Bispecific humanized single-domain antibody aiming at PD-L1 and CD47 and application of bispecific humanized single-domain antibody

Abstract

The invention relates to a bispecific humanized single-domain antibody aiming at PD-L1 and CD47 and application thereof, in particular to a humanized single-domain antibody which is bispecifically combined with PD-L1 and CD47 serving as immune checkpoint protein, and the in-vitro and in-vivo effects of the humanized single-domain antibody are confirmed. Therefore, the bispecific humanized single-domain antibody can be used as an immune checkpoint inhibitor to be effectively applied to immune anti-cancer therapy.

Description

Technical Field

[0001] The present invention relates to a bispecific humanized single-domain antibody targeting immune checkpoint proteins PD-L1 and CD47 and uses thereof. Background Art

[0002] In recent years, as the efficacy of newly developed immune anticancer therapies that utilize the human immune system has been confirmed, existing anticancer therapies using chemotherapy and targeted therapeutics are shifting to immune anticancer therapies using immunotherapeutics.

[0003] Basically, the immune cells of cancer patients develop tolerance to cancer antigens. Although they can recognize cancer cells, their functions are suppressed and they cannot effectively remove cancer cells. The core of immunotherapy is to induce activated immune cells by awakening tolerant immune cells, thereby destroying cancer cells. This immunotherapy includes cytokine therapeutics such as IFN-γ and IL-2, cancer vaccines using dendritic cells, cell therapeutics using T cells, and immune checkpoint inhibitors (ICI) that block immune checkpoint proteins. This therapeutic agent is generally called tumor immunotherapy (immuno-oncology therapy). Among them, the immune anticancer agents that multinational pharmaceutical companies are competing to develop are immune checkpoint inhibitors.

[0004] Immune checkpoint proteins are cell membrane proteins that inhibit the differentiation, proliferation, and activity of immune cells. Specifically, this protein is usually expressed in activated T cells, which reduces T cell proliferation, cytokine secretion, cytotoxicity, and has the function of inhibiting the excessive activity of T cells, so it is also called a co-inhibitory molecule. Usually, there are co-inhibitory receptors CTLA-4 and PD-1 expressed in T cells, so the activity of T cells is regulated by binding to B7.1 / 2 and PD-L1 as each ligand. At the same time, PD-L1 is expressed in cancer cells, which inactivates cancer-specific T cells and induces apoptosis, thereby protecting cancer cells from immune attacks by T cells, thereby serving as an immune escape mechanism for cancer. In addition, the report pointed out that the prognosis of cancer patients with ectopic expression of PD-L1 in cancer cells is worse than that of cancer patients without ectopic expression of PD-L1.

[0005] Immune checkpoint inhibitors are drugs that activate T cells to attack cancer cells by blocking the activity of immune checkpoint proteins (which are involved in the inhibition of such T cells), and antibodies that recognize CTLA-4, PD-1, and PD-L1 are usually used. Ipilimumab (Yervoy) is a CTLA-4 inhibitor, and among immune checkpoint inhibitors, it was the first second-line treatment for metastatic melanoma approved by the US FDA in 2011. Later in 2014, PD-1 blockers nivolumab (Opdivo) and pembrolizumab (Keytruda) were approved by the US FDA for metastatic melanoma. Subsequently, atezolizumab (Tecentriq), an immune checkpoint inhibitor targeting PD-L1, was approved by the US FDA for bladder cancer (2016), and in 2017 avelumab (Bavencio) and durvalumab (Imfinzi) were approved by the US FDA for the treatment of Merkel cell carcinoma (a type of skin cancer) and bladder cancer, respectively. In 2018, the PD-1 inhibitor cemiplimab (Libtayo) was approved by the US FDA for squamous cell carcinoma. Currently, these drugs are expanding their therapeutic indications and obtaining US FDA approval for more cancer treatments. As of 2019, 6 PD-1 / PD-L1 inhibitors have been approved by the FDA for a total of 18 cancers. In addition, immunomodulatory proteins such as B7-H4, ICOS, HVEM, PDL-2, PVRIG, etc. are in preclinical trials as new targets. Most of these therapeutic agents target targets expressed on T cells.

[0006] In order to overcome this biased target research trend centered on T cells, inhibitors targeting immune checkpoint proteins (which are expressed in myeloid cells such as macrophages and dendritic cells) are being developed in recent years. Among them, CSF1R, CD47, and TLR7 are becoming important targets.

[0007] Immune checkpoint proteins allow tumor cells to escape the attack of the immune system by inhibiting the activity of T cells. Programmed Death-Ligand 1 (PD-L1) is an immune checkpoint protein that is mainly expressed in leukocytes and non-hematopoietic cells of the lymphatic system and non-lymphoid tissues, and is also expressed on the surface of various tumor cells such as colon cancer, pancreatic cancer, melanoma, and cervical cancer. In particular, it interacts with programmed death-1 (PD-1) expressed on the surface of activated T cells, thereby causing TCR-mediated T-cell activation and inhibiting T-cell proliferation, thereby negatively regulating the immune response of T cells.

[0008] Cluster of Differentiation 47 (CD47), which was first identified as a tumor antigen in human ovarian cancer in 1980, is expressed in a variety of human tumor cells such as non-Hodgkin's 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, thrombospondin (TSP1) and integrins, thereby participating in cell death, proliferation and immune response. In particular, CD47 expressed in tumor cells interacts with SIRPα expressed on the surface of macrophages, sending a "don't eat me" signal, which not only evades the phagocytosis of macrophages, but also inhibits vascular proliferation and the action of effector T cells in the tumor environment, thereby promoting the proliferation and growth of tumor cells.

[0010] Magrolimab of the multinational pharmaceutical company Gilead is in the clinical trial stage as an antibody therapeutic targeting CD47, and in recent years multinational pharmaceutical companies are actively developing dual antibodies based on CD47 antibody therapeutics.

[0011] On the other hand, antibodies of non-human origin are often immunogenic, thus limiting their effectiveness and in some cases may cause dangerous allergic reactions. The immune response to such foreign antibodies promotes their clearance from the body and inhibits their ability to bind to the target antigen, thereby substantially reducing the effectiveness of the antibody. To overcome this problem, non-human antibodies can be humanized to reduce their immunogenicity to humans while maintaining the specificity and affinity of the parent non-human antibody.

[0012] In response to this, the inventors have developed a humanized antibody as an immune checkpoint inhibitor targeting immune checkpoint proteins PD-L1 and CD47, and thus proposed the present application.

[0013] [Prior art literature]

[0014] [Patent Document]

[0015] U.S. Patent Publication No. 2020-0181259

[0016] International Patent Publication No. 2011-143624

[0017] U.S. Patent No. 8,008,449

[0018] [Non-patent literature]

[0019] Pol Specenier(2016):Ipilimumab in melanoma,Expert Review of Anticancer Therapy.

[0020] DB Johnson,C Peng et al.Nivolumab in melanoma:latest evidence and clinical potential.Ther Adv Med Oncol 2015,Vol.7(2)97-106

[0021] Liu J, Wang L, Zhao F, Tseng S, Narayanan C, Shura L, et al. (2015) Pre-Clinical Development of a Humanized Anti-CD47 Antibody with Anti-CancerTherapeutic Potential.PLoS ONE 10(9):e0137345.

[0022] C Vincke et al.General Strategy to Humanize a Camelid Single-domainAntibody and Identification of a Universal Humanized Nanobody Scaffold.TheJournal of Biological Chemistry Vol.284,NO.5,3273-3284,January 30,2009. Summary of the invention

[0023] Technical issues

[0024] The present invention aims to provide a bispecific humanized single-domain antibody targeting immune checkpoint proteins PD-L1 and CD47 and uses thereof.

[0025] Technical Solution

[0026] In order to achieve the purpose of the present invention, the present invention provides a bispecific antibody that bispecifically binds to PD-L1 and CD47, comprising: a first humanized single domain antibody (first hsdAb) or an antigen-binding fragment thereof that specifically binds to PD-L1; and a second humanized single domain antibody (second hsdAb) or an antigen-binding fragment thereof that specifically binds to CD47.

[0027] In one embodiment of the present invention, the first hsdAb or antigen-binding fragment thereof may include a CDR1 consisting of the amino acid sequence shown in SEQ ID NO:2; a CDR2 consisting of the amino acid sequence shown in SEQ ID NO:3; and a CDR3 consisting of the amino acid sequence shown in SEQ ID NO:4.

[0028] Furthermore, the second hsdAb or antigen-binding fragment thereof may include a CDR1 consisting of the amino acid sequence shown in SEQ ID NO:9; a CDR2 consisting of the amino acid sequence shown in SEQ ID NO:10; and a CDR3 consisting of the amino acid sequence shown in SEQ ID NO:11.

[0029] Furthermore, the first hsdAb or the second hsdAb may include a heavy chain variable domain framework sequence having at least 95% sequence identity to SEQ ID NO:16, and a heavy chain variable domain framework sequence having at least 85% sequence identity to SEQ ID NO:17.

[0030] Furthermore, the first hsdAb or the second hsdAb may comprise a heavy chain variable domain framework sequence having 100% sequence identity to the sequence of SEQ ID NO:16 or SEQ ID NO: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 shown in any one of SEQ ID NOs: 19 and 23; (2) FR2 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 20 and 24; (3) FR3 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 21 and 25; and (4) a heavy chain variable domain framework sequence of FR4 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 22 and 26. More specifically, the first hsdAb or its antigen-binding fragment may include: FR1 consisting of the amino acid sequence shown in SEQ ID NO: 19; FR2 consisting of the amino acid sequence shown in SEQ ID NO: 20; FR3 consisting of the amino acid sequence shown in SEQ ID NO: 21; and a heavy chain variable domain framework sequence of FR4 consisting of the amino acid sequence shown in SEQ ID NO: 22. The second hsdAb or its antigen-binding fragment may include: FR1 consisting of the amino acid sequence shown in SEQ ID NO: 23; FR2 consisting of the amino acid sequence shown in SEQ ID NO: 24; and FR3 consisting of the amino acid sequence shown in SEQ ID NO: 25. The first hsdAb comprises an amino acid sequence as shown in SEQ ID NO: 16, and the second hsdAb comprises an amino acid sequence as shown in SEQ ID NO: 17.

[0031] In addition, the first hsdAb or its antigen binding fragment and / or the second hsdAb or its antigen binding fragment can be monovalent, divalent, trivalent, tetravalent or more. In addition, the first hsdAb or its antigen binding fragment can be fused to each other via a peptide linker. In addition, the second hsdAb or its antigen binding fragment can be fused to each other via a peptide linker.

[0032] In one aspect of the present invention, the first hsdAb or antigen-binding fragment thereof and the second hsdAb or antigen-binding fragment thereof may be fused to each other via a peptide linker, and in some embodiments comprise the amino acid sequence shown in SEQ ID NO:15.

[0033] In one aspect of the present invention, a heavy chain-only antibody (hcab) is provided, wherein an Fc fragment is fused to the first hsdAb or antigen-binding fragment thereof or the second hsdAb or antigen-binding fragment thereof.

[0034] In one aspect of the present invention, it may be a bispecific and multivalent (e.g., bivalent, trivalent, tetravalent or higher) HCAb comprising two or more copies of the first hsdAb or its antigen-binding fragment and / or the second hsdAb or its antigen-binding fragment and fused with an Fc fragment. In some embodiments, the HCAb may consist of the amino acid sequence shown in SEQ ID NO: 18.

[0035] In one aspect of the 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.

[0036] In one aspect of the present invention, the HCAb may include at least one amino acid substitution, which may be a conservative substitution and may be a substitution of an amino acid with a non-genetically encoded amino acid or a synthetic amino acid.

[0037] In one aspect of the invention, an immunomodulator, cytokine, cytotoxic drug, chemotherapeutic agent, diagnostic agent, antiviral agent, antimicrobial agent or drug may be conjugated. To this end, the invention provides an antibody conjugate comprising the bispecific antibody that bispecifically binds to PD-L1 and CD47 conjugated to an immunomodulator, cytokine, cytotoxic drug, chemotherapeutic agent, diagnostic agent, antiviral agent, antimicrobial agent or drug.

[0038] In addition, the present invention provides a nucleic acid molecule encoding the bispecific antibody that bispecifically binds to PD-L1 and CD47.

[0039] In addition, the present invention provides an expression vector comprising the nucleic acid molecule.

[0040] In addition, the present invention provides a host cell transformed by the expression vector.

[0041] In addition, the present invention provides a method for producing a bispecific antibody that bispecifically binds to PD-L1 and CD47.

[0042] It includes:

[0043] (a) culturing the host cell under conditions such that the bispecific antibody is expressed; and

[0044] (b) Recovering the expressed bispecific antibody.

[0045] In addition, the present invention provides the following: a pharmaceutical composition for preventing or treating cancer, comprising the bispecific antibody bispecifically binding to PD-L1 and CD47 or the antibody conjugate as an active ingredient; a method for preventing or treating cancer, comprising administering the bispecific antibody bispecifically binding to PD-L1 and CD47 or the antibody conjugate to an individual; use of the bispecific antibody bispecifically binding to PD-L1 and CD47 or the antibody conjugate in preventing or treating cancer; and use of the bispecific antibody bispecifically binding to PD-L1 and CD47 or the antibody conjugate in preparing a composition for preventing or treating cancer.

