Anti-CD3 and anti-CD123 bispecific antibodies and their uses

Abstract

The present application provides a bispecific antibody comprising an antigen-binding portion for human CD3E and / or an antigen-binding portion for human CD123. In addition, the present application further provides medical and biological uses of said bispecific antibody.

Description

Technology Field

[0001] This application generally relates to the field of antibody drugs, and specifically, this application provides a bispecific antibody comprising an antigen-binding portion for human CD3E and / or an antigen-binding portion for human CD123 and medical and biological uses thereof.

[0002] This application claims priority to a Chinese patent application filed on February 5, 2020, with application number 202010080449.9, all of which are incorporated into this application. Background Technology

[0003] Bispecific antibodies (BsAbs) are artificial antibodies containing two distinct antigen-binding sites. Bispecific antibodies are widely used in the biomedical field, particularly in oncology immunotherapy. One arm of a CD3-targeting bispecific antibody can bind to the CD3E subunit of the TCR receptor complex on the surface of T cells, while the other arm targets tumor antigens. Through this mechanism, bispecific antibodies can redirect T cells to specifically kill tumor cells in a manner independent of the major histocompatibility complex (MHC).

[0004] Bispecific antibody platforms are numerous and structurally complex. Based on their structure, they can be broadly classified into two types: antibodies containing an Fc fragment and antibodies without an Fc fragment. Bispecific antibodies without an Fc fragment consist of the VH and VL regions of two antibodies or are composed of a Fab fragment; representative examples of these bispecific antibodies include BiTE, DART, TandAbs, and bi-nanobodies. The advantage of these bispecific antibodies is the absence of light-heavy chain mismatch, while the disadvantage is that their short half-life makes clinical application inconvenient. Bispecific antibodies containing an Fc fragment maintain the structure of conventional monoclonal antibodies and can mediate the biological function of the Fc fragment. Representative examples of such bispecific antibodies include KIH IgG, crossmab, DVD-Ig, and Triomab, which have long half-lives in the body and can possess ADCC and CDC activities (Hongyan Liu, Abhishek Saxena, Sachdev S. Sidhu, et al. Fc engineering for Developing Therapeutic Bispecific Antibodies and Novel Scaffolds. Front. Immunol. 2017;8:38). The problem to be solved

[0005] Therefore, considering the broad applicability of bispecific antibodies in the industry, there is a need to develop new bispecific antibodies. means of solving the problem

[0006] According to the first aspect, the present application provides a bispecific antibody comprising an antigen-binding portion for human CD3E, wherein the antigen-binding portion for human CD3E is,

[0007] HCDR1 (heavy chain CDR1) indicated by sequence number 1,

[0008] HCDR2 (heavy chain CDR2) indicated by sequence number 2,

[0009] HCDR3 (heavy chain CDR3) indicated by sequence number 3,

[0010] LCDR1 (light chain CDR1) indicated by sequence number 4,

[0011] LCDR2 (light chain CDR2) indicated by sequence number 5, and

[0012] Includes LCDR3 (light chain CDR3) indicated by sequence number 6;

[0013] Here, HCDR and LCDR are defined according to Kabat.

[0014] According to a second aspect, the present application provides a bispecific antibody comprising an antigen-binding portion for human CD123, wherein the antigen-binding portion for human CD123 is,

[0015] HCDR1 (heavy chain CDR1) indicated by sequence number 7,

[0016] HCDR2 (heavy chain CDR2) indicated by sequence number 8,

[0017] HCDR3 (heavy chain CDR3) indicated by sequence number 9,

[0018] LCDR1 (light chain CDR1) indicated by sequence number 4,

[0019] LCDR2 (light chain CDR2) indicated by sequence number 5, and

[0020] Includes LCDR3 (light chain CDR3) indicated by sequence number 6;

[0021] Here, HCDR and LCDR are defined according to Kabat.

[0022] According to a third aspect, the present application provides a bispecific antibody comprising an antigen-binding portion for human CD3E and an antigen-binding portion for human CD123.

[0023] In a partial implementation method of the third aspect, the antigen binding portion for human CD3E is,

[0024] HCDR1, designated as sequence number 1,

[0025] HCDR2 designated by sequence number 2,

[0026] HCDR3, designated as sequence number 3,

[0027] LCDR1, designated as sequence number 4

[0028] LCDR2 indicated by sequence number 5, and

[0029] Includes LCDR3 indicated by sequence number 6;

[0030] Here, HCDR and LCDR are defined according to Kabat.

[0031] In a partial implementation method of the third aspect, the antigen binding portion for human CD123 is,

[0032] HCDR1, designated by sequence number 7,

[0033] HCDR2, designated by sequence number 8,

[0034] HCDR3, designated by sequence number 9,

[0035] LCDR1, designated as sequence number 4

[0036] LCDR2 indicated by sequence number 5, and

[0037] Includes LCDR3 indicated by sequence number 6;

[0038] Here, HCDR and LCDR are defined according to Kabat.

[0039] In a partial implementation of the third aspect, the antigen binding portion for human CD3E and the antigen binding portion for human CD123 include the same light chain variable region.

[0040] In a partial implementation of the third aspect, the dual-specificity antibody is an IgG1 antibody comprising two heavy chain constant regions having the same hinge region, and the amino acid sequence of the hinge region is denoted by SEQ ID NO. 10.

[0041] In a partial implementation of the third aspect, the bispecific antibody is an IgG1 antibody comprising a first heavy chain constant region and a second heavy chain constant region, wherein the 354th and 366th amino acids of the first heavy chain constant region are C and W, respectively, and the 349th, 366th, 368th and 407th amino acids of the second heavy chain constant region are C, S, A, and V, respectively; and the amino acid positions of the antibody constant region are determined according to EU numbering.

[0042] In a partial implementation of the third aspect, the bispecific antibody is an IgG1 antibody comprising a first heavy chain constant region and a second heavy chain constant region, wherein the 234th, 235th, and 331st amino acids of the first heavy chain constant region and the second heavy chain constant region are F, E, and S, respectively; and the amino acid positions of the antibody constant region are determined according to EU numbering.

[0043] In a partial implementation of the first aspect and the third aspect, the antigen binding portion for human CD3E includes a heavy chain variable region represented by SEQ ID NO. 11 and a light chain variable region represented by SEQ ID NO. 12.

[0044] In a partial implementation of the second and third aspects, the antigen binding portion for human CD123 includes a heavy chain variable region represented by SEQ ID NO. 13 and a light chain variable region represented by SEQ ID NO. 12.

[0045] In a partial implementation of any one of the above aspects, the antigen binding portion for human CD3E comprises a short-chain antibody (scfv) or a Fab fragment.

[0046] In a partial implementation of any one of the above aspects, the antigen binding portion for human CD123 comprises a short-chain antibody (scfv) or a Fab fragment.

[0047] In a partial implementation of a third aspect, the antibody has a first cancer and a second cancer, the first cancer includes an antigen binding portion for human CD3E, and the second cancer includes an antigen binding portion for human CD123.

[0048] The first arm comprises a heavy chain variable region amino acid sequence represented by SEQ ID NO. 11, a heavy chain constant region amino acid sequence represented by SEQ ID NO. 31, a light chain variable region amino acid sequence represented by SEQ ID NO. 12, and a light chain constant region amino acid sequence represented by SEQ ID NO. 32;

[0049] The second arm above includes a heavy chain variable region amino acid sequence represented by SEQ ID NO. 13, a heavy chain constant region amino acid sequence represented by SEQ ID NO. 30, a light chain variable region amino acid sequence represented by SEQ ID NO. 12, and a light chain constant region amino acid sequence represented by SEQ ID NO. 32.

[0050] According to the fourth aspect, the present application provides a pharmaceutical composition comprising a dual-specific antibody according to any one of the first to third aspects.

[0051] In a partial embodiment of the fourth aspect, the pharmaceutical composition is used to prevent or treat CD123-positive tumors.

[0052] According to the fifth aspect, the present application provides the use of a bispecific antibody according to any one of the first to third aspects, or a pharmaceutical composition according to the fourth aspect, in the manufacture of a drug for preventing or treating CD123-positive tumors.

