Bispecific humanized single domain antibodies against PD-L1 and CD47 and uses thereof

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] The present invention relates to a bispecific humanized single domain antibody against the immune checkpoint proteins PD-L1 and CD47 and its use.

Background Art

[0002] In recent years, the therapeutic effect of immunotherapy for cancer using the newly developed human immune system has been demonstrated, and there is a shift from conventional cancer treatment methods using chemotherapeutic agents and targeted therapeutic agents to immunotherapy for cancer using immunotherapeutic agents.

[0003] Generally, immune cells of cancer patients acquire tolerance to cancer antigens and remain functionally suppressed while recognizing cancer cells, so they cannot effectively eliminate cancer cells. The core of immunotherapy is to awaken and activate immune cells in a tolerant state to induce the destruction of cancer cells. Types of immunotherapy include cytokine therapy such as IFN-γ and IL-2 for treatment, dendritic cell-based cancer vaccines, T cell-based cell therapy, immune checkpoint inhibitors (ICI) that block immune checkpoint proteins, and the like. These treatment methods are generally called cancer immunotherapy. Among them, the cancer immunotherapy developed by international pharmaceutical companies competing with each other is immune checkpoint inhibitors.

[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, reduce the proliferation, cytokine secretion, and cytotoxicity of T cells, and suppress the excessive activity of T cells, so they are also called co-inhibitory molecules. In particular, T cells express co-inhibitory receptors such as CTLA-4 and PD-1, and bind to their respective ligands, B7.1 / 2 and PD-L1, to suppress the activity of T cells. On the other hand, PD-L1 expressed on cancer cells plays an important role as a molecular shield, inactivates cancer-specific T cells to induce apoptosis, protects cancer cells from immune attacks mediated by T cells, and contributes to the immune evasion mechanism of cancer. Furthermore, it has been reported that cancer patients with ectopic expression of PD-L1 on cancer cells have a worse prognosis than those without expression.

[0005] Immune checkpoint inhibitors are drugs that block the activity of immune checkpoint proteins involved in such 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, nivolumab (Opdivo) and pembrolizumab (Keytruda), both PD-1 blockers, obtained FDA approval for metastatic melanoma. Then, in 2016, atezolizumab (Tecentriq), a PD-L1, was approved for bladder cancer, avelumab (Bavencio) was approved for metastatic Merkel cell carcinoma, a type of skin cancer, in 2017, and durvalumab (Imfinzi) obtained approval for bladder cancer. In 2018, cemiplimab (Libtayo), a PD-1 inhibitor, received FDA approval as a treatment for cutaneous squamous cell carcinoma. Currently, these drugs are expanding their therapeutic indications and are obtaining FDA approval for an increasing number of cancer types. As of 2019, six PD-1 / PD-L1 inhibitors have obtained FDA approval for a total of 18 types of cancer. Furthermore, other immune regulatory 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 bias towards T cells in target discovery, recently, the development of inhibitors targeting immune checkpoint proteins expressed in myeloid cells such as macrophages and dendritic cells has been focused on. Among these, CSF1R, CD47, TLR7, etc. are emerging 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 cancers.

[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 sends 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 present inventors developed humanized antibodies targeting the immune checkpoint proteins PD-L1 and CD47 as immune checkpoint inhibitors and arrived at the present 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

Non-Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0015] An 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 uses thereof.

Means for Solving the Problems

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

[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] Also, 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 hsdAb or the second hsdAb may include a heavy-chain variable domain framework sequence having at least 95% sequence identity with the sequence of SEQ ID NO: 16, or may include a heavy-chain variable domain framework sequence having at least 85% sequence identity with the sequence of SEQ ID NO: 17.

[0020] In addition, the first hsdAb or the second hsdAb may include a heavy-chain variable domain framework sequence having 100% sequence identity with the sequence of SEQ ID NO: 16 or 17. Specifically, the first hsdAb or its antigen-binding fragment and / or the second hsdAb or its antigen-binding fragment may include: (1) FR1 consisting of the amino acid sequence represented by any one of SEQ ID NO: 19 or 23; (2) FR2 consisting of the amino acid sequence represented by any one of SEQ ID NO: 20 or 24; (3) FR3 consisting of the amino acid sequence represented by any one of SEQ ID NO: 21 or 25; and (4) a heavy-chain variable domain framework sequence of FR4 consisting of the amino acid sequence represented by any one of SEQ ID NO: 22 or 26. More specifically, the first hsdAb or its antigen-binding fragment may include FR1 consisting of the amino acid sequence represented by SEQ ID NO: 19; FR2 consisting 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 a heavy-chain variable domain framework sequence of FR4 consisting of the amino acid sequence represented by SEQ ID NO: 22, and the second hsdAb or its antigen-binding fragment may include FR1 consisting of the amino acid sequence represented by SEQ ID NO: 23; FR2 consisting 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 a heavy-chain variable domain framework sequence of FR4 consisting of the amino acid sequence represented by SEQ ID NO: 26. In a specific embodiment, the first hsdAb includes the amino acid sequence represented by SEQ ID NO: 16, and the second hsdAb includes 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 have a higher valence. In addition, the first hsdAb or its antigen-binding fragments may be fused to each other via a peptide linker. Furthermore, it may be fused to the second hsdAb or its antigen-binding fragment.

[0022] In one embodiment of the present invention, the 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 specific embodiment, it contains the amino acid sequence represented by SEQ ID NO: 15.

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

[0024] In one embodiment of the present invention, there is provided a bispecific and multivalent antibody consisting only of a heavy chain (HCAb) containing two or more copies of the first hsdAb or its antigen-binding fragment and / or the second hsdAb or its antigen-binding fragment. The HCAb may have a valence of divalent, trivalent, tetravalent, or higher, and may be an HCAb fused with an Fc fragment. In a specific embodiment, the HCAb may consist of the amino acid sequence represented by SEQ ID NO: 18.

[0025] In one embodiment of the present invention, the sdAb may be fused to 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 contains at least one or more amino acid substitutions, and the at least one or more amino acid substitutions are conservative substitutions and may be substitutions with non-genetically encoded amino acids or synthetic amino acids of amino acids.

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

[0028] The present invention also provides a nucleic acid molecule encoding a bispecific antibody that binds specifically to both the PD-L1 and CD47.

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

[0030] The present invention also provides a host cell transformed with the expression vector.

[0031] Furthermore, the present invention provides (a) a step of culturing a host cell under conditions that enable the expression of a bispecific antibody; and (b) a step of recovering the expressed bispecific antibody; and provides a method for producing a bispecific antibody that binds specifically to both PD-L1 and CD47, which includes the above steps.

[0032] The present invention also provides a pharmaceutical composition for preventing or treating cancer, which contains as an active ingredient the bispecific antibody that binds specifically to both the PD-L1 and CD47 or the antibody complex; a method for preventing or treating cancer, which includes a step of administering the bispecific antibody that binds specifically to both the PD-L1 and CD47 or the antibody complex to an individual; the use of the bispecific antibody that binds specifically to both the PD-L1 and CD47 or the antibody complex for use in a pharmaceutical composition for preventing or treating cancer; and the use of the bispecific antibody that binds specifically to both the PD-L1 and CD47 or the antibody complex for manufacturing a pharmaceutical composition for preventing or treating 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 cancer, 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 contain a pharmaceutically acceptable carrier.

Advantages of the Invention

[0035] In the present invention, a humanized single-domain antibody that binds specifically to both PD-L1 and CD47, which are immune checkpoint proteins, was developed, and its efficacy in vitro and in vivo was confirmed. Therefore, the bispecific humanized single-domain antibody can be usefully used in immune anti-cancer therapy as an immune checkpoint inhibitor.

Brief Description of the Drawings

[0036]

Figure 1A

Figure 1B

Figure 2A

Figure 2B

Figure 3A

Figure 3B

Figure 4A

Figure 4B

Figure 5A

Figure 5B

Figure 6A

Figure 6B

Figure 7A

Figure 7B

Figure 8

Figure 9A

Figure 9B

Figure 10A

Figure 10B

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15A

Figure 15B

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 present invention. These embodiments are provided to fully explain the present invention to those skilled in the art. Therefore, the 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 the present invention, the term "epitope" refers to a protein determinant that can specifically bind to an antibody. Epitopes generally consist of groups on the surface of a chemically active molecule such as an amino acid or sugar side chain, and generally have specific three-dimensional structural features and specific charge features.

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

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

[0041] The term "subject" or "solid" refers to mammals including, but not limited to, humans, cows, horses, cats, dogs, rodents, or primates. In certain embodiments, the individual is a human.

