdAbVHH domain comprise the amino acid sequences shown in SEQ ID NO:315 and SEQ ID NO:318. In some embodiments, the first sdAbVHH domain and the second sdAbVHH domain comprise the amino acid sequences shown in SEQ ID NO:251 and SEQ ID NO:455.

[0384] In some embodiments, the first VHH domain and the second VHH domain comprise an amino acid sequence selected from the following: SEQ ID NO:244 and SEQ ID NO:264; SEQ ID NO:314 and SEQ ID NO:318; SEQ ID NO:314 and SEQ ID NO:517; SEQ ID NO:244 and SEQ ID NO:306; SEQ ID NO:244 and SEQ ID NO:507; SEQ ID NO:314 and SEQ ID NO:306; SEQ ID NO:314 and SEQ ID NO:507; SEQ ID NO:314 and SEQ ID NO:299; SEQ ID NO:251 and SEQ ID NO:268; SEQ ID NO:251 and SEQ ID NO:267; SEQ ID NO:275 and SEQ ID NO:318; 275 and SEQ ID NO:517; SEQ ID NO:314 and SEQ ID NO:287; SEQ ID NO:314 and SEQ ID NO:264; ...69; SEQ ID NO:268; SEQ ID NO:269; SEQ ID NO:268; SEQ ID NO:269; SEQ ID NO:268; SEQ ID NO:269; SEQ ID NO:268; SEQ ID NO:269; SEQ ID NO:268; SEQ ID NO:2 SEQ ID NO:314 and SEQ ID NO:314; SEQ ID NO:315 and SEQ ID NO:264; SEQ ID NO:518 and SEQ ID NO:264; SEQ ID NO:316 and SEQ ID NO:318; SEQ ID NO:515 and SEQ ID NO:517; SEQ ID NO:318 and SEQ ID NO:318; SEQ ID NO:517 and SEQ ID NO:517; SEQ ID NO:317 and SEQ ID NO:318; SEQ ID NO:516 and SEQ ID NO:517; SEQ ID NO:251 and SEQ ID NO:455; SEQ ID NO:244 and SEQ ID NO:517; or SEQ ID NO:244 and SEQ ID NO:318. In some embodiments, the first VHH domain and the second VHH domain comprise the amino acid sequences shown in SEQ ID NO:315 and SEQ ID NO:318. In some embodiments, the first VHH domain and the second VHH domain comprise the amino acid sequences shown in SEQ ID NO:518 and SEQ ID NO:517. In some embodiments, the first VHH domain and the second VHH domain comprise the amino acid sequences shown in SEQ ID NO:251 and SEQ ID NO:455.

[0385] In some embodiments, the restricted multispecific polypeptide construct contains at least one DLL3 VHH domain, such as any of those provided herein, and at least one other antigen-binding domain specifically targeting another tumor-associated antigen (TAA). In some embodiments, the at least one other antigen-binding domain comprises one or more replicas of an antibody or antigen-binding fragment thereof selected from the group consisting of: Fab fragments, F(ab')2 fragments, Fv fragments, scFv, scAb, dAb, single-domain heavy chain antibodies, and single-domain light chain antibodies. In certain embodiments, the other TAA antigen-binding domain is a single-chain antibody. In some instances, the single chain is scFv, scAb, single-domain heavy chain antibody, or single-domain light chain antibody. For example, in some cases, the other TAA antigen-binding domain comprises one or more single-domain antibody (sdAb) fragments, such as VHH, VNAR, engineered VH, or VK domains. VHH can be generated from natural camelid heavy chain antibodies, from genetically modified rodents that produce single heavy chain antibodies, or from an initial / synthetic or humanized camelid single-domain antibody library. VNARs can be generated from cartilaginous fish heavy chain antibodies alone. Various methods have been implemented to generate monomeric sdAbs using known heterodimer VH and VK domains, including interface engineering and selection for specific germline families.

[0386] In some embodiments, other TAAs are selected from the group consisting of: 1-92-LFA-3, 5T3, α-4 integrin, α-V integrin, α4β1 integrin, α4β7 integrin, AGR2, anti-Lewis-Y, Apelin J receptor, APRIL, B7-H3, B7-H4, BAFF, BTLA, C5 complement, C-242, CA9, CA19-9, (Lewis a), carbonic anhydrase 9, CD2, CD3, CD6, CD9, CD11a, CD19, CD20, CD22, CD24, CD25, CD27, CD28, CD30, CD33, CD38, CD40, CD40L, CD41, CD44 , CD44v6, CD47, CD51, CD52, CD56, CD64, CD70, CD71, CD74, CD80, CD81, CD86, CD95, CD117, CD123, CD125, CD132, (IL-2RG), CD133, CD137, CD138, CD166, CD172A, CD248, CDH6, CEACAM5 (CEA), CEACAM6 (NCA-90), CLAUDIN-3, CLAUDIN-4, cMet, collagen, teratoma-derived growth factor (Cripto), CSFR, CSFR-1, CTLA-4, CTGF, CXCL10, CXCL13, CXCR1, CXCR2, CXCR4, CYR61, DL44, DLK1, DLL4, DPP-4 DSG1, EDA, EDB, EGFR, EGFRviii, endothelin B receptor (ETBR), ENPP3, EpCAM, EPHA2, EPHB2, ERBB3, RSV F protein, FAP, FGF-2, FGF8, FGFR1, FGFR2, FGFR3, FGFR4, FLT-3, folate receptor α (FRα), GAL3ST1, G-CSF, G-CSFR, GD2, GITR, GLUT1, GLUT4, GM-CSF, GM-CSFR, GPIIb / IIIa receptors, Gp130, GPIIB / IIIA, GPNMB, GRP78, HER2 / neu, HER3, HER4, HGF, hGH, HVEM, hyaluronidase, ICOS, IFNα, IFNβ, IFNγ, IgE, IgE receptor (FceRI), IGF, IGF1R, IL1B, IL1R, IL2, IL11, IL12, IL12p40, IL-12R IL-12Rβ1, IL13, IL13R, IL15, IL17, IL18, IL21, IL23, IL23R, IL27 / IL27R (wsx1), IL29, IL-31R, IL31 / IL31R, IL2R, IL4, IL4R, IL6, IL6R, insulin receptor, serrated ligand, serrated 1, serrated 2, KISS1-R, LAG-3, LIF-R, Lewis X, LIGHT, LRP4, LRRC26, Ly6G6D, LyPD1, MCSP, mesothelin, MRP4, MUC1, mucin-16 (MUC16, CA-125), Na / K ATPase, NGF, idiocytin, Notch receptor, Notch 1, Notch 2, Notch 3, Notch 4. NOV, OSM-R, OX-40, PAR2, PDGF-AA, PDGF-BB, PDGFRα, PDGFRβ, PD-1, PD-L1, PD-L2, phosphatidylinosyl-serine, P1GF, PSCA, PSMA, PSGR, RAAG12, RAGE, SLC44A4, sphingosine monophosphate, STEAP1, STEAP2, TAG-72, TAPA1, TEM-8, TGFβ, TIGIT, TIM-3, TLR2 TLR4, TLR6, TLR7, TLR8, TLR9, TMEM31, TNFα, TNFR, TNFRS12A, TRAIL-R1, TRAIL-R2, transferrin, transferrin receptor, TRK-A, TRK-B, uPAR, VAP1, VCAM-1, VEGF, VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGFR1, VEGFR2, VEGFR3, VISTA, WISP-1, WISP-2, and WISP-3.