[0053] According to the sixth aspect, the present application provides a method for preventing or treating a CD123-positive tumor, comprising the step of administering a dual-specific antibody according to any one of the first to third aspects, or a pharmaceutical composition according to the fourth aspect, to an individual in need thereof. Brief explanation of the drawing

[0054] Figure 1 shows the results of analyzing the specific binding of a recombinant anti-CD123 monoclonal antibody to CD123 on the cell surface using a flow cytometer. Figure 2 shows the results of analyzing the binding of the H7A3 humanized mutant H7A3-h2-m5+L27E5 to CD123 on the surface of MV-4-11 cells using a flow cytometer. Figure 3 shows the results of analyzing the binding of the dual-specific antibody CD3EХCD123 to the two antigens CD3E and CD123 using ELISA. Figure 4 shows the results of analyzing the binding ability of the dual-specific antibody CD3EХCD123 to CD3E on the surface of Jurkat-Dual cells using a flow cytometer. Figure 5 shows the results of analyzing the binding ability of the dual-specific antibody CD3EХCD123 to CD123 on the surface of MV-4-11 cells using a flow cytometer. Figure 6 shows the results of Jurkat-Dual activation by the bispecific antibody CD3XCD123 in the presence of CD123-positive tumor cells, where part A is the result of Jurkat-Dual activation by the bispecific antibody CD3XCD123, the anti-CD3E monoclonal antibody H3B8+L27E5, the anti-CD123 monoclonal antibody H7A3-h2-m5+L27E5, and the combination group of the anti-CD3E monoclonal antibody H3B8+L27E5 and the anti-CD123 monoclonal antibody H7A3-h2-m5+L27E5; Part B is the result of Jurkat-Dual activation by the dual-specific antibody CD3XCD123, the dual-specific antibody control sample Xmab14045, and HIgG in the presence of CD123-positive target cells, and the result of Jurkat-Dual activation by the dual-specific antibody CD3XCD123 and the dual-specific antibody control sample Xmab14045 in the presence of CD123-negative target cells. Figure 7 shows the killing results of purified T cells mediated by the dual-specific antibody CD3EХCD123 against CD123-positive tumor cells. Figure 8 shows the results of detecting specific activation of T cells and upregulation of CD69 expression by the dual-specific antibody CD3EXCD123 in the presence of CD123-positive tumor cells using a flow cytometer. Figure 9 shows the results of detecting the promotion of T cell proliferation by the bispecific antibody CD3EXCD123 in the presence of CD123-positive tumor cells using a flow cytometer, where part A is the result of the promotion of T cell proliferation by the bispecific antibody CD3XCD123, anti-CD3E monoclonal antibody H3B8+L27E5, anti-CD123 monoclonal antibody H7A3-h2-m5+L27E5, and the combination group of anti-CD3E monoclonal antibody H3B8+L27E5 and anti-CD123 monoclonal antibody H7A3-h2-m5+L27E5 in the presence of CD123-positive tumor cells; part B is the result of the promotion of T cell proliferation by the bispecific antibody CD3XCD123 and the bispecific antibody control sample Xmab14045 in the presence of CD123-positive tumor cells. Figure 10 shows hCD34 treated with the bispecific antibody CD3EXCD123. + This shows the change in tumor volume of a humanized MV-4-11 cell tumor model mouse. Explanation of the hierarchy Sequence No. 1 is the amino acid sequence of HCDR1 of the heavy chain variable region H3B8 of the anti-human CD3E monoclonal antibody H3B8+L27E5. Sequence No. 2 is the amino acid sequence of HCDR2 of the heavy chain variable region H3B8 of the anti-human CD3E monoclonal antibody H3B8+L27E5. Sequence No. 3 is the amino acid sequence of HCDR3 of the heavy chain variable region H3B8 of the anti-human CD3E monoclonal antibody H3B8+L27E5. Sequence No. 4 is the amino acid sequence of LCDR1 of the light chain variable region L27E5. Sequence No. 5 is the amino acid sequence of LCDR2 of the light chain variable region L27E5. Sequence No. 6 is the amino acid sequence of LCDR3 in the light chain variable region L27E5. Sequence No. 7 is the amino acid sequence of HCDR1 of the heavy chain variable region H7A3-h2-m5 of the anti-human CD123 monoclonal antibody H7A3-h2-m5+L27E5. Sequence No. 8 is the amino acid sequence of HCDR2 of the heavy chain variable region H7A3-h2-m5 of the anti-human CD123 monoclonal antibody H7A3-h2-m5+L27E5. Sequence No. 9 is the amino acid sequence of HCDR3 of the heavy chain variable region H7A3-h2-m5 of the anti-human CD123 monoclonal antibody H7A3-h2-m5+L27E5. Sequence No. 10 is the amino acid sequence of the hinge region. Sequence No. 11 is the amino acid sequence of H3B8, a humanized heavy chain mutant of the rat monoclonal antibody WM03-C6. Sequence No. 12 is the amino acid sequence of L27E5, a humanized light chain mutant of the rat monoclonal antibody WM03-C6. Sequence No. 13 is the amino acid sequence of the humanized version H7A3-h2-m5. Sequence No. 14 is the amino acid sequence of the human (Homo sapiens) CD3E extracellular domain (hCD3E). Sequence No. 15 is the amino acid sequence of the extracellular domain of human (Homo sapiens) CD3D (hCD3D). Sequence No. 16 is the amino acid sequence of the CD3E extracellular domain (mfCD3E) of the Philippine monkey (Macaca fascicularis). Sequence No. 17 is the amino acid sequence of the extracellular region of CD3D (mfCD3D) of the Philippine monkey (Macaca fascicularis). Sequence No. 18 is the amino acid sequence of the extracellular domain (mCD3E) of mouse (Mus musculus) CD3E. Sequence No. 19 is the amino acid sequence of the extracellular region (mCD3D) of mouse (Mus musculus) CD3D. Sequence No. 20 is the amino acid sequence of the human (Homo sapiens) CD123 subtype 1 extracellular region (hCD123-SP1). Sequence No. 21 is the amino acid sequence of the extracellular domain (mfCD123-SP1) of the CD123 subtype 1 of the Philippine monkey (Macaca fascicularis). Sequence No. 22 is the amino acid sequence of the extracellular region (mCD123-SP1) of mouse (Mus musculus) CD123 subtype 1. Sequence number 23 is the amino acid sequence of His tag. Sequence No. 24 is the amino acid sequence of the Fc fragment (mFc) of a mouse (Mus musculus) IgG2a antibody. Sequence No. 25 is the amino acid sequence of FcK, a human IgG1 subtype Fc mutant of heterodimer. Sequence No. 26 is the amino acid sequence of FcH, a human IgG1 subtype Fc mutant of heterodimer. Sequence No. 27 is the amino acid sequence of the heavy chain constant region of a human (Homo sapiens) IgG1 subtype antibody. SEQ ID NO 28 is the amino acid sequence of the human IgG1 subtype antibody heavy chain constant region mutant IgG1H. SEQ ID NO 29 is the amino acid sequence of the human IgG1 subtype antibody heavy chain constant region mutant IgG1K. SEQ ID NO 30 is the amino acid sequence of the human IgG1 subtype antibody heavy chain constant region mutant IgG1m3-H. SEQ ID NO 31 is the amino acid sequence of the human IgG1 subtype antibody heavy chain constant region mutant IgG1m3-K. SEQ ID NO 32 is the amino acid sequence of the light chain constant region of the human (Homo sapiens) κ subtype. Sequence No. 33 is the amino acid sequence of the light chain constant region of the human (Homo sapiens) λ subtype. Sequence No. 34 is the amino acid sequence of the heavy chain variable region of the monoclonal antibody WM03-C6. Sequence No. 35 is the amino acid sequence of the light chain variable region of the monoclonal antibody WM03-C6. Sequence No. 36 is the amino acid sequence of the anti-human CD123 short-chain antibody S8F3. SEQ ID NO 37 is the amino acid sequence of the heavy chain variable region S8F3VH of the anti-human CD123 short-chain antibody S8F3. Sequence No. 38 is the heavy chain amino acid sequence of the anti-CD123 antibody CSL362. Sequence No. 39 is the light chain amino acid sequence of the anti-CD123 antibody CSL362. SEQ ID NO 40 is the amino acid sequence of the CD123-binding heavy chain of Xmab14045 of the IgG1m3 subtype containing a Hole mutation. SEQ ID NO 41 is the amino acid sequence of the CD3E-binding scFv structure of Xmab14045 of the IgG1m3 subtype containing the Knob mutation. Sequence number 42 is the amino acid sequence of the light chain of Xmab14045. Specific details for implementing the invention

[0055] definition

[0056] The following definitions and methods are provided to better define this application and to guide those skilled in the art in practicing this application. Unless otherwise specified, terms used in this application have the meanings generally understood by those skilled in the art. All contents of all patent literature, academic papers, and other published materials cited herein are incorporated and incorporated into this application.

[0057] When describing antibody structures in this specification, descriptions related to amino acid position numbering refer to the definition of EU numbering for human IgG1 antibodies, which is known to those skilled in the art and readily available to them. Additionally, when describing mutations in conjunction with EU numbering positions in this specification, it refers to mutations occurring in the natural antibody sequence.

[0058] As used herein, the terms “Fc fragment,” “Fc domain,” “Fc portion,” or similar terms refer to a part of the antibody heavy chain constant region and include the hinge region and the CH2 and CH3 fragments of the constant region. Referring to the EU numbering definition of human IgG1 antibody, the Fc fragment is the 216th–447th amino acid sequence in the antibody constant region.

[0059] As used herein, the terms “Fab (fragment antigen binding) fragment,” “Fab portion,” or similar terms refer to an antibody fragment capable of binding to an antigen generated after treating a complete antibody with papain, and include a complete light chain (VL-CL), a variable region of the heavy chain, and a CH1 fragment (VH-CH1).

[0060] As used herein, the term "single chain fragment variable (scfv)" generally refers to an antibody having a single chain structure constructed by genetic engineering techniques and comprises a single polypeptide chain of a heavy chain variable region (VH) and a light chain variable region (VL). A flexible linker is typically designed between the heavy chain variable region and the light chain variable region so that the heavy chain variable region and the light chain variable region can be folded into a precise shape capable of binding to an antigen.

[0061] As used herein, the term "antigen binding site" refers to the portion of an antibody structure that determines antigen binding ability. Those skilled in the art will understand that since the CDR is the primary portion of an antibody structure that determines antigen binding ability, the CDR is also a core component of the antigen binding site. In the construction of bispecific antibodies, examples of the "antigen binding site" include, but are not limited to, short-chain antibodies (scfv) or Fab fragments.

[0062] As used herein, the term “bispecific antibody” means an antibody having the ability to bind to two different antigens, which may consist of two Fc fragments and two antigen-binding portions fused to each of them.