[0046] In one aspect of the invention, the cancer can be selected from the group consisting of melanoma, lung cancer, liver cancer, glioblastoma, ovarian cancer, colon cancer, head and neck cancer, bladder cancer, renal cell carcinoma, gastric cancer, breast cancer, metastatic cancer, prostate cancer, pancreatic cancer, non-Hodgkin's lymphoma, Hodgkin's lymphoma, multiple myeloma, leukemia, lymphoma, myelodysplastic syndrome, acute lymphocytic leukemia, acute myeloid leukemia, chronic lymphocytic leukemia, chronic myeloid leukemia, solitary myeloma and aplastic anemia.

[0047] In one aspect of the present invention, the pharmaceutical composition may further include a pharmaceutically acceptable carrier.

[0048] Effects of the Invention

[0049] In the present invention, a humanized single-domain antibody that bispecifically binds to the immune checkpoint proteins PD-L1 and CD47 was prepared, and its in vitro and in vivo efficacy was confirmed. Therefore, the bispecific humanized single-domain antibody can be effectively used as an immune checkpoint protein in immune anti-cancer therapy. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1A This is a graph showing the binding ability (EC50) of the PD-L1 antigen expressed on the cell surface by reacting the CHO-K1_PD-L1 cell line (CHO-K1 cells that induce the expression of the PD-L1 antigen) with different concentrations of the anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) prepared according to one embodiment of the present invention.

[0051] Figure 1B This is a graph confirming the inhibitory ability (IC50) of the anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) prepared according to one embodiment of the present invention on the PD-L1 / PD-1 interaction after combining the CHO-K1_PD-L1 cell line (CHO-K1 cells induced to express the PD-L1 antigen) with the PD-1 protein.

[0052] Figure 2A This is a graph confirming the binding ability (EC50) to the PD-L1 antigen expressed on the cell surface by reacting the CHO-K1_PD-L1 cell line (CHO-K1 cells that induce the expression of the PD-L1 antigen) with different concentrations of the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) prepared according to one embodiment of the present invention.

[0053] Figure 2B This is a graph confirming the inhibitory ability (IC50) of the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) prepared according to one embodiment of the present invention on the PD-L1 / PD-1 interaction after combining the CHO-K1_PD-L1 cell line (CHO-K1 cells induced to express the PD-L1 antigen) with the PD-1 protein.

[0054] Figure 3A This is a graph showing the binding ability (EC50) of the CD47 antigen expressed on the cell surface by reacting the Expi-CHO_CD47 cell line (Expi-CHO cells that induce CD47 antigen expression) with different concentrations of anti-CD47 HCAb (CD47 Nb-IgG4) prepared according to one embodiment of the present invention.

[0055] Figure 3B This is a graph confirming the inhibitory ability (IC50) of the anti-CD47 HCAb (CD47 Nb-IgG4) prepared according to one embodiment of the present invention on the CD47 / SIRPalpha interaction after combining the Expi-CHO_CD47 cell line (Expi-CHO cells induced to express CD47 antigen) with SIRPalpha protein.

[0056] Figure 4A This is a graph confirming the binding ability (EC50) to the PD-L1 antigen expressed on the cell surface by reacting the CHO-K1_PD-L1 cell line (CHO-K1 cells induced to express the PD-L1 antigen) with different concentrations of anti-PD-L1×CD47HCAb (PPC Nb-IgG4) prepared according to one embodiment of the present invention.

[0057] Figure 4B This is a graph confirming the binding ability (EC50) to the CD47 antigen expressed on the cell surface by reacting the Expi-CHO_CD47 cell line (Expi-CHO cells that induce CD47 antigen expression) with different concentrations of anti-PD-L1×CD47 HCAb (PPC Nb-IgG4) prepared according to one embodiment of the present invention.

[0058] Figure 5AThis is a graph confirming the inhibitory ability (IC50) of the anti-PD-L1×CD47 HCAb (PPC Nb-IgG4) prepared according to one embodiment of the present invention on the PD-L1 / PD-1 interaction after combining the CHO-K1_PD-L1 cell line (CHO-K1 cells induced to express the PD-L1 antigen) with the PD-1 protein.

[0059] Figure 5B This is a graph confirming the inhibitory ability (IC50) of the anti-PD-L1×CD47 HCAb (PPC Nb-IgG4) prepared according to one embodiment of the present invention on the CD47 / SIRPalpha interaction after combining the Expi-CHO_CD47 cell line (Expi-CHO cells that induce CD47 antigen expression) with the SIRPalpha protein.

[0060] Fig. 6A This is a graph showing the binding ability (EC50) to the PD-L1 antigen expressed on the cell surface by reacting the CHO-K1_PD-L1 cell line (CHO-K1 cells that induce the expression of the PD-L1 antigen) with different concentrations of the humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention.

[0061] Figure 6B This is a graph confirming the binding ability (EC50) to the CD47 antigen expressed on the cell surface by reacting the Expi-CHO_CD47 cell line (Expi-CHO cells that induce CD47 antigen expression) with different concentrations of humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention.

[0062] Fig. 7A This is a graph confirming the inhibitory ability (IC50) of the humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention on the PD-L1 / PD-1 interaction after combining the CHO-K1_PD-L1 cell line (CHO-K1 cells induced to express the PD-L1 antigen) with the PD-1 protein.

[0063] Figure 7B This is a graph confirming the inhibitory ability (IC50) of the humanized anti-PD-L1×CD47 HCAb (hPPCNb-IgG4) prepared according to one embodiment of the present invention on the CD47 / SIRPalpha interaction after combining the Expi-CHO_CD47 cell line (Expi-CHO cells that induce CD47 antigen expression) with the SIRPalpha protein.

[0064] Figure 8 is a diagram for confirming T cell activity by blocking cell-cell interaction with humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4), wherein the humanized anti-PD-L1×CD47 HCAb is prepared according to one embodiment of the present invention using CHO-K1 (which induces overexpression of PD-L1 protein) and Jurkat cells expressing PD-1.

[0065] Fig.9A and Fig. 9B This is a graph confirming the activation level of CD4+T cells in a mixed lymphocyte reaction of a humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention.

[0066] Fig. 10A and Fig. 10B This is a graph confirming the phagocytic effect of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention on macrophages.

[0067] Fig.11 This is a diagram confirming the human RBC (Red blood cell) binding ability of the humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention.

[0068] Fig.12 This is a diagram confirming the hemagglutination reaction of the humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) prepared according to one embodiment of the present invention.

[0069] Fig.13 The graph confirms the anti-tumor effect of humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) in a C57BL / 6 mouse model in which tumors expressing PD-L1 and CD47 antigens are formed.

[0070] Fig.14 The graph confirms the anti-tumor effect of humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) in a humanized NSG mouse model in which tumors expressing PD-L1 and CD47 antigens are formed.

[0071] Fig.15A and Fig. 15B The graph confirms the anti-tumor effect and survival rate of humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) in a humanized NSG mouse model of tumor formation induced by the human breast cancer cell line MDA-MB-231. DETAILED DESCRIPTION

[0072] Hereinafter, the embodiments of the present invention will be described in detail so that those skilled in the art can easily implement the present invention. The embodiments of the present invention are used to further fully describe the present invention to those of ordinary skill in the art. Therefore, the embodiments of the present invention can be modified into various other forms, and the scope of the present invention is not limited to the embodiments described below.

[0073] In the present invention, the term "epitope" refers to a protein determinant that can specifically bind to an antibody. An epitope is usually composed of chemically active surface groups of molecules (e.g., amino acids or sugar side chains) and usually has specific three-dimensional structural characteristics and specific charge characteristics.

[0074] The term "treat" means to slow down, interrupt, stop, control, stop, alleviate or improve the disorder or disease disclosed in this specification (e.g., disease symptoms or complications), or reverse its progress, but does not necessarily mean that all disease or disorder symptoms are completely removed, but refers to the entire process.

[0075] The term "preventing" means delaying the disease or disorder, eg, prophylactic treatment of a disease, or delaying the onset or progression of a disease or disorder.

[0076] The term "individual" or "subject" refers, without limitation, to mammals including humans, cows, horses, cats, dogs, rodents, or primates. In some embodiments, the individual is a human.

[0077] The term "antibody" is used in its broadest sense, including but not limited to monoclonal antibodies, polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), full-length antibodies, and various antibody structures of antigen-binding fragments thereof, as long as the desired antigen-binding activity can be exhibited. The term "antibody" includes existing four-chain antibodies, single-domain antibodies, and antigen-binding fragments thereof.

[0078] The basic four-chain antibody unit is a heterotetrameric glycoprotein composed of two identical light (L) chains and two identical heavy (H) chains. IgM antibodies are formed by an additional polypeptide called J chain and five basic heterotetrameric units and contain 10 antigen binding sites. In contrast, IgA antibodies include 2 to 5 basic four-chain units that can be polymerized in order to form multivalent aggregates by combining with J chains. For IgG, the four-chain unit is usually about 150,000 Daltons. Each L chain is connected to the H chain by one covalent disulfide bond, and the two H chains are connected to each other by one or more disulfide bonds depending on the H chain isotype. Each H chain and L chain also has regularly spaced intrachain disulfide bridges. Each H chain has three constant domains (CH) at the N-terminus, followed by a variable domain (VH) in the α chain and the γ chain, and four CH domains in the μ isotype and the ε isotype. Each L chain has a constant domain at the N-terminus and a variable domain (VL) attached at the other end. VL is aligned with VH, and CL is aligned with the first constant domain (CH1) of the heavy chain. The pairing of VH and VL simultaneously forms a single antigen binding site. In any vertebrate, the L chains are arranged in one of two distinct types called kappa or lambda based on the amino acid sequence of their constant domains. Immunoglobulins can be arranged in different classes or isotypes depending on the amino acid sequence of the constant domain (CH) of their heavy chains. There are five classes of immunoglobulins, α, δ, ε, γ, and μ, which are IgA, IgD, IgE, IgG, and IgM, respectively, with designated heavy chains. The γ and α classes are further divided into subclasses based on a few differences in CH sequence and function, for example, humans express the following subclasses: IgG1, IgG2A, IgG2B, IgG3, IgG4, IgA1, and IgA2.

[0079] The term "heavy chain only antibody" or "HCAb" refers to a functional antibody that contains heavy chains but lacks the light chains typically found in four-chain antibodies.

[0080] The term "single domain antibody", "nanoantibody" or "sdAb" refers to a single antigen-binding polypeptide having three complementary determining regions (CDRs). sdAb can bind to an antigen alone without pairing with a corresponding CDR-containing polypeptide. In some cases, single domain antibodies are obtained by engineering camelid HCAbs, whose heavy chain variable domains are referred to as "VHH" (variable domains of the heavy chain of a heavy chain antibody) in this application. The basic VHH has the following structure from N-terminus to C-terminus: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4, wherein FR1 to FR4 refer to framework regions 1 to 4, respectively, and CDR1 to CDR3 refer to complementary determining regions 1 to 3, respectively.

[0081] The "variable region" or "variable domain" of an antibody refers to the amino-terminal domain of the heavy or light chain of an antibody. The variable domains of the heavy and light chains may be referred to as "VH" and "VL", respectively. These domains are generally the largest variable part of an antibody (compared to other antibodies of the same class) and contain the antigen binding site. Heavy chain-only antibodies derived from Camelidae have a single heavy chain variable region referred to as "VHH".

[0082] The term "variable" refers to the fact that the specific segments of the variable domains are widely different in the sequence in the antibody. The V domain mediates antigen binding and defines the specificity of a specific antibody for a specific antigen. However, variability is not evenly distributed in the overall range of the variable domain. Instead, it is concentrated in 3 segments referred to as complementary determining regions (CDRs) or hypervariable regions (HVRs) in both the heavy chain variable domain and the light chain variable domain. The further highly protected part in the variable domain is referred to as the framework region (FR). The variable domain of each of the natural heavy and light chains includes 4 FR regions, which mainly adopt a β-folded structure connected by 3 CDRs to form a ring connection, and form a part of the β-folded structure in some cases. In each chain, the CDR is kept in a very close state by the FR region, and the CDR plays a role in the formation of the antigen binding site of the antibody from other chains (Kabat, Elvin A., Sequence of Immunological Interest, Fifth Edition, National Institute of Health, Bethesda, Md. (1991)). The constant domains are not directly involved in the binding of the antibody to the antigen, but exhibit various effector functions, such as participation of the antibody in antibody-dependent cellular cytotoxicity.

[0083] The term "constant domain" is the other part of the immunoglobulin containing the antigen binding site, and refers to the part of the immunoglobulin molecule that has more protected amino acid sequences than the variable domain. The constant domain contains the CH1, CH2 and CH3 domains (collectively referred to as CH) of the heavy chain and the CHL (or CL) domain of the light chain.