[0387] In some embodiments, an antigen-binding domain, such as the DLL3 VHH domain, is directly or indirectly linked to the Fc region and / or CD3-binding region via a linker. In some embodiments, the linker is via a linker. In some embodiments, the linker is a linker peptide (LP), which may include any of the flexible or rigid linkers. In some embodiments, the linker is selected from the group consisting of: GGSGGS, i.e., (GGS)2 (SEQ ID NO:1); GGSGGSGGS, i.e., (GGS)3 (SEQ ID NO:2); GGSGGSGGSGGS, i.e., (GGS)4 (SEQ ID NO:3); and GGSGGSGGSGGSGGS, i.e., (GGS)5 (SEQ ID NO:4). In some embodiments, the linker is a flexible linker containing glycine residues, such as by means of non-limiting examples GG, GGG, GGGG (SEQ ID NO:5), GGGGG (SEQ ID NO:6), and GGGGGG (SEQ ID NO:7). In some embodiments, the linker is (GGGGS)n, where n is 1 to 5 (SEQ ID NO: 123); (GGGGGS)n, where n is 1 to 4 (SEQ ID NO: 124); GGGGS (SEQ ID NO: 125); GGGGGS (SEQ ID NO: 126); GGGGGSGGGGGSGGGGS (SEQ ID NO: 127); GGGGSGGGGSGGGGS (SEQ ID NO: 128); or GGSGGGGSGGGGSGGGGS (SEQ ID NO: 129). In some embodiments, the linker comprises a combination of a GS linker and a glycine linker.

[0388] [b.Fc area]

[0389] Restricted multispecific polypeptide constructs include immunoglobulin Fc regions. Generally, restricted multispecific polypeptide constructs are dimers formed by polypeptides each containing an Fc region. The Fc polypeptide can be any polypeptide as described above. In certain embodiments, the Fc region is formed from Fc domains that are mutated or modified to promote heterodimerization, wherein different polypeptides can dimerize to produce a heterodimer. Therefore, in some embodiments, the dimer is a heterodimer in which the two polypeptide chains of the multispecific polypeptide construct are different.

[0390] Various methods for promoting the heterodimerization of complementary Fc peptides are known, see, for example, Ridgway et al., Protein Eng. 9:617-621 (1996); Merchant et al., Nat. Biotechnol. 16(7):677-81 (1998); Moore et al., (2011) MAbs, 3:546-57; Von Kreudenstein et al., MAbs, (2013) 5:646-54; Gunasekaran et al., (2010) J. Biol. Chem., 285:19637-46; Leaver-Fay et al., (2016) Structure, 24:641-51; Ha et al., (2016) Frontiers in Immunology, 7:1; Davis et al., (2010) Protein Eng Des Sel, 23:195-202; published international PCT applications No. WO 1998 / 050431, WO 2009 / 089004, WO2011143545, WO 2014 / 067011, WO 2012 / 058768, and WO2018027025; published U.S. patent applications No. US20140363426, US20150307628, US20180016354, and US20150239991; and U.S. patents No. US5731168, US7183076, US9701759, US9605084, and US9650446. Methods to promote Fc chain heterodimerization include Fc region mutagenesis, such as by incorporating a set of "mortar and pestle" mutations or by incorporating mutations that enable electrostatic manipulation of Fc to facilitate attractive interactions between different polypeptide chains. For example, in some embodiments, the Fc polypeptide in the heterodimer includes mutations that alter the charge polarity of the entire Fc dimer interface, such that the co-expression of electrostatically matched Fc chains supports favorable attractive interactions, thereby promoting the formation of the desired Fc heterodimer, while unfavorable repulsive charge interactions inhibit the formation of undesirable Fc homodimers (Guneskaran et al. (2010) JBC, 285:19637-19646). When co-expressed in cells, the chains may bind to each other, but self-binding is essentially not observed due to charge repulsion. Other strategies for generating heterodimeric Fcs include mixing human IgG with IgA CH3 domain segments to generate complementary CH3 heterodimers, termed SEED Fc.

[0391] The heterodimerization method and variants also include those described in the published International PCT application WO2014 / 145806, including the "mortar and pestle" mutation (also known as the "skew" variant), mutations related to "electrostatic manipulation" or "charge pairing," and pI variants. Heterodimer variants also include any variants described in U.S. Publication Applications US2012 / 0149876 or US2018 / 011883.

[0392] In some embodiments, to promote heterodimerization, both polypeptides of the Fc heterodimer contain paired or complementary amino acid modifications. Exemplary paired amino acid modifications of the polypeptides in the Fc fusion are described in... In [Table 3].

[0393]

[0394] In some embodiments, the modification includes introducing a protrusion (grooves) into a first Fc polypeptide and introducing a cavity (pothole) into a second Fc polypeptide, such that the protrusion can be positioned within the cavity to promote complexation of the first and second Fc-containing polypeptides. The amino acids targeted for substitution and / or modification to create protrusions or cavities in the polypeptide are typically interfacial amino acids that interact or contact one or more amino acids at the interface of the second polypeptide.

[0395] In some embodiments, the modified first Fc polypeptide containing a bulging (or bulging) amino acid comprises replacing the native or original amino acid with an amino acid having at least one side chain that protrudes from the interface of the first Fc polypeptide and is thus able to be positioned in a compensating cavity (or mortise) at the interface of an adjacent second polypeptide. The replaced amino acid is most often an amino acid with a side chain volume larger than the original amino acid residue. Those skilled in the art know how to determine and / or evaluate the characteristics of amino acid residues to identify those amino acids as ideal replaced amino acids for generating bulges. In some embodiments, the replaced residues used to form bulges are naturally occurring amino acid residues and include, for example, arginine (R), phenylalanine (F), tyrosine (Y), or tryptophan (W). In some instances, the original residues used for replacement are identified as amino acid residues with small side chains, such as alanine, aspartic acid, aspartic acid, glycine, serine, threonine, or valine.

[0396] In some embodiments, the modified second Fc polypeptide containing a cavity (cavity) is an Fc polypeptide comprising replacing the native or original amino acid with an amino acid having at least one side chain, the at least one side chain being recessed from the second polypeptide interface and thus capable of accommodating a corresponding protrusion from the first polypeptide interface. The replaced amino acid is most commonly an amino acid with a side chain volume smaller than the original amino acid residue. Those skilled in the art know how to determine and / or evaluate the characteristics of amino acid residues to identify those residues as ideal replacement residues for forming the cavity. Generally, the replacement residues used to form the cavity are naturally occurring amino acids and include, for example, alanine (A), serine (S), threonine (T), and valine (V). In some instances, the original amino acid used for replacement is identified as an amino acid with a large side chain, such as tyrosine, arginine, phenylalanine, or tryptophan.

