Bispecific humanized single-domain antibodies against PD-L1 and CD47 and their use

Abstract

The present invention relates to a bispecific humanized single-domain antibody targeting both PD-L1 and CD47 and its use. Specifically, a humanized single-domain antibody that binds bispecifically to the immune checkpoint proteins PD-L1 and CD47 has been developed, and its efficacy has been confirmed both in vitro and in vivo. Therefore, the bispecific humanized single-domain antibody can be usefully used in immuno-oncology therapy as an immune checkpoint inhibitor.

Description

[Technical Field]

[0001] This invention relates to a bispecific humanized single-domain antibody against the immune checkpoint proteins PD-L1 and CD47, and to the use of the same antibody. [Background technology]

[0002] In recent years, the therapeutic effects of newly developed immunotherapy using the human immune system have been demonstrated, and there is a shift from conventional cancer treatments using chemotherapy agents and targeted therapies to immunotherapy using immunotherapy agents.

[0003] Generally, immune cells in cancer patients acquire tolerance to cancer antigens and, while recognizing cancer cells, remain functionally suppressed, making it impossible to effectively eliminate them. The core of immunotherapy is to awaken and activate these resistant immune cells to induce the destruction of cancer cells. Types of immunotherapy include cytokine therapies such as IFN-γ and IL-2, dendritic cell-based cancer vaccines, T-cell-based cell therapies, and immune checkpoint inhibitors (ICIs) that block immune checkpoint proteins. These treatments are generally called immuno-oncology therapy. Among these, immune checkpoint inhibitors are the type of cancer immunotherapy that international pharmaceutical companies are competing to develop.

[0004] Immune checkpoint proteins are cell membrane proteins that suppress the differentiation, proliferation, and activation of immune cells. Specifically, these proteins are generally expressed on activated T cells and reduce T cell proliferation, cytokine secretion, and cytotoxicity, suppressing excessive T cell activity; therefore, they are also called co-inhibitory molecules. In particular, T cells express co-inhibitory receptors such as CTLA-4 and PD-1, and suppress T cell activity by binding to their respective ligands, B7.1 / 2 and PD-L1. On the other hand, PD-L1 expressed on cancer cells plays an important role as a molecular shield, inactivating cancer-specific T cells and inducing apoptosis, protecting cancer cells from T cell-mediated immune attacks and contributing to the immune evasion mechanism of cancer. Furthermore, it has been reported that cancer patients with ectopic expression of PD-L1 on their cancer cells have a worse prognosis than cancer patients without ectopic expression.

[0005] Immune checkpoint inhibitors are drugs that block the activity of immune checkpoint proteins involved in T cell suppression, thereby activating T cells to attack cancer cells. Representative antibodies used in this context include those targeting CTLA-4, PD-1, and PD-L1. Ipilimumab (Yervoy), a CTLA-4 inhibitor, was the first immune checkpoint inhibitor to receive FDA approval in 2011 as a second-line treatment for metastatic melanoma. Subsequently, in 2014, the PD-1 blockers nivolumab (Opdivo) and pembrolizumab (Keytruda) received FDA approval for metastatic melanoma, respectively. Then, in 2016, the PD-L1 inhibitor atezolizumab (Tecentriq) was approved for bladder cancer, avelumab (Bavencio) was approved in 2017 for metastatic Merkel cell carcinoma, a type of skin cancer, and durvalumab (Imfinzi) was approved for bladder cancer. In 2018, the PD-1 inhibitor cemiplimab (Libtayo) received FDA approval for the treatment of cutaneous squamous cell carcinoma. Currently, these drugs are expanding their indications and receiving FDA approval for an increasing number of cancer types. As of 2019, six PD-1 / PD-L1 inhibitors have received FDA approval for a total of 18 cancer types. Furthermore, other immunomodulatory proteins such as B7-H4, ICOS, HVEM, PDL-2, and PVRIG are also entering preclinical trials as new targets. Most of these therapies target molecules expressed on T cells.

[0006] To overcome the T-cell-centric bias in target discovery, recent efforts have focused on developing inhibitors that target immune checkpoint proteins expressed in myeloid cells such as macrophages and dendritic cells. Among these, CSF1R, CD47, and TLR7 have emerged as important targets.

[0007] PD-L1 (Programmed Cell Death Ligand 1) is an immune checkpoint protein that enables tumor cells to evade immune system attacks by suppressing the activity of T cells, and is mainly expressed in leukocytes of lymphoid and non-lymphoid tissues, as well as non-hematopoietic cells. Furthermore, PD-L1 is also expressed on the surface of various tumor cells such as colorectal cancer, pancreatic cancer, melanoma, and cervical cancer. In particular, PD-L1 interacts with PD-1 (Programmed Death-1) expressed on the surface of activated T cells, and negatively regulates the immune response of T cells by suppressing TCR-mediated activation of T cells, cytokine release, and T cell proliferation.

[0008] CD47 (Cluster of Differentiation 47), first identified as a tumor antigen of human ovarian cancer in 1980, is expressed in various human tumor cells such as non-Hodgkin lymphoma (NHL), multiple myeloma (MM), acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), chronic myeloid leukemia, bladder cancer, and solid tumors.

[0009] CD47 is expressed on the cell surface and interacts with SIRPα (Signal Regulatory Protein α), thrombospondin-1 (TSP1), and integrin proteins, and is involved in cell apoptosis and phagocytosis, proliferation, and immune response. Specifically, CD47 expressed on tumor cells interacts with SIRPα expressed on the surface of macrophages, and transmits a "don't eat me" signal so that tumor cells can avoid phagocytosis by macrophages. Furthermore, in the tumor microenvironment, CD47 inhibits angiogenesis and the function of effector T cells, and promotes the proliferation and growth of tumor cells.

[0010] Magrolimab is an antibody therapy targeting CD47 developed by Gilead Sciences, a global pharmaceutical company, and is currently in clinical trials. Recently, global multinational pharmaceutical companies have been actively working on the development of bispecific antibodies based on CD47 antibody therapy.

[0011] Non-human-derived antibodies are often immunogenic, which may limit their effectiveness and, in some cases, raise concerns about inducing harmful allergic reactions. The immune response to these foreign antibodies promotes their elimination from the body, suppresses their ability to bind to the target antigen, thereby significantly reducing the effectiveness of the antibody. To overcome these problems, it is possible to humanize non-human antibodies to reduce their immunogenicity in humans while maintaining the specificity and affinity of the parental non-human antibody.

[0012] Therefore, the inventors have developed humanized antibodies targeting the immune checkpoint proteins PD-L1 and CD47 as immune checkpoint inhibitors and have reached this application.

Prior Art Documents

Patent Documents

[0013]

Patent Document 1

Patent Document 2

Patent Document 3

Non-Patent Documents

[0014]

Non-Patent Document 1

Non-Patent Document 2

Non-Patent Document 3

[0015] The object of the present invention is to provide a bispecific humanized single-domain antibody that targets the immune checkpoint proteins PD-L1 and CD47, and the use thereof. [Means for solving the problem]

[0016] To achieve the objectives of the present invention, the present invention provides a bispecific antibody that bispecifically binds to PD-L1 and CD47, comprising a first humanized single-domain antibody (first hsdAb) or its antigen-binding fragment that specifically binds to PD-L1; and a second humanized single-domain antibody (second hsdAb) or its antigen-binding fragment that specifically binds to CD47.

[0017] In one embodiment of the present invention, the first hsdAb or its antigen-binding fragment may include CDR1 consisting of the amino acid sequence represented by SEQ ID NO: 2; CDR2 consisting of the amino acid sequence represented by SEQ ID NO: 3; and CDR3 consisting of the amino acid sequence represented by SEQ ID NO: 4.

[0018] Furthermore, the second hsdAb or its antigen-binding fragment may include CDR1 consisting of the amino acid sequence represented by SEQ ID NO: 9; CDR2 consisting of the amino acid sequence represented by SEQ ID NO: 10; and CDR3 consisting of the amino acid sequence represented by SEQ ID NO: 11.

[0019] Furthermore, the first or second hsdAb may include a heavy chain variable domain framework sequence having at least 95% sequence identity with sequence number 16, or a heavy chain variable domain framework sequence having at least 85% sequence identity with sequence number 17.

[0020] Furthermore, the first or second hsdAb may include a heavy chain variable domain framework sequence having 100% sequence identity with sequence number 16 or 17. Specifically, the first hsdAb or its antigen-binding fragment and / or the second hsdAb or its antigen-binding fragment may include a heavy chain variable domain framework sequence comprising (1) FR1 consisting of an amino acid sequence represented by either SEQ ID NO: 19 or 23; (2) FR2 consisting of an amino acid sequence represented by either SEQ ID NO: 20 or 24; (3) FR3 consisting of an amino acid sequence represented by either SEQ ID NO: 21 or 25; and (4) FR4 consisting of an amino acid sequence represented by either SEQ ID NO: 22 or 26. More specifically, the first hsdAb or its antigen-binding fragment may include a heavy chain variable domain framework sequence comprising FR1 consisting of an amino acid sequence represented by SEQ ID NO: 19; FR2 consisting of an amino acid sequence represented by SEQ ID NO: 20; FR3 consisting of an amino acid sequence represented by SEQ ID NO: 21; and FR4 consisting of an amino acid sequence represented by SEQ ID NO: 22, and the second hsdAb or its antigen-binding fragment may include a heavy chain variable domain framework sequence comprising FR1 consisting of an amino acid sequence represented by SEQ ID NO: 23; FR2 consisting of an amino acid sequence represented by SEQ ID NO: 24; FR3 consisting of an amino acid sequence represented by SEQ ID NO: 25; and FR4 consisting of an amino acid sequence represented by SEQ ID NO: 26. In a particular embodiment, the first hsdAb comprises the amino acid sequence represented by SEQ ID NO: 16, and the second hsdAb comprises the amino acid sequence represented by SEQ ID NO: 17.

[0021] Furthermore, the first hsdAb or its antigen-binding fragment and / or the second hsdAb or its antigen-binding fragment may be monovalent, divalent, trivalent, tetravalent, or more. Also, the first hsdAb or its antigen-binding fragment may be fused with each other via a peptide linker. Furthermore, it may be fused with the second hsdAb or its antigen-binding fragment.

[0022] In one embodiment of the present invention, an Fc fragment may be fused to the first hsdAb or its antigen-binding fragment and the second hsdAb or its antigen-binding fragment via a peptide linker. In a particular embodiment, the amino acid sequence is represented by SEQ ID NO: 15.

[0023] In one embodiment of the present invention, an antibody (HCAb) consisting only of a heavy chain in which an Fc fragment is fused to the first hsdAb or its antigen-binding fragment, or to the second hsdAb or its antigen-binding fragment, is provided.

[0024] In one embodiment of the present invention, a bispecific and polyvalent heavy chain-only antibody (HCAb) is provided, comprising two or more copies of the first hsdAb or its antigen-binding fragment and / or the second hsdAb or its antigen-binding fragment, wherein the HCAb may be bivalent, trivalent, tetravalent, or higher, and may be an HCAb fused with an Fc fragment. In certain embodiments, the HCAb may consist of an amino acid sequence represented by SEQ ID NO: 18.

[0025] In one embodiment of the present invention, the sdAb may fuse with the Fc fragment via a peptide linker, and the Fc fragment may be human IgG1, IgG2, IgG3, or IgG4.

[0026] In one embodiment of the present invention, the sdAb comprises at least one amino acid substitution, wherein the at least one amino acid substitution is a conservative substitution and may be a non-genetically coding amino acid or a synthetic amino acid.

[0027] In one embodiment of the present invention, an immunomodulator, cytokine, cytotoxic agent, chemotherapeutic agent, diagnostic agent, antiviral agent, antimicrobial agent, or drug may be conjugated. Accordingly, the present invention provides an antibody conjugate comprising a bispecific antibody that bispecifically binds to PD-L1 and CD47 conjugated to an immunomodulator, cytokine, cytotoxic agent, chemotherapeutic agent, diagnostic agent, antiviral agent, antimicrobial agent, or drug.

[0028] Furthermore, the present invention provides a nucleic acid molecule encoding a bispecific antibody that binds bispecifically to PD-L1 and CD47.

[0029] Furthermore, the present invention provides an expression vector containing the nucleic acid molecule.

[0030] Furthermore, the present invention provides host cells transformed with the expression vector.

[0031] Furthermore, the present invention is (a) A step of culturing host cells under conditions that enable the expression of bispecific antibodies; and (b) A step of recovering the expressed bispecific antibody; This invention provides a method for producing bispecific antibodies that bispecifically bind to PD-L1 and CD47, including [specific components of the antibody].

[0032] Furthermore, the present invention provides a pharmaceutical composition for the prevention or treatment of cancer containing a bispecific antibody or antibody complex that bispecifically binds to PD-L1 and CD47 as an active ingredient; a method for the prevention or treatment of cancer comprising the step of administering a bispecific antibody or antibody complex that bispecifically binds to PD-L1 and CD47 to an individual; the use of a bispecific antibody or antibody complex that bispecifically binds to PD-L1 and CD47 for use in a pharmaceutical composition for the prevention or treatment of cancer; and the use of a bispecific antibody or antibody complex that bispecifically binds to PD-L1 and CD47 for the manufacture of a pharmaceutical composition for the prevention or treatment of cancer.

[0033] In one embodiment of the present invention, the cancer may be selected from the group consisting of melanoma, lung cancer, liver cancer, glioblastoma, ovarian cancer, colorectal cancer, head and neck cancer, bladder cancer, renal cell carcinoma, gastric cancer, breast cancer, metastatic cancer, prostate cancer, pancreatic cancer, non-Hodgkin lymphoma, Hodgkin lymphoma, multiple myeloma, leukemia, lymphoma, myelodysplastic syndrome, acute lymphoblastic leukemia, acute myeloid leukemia, chronic lymphocytic leukemia, chronic myeloid leukemia, solitary myeloma, and aplastic anemia.