[0084] The terms "full-length antibody", "complete antibody" or "whole antibody" are used interchangeably with antibody fragments to refer to antibodies in substantially complete form. Specifically, full-length 4-chain antibodies include antibodies having heavy chains and light chains that include an Fc region. Full-length heavy-chain-only antibodies include heavy chain variable domains (e.g., VHH) and Fc regions. The constant domains may be native sequence constant domains (e.g., human native sequence constant domains) or amino acid sequence variants thereof. In some cases, a complete antibody may have one or more effector functions.

[0085] "Antibody fragments" or "antigen-binding fragments" include a portion of an intact antibody, preferably including the antigen-binding and / or variable regions of the intact antibody. Non-limiting examples of antibody fragments include 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 smallest antibody fragment containing a complete antigen recognition site and an antigen binding site. The fragment consists of a dense dimer, which is a dimer of one heavy chain and one light chain variable region domain in non-covalent bonding. "Single-chain Fv" or "sFv" or "scFv" for short is an antibody fragment that includes VH and VL antibody domains connected to a single polypeptide chain. Preferably, the scFv polypeptide additionally includes a polypeptide linker between the VH and VL domains, which enables the scFv to form the desired structure for binding to the antigen. "Diabodies" refer to small antibody fragments that are made by constructing sFv fragments with short linkers between the VH and VL domains to pair between chains of the V domains rather than within a chain, thereby achieving a bivalent fragment (i.e., a fragment with two antigen-binding sites). Bispecific diabodies are heterodimers of two "crossover" sFv fragments, in which the two antibody VH and VL domains are present on different polypeptide chains.

[0086] The term "humanized antibody" is used as a subset of "chimeric antibody".

[0087] "Humanized" forms of non-human (e.g., rhea or camelid) antibodies are chimeric antibodies containing minimal sequence derived from non-human immunoglobulins. In some embodiments, humanized antibodies are human immunoglobulins (acceptor antibody) in which residues from a CDR of an acceptor (defined below) are replaced by residues from a CDR of a non-human (donor antibody) (e.g., mouse, rat, rabbit, camel, rhea, alpaca, or non-human primate) having the desired specificity, affinity, and / or compatibility.

[0088] In some examples, human immunoglobulin framework region ("FR") residues are replaced by corresponding non-human residues. In addition, humanized antibodies may include residues not found in the recipient antibody or the donor antibody. These modifications may be implemented to further improve antibody performance (e.g., binding affinity).

[0089] The term "hypervariable region", "HVR" or "HV" as used in this application refers to the region of the antibody variable domain that has hypervariability in sequence and / or forms a structurally defined loop. Typically, a single domain antibody includes three HVRs (or CDRs): HVR1 (or CDR1), HVR2 (or CDR2), and HVR3 (or CDR3). HVR3 (or CDR3) represents the highest diversity of the three HVRs and is known to play an inherent role in conferring subtle specificity to antibodies. For example, see Hamers-Casterman et al., Nature 363:446-448 (1993); Sheriff et al., Nature Struct. Biol. 3:733-736 (1996).

[0090] As defined in the Kabat system, the term "complementarity determining region" or "CDR" is used to refer to the hypervariable region. See Kabat, Elvin A., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md. (1991). The Kabat complementarity determining region (CDR) is based on sequence variability and is the most commonly used.

[0091] As defined herein, the term "framework region" or "FR" residues are those variable domain residues other than the HVR residues.

[0092] The term "specificity" refers to the selective recognition of a particular epitope of an antigen by an antigen binding protein (eg, sdAb).

[0093] Natural antibodies are, for example, monospecific. As used herein, the term "multispecific" means that an antigen-binding protein has multiple epitope specificities (i.e., can specifically bind to 2, 3 or more different epitopes in one biomolecule, or can specifically bind to epitopes in 2, 3 or more different biomolecules). As used herein, "bispecific" means that an antigen-binding protein has 2 different antigen binding and specificities.

[0094] As used herein, the term "monospecific" refers to an antigen binding protein having one or more binding sites each that binds the same epitope of the same antigen.

[0095] The term "valency" indicates the definite number of binding sites present in an antigen binding protein. For example, the terms "divalent", "trivalent", "tetravalent", "pentavalent", and "hexavalent" indicate that there are 2 binding sites, 3 binding sites, 4 binding sites, 5 binding sites, and 6 binding sites in an antigen binding protein, respectively.

[0096] "Antibody effector function" refers to the biological activity of an antibody caused by its Fc region (a native sequence Fc region or an amino acid sequence variant Fc region), which varies depending on the isotype of the antibody. Examples of antibody effector functions include: Clq binding and complement-dependent cytotoxicity; Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; downregulation of cell surface receptors (e.g., B cell receptor); and B cell activation. "Complement-dependent cytotoxicity" or "CDC" refers to the lysis of target cells in the presence of complement. Activation of the classical complement pathway is initiated by binding of the first component of the complement system (Clq) to an antibody (of the appropriate subclass) bound to its cognate 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 specific cytotoxic cells (e.g., natural killer (NK) cells, neutrophils, and macrophages) enables these cytotoxic effector cells to specifically bind to antigen-bearing target cells, followed by killing of the target cells by cytotoxicity.

[0097] The term "Fc region" or "crystallizable fragment region" herein is used to define the C-terminal region of the immunoglobulin heavy chain including native sequence Fc regions and variant Fc regions. The native sequence Fc regions suitable for use in the antibodies of this specification include human IgG1, IgG2 (IgG2A, IgG2B), IgG3 and IgG4.

[0098] "Binding affinity" generally refers to the total strength of non-covalent interactions between a single binding site of a molecule (e.g., an antibody) and its binding partner (e.g., an antigen). As used herein, "binding affinity" refers to the intrinsic binding affinity that reflects the 1:1 interaction between binding pairs of members, unless otherwise specified. Binding affinity can be represented by Kd, Koff, Kon or Ka. As used herein, the term equilibrium dissociation constant "KD" or "Kd" refers to the dissociation constant of a specific antibody-antigen interaction, representing the antigen concentration required to account for one-half of the overall antibody binding domain present in the antibody molecule solution in equilibrium, and is represented in units of M. The determination of KD is based on the premise that all binding preparations are in solution. The dissociation constant (KD or Kd) is used as an index representing the affinity of an antibody to an antigen. For example, simple analysis can be achieved by Scatchard mapping (which utilizes antibodies labeled by a variety of marker preparations) method, and can also be achieved by conventional drugs and detection kits indicated by the instructions for use and experimental operation methods on the kit. The KD value that can be derived by utilizing these methods is expressed in the unit of M (Mols).

[0099] For peptide, polypeptide or antibody sequences, "percent (%) amino acid sequence identity" and "homology" are defined as the percentage of amino acid residues in a candidate sequence that are identical to the amino acid residues in a specific peptide or polypeptide sequence, after aligning the sequences and introducing gaps, if necessary, to achieve the highest percent sequence identity, without considering conservative substitutions as part of the sequence identity. Alignment for the purpose of determining percent amino acid sequence identity can be achieved in a variety of ways within the art, for example, using publicly available computer software such as BLAST, BLAST-2, ALIGN or MEGALIGN™ (DNATAR) software. Appropriate parameters for testing thereof can be determined by one skilled in the art, including any algorithm required to achieve maximum alignment over the full length of the sequences being compared.

[0100] The present invention relates to a bispecific antibody that bispecifically binds to PD-L1 and CD47, comprising a first humanized single domain antibody (first hsdAb) or an antigen-binding fragment thereof (anti-PD-L1 hsdAb) that specifically binds to PD-L1 (hereinafter referred to as "humanized anti-PD-L1×CD47 bsAb"), and a second humanized single domain antibody (second hsdAb) or an antigen-binding fragment thereof (anti-CD47-L1 hsdAb) that specifically binds to CD47, for example, humanized anti-PD-L1×CD47 bsAb, specifically, to a bispecific humanized single domain antibody fused with anti-PD-L1 hsdAb and anti-CD47 hsdAb, wherein the humanized anti-PD-L1×CD47 heavy chain-only antibody (HCAb) (for example, a crystal fragment (Fc fragment) of human immunoglobulin G (IgG)) is fused to the anti-PD-L1 hsdAb and / or the anti-CD47 Humanized anti-PD-L1×CD47 bsAb-Fc fusion protein of hsdAb), and preparation method and use thereof.

[0101] Thus, the present invention provides a bispecific antibody that bispecifically binds to PD-L1 and CD47, comprising a humanized anti-PD-L1×CD47 bsAb.

[0102] In the present invention, the bispecific antibody that bispecifically binds to PD-L1 and CD47 comprising a humanized anti-PD-L1×CD47 bsAb can be a humanized anti-PD-L1×CD47 bsAb, to which is fused a first humanized single domain antibody (first hsdAb) or an antigen-binding fragment thereof that specifically binds to PD-L1 through a first antigen-binding portion (hereinafter referred to as "anti-PD-L1 hsdAb"), and a second humanized single domain antibody (second hsdAb) or an antigen-binding fragment thereof that specifically binds to CD47 through a second antigen-binding portion (hereinafter referred to as "anti-CD47 hsdAb").

[0103] In the present invention, the anti-PD-L1 hsdAb comprises a CDR1 consisting of the amino acid sequence shown in SEQ ID NO: 2; a CDR2 consisting of the amino acid sequence shown in SEQ ID NO: 3; and a CDR3 consisting of the amino acid sequence shown in SEQ ID NO: 4.

[0104] Furthermore, the anti-CD47 hsdAb comprises a CDR1 consisting of the amino acid sequence shown in SEQ ID NO:9; a CDR2 consisting of the amino acid sequence shown in SEQ ID NO:10; and a CDR3 consisting of the amino acid sequence shown in SEQ ID NO:11.

[0105] The CDR sequences are shown in Tables 6 and 12.

[0106] In the present invention, the humanized anti-PD-L1×CD47 bsAb includes a human or humanized heavy chain domain framework region. The human or humanized heavy chain domain framework includes an amino acid sequence of a heavy chain variable domain (VH) framework derived from a human immunoglobulin framework. The heavy chain variable domain (VH) framework derived from a human immunoglobulin framework may include the same amino acid sequence, or contain an amino acid sequence change. In some 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.

[0107] Specifically, it may include a heavy chain variable domain framework sequence having at least 95%, 96%, 97%, 98%, 99% or 100% sequence identity with SEQ ID NO: 16. Or it may include a heavy chain variable domain framework sequence having at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% sequence identity with SEQ ID NO: 17.

[0108] The heavy chain variable domain framework (FR) sequences having 100% sequence identity with the sequence of SEQ ID NO: 16 or SEQ ID NO: 17 are shown in Tables 1 to 4.

[0109] [Table 1]

[0110] serial number FR1 SEQ ID NO: 1 EVQLVESGGGLVQPGGSLRLSCAAS 19 2 EVQLVESGGGLVQPGGSLNLSCAVS 23

[0111] [Table 2]

[0112] serial number FR2 SEQ ID NO: 1 MSWVRQAPGKGLEWVSD 20 2 VGWARQVPGKGLERVSR 24

[0113] [Table 3]

[0114]

[0115]

[0116] [Table 4]

[0117] serial number FR4 SEQ ID NO: 1 RGQGTQVTVSS 22 2 WGQGTQVTVSS 26

[0118] Specifically, the anti-PD-L1 hsdAb may include the following FR1, FR2, FR3 and FR4: FR1 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 19 and 23; FR2 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 20 and 24;

[0119] FR3 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 21 and 25; and

[0120] FR4 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 22 and 26.

[0121] More specifically, the anti-PD-L1 hsdAb may include the following FR1, FR2, FR3 and FR4:

[0122] (1) FR1 consisting of the amino acid sequence shown in SEQ ID NO: 19;

[0123] FR2 consisting of the amino acid sequence shown in SEQ ID NO: 20;

[0124] FR3 consisting of the amino acid sequence shown in SEQ ID NO: 21; and

[0125] FR4 consisting of the amino acid sequence shown in SEQ ID NO: 22; or

[0126] (2) FR1 consisting of the amino acid sequence shown in SEQ ID NO: 23;

[0127] FR2 consisting of the amino acid sequence shown in SEQ ID NO: 24;

[0128] FR3 consisting of the amino acid sequence shown in SEQ ID NO: 25; and

[0129] FR4 consists of the amino acid sequence shown in SEQ ID NO:26.

[0130] Further specifically, the anti-PD-L1 hsdAb may include the following FR1, FR2, FR3 and FR4:

[0131] FR1 consisting of the amino acid sequence shown in SEQ ID NO: 19;

[0132] FR2 consisting of the amino acid sequence shown in SEQ ID NO: 20;

[0133] FR3 consisting of the amino acid sequence shown in SEQ ID NO: 21; and

[0134] FR4 consists of the amino acid sequence shown in SEQ ID NO:22.

[0135] Furthermore, the anti-CD47 hsdAb may include the following FR1, FR2, FR3 and FR4:

[0136] FR1 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 19 and 23;

[0137] FR2 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 20 and 24;

[0138] FR3 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 21 and 25; and

[0139] FR4 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 22 and 26.