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

[0043] The basic four-chain antibody unit is a heterotetrameric glycoprotein consisting of two identical light (L) chains and two identical heavy (H) chains. IgM antibodies are composed of five of these basic heterotetramer units and an additional polypeptide called the J chain, and contain ten antigen-binding sites. In contrast, IgA antibodies contain two to five basic four-chain units that can polymerize in combination with the J chain to form multimeric assemblies. In the case of IgG, the molecular weight of the four-chain unit is typically about 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 spaced interchain disulfide bonds. Each H chain has at its N-terminus a variable domain (VH) corresponding to each of the α and γ chains, followed by three constant domains (CH), and four CH domains for the μ and ε types. Each L chain has a variable domain (VL) at its N-terminus, followed by a constant domain at the other end. VL aligns with VH, and CL aligns in a row with the first constant domain (CH1) of the heavy chain. The pairing of VH and VL forms a single antigen-binding site. The L chains 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 their constant domains. Immunoglobulins can be classified into different classes or isotypes based on the amino acid sequence of the constant domains (CH) of their heavy chains. There are five classes of immunoglobulins: IgA, IgD, IgE, IgG, and IgM, and the heavy chains are called α, δ, ε, γ, and μ, respectively. The γ and α classes are further divided into subclasses based on relatively minor differences in CH sequence and function. For example, in humans, the following subclasses are expressed: IgG1, IgG2A, IgG2B, IgG3, IgG4, IgA1, IgA2

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

[0045] The term "single domain antibody", "nanobody", or "sdAb" refers to a single antigen-binding polypeptide that contains three complementarity-determining regions (CDRs). An sdAb alone can bind to an antigen without pairing with the corresponding CDR-containing polypeptide. In some cases, single domain antibodies are made from the HCAb of camelids, and the variable domain of the heavy chain is referred to herein as "VHH" (variable domain of the heavy chain of the heavy chain antibody). A basic VHH has the following structure from the N-terminus to the 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 heavy or light chain of the antibody. The variable domains of the heavy and light chains are generally 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 only of the heavy chain of camelid species have a single variable domain of the heavy chain called "VHH".

[0047] The term "variable" refers to the fact that certain segments of the variable domain exhibit extensive sequence diversity among antibodies. The V domain mediates antigen binding and defines the specificity of a particular antibody for a particular antigen. However, the variability is not evenly distributed throughout the entire extent of the variable domain. Instead, in both the heavy and light chain variable domains, it is concentrated in three segments called complementarity-determining regions (CDRs) or hypervariable regions (HVRs). The more highly conserved portions of the variable domain are called framework regions (FRs). The variable domains of native heavy and light chains each contain four framework (FR) regions, which mainly adopt a beta-sheet structure and, in some cases, form part of a beta-sheet structure, with the three CDRs linked by loop connections. The CDRs of each chain are closely maintained by the FR regions, and the CDRs of different chains contribute to the formation of the antigen-binding site of the antibody (see Kabat, Elvin A., Sequences of Immunological Interest, Fifth Edition, National Institute of Health, Bethesda, Md. (1991)). The constant domain is not directly involved in the binding of antigen to antibody but is involved in various effector functions such as mediating antibody-dependent cell cytotoxicity.

[0048] The term "constant domain" refers to the portion of an immunoglobulin molecule that has an amino acid sequence more conserved than the variable domain (the other part of the immunoglobulin that includes the antigen-binding site). The constant domain includes the CH1, CH2, and CH3 domains of the heavy chain (collectively, 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 a substantially complete form, in contrast to antibody fragments. Specifically, a full-length four-chain antibody includes a heavy chain and a light chain that includes an Fc region. An antibody of only a full-length heavy chain 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. Optionally, a complete antibody may have one or more effector functions.

[0050] The term "antibody fragment" or "antigen-binding fragment" refers to a portion of a complete antibody, preferably including the antigen-binding and / or variable regions of a 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 the minimal antibody fragment that contains a complete antigen-recognition and binding site. This fragment consists of a dimer of one heavy and one light chain variable domain that are tightly, non-covalently associated. The term "single-chain Fv" is also abbreviated as "sFv" or "scFv" and is an antibody fragment that includes VH and VL antibody domains linked in a single polypeptide chain. Preferably, the scFv polypeptide further includes a polypeptide linker between the VH and VL domains such that the scFv can form the desired structure for antigen binding. The term "diabody" refers to a small antibody fragment produced by constructing an sFv fragment with a short linker (about 5-10 amino acid residues) between the VH and VL domains, thereby promoting pairing between chains rather than within chains of the V domains and resulting in a bivalent fragment containing two antigen-binding sites. A bispecific diabody is a heterodimer of two "crossed" sFv fragments in which the VH domains and VL domains of two different antibodies are present on different polypeptide chains.

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

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

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

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

[0055] The term "complementary determining region" or "CDR" is 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)). The Kabat complementary determining regions (CDRs) are most commonly used based on sequence variability.

[0056] The term "framework" or "FR" residues refers to variable-domain residues other than the HVR residues 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, for example, are monospecific. The term "multispecificity" as used herein refers to an antigen-binding protein having 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, "bispecificity" refers to an antigen-binding protein having two different antigen-binding specificities.

[0059] The term "monospecificity" as used herein refers to an antigen-binding protein having one or more binding sites that each specifically bind to the same epitope of the same antigen.

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

[0061] The term "antibody effector function" refers to the biological activities of an antibody that are attributable to the Fc region of the antibody (either the native sequence Fc region or an amino acid sequence variant Fc region), and 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; down-regulation of cell surface receptors (e.g., B cell receptors); and B cell activation, among others. "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 Ig bound to Fc receptors (FcRs) present on certain cytotoxic cells (e.g., natural killer (NK) cells, neutrophils, and macrophages) specifically binds these cytotoxic effector cells to antigen-bearing target cells, and then the target cells are killed with cytotoxins.

[0062] As used herein, the term "Fc region" or "fragment crystallizable region" is used to define the C-terminal region of an immunoglobulin heavy chain that includes 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 non-covalent binding interactions 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 a 1:1 interaction between the members of a binding pair. Binding affinity is represented by K d , Ko ff , K on , or K a . The equilibrium dissociation constant "K D " or "K dThe term "___" refers to the dissociation constant of a particular antibody-antigen interaction and represents the concentration of antigen, in units of M, required to occupy half of the antibody molecule binding domains present in solution at equilibrium. K D measurements assume 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 in the designated 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, the alignment can be achieved using a variety of methods known in the art, such as by using publicly available computer software such as, for example, BLAST, BLAST-2, ALIGN, or MEGALIGN® (DNATAR) software. One of ordinary skill in the art can determine appropriate parameters for measuring the alignment, including any algorithms necessary to achieve an optimal alignment over the entire length of the sequences being compared.

[0065] The present invention relates to a bispecific antibody that binds to PD-L1 and CD47 in a bispecific manner, for example, a humanized anti-PD-L1×CD47 bsAb (hereinafter sometimes referred to as "humanized anti-PD-L1×CD47 bsAb"), specifically, a bispecific humanized single-domain antibody in which an anti-PD-L1 hsdAb and an anti-CD47 hsdAb are fused, a humanized anti-PD-L1×CD47 heavy-chain only antibody (HCAb) (for example, a humanized anti-PD-L1×CD47 bsAb-Fc fusion protein in which a crystallizable fragment (Fc fragment) of human immunoglobulin G (IgG) is fused to an anti-PD-L1 hsdAb and / or an anti-CD47 hsdAb), and its production and use, which comprises a first humanized single-domain antibody (first hsdAb) or an antigen-binding fragment thereof (anti-PD-L1 hsdAb) that specifically binds to PD-L1; and a second humanized single-domain antibody (second hsdAb) or an antigen-binding fragment thereof (anti-CD47 hsdAb) that specifically binds to CD47.

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

[0067] In the present invention, the bispecific antibody that binds to PD-L1 and CD47 in a bispecific manner and comprises the humanized anti-PD-L1×CD47 bsAb may be a humanized anti-PD-L1×CD47 bsAb in which a first humanized single-domain antibody (first hsdAb) or an antigen-binding fragment thereof (hereinafter sometimes 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 an antigen-binding fragment thereof (hereinafter sometimes referred to as "anti-CD47 hsdAb") that specifically binds to CD47 as a second antigen-binding portion are fused.

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

[0069] In addition, the anti-CD47 hsdAb comprises 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.

[0070] The 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 the 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 may contain 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 can comprise a sequence having at least 95%, 96%, 97%, 98%, 99% or 100% sequence identity to the sequence of SEQ ID NO: 16. Alternatively, it can comprise a sequence having at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% sequence identity to the sequence of SEQ ID NO: 17.

[0073] The heavy chain variable domain framework (FR) sequences having 100% sequence identity to the sequence of SEQ ID NO: 16 or SEQ ID NO: 17 are shown in Tables 1 to 4 below.

[0074] [Table 1] [Table 2]

Table 3

Table 4

[0075] More specifically, the anti-PD-L1 hsdAb can include the following FR1, FR2, FR3, and FR4: (1) FR1 consisting of the amino acid sequence represented by SEQ ID NO: 19; FR2 consisting 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 SEQ ID NO: 23; FR2 consisting 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 consisting of the amino acid sequence represented by SEQ ID NO: 26

[0076] More specifically, the anti-PD-L1 hsdAb can include the following FR1, FR2, FR3, and FR4: FR1 consisting of the amino acid sequence represented by SEQ ID NO: 19; FR2 consisting 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

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

[0078] More specifically, the anti-CD47 hsdAb can include the following FR1, FR2, FR3, and FR4: (1) FR1 consisting of the amino acid sequence represented by SEQ ID NO: 19; FR2 consisting 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 SEQ ID NO: 23; FR2 consisting 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 consisting of the amino acid sequence represented by SEQ ID NO: 26

[0079] More specifically, the anti-CD47 hsdAb can include the following FR1, FR2, FR3, and FR4: FR1 consisting of the amino acid sequence represented by SEQ ID NO: 23; FR2 consisting 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 consisting 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] Also, the K of the binding of the humanized anti-PD-L1×CD47 bsAb to PD-L1 and CD47 D is respectively 10 -6 M to 10 -12 M, 10 -6 M to 10 -11 M, 10 -6 M to 10 -10 M, 10 -6 M to 10 -9 M, 10 -7 M to 10 -9 M, or 10 -8 M to 10 -9 M may be sufficient.