[0034] In one embodiment of the present invention, the pharmaceutical composition may further include a pharmaceutically acceptable carrier. [Effects of the Invention]

[0035] In this invention, we developed a humanized single-domain antibody that bispecifically binds to the immune checkpoint proteins PD-L1 and CD47, and confirmed its efficacy in vitro and in vivo. Therefore, this bispecific humanized single-domain antibody can be usefully used as an immune checkpoint inhibitor in immunotherapy for cancer. [Brief explanation of the drawing]

[0036] [Figure 1A] This figure shows the binding ability (EC50) of an anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) prepared according to one embodiment of the present invention to the PD-L1 antigen expressed on the cell surface. This binding ability was confirmed by reacting the anti-PD-L1 HCAb with the CHO-K1_PD-L1 cell line (CHO-K1 cells in which PD-L1 antigen expression has been induced) at various concentrations. [Figure 1B] This figure shows the inhibitory activity (IC50) of an anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1) prepared according to one embodiment of the present invention regarding PD-L1 / PD-1 interaction. This inhibitory activity was confirmed by evaluating the binding of the anti-PD-L1 HCAb to the CHO-K1_PD-L1 cell line (CHO-K1 cells in which PD-L1 antigen expression was induced) after binding of the PD-1 protein. [Figure 2A] This figure shows the binding ability (EC50) of anti-PD-L1 bivalent HCAb (PP Nb-IgG4) prepared according to one embodiment of the present invention to the PD-L1 antigen expressed on the cell surface. This binding ability was confirmed by reacting anti-PD-L1 bivalent HCAb with the CHO-K1_PD-L1 cell line (CHO-K1 cells in which PD-L1 antigen expression has been induced) at various concentrations. [Figure 2B]This shows the inhibitory activity (IC50) of the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) prepared according to one embodiment of the present invention against PD-L1 / PD-1 interaction. This inhibitory activity was confirmed by evaluating the binding of the anti-PD-L1 bivalent HCAb to the CHO-K1_PD-L1 cell line (CHO-K1 cells in which PD-L1 antigen expression was induced) after binding of the PD-1 protein. [Figure 3A] This shows the binding ability (EC50) of an anti-CD47 HCAb (CD47 Nb-IgG4) prepared according to one embodiment of the present invention to the CD47 antigen expressed on the cell surface. This binding ability was confirmed by reacting the anti-CD47 HCAb with the Expi-CHO_CD47 cell line (Expi-CHO cells in which CD47 antigen expression has been induced) at various concentrations. [Figure 3B] This figure shows the inhibitory activity (IC50) of an anti-CD47 HCAb (CD47 Nb-IgG4) prepared according to one embodiment of the present invention against the CD47 / SIRPα interaction. This inhibitory activity was confirmed by evaluating the binding of the anti-CD47 HCAb to the Expi-CHO_CD47 cell line (Expi-CHO cells in which CD47 antigen expression has been induced) after binding of the SIRPα protein. [Figure 4A] This figure shows the binding ability (EC50) of anti-PD-L1×CD47 HCAb (PPC Nb-IgG4), prepared according to one embodiment of the present invention, to the PD-L1 antigen expressed on the cell surface. This binding ability was confirmed by reacting anti-PD-L1×CD47 HCAb with the CHO-K1_PD-L1 cell line (CHO-K1 cells in which PD-L1 antigen expression has been induced) at various concentrations. [Figure 4B] This figure shows the binding ability (EC50) of anti-PD-L1×CD47 HCAb (PPC Nb-IgG4) prepared according to one embodiment of the present invention to the CD47 antigen expressed on the cell surface. This binding ability was confirmed by reacting anti-PD-L1×CD47 HCAb with the Expi-CHO_CD47 cell line (Expi-CHO cells in which CD47 antigen expression has been induced) at various concentrations. [Figure 5A]This figure shows the inhibitory activity (IC50) of anti-PD-L1×CD47 HCAb (PPC Nb-IgG4) against PD-L1 / PD-1 interaction, prepared according to one embodiment of the present invention. This inhibitory activity was confirmed by evaluating the binding of anti-PD-L1×CD47 HCAb to the CHO-K1_PD-L1 cell line (CHO-K1 cells in which PD-L1 antigen expression has been induced) after binding of the PD-1 protein. [Figure 5B] This figure shows the inhibitory activity (IC50) of anti-PD-L1×CD47 HCAb (PPC Nb-IgG4) against CD47 / SIRPα interaction, prepared according to one embodiment of the present invention. This inhibitory activity was confirmed by evaluating the binding of anti-PD-L1×CD47 HCAb to the Expi-CHO_CD47 cell line (Expi-CHO cells in which CD47 antigen expression has been induced) after binding of the SIRPα protein. [Figure 6A] This figure shows the binding ability (EC50) of a humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention to the PD-L1 antigen expressed on the cell surface. This binding ability was confirmed by reacting the humanized anti-PD-L1×CD47 HCAb with the CHO-K1_PD-L1 cell line (CHO-K1 cells in which PD-L1 antigen expression has been induced) at various concentrations. [Figure 6B] This figure shows the binding ability (EC50) of a humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention to the CD47 antigen expressed on the cell surface. This binding ability was confirmed by reacting the humanized anti-PD-L1×CD47 HCAb with the Expi-CHO_CD47 cell line (Expi-CHO cells in which CD47 antigen expression has been induced) at various concentrations. [Figure 7A]This figure shows the inhibitory activity (IC50) of humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) against PD-L1 / PD-1 interaction, prepared according to one embodiment of the present invention. This inhibitory activity was confirmed by evaluating the binding of humanized anti-PD-L1×CD47 HCAb to the CHO-K1_PD-L1 cell line (CHO-K1 cells in which PD-L1 antigen expression has been induced) after binding of the PD-1 protein. [Figure 7B] This figure shows the inhibitory activity (IC50) of humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) against CD47 / SIRPα interaction, prepared according to one embodiment of the present invention. This inhibitory activity was confirmed by evaluating the binding of humanized anti-PD-L1×CD47 HCAb to the Expi-CHO_CD47 cell line (Expi-CHO cells in which CD47 antigen expression has been induced) after binding of the SIRPα protein. [Figure 8] This figure shows T cell activation induced by interference with intercellular interactions using a humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention. This was confirmed using CHO-K1 cells that induce overexpression of PD-L1 protein and Jurkat cells that express PD-1. [Figure 9A] and [Figure 9B] This figure shows the activation of CD4+ T cells in a mixed lymphocyte reaction, confirming the degree of activation induced by humanized anti-PD-L1×CD47 HCAb(hPPC Nb-IgG4) prepared according to one embodiment of the present invention. [Figure 10A] This figure shows the degree of phagocytosis by macrophages, confirming the phagocytosis induced by humanized anti-PD-L1×CD47 HCAb(hPPC Nb-IgG4) prepared according to one embodiment of the present invention. [Figure 10B] This figure shows the degree of phagocytosis by macrophages, confirming the phagocytosis induced by humanized anti-PD-L1×CD47 HCAb(hPPC Nb-IgG4) prepared according to one embodiment of the present invention. [Figure 11]This figure shows the binding ability of humanized anti-PD-L1×CD47 HCAb(hPPC Nb-IgG4) prepared according to one embodiment of the present invention to human RBCs (red blood cells). [Figure 12] This figure shows the hemagglutination reaction of a humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention. [Figure 13] This figure shows the antitumor effect of humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) in a PD-L1 and CD47 antigen-expressing tumor-bearing C57BL / 6 mouse model. [Figure 14] This figure shows the antitumor effect of humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) in a PD-L1 and CD47 antigen-expressing tumor-bearing humanized NSG mouse model. [Figure 15A] This figure shows the antitumor effect and survival rate of humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) in a humanized NSG mouse model with tumorigenesis induced by the human breast cancer cell line MDA-MB-231. [Figure 15B] This figure shows the antitumor effect and survival rate of humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) in a humanized NSG mouse model with tumorigenesis induced by the human breast cancer cell line MDA-MB-231. [Best Mode for Carrying Out the Invention]

[0037] Hereinafter, embodiments of the present invention will be described in detail so that those skilled in the art can easily implement the invention. These embodiments are provided to fully illustrate the present invention to those skilled in the art. Accordingly, embodiments of the present invention can be modified in various ways, and the scope of the present invention is not limited to the embodiments described below.

[0038] In this invention, the term "epitope" refers to a protein binding determinant that can specifically bind to an antibody. Epitopes generally consist of a group of surfaces of chemically active molecules, such as amino acids or sugar side chains, and generally possess specific three-dimensional structural characteristics and particularly specific charge characteristics.

[0039] The term "treatment" refers to any process that reduces, interrupts, stops, controls, halts, alleviates or improves, or reverses the progression of, the symptoms or complications of any disorder or disease disclosed herein, but does not necessarily mean the complete elimination of all symptoms of the disease or disorder.

[0040] The term "prevention" refers to the preventative treatment of a disease or disorder, or the delay of the onset or progression of a disease or disorder.

[0041] The terms “subject” or “individual” refer to mammals, including but not limited to humans, cattle, horses, cats, dogs, rodents, or primates. In certain embodiments, the individual is a human.

[0042] The term "antibody" is used in its broadest sense and encompasses a variety of antibody structures, including monoclonal antibodies, polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), full-length antibodies, and their antigen-binding fragments, as long as they exhibit the desired antigen-binding activity. The term "antibody" also includes conventional four-chain antibodies, single-domain antibodies, and their antigen-binding fragments.

[0043] A basic four-chain antibody unit is a heterotetrameric glycoprotein consisting of two identical light (L) chains and two identical heavy (H) chains. IgM antibodies consist of five of these basic heterotetrameric units plus an additional polypeptide called a J chain, which contains 10 antigen-binding sites. In contrast, IgA antibodies contain 2 to 5 basic four-chain units that can polymerize with the J chain to form a multimeric assembly. For IgG, the molecular weight of a four-chain unit is typically around 150,000 daltons. Each L chain is linked to the H chain by one covalent disulfide bond, and the two H chains are linked to each other by one or more disulfide bonds depending on the H chain isotype. Each H and L chain also has regularly separated interchain disulfide bonds. Each H chain has a variable domain (VH) at its N-terminus, corresponding to the α and γ chains respectively, followed by three constant domains (CH), and four μ- and ε-type CH domains. Each light chain (L) has a variable domain (VL) at its N-terminus and a constant domain at the other end. The VL aligns with the VH, and the CL aligns with the first constant domain (CH1) of the heavy chain. The pairing of VH and VL forms a single antigen-binding site. The L chain of any vertebrate species can be assigned to one of two clearly defined types called kappa and lambda, based on the amino acid sequence of its constant domain. Immunoglobulins can be classified into different classes or isotypes based on the amino acid sequence of the constant domain (CH) of their heavy chain. There are five classes of immunoglobulins: IgA, IgD, IgE, IgG, and IgM, with their heavy chains called α, δ, ε, γ, and μ, respectively. The γ and α classes are further divided into subclasses based on relatively small differences in CH sequence and function. For example, humans express the following subclasses: IgG1, IgG2A, IgG2B, IgG3, IgG4, IgA1, and IgA2.

[0044] The term "heavy-chain-only antibody" or "HCAb" refers to a functional antibody that contains a heavy chain but lacks the light chain typically found in conventional four-chain antibodies.

[0045] The terms “single-domain antibody,” “nanobody,” or “sdAb” refer to a single antigen-binding polypeptide containing three complementarity-determining regions (CDRs). An sdAb alone can bind to an antigen without pairing with a corresponding CDR-containing polypeptide. In some cases, single-domain antibodies are produced from camelid HCAbs, and their heavy-chain variable domains are referred to herein as “VHH” (heavy-chain variable domains of heavy-chain antibodies). A basic VHH has the following structure from N-terminus to C-terminus: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4, where FR1 to FR4 refer to framework regions 1 to 4, respectively, and CDR1 to CDR3 refer to complementarity-determining regions 1 to 3.

[0046] The term "variable region" or "variable domain" of an antibody refers to the amino-terminal domain of the antibody's heavy or light chain. The variable domains of the heavy and light chains are commonly referred to as "VH" and "VL," respectively. These domains are generally the most variable parts of the antibody (compared to other antibodies of the same class) and contain the antigen-binding site. Antibodies consisting solely of the heavy chain of camelid species have a single heavy-chain variable region called "VHH."

[0047] The term "variable" refers to the fact that specific segments of the variable domain exhibit wide sequence diversity among antibodies. The V domain mediates antigen binding and defines the specificity of a particular antibody to a particular antigen. However, variability is not evenly distributed across the entire range of the variable domain. Instead, in both the heavy and light chain variable domains, it is concentrated into three segments called complementarity-determining regions (CDRs) or hypervariable regions (HVRs). A more highly conserved portion of the variable domain is called the framework region (FR). The variable domains of the natural heavy and light chains each contain four framework (FR) regions, which primarily employ a beta-sheet structure, sometimes forming part of a beta-sheet structure, and three CDRs linked by loop junctions. The CDRs of each chain are closely maintained by the FR regions, and the CDRs of different chains contribute to the formation of the antibody's antigen-binding site (see: Kabat, Elvin A., Sequences of Immunological Interest, Fifth Edition, National Institute of Health, Bethesda, Md. (1991)). The constant domain does not directly participate in antigen-antibody binding, but it is involved in various effector functions, such as mediating antibody-dependent cytotoxicity.

[0048] The term "constant domain" refers to a portion of an immunoglobulin molecule that has a more conserved amino acid sequence than the variable domain (the other parts of the immunoglobulin, including the antigen-binding site). Constant domains include the CH1, CH2, and CH3 domains of the heavy chain (collectively referred to as CH) and the CHL (or CL) domain of the light chain.

[0049] The terms "full-length antibody," "complete antibody," or "whole antibody" are used interchangeably to refer to an antibody in substantially complete form, as opposed to an antibody fragment. Specifically, a full-length four-chain antibody includes a heavy chain containing the Fc region and a light chain. A full-length heavy-chain-only antibody includes a heavy chain variable domain (e.g., VHH) and an Fc region. The constant domain may be a native sequence constant domain (e.g., a human native sequence constant domain) or an amino acid sequence variant thereof. In some cases, a complete antibody may have one or more effector functions.

[0050] The terms “antibody fragment” or “antigen-binding fragment” refer to a portion of a complete antibody, preferably including the antigen-binding and / or variable region of the complete antibody. Examples of antibody fragments include, but are not limited to, Fab, Fab', F(ab')2, and Fv fragments; diabodies; linear antibodies; single-chain antibody (scFv) molecules; single-domain antibodies (e.g., VHH); and multispecific antibodies formed from antibody fragments. “Fv” is a minimal antibody fragment containing the complete antigen recognition and binding site. This fragment consists of a compact, non-covalently tightly bound dimer of one heavy chain and one light chain variable region domain. The term “single-chain F5” is also abbreviated as “sFv” or “scFv” and is an antibody fragment containing VH and VL antibody domains linked to a single polypeptide chain. Preferably, the scFv polypeptide further includes a polypeptide linker between the VH and VL domains, allowing the scFv to form a desired structure for antigen binding. The term "diabody" refers to a small antibody fragment produced by constructing an sFv fragment with a short linker (approximately 5-10 amino acid residues) between the VH and VL domains. This facilitates inter-chain pairing rather than intra-chain pairing of the V domain, resulting in a bivalent fragment containing two antigen-binding sites. A bispecific diabody is a heterodimer of two "cross-reacting" sFv fragments, where the VH and VL domains of two different antibodies are located on different polypeptide chains.

[0051] The term "humanized antibody" is used as a subset of "chimeric antibody."