[0140] Further specifically, the anti-CD47 hsdAb may include the following FR1, FR2, FR3 and FR4:

[0141] (1) FR1 consisting of the amino acid sequence shown in SEQ ID NO: 19;

[0142] FR2 consisting of the amino acid sequence shown in SEQ ID NO: 20;

[0143] FR3 consisting of the amino acid sequence shown in SEQ ID NO: 21; and

[0144] FR4 consisting of the amino acid sequence shown in SEQ ID NO: 22; or

[0145] (2) FR1 consisting of the amino acid sequence shown in SEQ ID NO: 23;

[0146] FR2 consisting of the amino acid sequence shown in SEQ ID NO: 24;

[0147] FR3 consisting of the amino acid sequence shown in SEQ ID NO: 25; and

[0148] FR4 consists of the amino acid sequence shown in SEQ ID NO:26.

[0149] Further specifically, the anti-CD47 hsdAb may include the following FR1, FR2, FR3 and FR4:

[0150] FR1 consisting of the amino acid sequence shown in SEQ ID NO: 23;

[0151] FR2 consisting of the amino acid sequence shown in SEQ ID NO: 24;

[0152] FR3 consisting of the amino acid sequence shown in SEQ ID NO: 25; and

[0153] FR4 consists of the amino acid sequence shown in SEQ ID NO:26.

[0154] 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.

[0155] In addition, the KD of the humanized anti-PD-L1×CD47 bsAb binding to PD-L1 and CD47, respectively, can be 10-6M to 10-12M, 10-6M to 10-11M, 10-6M to 10-10M, 10-6M to 10-9M, 10-7M to 10-9M, or 10-8M to 10-9M.

[0156] In addition, the EC50 of the humanized anti-PD-L1×CD47 bsAb can be less than 500 nM in flow cytometric fluorescence sorting (FACS) analysis, specifically, 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 50 0 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 nM to 50 nM, or 1 nM to 10 nM.

[0157] In the present invention, the valence of the humanized anti-PD-L1×CD47 bsAb for each epitope of PD-L1 and CD47 can be any appropriate number. Specifically, the humanized anti-PD-L1×CD47 bsAb can have a bivalent, trivalent, tetravalent, pentavalent, hexavalent or higher valence for each of PD-L1 and CD47. For example, refer to P. Chames and D. Baty, Chapter 6. Bispecific Single Domain Antibodies, Springer-Verlag Berlin Heidelberg, 2011.

[0158] In addition, in the humanized anti-PD-L1×CD47 bsAb, the anti-PD-L1 hsdAb and the anti-CD47 hsdAb can be directly fused by a peptide bond or indirectly fused by a peptide linker. The length, degree of flexibility and / or other characteristics of the peptide linker may partially affect the characteristics of affinity, specificity or binding ability for one or more specific antigens or epitopes without limitation. For example, a longer peptide linker can be selected while ensuring that two adjacent domains do not affect each other in the three-dimensional structure. In some embodiments, the peptide linker includes flexible residues (e.g., glycine and serine) to allow free movement between adjacent domains. For example, a glycine-serine doublet can be an appropriate peptide linker. In addition, the peptide linker can have any suitable length, for example, at least about any 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 length. In some embodiments, the peptide linker can be a peptide linker consisting of any one of the amino acid sequences of SEQ ID NO: 27 to SEQ ID NO: 29.

[0159] Furthermore, the peptide linker may have a naturally occurring sequence or a non-naturally occurring sequence.

[0160] In some embodiments, the humanized anti-PD-L1×CD47bsAb may be bivalent for PD-L1 and monovalent for CD47, and the bivalent anti-PD-L1hsdAb may be directly fused to the monovalent anti-PD-L1hsdAb consisting of the amino acid sequence shown in SEQ ID NO: 16, or indirectly fused through a peptide linker. In addition, the humanized anti-PD-L1×CD47bsAb may consist of the amino acid sequence shown in SEQ ID NO: 15.

[0161] The single domain antibody (sdAb) according to the present invention can be derived from any species including, but not limited to, mouse, rat, human, camel, llama, lamprey, fish, shark, goat, rabbit and bovine. In addition, sdAb molecules naturally produced from species other than camelids can be included.

[0162] In addition, sdAbs are heavy chain antibodies lacking light chains and are derived from known naturally occurring single domain antigen binding molecules. Such single domain molecules are disclosed, for example, in WO 94 / 04678 Hamers-Casterman, et al., (1993) Nature 363: 446-448. The variable domains derived from naturally missing heavy chain molecules are denoted as VHH in this application, thereby distinguishing them from the existing VH of 4-chain immunoglobulins. Such VHH molecules can be derived from antibodies produced by species of the Camelidae family, such as camels, llamas, alpacas, dromedaries, alpacas and guanacos. Other species outside the Camelidae family can also produce heavy chain molecules with naturally missing light chains, and such VHH are also included in the scope of this application.

[0163] The humanized single domain antibody (hsdAb) according to the present invention is a humanized antibody, and the non-human antibody reduces the immunogenicity of the human after humanization, but maintains the specificity and affinity of the parent non-human antibody. Generally, the humanized antibody includes HVR, such as CDR (or a part thereof) derived from non-human antibody, and FR (or a part thereof) derived from at least one variable domain of human antibody sequence. In some embodiments, the humanized single domain antibody includes CDR (e.g., camelid species, such as rhea) derived from non-human single domain antibodies and heavy chain FR derived from human antibody sequences. In addition, in some embodiments, in the humanized antibody, some FR residues are replaced with corresponding residues from non-human antibodies (e.g., source antibodies of HVR residues) for example to restore or improve antibody specificity or affinity.

[0164] In the present invention, the bispecific antibody bispecifically binding to PD-L1 and CD47 including humanized anti-PD-L1×CD47 bsAb may be a humanized anti-PD-L1×CD47 HCAb or an antigen-binding fragment thereof.

[0165] Specifically, the humanized anti-PD-L1×CD47 HCAb may be a humanized anti-PD-L1×CD47 bsAb in the present specification fused with one or more CH2 and / or CH3 domains, for example, fused with an Fc fragment. In addition, the anti-PD-L1 hsdAb and / or anti-CD47 hsdAb in the present specification may be fused with one or more CH2 and / or CH3 domains, for example, fused with an Fc fragment.

[0166] The CH2 and / or CH3 domains are derived from immunoglobulins. The immunoglobulins may be IgA, IgD, IgE, IgG or IgM, and specifically, may be IgG. In some embodiments, humanized anti-PD-L1×CD47

[0167] HCAb may include an Fc fragment of IgG, for example, IgG1, IgG2, IgG3, or IgG4, and the Fc fragment may be a human Fc, for example, human IgG1 (hIgG1) Fc, hIgG2 Fc, hIgG3 Fc, or hIgG4 Fc.

[0168] The humanized anti-PD-L1×CD47 HCAb may be monomeric or multimeric. In addition, when it is multimeric, for example, it may be bispecific and multivalent (e.g., bivalent, trivalent, tetravalent or higher) comprising 2 or more copies of the anti-PD-L1 hsdAb and anti-CD47 hsdAb of the present application.

[0169] In the present invention, the humanized anti-PD-L1×CD47 bsAb, anti-PD-L1hsdAb or anti-CD47hsdAb can be fused through the CH2 and / or CH3 domains, specifically, the Fc fragment can be fused through a peptide linker. The length, degree of flexibility and / or other characteristics of the peptide linker may partially affect the characteristics of affinity, specificity or binding ability for one or more specific antigens or epitopes without limitation. For example, a longer peptide linker can be selected while ensuring that two adjacent domains do not affect each other in the three-dimensional structure. In some embodiments, the peptide linker includes flexible residues (e.g., glycine and serine) to allow free movement between adjacent domains. For example, a glycine-serine doublet can be an appropriate peptide linker. In addition, the peptide linker can be of any suitable length, for example, at least about any 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 in length.

[0170] In addition, the peptide linker can have a naturally occurring sequence or a non-naturally occurring sequence. For example, a sequence derived from the hinge region of an antibody having only a heavy chain can be used as a linker. For example, reference is made to patent WO1996 / 34103. In some embodiments, the peptide linker can be a hIgG1 hinge, a hIgG2 hinge, a hIgG3 hinge, a hIgG4 hinge, or a variant thereof. Alternatively, it can be a peptide linker consisting of the amino acid sequence of SEQ ID NO: 30.

[0171] In the present invention, the humanized anti-PD-L1×CD47 HCAb may include the amino acid sequence shown in SEQ ID NO: 18, or a variant thereof having at least 80% (e.g., at least any 80%, 88%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99%) sequence homology with the amino acid sequence.

[0172] In the present invention, the bispecific antibody comprising humanized anti-PD-L1×CD47bsAb that bispecifically binds to PD-L1 and CD47 can be a bispecific Fab-like antibody fragment (bsFab) in which the C-terminus of anti-PD-L1hsdAb and / or anti-CD47hsdAb is fused to the N-terminus of CH1 and Cκ domains.

[0173] The CH1 and / or 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.

[0174] According to the present invention, bsFab and its preparation technology reference P.Chames and D.Baty, Chapter 6. Bispecific Single Domain Antibodies, Springer-Verlag Berlin Heidelberg, 2011.

[0175] In the present invention, the bispecific antibody comprising a humanized anti-PD-L1×CD47 bsAb that bispecifically binds to PD-L1 and CD47 includes amino acid sequence variants. The amino acid sequence variants of the antibody can be prepared by introducing appropriate modifications or peptide synthesis into the nucleic acid sequence encoding the antibody. The above-mentioned deformation includes, for example, the deletion and / or insertion and / or substitution of residues in the amino acid sequence of the antibody. Any combination of deletion, insertion and substitution can be made into the realization of the final product, but the final product retains the desired characteristics, such as antigen binding. In some embodiments, substitutions, insertions or deletions may occur in one or more hypervariable regions (HVRs), as long as the above changes do not substantially reduce the antibody's ability to bind to antigens. For example, conservative changes that do not substantially reduce binding affinity can be implemented in HVRs. Such changes can be outside of HVR "hot spots" or CDRs.

[0176] In addition, the amino acid substitution may be at least one (e.g., any 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10) amino acid substitution. In addition, the at least one amino acid substitution may be a conservative substitution, which may be a substitution for a non-genetically encoded amino acid or a synthetic amino acid. In some embodiments, the amino acid substitution may include at least one (e.g., any 1, 2, 3 or 4) amino acid substitution in the CDR region, CDR1, CDR2 and / or CDR3. In some embodiments, the amino acid substitution may include at least one (e.g., any 1, 2, 3, 4, 5 or 6) amino acid substitution in the FR region, FR1, FR2, FR3 and / or FR4.

[0177] In addition, the amino acid sequence insertion includes amino- and / or carboxyl-terminal fusions of the length range of the polypeptide containing from 1 residue to more than 100 residues, and also includes insertions within the sequence of single or multiple amino acid residues. Examples of terminal insertions include antibodies with N-terminal methionyl residues. Other insertion variants of antibody molecules can include fusions of the N- or C-terminals of polypeptides or (e.g., ADEPT) enzymes that promote the serum half-life of antibodies.

[0178] In addition, one or more amino acid modifications can be introduced into the Fc region of a bispecific antibody (e.g., humanized anti-PD-L1×CD47 HCAb) that bispecifically binds to PD-L1 and CD47, including humanized anti-PD-L1×CD47 bsAb provided herein, thereby generating an Fc region variant. The Fc region variant may include a human Fc region sequence (e.g., human IgG1, IgG2, IgG3, or IgG4 Fc) comprising an amino acid modification (e.g., substitution) at one or more amino acid positions.

[0179] In the present invention, the bispecific antibody comprising a humanized anti-PD-L1×CD47 bsAb that bispecifically binds to PD-L1 and CD47 can be linked, fused, conjugated (e.g., covalently or non-covalently) to a diagnostic moiety or biocompatibility modifier, or otherwise associated therewith. For example, a peptide or polypeptide (e.g., a biotoxin, a biomarker, a purification tag, etc.), a protein, a polymer, a nucleic acid molecule, a small molecule, a mimetic, a synthetic drug, an inorganic molecule, an organic molecule, or a radioactive isotope can be conjugated or associated therewith.

[0180] In addition, the bispecific antibodies that bispecifically bind to PD-L1 and CD47, including humanized anti-PD-L1×CD47 bsAbs, can be conjugated or associated with diagnostic agents or detectable agonists, markers, or reporters that can be biological molecules (e.g., peptides or nucleotides), small molecules, fluorophores, or radioisotopes. Labeled regulators can be effectively used for PD-L1 and / or CD47-related diseases, such as monitoring the occurrence or progression of cancer, or in determining the efficacy of specific therapies including the antibodies disclosed in the present application (i.e., medical diagnosis (Theragnosis)), or as part of a clinical trial process to determine the subsequent treatment course. In addition, such markers or reporters can be used to purify the antibody efficacy disclosed in the present application.