[0063] In some embodiments, the term "bispecific antibody" in this specification refers to a bispecific antibody based on a human IgG1 antibody and has the basic characteristics and functions of a human IgG1 antibody, except for the modified structure described herein. It is known to those skilled in the art that the term "bispecific antibody" in this specification may be based on other immunoglobulin subtypes, such as human IgG2 antibodies.

[0064] It is known to those skilled in the art that the complementarity determining region (CDR, typically CDR1, CDR2, and CDR3) is the region that has the greatest influence on the affinity and specificity of the antibody in the variable region. There are two general methods for the CDR sequences of VH or VL (i.e., Kabat definition and Chothia definition) (see, e.g., Kabat, "Sequences of Proteins of Immunological Interest," National Institutes of Health, Bethesda, Md. (1991); A1-Lazikani et al., J. Mol. Biol. 273:927-948 (1997); and Martin et al., Proc. Natl. Acad. Sci.USA86:9268-9272 (1989)). For a given variable region sequence of an antibody, the sequence of the CDR region among the VH and VL sequences can be determined according to the Kabat definition or the Chothia definition. In the implementation method of the present application, the CDR sequence is defined using Kabat.

[0065] For a given variable region sequence of an antibody, the sequence of the CDR region within the variable region sequence can be analyzed through various methods, for example, using the online software Abysis (http: / / www.abysis.org / ).

[0066] The term “specific binding” as used in this specification refers to a non-random binding reaction between two molecules, such as the binding of an antibody to an antigen epitope.

[0067] The CD3 molecule is an important differentiation antigen of the T cell membrane and a characteristic marker of mature T cells. CD3 consists of a four-chain of γ, δ, ε, and ζ or a five-chain of γ, δ, ε, ζ, and η (ζ and η are homoisomers), and is expressed on the T cell membrane as three dimers: CD3γε, CD3δε, and CD3ζζ (or CD3ζη). The three chains of CD3γ, δ, and ε contain highly conserved acidic amino acid residues (γ is glutamic acid, and δ and ε are aspartic acid) and can form a stable TCR-CD3 complex structure by non-covalently linking to basic amino acid residues on the α and β chains of the T cell receptor (TCR) via salt bridges. This complex can transmit T cell activation signals and stabilize the TCR structure. The intracellular domain of each CD3 chain contains the ITAM (Immunoreceptor Tyrosine-Based Activation Motive) structure, which serves as the basis for the CD3 molecule mediating intracellular signal transduction. After the TCR specifically recognizes and binds to an antigen (antigen peptide presented by MHC molecules), an intracellular tyrosine protein kinase phosphorylates the tyrosine residues of the ITAM and recruits a tyrosine protein kinase (ZAP-70) containing an SH2 domain to transmit the signal into the cytoplasm of the T cell, thereby initiating the intracellular activation mechanism. Thus, CD3 functions to transmit the activation signal generated by the recognition of the antigen by the TCR and serves as the first signal inducing T cell activation.

[0068] CD123, also known as the human interleukin-3 (IL-3) receptor α chain, is a member of the cytokine receptor superfamily with a molecular weight of approximately 40 kDa and is a type I transmembrane glycoprotein. The interleukin-3 receptor is a heterodimer formed by the α chain (CD123) and the β chain (CD131). After IL-3 binds to CD123, CD131 provides signal transduction to regulate the functions of hematopoietic and immune cells and stimulate endothelial cell proliferation (Testa et al, Biomark Res.2:4(2014)).

[0069] CD123 is primarily expressed in myeloid progenitor cells, plasmacytic dendritic cells, monocytes, basophils, and a small subgroup of B cells (Munoz L et al, Haematologica. 86(12):1261-9). Approximately 80% of AML patients have blast cells that overexpress CD123, and studies have shown that overexpression of the CD123 antigen corresponds to a poor prognosis and low remission rate in AML (Testa U et al, Blood. 2002;100(8)). While most AML patients respond well to initial treatment, some patients (60-80%) require intensified therapy to achieve complete remission.

[0070] According to the first aspect, the present application provides a bispecific antibody comprising an antigen-binding portion for human CD3E, wherein the antigen-binding portion for human CD3E is,

[0071] HCDR1, designated as sequence number 1,

[0072] HCDR2 designated by sequence number 2,

[0073] HCDR3, designated as sequence number 3,

[0074] LCDR1, designated as sequence number 4

[0075] LCDR2 indicated by sequence number 5, and

[0076] Includes LCDR3 indicated by sequence number 6;

[0077] Here, HCDR and LCDR are defined according to Kabat.

[0078] According to a second aspect, the present application provides a bispecific antibody comprising an antigen-binding portion for human CD123, wherein the antigen-binding portion for human CD123 is,

[0079] HCDR1, designated by sequence number 7,

[0080] HCDR2, designated by sequence number 8,

[0081] HCDR3, designated by sequence number 9,

[0082] LCDR1, designated as sequence number 4

[0083] LCDR2 indicated by sequence number 5, and

[0084] Includes LCDR3 indicated by sequence number 6;

[0085] Here, HCDR and LCDR are defined according to Kabat.

[0086] According to a third aspect, the present application provides a bispecific antibody comprising an antigen-binding portion for human CD3E and an antigen-binding portion for human CD123.

[0087] In a partial implementation method of the third aspect, the antigen binding portion for human CD3E is,

[0088] HCDR1, designated as sequence number 1,

[0089] HCDR2 designated by sequence number 2,

[0090] HCDR3, designated as sequence number 3,

[0091] LCDR1, designated as sequence number 4

[0092] LCDR2 indicated by sequence number 5, and

[0093] Includes LCDR3 indicated by sequence number 6;

[0094] Here, HCDR and LCDR are defined according to Kabat.

[0095] In a partial implementation of a third aspect, the antigen binding portion for human CD123 is,

[0096] HCDR1, designated by sequence number 7,

[0097] HCDR2, designated by sequence number 8,

[0098] HCDR3, designated by sequence number 9,

[0099] LCDR1, designated as sequence number 4

[0100] LCDR2 indicated by sequence number 5, and

[0101] Includes LCDR3 indicated by sequence number 6;

[0102] Here, HCDR and LCDR are defined according to Kabat.

[0103] In a partial implementation of the third aspect, the antigen binding portion for human CD3E and the antigen binding portion for human CD123 include the same light chain variable region.

[0104] In a specific embodiment of a third aspect, the antigen binding portion for human CD3E and the antigen binding portion for human CD123 comprise the same light chain, and since such embodiment is advantageous for the accurate assembly of the light chain and heavy chain, it is an equally desirable embodiment.

[0105] In a partial embodiment of the third aspect, the bispecific antibody is an IgG1 antibody comprising two heavy chain constant regions having the same hinge region, the amino acid sequence of the hinge region is denoted by SEQ ID NO. 10, which replaces the 216th-230th sequence of the natural human IgG1 antibody constant region, and the amino acid position of the antibody constant region is determined according to EU numbering.

[0106] In a partial embodiment of the third aspect, the dual-specificity antibody is an IgG1 antibody comprising a first heavy chain constant region and a second heavy chain constant region, wherein the 354th and 366th amino acids of the first heavy chain constant region are C and W, respectively, and the 349th, 366th, 368th and 407th amino acids of the second heavy chain constant region are C, S, A, and V, respectively; and the amino acid positions of the antibody constant region are determined according to EU numbering.

[0107] When constructing a bispecific antibody that retains the antibody Fc domain, the structure of the bispecific antibody can be optimized from two perspectives: one is the heteromerization of the heavy chain, and the other is the precise assembly of the light and heavy chains. In some implementations, the two Fc fragments contain mutations that enable heavy chain heteromerization. The knob-in-hole (KIH) technique is a strategy to resolve heavy chain heteromerization. Generally, the KIH technique is a method that modifies the amino acid sequence of the CH3 region to form a structure favorable for pairing with heterogeneous anti-antibodies, thereby allowing the construction of a bispecific antibody while maintaining the structure of the normal antibody as much as possible. In some implementations, the KIH technique used includes a technique in which the 354th and 366th amino acids of one Fc fragment are C and W, respectively, and the 349th, 366th, 368th, and 407th amino acids of the other Fc fragment are C, S, A, and V, respectively. For guidance on KIH technology, refer to, for example, "An efficient route to human bispecific IgG", A. Margaret Merchant et al., Nature Biotechnology, Volume 16, 1998, the entirety of which is incorporated and incorporated into this application.

[0108] In a partial implementation of the third aspect, the bispecific antibody is an IgG1 antibody comprising a first heavy chain constant region and a second heavy chain constant region, wherein the 234th, 235th, and 331st amino acids of the first heavy chain constant region and the second heavy chain constant region are F, E, and S, respectively; and the amino acid positions of the antibody constant region are determined according to EU numbering.

[0109] In a partial embodiment of the third aspect, the 234th, 235th, and 331st amino acids of the CH2 fragment of the two heavy chain constant regions are F, E, and S, respectively, which can reduce antibody-dependent cytotoxicity (ADCC) mediated by the Fc fragment, thereby reducing adverse effects caused by bispecific antibodies in vivo. For guidance regarding the above mutation, refer, for example, to “The binding affinity of human IgG for its high affinity Fc receptor is determined by multiple amino acids in the CH2 domain and is modulated by the hinge region”, Stephen M. Canfield et al., J. Exp. Med. Volume 173, 1991, the entirety of which is incorporated and incorporated into this application.