[0052] The "humanized" form of non-human (e.g., llama or camelid) antibodies is a chimeric antibody containing a minimal sequence derived from a non-human immunoglobulin. In some prevailing forms, a humanized antibody is an immunoglobulin in which residues from a donor species CDR (e.g., mouse, rat, rabbit, camel, llama, alpaca, or non-human primate CDR) are replaced with residues from a human immunoglobulin (recipient antibody) CDR that have the desired specificity, affinity, and / or functionality.

[0053] In some cases, framework ("FR") residues of human immunoglobulins are replaced with corresponding non-human residues. Furthermore, humanized antibodies may contain residues not found in either the recipient or donor antibody. These modifications may be made to further improve antibody performance, such as by enhancing binding affinity.

[0054] The terms “hypervariable region,” “HVR,” or “HV,” as used herein, refer to regions of the antibody variable domain that exhibit hypervariability and / or form structurally defined loops. Generally, single-domain antibodies contain three HVRs (or CDRs): HVR1 (or CDR1), HVR2 (or CDR2), and HVR3 (or CDR3). HVR3 (or CDR3) is the most diverse of the three HVRs and is known to play a unique role in conferring superior specificity to the antibody. For example, see Hamers-Casterman et al., Nature 363:446-448 (1993); Sheriff et al., Nature Struct. Biol. 3:733-736 (1996).

[0055] The terms "complementarity-determining region" or "CDR" are used to refer to hypervariable regions defined by the Kabat system (see: Kabat, Elvin A., Sequences of Proteins of Immunological Interest, 5th Ed., Public Health Service, National Institutes of Health, Bethesda, Md. (1991)). Kabat complementarity-determining regions (CDRs) are most commonly used based on sequence variability.

[0056] The terms "framework" or "FR" residues refer to variable-domain residues other than HVR residues as defined herein.

[0057] The term "specific" refers to the selective recognition of a particular epitope of an antigen by an antigen-binding protein (e.g., sdAb).

[0058] Natural antibodies are, for example, monospecific. As used herein, the term “multispecific” refers to an antigen-binding protein that has polyepitope specificity (i.e., the ability to specifically bind to two, three, or more different epitopes on a single biological molecule, or the ability to specifically bind to epitopes on two, three, or more different biological molecules). As used herein, “bispecific” refers to an antigen-binding protein that has two different antigen-binding specificities.

[0059] As used herein, the term "single specificity" refers to an antigen-binding protein that has one or more binding sites, each specifically binding to the same epitope of the same antigen.

[0060] The term "valency" refers to a specified number of binding sites present in an antigen-binding protein. For example, the terms "divalent," "trivalent," "tetravalent," "pentavalent," and "hexavalent" refer to antigen-binding proteins having 2, 3, 4, 5, and 6 binding sites, respectively.

[0061] The term "antibody effector function" refers to the biological activity of an antibody that is attributed to the antibody's Fc region (either the native sequence Fc region or the amino acid sequence variant Fc region), and which varies depending on the antibody isotype. Examples of antibody effector functions include: C1q binding and complement-dependent cytotoxicity; Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; downregulation of cell surface receptors (e.g., B cell receptors); and B cell activation. "Complement-dependent cytotoxicity" or "CDC" refers to the lysis of target cells in the presence of complement. Activation of the classical complement pathway is initiated when the first component of the complement system (C1q) binds to an appropriate subclass of antibody, and that antibody binds to a homologous antigen. "Antibody-dependent cell-mediated cytotoxicity" or "ADCC" refers to a form of cytotoxicity in which secreted immunoglobulin (Ig) bound to Fc receptors (FcRs) present on certain cytotoxic cells (e.g., natural killer (NK) cells, neutrophils, and macrophages) causes these cytotoxic effector cells to specifically bind to antigen-carrying target cells, which then kill the target cells with cytotoxins.

[0062] In this specification, the terms “Fc region” or “fragment crystallizable region” are used to define the C-terminal region of an immunoglobulin heavy chain, including both the native sequence Fc region and its variant Fc regions. Native sequence Fc regions suitable for use in the antibodies described herein include human IgG1, IgG2 (IgG2A, IgG2B), IgG3, and IgG4.

[0063] The term "binding affinity" generally refers to the overall strength of the non-covalent interaction between a single binding site of a molecule (e.g., an antibody) and its binding partner (e.g., an antigen). Unless otherwise specified, "binding affinity" as used herein refers to the intrinsic binding affinity that reflects the 1:1 interaction between the members of a binding pair. Binding affinity is K d , Ko ff , K on , or K a It is expressed as follows. The equilibrium dissociation constant "K" used herein D " or "K dThe term "___" refers to the dissociation constant of a specific antibody-antigen interaction, which represents the concentration of antigen required to occupy half of the antibody molecule binding domains present in solution at equilibrium, expressed in units of M. K D measurement assumes that all binding reagents are in solution. The dissociation constant (K D or K d ) serves as an indicator of the affinity of the antibody for the antigen. For example, easy analysis can be performed using the Scatchard method with antibodies recognized by various marker agents or using commercially available measurement kits, following the attached instruction manuals and experimental operation procedures. The K D values obtained by these methods are expressed in units of M (moles).

[0064] The terms "percent (%) amino acid sequence identity" and "homology" with respect to a peptide, polypeptide, or antibody sequence are defined as the percentage of amino acid residues in a candidate sequence that are identical to the amino acid residues within the specified peptide or polypeptide sequence, without considering conservative substitutions as part of sequence identity, after performing sequence alignment and introducing gaps as necessary to achieve the maximum percent sequence identity. For the purpose of determining the percent amino acid sequence identity, alignment can be achieved using various methods known in the art, for example, by using publicly available computer software such as BLAST, BLAST-2, ALIGN, or MEGALIGN® (DNATAR) software. One of ordinary skill in the art can determine appropriate parameters for measuring alignment, including any algorithm necessary to achieve optimal alignment over the entire length of the sequences being compared.

[0065] The present invention relates to a bispecific antibody that bispecifically binds to PD-L1 and CD47, comprising a first humanized single-domain antibody (first hsdAb) or its antigen-binding fragment (anti-PD-L1 hsdAb) that specifically binds to PD-L1, and a second humanized single-domain antibody (second hsdAb) or its antigen-binding fragment (anti-CD47 hsdAb) that specifically binds to CD47, for example, a humanized anti-PD-L1×CD47 bsAb, specifically a bispecific humanized single-domain antibody in which anti-PD-L1 hsdAb and anti-CD47 hsdAb are fused, and a humanized anti-PD-L1×CD47 heavy chain-only antibody (HCAb) (for example, a crystalline fragment (Fc fragment) of human immunoglobulin G (IgG) that is anti-PD-L1 hsdAb and / or anti-CD47). This invention relates to a humanized anti-PD-L1 × CD47 bsAb-Fc fusion protein fused to hsdAb, as well as its manufacture and use.

[0066] Accordingly, the present invention provides a bispecific antibody that bispecifically binds to PD-L1 and CD47, comprising a humanized anti-PD-L1×CD47 bsAb.

[0067] In the present invention, the bispecific antibody that bispecifically binds to PD-L1 and CD47, including the humanized anti-PD-L1×CD47 bsAb, may be a humanized anti-PD-L1×CD47 bsAb formed by the fusion of a first humanized single-domain antibody (first hsdAb) or its antigen-binding fragment (hereinafter referred to as "anti-PD-L1 hsdAb") that specifically binds to PD-L1 as a first antigen-binding portion, and a second humanized single-domain antibody (second hsdAb) or its antigen-binding fragment (hereinafter referred to as "anti-CD47 hsdAb") that specifically binds to CD47 as a second antigen-binding portion.

[0068] In the present invention, the anti-PD-L1 hsdAb comprises CDR1 consisting of the amino acid sequence represented by SEQ ID NO: 2; CDR2 consisting of the amino acid sequence represented by SEQ ID NO: 3; and CDR3 consisting of the amino acid sequence represented by SEQ ID NO: 4.

[0069] Furthermore, the anti-CD47 hsdAb includes CDR1, which consists of the amino acid sequence represented by SEQ ID NO: 9; CDR2, which consists of the amino acid sequence represented by SEQ ID NO: 10; and CDR3, which consists of the amino acid sequence represented by SEQ ID NO: 11.

[0070] The aforementioned CDR sequences are shown in Tables 6 and 12.

[0071] In the present invention, the humanized anti-PD-L1×CD47 bsAb comprises a human or humanized heavy chain domain framework region. The human or humanized heavy chain domain framework is a framework comprising the amino acid sequence of a heavy chain variable domain (VH) framework derived from a human immunoglobulin framework. The heavy chain variable domain (VH) framework derived from a human immunoglobulin framework may comprise an amino acid sequence containing the same amino acid sequence, or it may comprise amino acid changes. In certain embodiments, the number of amino acid changes is 10 or less, 9 or less, 8 or less, 7 or less, 6 or less, 5 or less, 4 or less, 3 or less, or 2 or less.

[0072] Specifically, the heavy chain variable domain framework sequence may include sequences that have at least 95%, 96%, 97%, 98%, 99%, or 100% sequence identity with sequence number 16. Or, it may include sequences that have at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity with sequence number 17.

[0073] Heavy chain variable domain framework (FR) sequences having 100% sequence identity with the sequence of sequence number 16 or sequence number 17 are shown in Tables 1 to 4 below.

[0074] [Table 1] [Table 2] [Table 3] [Table 4] Specifically, the anti-PD-L1 hsdAb may include the following FR1, FR2, FR3, and FR4: FR1 consists of the amino acid sequence represented by either SEQ ID NO: 19 or 23; FR2 consists of the amino acid sequence represented by either SEQ ID NO: 20 or 24; FR3 consisting of an amino acid sequence represented by either SEQ ID NO: 21 or 25; and FR4 consists of the amino acid sequence represented by either SEQ ID NO: 22 or 26.

[0075] More specifically, the anti-PD-L1 hsdAb may include the following FR1, FR2, FR3, and FR4: (1) FR1 consisting of the amino acid sequence represented by SEQ ID NO: 19; FR2 consists of the amino acid sequence represented by SEQ ID NO: 20; FR3 consisting of the amino acid sequence represented by SEQ ID NO: 21; and FR4 consisting of the amino acid sequence represented by SEQ ID NO: 22; or (2) FR1 consisting of the amino acid sequence represented by Sequence ID No. 23; FR2 consists of the amino acid sequence represented by SEQ ID NO: 24; FR3 consisting of the amino acid sequence represented by SEQ ID NO: 25; and FR4 consists of the amino acid sequence represented by SEQ ID NO: 26

[0076] More specifically, the anti-PD-L1 hsdAb may include the following FR1, FR2, FR3, and FR4: FR1 consists of the amino acid sequence represented by SEQ ID NO: 19; FR2 consists of the amino acid sequence represented by SEQ ID NO: 20; FR3 consisting of the amino acid sequence represented by SEQ ID NO: 21; and FR4 consists of the amino acid sequence represented by Sequence ID No. 22.

[0077] Furthermore, the anti-CD47 hsdAb can include the following FR1, FR2, FR3, and FR4: FR1 consists of an amino acid sequence represented by either SEQ ID NO: 19 or 23; FR2 consisting of an amino acid sequence represented by either SEQ ID NO: 20 or 24; FR3 consisting of an amino acid sequence represented by either SEQ ID NO: 21 or 25; and FR4 consists of the amino acid sequence represented by either SEQ ID NO: 22 or 26.

[0078] More specifically, the anti-CD47 hsdAb may include the following FR1, FR2, FR3, and FR4: (1) FR1 consisting of the amino acid sequence represented by SEQ ID NO: 19; FR2 consists of the amino acid sequence represented by SEQ ID NO: 20; FR3 consisting of the amino acid sequence represented by SEQ ID NO: 21; and FR4 consisting of the amino acid sequence represented by SEQ ID NO: 22; or (2) FR1 consisting of the amino acid sequence represented by Sequence ID No. 23; FR2 consists of the amino acid sequence represented by SEQ ID NO: 24; FR3 consisting of the amino acid sequence represented by SEQ ID NO: 25; and FR4 consists of the amino acid sequence represented by SEQ ID NO: 26

[0079] More specifically, the anti-CD47 hsdAb may include the following FR1, FR2, FR3, and FR4: FR1 consists of the amino acid sequence represented by SEQ ID NO: 23; FR2 consists of the amino acid sequence represented by SEQ ID NO: 24; FR3 consisting of the amino acid sequence represented by SEQ ID NO: 25; and FR4 consists of the amino acid sequence represented by SEQ ID NO: 26

[0080] In the present invention, the anti-PD-L1 hsdAb binds to the epitope of PD-L1, and the anti-CD47 hsdAb binds to the epitope of CD47.

[0081] Furthermore, the binding of the humanized anti-PD-L1×CD47 bsAb to PD-L1 and CD47 is K D Each is 10 -6 M~10 -12 M, 10 -6 M~shi10 -11 M, 10 -6 M~10 -10 M, 10 -6 M~10 -9 M, 10 -7 M~10 -9 M, or 10 -8 M~10 -9 M is fine.

[0082] Furthermore, the EC50 of the humanized anti-PD-L1×CD47 bsAb in FACS analysis was less than 500 nM, specifically in the following ranges: 0.01 nM to 500 nM, 0.01 nM to 400 nM, 0.01 nM to 300 nM, 0.01 nM to 200 nM, 0.01 nM to 100 nM, 0.01 to 50 nM, 0.01 to 10 nM, 0.1 nM to 500 nM, and 0.1 nM to 4 The values ​​may be 00nM, 0.1nM~300nM, 0.1nM~200nM, 0.1nM~100nM, 0.1~50nM, 0.1~10nM, 1nM~500nM, 1nM~400nM, 1nM~300nM, 1nM~200nM, 1nM~100nM, 1~50nM, or 1~10nM.

[0083] In the present invention, the humanized anti-PD-L1×CD47 bsAb can have any appropriate valency for each of the PD-L1 and CD47 epitopes. Specifically, the humanized anti-PD-L1×CD47 bsAb can have valencies of 2, 3, 4, 5, 6, or more for each of PD-L1 and CD47. See, for example, P. Chames and D. Baty, Chapter 6. Bispecific Single Domain Antibodies, Springer-Verlag Berlin Heidelberg, 2011.

[0084] Furthermore, the humanized anti-PD-L1×CD47 bsAb may be formed by direct binding of anti-PD-L1 hsdAb and anti-CD47 hsdAb via peptide bonds, or by indirect fusion via a peptide linker. The length, flexibility, and / or other properties of the peptide linker may affect certain properties, including but not limited to affinity, specificity, or binding ability to one or more specific antigens or epitopes. For example, a longer peptide linker may be selected to prevent steric interference between two adjacent domains. In certain embodiments, the peptide linker may include flexible residues (e.g., glycine and serine) to allow adjacent domains to move freely from one another. For example, a glycine-serine doublet may be a suitable peptide linker. Furthermore, the peptide linker may be any suitable length, for example, any length of at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 50, 75, 100 or more amino acid residues. In some embodiments, the peptide linker may be a peptide linker consisting of any one of the amino acid sequences from SEQ ID NOs. 27 to SEQ ID NOs. 29.