[0181] In addition, the bispecific antibody comprising a humanized anti-PD-L1×CD47 bsAb that bispecifically binds to PD-L1 and CD47 can be conjugated to an immunomodulator, a cytokine, a cytotoxic drug, a chemotherapeutic agent, a diagnostic agent, an antiviral agent, an antimicrobial agent, or a drug. To this end, the present invention provides an antibody conjugate comprising a bispecific antibody comprising a humanized anti-PD-L1×CD47 bsAb that bispecifically binds to PD-L1 and CD47 according to the present invention, conjugated to an immunomodulator, a cytokine, a cytotoxic drug, a chemotherapeutic agent, a diagnostic agent, an antiviral agent, an antimicrobial agent, or a drug.

[0182] In addition, the present invention provides a nucleic acid molecule encoding a bispecific antibody comprising a humanized anti-PD-L1×CD47 bsAb that bispecifically binds to PD-L1 and CD47 disclosed in the present application, an expression vector comprising the nucleic acid molecule, and a host cell transformed by the expression vector.

[0183] In addition, the present invention provides a method for producing a bispecific antibody, comprising: (a) culturing the host cell under conditions that allow the bispecific antibody to be expressed; and (b) recovering the expressed bispecific antibody.

[0184] In the present invention, DNA encoding bispecific antibodies that bispecifically bind to PD-L1 and CD47, including humanized anti-PD-L1×CD47 bsAb disclosed in the present application, can be easily isolated and sequenced by conventional steps (e.g., using oligonucleotide probes that can specifically bind to genes encoding antibody heavy chains and light chains). Isolated and subcloned hybridoma cells (or phage or yeast-derived colonies) can be used as a preferred source of supply for the DNA. In particular, the isolated DNA (variable) can be used in the cloning of constant region sequences and variable region sequences to prepare antibodies.

[0185] An exemplary method is accompanied by extracting RNA from the selected cells, converting to cDNA and amplification caused by PCR using antibody-specific primers. Suitable primers are widely known in the art to which the present invention belongs, and can be obtained and utilized from the many commercial supply sources cited in the present application. In order to express the recombinant human antibody or non-human antibody separated by the screening of the combinatorial library, the DNA encoding the antibody is cloned into a recombinant expression vector, which is then imported into a host cell including mammalian cells, insect cells, plant cells, yeast and bacteria. In some embodiments, a regulator is introduced into a monkey COS cell, a NS0 cell, a Chinese hamster ovary (CHO) cell or a myeloma cell that does not generate a construct for other purposes to express.

[0186] In the present invention, the nucleic acid molecule coexists with a promoter controlling nucleic acid expression in a vector under appropriate circumstances. The vector is used in its most general sense, for example, including any intermediate vector for nucleic acid, which allows nucleic acid to be introduced into prokaryotic cells and / or eukaryotic cells and can be fused with the genome under appropriate circumstances. Such vectors are preferably replicated and / or expressed in cells. The vector may include a plasmid, a phagemid, a phage or a viral genome. The plasmid is usually associated with an extrachromosomal genetic material construct that can be replicated independently of chromosomal DNA, usually a circular DNA duplex.

[0187] Expression vectors comprising antibody coding sequences and appropriate transcriptional and translational control signals can be constructed using methods known in the art to which the present invention pertains, including, for example, in vitro recombinant DNA techniques, synthetic techniques, and in vivo gene recombination.

[0188] In the present invention, the host cell or recombinant host cell refers to a cell into which an expression vector is introduced. Recombinant host cells and host cells refer not only to specific target cells, but also to the offspring of these cells. The offspring may undergo specific deformation due to mutation or environmental influences, so such offspring may be substantially different from the parent cell, but it is still included in the scope of the term host cell in this application. Such cells may include vectors as described above.

[0189] In addition, the molecular biology techniques and current protein expression methodologies known in the relevant field can be used to produce substantial amounts of the antibodies disclosed herein. More specifically, the nucleic acid molecules encoding the antibodies can be integrated into known commercially available protein production systems including various types of host cells, thereby providing preclinical, clinical or commercial pharmaceutical products sought after. In some embodiments, the nucleic acid molecules encoding the antibodies are engineered into a vector or expression vector that provides effective integration into a selected host cell and subsequent high expression levels of the antibodies.

[0190] Preferably, nucleic acid molecules encoding antibodies disclosed herein and vectors comprising the nucleic acid molecules can be used for transfection of mammalian, plant, bacterial or yeast host cells, however, prokaryotic systems can also be used. Transfection can be achieved by any known method for introducing polynucleotides into host cells. Methods for introducing heterologous polynucleotides into mammalian cells are widely known in the art to which the present invention belongs, including dextran-mediated transfection, 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 by viral vectors. Methods for transforming mammalian cells are widely known in the art to which the present invention belongs. In addition, methods for transforming plant cells are widely known in the art to which the present invention belongs, including, for example, Agrobacterium-mediated transformation, biotransformation, direct injection, electroporation and viral transformation. Methods for transforming viruses and yeast cells are widely known in the art to which the present invention belongs.

[0191] Many commercially available host-expression vector systems can be used to express the antibodies disclosed in the present application. Such host-expression systems not only show vectors that express the coding sequence of interest and can be purified in the subsequent, but also show cells that can express the molecules of the present invention in the system when transformed or transfected with appropriate nucleotide coding sequences. Such systems include, but are not limited to, microorganisms transformed with recombinant phage DNA, plasmid DNA or cosmid DNA expression vectors containing regulatory subcoding sequences, such as bacteria (e.g., Escherichia coli (E. coli), Bacillus subtilis (B. subtilis), Streptomyces (streptomyces)); yeast (e.g., Saccharomyces, Pichia) transfected with recombinant yeast expression vectors containing regulatory subcoding sequences; insect cell systems infected with recombinant viral expression vectors (e.g., baculovirus) containing regulatory subcoding sequences; insect cell systems infected with recombinant viral expression vectors (e.g., baculovirus) containing regulatory subcoding sequences; Plant cell systems (e.g., Nicotiana, Arabidopsis, Lemna, corn, wheat, potato, etc.) infected with cauliflower mosaic virus (CaMV) or transfected with a recombinant plasmid expression vector (e.g., Ti plasmid) containing a regulator coding sequence; or mammalian cell systems (e.g., COS, CHO, BHK, 293, 3T3 cells) containing a recombinant expression construct containing a promoter derived from the genome of mammalian cells (e.g., metallothionein promoter) or derived from mammalian viruses (e.g., adenovirus late promoter; vaccinia virus 7.5K promoter).

[0192] If the antibodies disclosed in the present application are produced by recombinant expression or any of the other techniques disclosed in the present application, they can be purified by any method known in the art for purifying immunoglobulins, or by any other standard technique for purifying proteins.

[0193] In addition, the present invention provides a pharmaceutical composition for preventing or treating cancer, comprising a bispecific antibody comprising a humanized anti-PD-L1×CD47 bsAb that bispecifically binds to PD-L1 and CD47 disclosed in the present application, or an antibody conjugate comprising the bispecific antibody as an active ingredient.

[0194] In addition, the present invention provides a method for preventing or treating cancer, comprising: administering to an individual a pharmaceutically effective amount of a pharmaceutical composition, wherein the pharmaceutical composition comprises a bispecific antibody that bispecifically binds to PD-L1 and CD47 including a humanized anti-PD-L1×CD47 bsAb disclosed in the present specification, or an antibody conjugate comprising the bispecific antibody.

[0195] In addition, the present invention provides use of an antibody conjugate in preventing or treating cancer, wherein the antibody conjugate comprises a bispecific antibody bispecifically binding to PD-L1 and CD47 including a humanized anti-PD-L1×CD47 bsAb disclosed in the present specification, or an antibody conjugate comprising the bispecific antibody.

[0196] In addition, the present invention provides use of an antibody conjugate in preparing a composition for preventing or treating cancer, wherein the antibody conjugate comprises a bispecific antibody bispecifically binding to PD-L1 and CD47 including a humanized anti-PD-L1×CD47 bsAb disclosed in the present specification, or an antibody conjugate comprising the bispecific antibody.

[0197] In the present invention, the cancer can be a cancer that requires blocking the activity of immune checkpoint proteins to activate T cells, for example, selected from the group consisting of melanoma, lung cancer, liver cancer, glioblastoma, ovarian cancer, colorectal cancer, head and neck cancer, bladder cancer, renal cell carcinoma, gastric cancer, breast cancer, metastatic cancer, prostate cancer, pancreatic cancer, non-Hodgkin's lymphoma, Hodgkin's lymphoma, multiple myeloma, leukemia, lymphoma, myelodysplastic syndrome, acute lymphocytic leukemia, acute myeloid leukemia, chronic lymphocytic leukemia, chronic myeloid leukemia, solitary myeloma and aplastic anemia, but not limited thereto.

[0198] In the present invention, the bispecific antibody comprising humanized anti-PD-L1×CD47bsAb that bispecifically binds to PD-L1 and CD47 or the antibody conjugate comprising the bispecific antibody disclosed in the specification is as described above, and the specific description refers to the above content. Only the inherent structure of the pharmaceutical composition and use is described below.

[0199] The pharmaceutical composition according to the present invention may include one or more (e.g., 2 or 3) bispecific antibodies that bispecifically bind to PD-L1 and CD47 including humanized anti-PD-L1×CD47 bsAb disclosed in the present specification, or an antibody conjugate comprising the bispecific antibody.

[0200] The pharmaceutical composition according to the present invention can prevent or treat cancer by administering to an individual, specifically, to a cancer patient.

[0201] In addition, the pharmaceutical composition according to the present invention can be prepared in a desired manner according to the form of the antibody of the application, the desired mode of delivery, and many other variables, using the technology recognized in the technical field of the present invention to be prepared in a desired manner. In addition, it can be configured to contain a suitable carrier widely known and pharmaceutically acceptable in the field of the present invention, the carrier includes excipients and adjuvants as relatively inactive substances, and the excipients and adjuvants assist the processing of the active compound with the pharmaceutically optimized preparation, so as to apply and transmit. For example, a variety of pharmaceutically acceptable carriers including carriers, adjuvants and diluents can be easily obtained from many commercial supply sources. In addition, pharmaceutically acceptable auxiliary material classifications can also be obtained, for example, pH adjusting agents, buffers, surface tension regulators, stabilizers, wetting agents, etc. Specific non-limiting exemplary carriers include physiological saline, buffered physiological saline, dextrose, water, glycerol, ethanol and combinations thereof.

[0202] In addition, the pharmaceutical composition according to the present invention can be formulated for enteral, parenteral or topical administration. In fact, all three types of preparations can be used simultaneously to achieve systemic administration of active ingredients. It is known in the art to which the present invention belongs to that excipients and preparations are used to deliver parenteral and non-parenteral drugs. Preparations suitable for parenteral administration are active compounds in water-soluble forms, for example, aqueous solutions including water-soluble salts. In addition, a suspension of a suitable active compound can be placed in an oily injection suspension. Suitable lipophilic solvents or carriers include fatty oils, such as sesame oil or synthetic fatty acid esters (e.g., ethyl oleate or triglycerides). Aqueous injection suspensions can contain substances that increase the viscosity of the suspension, for example, including sodium carboxymethylcellulose, sorbitol and / or dextran. Arbitrarily, the suspension can also contain a stabilizer. In addition, liposomes can be utilized to encapsulate the agonist in order to deliver it to cells.

[0203] Suitable preparations for enteral administration include hard or soft gelatin capsules, pills, tablets including coated tablets, elixirs, suspensions, syrups or inhalants and controlled release forms thereof.

[0204] Generally, the antibodies disclosed in the present application can be administered to a subject in need thereof by various routes, for example, including but not limited to oral, intravenous, intraarterial, subcutaneous, parenteral, intranasal, intramuscular, intracardiac, intraventricular, intratracheal, buccal, rectal, intraperitoneal, intradermal, topical, percutaneous, and intraspinal, or ectopic transplantation or inhalation. Suitable preparations and routes for administration can be selected according to the intended use and therapy.

[0205] The pharmaceutical composition according to the present invention is administered in a pharmaceutically effective amount for treating or preventing cancer. The pharmaceutically effective amount refers to the amount of the antibody or pharmaceutical composition comprising the antibody that is expected by a doctor or other clinician to induce a biological or medical response in a subject. In addition, in order to achieve the amount of therapy with preventive and / or therapeutic effects, multiple doses of the antibody or pharmaceutical composition comprising the antibody can be administered at a specific frequency.

[0206] The pharmaceutically effective amount is determined according to the body weight of a typical subject to be treated, its physical state, a wide range of states to be treated, and the age of the subject to be treated. Typically, the antibodies disclosed herein can be administered in an amount ranging from 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, but not limited thereto. In addition, the antibody can be administered in a dosage 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 not limited thereto.

[0207] In addition, the pharmaceutical composition according to the present invention can be administered in a dose of about 100 mg to about 10000 mg, about 200 mg to about 9000 mg, about 300 mg to about 8000 mg, about 400 mg to 7000 mg, 500 mg to 5000 mg, but is not limited thereto.