[0110] In a partial implementation of the first aspect and the third aspect, the antigen binding portion for human CD3E includes a heavy chain variable region represented by SEQ ID NO. 11 (including HCDR1 represented by SEQ ID NO. 1, HCDR2 represented by SEQ ID NO. 2, and HCDR3 represented by SEQ ID NO. 3) and a light chain variable region represented by SEQ ID NO. 12 (including LCDR1 represented by SEQ ID NO. 4, LCDR2 represented by SEQ ID NO. 5, and LCDR3 represented by SEQ ID NO. 6).

[0111] In a partial implementation of the second aspect and the third aspect, the antigen binding portion for human CD123 includes a heavy chain variable region represented by SEQ ID NO. 13 (including HCDR1 represented by SEQ ID NO. 7, HCDR2 represented by SEQ ID NO. 8, and HCDR3 represented by SEQ ID NO. 9) and a light chain variable region represented by SEQ ID NO. 12 (including LCDR1 represented by SEQ ID NO. 4, LCDR2 represented by SEQ ID NO. 5, and LCDR3 represented by SEQ ID NO. 6).

[0112] In a partial implementation of any one of the above aspects, the antigen binding portion for human CD3E comprises a short-chain antibody (scfv) or a Fab fragment.

[0113] In a partial implementation of any one of the above aspects, the antigen binding portion for human CD123 comprises a short-chain antibody (scfv) or a Fab fragment.

[0114] Bispecific antibodies have two different antigen-binding sites for two different antigens, and since the antigen-binding sites can include two forms of short-chain antibodies (scfv) or Fab fragments, for a given two antigens, the composition of the antigen-binding sites of the bispecific antibody can have four combinations: Fab+Fab, Fab+scfv, scfv+Fab, and scfv+scfv.

[0115] In a specific embodiment of any one of the above aspects, the antigen binding portion for human CD3E comprises a Fab fragment, and the antigen binding portion for human CD123 comprises a Fab fragment.

[0116] In a specific embodiment of any one of the above aspects, the antigen binding portion for human CD3E comprises a Fab fragment, and the antigen binding portion for human CD123 comprises a short-chain antibody (scfv).

[0117] In some specific implementation of any one of the above aspects, the antigen binding portion for human CD3E comprises a short-chain antibody (scfv), and the antigen binding portion for human CD123 comprises a Fab fragment.

[0118] In a specific embodiment of any one of the above aspects, the antigen binding portion for human CD3E comprises a short-chain antibody (scfv), and the antigen binding portion for human CD123 comprises a short-chain antibody (scfv).

[0119] In this specification, a bispecific antibody is also described as having two "arms," ​​and the bispecific antibody may be divided into two arms with a middle boundary. The arms of the bispecific antibody may consist of an Fc fragment and an antigen-binding portion (a Fab fragment or a short-chain antibody). In the case of an arm composed of an Fc fragment and a Fab fragment, since the structure is similar to that of a general antibody and includes a complete heavy chain and a light chain, the structure of such an arm may be represented as Fc+Fab or as a heavy chain (a variable region of the heavy chain and a CH1 fragment in Fc+Fab) + a light chain (a portion of the light chain in Fab). When both arms include an antigen-binding portion formed by the Fab fragment, the structure of the bispecific antibody thus formed is close to that of a natural antibody, so this is a preferred embodiment.

[0120] In a partial embodiment of a third aspect, the antibody has a first cancer and a second cancer, wherein the first cancer comprises an antigen-binding portion for human CD3E and the second cancer comprises an antigen-binding portion for human CD123.

[0121] The first arm comprises a heavy chain variable region amino acid sequence represented by SEQ ID NO. 11, a heavy chain constant region amino acid sequence represented by SEQ ID NO. 31, a light chain variable region amino acid sequence represented by SEQ ID NO. 12, and a light chain constant region amino acid sequence represented by SEQ ID NO. 32;

[0122] The second arm above includes a heavy chain variable region amino acid sequence represented by SEQ ID NO. 13, a heavy chain constant region amino acid sequence represented by SEQ ID NO. 30, a light chain variable region amino acid sequence represented by SEQ ID NO. 12, and a light chain constant region amino acid sequence represented by SEQ ID NO. 32.

[0123] In some implementations of any one of the above aspects, the heavy chain constant region of the bispecific antibody is a human IgG1 subtype such as IgG1H, IgG1K, IgG1m3-H, or IgG1m3-K, or various mutants of a selected human IgG1 subtype.

[0124] In a partial implementation of any one of the above aspects, the light chain constant region of the bispecific antibody is a human κ subtype or a human λ subtype, preferably a human κ subtype.

[0125] According to the fourth aspect, the present application provides a pharmaceutical composition comprising a dual-specific antibody according to any one of the first to third aspects.

[0126] In some implementations, the pharmaceutical composition further comprises pharmaceutically acceptable carriers, excipients, diluents, etc.

[0127] In some embodiments, the pharmaceutical composition is used to prevent or treat CD123-positive tumors such as acute myeloid leukemia (AML) and blastocystoid dendritic cell tumors (BPDCN).

[0128] In some embodiments, the pharmaceutical composition may further include lubricants such as talc powder, magnesium stearate, and mineral oil; wetting agents; emulsifiers; suspending agents; preservatives such as benzoic acid, sorbic acid, and calcium propionate; sweeteners and / or flavoring agents, etc.

[0129] In some embodiments, the pharmaceutical composition of the present application may be formulated into the form of tablets, pills, powders, lozenges, elixirs, suspensions, emulsions, solutions, syrups, suppositories, or capsules.

[0130] In some embodiments, the pharmaceutical composition of the present application may be delivered using any physiologically acceptable method of administration, and such method of administration includes, but is not limited to, oral administration, parenteral administration, nasal administration, rectal administration, intraperitoneal administration, intravascular injection, subcutaneous administration, transdermal administration, inhalation administration, etc.

[0131] In some implementations, a pharmaceutical composition for therapeutic use can be prepared and stored in the form of a lyophilized preparation or an aqueous solution by mixing a reagent of desired purity with a pharmaceutically acceptable carrier, excipient, etc., depending on the case.

[0132] According to the fifth aspect, the present application provides the use of a bispecific antibody according to any one of the first to third aspects, or a pharmaceutical composition according to the fourth aspect, in the manufacture of a drug for preventing or treating CD123-positive tumors.

[0133] In a partial implementation of the fifth aspect, the CD123-positive tumor is selected from acute myeloid leukemia (AML) and blastocystoid dendritic tumor (BPDCN).

[0134] According to the sixth aspect, the present application provides a method for the prevention or treatment of a CD123-positive tumor comprising the step of administering a dual-specific antibody according to any one of the first to third aspects, or a pharmaceutical composition according to the fourth aspect, to an individual in need thereof.

[0135] In a partial implementation of the sixth aspect, the CD123-positive tumor is selected from acute myeloid leukemia (AML) and blastocystoid dendritic tumor (BPDCN).

[0136] It should be understood that the above detailed description is intended only to enable those skilled in the art to more clearly understand the contents of this application and is not intended to limit it in any aspect. Those skilled in the art may make various modifications and variations to the above-mentioned implementation.

[0137] The following examples are for illustrative purposes only and are not intended to limit the scope of the present application.

[0138] Examples

[0139] Example 1: Preparation of recombinant protein

[0140] In the process of preparing and identifying CD3EХCD123 bispecific antibodies, it is necessary to use various different recombinant proteins including human CD3E extracellular domain (hCD3E, SEQ NO. 14), human CD3D extracellular domain (hCD3D, SEQ NO. 15), monkey CD3E extracellular domain (mfCD3E, SEQ NO. 16), monkey CD3D extracellular domain (mfCD3D, SEQ NO. 17), mouse CD3E extracellular domain (mCD3E, SEQ NO. 18), mouse CD3D extracellular domain (mCD3D, SEQ NO. 19), and human CD123 subtype 1 extracellular domain (hCD123-SP1, SEQ NO. 20), monkey CD123 subtype 1 extracellular domain (mfCD123-SP1, SEQ NO. 21), and mouse CD123 subtype 1 extracellular domain (mCD123-SP1, SEQ NO. 22). Since all of these recombinant proteins have many post-translational modifications (e.g., glycosylation or disulfide bonds), using a mammalian cell expression system is more advantageous for maintaining the structure and function of the recombinant proteins. In addition, for convenience of purification, non-antibody recombinant proteins are formed by adding a His tag (SEQ No. 23) at the C-terminus or an Fc fragment of mouse antibody IgG2a (mFc, SEQ No. 24), or by forming a heterodimer Fc mutant of a human IgG1 subtype (FcK, SEQ No. 25 or FcH, SEQ No. 26) based on Knob-Into-Hole (KIH) technology. When producing a recombinant antibody, the antibody heavy chain constant region may be a human IgG1 subtype (SEQ No. 27) or various mutants of a selected human IgG1 subtype, for example, IgG1H (SEQ No. 28), IgG1K (SEQ No. 29), IgG1m3-H (SEQ No. 30) or IgG1m3-K (SEQ No. 31), and the light chain constant region is a human κ subtype (SEQ No. 32) or a human λ subtype (SEQ No. 33).