[0397] The CH3 interface of human IgG1 involves, for example, sixteen residues located on each of four antiparallel β-terminal domains, each of which has a surface hidden at 1090 Ų (see, for example, Deisenhofer et al., (1981) Biochemistry, 20:2361-2370; Miller et al., (1990) J Mol. Biol., 216, 965-973; Ridgway et al., (1996) Prot. Engin., 9:617-621; U.S. Patent No. 5,731,168). Modification of the CH3 domain to create bulges or cavities is described, for example, in U.S. Patent No. 5,731,168; International Patent Applications WO98 / 50431 and WO 2005 / 063816; and Ridgway et al., (1996) Prot. Engin., 9:617-621. In some instances, modification of the CH3 domain to create bulges or cavities typically targets residues located on two antiparallel β-termini. The aim is to minimize the risk of the created bulges being contained in the surrounding solvent rather than by compensating cavities within the CH3 domain of the ligand.

[0398] For example, in some embodiments, the heterodimer Fc comprises a polypeptide having an amino acid modification located within the CH3 domain of Thr366, which, when replaced by a larger-volume amino acid (e.g., Try(T366W)), preferentially pairs with a second CH3 domain having smaller-volume amino acid modifications (e.g., Ser, Ala, and Val) at positions Thr366, Leu368, and Tyr407 (T366S / L368A / Y407V). The heterodimer achieved via CH3 modification can be further stabilized by introducing disulfide bonds, for example by changing Ser354 on the opposing CH3 domain to Cys(S354C) and Tyr349 to Cys(Y349C) (reviewed in Carter, 2001 Journal of Immunological Methods, 248:7-15).

[0399] In certain embodiments, the multispecific polypeptide construct contains first and second Fc peptides capable of mediating Fc heterodimerization, and contains a first Fc polypeptide having mutants T366W and S354C and a second Fc polypeptide having mutants T366S, L368A, Y407V, and Y349C. In some embodiments, the first Fc polypeptide is selected from Fc polypeptides containing the sequence described in SEQ ID NO: 440 or 446 and the second Fc polypeptide is selected from Fc polypeptides containing the sequence described in SEQ ID NO: 441, 444, or 448. In some embodiments, the first Fc polypeptide is or contains an amino acid sequence shown in any one of SEQ ID NO: 103, 107, 115, or 117, and the second Fc polypeptide is or contains an amino acid sequence shown in any one of SEQ ID NO: 104, 108, 111, 113, 119, or 121.

[0400] In some embodiments, the Fc peptide exhibits characteristics of Fc-mediated effector function. In a particular example, the first Fc peptide is or contains the sequence described in SEQ ID NO: 440 and the second Fc peptide is or contains SEQ ID NO: 441 or 444. In some embodiments, the first Fc peptide is or contains the sequence described in SEQ ID NO: 103 and the second Fc peptide is or contains the sequence described in SEQ ID NO: 104 or 111. In some embodiments, the first Fc peptide is or contains the sequence described in SEQ ID NO: 107 and the second Fc peptide is or contains the sequence described in SEQ ID NO: 108 or 113. The first and second Fc peptides may be formatted on either polypeptide chain of the construct.

[0401] In some embodiments, one or both of the first and second Fc peptides may further include one or more amino acid mutations to further reduce one or more Fc effector functions, such as reducing Fc receptor binding. Exemplary mutations reducing Fc effector function include any mutations as described. In some embodiments, the modification may be the deletion of one or more positions of Glu233 (E233), Leu234 (L234), or Leu235 (L235), such as the deletion of the amino acids Glu233 (E233), Leu234 (L234), and Leu235 (L235). In some embodiments, the first Fc peptide is selected from Fc peptides comprising the sequence described in SEQ ID NO: 442 or 447, and the second Fc peptide is selected from Fc peptides comprising the sequence described in SEQ ID NO: 443, 445, or 449. In some embodiments, the first Fc polypeptide is or contains an amino acid sequence shown in any one of SEQ ID NO: 105, 109, 116 or 118, and the second Fc polypeptide is or contains an amino acid sequence shown in any one of SEQ ID NO: 106, 110, 112, 114, 120 or 122.

[0402] In certain instances, the first Fc polypeptide is or contains the sequence described in SEQ ID NO: 442 and the second Fc polypeptide is or contains SEQ ID NO: 443 or 445. In some embodiments, the first Fc polypeptide is or contains the sequence described in SEQ ID NO: 105 and the second Fc polypeptide is or contains the sequence described in SEQ ID NO: 106 or 112. In some embodiments, the first Fc polypeptide is or contains the sequence described in SEQ ID NO: 109 and the second Fc polypeptide is or contains the sequence described in SEQ ID NO: 110 or 114. The first and second Fc polypeptides may be formatted on either polypeptide chain of the construct.

[0403] In some embodiments, the first Fc polypeptide or the second Fc polypeptide further comprises the mutant M252Y and / or M428V. In a particular example, the first Fc polypeptide is or contains the sequence described in SEQ ID NO:446 and the second Fc polypeptide is or contains the sequence described in SEQ ID NO:448. In some embodiments, the first Fc polypeptide is or contains the sequence described in SEQ ID NO:115 and the second Fc polypeptide is or contains the sequence described in SEQ ID NO:119. In some embodiments, the first Fc polypeptide is or contains the sequence described in SEQ ID NO:117 and the second Fc polypeptide is or contains the sequence described in SEQ ID NO:121. In other embodiments, the first Fc polypeptide is or contains the sequence described in SEQ ID NO:447 and the second Fc polypeptide is or contains the sequence described in SEQ ID NO:449. In some embodiments, the first Fc polypeptide is or contains the sequence described in SEQ ID NO:116 and the second Fc polypeptide is or contains the sequence described in SEQ ID NO:120. In some embodiments, the first Fc polypeptide is or contains the sequence described in SEQ ID NO:118 and the second Fc polypeptide is or contains the sequence described in SEQ ID NO:122. The first and second Fc polypeptides may be formatted on either polypeptide chain of the construct.

[0404] Other variants that may contribute to the formation of heterodimers include any combination or pair of spatial variants (e.g., skew variants) of the first Fc polypeptide and the second Fc polypeptide, selected from: S364K / E357Q and L368D / K370S; L368D / K370S and S364K; L368E / K370S and S364K; T411T / E360E / Q362E and D401K; L368D / K370S and S364K / E357L, K370S and S364K / E357Q and T366S / L368A / Y407V and T366W or 366S / L368A / Y407V / Y349C and T366W / S354C), where each pair represents a mutation in the first Fc polypeptide and the second Fc polypeptide. In a particular embodiment, the provided construct contains first and second Fc polypeptides having mutant pairs L368D / K370S and S364K and E357Q.

[0405] Another mechanism that can be used to generate heterodimers is sometimes referred to as “electrostatic manipulation,” as described by Gunasekaran et al., J. Biol. Chem. 285(25):19637 (2010). This is sometimes referred to herein as “charge pair.” In this embodiment, electrostatic bias is used to form heterodimerization. As those skilled in the art will understand, such mechanisms can also affect pI and thus purification, and therefore can be considered pI variants in some cases. However, since such variant systems are generated to force heterodimerization and are not used as purification tools, they are classified as “spatial variants.” In one embodiment, the first Fc polypeptide may contain the mutant D221E / P228E / L368E and the second Fc polypeptide may contain the mutant D221R / P228R / K409R. In another embodiment, the first Fc polypeptide may contain the mutant C220E / P228E / 368E and the second Fc polypeptide may contain the mutant C220R / E224R / P228R / K409R.