[0082] Also, the EC50 of the humanized anti-PD-L1×CD47 bsAb in FACS analysis is less than 500 nM, specifically, it may be 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, 0.1 nM to 400 nM, 0.1 nM to 300 nM, 0.1 nM to 200 nM, 0.1 nM to 100 nM, 0.1 to 50 nM, 0.1 to 10 nM, 1 nM to 500 nM, 1 nM to 400 nM, 1 nM to 300 nM, 1 nM to 200 nM, 1 nM to 100 nM, 1 to 50 nM or 1 to 10 nM.

[0083] In the present invention, the humanized anti-PD-L1×CD47 bsAb can have any suitable valence with respect to the epitopes of PD-L1 and CD47, respectively. Specifically, the humanized anti-PD-L1×CD47 bsAb can have a valence of 2, 3, 4, 5, 6, or more with respect to 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] Alternatively, the humanized anti-PD-L1×CD47 bsAb may be formed by direct binding of an anti-PD-L1 hsdAb and an anti-CD47 hsdAb by a peptide bond, or by indirect fusion via a peptide linker. The length, flexibility and / or other properties of the peptide linker may affect specific properties including, but not limited to, the affinity, specificity or binding ability for one or more specific antigens or epitopes. For example, a longer peptide linker can be selected to prevent steric hindrance between two adjacent domains. In certain embodiments, the peptide linker includes flexible residues (such as glycine and serine) to allow adjacent domains to move freely relative to each other. For example, a glycine-serine doublet may be a suitable peptide linker. Also, the peptide linker can be of any suitable length, for example, any one 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 length. In some embodiments, the peptide linker may be a peptide linker consisting of any one of the amino acid sequences of SEQ ID NO: 27 to SEQ ID NO: 29.

[0085] In addition, 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 bivalent with respect to PD-L1 and monovalent with respect to CD27, wherein the bivalent anti-PD-L1 hsdAb can be indirectly fused to the monovalent anti-PD-L1 hsdAb represented by amino acid sequence number 16, either directly or via a peptide linker. Further, the humanized anti-PD-L1×CD47 bsAb may consist of the amino acid sequence represented by SEQ ID NO: 15.

[0087] The single domain antibodies (sdAbs) according to the present invention may be derived from any species including, but not limited to, mouse, rat, human, camel, llama, nurse shark, fish, goat, rabbit, and bovine. Also, the naturally occurring sdAb molecules may be derived from species other than camelids.

[0088] Also, an sdAb is a heavy chain antibody lacking a light chain derived from a known naturally occurring single domain antigen-binding molecule. Such single domain molecules are disclosed, for example, in WO 94 / 04678 and Hamers-Casterman et al. (1993), Nature 363:446-448. The variable domain derived from a heavy chain molecule that naturally lacks a light chain is disclosed herein as VHH and is distinguished from the conventional VH of a four-chain immunoglobulin. Such VHH molecules can be derived from antibodies produced by animals of the camelid family such as camel, llama, vicuña, dromedary camel, alpaca and guanaco. Other species capable of producing heavy chain molecules that naturally lack a light chain, other than camelids, are also included within the scope of the present 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, which retains the specificity and affinity of the parental non-human antibody while reducing its immunogenicity in humans. Generally, a humanized antibody is composed of one or more variable domains in which hypervariable regions (HVRs), such as complementarity-determining regions (CDRs), or a part thereof are derived from a humanized antibody sequence. In some embodiments, the humanized single-domain antibody comprises CDRs derived from a non-human single-domain antibody (e.g., CDRs derived from a camelid such as a llama) and heavy-chain FRs derived from a human antibody sequence. In some specific embodiments, certain FR residues in the humanized antibody are substituted with corresponding residues of a non-human antibody (e.g., the antibody from which the HVR residues are derived), so that the specificity or affinity of the antibody is restored or enhanced.

[0090] In the present invention, the bispecific antibody that binds to PD-L1 and CD47 specifically and contains the humanized anti-PD-L1×CD47 bsAb may be a humanized anti-PD-L1×CD47 HCAb or an antigen-binding fragment thereof.

[0091] Specifically, the humanized anti-PD-L1×CD47 HCAb may be a fusion of the humanized anti-PD-L1×CD47 bsAb described herein with one or more CH2 and / or CH3 domains such as an Fc fragment. Also, the anti-PD-L1 hsdAb and / or anti-CD47 hsdAb described herein may be a fusion with one or more CH2 and / or CH3 domains such as an Fc fragment.

[0092] The CH2 and / or CH3 domains are derived from an immunoglobulin and may be IgA, IgD, IgE, IgG or IgM, and specifically may be IgG. In some embodiments, the humanized anti-PD-L1×CD47 HCAb may contain an Fc fragment of IgG, e.g., IgG1, IgG2, IgG3 or IgG4, and the Fc fragment may be a human Fc, e.g., a human IgG1 (hIgG1) Fc, hIgG2 Fc, hIgG3 Fc or hIgG4 Fc.

[0093] The humanized anti-PD-L1×CD47 HCAb may be a monomer or a multimer. Also, in the case of a multimer, it may be bispecific and multivalent (e.g., bivalent, trivalent, tetravalent, or more valent), 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 domain, specifically the Fc fragment, via a peptide linker. The length, flexibility, and / or other properties of the peptide linker may affect specific properties including, but not limited to, the 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 contains flexible residues (e.g., glycine and serine) to allow adjacent domains to move freely relative to each other. For example, a glycine-serine doublet may be a suitable peptide linker. Also, the peptide linker may be of any suitable length, such as 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 acids.

[0095] Furthermore, the peptide linker may have a natural sequence or a non-natural sequence. For example, a sequence derived from the hinge region of an antibody of only the heavy chain may be used as the linker. See, for example, WO1996 / 34103. In some embodiments, the peptide linker may be an hIgG1 hinge, hIgG2 hinge, hIgG3 hinge, hIgG4 hinge, or variants 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 can include the amino acid sequence represented by SEQ ID NO: 18, or a variant thereof showing at least 80% sequence homology (for example, at least 80%, 88%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99%) to the amino acid sequence.

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

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

[0099] The bsFab according to the present invention and its manufacturing technology 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 antibody that binds to PD-L1 and CD47 in a dual-specific manner and includes the humanized anti-PD-L1×CD47 bsAb includes amino acid sequence variants. The amino acid sequence variants of the antibody can be produced by introducing appropriate modifications into 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. The final product can be obtained by any combination of deletions, insertions, and substitutions, provided that the final product retains the desired properties such as antigen binding. In some embodiments, the substitution, insertion, or deletion may occur within one or more hypervariable regions (HVRs) as long as the binding ability of the antibody to the antigen is not substantially reduced by such changes. For example, conservative changes that do not significantly reduce the binding affinity can be made within the HVRs. Such changes may also occur outside the HVR "hot spots" or CDRs.

[0101] Also, 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). Further, the at least one amino acid substitution may be a conservative substitution or 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 may include at least one substitution (e.g., any 1, 2, 3, or 4 amino acids) in CDR1, CDR2, and / or CDR3. In certain embodiments, the amino acid substitution occurs in the FR region and can include at least one chi n (e.g., any 1, 2, 3, 4, 5, or 6 amino acids) in FR1, FR2, FR3, and / or FR4.

[0102] In addition, the amino acid sequence insertions include amino-terminal and / or carboxyl-terminal fusions ranging in length from one residue to polypeptides containing 100 or more residues, as well as in-sequence insertions of single or multiple amino acid residues. Examples of terminal insertions include antibodies having an N-terminal methionyl residue. Other insertion variants of antibody molecules may include fusions of polypeptides that extend the serum half-life of the antibody or (e.g., in the case of ADEPT) enzymes to the N- or C-terminus of the antibody.

[0103] Furthermore, as described herein, one or more amino acid modifications can be introduced into the Fc region of a bispecific antibody (e.g., a humanized anti-PD-L1×CD47 HCAb) that binds bispecifically to PD-L1 and CD47 and includes the humanized anti-PD-L1×CD47 bsAb to generate an Fc region variant. The Fc region variant can include a human Fc region sequence (e.g., human IgG1, IgG2, IgG3, or IgG4 Fc) that contains amino acid modifications (e.g., substitutions) at one or more amino acid positions.

[0104] In the present invention, the bispecific antibody that binds bispecifically to PD-L1 and CD47 and includes the humanized anti-PD-L1×CD47 bsAb can be linked, fused, bound (e.g., covalently or non-covalently), or otherwise associated with a diagnostic moiety or a biocompatibility modifier. For example, a peptide or polypeptide (e.g., a biotoxin, biomarker, tablet tag, etc.), protein, polymer, nucleic acid molecule, small molecule, mimetic, synthetic agent, inorganic molecule, organic molecule, or radioisotope can bind or associate with them.