[0141] Based on the amino acid sequences of various target recombinant proteins in the Uniprot database, genes for said various recombinant proteins (including His tag, mFc, or Fc coding genes) were designed and synthesized. Using conventional molecular biology techniques, the synthesized genes for the various recombinant proteins were cloned into a suitable eukaryotic expression vector (e.g., pcDNA3.1, Inc. of Invitrogen, etc.). Then, the prepared recombinant protein expression plasmids were transfected into HEK293 cells (e.g., HEK293F, Inc. of Invitrogen) using liposomes (e.g., 293fectin, Inc. of Invitrogen, etc.) or other transfection reagents (e.g., PEI, etc.) and cultured for 3-5 days under serum-free suspension conditions. Next, the culture supernatant was obtained by means such as centrifugation. The recombinant proteins expressed by His tag fusion were purified in a single step from the culture supernatant using a metal chelate affinity chromatography column (e.g., HisTrap FF, Inc. of GE, etc.). The recombinant protein and recombinant antibody expressed by mFc fusion were purified in a single step using a Protein A / G affinity chromatography column (e.g., GE’s Mabselect SURE). Next, the recombinant protein stock buffer was replaced with PBS (pH 7.0) or another suitable buffer using a desalting column (e.g., GE’s Hitrap desaulting). If necessary, the antibody samples were filtered, sterilized, aliquoted, and stored at -20°C.

[0142] Example 2: Construction of a CD123 mouse immune library based on light chain immobilization

[0143] To construct CD3EХCD123 bispecific antibodies based on a common light chain, a specific anti-CD3E monoclonal antibody light chain variable region was selected to match the mouse heavy chain variable region that had undergone CD123 antigen in vivo affinity maturation, and a library of short-chain antibodies (scFv) for screening specific antibodies against CD123 was constructed using conventional molecular biological means.

[0144] After cross-immunizing 6-8 week old BALB / c mice with hCD123-SP1-His and mfCD123-SP1-His recombinant proteins, splenocytes were collected. Mouse splenocyte lymphocytes were isolated using a mouse lymphocyte isolate (Dakewe Biotech Co., Ltd., CAT#DKW33-R0100), and total RNA was extracted from the isolated lymphocytes using a cell total RNA extraction kit (Tiangen Biochemical Technology (Beijing) Co., Ltd., CAT#DP430). The extracted total RNA was used as a template, and the heavy chain variable region of the antibody was synthesized using a first-strand cDNA synthesis kit (Thermo scientific, CAT#K1621). Using conventional molecular biological means, a rat monoclonal antibody WM03-C6 (refer to the sequence of monoclonal antibody 20E5-F10 in Patent WO_2016_116626_A1, where the amino acid sequence of the heavy chain variable region is denoted by SEQ ID NO. 34 and the amino acid sequence of the light chain variable region is denoted by SEQ ID NO. 35) that specifically recognizes human and Philippine monkey CD3E was obtained by PCR amplification, and a mouse heavy chain variable region obtained by immunization with CD123 recombinant antigen was obtained, then a short-chain antibody (scFv) was constructed using overlap expansion PCR technology, and the gene of the prepared mouse short-chain antibody was cloned into the carrier pADSCFV-S (for the experimental technical process, refer to Example 1 of Chinese Patent Application No. 201510097117.0) to construct an scFv library. The library capacity of this antibody library is 1.2X10E8, and the accuracy is 65%.

[0145] Example 3: Screening of light chain immobilized CD123 mouse immune library

[0146] 3.1 Screening of Anti-Human CD123 Mouse Short-Chain Antibodies

[0147] Using the recombinant hCD123-SP1-his prepared in Example 1 as an antigen, and using a solid-phase screening strategy (for experimental methods, refer to Phage Display: General Experimental Guide / edited by Clarkson (T.), Lowman (HB) (USA); translated by MaLan et al., Chemical Industry Publishing Co., 2008.5), a phage library displaying the mouse short-chain antibody constructed in Example 2 was screened, and a total of three screenings were performed in the manner of binding, elution, neutralization, infection, and amplification to finally obtain a single short-chain antibody S8F3 (SEQ No. 36) that specifically binds to human CD123.

[0148] Using conventional molecular biological means, the nucleotide sequences encoding the heavy chain variable region S8F3VH (SEQ No. 37) and the light chain variable region WM03-C6VK (SEQ No. 35) of S8F3 were each cloned into a eukaryotic expression vector (e.g., pcDNA3.1 of Invitrogen, etc.) fused with nucleotide sequences encoding the human heavy chain constant region and the light chain constant region, and the whole antibody S8F3VH+C6VK was combined and expressed. At the same time, the anti-CD123 antibody CSL362 (heavy chain amino acid sequence indicated by SEQ No. 38 and light chain amino acid sequence indicated by SEQ No. 39) was prepared by reference to U.S. Patent US_2014_0178364_A1 and used as a positive control antibody for subsequent studies.

[0149] 3.2 Affinity Analysis of Recombinant Anti-CD123 Monoclonal Antibody

[0150] The affinity of the anti-CD123 antibody was measured using surface plasmon resonance technology with a Biacore X100. All related reagents and consumables, including the amino coupling kit (BR-1000-50), human antibody capture kit (BR-1008-39), CM5 chip (BR100012), and 10Х HBS-EP at pH 7.4 (BR100669), were purchased from GE healthcare. According to the kit instructions, the carboxylated CM5 chip surface was activated using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS), and an anti-human IgG(Fc) antibody (capture antibody) was diluted to 25 μg / mL with 10 mM pH 5.0 sodium acetate and injected at a flow rate of 10 μL / min to achieve a coupling amount of approximately 10,000 response units (RU). After injecting the capture antibody, 1 M ethanolamine was injected to block the unreacted group. For kinetic measurements, anti-CD123 antibodies (S8F3VH+C6VK and CSL362) were diluted to 0.5–1 μg / mL and injected at 10 μL / min to ensure that approximately 100 RU of the antibody was captured by the anti-human Fc antibody. Next, hCD123-SP1-his was set to a series of concentration gradients (e.g., 2.47 nM, 7.4 nM, 22.2 nM, 66.7 nM, 200 nM) and injected from low to high concentrations at 30 μL / min at 25 °C, with a binding time of 120 s and a dissociation time of 600–2400 s, and the chip surface was regenerated for 30 s by injecting a 3 M MgCl2 solution at 10 μL / min. Coupling velocity (K) was fitted using a 1:1 coupling model with Biacore X100 evaluation software version 2.0.1. on) and dissociation rate (K off) was calculated. Ratio K off / K on The dissociation equilibrium constant (KD) was calculated. The fitting results are shown in Table 1.

[0151] Binding affinity constant of recombinant anti-CD123 monoclonal antibody against hCD123-SP1-his K on K off KD S8F3VH+C6VK 4.546E+5 3.651E-3 8.030E-9 CSL362 1.324E+5 7.931E-5 5.991E-10

[0152] 3.3 Identification of Recombinant Anti-CD123 Monoclonal Antibody Binding to Cell Surface CD123 Antigen

[0153] KG-1a cells in the logarithmic growth phase (human acute myeloid leukemia cells, purchased from the Cell Resource Center of the Institute of Basic Medicine, Chinese Academy of Medical Sciences) were taken, centrifuged, and then 2×10⁻¹⁰ in PBS buffer containing 1% BSA 6 The sample was resuspended at 100 μL / well and distributed into a 96-well V-bottom plate. After centrifugation, the supernatant was removed. Sample S8F3VH+C6VK, control sample CSL362, and HighG-independent antibody (GenScript A01006, human IgG control (whole molecule), Purifie) to be measured were prepared with PBS at final concentrations of 5 μg / mL and 0.5 μg / mL, respectively. These were added to the wells containing the cells and incubated at 4°C for 1 hour. Next, the sample was washed three times with 200 μL of PBS, and a positive-anti-human IgG-FITC (Zhongshan Jinqiao, ZF-0308) secondary antibody (100 μL / well) was added. The sample was then incubated under dark conditions at 4°C for 30 minutes. Next, the sample was washed three times with 200 μL of PBS and resuspended in 100 μL of PBS, after which the FITC channel was detected using a flow cytometer (ACEA, Novocyte). The results showed that S8F3VH+C6VK could specifically bind to CD123-positive KG-1a cells (Fig. 1).

[0154] Example 4: Humanization and Activity Identification of WM03-C6 Monoclonal Antibody

[0155] 4.1 WM03-C6 Humanization Design

[0156] Humanization studies were performed on the rat monoclonal antibody WM03-C6 to reduce its immunogenicity. The humanization method utilized a classical framework transfer strategy (J immunol. 169, 1119-1125, 2002). The heavy chain variable region and light chain variable region of WM03-C6 were compared, respectively, with human antibody germline gene sequences from the IMGT database. Suitable germline gene sequences were selected to provide framework regions 1 to 3 (FR1+FR2+FR3) of the antibody, and suitable J region gene sequences were selected to provide framework region 4 (FR4). These templates can be selected based on various factors such as the relative total length of the antibody, the size of the CDR, amino acid residues located at the junction between the antibody's framework region (FR) and hypervariable region (CDR), and overall sequence homology. The selected templates may be a mixture of multiple sequences or a common template to maintain a suitable arrangement of the parental complementarity determining region (CDR) as much as possible. At the same time, to avoid protein heterogeneity that may occur due to the deamination site (NG) in the hypervariable region (CDR) of the antibody, the light and heavy chains of the humanized antibody were designed with mutations. Finally, the humanized light chain mutant L27E5 (SEQ No. 12) and the humanized heavy chain mutant H3B8 (SEQ No. 11) were obtained.