[0085] Furthermore, the peptide linker may have a naturally occurring sequence or a non-naturally occurring sequence.

[0086] In some embodiments, the humanized anti-PD-L1×CD47 bsAb may be divalent with respect to PD-L1 and monovalent with respect to CD27, where the divalent anti-PD-L1 hsdAb can be directly or indirectly fused to the monovalent anti-PD-L1 hsdAb represented by amino acid sequence number 16 via a peptide linker. Furthermore, the humanized anti-PD-L1×CD47 bsAb may consist of the amino acid sequence represented by sequence number 15.

[0087] The single-domain antibodies (sdAbs) according to the present invention may be derived from any species, including but not limited to mice, rats, humans, camels, llamas, lampreys, cartilaginous fish, fish, goats, rabbits, and Bovidae. Furthermore, naturally occurring sdAb molecules may be derived from species other than camelidae.

[0088] Furthermore, sdAb is a light-chain-deficient heavy-chain antibody derived from known naturally occurring single-domain antigen-binding molecules. Such single-domain molecules are disclosed, for example, in WO 94 / 04678 and Hamers-Casterman et al. (1993), Nature 363:446-448. Variable domains derived from naturally light-chain-deficient heavy-chain molecules are disclosed herein as VHH and are distinguished from conventional VH of quadri-chain immunoglobulins. Such VHH molecules can be derived from antibodies produced by camelid animals such as camels, llamas, picunyas, dromedaries, alpacas, and guanacos. Other species outside the camelid family that can naturally produce light-chain-deficient heavy-chain molecules are also included in the scope of this invention.

[0089] According to the present invention, a humanized single-domain antibody (hsdAb) is typically a humanized antibody obtained by humanizing a non-human antibody to reduce its immunogenicity in humans while retaining the specificity and affinity of the parent non-human antibody. Generally, a humanized antibody consists of one or more variable domains, the hypervariable region (HVR), such as a complementarity-determining region (CDR), or a portion thereof, derived from the humanized antibody sequence. In some embodiments, a humanized single-domain antibody includes a CDR derived from a non-human single-domain antibody (e.g., a CDR derived from a camelid such as a llama) and a heavy chain FR derived from the human antibody sequence. In some specific embodiments, certain FR residues in the humanized antibody are substituted, for example, with corresponding residues from a non-human antibody (e.g., an antibody from which the HVR residues are derived), thereby restoring or enhancing the specificity or affinity of the antibody.

[0090] In the present invention, the bispecific antibody that bispecifically binds to PD-L1 and CD47, including the humanized anti-PD-L1×CD47 bsAb, may also be a humanized anti-PD-L1×CD47 HCAb or its antigen-binding fragment.

[0091] Specifically, the humanized anti-PD-L1×CD47 HCAb may be obtained by fusing the humanized anti-PD-L1×CD47 bsAb described herein to one or more CH2 and / or CH3 domains such as an Fc fragment. Furthermore, the anti-PD-L1 hsdAb and / or anti-CD47 hsdAb described herein may be obtained by fusing one or more CH2 and / or CH3 domains such as an Fc fragment.

[0092] The CH2 and / or CH3 domains are derived from immunoglobulin and may be IgA, IgD, IgE, IgG, or IgM, and may specifically be IgG. In some embodiments, the humanized anti-PD-L1×CD47 HCAb may include an Fc fragment of IgG, for example IgG1, IgG2, IgG3, or IgG4, and the Fc fragment may be human Fc, for example human IgG1(hIgG1)Fc, hIgG2Fc, hIgG3Fc, or hIgG4Fc.

[0093] The humanized anti-PD-L1×CD47 HCAb may be a monomer or a polymer. In the case of a polymer, it may be bispecific and polyvalent (e.g., divalent, trivalent, tetravalent, or more valencies), and may contain two or more copies of the anti-PD-L1 hsdAb and anti-CD47 hsdAb described herein.

[0094] In the present invention, the humanized anti-PD-L1×CD47 bsAb, anti-PD-L1 hsdAb, or anti-CD47 hsdAb may be fused to the CH2 and / or CH3 domains, specifically the Fc fragment, via a peptide linker. The length, flexibility, and / or other properties of the peptide linker may affect certain properties, including but not limited to affinity, specificity, or binding ability to one or more specific antigens or epitopes. For example, a longer peptide linker may be selected so that two adjacent domains do not sterically interfere with each other. In some embodiments, the peptide linker includes flexible residues (e.g., glycine and serine) to allow adjacent domains to move freely from one another. For example, a glycine-serine doublet may be a suitable peptide linker. Furthermore, the peptide linker may be of any suitable length, such as at least approximately 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 50, 75, 100 or more amino acids.

[0095] Furthermore, the peptide linker may have a natural or non-natural sequence. For example, a sequence derived from the hinge region of a heavy-chain-only antibody may be used as the linker. See, for example, WO1996 / 34103. In some embodiments, the peptide linker may be an hIgG1 hinge, an hIgG2 hinge, an hIgG3 hinge, an hIgG4 hinge, or a variant thereof. Alternatively, it may be a peptide linker consisting of the amino acid sequence of SEQ ID NO: 30.

[0096] In the present invention, the humanized anti-PD-L1×CD47 HCAb may include the amino acid sequence represented by SEQ ID NO: 18, or a variant thereof that exhibits at least 80% sequence homology to the amino acid sequence (for example, at least 80%, 88%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%).

[0097] In the present invention, the bispecific antibody that bispecifically binds to PD-L1 and CD47, including the humanized anti-PD-L1×CD47 bsAb, may be a bispecific Fab-like antibody fragment (bsFab) in which the C-terminus of anti-PD-L1 hsdAb and / or anti-CD47 hsdAb is fused to the N-terminus of the CH1 and Cκ domains.

[0098] The CH1 and Cκ domains are derived from an immunoglobulin. The immunoglobulin may be IgA, IgD, IgE, IgG, or IgM, and specifically, IgG. The IgG may be IgG1, IgG2, IgG3, or IgG4, and may be human IgG1, IgG2, IgG3, or IgG4.

[0099] The bsFab and its manufacturing technology according to the present invention are cited from P. Chames and D. Baty, Chapter 6. Bispecific Single Domain Antibodies, Springer-Verlag Berlin Heidelberg, 2011.

[0100] In the present invention, the bispecific antibodies that bispecifically bind to PD-L1 and CD47, including the humanized anti-PD-L1×CD47 bsAb, include amino acid sequence variants. Amino acid sequence variants of antibodies can be produced by introducing appropriate modifications to the nucleic acid sequence encoding the antibody, or by peptide synthesis. Such modifications include, for example, deletions, and / or insertions, and / or substitutions of residues within the amino acid sequence of the antibody. Any combination of deletions, insertions, and substitutions can yield the final product, provided that the final product retains desired properties, such as antigen binding. In some embodiments, substitutions, insertions, or deletions may occur within one or more hypervariable regions (HVRs), provided that such modifications do not substantially reduce the antibody's ability to bind to the antigen. For example, conservative modifications that do not significantly reduce binding affinity may be made within HVRs. Such modifications may also occur outside of HVR "hot spots" or CDRs.

[0101] Furthermore, the amino acid substitution may be at least one substitution (e.g., any 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids). Also, the at least one amino acid substitution may be a conservative substitution, or it may be a substitution with a non-genetically encoded amino acid or a synthetic amino acid. In some embodiments, the amino acid substitution occurs within the CDR region, and CDR1, CDR2 and / or CDR3 may contain at least one substitution (e.g., any 1, 2, 3, or 4 amino acids). In certain embodiments, the amino acid substitution occurs within the FR region, and FR1, FR2, FR3 and / or FR4 may contain at least one substitution (e.g., any 1, 2, 3, 4, 5, or 6 amino acids).

[0102] Furthermore, the aforementioned amino acid sequence insertions include amino-terminus and / or carboxyl-terminus fusions ranging in length from one residue to polypeptides containing 100 or more residues, as well as intrasequence insertions of single or multiple amino acid residues. An example of a terminal insertion is an antibody having an N-terminal methionyl residue. Other insertion variants of antibody molecules may include fusion of the antibody to the N or C terminus of a polypeptide or (e.g., ADEPT) an enzyme that extends the serum half-life of the antibody.

[0103] Furthermore, as described herein, Fc region variants can also be generated by introducing one or more amino acid modifications into the Fc region of a bispecific antibody (e.g., humanized anti-PD-L1×CD47 HCAb) that bispecifically binds to PD-L1 and CD47, including humanized anti-PD-L1×CD47 bsAb. Fc region variants may include human Fc region sequences (e.g., human IgG1, IgG2, IgG3, or IgG4 Fc) that include one or more amino acid modifications (e.g., substitutions) at amino acid positions.

[0104] In the present invention, the bispecific antibody, comprising the humanized anti-PD-L1×CD47 bsAb, which bispecifically binds to PD-L1 and CD47, can be linked, fused, conjugated (e.g., covalently or non-covalently), or otherwise associated with a diagnostic moiety or biocompatibility modifier. For example, peptides or polypeptides (e.g., biotoxins, biomarkers, tablet tags, etc.), proteins, polymers, nucleic acid molecules, small molecules, mimetic agents, synthetic drugs, inorganic molecules, organic molecules, or radioisotopes can be linked to or associated with them.

[0105] Furthermore, bispecific antibodies that bispecifically bind to PD-L1 and CD47, including the humanized anti-PD-L1×CD47 bsAb, can bind to or associate with diagnostic agents or detectable drugs, markers, or reporters, which may be biological molecules (e.g., peptides or nucleotides), small molecules, fluorophores, or radioisotopes. Labeled modulators may also be useful as part of clinical trial procedures for monitoring PD-L1 and / or CD47-associated diseases, such as the onset or progression of cancer, or for determining the efficacy of certain therapies (i.e., theragnosis) containing the antibodies disclosed herein, or for determining future treatment outcomes. These markers or reporters are also useful for purifying the antibodies disclosed herein.

[0106] Furthermore, the bispecific antibody, which contains the humanized anti-PD-L1×CD47 bsAb and bispecifically binds to PD-L1 and CD47, can be conjugated to immunomodulators, cytokines, cytotoxic agents, chemotherapeutic agents, diagnostic agents, antiviral agents, antimicrobial agents, or other drugs. In this regard, the present invention provides an antibody conjugate containing a bispecific antibody, which contains the humanized anti-PD-L1×CD47 bsAb of the present invention and bispecifically binds to PD-L1 and CD47, conjugated to an immunomodulator, cytokine, cytotoxic agent, chemotherapeutic agent, diagnostic agent, antiviral agent, antimicrobial agent, or other drug.

[0107] Furthermore, the present invention provides a nucleic acid molecule encoding a bispecific antibody that bispecifically binds to PD-L1 and CD47, including the humanized anti-PD-L1×CD47 bsAb disclosed herein, an expression vector containing the nucleic acid molecule, and host cells transformed with the expression vector.

[0108] Furthermore, the present invention also includes (a) the step of culturing host cells under conditions that enable the expression of bispecific antibodies; and (b) A method for producing a bispecific antibody is provided, which includes the step of recovering the expressed bispecific antibody.

[0109] In the present invention, the DNA encoding the bispecific antibodies that bispecifically bind to PD-L1 and CD47, including the humanized anti-PD-L1×CD47 bsAb disclosed herein, can be readily isolated and sequence-analyzed using conventional procedures (e.g., by using oligonucleotide probes that can specifically bind to the genes encoding the heavy and light chains of the antibody). Isolated and subcloned hybridoma cells (or colonies derived from phages or yeast) can serve as a preferred source of DNA. More specifically, the isolated DNA (which may be deformed) can be used to clone constant and variable region sequences for antibody production.

[0110] Exemplary methods include extracting RNA from selected cells, converting it to cDNA, and amplifying it by PCR using antibody-specific primers. Suitable primers are well known in the relevant field and are readily available from numerous commercial suppliers, as illustrated herein. To express recombinant human or non-human antibodies isolated by screening of combined libraries, the DNA encoding the antibodies is cloned into a recombinant expression vector and introduced into host cells, including mammalian cells, insect cells, plant cells, yeast, and bacteria. In some embodiments, the modulator is introduced and expressed in cells that do not otherwise produce constructive depression, such as monkey COS cells, NS0 cells, Chinese hamster ovary (CHO) cells, or myeloma cells.

[0111] In the present invention, the nucleic acid molecule resides within a vector and, where appropriate, includes a promoter for controlling the expression of the nucleic acid. The vector, used in the most common sense, includes any intermediate vehicle for the nucleic acid that enables the nucleic acid to be introduced into prokaryotic and / or eukaryotic cells and, where appropriate, incorporated into the genome. Such vectors are preferably capable of replication and / or expression within cells. The vector may include plasmids, phagemids, bacteriophages, or viral genomes. The plasmid is an extrachromosomal genetic material construct that can replicate independently of chromosomal DNA and usually exists in the form of a circular DNA double helix.

[0112] Expression vectors containing antibody coding sequences and appropriate retrieval and translation control signals can be constructed using methods widely known to those skilled in the art. These methods include, for example, in vitro recombinant DNA techniques, synthetic techniques, and in vivo genetic recombination.

[0113] In the present invention, the host cell or recombinant host cell refers to a cell into which an expression vector has been introduced. Recombinant host cells and host cells refer not only to specific target cells but also to their offspring. These offspring may not be identical to the parent cells because certain deformations may occur in subsequent generations due to mutations or environmental influences, but they still fall within the scope of the term “host cell” as used herein. Such cells may contain the vector.

[0114] Furthermore, by using molecular biology techniques and current protein expression methodologies recognized in the art, the antibodies disclosed herein can be produced in substantial quantities. More specifically, the nucleic acid molecules encoding the antibodies can be integrated into widely known and commercially available protein production systems, including various types of host cells, to provide desired pharmaceuticals in preclinical, clinical, or commercial quantities. In some embodiments, the nucleic acid molecules encoding the antibodies are manipulated within a vector or expression vector to enable efficient integration into selected host cells and subsequently provide high levels of antibody expression.