[0208] Pharmaceutical compositions according to the present invention are usually applied to patients multiple times. Exemplary therapy includes administration once in two weeks, once in a month, or once in 3 months to 6 months. For example, as a cycle, antibodies (e.g., intravenous preparations) are administered to patients once every 4 weeks, for example, every 28 days. The frequency of administration can be adjusted according to the pharmacokinetic graph of the antibody in the patient. For example, the half-life of the antibody may require a frequency of administration of 2 weeks. In some methods, 2 or more antibodies with different binding specificities can be administered simultaneously, and the dosage of each antibody administered in this case is included in the scope provided.

[0209] Dosage and frequency vary according to the half-life of the antibody in the patient. Generally, human antibodies show the longest half-life, followed by humanized antibodies, chimeric antibodies, and non-human antibodies. Dosage and frequency of administration can vary according to the preventive or therapeutic properties of the treatment.

[0210] The duration of the treatment is determined according to the disease to be treated, the age and condition of the patient, the course and type of the disease suffered by the patient, the patient's response to the treatment, etc. The clinician closely observes the effect of the therapy and can make adjustments as needed. When an agonist is used in combination, two or more therapeutic agents can be administered simultaneously or sequentially in any order, that is, the antibody disclosed in the present application can be administered before the administration of the second therapeutic agent, at the same time as the second therapeutic agent, or after the administration of the second therapeutic agent.

[0211] Hereinafter, the present invention will be described in detail through examples and experimental examples.

[0212] However, the following examples and experimental examples are only for illustrating the present invention, and the contents of the present invention are not limited to the following examples and experimental examples.

[0213] <Example 1> Immunization and blood collection

[0214] Human PD-L1 protein or human CD47 protein was used as an immune antigen and mixed with an immune adjuvant (GERBU) and injected into an alpaca intramuscularly three times for immunization. Three immunizations were performed, and 10 ml of blood was collected from each alpaca on the 14th day after the last immunization, and the immune response was analyzed by ELISA. In order to determine whether antibodies were produced, the immune antigen was inoculated into a 96-well microplate at a concentration of 1 μg / ml using a coating buffer and coated overnight at 4°C. After washing the 96-well microplate three times with PBST, in order to inhibit nonspecific binding, it was blocked by treating with 5% skim milk at room temperature for 2 hours. After that, it was washed three times with PBST, and the serum samples obtained before immunization (day 0), 14 days (day 14), 28 days (day 28), and 42 days (day42) after immunization were treated according to different gradient dilution concentrations. Then, the 96-well microplate was washed five times with PBST, and after reacting with goat anti-Llama IgG horseradish peroxidase (goat anti-Llama IgG HRP) antibody at room temperature for 1 hour, whether the antibody was bound to the antibody was confirmed by TMB reaction.

[0215] <Example 2> Library preparation and evaluation

[0216] The library was constructed by gene amplification, which encodes a single domain antibody that binds to the immune antigen confirmed by Example 1 above. To construct the library, peripheral blood mononuclear cells (PBMC) were separated from the blood using Ficoll. The gene fragment encoding the single domain antibody was amplified from the total RNA extracted from the isolated peripheral blood mononuclear cells (PBMC) using specific primers and cloned into the pComb3x vector. The size of the prepared immune library was 5.4×10 8 .

[0217] <Example 3> Library Amplification

[0218] The immune library prepared in Example 2 was transformed into the XL1-blue strain. The transformed XL1-blue strain was added to 10 ml of 2x YT medium containing 2% glucose and 100 μg / ml ampicillin, and cultured in a shaking mixer at 37°C. The culture was continued until the OD was 0. 600 The absorbance at 0.5 was 0.5, and M13K07 phage (Invitrogen) was added until it reached 1×10 11pfu / ml. Then, after static culture at 37°C for 30 minutes, additional culture was performed at 200 rpm in a 37°C shaking mixer for 30 minutes. The culture solution was centrifuged at 4000 rpm for 15 minutes at room temperature, and the supernatant was removed. Then, 10 ml of 2x YT medium containing 100 μg / ml ampicillin and 50 μg / ml kanamycin was added to resuspend the pellet of the culture solution, and cultured overnight at 250 rpm in a 30°C shaking incubator. Then, centrifuged at 4°C and 4000 rpm for 30 minutes. The supernatant was precipitated by PEG precipitation method and centrifuged at 4°C and 12000 rpm for 30 minutes. The pellet was resuspended in PBS, centrifuged at 4°C and 13000 rpm for 5 minutes, and the supernatant was transferred to a new test tube and stored at 4°C before use.

[0219] <Example 4> Bio-panning

[0220] In order to screen out single domain antibodies that are specific to the immune antigen, the immune antigen is inoculated into a 96-well microplate at a concentration of 5 μg / ml using a coating buffer and coated overnight at 4°C. The library for screening single domain antibodies (the library in Example 3 above) is inoculated into a 96-well microplate and reacted at room temperature for 30 minutes. Then, the library is transferred to a new well, and the reaction at room temperature for 30 minutes is repeated 4 times. This operation is performed to reduce the library that nonspecifically binds to the wells of the microplate. The library is transferred to a 1.7 ml test tube and stored at 4°C before use. The microplate coated with the immune antigen is washed 5 times with PBST, 5% skim milk is added and blocked at room temperature for 2 hours. Then, it is washed 5 times with PBST, and the library with a reduced nonspecific binding rate is mixed with the binding solution (2.5% skim milk, 0.5% Tween 20) at 5×10 12 virions / well were inoculated and reacted at room temperature for 30 minutes. Then, after washing 10 times with a washing solution (PBS, 0.5% Tween 20), it was washed an additional 3 times with PBST. For single-domain antibodies that specifically bind to the immune antigen, the immune antigen was added in an amount of 5 μg per well and reacted at room temperature and 500 rpm for 30 minutes for selective elution. The eluted phage was infected with XL-1blue cells in the logarithmic growth phase and spread on 2x YT agar medium. For the second screening, the operation was repeated under the same conditions as above. Single phage clones generated in the agar medium were amplified separately and screened by FACS.

[0221] <Example 5> Phage screening

[0222] In order to induce temporary overexpression of immune antigens in Expi-CHO cells, the gene encoding the immune antigen was inserted into the pCMV6-GFP vector, thereby constructing the pCMV6-immune antigen-GFP plasmid. Expi-CHO cells were washed with DPBS and centrifuged at room temperature and 1200 rpm for 3 minutes. The supernatant was removed, and the cells were resuspended with 2% skim milk and blocked at 4°C for 30 minutes. After centrifuging the cells at room temperature and 1200 rpm for 3 minutes, the supernatant was removed, washed twice with DPBS, and inoculated into a 96-well microplate until 3×10 5 cells / 100μl / well. A single clone phage was added to each well, reacted at 4°C for 1 hour, and then washed twice with DPBS. The cells were inoculated with an antibody that specifically binds to the phage (M13 major coat protein Alexa Flour 647 (Santacruz)) and reacted at 4°C for 30 minutes in a light-proof environment. The cells were washed twice with DPBS, resuspended with new DPBS, and analyzed by FACS using an Accuri C6 (BD) device. The screened clones were selected using the FACS system and subjected to base sequence analysis.

[0223] <Example 6> Expression and purification of single-domain antibodies fused with human IgG Fc domain

[0224] <6-1> Expression and purification of monovalent single-domain antibodies fused with human IgG Fc domain

[0225] The clones selected in the above-mentioned <Example 5> were cloned into TGEX-Fc (IgG1) or TGEX-Fc (IgG4) expression vectors. In order to express single-domain antibodies fused with human IgG Fc domains, Expi-CHO cells with a viability of 95% to 99% were counted and 7×10 6 Each cell was added to 25 ml of culture medium (Expi-CHO expression medium (Gibco)) and cultured overnight in a shaking incubator at 37°C and 125 rpm with 8% CO2. TM CHOReagent (Gibco, 100033021) and 920 μl of OptiPRO TM Add 20 μg of plasmid DNA encoding a single domain antibody fused to the human IgG Fc domain and 1 ml of OptiPRO TMThe mixture of medium was reacted at room temperature for 5 minutes and then added to the cultured cells. The cells were cultured at 125 rpm in a shaking incubator with 8% CO2 for 20 hours. Then, 150 μl of ExpiFectamine was added to enhance the expression of single-domain antibodies fused with human IgG Fc domains. TM CHO enhancer (Gibco) and 6 ml of ExpiCHO Feed (Gibco), and then cultured at 125 rpm in a shaking incubator maintained at 5% CO2 for 5 days. After centrifuging the cultured cells at 4000 rpm for 30 minutes at 4°C, the supernatant was filtered with a 0.2 μm syringe filter. Then, the supernatant was loaded into a HiTrap protein G HP column (GE Healthcare) and 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). After adding 1M Tris-HCl (pH 9.0) to the eluted sample for neutralization, it was stored at 4°C before use.

[0226] <6-2> Expression and purification of bivalent single-domain antibodies fused with human IgG4 Fc domain

[0227] The clones selected in Example 5 above were connected using two G2S linkers (GGSGGS) to prepare a bivalent single-domain antibody. The nucleotides encoding the bivalent single-domain antibody were obtained by gene synthesis (MARCROGEN, South Korea). In order to express and purify a single-domain antibody fused with a human IgG4 Fc domain, the synthesized gene was cloned into a TGEX-Fc (IgG4) expression vector. Then, expression and purification were performed in the same manner as in Example 6-1 above.

[0228] <Example 7> Expression and purification of bispecific single domain antibodies fused with human IgG4 Fc domain

[0229] From the clones selected in Example 5 above, one single-domain antibody each specifically binding to the PD-L1 and CD47 antigens was screened. The nucleotide sequence encoding each single-domain antibody was connected using a peptide linker, and two single-domain antibodies specifically binding to PD-L1 and single-domain antibodies specifically binding to CD47 (anti-PD-L1 sdAb×anti-PD-L1 sdAb×anti-CD47 sdAb) were connected in sequence. The nucleotide sequence encoding this trivalent bispecific single-domain antibody was obtained by gene synthesis (MARCROGEN, South Korea). In order to express and purify the bispecific single-domain antibody fused with the human IgG4 Fc domain, the synthesized gene was cloned into the TGEX-Fc (IgG4) expression vector, and then expressed and purified by the same method as in Example 6-1 above.

[0230] <Example 8> Expression and purification of bispecific humanized single domain antibodies fused with human IgG4 Fc domain

[0231] In order to humanize the trivalent bispecific single domain antibody prepared in Example 7, the FR (frame region) sequence of human germline VH segment DP-47 (J Mol Biol. 1992 Oct 5; 227 (3): 776-98) was compared for homology. In the amino acid sequences of one of the clones selected in Example 5 that specifically bind to PD-L1 and CD47 antigens, respectively (PDL1 Nb #01 and CD47 Nb #01), glutamine (Q) as the first amino acid in the FR1 (frame region) sequence was replaced with glutamic acid (E), thereby preparing 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) with improved homology to human antibody DP-47. The amino acid substitution mutation was induced using primers that specifically bind to the Q5 Site directed mutagenesis kit (NEB), and then expressed and purified by the same method as in Example 6 above.

[0232] Example 9. Evaluation of the ability of single domain antibodies to bind to immune antigens using FACS

[0233] The binding ability of the anti-PD-L1 HCAb (PDL1 Nb#01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-CD47 HCAb (CD47 Nb-IgG4), anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4), and humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in the above Examples 6, 7, and 8 to the immune antigen was confirmed by FACS.

[0234] Specifically, after washing the CHO-K1_PD-L1 cell line (CHO-K1 cells overexpressing the PD-L1 antigen) or Expi-CHO_CD47 cell line (Expi-CHO cells overexpressing the CD47 antigen) with DPBS, centrifuge at 1200 rpm for 3 minutes at room temperature. Remove the supernatant, resuspend the cells with 2% skim milk and block at 4°C for 30 minutes. After centrifugation at 1200 rpm for 3 minutes at room temperature to remove the supernatant, wash the cells twice with DPBS. Then, at 3×105 cells / 100 After the cells were seeded in each well, anti-PD-L1 HCAb (PDL1 Nb#01-IgG1), anti-PD-L1 bivalent HCAb (PPC 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) were treated at different concentrations, and the isotype control antibody was used as a negative control group. After reacting at 4°C for 1 hour, the cells were washed twice with DPBS. Then, an antibody that specifically binds to the human Fc domain (Anti-human IgG Fc APC antibody (Biolegend)) was treated and reacted for 30 minutes at 4°C in a light-protected environment. The cells were washed twice with DPBS, resuspended with 100 ml of DPBS, and FACS analysis was performed using an Accuri C6 (BD) device.

[0235] <Example 10> Evaluation of the inhibitory ability of single domain antibodies on immune antigen interactions using FACS

[0236] The inhibitory ability of the anti-PD-L1 HCAb (PDL1 Nb#01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-CD47 HCAb (CD47 Nb-IgG4), anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4), and humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in the above Examples 6, 7, and 8 on the PD-1 / PD-L1 or CD47 / SIRPalpha interactions was evaluated.