[0157] 4.2 Affinity Analysis of WM03-C6 Humanized Monoclonal Antibody

[0158] Referring to Example 3.2, an anti-human IgG(Fc) antibody was coupled to the surface of a CM5 chip, and an anti-CD3E antibody (WM03-C6 and H3B8+L27E5) was diluted to 0.5–1 μg / mL and injected at 10 μL / min to ensure that approximately 350–400 RU of antibody was captured by the anti-human Fc antibody. Next, hCD3E-his was set to a series of concentration gradients (e.g., 6.17 nM, 18.5 nM, 55.6 nM, 167 nM, 500 nM) and injected from low to high concentration at 30 μL / min at 25 °C, with a binding time of 120 s and a dissociation time of 600–1800 s, and a 3 M MgCl2 solution was injected at 10 μL / min to regenerate the chip surface for 30 s. Coupling velocity (K) was fitted using a 1:1 coupling model with Biacore X100 evaluation software version 2.0.1. on ) and dissociation rate (K off ) was calculated. Ratio K off / K on The dissociation equilibrium constant (KD) was calculated. The fitting results are shown in Table 2.

[0159] Binding affinity constant of WM03-C6 humanized monoclonal antibody against hCD3E-his K on K off KD WM03-C6 2.716E+4 2.879E-4 1.060E-8 H3B8+L27E5 2.084E+4 9.928E-4 4.763E-8

[0160] Example 5: In vitro affinity maturation of S8F3

[0161] To enhance the specificity of bispecific antibodies against target antigens and to improve tissue distribution and cytotoxicity during administration, in vitro affinity maturation was performed on the S8F3 heavy chain variable region. A CDR3 mutant library based on S8F3VH was constructed by introducing mutations into CDR3 of the S8F3 heavy chain variable region using conventional molecular biological methods. The designed mutation scheme is shown in Table 3, and the constructed library capacity is 1.7X10E8, with an accuracy of 86%.

[0162] Design plan for the S8F3VH-CDR3 mutant library initial amino acids Designed amino acids Consent codon L L, F, V, or I NTC R R, T, K, S, or N AVW Y Y or F TWC G G, S, D, N, I or V RDT N N, D, T, or A RMC Y Y or F TWC G G, V, A or D GNT D D, N, Y, S, T, or A DMC A A, T, N, D, I or V RHT M M, V, T, A, K or E RHG D D, N, Y, S, T, or A DMC D D, Y, F, or V KWT

[0163] Based on a dual carrier phage display system (for the experimental technical process, refer to Example 5 of Chinese Patent Application No. 201510097117.0), the S8F3VH-CDR3 mutant library constructed using the hCD123-SP1 antigen was screened and concentrated three times via a solid-phase screening method to finally obtain the heavy chain variable region mutant H7A3 with enhanced affinity. The nucleotide sequence encoding the obtained H7A3 variable region was cloned into a eukaryotic expression vector fused with the nucleotide sequence encoding the human heavy chain constant region, and combined with the L27E5 light chain expression vector to express the entire antibody.

[0164] Referring to Example 3.2, the affinity of the S8F3 heavy chain mutant H7A3 was measured using Biacore X100, and the results are as shown in Table 4.

[0165] Binding affinity constant of S8F3 heavy chain mutant for hCD123-SP1-his K on K off KD S8F3VH+L27E5 4.384E+4 1.455E-4 3.319E-9 H7A3+L27E5 2.698E+5 6.638E-5 2.46E-10

[0166] Example 6: Humanization of H7A3 and Identification of Its Activity

[0167] A classical framework transplantation strategy was used for the humanization of the anti-human CD123 mouse monoclonal antibody heavy chain H7A3. Referring to Example 4, a CDR transplantation was performed on the heavy chain variable region of H7A3 to obtain the humanized version H7A3VH-h2, and simultaneously, to ensure the arrangement and affinity of the antibody, some key amino acids (e.g., I69, R71, T73, A75) of the humanized antibody framework region were reverse-mutated to finally obtain the humanized version H7A3-h2-m5 (Sequence No. 13). The antibody variable region gene was designed and synthesized according to the amino acid sequence of the humanized antibody, cloned into a eukaryotic expression vector, and combined with the common light chain L27E5 to express the human IgG1 version whole antibody.

[0168] Referring to Example 3.2, the affinity of the humanized version of H7A3, H7A3-h2-m5, was measured using Biacore X100, and the results are as shown in Table 5.

[0169] Binding affinity constant of the H7A3 humanized version for hCD123-SP1-his K on K off KD H7A3+L27E5 3.054E+5 6.889E-5 2.256E-10 H7A3-h2-m5+L27E5 3.144E+5 7.671E-5 2.440E-10

[0170] MV-4-11 cells in the logarithmic growth phase (human acute mononuclear leukemia cells, purchased from NANJING COBIOER BIOSCIENCES CO., LTD) were taken, centrifuged, and then 2 x 10⁻¹⁰ in PBS buffer containing 1% BSA 6The samples were resuspended at 1 / mL and distributed at 100 μL / well in a 96-well V-bottom plate; the supernatant was removed after centrifugation. Sample H7A3-h2-m5+L27E5, control sample CSL362, and HIgG-independent antibodies (GenScript A01006, human IgG control (whole molecule), Purifie) to be measured were all prepared at a total of 9 concentrations by 3-fold gradient dilution starting from a final concentration of 100 nM using PBS, added to wells containing cells, and incubated at 4°C for 1 hour. Next, the samples were washed three times with 200 μL of PBS, and positive-anti-human IgG-FITC (Zhongshan Jinqiao, ZF-0308) was added at 100 μL / well, followed by incubation at 4°C under dark conditions for 30 minutes. Next, the sample was washed three times with 200 μL of PBS and resuspended in 100 μL of PBS, after which the FITC channel was detected using a flow cytometer (ACEA, Novocyte). The results are as shown in Figure 2, and H7A3-h2-m5+L27E5 can bind well to CD123-positive cells MV-4-11, with a KD value of 3.1 nM.

[0171] Example 7: Preparation of a bispecific antibody

[0172] A common light chain-based heterodimer was constructed by cloning the nucleotide sequences encoding the heavy chain variable region H3B8 of the CD3E monoclonal antibody and the heavy chain variable region H7A3-h2-m5 of the CD123 monoclonal antibody into suitable eukaryotic expression vectors, respectively. Specifically, the nucleotide sequence encoding the heavy chain variable region of the CD3E antibody was cloned into a eukaryotic expression vector fused with the nucleotide sequence encoding the Knob-mutated IgG1 constant region IgG1m3-K, the nucleotide sequence encoding the heavy chain variable region of the CD123 antibody was cloned into a eukaryotic expression vector containing the nucleotide sequence encoding the Hole-mutated IgG1 constant region IgG1m3-H, and the nucleotide sequence encoding the variable region VK of the common light chain L27E5 was cloned into a eukaryotic expression vector fused with the nucleotide sequence encoding the human light chain constant region CK. At the same time, to compare the biological activity of the candidate molecule with that of the CD3EХCD123 bispecific antibody currently under clinical study, Xmab14045 (SEQ No. 40, SEQ No. 41, and SEQ No. 42), based on the same Fc structure, was constructed by referring to patent WO2017210443.

[0173] Three constructed eukaryotic expression vectors expressing H3B8-IgG1m3-K, H7A3-h2-m5-IgG1m3-H, and L27E5-CK were co-transfected into HEK293F cells via liposomes and cultured for 3–5 days under serum-free suspension conditions, after which the culture supernatant was obtained by means such as centrifugation. The bispecific antibodies in the culture supernatant were purified using a Protein A / G affinity chromatography column (e.g., GE’s Mabselect SURE, etc.), and then the recombinant protein stock buffer was replaced with PBS (pH 7.0) or another suitable buffer using a desalting column (e.g., GE’s Hitrap Desaulting, etc.). After desalting, the protein solution was purified using size exclusion chromatography (SEC) with a Superdex200 (GE) to obtain the target protein. If necessary, antibody samples were filtered, sterilized, divided, and stored at -20°C for stockpiling.

[0174] Example 8: Analysis of the affinity of bispecific antibodies

[0175] Referring to Examples 3.2 and 4.2, the affinities of the anti-CD3E monoclonal antibody H3B8+L27E5, the anti-CD123 monoclonal antibody H7A3-h2-m5+L27E5, the CD3ХCD123 bispecific antibody (CD3ХCD123 BsAb), and the Xmab14045 control antibody were measured using surface plasmon resonance technology with Biacore X100, and the affinity fitting results are as shown in Tables 6 and 7.

[0176] Binding affinity constant of CD3XCD123 bispecific antibody against hCD123-SP1-his K on K off KD H7A3-h2-m5+L27E5 2.337E+5 5.137E-5 2.198E-10 CD3ХCD123 BsAb 2.254E+5 4.015E-5 1.781E-10 Xmab14045 9.671E+4 8.188E-5 8.467E-10

[0177] Binding affinity constant of CD3XCD123 bispecific antibody against CD3E hCD3E-his mfCD3E-his K on K off KD K on K off KD H3B8+L27E5 5.607E+4 7.108E-4 1.268E-8 1.233E+5 6.492E-4 5.263E-9 CD3ХCD123BsAb 5.646E+4 7.967E-4 1.411E-8 1.368E+5 1.571E-3 1.149E-8 Xmab14045 5.648E+5 3.674E-4 6.505E-10 8.633E+5 8.074E-4 9.352E-10

[0178] Example 9: Identification of the ability of a bispecific antibody to simultaneously recognize two antigens, CD3E and CD123

[0179] The co-binding of the CD3XCD123 bispecific antibody (CD3XCD123 BsAb) to the two antigens CD3E and CD123 was detected using a conventional ELISA method. The CD123-SP1-mFc antigen was coated onto 96-well ELISA plates (3 μg / mL, 100 μL / well) overnight at 4°C. After blocking with PBS-0.1% Tween 20-3% milk blocking solution at 37°C for 1 hour, the anti-CD123 monoclonal antibody H7A3-h2-m5+L27E5, the anti-CD3E monoclonal antibody H3B8+L27E5, and the CD3XCD123 bispecific antibody (10 μg / mL, 100 μL / well) were added, two duplicate wells were set up for each, and the plates were incubated at 37°C for 1 hour. After washing the ELISA plates with PBS-0.1%Tween20, CD3E-his antigen (1 μg / mL, 100 μL / well) was added and incubated at 37°C for 1 hour. After washing the ELISA plates with PBS-0.1%Tween20, HRP mouse anti-his IgG (Beijing ComWin Biotech Co., Ltd., cw0285M) was added and incubated at 37°C for 1 hour. After washing the ELISA plates with PBS-0.1%Tween20, OPD substrate chromogenic solution was added, and after 5-10 minutes, color development was stopped using 1 M H2SO4, and optical density values ​​were measured at 492 nm / 630 nm dual wavelengths using a microplate reader. The ELISA analysis results are as shown in Figure 3, and the CD3XCD123 bispecific antibody can simultaneously recognize two antigens, CD3E and CD123.