[0105] In addition, the bispecific antibody that binds to PD-L1 and CD47 in a dual-specific manner and includes the humanized anti-PD-L1×CD47 bsAb can bind or associate with a diagnostic agent or a detectable agent, a marker or a reporter that is a biological molecule (e.g., a peptide or a nucleotide), a small molecule, a phosphor, or a radioisotope. The labeled modulator can be useful for monitoring diseases associated with PD-L1 and / or CD47 such as the onset or progression of cancer, or for determining the effectiveness of a specific therapeutic method (i.e., seragnosis) including the antibodies disclosed herein, or as part of a clinical trial procedure for determining a future treatment course. These markers or reporters are also useful for the purification of the antibodies disclosed herein.

[0106] Furthermore, the bispecific antibody that binds to PD-L1 and CD47 in a dual-specific manner and includes the humanized anti-PD-L1×CD47 bsAb can bind to an immunomodulatory agent, a cytokine, a cytotoxic agent, a chemotherapeutic agent, a diagnostic agent, an antiviral agent, an antimicrobial agent or a drug. In this regard, the present invention provides an antibody complex including the bispecific antibody that binds to PD-L1 and CD47 in a dual-specific manner and includes the humanized anti-PD-L1×CD47 bsAb of the present invention bound to an immunomodulatory agent, a cytokine, a cytotoxic agent, a chemotherapeutic agent, a diagnostic agent, an antiviral agent, an antimicrobial agent or a drug.

[0107] In addition, the present invention provides a nucleic acid molecule encoding a bispecific antibody that binds to PD-L1 and CD47 in a dual-specific manner and includes the humanized anti-PD-L1×CD47 bsAb disclosed herein, an expression vector including the nucleic acid molecule, and a host cell transformed with the expression vector.

[0108] In addition, the present invention provides a method for producing a bispecific antibody, including: (a) culturing a host cell under conditions that enable the expression of the bispecific antibody; and (b) recovering the expressed bispecific antibody.

[0109] In the present invention, the DNA encoding the bispecific antibody that binds to PD-L1 and CD47 in a bispecific manner and contains the humanized anti-PD-L1×CD47 bsAb disclosed herein can be easily isolated and sequenced 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 subcloning hybridoma cells (or colonies derived from phage or yeast) can serve as a preferred source of DNA. More specifically, the isolated DNA (which may be modified) can be used for cloning the constant and variable region sequences for antibody production.

[0110] Exemplary methods include extraction of RNA from selected cells, conversion to cDNA, and amplification by PCR using antibody-specific primers. Appropriate primers are well known in the art and are readily available from a number of commercial sources, as exemplified herein. To express recombinant human or non-human antibodies isolated by screening combinatorial libraries, the DNA encoding the antibody 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 into and expressed in cells such as simian COS cells, NS0 cells, Chinese hamster ovary (CHO) cells or myeloma cells that do not produce construction depression by other methods.

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

[0112] Expression vectors containing the antibody coding sequence and appropriate transcriptional and translational control signals can be constructed using methods well 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 also refer to not only the specific target cells but also their progeny. Due to mutations or environmental influences, specific modifications may occur in subsequent generations, and these progeny may not be identical to the parental cells, but they are still within the scope of the term "host cell" as used herein. Such cells can contain the vector.

[0114] Also, by using molecular biology techniques and current protein expression methodologies recognized in the art, a substantial amount of the antibodies disclosed herein can be produced. More specifically, the nucleic acid molecule encoding the antibody can be integrated into a widely known and commercially available protein production system, including various types of host cells, to provide the desired pharmaceutical in preclinical, clinical, or commercial quantities. In some embodiments, the nucleic acid molecule encoding the antibody is engineered into a vector or expression vector that enables efficient integration into the selected host cell and subsequent high-level antibody expression.

[0115] Preferably, the nucleic acid molecules encoding the 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 prokaryotic systems can also be used. Transfection 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 transfection, protoplast fusion, electroporation, encapsulation of polynucleotides in liposomes, and direct microinjection of DNA into the nucleus. In addition, nucleic acid molecules can 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, biolistic transformation, direct injection, electroporation, and viral transformation. Methods for transforming bacterial and yeast cells are also widely known in the art.

[0116] To express the antibodies disclosed herein, a variety of commercially available host expression vector systems can be used. These host expression systems not only represent vehicles in which the coding sequence of interest is expressed and then purified, but also cells that can express the molecules of the present invention in vivo 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 a modulator coding sequence, such as bacteria (e.g., Escherichia coli, Bacillus subtilis, Streptomyces); yeast infected with recombinant yeast expression vectors containing a modulator coding sequence (e.g., Saccharomyces, Pichia); insect cell lines infected with recombinant virus expression vectors (e.g., baculovirus) containing a modulator coding sequence; plant cell lines (e.g., Nicotiana, Arabidopsis, duckweed, corn, wheat, potato, etc.) infected with recombinant virus expression vectors (e.g., cauliflower mosaic virus, CaMV; tobacco mosaic virus, TMV), or recombinant plasmid expression vectors (e.g., Ti plasmid) containing a modulator coding sequence; or mammalian cell lines (e.g., COS, CHO, BHK, 293, 3T3 cells) containing a recombinant expression construct containing a promoter derived from a mammalian genome (e.g., metallothionein promoter) or a promoter derived from a mammalian virus (e.g., adenovirus late promoter; vaccinia virus 7.5K promoter), but are not limited to these examples.

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

[0118] The present invention also provides a pharmaceutical composition for preventing or treating cancer, comprising as an active ingredient a bispecific antibody that binds to PD-L1 and CD47 in a bispecific manner and contains the humanized anti-PD-L1×CD47 bsAb disclosed herein, or an antibody complex 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 an effective amount of a bispecific antibody that binds to PD-L1 and CD47 in a bispecific manner and contains the humanized anti-PD-L1×CD47 bsAb disclosed herein, or an antibody complex containing the bispecific antibody.

[0120] Furthermore, the present invention also provides the use of a bispecific antibody that binds to PD-L1 and CD47 in a bispecific manner and contains the humanized anti-PD-L1×CD47 bsAb disclosed herein, or an antibody complex containing the bispecific antibody, for preventing or treating cancer.

[0121] The present invention further provides the use of a bispecific antibody that binds to PD-L1 and CD47 in a bispecific manner and contains the humanized anti-PD-L1×CD47 bsAb disclosed herein, or an antibody complex containing the bispecific antibody, for manufacturing a composition for preventing or treating cancer.

[0122] In the present invention, the cancer is a cancer that requires activation of T cells by blocking the activity of immune checkpoint proteins, and examples thereof include, but are not limited to, melanoma, lung cancer, liver cancer, glioblastoma, ovarian cancer, colorectal cancer, head and neck cancer, bladder cancer, renal cell cancer, 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 content related to the bispecific antibody that binds to PD-L1 and CD47 in a bispecific manner and contains the humanized anti-PD-L1×CD47 bsAb disclosed in the present specification, or the antibody complex containing the bispecific antibody, is the same as the foregoing description. Therefore, the specific description thereof is incorporated by reference, and hereinafter, only the specific configuration of the pharmaceutical composition and its use will be described.

[0124] The pharmaceutical composition according to the present invention may contain one or more (for example, 2 or 3 kinds) of the bispecific antibodies that bind to PD-L1 and CD47 in a bispecific manner and contain the humanized anti-PD-L1×CD47 bsAb disclosed in the present specification, 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] In addition, the pharmaceutical composition according to the present invention can be formulated as desired using techniques recognized in the relevant technical field according to the form of the antibody described in the present specification, the intended delivery method, and various other variables. It can also be formulated to contain a suitable pharmaceutically acceptable carrier including 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 processing it into a pharmaceutically optimized formulation for delivery. For example, various pharmaceutically acceptable carriers including vehicles, adjuvants, and diluents are readily available from a number of commercial manufacturers. Furthermore, classifications of pharmaceutically acceptable auxiliary substances 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] In addition, the pharmaceutical composition 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 administration 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 water-soluble form, for example, aqueous solutions of water-soluble salts. Furthermore, appropriate active compounds can also be suspended in an oily injection suspension for administration. Suitable lipophilic solvents or vehicles include fatty oils such as sesame oil, or synthetic fatty acid esters such as ethyl oleate and triglycerides. Aqueous injection suspensions can contain agents that increase the viscosity of the suspension, such as sodium carboxymethyl cellulose, sorbitol, and / or dextran. Optionally, the suspension can 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, tablets including coated tablets, elixirs, suspensions, syrups or inhalants, and controlled release forms thereof.

[0129] In general, the antibodies disclosed herein can be administered to a subject in need thereof in vivo via various routes including, but not limited to, oral, intravenous, intraarterial, subcutaneous, parenteral, intranasal, intramuscular, intracardiac, intraventricular, intratracheal, buccal, rectal, intraperitoneal, intradermal, topical, transdermal, and intrathecal administration, or via implantation or inhalation. Appropriate formulations and administration routes can be selected according to the purpose of use and the therapeutic treatment.