[0115] Preferably, nucleic acid molecules encoding antibodies disclosed herein and vectors containing these nucleic acid molecules can be used for the transformation of suitable mammalian, plant, bacterial, or yeast host cells, but prokaryotes can also be used. Transfusion can be carried out by known methods for introducing polynucleotides into host cells. Methods for introducing heterologous polynucleotides into mammalian cells are well known in the art and include techniques such as dextran-mediated transformation, calcium phosphate precipitation, polybrene-mediated transfusion, protoplast fusion, electroporation, encapsulation of polynucleotides into liposomes, and direct microinjection of DNA into the nucleus. Nucleic acid molecules can also be introduced into mammalian cells using viral vectors. Methods for transforming mammalian cells are widely known in the art. Methods for transforming plant cells are also widely known in the art and include, for example, Agrobacterium-mediated transformation, bioristic transformation, direct injection, electroporation, and viral transformation. Methods for transforming bacterial and yeast cells are also widely known in the art.

[0116] Various commercially available host expression vector systems can be used to express the antibodies disclosed herein. These host expression systems not only represent vehicles that express the desired coding sequence and are subsequently purified, but are also cells that can in vivo express the molecules of the present invention when transformed or transfected with the appropriate nucleotide coding sequence. These systems include microorganisms transformed with recombinant bacteriophage DNA, plasmid DNA, or cosmid DNA expression vectors containing the modulator coding sequence, e.g., bacteria (e.g., Escherichia coli, Bacillus subtilis, Streptomyces); yeast transfected with recombinant yeast expression vectors containing the modulator coding sequence (e.g., Saccharomyces, Pichia); insect cell lines infected with recombinant virus expression vectors containing the modulator coding sequence (e.g., baculovirus); recombinant virus expression vectors (e.g., cauliflower mosaic virus, CaMV; tobacco mosaic virus). Examples include, but are not limited to, plant cell lines (e.g., Nicotiana, Arabidopsis, duckweed, maize, wheat, potato, etc.) transfected with recombinant plasmid expression vectors containing modulator coding sequences (e.g., Ti plasmid), or mammalian cell lines (e.g., COS, CHO, BHK, 293, 3T3 cells) containing recombinant expression constructs containing mammalian genome-derived promoters (e.g., metallothionein promoter) or mammalian virus-derived promoters (e.g., adenovirus late promoter; vaccinia virus 7.5K promoter).

[0117] If the antibodies disclosed herein are produced by recombinant expression or other techniques disclosed herein, they may be purified by any method known in the art for purifying immunoglobulins, or more generally, by any other standard technique for protein purification.

[0118] The present invention also provides a pharmaceutical composition for the prevention or treatment of cancer, comprising, as an active ingredient, a bispecific antibody that bispecifically binds to PD-L1 and CD47, including the humanized anti-PD-L1×CD47 bsAb disclosed herein, or an antibody conjugate containing the bispecific antibody.

[0119] The present invention also provides a method for preventing or treating cancer, comprising the step of administering to an individual a pharmaceutical composition containing a bispecific antibody that bispecifically binds to PD-L1 and CD47, including the humanized anti-PD-L1×CD47 bsAb disclosed herein, or an effective amount of an antibody conjugate containing the bispecific antibody.

[0120] Furthermore, the present invention also provides the use of a bispecific antibody, or an antibody conjugate, comprising the humanized anti-PD-L1×CD47 bsAb disclosed herein, which bispecifically binds to PD-L1 and CD47, for the prevention or treatment of cancer.

[0121] The present invention further provides the use of a bispecific antibody, including the humanized anti-PD-L1×CD47 bsAb disclosed herein, that bispecifically binds to PD-L1 and CD47, or an antibody conjugate containing the bispecific antibody, for the production of a composition for the prevention or treatment of cancer.

[0122] In the present invention, the cancer is a cancer that requires the activation of T cells by blocking the activity of immune checkpoint proteins, and is selected from, but is not limited to, the group consisting of, for example, melanoma, lung cancer, liver cancer, glioblastoma, ovarian cancer, colorectal cancer, head and neck cancer, bladder cancer, renal cell carcinoma, gastric cancer, breast cancer, metastatic cancer, prostate cancer, pancreatic cancer, non-Hodgkin lymphoma, Hodgkin lymphoma, multiple myeloma, leukemia, lymphoma, myelodysplastic syndrome, acute lymphoblastic leukemia, acute myeloid leukemia, chronic lymphocytic leukemia, chronic myeloid leukemia, solitary myeloma, and aplastic anemia.

[0123] In the present invention, the description of the bispecific antibody that bispecifically binds to PD-L1 and CD47, including the humanized anti-PD-L1×CD47 bsAb disclosed herein, or the antibody conjugate containing the bispecific antibody, is the same as described above. Therefore, the specific description is provided for reference only, and below, only the specific composition of the pharmaceutical composition and its use will be described.

[0124] The pharmaceutical composition according to the present invention may include one or more (e.g., two or three) of the humanized anti-PD-L1×CD47 bsAb disclosed herein, bispecific antibodies that bispecifically bind to PD-L1 and CD47, or an antibody complex containing the bispecific antibody.

[0125] By administering the pharmaceutical composition according to the present invention to an individual, specifically a cancer patient, cancer can be prevented or treated.

[0126] Furthermore, the pharmaceutical compositions according to the present invention can be formulated as desired using techniques recognized in the relevant art, depending on the form of the antibody described herein, the intended delivery method, and various other variables. They can also be formulated to include suitable pharmaceutically acceptable carriers containing relatively inert substances such as excipients and adjuvants, which are well known in the art and facilitate the administration of the active compound or assist in the formulation of a pharmaceutically optimized formulation for delivery. For example, various pharmaceutically acceptable carriers, including vehicles, adjuvants, and diluents, are readily available from numerous commercial manufacturers. In addition, classifications of pharmaceutically acceptable adjuvants such as pH adjusters and buffers, tonicity adjusters, stabilizers, and wetting agents are also available. Specific non-limiting examples of such carriers include physiological saline, buffered physiological saline, dextrose, water, glycerol, ethanol, and combinations thereof.

[0127] Furthermore, the pharmaceutical compositions according to the present invention can be formulated for enteral, parenteral, or topical administration. In fact, all three of these formulations can be used simultaneously to achieve systemic delivery of the active ingredient. Excipients and formulations for parenteral and parenteral drug delivery are known in the relevant art. Formulations suitable for parenteral administration include aqueous solutions of the active compound in a water-soluble form, such as an aqueous solution of a water-soluble salt. In addition, a suitable active compound can also be administered suspended in an oily injection suspension. Suitable lipophilic solvents or vehicles include fatty oils such as sesame oil, or synthetic fatty acid esters such as ethyl oleate and triglycerides. The aqueous injection suspension may contain agents that increase the viscosity of the suspension, such as sodium carboxymethylcellulose, sorbitol, and / or dextran. Optionally, the suspension may also contain stabilizers. Furthermore, liposomes may be used to encapsulate the agonist for delivery to cells.

[0128] Formulations suitable for enteral administration include hard or soft gelatin capsules, tablets, coated tablets, elixirs, suspensions, syrups or inhalants, and controlled-release forms thereof.

[0129] In general, the antibodies disclosed herein can be administered in vivo to subjects requiring them via various routes, including, but not limited to, oral, intravenous, intra-arterial, subcutaneous, parenteral, intranasal, intramuscular, intracardiac, intraventricular, intratracheal, buccal, rectal, intraperitoneal, intradermal, topical, transdermal, and intrathecal administration, or via transplantation or inhalation. The appropriate formulation and route of administration can be selected depending on the intended use and therapeutic approach.

[0130] The pharmaceutical composition according to the present invention is administered in a pharmaceutically effective dose for the treatment or prevention of cancer. The pharmaceutically effective dose refers to the amount of antibody or antibody-containing pharmaceutical composition required to induce a biological or medical response in a subject, as determined by a physician or other clinician. Furthermore, the antibody or antibody-containing pharmaceutical composition may be administered multiple times at a specific frequency to achieve a therapeutic dose that has a preventive and / or therapeutic effect.

[0131] The pharmaceutically effective dose typically depends on factors such as the body weight of the subject being treated, their physical condition, the severity of the condition being treated, and the age of the subject being treated. Generally, the antibodies disclosed herein are administered in doses ranging from approximately 10 ng / kg body weight to approximately 100 mg / kg body weight, approximately 50 μg / kg body weight to approximately 5 mg / kg body weight, approximately 100 μg / kg body weight to approximately 10 mg / kg body weight, approximately 100 μg / kg body weight to approximately 20 mg / kg body weight, and 0.5 mg / kg body weight to approximately 20 mg / kg body weight. Furthermore, the antibodies may be administered in doses of at least approximately 100 μg / kg body weight, at least approximately 250 μg / kg body weight, at least approximately 750 μg / kg body weight, at least approximately 3 mg / kg body weight, at least approximately 5 mg / kg body weight, or at least approximately 10 mg / kg body weight, but are not limited to these.

[0132] Furthermore, the pharmaceutical composition according to the present invention is administered in doses of approximately 100 mg to approximately 10,000 mg, approximately 200 mg to approximately 9,000 mg, approximately 300 mg to approximately 8,000 mg, approximately 400 mg to approximately 7,000 mg, and 500 mg to 5,000 mg, but is not limited thereto.

[0133] The pharmaceutical compositions according to the present invention are typically administered to patients multiple times. Exemplary therapeutic regimens include administration every two weeks, once a month, or once every three to six months. For example, a patient may receive the antibody (e.g., as an intravenous formulation) every four weeks, or, for example, every 28 days. The frequency of administration can be adjusted according to the pharmacokinetic profile of the antibody in the patient. For example, an administration frequency of two weeks may be required if the antibody has a half-life of two weeks. Depending on the method, two or more antibodies with different binding specificities may be administered simultaneously, in which case the dose of each antibody administered will remain within the indicated range.

[0134] The dosage and frequency of administration are determined by the half-life of the antibody in the patient. Generally, human antibodies have the longest half-life, followed by humanized antibodies, chimeric antibodies, and non-human antibodies. The dosage and frequency of administration also differ depending on whether the treatment is prophylactic or therapeutic.

[0135] The duration of therapeutic treatment is influenced by the disease being treated, the patient's age and condition, the stage and type of the patient's disease, and the patient's tolerance to the treatment. Clinicians can carefully observe the effects of the therapy and make necessary adjustments as needed. When activators are used in combination, two or more therapeutic agents may be administered simultaneously or in any order. That is, the antibodies disclosed herein may be administered before the administration of the second therapeutic agent, simultaneously with the second therapeutic agent, or after the administration of the second therapeutic agent.

[0136] The present invention will be described in detail below with reference to examples and experimental examples.

[0137] However, please understand that the following examples and experimental cases are merely illustrative and are not intended to limit the scope of the present invention.

[0138] Example 1: Immunization and blood collection Human PD-L1 protein or human CD47 protein, which are immune antigens, were mixed with an immunosuppressant (GERBU) and administered to one alpaca three times by intramuscular injection. Immunization was performed in three separate doses, and 10 mL of blood was collected from each alpaca 14 days after the final immunization. The immune response was analyzed using ELISA. To determine whether antibodies were produced, the immune antigen was dispensed at a concentration of 1 μg / mL into 96-well microplates using coating buffer and coated overnight at 4°C. The 96-well microplates were washed three times with PBST and then blocked by treating with 5% skim milk at room temperature for 2 hours to inhibit nonspecific binding. After washing three times with PBST, serum samples obtained pre-immunization (day 0), 14 days post-immunization (day 14), 28 days post-immunization (day 28), and 42 days post-immunization (day 42) were processed at serial dilution concentrations. Subsequently, the 96-well microplate was washed five times with PBST, and then the goat anti-llama IgG HRP antibody was reacted at room temperature for 1 hour. The TMB reaction was then used to confirm whether the antibody had bound to the immune antigen.

[0139] Example 2: Library Construction and Evaluation An immunotherapy library was constructed by amplifying the gene encoding a single-domain antibody that binds to the immune antigen confirmed in Example 1. To construct the library, peripheral blood mononuclear cells (PBMCs) were isolated from blood using Ficol. From the isolated peripheral blood mononuclear cells (PBMCs), a gene fragment encoding a single-domain antibody was amplified using specific primers and cloned into a pComb3X vector. The size of the constructed immunotherapy library was 5.4 × 10⁶. 8 That was the case.

[0140] Example 3: Library Amplification The immunotherapy library prepared in Example 2 was transformed into the XL1-blue strain. The transformed XL1-blue strain was added to 10 mL of 2xYT medium containing 2% glucose and 100 μg / mL ampicillin, and cultured in a shaking incubator at 37°C. Optical density (OD) at 600 nm 600 The culture is continued until the ratio reaches 0.5, at which point M13K07 phage (Invitrogen) is divided into 1 × 10⁻¹⁶ cells.11 The culture was then incubated at 37°C for 30 minutes, followed by further incubation in a shaking incubator at 37°C and 200 rpm for 30 minutes. The culture medium was then centrifuged at room temperature at 4,000 rpm for 15 minutes, and the supernatant was removed. Next, 10 mL of 2x YT medium containing 100 μg / mL ampicillin and 50 μg / mL kanamycin was added, and the culture pellet was resuspended. The culture was incubated overnight at 30°C and 250 rpm in a shaking incubator. After this, it was centrifuged at 4°C and 4,000 rpm for 30 minutes. The supernatant was precipitated by PEG precipitation and centrifuged at 4°C and 12,000 rpm for 30 minutes. The resulting pellet was resuspended in PBS and centrifuged at 4°C and 13,000 rpm for 5 minutes. The supernatant was transferred to a new tube and stored at 4°C until use.

[0141] Example 4: Bio-panning To select single-domain antibodies specific to the immunoantigen, the antigen was dispensed into a 96-well microplate at a concentration of 5 μg / mL using coating buffer and coated overnight at 4°C. The library used to select single-domain antibodies (the library from Example 3) was dispensed into a 96-well microplate and reacted at room temperature for 30 minutes. Next, the library was transferred to a new well and reacted at room temperature for 30 minutes. This procedure was repeated four times to reduce the nonspecific binding of the library to the microplate wells. The library was then transferred to a 1.7 mL tube and stored at 4°C until use. The microplate coated with immunoantigen was washed five times with PBST and blocked with 5% skim milk at room temperature for 2 hours. After washing five times with PBST, the library with reduced nonspecific binding was treated with a binding solution (2.5% skim milk, 0.5% tween20) in 5 × 10⁻¹⁶ solutions. 12The phages were dispensed at a bilion / well concentration and incubated at room temperature for 30 minutes. The plates were then washed 10 times with washing solution (PBS, 0.5% tween20) and then 3 times with PBST. Single-domain antibodies specifically bound to the immunoantigen were selectively eluted by adding 5 μg of immunoantigen per well and culturing at room temperature at 500 rpm with shaking for 30 minutes. The eluted phages infected exponentially growing XL1-blue cells and streaked onto 2xYT agar. Panning for a second selection was repeated under the same conditions. Each single-phage clone generated on the agar was amplified and screened using FACS.