[0180] Example 10: Identification of the Recognition Ability of Dual-Specific Antibodies Against Cell Surface CD3E and CD123

[0181] 10.1 Identification of Recognition Ability of Bispecific Antibodies Against Cell Surface CD3E

[0182] Jurkat-Dual cells in the logarithmic growth phase (purchased from Invivogen, jktd-isnf) were taken, centrifuged, and then 2 x 10⁻³ of 1% BSA-containing PBS buffer was added. 6 The solution was resuspended at 100 μL / well and distributed into a 96-well V-bottom plate, and the supernatant was removed after centrifugation. Using PBS, the samples, the bispecific antibody control sample Xmab14045, the anti-CD3E monoclonal antibody H3B8+L27E5, the HIgG-independent antibody (GenScript A01006), the human IgG control (whole molecule, Purifie), and the CD3XCD123 bispecific antibody (CD3XCD123 BsAb) were all set to a starting concentration of 400 nM. The first three concentration points were diluted twofold, and the last five concentration points were diluted threefold, resulting in a total of eight concentrations. These were added to wells containing cells and incubated at 4°C for 1 hour. Next, the cells were washed three times with 200 μL of PBS, and anti-human IgG-FITC (Zhongshan Jinqiao, ZF-0308) was added at a rate of 100 μL / well. The cells were incubated under dark conditions at 4°C for 30 minutes. Subsequently, 200 μL of PBS was used. After washing three times and resuspending with 100 μL of PBS, FITC channels were detected using a flow cytometer (ACEA, Novocyte). The results show that the CD3ХCD123 dual-specific antibody can bind to CD3E on the surface of Jurkat-Dual cells, and the binding strength is weaker than that of the control sample Xmab14045, consistent with the affinity measurement results of Example 9 (Fig. 4).

[0183] 10.2 Identification of the Recognition Ability of Bispecific Antibodies Against Cell Surface CD123

[0184] MV-4-11 cells in the logarithmic growth phase were taken, centrifuged, and then 2 x 10⁻³ in PBS buffer containing 1% BSA 6The solution was resuspended at 100 μL / well and distributed into a 96-well V-bottom plate, and the supernatant was removed after centrifugation. The samples, the bispecific antibody control sample Xmab14045, the anti-CD123 monoclonal antibody H7A3-h2-m5+L27E5, the CD3XCD123 bispecific antibody (CD3XCD123 BsAb), the HIgG-independent antibody (GenScript A01006), and the human IgG control (whole molecule, Purifie) were all diluted in a 3-fold gradient starting from a final concentration of 100 nM to a total of 9 concentrations using PBS, added to 96-well plates containing cells, and incubated at 4°C for 1 hour. Next, the cells were washed three times with 200 μL of PBS, and anti-human IgG-FITC (Zhongshan Jinqiao, ZF-0308) was added at a rate of 100 μL / well, followed by incubation under dark conditions at 4°C for 30 minutes. Subsequently, the cells were washed three times with 200 μL of PBS, and After resuspending in 100 μL of PBS, FITC channels were detected using a flow cytometer (ACEA, Novocyte). The results showed that the CD3XCD123 bispecific antibody could bind well to CD123 on the surface of CD123-positive cells MV-4-11, with a KD value of 7.08 nM, which is comparable to Xmab14045 (KD=6.07 nM) (Fig. 5).

[0185] Example 11: Mediation of a dual-specific antibody for Jurkat-Dual cell-specific activation

[0186] MV-4-11 cells in the logarithmic growth phase (CD123 + Collect cells, centrifuge, and then transfer to 1640 aliquot medium at 2x10°C. 6 The mixture was resuspended at 1 / mL, distributed into cell plates at 50 μL / well, and the supernatant was removed after centrifugation. Jurkat-Dual cells (purchased from Invivogen) in the logarithmic growth phase were collected and plated in 1640 medium at a rate of 2×10⁻¹⁰�⁰⁰⁰⁰⁰⁰⁰⁰⁰⁰⁰⁰⁰⁰ 6The mixture was resuspended at 6 / mL and added to a cell plate at 50 μL / well to obtain a final E:T ratio of 1:1. Next, CD3XCD123 bispecific antibody (CD3XCD123 BsAb, 50 μL / well) was added at 8 concentration points with a starting concentration of 6 nM and a 4-fold gradient dilution. In the system control, the dual-specificity antibody control sample Xmab14045, the anti-CD3E monoclonal antibody H3B8+L27E5, the anti-CD123 monoclonal antibody H7A3-h2-m5+L27E5, the combination group of H3B8+L27E5 and H7A3-h2-m5+L27E5 (H3B8+L27E5 & H7A3-h2-m5+L27E5), the CD123 negative target cell (BAF3 cell) control, and the HIgG-independent antibody (GenScript A01006, human IgG control (whole molecule), Purifie) were set up, and the concentrations used were all equal to the concentration of the CD3XCD123 dual-specificity antibody. After culturing for 20 hours, the supernatant was collected, and tumor cell mediation of the specific activation of Jurkat-Dual cells under different conditions was detected and analyzed in reference to QUANTI-Luc’s guidelines (QUANTI-Luc, Invivogen, rep-qlc2) (parts A and B of Figure 6).

[0187] Example 12: Expression of purified T cell surface-activating molecules and mediation of CD3XCD123 bispecific antibodies for the killing of CD123-positive tumor cells

[0188] 12.1 Isolation of Human Peripheral Blood Mononuclear Cells (PBMCs)

[0189] Blood was collected from normal volunteers (50 mL each), and all volunteers signed an informed consent form.

[0190] The criteria for selecting applicants are as follows.

[0191] 1. 18 years of age or older;

[0192] 2. No HIV or HBV infection;

[0193] 3. Regular blood test normal;

[0194] 4. Not a pregnant or breastfeeding woman;

[0195] PBMCs were isolated from volunteer whole blood cells by Ficoll density gradient centrifugation and cultured in 1640 medium.

[0196] 12.2 Detection of bispecific antibodies for the killing of CD123-positive tumor cells by purified T cells

[0197] MV-4-11 cells (CD123+ cells) in the logarithmic growth phase were collected, centrifuged, and then placed in 1x10 oz of 1640 agar medium. 6 The mixture was resuspended at 1 / mL and distributed into cell plates at 50 μL / well. Next, CD3XCD123 bispecific antibody (CD3XCD123 BsAb, 50 μL / well) was added at 8 concentration points with a starting concentration of 1 nM and a 4-fold gradient dilution. T cells (5X10) purified from PBMCs were prepared by referring to the T cell negative classification kit guidelines (BD IMaq human T lymphocyte enrichment set-DM, BD, 557874). 6 Body / mL, 50 μL / well) was obtained by sorting, and the final effect-to-target ratio was 5:1. Simultaneously, a single target cell control (MV-4-11 cells), a single effector cell control (T), and a single medium blank control were established and supplemented with medium up to a volume of 150 μL. After incubation for 20 hours, the supernatant was taken, and cytoTox96 ® Guidelines for Non-Radioactive Cytotoxicity Detection Reagents (cytoTox96 ® The killing rate of T cells against tumor cells mediated by a CD3XCD123 bispecific antibody was detected and analyzed with reference to Non-Radioactive Cytotoxicity Assay (promega, G1780). The results show that the CD3XCD123 bispecific antibody can specifically mediate the killing of T cells against CD123-positive tumor cells and does not exhibit a killing effect against CD123-negative tumor cells (BAF3 cells) (Fig. 7).

[0198] 12.3 Identification of Surface Activation Molecule Expression After T Cells Are Specifically Activated by Bispecific Antibodies

[0199] 12.2 In the experiment, supernatant cells were removed, washed twice with PBS, and anti-human CD3-APC (ebioscience, 17-0037-42) and anti-human CD69-PE (ebioscience, 11-0069-42) flow antibodies were added and cultured for 30 minutes under dark conditions at 4°C, then washed twice with PBS, resuspended in 100 μL of PBS, and detected using a flow cytometer (ACEA, Novocyte). The difference in the expression of the activation marker CD69 among CD3-positive cell populations (MV-4-11 cells) treated with CD3-XCD123 bispecific antibody (CD3-XCD123 BsAb) was compared. The results show that in the presence of positive tumor cells, a CD3XCD123 bispecific antibody can specifically upregulate the expression of T cell surface CD69, and that CD123-negative tumor cells (BAF3 cells) cannot upregulate the expression of T cell surface CD69 (Fig. 8).