[0406] In some embodiments, pI variants can promote heterodimerization. In some states, pI variants may include variants that increase the protein pI (basicity change). In other states, pI variants may include variants that decrease the protein pI (acidicity change). In some cases, all combinations of such variants can be achieved, including combinations where one Fc polypeptide is wild-type, or a variant showing no significant difference in pI from the wild-type, and another Fc polypeptide is either more basic or more acidic. Alternatively, the Fc polypeptides can be modified, one to be more basic and the other to be more acidic. In some embodiments, at least one Fc polypeptide is a negative pI variant Fc containing the mutant Q295E / N384D / Q418E / N421D.

[0407] In some embodiments, spatial heterodimer variants (e.g., pestle and mortar) can be used in combination with pI or charge-pair variants.

[0408] In a particular embodiment, the provided construct comprises: (a) a first Fc polypeptide comprising the skewed variant S364K / E357Q; and (b) a second Fc polypeptide comprising the skewed variant L368D / K370S and the pI variant N208D / Q295E / N384D / Q418E / N421D. In some embodiments, one or both of the first and second polypeptides may contain other mutations that reduce Fc effector activity, such as the exemplary mutations E233P / L234V / L235A / G236del / S267K. Examples of such first and second Fc polypeptides capable of mediating Fc heterodimerization include the sequences described in SEQ ID NO: 472 and 473. The first and second Fc polypeptides may be formatted on either polypeptide chain of the construct.

[0409] The resulting restricted multispecific polypeptide constructs can be purified by any suitable method, such as affinity chromatography using protein A or protein G columns. In the case of transformation of two nucleic acid molecules encoding different polypeptides in the cell, homodimers and heterodimers will be formed. Conditions used for expression can be adjusted to favor heterodimer formation rather than homodimer formation.

[0410] Techniques for recovering heterodimers from homodimers based on the differential affinity of heterodimers for affinity reagents are known. In some cases, such techniques involve designing heterodimers such that one of the Fc polypeptide chains does not bind to the affinity reagent protein A. In some cases, one of the polypeptide chains may contain one or more amino acid substitutions to eliminate or reduce the affinity of one of the polypeptides in the Fc heterodimer for the protein A reagent, see, for example, WO2017134440, WO2010151792, Jendeberg et al. (Jendeberg et al., (1997) J. Immunol. Methods, 201(1):25-34). In some of these embodiments, the Fc region may be modified at the protein A binding site of one member of the heterodimer to prevent protein A binding and thereby enable more efficient purification. Heterodimeric fusion proteins. An exemplary modification within this binding site is Ile253, such as Ile253Arg (I253R). In some embodiments, the modification may be H435R or H435R / Y436F. In some embodiments, the Fc polypeptide in the Fc heterodimer may contain modifications that enable it to bind protein A, rather than protein G (pA+ / pG-). Exemplary pA+ / pG amino acid modifications include, with reference to human IgG1, the Fc containing serine at position 428, serine at position 434, and histidine at position 436, if applicable, or containing such residues at corresponding positions in human IgG2, 3, or 4. In some forms, such amino acid modifications at positions 428, 434, and, if applicable, 436 of an IgG Fc polypeptide reduce or prevent protein G binding, thereby enhancing protein purification.

[0411] In some embodiments, any such modification imparting differential affinity to the affinity reagent may be combined with any one or more other amino acid modifications described above. For example, the I253R modification may be combined with the T366S / L368A / Y407V modification or with the T366W modification. Fc modified with T366S / L368A / Y407V can form a homodimer due to the absence of steric hindrance at the dimer interface, whereas T336W-modified Fc exhibits steric hindrance at the dimer interface. Therefore, in some embodiments, combining the I253R modification with the T366S / L368A / Y407V-modified Fc prevents the purification of any homodimer Fc that may have formed. Similar modifications can be used by combining T366S / L368A / Y407V and H453R.

[0412] In some embodiments, the Fc region of the heterodimer molecule may additionally contain one or more other Fc mutations, such as any of the mutations described above. In some embodiments, the heterodimer molecule contains an Fc region with a mutation that reduces effector function. In some embodiments, the Fc region is modified to reduce Fc-mediated effector function, such as via reduced Fc receptor binding, for example, binding to FcγR, but not typically FcRn binding.

[0413] In some embodiments, the Fc region is mutated at one or more of the following locations to reduce Fc receptor binding: Glu233 (E233), Leu234 (L234), or Leu235 (L235). One or more mutations may include E233P, L234V, and / or L235A.

[0414] In certain embodiments, mutations in the Fc region that reduce Fc effector function (e.g., by reducing Fc receptor binding to FcγR) include mutations selected from the following: G236R / L328R, E233P / L234V / L235A / G236del / S239K, E233P / L234V / L235A / G236del / S267K, E233P / L234V / L235A / G236del / S239K / A327G, E23 3P / L234V / L235A / G236del / S267K / A327G or E233P / L234V / L235A / G236del, D265A / P329A, D265A / P329G, D265A / N297A, L234V / L235A / D265A, L234V / L235A / N297A, L234V / L235A / P329A, or L234V / L235A / P329G.

[0415] In some embodiments, one Fc polypeptide in the heterodimer Fc contains the amino acid sequence shown in any one of SEQ ID NO: 440 (e.g., SEQ ID NO: 103 or 107) and 446 (e.g., SEQ ID NO: 115 or 117), and the other Fc polypeptide in the heterodimer Fc contains the amino acid sequence shown in any one of SEQ ID NO: 441 (e.g., SEQ ID NO: 104 or 108), 444 (e.g., SEQ ID NO: 111 or 113), and 448 (e.g., SEQ ID NO: 119 or 121). In some embodiments, one Fc polypeptide in the heterodimer Fc contains the amino acid sequence shown in any one of SEQ ID NO:442 (e.g., SEQ ID NO:105 or 109) and 447 (e.g., SEQ ID NO:116 or 118), and the other Fc polypeptide in the heterodimer Fc contains the amino acid sequence shown in any one of SEQ ID NO:443 (e.g., SEQ ID NO:106 or 110), 445 (e.g., SEQ ID NO:112 or 114), and 449 (SEQ ID NO:120 or 122).

[0416] In some embodiments, the Fc region of the provided multispecific polypeptide construct exhibits one or more effector functions. In some cases, the Fc region can provide Fc-mediated effector functions, such as ADCC (e.g., NK cell-released granzyme B), ADCP, and / or CDC. Generally, the Fc region is responsible for effector functions, such as complement-dependent cytotoxicity (CDC) and antibody-dependent cytotoxicity (ADCC), and also for antigen-binding capacity, which is a major function of immunoglobulins. In addition, the FcRn sequence present in the Fc region increases the in vivo half-life by binding to the in vivo FcRn receptor, thereby regulating the serum IgG content. In some embodiments where the multispecific polypeptide construct contains a cleavable linker, linker cleavage can produce two biologically active components: a CD3-binding region that binds to and conjugates CD3 on T cells, which in some states may also contain a CRBR for inducing co-stimulatory signals to T cells and / or an IRBR for inducing inhibitory signals to T cells; and an Fc region connected to the DLL3 VHH domain, which can exhibit target-specific effector functions. In the specific embodiments provided herein, the multispecific polypeptide constructs contain non-cleavable linkers and, in some states, may not exhibit independent Fc-mediated effector functionality.