[0142] Example 5: Phage Screening To induce transient overexpression of an immune antigen in Expi-CHO cells, the gene encoding the immune antigen was inserted into a pCMV6-GFP vector, and a pCMV6-immune antigen-GFP plasmid was constructed. Expi-CHO cells were washed with DPBS and centrifuged at room temperature at 1,200 rpm for 3 minutes. The supernatant was removed, and the cells were resuspended in 2% skim milk and blocked at 4°C for 30 minutes. The cells were then centrifuged at room temperature at 1,200 rpm for 3 minutes, and the supernatant was removed. The cells were washed twice with DPBS and placed in 3 × 10⁶ wells of a 96-well microplate. 5 Cells were dispensed in 100 μL / well. A single cloned phage was added to each well and cultured at 4°C for 1 hour, then washed twice with DPBS. An antibody that specifically binds to the phage (M13 major coat protein Alexa Fluor 647 (Santa cruz)) was dispensed into the cells and cultured in the dark at 4°C for 30 minutes. The cells were washed twice with DPBS, resuspended in fresh DPBS, and analyzed by FACS using an Accuri C6 (BD) instrument. Clones screened by the FACS system were selected and sequenced.

[0143] Example 6: Expression and purification of a single-domain antibody fused with a human IgG Fc domain. 6-1: Expression and purification of monovalent single-domain antibodies fused with human IgG Fc domains

[0144] The clones selected in Example 5 were cloned into TGEX-Fc(IgG1) or TGEX-Fc(IgG4) expression vectors. To express a single-domain antibody fused with a human IgG Fc domain, Expi-CHO cells with a viability of 95-99% were counted, and 7 × 10⁶ cells were used. 6 Cells were added to 25 mL of culture medium (Expi-CHO expression medium (Gibco)). These cells were cultured overnight in a shaking incubator at 37°C, 125 rpm, and 8% CO2. Subsequently, 80 μL of ExpiFectamine® CHO Reagent (Gibco, 100033021) was mixed with 920 μL of OptiPRO® medium, and 20 μg of plasmid DNA encoding a single-domain antibody fused with a human IgG Fc domain was added, followed by 1 mL of OptiPRO® medium. The mixture was allowed to react at room temperature for 5 minutes before being added to the cultured cells. The cells were cultured for 20 hours in a shaking incubator at 125 rpm, and 8% CO2. To enhance the expression of single-domain antibodies fused with human IgG Fc domains, 150 μL of ExpiFectamine® CHO enhancer (Gibco) and 6 mL of ExpiCHO Feed (Gibco) were added, and the cells were cultured for 5 days in a shaking incubator at 32°C, 125 rpm, and 5% CO2. The cultured cells were centrifuged at 4°C and 4,000 rpm for 30 minutes, and the supernatant was filtered through a 0.2 μm syringe filter. The supernatant was loaded onto a HiTrap protein GHP column (GE Healthcare), washed with PBS, and the single-domain antibodies fused with human IgG Fc domains were eluted from the column using IgG elution buffer (Thermo). The eluted samples were neutralized with 1 M Tris-HCl (pH 9.0) and stored at 4°C until use.

[0145] 6-2: Expression and purification of a bivalent monodomain antibody fused with a human IgG4 Fc domain. The clones selected in Example 5 were linked using two G2S linkers (GGSGGS) to produce a bivalent single-domain antibody. The nucleotide encoding the bivalent single-domain antibody was obtained by gene synthesis (Macrogen, South Korea). For the expression and purification of the single-domain antibody fused with the human IgG4 Fc domain, the synthesized gene was cloned into a TGEX-Fc(IgG4) expression vector. Subsequently, expression and purification were performed in the same manner as described in Example 6-1.

[0146] Example 7: Expression and purification of a bispecific single-domain antibody fused with a human IgG4 Fc domain. From the clones selected in Example 5, one single-domain antibody specifically binding to PD-L1 and one single-domain antibody specifically binding to CD47 antigen were selected. The nucleotide sequences encoding each single-domain antibody were linked with a peptide linker, and two single-domain antibodies specifically binding to PD-L1 and one single-domain antibody specifically binding to CD47 were linked in sequence (anti-PD-L1 sdAb × anti-PD-L1 sdAb × anti-CD47 sdAb). The nucleotide sequence encoding this trivalent bispecific single-domain antibody was obtained by gene synthesis (Macrogen, South Korea). For the expression and purification of the bispecific single-domain antibody fused with a human IgG4 Fc domain, the synthesized gene was cloned into a TGEX-Fc(IgG4) expression vector. Expression and purification were performed in the same manner as described in Example 6-1.

[0147] Example 8: Expression and purification of a bispecific humanized single-domain antibody fused with a human IgG4 Fc domain. To humanize the trivalent bispecific single-domain antibody prepared in Example 7, its homology (identity) was evaluated by comparing it with the FR (framework region) sequence of the human germline VH fragment DP-47 (J Mol Biol. 1992 Oct 5;227(3):776-98). Among the clones selected in Example 5, the amino acid sequences of the PD-L1-specific single-domain antibody and the CD47-specific single-domain antibody (PDL1 Nb#01 and CD47 Nb#01) were analyzed. The first amino acid of the FR1 (framework region) sequence, glutamine (Q), was replaced with glutamic acid (E) to improve homology with the human antibody DP-47. As a result, trivalent PD-L1 and CD47 bispecific humanized single-domain antibodies (humanized anti-PD-L1 sdAb × humanized anti-PD-L1 sdAb × humanized anti-CD47 sdAb) were prepared. Amino acid substitutions were performed using the Q5 Site-Specific Mutagenesis Kit (NEB) and target mutagenesis primers. Expression and purification were performed using the same method as described in Example 6.

[0148] Example 9: Evaluation of the binding ability of single-domain antibodies to immune antigens using FACS The binding ability of the anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-CD47 HCAb (CD47 Nb-IgG4), anti-PD-L1 × CD47 trivalent HCAb (PPC Nb-IgG4), and humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Examples 6, 7, and 8 to immunoantigens was confirmed using FACS.

[0149] Specifically, CHO-K1_PD-L1 cell lines (CHO-K1 cells overexpressing PD-L1 antigen) or Expi-CHO_CD47 cell lines (Expi-CHO cells overexpressing CD47 antigen) were washed with DPBS and centrifuged at room temperature at 1,200 rpm for 3 minutes. The supernatant was removed, the cells were resuspended in 2% skim milk, and blocked at 4°C for 30 minutes. Subsequently, the cells were centrifuged at room temperature at 1,200 rpm for 3 minutes, the supernatant was removed, and the cells were washed twice with DPBS. Next, the cells were placed in each well at a rate of 3 × 10⁶5 Cells were dispensed at a density of 100 μl / well and treated with anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-CD47 HCAb (CD47 Nb-IgG4), anti-PD-L1 × CD47 trivalent HCAb (PPC Nb-IgG4), or humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) at different concentrations. Isotype control antibodies were used as a negative control group. After culturing cells at 4°C for 1 hour, they were washed twice with DPBS. Subsequently, they were treated with an antibody that specifically binds to the human Fc domain (anti-human IgG Fc APC antibody (Biolegend)) and cultured in the dark at 4°C for 30 minutes. After washing the cells two more times with DPBS, they were resuspended in 100 μL of DPBS and analyzed by FACS using an Accuri C6 (BD) instrument.

[0150] Example 10: Evaluation of the inhibitory activity of single-domain antibodies against immune antigen interactions using FACS The inhibitory activity of the anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-CD47 HCAb (CD47 Nb-IgG4), anti-PD-L1 × CD47 trivalent HCAb (PPC Nb-IgG4), and humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Examples 6, 7, and 8 against PD-1 / PD-L1 or CD47 / SIRPα interactions was evaluated.

[0151] Specifically, to evaluate anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-PD-L1 × CD47 trivalent HCAb (PPC Nb-IgG4), or humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) against PD-1 / PD-L1 interactions, CHO-K1_PD-L1 cell line (CHO-K1 cells that stably express the PD-L1 antigen) was placed in a 96-well microplate at a rate of 2 × 10⁶ cells per well. 5Cells were aliquoted and treated with 10 μg / mL human PD-1-His protein. Next, they were treated with anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-PD-L1 × CD47 trivalent HCAb (PPC Nb-IgG4), or humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) at different concentrations. Isotype control antibodies were used as a negative control group. Cells were cultured at 4°C for 1 hour and washed three times with DPBS. Subsequently, an antibody that specifically binds to the His antigen (goat anti-His APC) was added and reacted at 4°C in the dark for 30 minutes. Cells were washed three times with DPBS, resuspended in 100 μL of DPBS, and the amount of residual PD-1-His protein in CHO-K1_PD-L1 cells (CHO-K1 cells that stably express the PD-L1 antigen) was checked using an Accuri C6 (BD) instrument. The inhibitory activity of anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4), or humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) on PD-1 / PD-L1 interaction was evaluated by measuring residual PD-1-His protein.

[0152] Furthermore, to evaluate the effects of anti-CD47 sdAb (CD47 Nb-IgG4), anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4), or humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) on CD47 / SIRPα interaction, Expi-CHO_CD47 cells (Expi-CHO cells that stably express the CD47 antigen) were aliquoted into 96-well microplates at a rate of 2 × 10⁵ cells per well and treated with 10 μg / mL human SIRPα-His protein. Subsequently, the cells were treated with anti-CD47 sdAb (CD47 Nb-IgG4), anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4), or humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) at different concentrations, and isotype control antibodies were used as negative controls. Next, the amount of residual SIRPα-His protein in Expi-CHO_CD47 cells (Expi-CHO cells that stably express the CD47 antigen) was confirmed using the Accuri C6(BD) instrument in the same manner as described above. The inhibitory ability of anti-CD47 sdAb (CD47 Nb-IgG4), anti-PD-L1×CD47 HCAb (PPC Nb-IgG4), or humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) on CD47 / SIRPα interaction was evaluated by measuring the residual amount of SIRPα-His protein.

[0153] Example 11: Evaluation of affinity between single-domain antibody and immune antigen Using an Octet RED 96e (ForteBio) instrument, the affinity (K) between the anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-CD47 HCAb (CD47 Nb-IgG4), anti-PD-L1 × CD47 trivalent HCAb (PPC Nb-IgG4), or humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Examples 6, 7, and 8 and the immunoantigen protein was measured. d ) was measured.

[0154] Specifically, biosensor chips coated with anti-human Fc (Fortebio) were saturated with 5 μg / mL of anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-CD47 HCAb (CD47 Nb-IgG4), anti-CD47 HCAb (CD47 Nb-IgG4), anti-PD-L1 × CD47 trivalent HCAb (PPC Nb-IgG4), or humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) in 96-well microplates (Greiner) at a 1.5 nm level. PD-L1 antigen and CD47 antigen were serially diluted 2-fold from 10 to 400 nM using 1X kinetic buffer (ForteBio) and reacted with coated biosensors at 30°C and 1,000 rpm. Binding and dissociation reactions of the samples were analyzed over 200 and 400 seconds, respectively. The results were analyzed using a 1:1 interaction model (global fitting).

[0155] Furthermore, the affinity (K) of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 for heterologous immunogenicity antigen proteins d The values ​​were measured using an Octet RED 96e (ForteBio) instrument.

[0156] Specifically, a biosensor chip coated with anti-human Fc (Fortebio) was saturated with 20 μg / mL of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) at a 1.5 nm level in a 96-well microplate (Greiner). Human, mouse, and cynomolgus PD-L1 or CD47 antigens were serially diluted 2-fold from 12.5 to 200 nM using 1X kinetic buffer (ForteBio) and reacted with the coated biosensor while stirring at 30°C and 1,000 rpm. The binding and dissociation reactions of the samples were analyzed over 200 seconds and 400 seconds, respectively. The resulting data were analyzed using a 1:1 interaction model (global fitting) method.

[0157] Example 12: In vitro efficacy evaluation of a humanized single-domain antibody The in vitro efficacy of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 was evaluated in relation to T cell activity and mixed lymphocyte reactions. + This was evaluated by assessing the degree of T cell activation and the degree of macrophage-mediated phagocytosis activation.

[0158] 12-1: Evaluation of T cell activity of humanized single-domain antibodies The T cell activity efficacy of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 was evaluated by measuring the IL-2 concentration expressed by Jurkat cells. This was done by evaluating the inhibition of the interaction between CHO-K1 cells expressing overexpression of PD-L1 protein and the Jurkat cells expressing PD-1.

[0159] Specifically, CHO-K1 cells overexpressing PD-L1 protein were placed at a rate of 2 × 10⁶ per well. 4The cells were dispensed individually and cultured for 16 hours in an incubator maintained at 5% CO2. Afterward, the culture medium was removed, and humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC-Nb-IgG4) was added at various concentrations and incubated for 1 hour. Jurkat cells were divided into 5 × 10⁶ cells. 5 Cells were prepared at 100 μL and treated with PHA at a final concentration of 0.5 mg / mL. Subsequently, CHO-K1 cells treated with humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC-Nb-IgG4) were added and cultured for 48 hours. Next, the IL-2 concentration in the supernatant was measured by ELISA to evaluate the T cell activity efficacy of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC-Nb-IgG4).

[0160] 12-2: CD4 in mixed lymphocyte reactions of humanized single-domain antibodies + Evaluation of T cell activation CD4 in mixed lymphocyte reaction (MLR) analysis + T cell activation was evaluated using the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8.

[0161] Specifically, monocyte-induced dendritic cells (MoDCs) derived from donor A were placed in 5 × 10⁶ wells of a 96-well flat-bottom plate. 4 Dispense into cells / wells, and use CD4 of Ner B. + T cells are placed in each well containing MoDC in a 4x10 5 Cells were added to each well and mixed. Next, humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC-Nb-IgG4) was added at various concentrations, and the cells were cultured for 5 days (120 hours) in a 37°C incubator maintaining 5% CO2. Then, on day 3 of culture, supernatant was collected for IL-2 measurement, and on day 5, supernatant was collected from the same well for IFN-γ measurement. Subsequently, the concentrations of IL-2 and IFN-γ were measured using ELISA.

[0162] MoDCs used in mixed lymphocyte reaction analysis express PD-L1 and CD4 +T cells express PD-1. Inhibiting the interaction between PD-L1 and PD-1 with an antibody activates the suppressed activity of T cells. This is similar to how PD-L1 expressed on cancer cells binds to PD-1 expressed on T cells, suppressing immune activity and thus inhibiting the system by which cancer cells survive within the immune system. Therefore, the effectiveness of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) was evaluated by performing mixed lymphocyte reaction analysis and measuring the concentrations of IL-2 and IFN-γ, which are indicators of T cell activation.