[0200] Example 13: In vitro proliferation experiment of T cells mediated by bispecific antibodies

[0201] 13.1 Isolation of Human Peripheral Blood Mononuclear Cells (PBMCs)

[0202] PBMC was obtained by classification with reference to Example 12.1.

[0203] 13.2 Experiment on T cell in vitro proliferation mediated by bispecific antibodies

[0204] T cells purified from PBMCs were obtained by sorting them according to the T cell-negative isolation kit instructions (BD IMaq human T lymphocyte enrichment set-DM, BD, 557874). The purified T cells were stained according to the CFSE staining instructions (CFSE, eBioscience, 65-0850-84) and placed in 2×10⁻¹⁰ medium at 1640. 6It was resuspended at 1 / mL and added to cell culture plates at 50 μL / well. MV-4-11 cells (CD123 in the logarithmic growth phase) + Collect cells, centrifuge, and then transfer to 1640 aliquot medium at 5x10°C. 5 The solution was resuspended at 1 / mL and distributed into a cell plate at 50 μL / well, with a final effect-to-target ratio of 4:1. Next, eight concentration points of CD3XCD123 bispecific antibody (CD3XCD123 BsAb, 50 μL / well) and Xmab14045 were added at a starting concentration of 0.25 nM and 4-fold gradient dilution. Additionally, controls were established at a concentration point of 0.25 nM, consisting of the single anti-CD3E monoclonal antibody H3B8+L27E5, the single anti-CD123 monoclonal antibody H7A3-h2-m5+L27E5, and the combination of H3B8+L27E5 and H7A3-h2-m5+L27E5 (H3B8+L27E5 & H7A3-h2-m5+L27E5). In addition, a control was established at a concentration point of 0.25 nM of the CD3XCD123 bispecific antibody in the presence of CD123-negative tumor cells (BAF3 cells). After culturing for 5 days, the cells were washed twice with PBS, incubated with the anti-human CD3-APC (ebioscience, 17-0037-42) flow antibody at 4°C under dark conditions for 30 minutes, washed twice with PBS, resuspended in 100 μL of PBS, and then T cell proliferation was detected using a flow cytometer (ACEA, Novocyte) (parts A and B of Fig. 9).

[0205] Example 14: CD34 + Identification of the activity of bispecific antibodies in a mouse model

[0206] 42 female hCD34s aged 20-24 weeks (32 grouped, 10 reserved) + Select humanized mice (purchased from Peng Li Biomedical Technology (Shanghai) Co., Ltd.) and 100 μL of 1×10 7Canine MV-4-11 cells were uniformly mixed with 100 μL of Matrigel and subcutaneously injected into the right back of mice. The mice were anesthetized with 3-4% isoflurane prior to inoculation. The tumors averaged approximately 50-80 mm. 3 When grown, 32 tumor-bearing mice had hCD45 in their peripheral blood + The animals were randomly divided into four groups of eight animals each based on the ratio and tumor volume. The day of administration was defined as Day 0. The experiment was conducted by dividing the animals into four groups: CD3XCD123 bispecific antibody 0.01 mg / kg, 0.1 mg / kg, 0.5 mg / kg, and a negative control group (IgG1m3, 0.5 mg / kg). Eight animals were placed in each group, and the antibody was administered via tail vein injection on Day 0, Day 3, Day 7, Day 14, and Day 21, for a total of five administrations. The therapeutic effect was evaluated based on the relative tumor suppression rate (TGI), and safety was assessed based on changes in animal body weight and mortality rates.

[0207] During the experiment, the mental state of the animals was generally good, and at the end of the in vivo experiment (day 24), there was no significant difference in body weight between the animals in the administered group and the negative control group (IgG1m3 IV 0.5 mg / kg dose group, G1 group) (P>0.05). The CD3XCD123 bispecific antibody significantly inhibited tumor growth at a dose of 0.5 mg / kg (G2 group), with a relative tumor inhibition rate (TGI) of 97.35%. Furthermore, the tumors in six animals disappeared almost completely, showing an extremely significant difference compared to the negative control group (G1 group) over the same period (p<0.001). After administration of the CD3-XCD123 bispecific antibody at a dose of 0.1 mg / kg to animals in the group (Group G3), tumor volume growth slowed, and the relative tumor inhibition rate (TGI) was 52.08%, showing a significant difference compared to the negative control group (Group G1) at the same time (p<0.05). G4 is the CD3-XCD123 bispecific antibody at a dose of 0.01 mg / kg. There is a significant dose-effect relationship between the CD3-XCD123 bispecific antibody and tumor growth, and the tumor growth status of animals in each group at each time point is illustrated in Figure 10. Here, when comparing with the G1 group, using one-way ANOVA / Dunnett t-test statistics, * indicates P<0.05, ** indicates P<0.01, and *** indicates P<0.001. When using repeated measures / Bonferroni corrected statistics, ### indicates P<0.001.

[0208] Although the present application has been described in detail above in general terms and specific implementations, it is obvious to those skilled in the art that modifications or improvements may be made based on the present application. Accordingly, all such modifications or improvements made without departing from the spirit of the present application fall within the scope of protection of the present application.

Claims

Claim 1 A bispecific antibody comprising an antigen-binding portion for human CD3E and an antigen-binding portion for human CD123, wherein the antigen-binding portion for human CD3E comprises HCDR1 represented by SEQ ID NO. 1, HCDR2 represented by SEQ ID NO. 2, HCDR3 represented by SEQ ID NO. 3, LCDR1 represented by SEQ ID NO. 4, LCDR2 represented by SEQ ID NO. 5, and LCDR3 represented by SEQ ID NO. 6; wherein the antigen-binding portion for human CD123 comprises HCDR1 represented by SEQ ID NO. 7, HCDR2 represented by SEQ ID NO. 8, HCDR3 represented by SEQ ID NO. 9, LCDR1 represented by SEQ ID NO. 4, LCDR2 represented by SEQ ID NO. 5, and LCDR3 represented by SEQ ID NO. 6; and wherein HCDR and LCDR are defined according to Kabat. Claim 2 In claim 1, the bispecific antibody is an IgG1 antibody comprising two heavy chain constant regions having the same hinge region, and the amino acid sequence of the hinge region is represented by SEQ ID NO.

10. Claim 3 A dual-specific antibody according to claim 1, wherein the antigen-binding portion for human CD3E comprises a heavy chain variable region represented by SEQ ID NO. 11 and a light chain variable region represented by SEQ ID NO. 12; and the antigen-binding portion for human CD123 comprises a heavy chain variable region represented by SEQ ID NO. 13 and a light chain variable region represented by SEQ ID NO.

12. Claim 4 A dual-specific antibody according to claim 1, wherein the antigen-binding portion for human CD3E and the antigen-binding portion for human CD123 comprise the same light chain. Claim 5 In claim 1, the bispecific antibody is an IgG1 antibody comprising a first heavy chain constant region and a second heavy chain constant region, wherein the 354th and 366th amino acids of the first heavy chain constant region are C and W, respectively, and the 349th, 366th, 368th and 407th amino acids of the second heavy chain constant region are C, S, A, and V, respectively; the 234th, 235th and 331st amino acids of the first heavy chain constant region and the second heavy chain constant region are F, E, and S, respectively; and the amino acid positions of the antibody constant region are determined according to EU numbering. Claim 6 A dual-specific antibody according to claim 1, wherein the antigen-binding portion for human CD3E and the antigen-binding portion for human CD123 comprise a short-chain antibody (scfv) or a Fab fragment. Claim 7 A dual-specific antibody according to claim 1, wherein the antigen-binding portion for human CD3E comprises a Fab fragment and the antigen-binding portion for human CD123 comprises a Fab fragment; wherein the antigen-binding portion for human CD3E comprises a Fab fragment and the antigen-binding portion for human CD123 comprises a short-chain antibody (scfv); wherein the antigen-binding portion for human CD3E comprises a short-chain antibody (scfv) and the antigen-binding portion for human CD123 comprises a Fab fragment; or wherein the antigen-binding portion for human CD3E comprises a short-chain antibody (scfv) and the antigen-binding portion for human CD123 comprises a short-chain antibody (scfv). Claim 8 A bispecific antibody according to claim 1, wherein the antibody has a first arm and a second arm, wherein the first arm comprises an antigen-binding portion for human CD3E and the second arm comprises an antigen-binding portion for human CD123, and the first arm comprises a heavy chain variable region amino acid sequence represented by SEQ ID NO. 11, a heavy chain constant region amino acid sequence represented by SEQ ID NO. 31, a light chain variable region amino acid sequence represented by SEQ ID NO. 12, and a light chain constant region amino acid sequence represented by SEQ ID NO. 32; and the second arm comprises a heavy chain variable region amino acid sequence represented by SEQ ID NO. 13, a heavy chain constant region amino acid sequence represented by SEQ ID NO. 30, a light chain variable region amino acid sequence represented by SEQ ID NO. 12, and a light chain constant region amino acid sequence represented by SEQ ID NO.

32. Claim 9 A pharmaceutical composition comprising a dual-specific antibody according to claim 1 and used to prevent or treat CD123-positive tumors. Claim 10 A pharmaceutical composition according to claim 9, wherein the CD123-positive tumor is selected from acute myeloid leukemia (AML) and blastocystoid dendritic tumor (BPDCN). Claim 11 delete Claim 12 delete Claim 13 delete Claim 14 delete Claim 15 delete Claim 16 delete

Images