[0237] Specifically, to evaluate the PD-1 / PD-L1 interaction of anti-PD-L1 HCAb (PDL1 Nb#01-IgG1), anti-PD-L1 bivalent HCAb (PPNb-IgG4), anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4), or humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), CHO-K1_PD-L cell line (CHO-L1 cells that constantly express PD-L1 antigen) were seeded into 96-well microplates at 2×105 cells per well and then treated with 10 μg / ml human PD-1-His protein. Then, anti-PD-L1 HCAb (PDL1Nb#01-IgG1), anti-PD-L1 bivalent HCAb (PPNb-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 the negative control group was treated with isotype control antibody. The cells were reacted at 4°C for 1 hour, washed 3 times with DPBS, and then the antibody specifically binding to the His antigen (Goat anti-His APC) was added and reacted at 4°C in the dark for 30 minutes. After washing the cells three times with DPBS, they were resuspended in 100 μl of DPBS and the amount of remaining PD-1-His protein in CHO-K1_PD-L1 cells (CHO-K1 cells that constantly express the PD-L1 antigen) was detected using an Accuri C6 (BD) apparatus to evaluate the inhibitory ability of anti-PD-L1 HCAb (PDL1 Nb#01-IgG1), anti-PD-L1 divalent HCAb (PP Nb-IgG4), anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) or humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) on the PD-1 / PD-L1 interaction.

[0238] In addition, to evaluate the effect of anti-CD47 sdAb (CD47Nb-IgG4), anti-PD-L1×CD47 trivalent HCAb (PPCNb-IgG4) or humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) on CD47 / SIRPalpha interaction, Expi-CHO_CD47 cells (Expi-CHO cells that constantly express CD47 antigen) were seeded into 96-well microplates at 2×105cells per well and treated with 10μg / ml human SIRPalpha-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 with different concentrations, and isotype control antibody was treated for negative control group. Then, the amount of remaining SIRPalpha-His protein in Expi-CHO_CD47 cells (Expi-CHO cells that constantly express the CD47 antigen) was confirmed using an Accuri C6 (BD) instrument in the same manner as described above, thereby evaluating the inhibitory ability of anti-CD47 sdAb (CD47Nb-IgG4), anti-PD-L1×CD47 HCAb (PPC Nb-IgG4) or humanized anti-PD-L1×CD47 HCAb (hPPC Nb-IgG4) on the CD47 / SIRPalpha interaction.

[0239] <Example 11> Evaluation of the affinity of single domain antibodies to immune antigens

[0240] The affinity (Kd) 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) or humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Examples 6, 7 and 8 to the immune antigen protein was determined using the Octet RED 96e (ForteBio) instrument.

[0241] Specifically, the biosensor chip coated with anti-human Fc (Fortebio) was saturated at 1.5 nM in a 96-well microplate (Greiner) inoculated with 5 μg / ml of anti-PD-L1 HCAb (PDL1 Nb#01-IgG1), anti-PD-L1 bivalent HCAb (PP Nb-IgG4), anti-CD47 HCAb (CD47 Nb-IgG4), anti-CD47 HCAb (CD47Nb-IgG4), anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) or humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4). PD-L1 and CD47 antigens were gradiently diluted 2 times to 10 nM to 400 nM using 1X kinetic buffer (ForteBio), stirred and reacted at 30°C at 1000 rpm. The binding and dissociation reactions of the samples were analyzed for 200 seconds and 400 seconds, respectively. The result data were analyzed using the 1:1 interaction model (Global fitting) method.

[0242] In addition, the affinity (Kd) of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 to the heterologous immune antigen protein was determined using the Octet RED 96e (ForteBio) instrument.

[0243] Specifically, the biosensor chip coated with Anti-human Fc (Fortebio) was saturated at 1.5nM in a 96-well microplate (Greiner) inoculated with 20μg / ml humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4). The PD-L1 or CD47 antigens of humans, mice, and monkeys were gradiently diluted 2-fold to 12.5nM to 200nM using 1Xkinetic buffer (ForteBio), stirred and reacted at 1000rpm at 30°C. The binding and dissociation reactions of the samples were analyzed for 200 seconds and 400 seconds, respectively. The result data were analyzed using the 1:1 interaction model (Global fitting) method.

[0244] <Example 12> Evaluation of in vitro effectiveness of humanized single domain antibodies

[0245] By confirming T cell activity, CD4 +The degree of activation of T cells and the degree of activation of macrophage phagocytosis were evaluated to evaluate the in vitro effectiveness of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 above.

[0246] <12-1> Evaluation of T cell activity of humanized single domain antibodies

[0247] In CHO-K1 cells induced to overexpress PD-L1 protein and Jurkat cells expressing PD-1, the concentration of 1L-2 expressed from Jurkat cells due to the inhibition of the cell-cell interaction by the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in the above Example 8 was measured, thereby evaluating the T cell activation efficacy.

[0248] Specifically, CHO-K1 cells overexpressing PD-L1 protein were cultured at 2×10 4 cells were inoculated and cultured in a culture machine maintained at 5% CO2 for 16 hours. The culture medium was then removed and humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was treated at different concentrations and reacted for 1 hour. 5 After the Jurkat cells were prepared, PHA was treated until it reached 0.5 mg / ml, and then added to the CHO-K1 cells of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) and reacted for 48 hours. Then, 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).

[0249] <12-2> Evaluation of CD4+ T cell activation of humanized single domain antibodies in mixed lymphocyte reactions

[0250] The CD4+ / -1β-adrenergic reaction (CD4+ / -2β-adrenergic reaction) of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 was evaluated by mixed lymphocyte reaction (MLR) analysis. + Activation of T cells.

[0251] Specifically, monocyte-induced dendritic cells (MoDC) of donor A were inoculated into a 96-well plate at 5x104 cells / well, and CD4+T cells of donor B were inoculated into the wells inoculated with MoDC at 4x105 cells / well to mix them, and then humanized anti-PD-L1×CD47 trivalent HCAb (hPPC-Nb-IgG4) was treated at different concentrations and cultured for 5 days (120 hours) at 37°C with 5% CO2. Then, the supernatant for measuring IL-2 production was obtained on the third day after culture, and the supernatant for measuring IFN-γ production was obtained from the same well, and the IL-2 and IFN-γ concentrations were measured by ELISA.

[0252] MoDCs used for mixed lymphocyte reaction analysis express PD-L1, and CD4+T cells express PD-1. The activity of suppressed T cells can be activated by interfering with the interaction between PD-L1 and PD-1 through antibodies. This is equivalent to a system in which immune activity is suppressed by the binding of PD-L1 expressed in cancer cells and PD-1 expressed in T cells, thereby suppressing the survival of cancer cells from the immune system. For this purpose, 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 effectiveness of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4).

[0253] <12-3> Evaluation of humanized single domain antibody activation of macrophage phagocytosis

[0254] Phagocytosis is activated by specific binding of antigens to the IgG4 Fc domain. Antigen-specific binding is affected by the physical environment (intercellular distance), chemical environment (chemokine), and biological environment (cytokine) between target cells and effector cells. PD-L1 and CD47 antigen-specific humanized antibodies specifically bind to hPDL1 and hCD47 expressed in target cells, and the Fc region binding portion binds to the Fc receptor on the surface of effector cells, thereby promoting the binding of target cells to effector cells, thereby inducing phagocytosis of macrophages.

[0255] Therefore, the activation of macrophage phagocytosis was evaluated by CD47 antigen-specific binding of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 above.

[0256] Specifically, 5×106cell THP-1 cells were inoculated into a 100mm culture dish as effector cells and cultured in a culture machine maintained at 5% CO2 for 24 hours. After reacting with 40nM PMA for 24 hours, the culture medium was removed and stained with 1uM crimson dye. After staining, the culture medium was replaced and allowed to stand for 48 hours. THP-1 cells removed by trypsin were used as target cells and mixed with prey cells (Raji cells) stained with 3uM CFSE at a ratio of 8:1. After being treated with humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) at different concentrations, they were co-cultured for 4 hours. Then, the degree of phagocytosis was evaluated by FACS.

[0257] <12-4> Evaluation of humanized single domain antibodies for activation of phagocytosis of PBMC-derived macrophages

[0258] Monocytes isolated from human PBMCs were differentiated into macrophages and co-cultured with target cells (Raji_PDL1), thereby evaluating the activation of human monocyte-derived macrophage phagocytosis by specific binding of hPD-L1 and hCD47 antigens by the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 above.

[0259] Specifically, PBMCs were isolated from whole blood using a ficoll gradient and CD14 + Monocytes were isolated using MACS kit and then cultured at 2x10 in 100 mm culture dishes. 6 cells / ml were treated with 50ng / ml M-CSF and differentiated for 6 days. The differentiated macrophages were used as effector cells and stained with 5uM CFSE, and the target cells (Raji_PDL1) were stained with 10uM deep red dye. Target cells were stained with 2x10 4 Using cells / well as the standard, the mixing ratio of effector cells to target cells was set at 8:1 for co-culture. In addition, the concentration of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was diluted 1 / 5 from 1000nM and evaluated at 15 points, and co-cultured for 4 hours. After co-culture, the expression of Deep Red APC and CFSE FITC was confirmed by FACS analysis, and the APC expressing cell population was confirmed based on the expression of CFSE, thereby evaluating the phagocytosis level.

[0260] <Example 13> Evaluation of the in vitro safety of humanized single domain antibodies

[0261] Human RBC binding and hemagglutination reactions were confirmed, thereby evaluating the in vitro safety of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 above.

[0262] <13-1> Evaluation of RBC binding of humanized single domain antibodies

[0263] The binding of human RBCs by the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 above due to specific binding to the CD47 antigen was evaluated.

[0264] Specifically, in order to confirm the binding ability of human RBC (Red blood cell) to humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), RBC was washed 7 times with DPBS and diluted to 12% (v / v) with DPBS. After inoculating 50ul of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) in a 96-well V-plate, 50ul of 12% (v / v) RBC was added and reacted at 4°C for 1 hour. The cells were washed with DPBS, treated with anti-human IgG4 as a secondary antibody, and reacted at 4°C for 1 hour. After washing the cells with DPBS, the level of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) bound to RBC was evaluated by FACS.

[0265] <13-2> Evaluation of hemagglutination reactions induced by humanized single domain antibodies

[0266] The hemagglutination reaction caused by the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 above due to specific binding to the CD47 antigen was evaluated.

[0267] Specifically, in order to evaluate the hemagglutination reaction of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), RBCs were washed 7 times with DPBS and diluted to 6% (v / v) with DPBS. After inoculating 50ul of 2X humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) in a 96-well U-shaped plate, 50ul of 6% (v / v) RBCs were added. After reacting at room temperature for 1 hour, the hemagglutination reaction was observed with the naked eye.

[0268] <Example 14> Evaluation of in vivo effectiveness of humanized single domain antibodies

[0269] The antitumor effect of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Example 8 above was confirmed in C57BL / 6 mice and humanized NSG mice formed by injecting tumors into tumor cell lines expressing human PD-L1 and human CD47, thereby evaluating the in vivo effectiveness.

[0270] <14-1> C57BL / 6 mouse model using B16F10_PD-L1_CD47 tumor

[0271] 8×105cell / 100ul of B16F10 cell line inducing overexpression of human PD-L1 and human CD47 were injected into 6- to 8-week-old C57BL / 6 mice. Tumor formation was induced until the size of the tumor's horizontal axis (long axis) × vertical axis (short axis) reached 3×3mm. Then, 10mpk of humanized anti-PD-L1×CD47 trivalent HCAb (hPPCNb-IgG4) was intraperitoneally administered 7 times every 2 days, and the tumor size was measured. After the last intraperitoneal administration, the tumor volume was measured every 2 days for a total of 2 weeks. The tumor volume is measured as the horizontal axis (long axis) × vertical axis (short axis) × vertical axis (short axis) / 2.

[0272] <14-2> Humanized NSG mouse model using Raji_PD-L1 tumor

[0273] Humanized NSG mice were made by intraperitoneal injection of 2×107cell / 100ul human PBMC into female NSG mice aged 6 to 8 weeks. Human Raji (PD-L1) lymphoma cell line induced to overexpress human PD-L1 was injected with 2×106cell / 100ul. Tumor formation was induced until the size of the tumor horizontal axis (long axis) × vertical axis (short axis) reached 3×3mm. Then, 10mpk of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was intraperitoneally administered every 3 weeks, and the tumor volume was measured. The experiment was terminated when the tumor volume reached 2000mm3 after the last intraperitoneal administration.

[0274] <14-3> Humanized NSG mouse model using MDA-MB-231 tumor

[0275] Humanized NSG mice were prepared by intravenously injecting 2×107cell / 100ul human PBMC into female NOD scid NSG mice aged 6 to 8 weeks. After confirming the expression of hCD45 and mCD45, 2×106cell / 100ul of human breast cancer cell line MDA-MB-231 was subcutaneously injected under the right rib. Tumor formation was induced until the size of the tumor horizontal axis (long axis) × vertical axis (short axis) reached 3×3mm. Then, 1mpk, 5mpk, and 10mpk of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) were intraperitoneally administered every 3 weeks, and the tumor volume was measured. In addition, when the tumor volume of the humanized NSG mouse reached 100mm3, Kaplan-Meier survival analysis (Kaplan-meier estimatorsurvival) was performed based on the individual death criterion to predict the survival rate.