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

[0131] The pharmaceutically effective amount typically depends on factors such as the weight of the subject being treated, its physical condition, the degree of the condition being treated, and the age of the subject being treated. Generally, the antibodies disclosed herein are administered in an amount in the range of about 10 ng / kg body weight to about 100 mg / kg body weight, about 50 μg / kg body weight to about 5 mg / kg body weight, about 100 μg / kg body weight to about 10 mg / kg body weight, about 100 μg / kg body weight to about 20 mg / kg body weight, 0.5 mg / kg body weight to about 20 mg / kg body weight per dose, but is not limited thereto. Also, the antibody can be administered at a dose of at least about 100 μg / kg body weight, at least about 250 μg / kg body weight, at least about 750 μg / kg body weight, at least about 3 mg / kg body weight, at least about 5 mg / kg body weight, or at least about 10 mg / kg body weight, but is not limited thereto.

[0132] Also, the pharmaceutical composition according to the present invention is administered at a dose of about 100 mg to about 10,000 mg, about 200 mg to about 9,000 mg, about 300 mg to about 8,000 mg, about 400 mg to 7,000 mg, 500 mg to 5,000 mg, but is not limited thereto.

[0133] The pharmaceutical composition according to the present invention is usually administered to a patient multiple times. Exemplary treatment regimens include administration once every two weeks, once a month, or once every three to six months. For example, a patient can be administered an antibody (e.g., as an intravenous formulation) once every four weeks, e.g., once every 28 days. The dosing frequency can be adjusted according to the pharmacokinetic profile of the antibody in the patient. For example, a dosing frequency of once every two weeks may be required if the half-life of the antibody is two weeks. In some methods, two or more antibodies with different binding specificities may be administered simultaneously, in which case the dose of each antibody administered falls within the indicated range.

[0134] The dosage and dosing frequency 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 dosing frequency vary depending on whether the treatment is prophylactic or therapeutic.

[0135] The duration of the treatment therapy is also affected by factors such as the disease to be treated, the age and condition of the patient, the stage and type of the patient's disease, and the acceptance of the patient's response to the treatment. The clinician can carefully observe the effect of the therapy and make necessary adjustments if required. When active agents 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, simultaneously with, or after the administration of the second therapeutic agent.

[0136] Hereinafter, the present invention will be described in detail with reference to Examples and Experimental Examples.

[0137] However, it should be understood that the following Examples and Experimental Examples are merely illustrative and are not intended to limit the scope of the present invention to their content.

[0138] Example 1: Immunization and Blood Sampling The human PD-L1 protein or human CD47 protein, which is an immunizing antigen, was mixed with an adjuvant (GERBU) and administered to one alpaca by intramuscular injection three times. The immunization was carried out in three divided doses, and 10 mL of blood was collected from the alpaca 14 days after the final immunization. The immune response was analyzed using ELISA. To examine whether antibodies were produced, the immunizing antigen was dispensed into a 96-well microplate using coating buffer at a concentration of 1 μg / mL and coated overnight at 4°C. The 96-well microplate was washed three times with PBST and then blocked by treating with 5% skim milk at room temperature for 2 hours to inhibit non-specific binding. After washing three times with PBST, serum samples obtained before immunization (day 0), 14 days after immunization (day 14), 28 days after immunization (day 28), and 42 days after immunization (day 42) were treated at serial dilution concentrations. Thereafter, the 96-well microplate was washed five times with PBST, and then reacted with goat anti-lama IgG HRP antibody at room temperature for 1 hour, and it was confirmed by TMB reaction whether the antibody bound to the immunizing antigen.

[0139] Example 2: Construction and evaluation of library The gene encoding the single-domain antibody that binds to the immunizing antigen confirmed in Example 1 above was amplified to construct an immune library. For the construction of the library, peripheral blood mononuclear cells (PBMCs) were separated from the blood using Ficoll. A gene fragment encoding the single-domain antibody was amplified from the total RNA extracted from the separated peripheral blood mononuclear cells (PBMCs) using specific primers and cloned into the pComb3X vector. The size of the prepared immune library was 5.4×10 8 It was.

[0140] Example 3: Library amplification The immune library prepared in Example 2 above was transformed into 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 shaker incubator at 37°C. Cultured until the optical density (OD 600 ) at 600 nm reached 0.5, and at that time, M13K07 phage (manufactured by Invitrogen) was added at 1×1011 It was added until it reached pfu / mL. Then, it was statically cultured at 37°C for 30 minutes, and subsequently further cultured at 37°C and 200 rpm for 30 minutes in a shaking incubator. Then, the culture solution was centrifuged at room temperature at 4,000 rpm for 15 minutes, and the supernatant was removed. Next, 10 mL of 2xYT medium containing 100 μg / mL ampicillin and 50 μg / mL kanamycin was added, the pellet of the culture solution was resuspended, and it was cultured overnight at 30°C and 250 rpm in a shaking incubator. After that, it was centrifuged at 4°C and 4,000 rpm for 30 minutes. The supernatant was precipitated by the PEG precipitation method and centrifuged at 4°C and 12,000 rpm for 30 minutes. The obtained 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 immunizing antigen, 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 of 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 4 times to reduce the non-specific binding of the library to the microplate wells. Then, the library was transferred to a 1.7 mL tube and stored at 4°C until use. The microplate coated with the immunizing antigen was washed 5 times with PBST and blocked with 5% skim milk at room temperature for 2 hours. After washing 5 times with PBST, binding solution (2.5% skim milk, 0.5% tween20) was added to the library with reduced non-specific binding at 5×10 12Aliquoted at the concentration of virions per well and reacted at room temperature for 30 minutes. Then, the plate was washed 10 times with a washing solution (PBS, 0.5% tween20) and further washed 3 times with PBST. The single-domain antibody specifically bound to the immunogen was selectively eluted by adding 5 μg of the immunogen per well and culturing with shaking at room temperature at 500 rpm for 30 minutes. The eluted phage was infected with exponentially growing XL1-blue cells and spread on 2xYT agar medium. Panning for the second selection was repeated under the same conditions as above. The single phage clones generated on the agar medium were each amplified and screened using FACS.

[0142] Example 5: Phage Screening To induce transient overexpression of the immunogen in Expi-CHO cells, the gene encoding the immunogen was inserted into the pCMV6-GFP vector to construct the pCMV6-immunogen-GFP plasmid. 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. Next, the cells were centrifuged at room temperature at 1200 rpm for 3 minutes, and the supernatant was removed. The cells were washed twice with DPBS and aliquoted into a 96-well microplate at 3×10 5 cells / 100 μL / well. Single clone phage was added to each well, cultured at 4°C for 1 hour, and then washed twice with DPBS. An antibody that specifically binds to the phage (M13 major coat protein Alexa Fluor 647 (Santa cruz)) was aliquoted 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 subjected to sequence analysis.

[0143] Example 6: Expression and Purification of a Single-Domain Antibody Fused with the Human IgG Fc Domain 6-1: Expression and Purification of a Monovalent Single-Domain Antibody Fused with the Human IgG Fc Domain

[0144] The clone selected in Example 5 was cloned into a TGEX-Fc (IgG1) or TGEX-Fc (IgG4) expression vector. For the expression of a single-domain antibody fused with the human IgG Fc domain, Expi-CHO cells with a viability of 95-99% were counted, and 7×10 6 cells were added to 25 mL of culture medium (Expi-CHO Expression Medium (manufactured by Gibco)). The cells were cultured overnight in a shaking incubator at 37 °C, 125 rpm, and 8% CO2. Then, 80 μL of ExpiFectamine® CHO Reagent (manufactured by Gibco, 100033021) was added to 920 μL of OptiPRO® medium, 20 μg of plasmid DNA encoding a single-domain antibody fused with the human IgG Fc domain was added, and subsequently, 1 mL of OptiPRO® medium was added. After reacting the mixture at room temperature for 5 minutes, it was added to the cultured cells. The cells were cultured in a shaking incubator at 125 rpm and 8% CO2 for 20 hours. To enhance the expression of the single-domain antibody fused with the human IgG Fc domain, 150 μL of ExpiFectamine® CHO Enhancer (Gibco) and 6 mL of ExpiCHO Feed (Gibco) were added, and the cells were cultured in a shaking incubator at 32 °C, 125 rpm, and 5% CO2 for 5 days. 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 G HP column (GE Healthcare), washed with PBS, and the single-domain antibody fused with the human IgG Fc domain was eluted from the column using IgG elution buffer (Thermo). The eluted sample was neutralized by adding 1 M Tris-HCl (pH 9.0) and stored at 4 °C until use.

[0145] 6-2: Expression and Purification of a Bivalent Single-Domain Antibody Fused with the Human IgG4 Fc Domain The clones selected in Example 5 were linked using two G2S linkers (GGSGGS) to prepare a bivalent single-domain antibody. The nucleotide encoding the bivalent single-domain antibody was obtained by gene synthesis (manufactured by Macrogen, 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, it was expressed and purified in the same manner as described in Example 6-1.