[0417] In some embodiments, the Fc region includes an Fc polypeptide that has been mutated or modified to alter one or more effector functions. Therefore, in some cases, the provided restricted multispecific polypeptide constructs can be used to alter (e.g., reduce or enhance) the effector function of the Fc, such as one or more of ADCC, ADCP, and / or CDC. Exemplary mutations that reduce effector function include any of those described above.

[0418] In some embodiments, IgG1 Fc peptides or variants thereof, such as those described below, can be prepared using G1 m1 or G1 m3 isoforms. In some embodiments, the Fc region may contain amino acids of the human G1 m1 isoform, such as residues containing Asp(D) and Leu(L) at positions 356 and 358, as described in SEQ ID NO: 8. In some cases, the Fc peptide may contain amino acid substitutions for E356D and M358L to reconstruct the residues of the isoform G1 m1. In other embodiments, the Fc region may contain amino acids of the human G1 m3 isoform, such as residues Glu(E) and Met(M) at positions 356 and 358 according to EU numbers, as described in SEQ ID NO: 472 and 473. In some cases, the Fc peptide may contain amino acid substitutions for D356E and L358M to reconstruct the residues of the isoform G1 m3.

[0419] [c.CD3 associative domain]

[0420] Restricted multispecific polypeptide constructs include one or more replicas of an anti-CD3 binding domain. The anti-CD3 binding domain of the present invention activates T cells by binding to CD3 or a member of the CD3 complex on T cells. In a preferred embodiment, the anti-CD3 binding domain of the present invention specifically binds to the ε chain of CD3, also referred to as CD3ε. The anti-CD3ε binding domain of the present invention activates T cells by binding to CD3ε on T cells. The anti-CD3 binding domain of the present invention promotes, stimulates, activates, and / or otherwise enhances CD3-mediated T cell activation. The biological activities of CD3 include, for example, T cell activation and other signal transductions achieved through the interaction between CD3 and the antigen-binding subunit of the T cell receptor (TCR). For example, the anti-CD3 binding domain of the present invention fully or partially activates T cells by binding to CD3ε on T cells, thereby partially or completely modulating (e.g., promoting, stimulating, activating, or otherwise enhancing) CD3-mediated T cell activation.

[0421] The CD3 binding domain may be any of those described above. In a particular embodiment, the CD3 binding domain is an Fv antibody fragment that binds to CD3ε (referred to herein as an anti-CD3ε Fv fragment). In some embodiments, the anti-CD3ε Fv antibody fragment is a disulfide-stabilized anti-CD3 binding Fv fragment (dsFv). In some embodiments, the anti-CD3 binding domain is monovalent with respect to binding CD3.

[0422] In some embodiments, the CD3 binding region is an Fv antibody fragment containing a variable heavy chain (Hv, also known as VH) and a variable light chain (Lv, also known as VL), such as any of the above. In such embodiments, the immunoglobulin Fc region is a heterodimeric Fc region containing two different Fc polypeptides, which enables heterodimeric binding between the two polypeptides in the Fc heterodimer, such as any of the above. In such embodiments, the variable heavy chain (VH) and variable light chain (VL) of the CD3 binding region are linked by opposite chains of the heterodimeric Fc.

[0423] In some embodiments, the CD3 binding region is the Fv or dsFv of SP34 (Pessano et al., The EMBO Journal.4:337-344, 1985) or a humanized variant of SP34 (WO2015001085).

[0424] In some embodiments, the anti-CD3ε binding domain is an Fv, such as a dsFv fragment, which includes a combination of heavy chain variable region amino acid sequences and light chain variable region amino acid sequences. In some embodiments, the CD3 binding domain is an Fv or dsFv fragment containing at least a VH CDR1 sequence including the amino acid sequence TYAMN (SEQ ID NO: 29); at least a VH CDR2 sequence including the amino acid sequence RIRSKYNNYATYYADSVKD (SEQ ID NO: 30); at least a VH CDR3 sequence including the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); at least a VL CDR1 sequence including the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO: 32); at least a VL CDR2 sequence including the amino acid sequence GTNKRAP (SEQ ID NO: 33); and at least a VL CDR3 sequence including the amino acid sequence ALWYSNLWV (SEQ ID NO: 34). In some embodiments, the anti-CD3ε binding domain is Fv, such as the dsFv fragment, which includes a heavy chain variable region amino acid sequence selected from the group consisting of SEQ ID NO: 35-65 and a light chain variable region amino acid sequence selected from the group consisting of SEQ ID NO: 66-84 or 368.

[0425] In some embodiments, the anti-CD3ε binding domain includes a VH CDR1 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence TYAMN (SEQ ID NO: 29); a VH CDR2 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence RIRSKYNNYATYYADSVKD (SEQ ID NO: 30); a VH CDR3 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); and a VH CDR3 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO: 29). The VL CDR1 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GTNKRAP (SEQ ID NO:33); the VL CDR2 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence ALWYSNLWV (SEQ ID NO:34); and the VL CDR3 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence ALWYSNLWV (SEQ ID NO:34).

[0426] In some embodiments, the anti-CD3ε binding domain includes: a VH CDR1 sequence comprising at least the amino acid sequence GFTNTYAMN (SEQ ID NO: 461); a VH CDR2 sequence comprising at least the amino acid sequence RIRSKYNNYATY (SEQ ID NO: 462); a VH CDR3 sequence comprising at least the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); a VL CDR1 sequence comprising at least the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO: 32); a VL CDR2 sequence comprising at least the amino acid sequence GTNKRAP (SEQ ID NO: 33); and a VL CDR3 sequence comprising at least the amino acid sequence ALWYSNLWV (SEQ ID NO: 34).

[0427] In some embodiments, the anti-CD3ε binding domain includes a VH CDR1 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GFTNTYAMN (SEQ ID NO: 461); a VH CDR2 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence RIRSKYNNYATY (SEQ ID NO: 462); a VH CDR3 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); and a VH CDR3 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO: 461). The VL CDR1 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GTNKRAP (SEQ ID NO:33); the VL CDR2 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence ALWYSNLWV (SEQ ID NO:34); and the VL CDR3 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence ALWYSNLWV (SEQ ID NO:34).

[0428] In some embodiments, the anti-CD3ε binding domain includes: a VH CDR1 sequence comprising at least the amino acid sequence GFTNTYAMN (SEQ ID NO: 461); a VH CDR2 sequence comprising at least the amino acid sequence RIRSKYNNYATY (SEQ ID NO: 462); a VH CDR3 sequence comprising at least the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); a VL CDR1 sequence comprising at least the amino acid sequence GSSTGAVTTSNYAN (SEQ ID NO: 468); a VL CDR2 sequence comprising at least the amino acid sequence GTNKRAP (SEQ ID NO: 469); and a VL CDR3 sequence comprising at least the amino acid sequence ALWYSNHWV (SEQ ID NO: 464).

[0429] In some embodiments, the anti-CD3ε binding domain includes a VH CDR1 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GFTNTYAMN (SEQ ID NO: 461); a VH CDR2 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence RIRSKYNNYATY (SEQ ID NO: 462); a VH CDR3 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); and a VH CDR3 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GSSTGAVTTSNYAN (SEQ ID NO: 461). The VL CDR1 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GTNKRAP (SEQ ID NO:469); the VL CDR2 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence ALWYSNHWV (SEQ ID NO:464); and the VL CDR3 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence ALWYSNHWV (SEQ ID NO:464).