[0163] 12-3: Evaluation of phagocytic activation of humanized single-domain antibodies by macrophages Phagocytosis is activated by specific antigen binding and the IgG4 Fc domain. Antigen-specific binding is influenced by the physical (intercellular distance), chemical (chemokine), and biological (cytokine) environment between target cells and effector cells. PD-L1 and CD47 antigen-specific humanized antibodies specifically bind to hPDL1 and hCD47 expressed on target cells, and their Fc domain junctions bind to Fc receptors on the surface of effector cells, promoting interaction between target cells and effector cells and inducing phagocytosis by macrophages.

[0164] Therefore, we evaluated the activation of phagocytosis by macrophages via CD47 antigen-specific binding of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8.

[0165] Specifically, 5 x 10 6THP-1 cells were seeded in a φ100 dish and cultured in a 5% CO2 incubator for 24 hours. After 24 hours, the cells were treated with 40 nM PMA for 24 hours, the culture medium was removed, and the cells were then stained with 1 μM crimson dye. After staining, the cells were replaced with fresh medium and allowed to rest for 48 hours. THP-1 cells detached with trypsin were mixed with target cells (Raji cells) stained with 3 μM CFSE in an 8:1 ratio and treated with various concentrations of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4). The cells were co-cultured for 4 hours. Phagocytosis was then evaluated using FACS.

[0166] 12-4: Evaluation of phagocytic activation of humanized single-domain antibodies by PBMC-derived macrophages. Monocytes isolated from human PBMCs were differentiated into macrophages and co-cultured with target cells (Raji_PDL1). The activation of phagocytosis by human monocyte-derived macrophages via the hPD-L1 and hCD47 antigen-specific binding of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 was evaluated.

[0167] Specifically, CD14 + Using the MACS kit, monocytes were isolated from PBMCs separated from whole blood via a Ficol gradient and stored in 2 × 10⁶ dishes in 100 mm dishes. 6 Macrophages were differentiated by treatment with 50 ng / mL M-CSF at a concentration of cells / mL for 6 days. After differentiation, macrophages were stained with 5 μM CFSE as effector cells, and target cells (Raji_PDL1) were stained with 10 μM deep red. Target cells were 2 × 10⁶ 4 Cells were seeded in a cell / well ratio, and the ratio of effector cells to target cells during co-culture was 8:1. Humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) was serially diluted from 1000 nM to 1 / 5 and evaluated at 15 concentration points. After 4 hours of co-culture, FACS analysis was performed to confirm the expression of deep red (APC) and CFSE (FITC). Phagocytosis was evaluated by identifying the APC-expressing cell population based on CFSE expression.

[0168] Example 13: In vitro safety evaluation of humanized single-domain antibodies The in vitro safety of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 was evaluated by confirming human RBC binding and hemagglutination reactions.

[0169] 13-1: Evaluation of RBC binding of humanized single-domain antibodies The human RBC binding that occurs when the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 specifically binds to the CD47 antigen was evaluated.

[0170] Specifically, to confirm the binding ability of human RBCs (red blood cells) to humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), RBCs were washed seven times with DPBS and then diluted to 12% (v / v) with DPBS. 50 μL of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was dispensed into a 96-well flat-bottom plate, followed by the addition of 50 μL of 12% (v / v) RBCs, and the cells were incubated at 4°C for 1 hour. After washing the cells with DPBS, they were treated with anti-human IgG4 using a secondary antibody and incubated at 4°C for 1 hour. Finally, after washing with DPBS, the degree of binding of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) to RBCs was evaluated by FACS.

[0171] 13-2: Evaluation of hemagglutination reactions using humanized single-domain antibodies The hemagglutination reaction induced by the specific binding of the purified humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) from Example 8 to the CD47 antigen was evaluated.

[0172] Specifically, to evaluate the hemagglutination reaction of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4), RBCs were washed seven times with DPBS and then diluted to 6% (v / v) with DPBS. 50 μL of 2X humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) was dispensed into a 96-well flat-bottom plate, followed by the addition of 50 μL of 6% (v / v) RBCs. After 1 hour of reaction at room temperature, the hemagglutination reaction was observed visually.

[0173] Example 14: Evaluation of the in vivo efficacy of a humanized single-domain antibody The in vivo efficacy of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 was evaluated by confirming its antitumor effect in C57BL / 6 mice and humanized NSG mice that had tumors formed by injection of tumor cell lines expressing human PD-L1 and human CD47.

[0174] 14-1: C57BL / 6 mouse model using B16F10_PD-L1_CD47 tumor 8 × 10¹⁶ C57BL / 6 mice aged 6-8 weeks were given B16F10 cell line overexpressing human PD-L1 and human CD47. 5 Cells were injected at a dose of 100 μL. Tumor formation was induced until the tumor reached a size of 3 × 3 mm in length (long axis) × width (short axis). Subsequently, humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) was administered intraperitoneally at a dose of 10 mg / kg at 2-day intervals for 7 doses, and the tumor size was measured. After the last intraperitoneal administration, the tumor volume was measured at 2-day intervals for 2 weeks. Tumor volume was calculated using the formula: length (long axis) × width (short axis) × height (short axis) / 2.

[0175] 14-2: Humanized NSG mouse model using Raji_PD-L1 tumor Female NSG mice aged 6-8 weeks were given 2 x 10⁶ doses. 7 Human PBMCs were injected intraperitoneally at a dose of 100 μL per cell to generate humanized NSG mice. Subsequently, 2 × 10⁶ human Raji(PD-L1) lymphoma cell lines that induce overexpression of human PD-L1 were used.6 Cells were injected at a rate of 100 μL. Tumor formation was induced until the tumor reached a size of 3 × 3 mm (length × width). Subsequently, humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) 10 mpk was administered intraperitoneally three times a week, and the tumor volume was measured. After the final intraperitoneal administration, the tumor volume was 2000 mm³. 3 The experiment was terminated when the target was reached.

[0176] 14-3: Humanized NSG mouse model using MDA-MB-231 tumor 6-8 week old female NOD scid NSG mice were given 2 × 10⁶ doses. 7 Humanized NSG mice were generated by intravenous injection of human PBMCs at a rate of cells / 100 μL. After confirming the expression of hCD45 and mCD45, 2 × 10⁶ human breast cancer cell line MDA-MB-231 was used. 6 Cells were administered subcutaneously at a dose of 100 μL into the right flank. Tumor formation was induced until the tumor reached a size of 3 × 3 mm (length × width). Subsequently, humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) was administered intraperitoneally at doses of 1, 5, and 10 mpk three times a week, and tumor volume was measured. Furthermore, when the tumor volume of humanized NSG mice reached 100 mm... 3 Once this point was reached, a Kaplan-Meier survival analysis was performed to predict the survival rate based on the mortality rate of the animals.

[0177] Experimental Example 1: Preparation of anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1) and in vitro characterization 1-1: Preparation of anti-PD-L1 HCAb (PDL1 Nb#01-IgG1)

[0178] Human PD-L1 antigen was used as the immunoantigen, and a single-domain antibody clone specific to the PD-L1 antigen was selected using the same method as in Example 5, and sequence analysis was performed. The amino acid sequences of the selected anti-PD-L1 sdAb (PDL1 Nb♯01) are shown in Tables 5 and 6 below.

[0179] Furthermore, using the selected anti-PD-L1 sdAb (PDL1 Nb♯01), a PD-L1-specific monovalent single-domain antibody containing a human IgG1Fc domain was expressed and purified in the same manner as in Example 6-1, and the purified monovalent single-domain antibody was named anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1). The amino acid sequence of anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1) containing a human IgG1Fc domain is shown in Table 7 below.

[0180] Furthermore, using the selected anti-PD-L1 sdAb (PDL1 Nb#01) clone, a PD-L1-specific bivalent single-domain antibody containing human IgG4 Fc was expressed and purified in the same manner as in Example 6-2, and the purified bivalent single-domain antibody was named anti-PD-L1 bivalent HCAb (PP Nb-IgG4). The amino acid sequence of anti-PD-L1 bivalent sdAb (PPNb) excluding the human IgG4 Fc domain is shown in Table 8 below, and the amino acid sequence of anti-PD-L1 bivalent HCAb (PP Nb-IgG4) containing the human IgG4 Fc domain is shown in Table 9 below.

[0181] [Table 5]

[0182] [Table 6]

[0183] [Table 7]

[0184] [Table 8]

[0185] [Table 9]

[0186] 1-2: Evaluation of antigen-binding ability and PD-1 / PD-L1 interaction inhibitory ability of anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1) and anti-PD-L1 bivalent HCAb (PP Nb-IgG4)

[0187] The antigen-binding ability (Figures 1A and 2A) and PD-1 / PD-L1 interaction inhibitory ability (Figures 1B and 2B) of the anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1) and anti-PD-L1 bivalent HCAb (PP Nb-IgG4) purified in Experimental Example 1 were evaluated by FACS using the methods described in Examples 9 and 10.

[0188] As a result, as shown in Figures 1A and 1B, the anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1) showed an antigen-binding ability of 23.31 nM (EC50) and PD-1 / PD-L1 interaction inhibitory activity (IC50: 4.60 nM) in CHO-K1_PD-L1 cells (CHO-K1 cells that express a constant amount of PD-L1 antigen).

[0189] Furthermore, as shown in Figures 2A and 2B, the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) showed an antigen-binding capacity (EC50) of 1.93 nM and an inhibitory capacity (IC50) of 2.86 nM in CHO-K1_PD-L1 cells (CHO-K1 cells that consistently express the PD-L1 antigen).

[0190] 1-3: Evaluation of affinity between anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1) and anti-PD-L1 bivalent HCAb (PP Nb-IgG4) and immunoantigens.

[0191] The affinity of the anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1) and anti-PD-L1 bivalent HCAb (PP Nb-IgG4) purified in Experimental Example 1 for the PD-L1 antigen was evaluated using the method described in Example 10.

[0192] As a result, as shown in Table 10, the anti-PD-L1 HCAb (PDL1 Nb♯01-IgG1) showed an antigen affinity of 7.08 nM for the PD-L1 antigen, while the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) showed an antigen affinity of 4.58 nM for the PD-L1 antigen.

[0193] [Table 10]

[0194] Experimental Example 2: Preparation of anti-CD47 HCAb (CD47 Nb-IgG4) and in vitro characterization 2-1: Preparation of anti-CD47 HCAb (CD47 Nb-IgG4) Human CD47 antigen was used as the immunogen antigen, and a single-domain antibody clone specific to the CD47 antigen was selected using the same method as in Example 5, and sequence analysis was performed. The amino acid sequences of the selected anti-CD47 sdAb (CD47 Nb#01) are shown in Tables 11 and 12 below.

[0195] Furthermore, using the selected anti-CD47 sdAb (CD47_Nb_#01), a CD47-specific single-domain antibody containing the human IgG4 Fc domain was expressed and purified in the same manner as in Example 6-1, and the purified single-domain antibody was named anti-CD47 HCAb (CD47 Nb-IgG4). The amino acid sequence of anti-CD47 HCAb (CD47 Nb-IgG4) containing the human IgG4 Fc domain is shown in Table 13 below.

[0196] [Table 11]

[0197] [Table 12]

[0198] [Table 13]

[0199] 2-2: Evaluation of the antigen-binding ability and CD47 / SIRPα interaction inhibitory ability of anti-CD47 HCAb (CD47 Nb-IgG4)

[0200] The antigen-binding ability (Figure 3A) and CD47 / SIRPα interaction inhibitory ability (Figure 3B) of the anti-CD47 HCAb (CD47 Nb-IgG4) purified in Experimental Example 2-1 were evaluated by FACS in the same manner as described in Examples 9 and 10.

[0201] As a result, as shown in Figures 3A and 3B, the anti-CD47 HCAb (CD47 Nb-IgG4) showed an antigen-binding ability (EC50) of 3.78 nM and an inhibitory ability (IC50) of 7.28 nM against CD47 / SIRPα interaction in Expi-CHO_CD47 cells (Expi-CHO cells that consistently express the CD47 antigen).

[0202] 2-3: Evaluation of the affinity of anti-CD47 HCAb (CD47 Nb-IgG4) for immune antigens

[0203] The affinity between the anti-CD47 HCAb (CD47 Nb-IgG4) purified in Experimental Example 2-1 and the CD47 antigen was evaluated using the same method as in Example 11.

[0204] As a result, as shown in Table 14, the anti-CD47 HCAb (CD47 Nb-IgG4) showed an antigen affinity of 2.78 nM for the CD47 antigen.

[0205] [Table 14]

[0206] Experimental Example 3: Preparation of anti-PD-L1 × CD47 trivalent HCAb (PPC Nb-IgG4) and in vitro characterization 3-1: Preparation of anti-PD-L1 × CD47 trivalent HCAb (PPC Nb-IgG4) Using the clone selected in Experimental Example 1-1, a bispecific single-domain antibody targeting both PD-L1 and CD47 antigens as an immunizing antigen and containing the human IgG4 Fc domain was prepared. Using the nucleotide sequence encoding the anti-PD-L1 sdAb (PDL1 Nb#01) selected in Experimental Example 1-1 and the nucleotide sequence encoding the anti-CD47 sdAb (CD47 Nb#01) selected in Experimental Example 2-1, genes were synthesized in the same manner as in Example 7, cloned into a TGEX-Fc(IgG4) expression vector, and then expressed and purified. The purified single-domain antibody was named anti-PD-L1×CD47 HCAb (PPC Nb-IgG4).

[0207] The amino acid sequence of the PD-L1×CD47 trivalent sdAb excluding the human IgG4 Fc domain is shown in Table 15 below, and the amino acid sequence of the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) containing the human IgG4 Fc domain is shown in Table 16 below.

[0208] [Table 15]

[0209] [Table 16]

[0210] 3-2: Evaluation of the immunizing antigen binding ability of anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4)

[0211] The antigen binding ability of the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) purified in Experimental Example 3-1 to the PD-L1 antigen and CD47 was confirmed by FACS in the same manner as in Example 9.

[0212] As a result, as shown in FIGS. 4A and 4B, the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) showed an antigen-binding ability (EC50) of 13.33 nM in CHO-K1_PD-L1 cells (CHO-K1 cells that constantly express the PD-L1 antigen) and an antigen-binding ability (EC50) of 18.80 nM in Expi-CHO_CD47 cells (Expi-CHO cells that constantly express the CD47 antigen).

[0213] 3-3: Evaluation of the ability of anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) to inhibit PD-1 / PD-L1 and CD47 / SIRPα interactions The inhibitory ability of the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) purified in Experimental Example 3-1 against PD-1 / PD-L1 and CD47 / SIRPα interactions was evaluated by FACS in the same manner as in Example 10.

[0214] As a result, as shown in FIGS. 5A and 5B, the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) showed an inhibitory ability (IC50) of 9.15 nM against the PD-1 / PD-L1 interaction and an inhibitory ability (IC50) of 22.44 nM against the CD47 / SIRPα interaction.