[0146] Example 7: Expression and Purification of Bispecific Single-Domain Antibody Fused with Human IgG4 Fc Domain From among the clones selected in Example 5, one single-domain antibody specifically binding to each of the PD-L1 and CD47 antigens was 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 the 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 (manufactured by Macrogen, Korea). For the expression and purification of the bispecific single-domain antibody fused with the human IgG4 Fc domain, the synthesized gene was cloned into a TGEX-Fc(IgG4) expression vector. It was expressed and purified in the same manner as described in Example 6-1.

[0147] Example 8: Expression and Purification of Bispecific Humanized Single-Domain Antibody Fused with Human IgG4 Fc Domain For the humanization of the trivalent bispecific single-domain antibody prepared in Example 7, comparison was made with the FR (framework region) sequence of the human germline VH fragment DP-47 (J Mol Biol. 1992 Oct 5;227(3):776-98), and homology (identity) was evaluated. Among the clones selected in Example 5, the amino acid sequences of the single-domain antibody specific for PD-L1 and the single-domain antibody specific for CD47 (PDL1 Nb#01 and CD47 Nb#01) were analyzed. Glutamine (Q), which is the first amino acid of the FR1 (framework region) sequence, was substituted with glutamic acid (E) to improve homology with the human antibody DP-47. As a result, a trivalent PD-L1 and CD47 bispecific humanized single-domain antibody (humanized anti-PD-L1 sdAb × humanized anti-PD-L1 sdAb × humanized anti-CD47 sdAb) was prepared. Amino acid substitution was performed using a Q5 Site-Directed Mutagenesis Kit (NEB) and primers for target mutagenesis. It was expressed and purified by 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 abilities 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 immune antigens were confirmed using FACS.

[0149] Specifically, the CHO-K1_PD-L1 cell line (CHO-K1 cells overexpressing the PD-L1 antigen) or the Expi-CHO_CD47 cell line (Expi-CHO cells overexpressing the CD47 antigen) was 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. Then, 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, 3×105 Dispense at a density of cells / 100 μl / well, and treat 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. An isotype control antibody was used as a negative control group. After culturing the cells at 4°C for 1 hour, they were washed twice with DPBS. Then, they were treated with an antibody specifically binding 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 twice more with DPBS, they were resuspended in 100 μL of DPBS and analyzed by FACS using an Accuri C6 (BD) device.

[0150] Example 10: Evaluation of the inhibitory ability of single-domain antibodies against immune antigen interactions using FACS 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 were evaluated for their inhibitory ability against the PD-1 / PD-L1 or CD47 / SIRPα interaction.

[0151] Specifically, for the evaluation 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) against the PD-1 / PD-L1 interaction, the CHO-K1_PD-L1 cell line (CHO-K1 cells stably expressing the PD-L1 antigen) was seeded at 2×10 5Cells were dispensed and treated with 10 μg / mL of human PD-1-His protein. Next, 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) were treated at different concentrations. An isotype control antibody was used as a negative control group. The cells were cultured at 4°C for 1 hour and washed three times with DPBS. Subsequently, an antibody specifically binding to the His antigen (goat anti-His APC) was added and reacted at 4°C in the dark for 30 minutes. The cells were washed three times with DPBS, resuspended in 100 μL of DPBS, and the amount of PD-1-His protein remaining in CHO-K1_PD-L1 cells (CHO-K1 cells stably expressing the PD-L1 antigen) was confirmed using an Accuri C6 (BD) device. The inhibitory ability 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) against the PD-1 / PD-L1 interaction was evaluated by measuring the residual PD-1-His protein.

[0152] In addition, to evaluate the effect 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 the CD47 / SIRPα interaction, Expi-CHO_CD47 cells (Expi-CHO cells stably expressing the CD47 antigen) were dispensed into a 96-well microplate at 2×105 cells per well and treated with 10 μg / mL of human SIRPα-His protein. Then, 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) were treated at different concentrations, and an isotype control antibody was used as a negative control group. Next, the amount of SIRPα-His protein remaining in Expi-CHO_CD47 cells (Expi-CHO cells stably expressing the CD47 antigen) was confirmed using an Accuri C6 (BD) device 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 the CD47 / SIRPα interaction was evaluated by measuring the remaining amount of SIRPα-His protein.

[0153] Example 11: Evaluation of the Affinity between a Single-Domain Antibody and an Immunoantigen Using an Octet RED 96e (ForteBio) device, the affinity (K d ) 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.

[0154] Specifically, a biosensor chip (Fortebio) coated with anti-human Fc was saturated and bound at the 1.5 nm level with a 96-well microplate (manufactured by Greiner) in which 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) was dispensed respectively. The PD-L1 antigen and the CD47 antigen were serially diluted 2-fold up to 10 - 400 nM using 1X kinetic buffer (manufactured by ForteBio), and reacted with the biosensor coated at 30°C and 1,000 rpm. The binding reaction and dissociation reaction of the sample were analyzed over 200 seconds and 400 seconds respectively. The resulting data was analyzed using the 1:1 interaction model (Global fitting) method.

[0155] Furthermore, the affinity (K d ) of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 above for the heterologous immunogenic antigen protein was measured using an Octet RED 96e (manufactured by ForteBio) apparatus.

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

[0157] Example 12: In vitro efficacy evaluation of humanized single-domain antibody The in vitro efficacy of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 above was evaluated by assessing the degree of activation of T cells and the degree of activation of phagocytosis by macrophages in the mixed lymphocyte reaction. + This was evaluated by measuring the concentration of IL-2 expressed by Jurkat cells. This was done by evaluating the inhibition of the interaction between CHO-K1 cells overexpressing the PD-L1 protein and the above-mentioned Jurkat cells expressing PD-1.

[0158] 12-1: Evaluation of T cell activity of humanized single-domain antibody Specifically, 2×10 CHO-K1 cells overexpressing the PD-L1 protein were seeded per well.

[0159] Specifically, 2×10 4They were dispensed individually and cultured for 16 hours in an incubator maintained at 5% CO2. Then, the medium was removed, and humanized anti-PD-L1×CD47 trivalent HCAb (hPPC-Nb-IgG4) was added at various concentrations and reacted for 1 hour. Jurkat cells were adjusted to 5×10 5 cells / 100 μL and treated with PHA at a final concentration of 0.5 mg / mL. Then, they were added to CHO-K1 cells treated with humanized anti-PD-L1×CD47 trivalent HCAb (hPPC-Nb-IgG4) and cultured for 48 hours. Next, the IL-2 concentration in the supernatant was measured by ELISA to evaluate the T cell activation efficacy of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC-Nb-IgG4).

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

[0161] Specifically, monocytes-derived dendritic cells (MoDC) from donor A were dispensed into 96-well flat-bottom plates at 5×10 4 cells / well, and CD4 + T cells of donor B were added to each well containing MoDC at 4×10 5 cells / well and mixed. Next, humanized anti-PD-L1×CD47 trivalent HCAb (hPPC-Nb-IgG4) was added at various concentrations, and the cells were cultured in a 37 °C incubator maintained at 5% CO2 for 5 days (120 hours). Then, on the 3rd day of culture, the supernatant was collected for IL-2 measurement, and on the 5th day, the supernatant was collected from the same wells for IFN-γ measurement. Thereafter, the concentrations of IL-2 and IFN-γ were measured using the ELISA method.

[0162] The MoDC used in the mixed lymphocyte reaction analysis expresses PD-L1 and CD4 +T cells express PD-1. Inhibiting the interaction between PD-L1 and PD-1 with an antibody activates the suppressed T cell activity. This is similar to the system where cancer cells survive within the immune system by suppressing the immune activity through the binding of PD-L1 expressed on cancer cells and PD-1 expressed on T cells. Therefore, the mixed lymphocyte reaction analysis was performed, and the concentrations of IL-2 and IFN-γ, which are indicators of T cell activation, were measured to evaluate the efficacy of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4).

[0163] 12-3: Evaluation of the activation of phagocytosis by macrophages of humanized single-domain antibodies Phagocytosis is activated by specific binding of antigens and the IgG4 Fc domain. Specific antigen binding is affected by the physical (intercellular distance), chemical (chemokine), and biological (cytokine) environments between the target cell and the effector cell. The PD-L1- and CD47-antigen-specific humanized antibodies specifically bind to hPDL1 and hCD47 expressed on the target cell, the Fc region binding site binds to the Fc receptor on the surface of the effector cell, promotes the interaction between the target cell and the effector cell, and induces phagocytosis by macrophages.

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

[0165] Specifically, 5×10 6Individual THP-1 cells were seeded in a φ100 dish and cultured in an incubator with 5% CO2 for 24 hours. After 24 hours, the cells were treated with 40 nM PMA for 24 hours, the medium was removed, and then the cells were stained with 1 μM deep red pigment. 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, pre-treated cells (Raji cells) stained with 3 μM CFSE, at a ratio of 8:1 and treated with various concentrations of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4). The cells were co-cultured for 4 hours. Then, phagocytosis was evaluated using FACS.

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

[0167] Specifically, CD14 + Using a MACS kit, monocytes were isolated from PBMCs separated from whole blood via a Ficoll gradient, and differentiated into macrophages by treating them with 50 ng / mL M-CSF at a concentration of 2×10 6 cells / mL for 6 days. After differentiation, the macrophages were stained with 5 μM CFSE as effector cells, and the target cells (Raji_PDL1) were stained with 10 μM deep red. The target cells were seeded at 2×10 4 cells / well, and the ratio of effector cells to target cells during co-culture was 8:1. Also, the 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), and phagocytosis was evaluated by confirming the APC-expressing cell population based on the expression of CFSE.