[0430] In some embodiments, the anti-CD3ε binding domain includes: a VH CDR1 sequence comprising at least the amino acid sequence GFTFSTYAMN (SEQ ID NO: 466); a VH CDR2 sequence comprising at least the amino acid sequence RIRSKYNNYATY (SEQ ID NO: 467); a VH CDR3 sequence comprising at least the amino acid sequence HGNFGDSYVSWFAY (SEQ ID NO: 463); a VL CDR1 sequence comprising at least the amino acid sequence GSSTGAVTTSNYAN (SEQ ID NO: 468); a VL CDR2 sequence comprising at least the amino acid sequence GTNKRAP (SEQ ID NO: 469); and a VL CDR3 sequence comprising at least the amino acid sequence ALWYSNHWV (SEQ ID NO: 464).

[0431] In some embodiments, the anti-CD3ε binding domain includes a VH CDR1 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GFTFSTYAMN (SEQ ID NO: 466); a VH CDR2 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence RIRSKYNNYATY (SEQ ID NO: 467); a VH CDR3 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence HGNFGDSYVSWFAY (SEQ ID NO: 463); and a VH CDR3 sequence that is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GSSTGAVTTSNYAN (SEQ ID NO: 466). The VL CDR1 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence GTNKRAP (SEQ ID NO:469); the VL CDR2 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence ALWYSNHWV (SEQ ID NO:464); and the VL CDR3 sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to the amino acid sequence ALWYSNHWV (SEQ ID NO:464).

[0432] In some embodiments, the anti-CD3ε binding domain comprises a CDR3 comprising at least the amino acid VLWYSNRWV (SEQ ID NO: 465). In some embodiments, the anti-CD3ε binding domain comprises a CDR3 that is identical to at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more of the amino acid VLWYSNRWV (SEQ ID NO: 465).

[0433] In some embodiments, the anti-CD3ε binding domain includes one or more replicas of an antibody or its antigen-binding fragment selected from the group consisting of: Fab fragment, F(ab')2 fragment, Fv fragment, scFv, scAb, dAb, single-domain heavy chain antibody, and single-domain light chain antibody. In some embodiments, the anti-CD3 binding domain includes an Fv antibody fragment that binds to CD3ε (referred to herein as the anti-CD3ε Fv fragment). In some embodiments, the anti-CD3ε Fv antibody fragment is a disulfide-stabilized anti-CD3 binding Fv fragment (dsFv). In some embodiments, the anti-CD3 binding domain is monovalent with respect to binding CD3.

[0434] In some embodiments, the CD3 binding region is not a single-chain antibody. For example, in some samples, the CD3 binding region is not a single-chain variable fragment (scFv).

[0435] In some embodiments, the CD3 binding region is an Fv antibody fragment containing a variable heavy chain (Hv, also known as VH) and a variable light chain (Lv, also known as VL), such as any of those described above. In such embodiments, the immunoglobulin Fc region is a heterodimeric Fc region containing two different Fc polypeptides, capable of achieving heterodimeric binding of the two polypeptides in the Fc heterodimer, such as any of those described in Section III.C.2.b. In such embodiments, the variable heavy chain (VH) and variable light chain (VL) of the CD3 binding region are linked by opposite chains of the heterodimeric Fc.

[0436] In some embodiments, the anti-CD3ε binding domain comprises a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence and the light chain variable region amino acid sequence comprise amino acid sequences selected from the group consisting of SEQ ID NO: 27, 28, 35-84, 368, 451-454, and 460. In some embodiments, the anti-CD3ε binding domain comprises a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence is selected from the group consisting of SEQ ID NO: 27, 35-65, 453, 454, and 460, and the light chain variable region amino acid sequence comprises amino acid sequences selected from the group consisting of SEQ ID NO: 28, 66-84, 368, 451, and 452.

[0437] In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence. In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence and the light chain variable region amino acid sequence contain amino acid sequences that are at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to amino acid sequences selected from the group consisting of SEQ ID NO: 35-84, 368, 451-454 and 460. In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence and the light chain variable region amino acid sequence comprise amino acid sequences selected from the group consisting of SEQ ID NO: 35-84, 368, 451-454, and 460. In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence is selected from the group consisting of SEQ ID NO: 35-65, 453, 454, and 460, and the light chain variable region amino acid sequence is selected from the group consisting of SEQ ID NO: 66-84, 368, 451, and 452. In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to amino acid sequences selected from the group consisting of SEQ ID NO: 35-65, 453, 454 and 460, and the light chain variable region amino acid sequence is at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more identical to amino acid sequences selected from the group consisting of SEQ ID NO: 66-84, 368, 451 and 452.

[0438] In some embodiments, the anti-CD3ε binding domain is an Fv or dsFv fragment comprising a heavy chain variable region amino acid sequence that is identical to at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more of the amino acid sequences selected from the group consisting of SEQ ID NO: 35-65, 453, 454 and 460, and an amino acid sequence that is identical to at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more of the amino acid sequences selected from the group consisting of SEQ ID NO: 66-84 or 368, 451 and 452. In some embodiments, the anti-CD3 binding domain is an Fv or dsFv fragment containing a variable heavy chain (VH) comprising the amino acid sequence of SEQ ID NO: 47 and a variable light chain (VL) comprising the amino acid sequence of SEQ ID NO: 75. In some embodiments, the anti-CD3 binding domain is Fv or dsFv, which contains a variable heavy chain (VH) containing the amino acid sequence of SEQ ID NO:47 and a variable light chain (VL) containing the amino acid sequence of SEQ ID NO:368.

[0439] In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence. In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence and the light chain variable region amino acid sequence contain at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more of an amino acid sequence selected from the group consisting of SEQ ID NO: 27, 28, 35-84, 368, 451-454, 460. In some embodiments, the anti-CD3ε binding domain is an Fv fragment comprising a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence and the light chain variable region amino acid sequence comprise an amino acid sequence selected from the group consisting of SEQ ID NO: 27, 28, 35-84, 368, 451-454, 460. In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence is identical to at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more of the amino acid sequences selected from the group consisting of SEQ ID NO: 27, 35-65, 453, 454 and 460, and the light chain variable region amino acid sequence is identical to at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more of the amino acid sequences selected from the group consisting of SEQ ID NO: 28, 66-84, 368, 451 and 452. In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence is selected from the group consisting of SEQ ID NO: 27, 35-65, 453, 454 and 460, and the light chain variable region amino acid sequence is selected from the group consisting of SEQ ID NO: 28, 66-84, 368, 451 and 452.

[0440] In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence is identical to at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more of the amino acid sequences selected from the group consisting of SEQ ID NO: 27, 35-46, 48-50, 453, 454 and 460, and the light chain variable region amino acid sequence is identical to at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more of the amino acid sequences selected from the group consisting of SEQ ID NO: 28, 66, 68-74, 76, 78, 80, 368, 451 and 452. In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence is selected from the group consisting of SEQ ID NO: 27, 35-46, 48-50, 453, 454 and 460, and the light chain variable region amino acid sequence is selected from the group consisting of SEQ ID NO: 28, 66, 68-74, 76, 78, 80, 451 and 452.