[0215] 3-4: Evaluation of the affinity of anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) for the immunizing antigen The affinity of the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) purified in Experimental Example 3-1 for the PD-L1 antigen and the CD47 antigen was evaluated in the same manner as in Example 11.

[0216] As a result, as shown in Table 17, the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) showed excellent affinities of 6.84 nM for the PD-L1 antigen and 3.62 nM for the CD47 antigen.

[0217]

Table 17

[0218] Experimental Example 4: Preparation of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) and in vitro characterization

[0219] 4-1: Preparation of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4)

[0220] The anti-PD-L1×CD47 trivalent HCAb (PPC-Nb-IgG4) prepared in Experimental Example 3 was humanized by site-directed mutagenesis using the same method as in Example 7, and named humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4).

[0221] As shown in Table 18, the homology of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) to the FR (framework region) sequence of the human germline VH segment DP-47 was compared. The homology of the FR1, FR2, FR3, and FR4 sequences of the humanized anti-PD-L1 sdAb (hPDL1 Nb#01) within the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was confirmed to be 98.4% on average, and the homology of the FR1, FR2, FR3, and FR4 sequences of the humanized anti-CD47 sdAb (hCD47 Nb#01) within the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was confirmed to be 89.6% on average.

[0222] Table 19 below shows the amino acid sequences of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb) obtained by removing the human IgG4 Fc domain from the aforementioned humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), as well as the amino acid sequences of the humanized anti-PD-L1 sdAb (hPDL1 Nb#01) and the humanized anti-CD47 sdAb site (hCD47 Nb#01). Table 20 below shows the amino acid sequence of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) containing the human IgG4 Fc domain.

[0223] [Table 18]

[0224]

Table 19

[0225]

Table 20

[0226] Also, for comparison with the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) prepared in the above Experimental Examples 1-3 was humanized by substituting amino acids by site-specific mutagenesis in the same manner as above, and named humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4). Furthermore, the anti-CD47 HCAb (CD47 Nb-IgG4) prepared in Experimental Example 2 was humanized by substituting amino acids by site-specific mutagenesis in the same manner as above, and named humanized anti-CD47 HCAb (hCD47 Nb-IgG4). The amino acid sequences of these antibodies are shown in Table 21 below.

[0227]

Table 21

[0228] 4-2: Evaluation of the Immunoantigen Binding Ability of Humanized Anti-PD-L1×CD47 Trivalent HCAb (hPPC Nb-IgG4)

[0229] In the same manner as in Example 9 above, the PD-L1 antigen and CD47 antigen binding abilities of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 were confirmed using FACS.

[0230] As a result, as shown in Figures 6A and 6b, humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) showed an antigen-binding capacity (EC50) of 11.48 nM in CHO-K1_PD-L1 cells (CHO-K1 cells that consistently express the PD-L1 antigen) and an antigen-binding capacity (EC50) of 22.68 nM in Expi-CHO_CD47 cells (Expi-CHO cells that consistently express the CD47 antigen).

[0231] 4-3: Evaluation of the inhibitory activity of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) on PD-1 / PD-L1 and CD47 / SIRPα interactions.

[0232] Using the same method as in Example 10, the inhibitory activity of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 against PD-1 / PD-L1 and CD47 / SIRPα interactions was evaluated using FACS.

[0233] As a result, as shown in Figures 7A and 7B, the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) showed an inhibitory activity (IC50) of 6.83 nM against PD-1 / PD-L1 interaction and an inhibitory activity (IC50) of 33.04 nM against CD47 / SIRPα interaction.

[0234] 4-4: Evaluation of affinity between humanized anti-PD-L1 x CD47 trivalent HCAb (hPPC Nb-IgG4) and immune antigens

[0235] The affinity of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 to the PD-L1 antigen and the CD47 antigen was evaluated using the same method as in Example 11.

[0236] As a result, as shown in Table 21, it was confirmed that the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) showed excellent affinity for the PD-L1 antigen (6.77 nM) and the CD47 antigen (3.34 nM).

[0237] [Table 22]

[0238] 4-5: Evaluation of affinity between humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) and interspecies immune antigens.

[0239] The affinity of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 for interspecies PD-L1 antigen and CD47 antigen was evaluated using the same method as in Example 11.

[0240] As a result, as shown in Table 23, the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC-Nb-IgG4) showed antigen affinities of 6.77 nM and 3.34 nM against human PD-L1 and human CD47 antigens, respectively, and 8.17 nM and 6.91 nM against monkey PD-L1 and monkey CD47 antigens, respectively. No antigen affinity was observed against mouse PD-L1 and mouse CD47 antigens.

[0241] [Table 23]

[0242] Experimental Example 5: In vitro evaluation of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4)

[0243] 5-1: Evaluation of T cell activity of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4)

[0244] The T cell activation of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 was evaluated by measuring the IL-2 concentration expressed from jurkat cells. This was done by inhibiting the interaction between PD-L1-expressing CHO-K1 cells and PD-1-expressing jurkat cells, using the same method as in Example 12-1.

[0245] As a result, as shown in Figure 8, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) showed efficacy of 1.47 nM (IC50).

[0246] 5-2: CD4 in mixed lymphocyte reactions of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) + Evaluation of T cell activation

[0247] CD4 + T cell activation was evaluated by performing mixed lymphocyte reaction (MLR) analysis in the same manner as in Example 12-2. As a comparative example, humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4), humanized anti-CD47 HCAb (hCD47 Nb-IgG4) prepared in Experimental Example 4-1, and a 1:1 mixture of these antibodies were used.

[0248] As a result, as shown in Figures 9A and 9B, humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) is CD4 + The antibody demonstrated efficacy of 0.08 (EC50) and 0.18 nM (EC50) in T cell IL-2 and IFN-γ production, respectively. Furthermore, it was confirmed that this antibody showed efficacy equivalent to or better than humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4), humanized anti-CD47 HCAb (hCD47 Nb-IgG4), and combination therapies thereof.

[0249] 5-3: Evaluation of phagocytic activation of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) by macrophages.

[0250] The degree of phagocytosis by macrophages was evaluated by selectively binding the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 to CD47 antigen-expressing prey cells (Raji cells) using THP-1 cells differentiated into macrophages by PMA, in the same manner as in Example 12-3.

[0251] As a result, as shown in Figure 10A, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) treated group showed significantly superior phagocytic effects compared to the control group.

[0252] The degree of phagocytic activity by macrophages was evaluated using the same method as in Example 12-4, by selectively binding the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 to PD-L1 and CD47 antigen-expressing prey cells (Raji_PDL1) using macrophages differentiated from monocytes isolated from human PBMCs. For comparative examples, the humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4), humanized anti-CD47 HCAb (hCD47 Nb-IgG4) prepared in Experimental Example 4-1, and a 1:1 mixture thereof were used.

[0253] As a result, as shown in Figure 10B, the group treated with humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) effectively induced phagocytosis at a lower concentration (0.005 pM) compared to the group treated with humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4) alone (0.522 pM), the group treated with humanized anti-CD47 HCAb (hCD47 Nb-IgG4) alone (0.181 pM), and the group treated with a combination of these (0.159 pM).

[0254] Experimental Example 6: In vitro safety evaluation of humanized anti-PD-L1 x CD47 trivalent HCAb (hPPC Nb-IgG4)

[0255] 6-1: Evaluation of the human RBC binding ability of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4)

[0256] The human RBC binding ability of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 was evaluated using the same method as in Example 13-1.

[0257] As a result, as shown in Figure 11, humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4), unlike the CD47 monoclonal antibody used as a benign control, did not show RBC binding at a concentration of 3 μM.

[0258] 6-2: Evaluation of the hemagglutination reaction of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4)

[0259] The hemagglutination reaction of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 was evaluated using the same method as in Example 13-2.

[0260] As a result, as shown in Figure 12, unlike the CD47 monoclonal antibody used as a benign control, humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) did not show hemagglutination at a concentration of 3 μM.

[0261] Experimental Example 7: Evaluation of the in vivo efficacy of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4).

[0262] The antitumor effect of the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 was evaluated in the same manner as in Example 14 using a C57BL / 6 mouse model in which tumors were formed by injecting tumor cells expressing pre-PD-L1 and CD47 antigens (b16F10 cells), a humanized NSG mouse model in which tumors were formed by injecting tumor cells expressing PD-L1 and CD47 antigens (Raji_PD-L1 cells), and a humanized NSG mouse model in which tumors were formed by injecting the human breast cancer cell line MDA-MB-231.

[0263] As a result, as shown in Figure 13, humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) showed an antitumor effect of approximately 73.61% compared to the negative control group (isotype) in a C57BL / 6 mouse model with tumors.

[0264] Furthermore, as shown in Figure 14, the humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) showed an antitumor effect of approximately 73.41% compared to the negative control group (Isotype) in a humanized NSG mouse model with tumors.

[0265] Furthermore, as shown in Figures 15A and 15B, humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) showed an antitumor effect of approximately 81.98% in a humanized NSG mouse model with tumors compared to the negative control group (isotype). In addition, the expected survival rate in the group administered 10 mpk of humanized anti-PD-L1 × CD47 trivalent HCAb (hPPC Nb-IgG4) showed a significant difference compared to the negative control group (isotype). [Industrial applicability]

[0266] The humanized single-domain antibody according to the present invention, which bispecifically binds to the immune checkpoint proteins PD-L1 and CD47, exhibits efficacy both in vitro and in vivo, and can therefore be effectively used as an immune checkpoint inhibitor in immunotherapy for cancer.

Claims

1. A first humanized single-domain antibody (first hsdAb) that specifically binds to PD-L1 or its antigen-binding fragment; and A second humanized single-domain antibody (second hsdAb) that specifically binds to CD47, or its antigen-binding fragment; Includes, The first hsdAb or its antigen-binding fragment, CDR1 consisting of the amino acid sequence represented by Sequence ID No. 2; CDR2 consisting of the amino acid sequence represented by Sequence ID No. 3; and CDR3 consisting of the amino acid sequence represented by Sequence ID No. 4; Includes, The second hsdAb or its antigen-binding fragment is CDR1 consisting of the amino acid sequence represented by Sequence ID No. 9; CDR2 consisting of the amino acid sequence represented by Sequence ID No. 10; and CDR3 consisting of the amino acid sequence represented by Sequence ID No. 11; A bispecific antibody that contains and bispecifically binds to PD-L1 and CD47.

2. The first hsdAb or its antigen-binding fragment comprises the following framework region, and the bispecific antibody according to claim 1 is bispecifically bound to PD-L1 and CD47: FR1 consists of the amino acid sequence represented by Sequence ID No. 19; FR2 consists of the amino acid sequence represented by Sequence ID No. 20; FR3 consisting of the amino acid sequence represented by SEQ ID NO: 21; and FR4 consisting of the amino acid sequence shown in Sequence ID No. 22

3. The bispecific antibody according to claim 1, wherein the second hsdAb or its antigen-binding fragment comprises the following framework region, which is bispecifically bound to PD-L1 and CD47: FR1 consists of the amino acid sequence represented by SEQ ID NO: 23; FR2 consists of the amino acid sequence represented by Sequence ID No. 24; FR3 consisting of the amino acid sequence represented by Sequence ID No. 25; and FR4 consisting of the amino acid sequence represented by SEQ ID NO: 26

4. A bispecific antibody according to claim 1, wherein the first hsdAb comprises the amino acid sequence represented by SEQ ID NO: 16, and bispecifically binds to PD-L1 and CD47.

5. A bispecific antibody according to claim 1, wherein the second hsdAb comprises the amino acid sequence represented by SEQ ID NO: 17 and bispecifically binds to PD-L1 and CD47.

6. A bispecific antibody according to claim 1, wherein the first hsdAb or its antigen-binding fragment is fused with a second hsdAb or its antigen-binding fragment via a peptide linker, thereby bispecifically binding to PD-L1 and CD47.

7. A bispecific antibody according to claim 6, wherein the first hsdAb and the second hsdAb fused via the peptide linker have an amino acid sequence represented by SEQ ID NO: 15, and bispecifically bind to PD-L1 and CD47.

8. A bispecific antibody according to claim 1, wherein an Fc fragment is fused to the bispecific antibody, and the bispecific antibody bispecifically binds to PD-L1 and CD47.

9. A bispecific antibody according to claim 8, wherein the Fc fragment is an Fc fragment of human IgG1, IgG2, IgG3, or IgG4, that bispecifically binds to PD-L1 and CD47.

10. The bispecific antibody according to claim 8, wherein the bispecific antibody comprises the amino acid sequence represented by Sequence ID No. 18 and bispecifically binds to PD-L1 and CD47.

11. A bispecific antibody according to claim 1 that is conjugated to an immunomodulator, cytokine, cytotoxic agent, chemotherapeutic agent, diagnostic agent, antiviral agent, antimicrobial agent, or drug, and bispecifically conjugates to PD-L1 and CD47.

12. An antibody conjugate comprising an immunomodulator, cytokine, cytotoxic agent, chemotherapeutic agent, diagnostic agent, antiviral agent, antimicrobial agent, or drug to which a bispecific antibody that bispecifically binds to PD-L1 and CD47 as described in claim 1 is conjugated.

13. A nucleic acid molecule encoding a bispecific antibody that bispecifically binds to PD-L1 and CD47 as described in claim 1.

14. An expression vector comprising the nucleic acid molecule described in Claim 13.

15. A host cell transformed with the expression vector described in Claim 14.

16. (a) A step of culturing the host cells described in claim 15 under conditions that enable the expression of a bispecific antibody; and (b) A step of recovering the expressed bispecific antibody; A method for producing a bispecific antibody that bispecifically binds to PD-L1 and CD47, which include [the specified substance].

17. A pharmaceutical composition for the prevention or treatment of cancer, comprising as an active ingredient a bispecific antibody that bispecifically binds to PD-L1 and CD47 as described in claim 1, or an antibody complex as described in claim 12.

18. The pharmaceutical composition according to claim 17, characterized in that the cancer is selected from the group consisting of melanoma, lung cancer, liver cancer, glioblastoma, ovarian cancer, colorectal cancer, head and neck cancer, bladder cancer, renal cell carcinoma, gastric cancer, breast cancer, metastatic cancer, prostate cancer, pancreatic cancer, non-Hodgkin lymphoma, Hodgkin lymphoma, multiple myeloma, leukemia, lymphoma, myelodysplastic syndrome, acute lymphoblastic leukemia, acute myeloid leukemia, chronic lymphocytic leukemia, chronic myeloid leukemia, solitary myeloma, and aplastic anemia.

19. The pharmaceutical composition according to claim 17, characterized in that the pharmaceutical composition further comprises a pharmaceutically acceptable carrier.

Images