[0168] Example 13: In Vitro Safety Evaluation of Humanized Single-Domain Antibody 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 reaction.

[0169] 13-1: Evaluation of RBC Binding of Humanized Single-Domain Antibody 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) and the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), the RBCs were washed 7 times with DPBS and then diluted to 12% (v / v) with DPBS. 50 μL of the 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 mixture was cultured at 4°C for 1 hour. After washing the cells with DPBS, they were treated with the secondary antibody anti-human IgG4 and reacted at 4°C for 1 hour. Finally, after washing with DPBS, the degree of binding of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) to RBCs was evaluated by FACS.

[0171] 13-2: Evaluation of Hemagglutination Reaction by Humanized Single-Domain Antibody The hemagglutination reaction induced by the specific binding of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 and the CD47 antigen was evaluated.

[0172] Specifically, to evaluate the hemagglutination reaction of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), RBCs were washed 7 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 reacting at room temperature for 1 hour, the hemagglutination reaction was observed with the naked eye.

[0173] Example 14: In Vivo Efficacy Evaluation of Humanized Single-Domain Antibodies The in vivo efficacy of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 was evaluated by confirming the antitumor effect in C57BL / 6 mice and humanized NSG mice bearing tumors formed by the injection of tumor cell lines expressing human PD-L1 and human CD47.

[0174] 14-1: C57BL / 6 Mouse Model Using B16F10_PD-L1_CD47 Tumors C57BL / 6 mice at 6 - 8 weeks of age were injected with 8×10 5 cells / 100 μL of the B16F10 cell line overexpressing human PD-L1 and human CD47. Tumor formation was induced until the tumor reached a size of 3×3 mm in dimensions of length (major axis) × width (minor axis). Subsequently, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was intraperitoneally administered 7 times at a dose of 10 mg / kg at 2-day intervals, and the tumor size was measured. After the last intraperitoneal administration, the tumor volume was measured at 2-day intervals for 2 weeks. The tumor volume was calculated using the formula length (major axis) × width (minor axis) × height (minor axis) / 2.

[0175] 14-2: Humanized NSG Mouse Model Using Raji_PD-L1 Tumors Six - to eight-week-old female NSG mice were intraperitoneally injected with 2×10 7 cells / 100 μL of human PBMC to generate humanized NSG mice. Subsequently, 2×10 of the human Raji (PD-L1) lymphoma cell line that induces overexpression of human PD-L16 Cells / 100 μL were injected. Tumor formation was induced until the tumor reached a size of 3×3 mm (length×width). Thereafter, humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) at 10 mpk was intraperitoneally administered at 3-week intervals, and the tumor volume was measured. After the final intraperitoneal administration, the experiment was terminated when the tumor volume reached 2000 mm 3 ³.

[0176] 14-3: Humanized NSG mouse model using MDA-MB-231 tumors Female NOD scid NSG mice at 6-8 weeks of age were intravenously injected with 2×10 7 cells / 100 μL of human PBMC to generate humanized NSG mice. After confirming the expression of hCD45 and mCD45, 2×10 6 cells / 100 μL of the human breast cancer cell line MDA-MB-231 were subcutaneously administered to the right flank. Tumor formation was induced until the tumor reached a size of 3×3 mm (length×width). Thereafter, humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was intraperitoneally administered at doses of 1, 5, and 10 mpk three times a week, and the tumor volume was measured. Furthermore, when the tumor volume of the humanized NSG mice reached 100 mm 3 ³, Kaplan-Meier survival analysis was performed to predict the survival rate based on the mortality rate of the animals.

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

[0178] Using human PD-L1 antigen as the immunizing antigen, a single-domain antibody clone specific for the PD-L1 antigen was selected by the same method as in Example 5 above, 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] Also, using the selected anti-PD-L1 sdAb (PDL1 Nb#01), a monovalent single-domain antibody specific for PD-L1 containing the human IgG1 Fc 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 the human IgG1 Fc domain is shown in Table 7 below.

[0180] Also, using the selected anti-PD-L1 sdAb (PDL1 Nb#01) clone, a bivalent single-domain antibody specific for PD-L1 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 the 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] By the methods described in Example 9 and Example 10 above, 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.

[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 a PD-1 / PD-L1 interaction inhibitory ability (IC50: 4.60 nM) in CHO-K1_PD-L1 cells (CHO-K1 cells in which the PD-L1 antigen is constantly expressed).

[0189] Also, as shown in Figures 2A and 2B, the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) showed an antigen-binding ability (EC50) of 1.93 nM and a PD-1 / PD-L1 interaction inhibitory ability (IC50) of 2.86 nM in CHO-K1_PD-L1 cells (CHO-K1 cells in which the PD-L1 antigen is constantly expressed).

[0190] 1-3: Evaluation of the affinity of anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) and anti-PD-L1 bivalent HCAb (PP Nb-IgG4) for the immunizing antigen

[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 by the method described in Example 10 above.

[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, and 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 and in vitro Characterization of Anti-CD47 HCAb (CD47 Nb-IgG4) 2-1: Preparation of Anti-CD47 HCAb (CD47 Nb-IgG4) Using human CD47 antigen as the immunogenic antigen, a single-domain antibody clone specific for the CD47 antigen was selected and subjected to sequence analysis in the same manner as in Example 5. The amino acid sequences of the selected anti-CD47 sdAb (CD47 Nb#01) are shown in Tables 11 and 12 below.

[0195] Also, using the selected anti-CD47 sdAb (CD47_Nb_#01), a single-domain antibody specific for CD47 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 the 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 of anti-CD47 HCAb (CD47 Nb-IgG4) and its ability to inhibit CD47 / SIRPα interaction

[0200] The antigen-binding ability (Figure 3A) and the ability to inhibit CD47 / SIRPα interaction (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 the methods 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 in Expi-CHO_CD47 cells (Expi-CHO cells that stably express CD47 antigen), and an inhibitory ability (IC50) of 7.28 nM against the CD47 / SIRPα interaction.

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

[0203] The affinity between the anti-CD47 HCAb (CD47 Nb-IgG4) purified in Experimental Example 2-1 and the CD47 antigen was evaluated in the same manner 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 immunogen 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 immunogen binding ability of anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4)

[0211] The antigen binding abilities of the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) purified in Experimental Example 3-1 to the PD-L1 antigen and CD47 were 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 constitutively express the PD-L1 antigen) and an antigen-binding ability (EC50) of 18.80 nM in Expi-CHO_CD47 cells (Expi-CHO cells that constitutively 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 the PD-1 / PD-L1 and CD47 / SIRPα interactions was evaluated by FACS in the same manner as in Example 10 above.

[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 antigens 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 above.

[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 Evaluation of Its Characteristics in vitro

[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 substituting amino acids through site-specific mutagenesis in the same manner as in Example 7, and was named humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4).

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

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

[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 through site-specific mutagenesis in the same manner as described 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 through site-specific mutagenesis in the same manner as described 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 Immune Antigen Binding Ability of Humanized Anti-PD-L1×CD47 Trivalent HCAb (hPPC Nb-IgG4)

[0229] Using the same method 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 by FACS.

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

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

[0232] Using the same method as in Example 10 above, the inhibitory ability 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 by FACS.

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

[0234] 4-4: Evaluation of the affinity of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) for the immunizing antigen

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

[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 at 6.77 nM and for the CD47 antigen at 3.34 nM.

[0237]

Table 22

[0238] 4 - 5: Evaluation of the 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 by the same method as in Example 11 above.

[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 for human PD-L1 and human CD47 antigens, respectively, and antigen affinities of 8.17 nM and 6.91 nM for cynomolgus monkey PD-L1 and cynomolgus monkey CD47 antigens, respectively. No antigen affinity was observed for 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 the 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 concentration of IL-2 expressed from Jurkat cells. It was performed by inhibiting the interaction between CHO-K1 cells expressing PD-L1 and Jurkat cells expressing PD-1 in the same manner 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 an efficacy of 1.47 nM (IC50).

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

[0247] The CD4 T cell activation in the mixed lymphocyte reaction of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 was evaluated by performing a mixed lymphocyte reaction (MLR) analysis in the same manner as in Example 12-2. As a comparative example, the humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4), humanized anti-CD47 HCAb (hCD47 Nb-IgG4), and a 1:1 mixture of these antibodies prepared in Experimental Example 4-1 were used. + As a result, as shown in Figures 9A and 9B, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) showed efficacies of 0.08 (EC50) and 0.18 nM (EC50) in the production of IL-2 and IFN-γ of CD4 T cells, respectively. Furthermore, it was confirmed that this antibody showed an efficacy equal to or higher than that of the humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4), humanized anti-CD47 HCAb (hCD47 Nb-IgG4), and their combination therapy.

[0248] As a result, as shown in Figures 9A and 9B, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) + showed efficacies of 0.08 (EC50) and 0.18 nM (EC50) in the production of IL-2 and IFN-γ of CD4 T cells, respectively. Furthermore, it was confirmed that this antibody showed an efficacy equal to or higher than that of the humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4), humanized anti-CD47 HCAb (hCD47 Nb-IgG4), and their combination therapy.