[0441] In some embodiments, the anti-CD3ε Fv antibody fragment comprises a combination of a heavy chain variable region amino acid sequence and a light chain variable region amino acid sequence, wherein the heavy chain variable region amino acid sequence is identical to at least 90%, 91%, 92%, ...

Claims

1. A DLL3-binding polypeptide construct comprising at least one heavy-chain-only variable domain (DLL3 VHH domain), the DLL3 VHH domain specifically binding to DLL3 and comprising complementarity-determining region 1 (CDR1), complementarity-determining region 2 (CDR2), and complementarity-determining region 3 (CDR3), wherein CDR1, CDR2, and CDR3 each comprise the following amino acid sequences: i. SEQ ID NO: 319, 336, and 354; ii. SEQ ID NO: 319, 337, and 354; iii. SEQ ID NO: 319, 338, and 354; iv. SEQ ID NO: 320, 338, and 354; v. SEQ ID NO: 321, 338, and 354; vi. SEQ ID NO: 322, 338, and 354; vii. SEQ ID NO: 323, 338, and 354; viiii. SEQ ID NO: 324, 338, and 354; ix. SEQ ID NO: 319, 336, and 354; NO: 325, 338 and 354; or x.SEQ ID NO: 326, 338 and 354.

2. The DLL3-binding polypeptide construct of claim 1, comprising one or more binding domains to a target other than DLL3.

3. The DLL3-binding polypeptide construct of claim 1 or 2, wherein the DLL3 is human DLL3.

4. The DLL3-binding polypeptide construct of claim 2, wherein one or more other binding domains bind to an activating receptor on an immune cell.

5. The DLL3-binding polypeptide construct of claim 4, wherein the immune cell is a T cell or a natural killer (NK) cell.

6. The DLL3-binding polypeptide construct of claim 4, wherein the activating receptor is CD3 or CD16.

7. The DLL3-binding polypeptide construct of claim 1 or 2 further comprises a radioactive reagent.

8. The DLL3-binding polypeptide construct of claim 2, wherein the one or more other binding domains are interleukins or truncated fragments or variants thereof capable of binding to interleukin receptors.

9. The DLL3-binding polypeptide construct of claim 2, wherein the one or more other binding domains comprise an antibody or an antigen-binding fragment thereof, wherein the antibody or antigen-binding fragment thereof is Fv, a disulfide-stabilized Fv (dsFv), scFv, Fab, a single-domain antibody (sdAb), VNAR, or VHH.

10. The DLL3-binding polypeptide construct of claim 9, wherein the DLL3-binding polypeptide construct includes an immunoglobulin Fc region, or wherein the immunoglobulin Fc region connects the at least one VHH domain to the one or more other binding domains.

11. The DLL3-binding polypeptide construct of claim 10, wherein the Fc region of the immunoglobulin is a heterodimeric Fc region.

12. The DLL3-binding polypeptide construct as requested in item 1 or 2, wherein: The at least one DLL3 VHH domain contains an amino acid sequence that exhibits at least 95% sequence identity with any of SEQ ID NO: 245-257 and binds to DLL3.

13. The DLL3-binding polypeptide construct as requested in item 1 or 2, wherein: The at least one DLL3 VHH domain contains the amino acid sequence shown in any of SEQ ID NO: 245-257.

14. The DLL3-binding polypeptide construct of claim 1 or 2, comprising: (a) a first component comprising a heterodimeric Fc region containing a first Fc polypeptide and a second Fc polypeptide; and (b) a second component comprising an anti-CD3 antibody or antigen-binding fragment containing a heavy chain variable region (VH) and a light chain variable region (VL), wherein: The VH and VL constituting the anti-CD3 antibody or antigen-binding fragment are linked to the corresponding polypeptide of the heterodimer Fc; the first component and the second component are coupled by a linker, wherein the heterodimer Fc region is located at the N-terminus of the anti-CD3 antibody; and one or both of the first and second components contain at least one DLL3 VHH domain.

15. The DLL3-binding polypeptide construct of claim 14, wherein one or both of the first Fc polypeptide and the second Fc polypeptide of the heterodimeric Fc region contain at least one modification that induces heterodimerization compared to the polypeptide of the homodimeric Fc region.

16. The DLL3-binding polypeptide construct of claim 14, wherein one or both of the first Fc polypeptide and the second Fc polypeptide of the heterodimer Fc region contain at least one modification that induces heterodimerization compared to the Fc polypeptide of SEQ ID NO: 8 or its immunoactive fragment.

17. The DLL3-binding polypeptide construct of claim 14, wherein each of the first and second Fc polypeptides in the Fc region of the heterodimer comprises a mortar and pestle modification or comprises a charge mutation that enhances the electrostatic complementarity of the polypeptides.

18. The DLL3-binding polypeptide construct of claim 14, wherein the anti-CD3 antibody or antigen-binding fragment is an Fv antibody fragment.

19. The DLL3-binding polypeptide construct of claim 18, wherein the Fv antibody fragment comprises a disulfide-stabilized anti-CD3-binding Fv fragment (dsFv).

20. The DLL3-binding polypeptide construct of claim 14, wherein the anti-CD3 antibody or antigen-binding fragment comprises: (a) VH CDR1, comprising the amino acid sequence TYAMN (SEQ ID NO: 29); VH CDR2, comprising the amino acid sequence RIRSKYNNYATYYADSVKD (SEQ ID NO: 30); VH CDR3, comprising the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); VL CDR1, comprising the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO: 32); VL CDR2, comprising the amino acid sequence GTNKRAP (SEQ ID NO: 33); and VL CDR3, comprising the amino acid sequence ALWYSNLWV (SEQ ID NO: 34); (b) VH CDR1, comprising the amino acid sequence GFTNTYAMN (SEQ ID NO: 461); VH CDR2, comprising the amino acid sequence RIRSKYNNYATY (SEQ ID NO: 462); VH CDR3, comprising the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); VL CDR1, comprising the amino acid sequence RSSTGAVTTSNYAN (SEQ ID NO: 32); VL CDR2, comprising the amino acid sequence GTNKRAP (SEQ ID NO: 33); and VL CDR3, comprising the amino acid sequence ALWYSNLWV (SEQ ID NO: 34); (c) VH CDR1, comprising the amino acid sequence GFTNTYAMN (SEQ ID NO: 461); VH CDR2, comprising the amino acid sequence RIRSKYNNYATY (SEQ ID NO: 462); VH CDR3, comprising the amino acid sequence HGNFGNSYVSWFAY (SEQ ID NO: 31); VL CDR1, comprising the amino acid sequence GSSTGAVTTSNYAN (SEQ ID NO: 468); VL CDR2, comprising at least the amino acid sequence GTNKRAP (SEQ ID NO: 469); and VL CDR3, comprising the amino acid sequence ALWYSNHWV (SEQ ID NO: 464); or (d) VH CDR1, comprising the amino acid sequence GFTFSTYAMN (SEQ ID NO: 466); VH CDR2, comprising the amino acid sequence RIRSKYNNYATY (SEQ ID NO: 467); and VH CDR3, comprising the amino acid sequence HGNFGDSYVSWFAY (SEQ ID NO: 463);VL CDR1, comprising the amino acid sequence GSSTGAVTTSNYAN (SEQ ID NO: 468); VL CDR2, comprising the amino acid sequence GTNKRAP (SEQ ID NO: 469); and VL CDR3, comprising the amino acid sequence ALWYSNHWV (SEQ ID NO: 464).