[0249] 5-3: Evaluation of the phagocytosis 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 prey cells (Raji cells) expressing the CD47 antigen 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) treatment group showed a significantly superior phagocytosis effect compared to the control group.

[0252] The degree of phagocytosis activity by macrophages was evaluated through the selective binding of 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 in the same manner as in Example 12-4. As comparative examples, the humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4), humanized anti-CD47 HCAb (hCD47 Nb-IgG4), and a 1:1 mixture thereof prepared in Experimental Example 4-1 were used.

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

[0254] Experimental Example 6: In Vitro Safety Evaluation of Humanized Anti-PD-L1×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] Using the same method as in Example 13-1, 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.

[0257] As a result, as shown in Figure 11, unlike the CD47 monoclonal antibody used as the positive control group, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) 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] Using the same method as in Example 13-2, the hemagglutination reaction of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 was evaluated.

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

[0261] Experimental Example 7: In Vivo Efficacy Evaluation 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 above was evaluated in a C57BL / 6 mouse model in which tumors were formed by injecting tumor cells (b16F10 cells) expressing the pre-PD-L1 and CD47 antigens, a humanized NSG mouse model in which tumors were formed by injecting tumor cells (Raji_PD-L1 cells) expressing the PD-L1 and CD47 antigens, and a humanized NSG mouse model in which tumors were formed by injecting the human breast cancer cell line MDA-MB-231, in the same manner as in Example 14 below.

[0263] As a result, as shown in Fig. 13, the 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 the C57BL / 6 mouse model with tumors.

[0264] Also, as shown in Fig. 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 the humanized NSG mouse model with tumors.

[0265] Furthermore, as shown in Figs. 15A and 15B, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) showed an antitumor effect of approximately 81.98% compared to the negative control group (isotype) in the humanized NSG mouse model with tumors. Also, the expected survival rate of the group administered with 10 mpk of the 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 that binds specifically to both PD-L1 and CD47 of the immune checkpoint protein according to the present invention exhibits efficacy both in vitro and in vivo, and thus can 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 an antigen-binding fragment thereof; and A second humanized single domain antibody (second hsdAb) that specifically binds to CD47 or an antigen-binding fragment thereof; A bispecific antibody that binds bispecifically to PD-L1 and CD47, comprising the above.

2. The bispecific antibody that binds bispecifically to PD-L1 and CD47 according to claim 1, wherein the first hsdAb or an antigen-binding fragment thereof 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.

3. The bispecific antibody that binds bispecifically to PD-L1 and CD47 according to claim 1, wherein the second hsdAb or an antigen-binding fragment thereof comprises: 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.

4. The bispecific antibody that binds bispecifically to PD-L1 and CD47 according to claim 1, wherein the first hsdAb or the second hsdAb comprises a heavy chain variable domain framework sequence having at least 95% sequence identity with the sequence of SEQ ID NO:

16.

5. The bispecific antibody that binds bispecifically to PD-L1 and CD47 according to claim 1, wherein the first hsdAb or an antigen-binding fragment thereof, or the second hsdAb or an antigen-binding fragment thereof, comprises a heavy chain variable domain framework sequence having at least 85% sequence identity with the sequence of SEQ ID NO:

17.

6. The bispecific antibody that binds bispecifically to PD-L1 and CD47 according to claim 1, wherein the first hsdAb or an antigen-binding fragment thereof, or the second hsdAb or an antigen-binding fragment thereof, comprises a heavy chain variable domain framework sequence having 100% sequence identity with the sequence of SEQ ID NO: 16 or 17.

7. The bispecific antibody that binds bispecifically to PD-L1 and CD47 according to claim 6, wherein the first hsdAb or an antigen-binding fragment thereof comprises the following heavy chain variable domain framework sequence: (1) FR1 consisting of the amino acid sequence represented by any one of SEQ ID NO: 19 or 23; ​ ​ (2) FR2 consisting of the amino acid sequence represented by either SEQ ID NO: 20 or 24; (3) FR3 consisting of the amino acid sequence represented by either SEQ ID NO: 21 or 25; and (4) FR4 consisting of the amino acid sequence represented by either SEQ ID NO: 22 or 26

8. The bispecific antibody according to claim 7, wherein the first hsdAb or an antigen-binding fragment thereof specifically binds to both PD-L1 and CD47 and comprises the following heavy chain variable domain framework sequences: FR1 consisting of the amino acid sequence represented by SEQ ID NO: 19; FR2 consisting 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

9. The bispecific antibody according to claim 6, wherein the second hsdAb or an antigen-binding fragment thereof specifically binds to both PD-L1 and CD47 and comprises the following heavy chain variable domain framework sequences: (1) FR1 consisting of the amino acid sequence represented by either SEQ ID NO: 19 or 23; (2) FR2 consisting of the amino acid sequence represented by either SEQ ID NO: 20 or 24; (3) FR3 consisting of the amino acid sequence represented by either SEQ ID NO: 21 or 25; and (4) FR4 consisting of the amino acid sequence represented by either SEQ ID NO: 22 or 26

10. The bispecific antibody according to claim 9, wherein the second hsdAb or an antigen-binding fragment thereof specifically binds to both PD-L1 and CD47 and comprises the following heavy chain variable domain framework sequences: FR1 consisting of the amino acid sequence represented by SEQ ID NO: 23; FR2 consisting 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 consisting of the amino acid sequence represented by SEQ ID NO: 26

11. The bispecific antibody according to claim 1, wherein the first hsdAb consists of the amino acid sequence represented by SEQ ID NO: 16 and specifically binds to both PD-L1 and CD47.

12. The bispecific antibody according to claim 7, wherein the second hsdAb consists of the amino acid sequence represented by SEQ ID NO: 17 and specifically binds to both PD-L1 and CD47.

13. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 1, wherein the first hsdAb or an antigen-binding fragment thereof is fused to a second hsdAb or an antigen-binding fragment thereof via a peptide linker.

14. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 13, wherein the first hsdAb and the second hsdAb fused via the peptide linker consist of the amino acid sequence represented by SEQ ID NO:

15.

15. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 1, which is an antibody consisting only of a heavy chain (HCAb) in which an Fc fragment is fused to the first hsdAb or an antigen-binding fragment thereof, or the second hsdAb or an antigen-binding fragment thereof.

16. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 15, wherein the HCAb is a monomer or a multimer.

17. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 15, wherein the Fc fragment is human IgG1, IgG2, IgG3, or IgG4.

18. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 15, wherein the first hsdAb or an antigen-binding fragment thereof and the second hsdAb or an antigen-binding fragment thereof are fused to an Fc fragment via a peptide linker.

19. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 15, wherein the HCAb consists of the amino acid sequence represented by SEQ ID NO:

18.

20. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 15, which contains at least one or more amino acid substitutions.

21. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 20, wherein at least one or more amino acid substitutions are conservative substitutions.

22. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 21, wherein at least one amino acid substitution is a substitution with a non-genetically encoded amino acid or a synthetic amino acid of an amino acid.

23. The bispecific antibody that binds specifically to both PD-L1 and CD47 according to claim 1, which is bound to an immunomodulator, a cytokine, a cytotoxic agent, a chemotherapeutic agent, a diagnostic agent, an antiviral agent, an antimicrobial agent, or a drug.

24. An antibody complex comprising a bispecific antibody that binds bispecifically to the PD-L1 and CD47 according to claim 1, which is conjugated to an immunomodulator, cytokine, cytotoxic agent, chemotherapeutic agent, diagnostic agent, antiviral agent, antimicrobial agent or drug.

25. A nucleic acid molecule encoding a bispecific antibody that binds bispecifically to the PD-L1 and CD47 according to claim 1.

26. An expression vector comprising the nucleic acid molecule according to claim 25.

27. A host cell transformed with the expression vector according to claim 26.

28. (a) culturing the host cell according to claim 26 under conditions that allow the expression of the bispecific antibody; and (b) recovering the expressed bispecific antibody; A method for producing a bispecific antibody that binds bispecifically to PD-L1 and CD47, comprising the steps of:

29. A pharmaceutical composition for the prevention or treatment of cancer, comprising as an active ingredient the bispecific antibody that binds bispecifically to the PD-L1 and CD47 according to claim 1 or the antibody complex according to claim 24.

30. The pharmaceutical composition according to claim 29, wherein 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 cancer, 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.

31. The pharmaceutical composition according to claim 29, wherein the pharmaceutical composition further comprises a pharmaceutically acceptable carrier.

32. A method for the prevention or treatment of cancer, comprising the step of administering to an individual the bispecific antibody that binds bispecifically to the PD-L1 and CD47 according to claim 1 or the antibody complex according to claim 24.

33. Use of the bispecific antibody that binds bispecifically to the PD-L1 and CD47 according to claim 1 or the antibody complex according to claim 24 for use in a pharmaceutical composition for the prevention or treatment of cancer.

34. Use of the bispecific antibody that binds bispecifically to the PD-L1 and CD47 according to claim 1 or the antibody complex according to claim 24 for the manufacture of a pharmaceutical composition for the prevention or treatment of cancer.

Images