21. The DLL3-binding polypeptide construct of claim 14, wherein the anti-CD3 antibody or antigen-binding fragment comprises: VH having an amino acid sequence of any one of SEQ ID NO: 35-65; and VL having an amino acid sequence of any one of SEQ ID NO: 66-84 and 368.

22. The DLL3-binding polypeptide construct of claim 14, wherein the at least one DLL3 VHH domain is located at the amino terminus relative to the heterodimeric Fc region of the DLL3-binding polypeptide construct and / or at the carboxyl terminus relative to the anti-CD3 antibody or antigen-binding fragment of the multi-DLL3-binding polypeptide construct.

23. The DLL3-binding polypeptide construct of claim 14, wherein the DLL3-binding polypeptide construct includes a first DLL3 VHH domain specifically binding to DLL3 and a second DLL3 VHH domain specifically binding to DLL3.

24. The DLL3-binding polypeptide construct of claim 14, wherein the DLL3-binding polypeptide construct comprises a first DLL3 VHH domain specifically binding to DLL3 and a second DLL3 VHH domain specifically binding to DLL3; and (a) the first DLL3 VHH domain is the same as the second DLL3 VHH domain; or (b) the first DLL3 VHH domain is different from the second DLL3 VHH domain.

25. The DLL3-binding polypeptide construct of claim 24, wherein the first DLL3 VHH domain and the second DLL3 VHH domain bind to different or non-overlapping antigenic determinants of DLL3 and / or do not compete for binding to DLL3.

26. The DLL3-binding polypeptide construct of claim 14, wherein one or both of the first and second components comprises at least one costimulatory receptor binding region (CRBR) that binds a costimulatory receptor.

27. The DLL3-binding polypeptide construct of claim 26, wherein the at least one co-stimulatory receptor binding region (CRBR) is located at the amino terminus relative to the heterodimeric Fc region of the DLL3-binding polypeptide construct and / or at the carboxyl terminus relative to the anti-CD3 antibody or antigen-binding fragment of the DLL3-binding polypeptide construct.

28. The DLL3-binding polypeptide construct of claim 26, wherein the at least one co-stimulatory receptor binding region (CRBR) is an antibody or antigen-binding fragment thereof selected from the group consisting of: Fv, scFv, Fab, single-domain antibody (sdAb), VNAR, and VHH.

29. The DLL3-binding polypeptide construct of claim 26, wherein the at least one co-stimulatory receptor binding region (CRBR) binds to a co-stimulatory receptor selected from the following: 41BB (CD137), OX40 (CD134), CD27, glucocorticoid-induced TNFR-related protein (GITR), CD28, ICOS, CD40, B cell activating factor receptor (BAFF-R), B cell maturation antigen (BCMA), transmembrane activating factor and CAML interactor (TACI), and NKG2D.

30. The DLL3-binding polypeptide construct of claim 29, wherein at least one co-stimulatory receptor-binding region (CRBR) binds 41BB.

31. The DLL3-binding polypeptide construct of claim 14, wherein one or both of the first and second components comprises at least one inhibitory receptor binding region (IRBR) that binds an inhibitory receptor.

32. The DLL3-binding polypeptide construct of claim 31, wherein the at least one inhibitory receptor binding region (IRBR) is located at the amino terminus relative to the heterodimeric Fc region of the DLL3-binding polypeptide construct and / or at the carboxyl terminus relative to the anti-CD3 antibody or antigen-binding fragment of the DLL3-binding polypeptide construct.

33. The DLL3-binding polypeptide construct of claim 31, wherein the at least one inhibitory receptor binding region (IRBR) is an antibody or antigen-binding fragment thereof selected from the group consisting of: Fv, scFv, Fab, single-domain antibody (sdAb), VNAR, and VHH.

34. The DLL3-binding polypeptide construct of claim 31, wherein the at least one inhibitory receptor binding region (IRBR) binds to an inhibitory receptor selected from PD-1, CTLA-4, TIGIT, VISTA, and TIM3.

35. The DLL3-binding polypeptide construct of claim 14, wherein the linker is a non-cleavable linker.

36. The DLL3-binding polypeptide construct of claim 35, wherein the non-cleavable linker comprises GS, GGS, GGGGS (SEQ ID NO: 125), GGGGGS (SEQ ID NO: 126) or a combination thereof.

37. The DLL3-binding polypeptide construct of claim 35, wherein the non-cleavable linker is a sequence of GGGGGSGGGGGSGGGGS (SEQ ID NO: 127) or contains a sequence of GGGGGSGGGGSGGGGS (SEQ ID NO: 127).

38. A separated single-domain antibody binding to DLL3, comprising: complementarity-determining region 1 (CDR1), complementarity-determining region 2 (CDR2), and complementarity-determining region 3 (CDR3), wherein CDR1, CDR2, and CDR3 each comprise the following amino acid sequences: i. SEQ ID NO: 319, 336, and 354; ii. SEQ ID NO: 319, 337, and 354; iii. SEQ ID NO: 319, 338, and 354; iv. SEQ ID NO: 320, 338, and 354; v. SEQ ID NO: 321, 338, and 354; vi. SEQ ID NO: 322, 338, and 354; vii. SEQ ID NO: 323, 338, and 354; viiii. SEQ ID NO: 324, 338, and 354; ix. SEQ ID NO: 325, 338, and 354; or x. SEQ ID NO: 325, 338, and 354; NO: 326, 338 and 354.

39. A polynucleotide encoding a DLL3-binding polypeptide construct as claimed in any one of claims 1 to 37.

40. A polynucleotide encoding a single-domain antibody as claimed in claim 38.

41. A carrier comprising a polynucleotide as claimed in claim 39 or 40.

42. A cell comprising a polynucleotide as claimed in claim 39 or 40 or a carrier as claimed in claim 41.

43. An engineered immune cell comprising a chimeric antigen receptor, the chimeric antigen receptor comprising: an extracellular domain comprising a single-domain antibody as claimed in claim 38; a transmembrane domain; and an intracellular signal transduction domain.

44. A pharmaceutical composition comprising a DLL3-binding polypeptide construct as claimed in any one of claims 1 to 37, a single-domain antibody as claimed in claim 38, or engineered immune cells as claimed in claim 43, and a pharmaceutically acceptable carrier.

45. Use of a DLL3-binding polypeptide construct as claimed in any one of claims 1 to 37, a single-domain antibody as claimed in claim 38, an engineered immune cell as claimed in claim 43, or a pharmaceutical composition as claimed in claim 44, for the purpose of preparing a pharmaceutical product that stimulates or induces an immune response in an individual.

46. ​​Use of a DLL3-binding polypeptide construct as claimed in any one of claims 1 to 37, a single-domain antibody as claimed in claim 38, an engineered immune cell as claimed in claim 43, or a pharmaceutical composition as claimed in claim 44, for the purpose of preparing a pharmaceutical product for treating an individual with DLL3-positive cancer.