[0276] <Experimental Example 1> Preparation of anti-PD-L1 HCAb (PDL1Nb#01-IgG1) and evaluation of in vitro properties

[0277] <1-1> Preparation of anti-PD-L1 HCAb (PDL1 Nb#01-IgG1)

[0278] Human PD-L1 antigen was used as an immune antigen, and single domain antibody clones specific to PD-L1 antigen were screened by the same method as in Example 5, and base sequence analysis was performed. The amino acid sequence of the screened anti-PD-L1sdAb (PDL1 Nb#01) is shown in Tables 5 and 6 below.

[0279] In addition, the screened anti-PD-L1 sdAb (PDL1 Nb#01) was used to express and purify a monovalent single-domain antibody specific for PD-L1 including a human IgG1 Fc domain 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 the anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) including a human IgG1 Fc domain is shown in Table 7 below.

[0280] In addition, the anti-PD-L1 sdAb (PDL1 Nb#01) clone screened was used to express and purify a bivalent single-domain antibody specific for PD-L1 including a human IgG4 Fc domain by the same method 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 the anti-PD-L1 bivalent sdAb (PP Nb) after removing the human IgG4 Fc domain from the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) is shown in Table 8 below, and the amino acid sequence of the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) including a human IgG4 Fc domain is shown in Table 9 below.

[0281] [Table 5]

[0282]

[0283] [Table 6]

[0284]

[0285] [Table 7]

[0286]

[0287] [Table 8]

[0288]

[0289]

[0290] [Table 9]

[0291]

[0292] <1-2> Evaluation of the antigen binding ability of anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) and anti-PD-L1 bivalent HCAb (PP Nb-IgG4) and their inhibitory ability on PD-1 / PD-L1 interaction

[0293] In the same manner as in Examples 9 and 10, the antigen binding abilities of the anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) and the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) purified in Experimental Example 1 were evaluated by FACS ( Figure 1A and Figure 2A ) and the ability to inhibit PD-1 / PD-L1 interaction ( Figure 1B and 2B ).

[0294] Results, such as Figure 1A and 1B As shown, it was confirmed that the anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) had an antigen binding ability of 23.31nM (EC50) in CHO-K1_PD-L1 cells (CHO-K1 cells that constantly express the PD-L1 antigen), and the inhibitory ability against the PD-1 / PD-L1 interaction was 4.60nM (IC50).

[0295] In addition, if Figure 2A and 2B As shown, it was confirmed that the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) had an antigen binding ability of 1.93nM (EC50) in CHO-K1_PD-L1 cells (CHO-K1 cells that constantly express the PD-L1 antigen), and an inhibitory ability on the PD-1 / PD-L1 interaction of 2.86nM (IC50).

[0296] <1-3> Evaluation of the affinity of anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) and anti-PD-L1 bivalent HCAb (PP Nb-IgG4) to the immune antigen

[0297] The affinity of the anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) and anti-PD-L1 bivalent HCAb (PP Nb-IgG4) purified in the above Experimental Example 1 to PD-L1 was evaluated by the same method as in the above Example 10.

[0298] As a result, as shown in Table 10, it was confirmed that the anti-PD-L1 HCAb (PDL1 Nb#01-IgG1) had an antigen affinity of 7.08 nM for the PD-L1 antigen, and the anti-PD-L1 bivalent HCAb (PP Nb-IgG4) had an antigen affinity of 4.58 nM for the PD-L1 antigen.

[0299] [Table 10]

[0300]

[0301] <Experimental Example 2> Preparation of anti-CD47 HCAb (CD47 Nb-IgG4) and evaluation of in vitro characteristics

[0302] <2-1> Preparation of anti-CD47 HCAb (CD47 Nb-IgG4)

[0303] Human CD47 antigen was used as an immune antigen, and single domain antibody clones specific to CD47 antigen were screened by the same method as in Example 5, and base sequence analysis was performed. The amino acid sequence of the screened anti-CD47 sdAb (CD47 Nb#01) is shown in Tables 11 and 12 below.

[0304] In addition, the screened anti-CD47 sdAb (CD47 Nb_#01) was used to express and purify a single domain antibody specific for CD47 including a human IgG4 Fc domain in the same manner as in Example 6-1, and the purified single domain antibody was named anti-CD47 HCAb (CD47Nb-IgG4). The amino acid sequence of the anti-CD47 HCAb (CD47 Nb-IgG4) including a human IgG4 Fc domain is shown in Table 13 below.

[0305] [Table 11]

[0306]

[0307] [Table 12]

[0308]

[0309] [Table 13]

[0310]

[0311]

[0312] <2-2> Evaluation of the antigen binding ability of anti-CD47 HCAb (CD47 Nb-IgG4) and its inhibitory ability on CD47 / SIRPalpha interaction

[0313] In the same manner as in Examples 9 and 10, the antigen binding ability of the anti-CD47 HCAb (CD47 Nb-IgG4) purified in Experiment 2-1 was evaluated by FACS ( Figure 3A ) and the ability to inhibit CD47 / SIRPalpha interaction ( Figure 3B ).

[0314] Results, such as Figure 3A and 3BAs shown, it was confirmed that anti-CD47 HCAb (CD47 Nb-IgG4) had an antigen binding ability of 3.78 nM (EC50) in Expi-CHO_CD47 cells (Expi-CHO cells that constantly express CD47 antigen) and an inhibitory ability on CD47 / SIRPalpha interaction of 7.28 nM (IC50).

[0315] <2-3> Evaluation of the affinity of anti-CD47 HCAb (CD47 Nb-IgG4) to the immune antigen

[0316] The affinity of the anti-CD47 HCAb (CD47 Nb-IgG4) purified in the above Experimental Example 2-1 to the CD47 antigen was evaluated by the same method as in Example 11.

[0317] As a result, as shown in Table 14, it was confirmed that the anti-CD47 HCAb (CD47Nb-IgG4) had an antigen affinity of 2.78 nM for the CD47 antigen.

[0318] [Table 14]

[0319]

[0320] <Experimental Example 3> Preparation of anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) and evaluation of in vitro characteristics

[0321] <3-1> Preparation of anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4)

[0322] A single domain antibody including a human IgG4 Fc domain and having bispecificity for PD-L1 and CD47 antigens as an immune antigen was prepared using the clone screened in the above Experimental Example 1-1. Using the nucleotide sequence encoding the anti-PD-L1 sdAb (PDL1 Nb#01) screened in the above Experimental Example 1-1 and the nucleotide sequence encoding the anti-CD47 sdAb (CD47 Nb#01) screened in the above Experimental Example 2-1, the gene was synthesized by the same method as in the above Example 7 and cloned into the TGEX-Fc (IgG4) expression vector for expression and purification, and the purified single domain antibody was named anti-PD-L1×CD47 HCAb (PPC Nb-IgG4).

[0323] The amino acid sequence of the PD-L1×CD47 trivalent sdAb after removing the human IgG4 Fc domain from the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) is shown in Table 15 below, and the amino acid sequence of the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) including the human IgG4 Fc domain is shown in Table 16 below.

[0324] [Table 15]

[0325]

[0326]

[0327] [Table 16]

[0328]

[0329]

[0330] <3-2> Evaluation of the immune antigen binding ability of anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4)

[0331] The ability of the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) purified in Experimental Example 3-1 to bind to the PD-L1 antigen and the CD47 antigen was confirmed by FACS using the same method as in Example 9.

[0332] Results, such as Figure 4A and 4B As shown, it was confirmed that the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) had an antigen binding ability of 13.33nM (EC50) in CHO-K1_PD-L1 cells (CHO-K1 cells that constantly express the PD-L1 antigen) and 18.80nM (EC50) in Expi-CHO_CD47 cells (Expi-CHO cells that constantly express the CD47 antigen).

[0333] <3-3> Evaluation of the inhibitory ability of anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) on the interaction between PD-1 / PD-L1 and CD47 / SIRPalpha

[0334] The anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) purified in Experimental Example 3-1 was confirmed to be responsive to PD-1 / PD-L1 by FACS in the same manner as in Example 10. Inhibitory capacity of CD47 / SIRPalpha interaction.

[0335] Results, such as Figure 5A and 5B As shown, it was confirmed that the inhibitory ability of anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) on PD-1 / PD-L1 interaction was 9.15 nM (IC50), and the inhibitory ability on CD47 / SIRPalpha interaction was 22.44 nM (IC50).

[0336] <3-4> Evaluation of the affinity of anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) to the immune antigen

[0337] The affinity of the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) purified in the above Experimental Example 3-1 to the PD-L1 antigen and the CD47 antigen was evaluated by the same method as in the above Example 11.

[0338] As a result, as shown in FIG. 17 , it was confirmed that the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) had an excellent affinity for the PD-1 antigen of 6.84 nM and an affinity for the CD47 antigen of 3.62 nM.

[0339] [Table 17]

[0340]

[0341] <Experimental Example 4> Preparation of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) and evaluation of in vitro properties

[0342] <4-1> Production of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)

[0343] For the humanized antibody obtained by humanizing the anti-PD-L1×CD47 trivalent HCAb (PPC Nb-IgG4) prepared in the above-mentioned Experimental Example 3, the amino acid 2 was replaced by site-directed mutagenesis using the same method as in the above-mentioned Example 7, and was named humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4).

[0344] For humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), the homology was compared with the FR (frame region) sequence of the human germline VH segment DP-47 and is shown in Table 18. The homology of the FR1, FR2, FR3, and FR4 sequences of humanized anti-PD-L1sdAb (hPDL1 Nb#01) in humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) averaged 98.4%, and the homology of the FR1, FR2, FR3, and FR4 sequences of humanized anti-CD47sdAb (hCD47 Nb#01) in humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) averaged 89.6%.

[0345] The amino acid sequence of the humanized anti-PD-L1×CD47 trivalent sdAb (hPPC Nb) after removing the human IgG4 Fc domain in the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), the amino acid sequence of the humanized anti-PD-L1 sdAb site (hPDL1Nb#01) and the humanized anti-CD47 sdAb site (hCD47Nb#01) are shown in Table 19 below, and the amino acid sequence of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) including the human IgG4 Fc domain is shown in Table 20 below.

[0346] [Table 18]

[0347]

[0348]

[0349] [Table 19]

[0350]

[0351]

[0352] [Table 20]

[0353]

[0354]

[0355] In addition, for comparison with the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4), the humanized antibody prepared in the above Experimental Example 3 was humanized by replacing amino acids by site-directed mutagenesis (site directed mutagenesis) as described above, and was named humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4). In addition, the humanized antibody prepared in the above Experimental Example 2 was humanized by replacing amino acids by site-directed mutagenesis (site directed mutagenesis) as described above, and was named humanized anti-CD47 HCAb (hCD47 Nb-IgG4). Its amino acid sequence is shown in Table 21 below.

[0356] [Table 21]

[0357]

[0358]

[0359] <4-2> Evaluation of the immune antigen binding ability of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)

[0360] The binding ability of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in the above Experimental Example 4-1 to the PD-L1 antigen and the CD47 antigen was confirmed by FACS using the same method as in the above Example 9.

[0361] Results, such as Fig. 6A and 6B As shown, it was confirmed that the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) had an antigen binding ability of 11.48nM (EC50) in CHO-K1_PD-L1 cells (CHO-K1 cells that constantly express the PD-L1 antigen) and 22.68nM (EC50) in Expi-CHO_CD47 cells (Expi-CHO cells that constantly express the CD47 antigen).

[0362] <4-3> Evaluation of the inhibitory ability of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) on PD-1 / PD-L1 and CD47 / SIRPalpha interactions

[0363] The PD-1 / PD-L1 activity of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 was confirmed by FACS using the same method as in Example 10. Inhibitory capacity of CD47 / SIRPalpha interaction.

[0364] Results, such as Fig. 7A and 7B As shown, it was confirmed that the inhibitory ability of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) on PD-1 / PD-L1 interaction was 6.83nM (IC50), and the inhibitory ability on CD47 / SIRPalpha interaction was 33.04nM (IC50).

[0365] <4-4> Evaluation of the affinity of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) for immune antigens

[0366] The affinity of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in the above Experimental Example 4-1 to the PD-L1 antigen and the CD47 antigen was evaluated by the same method as in the above Example 11.

[0367] As a result, as shown in FIG. 21 , it was confirmed that the affinity of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) for the PD-1 antigen was 6.77 nM and the affinity for the CD47 antigen was 3.34 nM, which were excellent.

[0368] [Table 22]

[0369]

[0370] <4-5> Evaluation of the affinity of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) for heterologous immune antigens

[0371] The affinity of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in the above Experimental Example 4-1 to the heterologous PD-L1 antigen and CD47 antigen was evaluated by the same method as in the above Example 11.

[0372] As a result, as shown in Figure 23, it was confirmed that the affinity of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) for human PD-1 antigen and human CD47 antigen was 6.77nM and 3.34nM, respectively, and the affinity for monkey PD-1 antigen and monkey CD47 antigen was 8.17nM and 6.91nM, respectively. The affinity for mouse PD-1 antigen and mouse CD47 antigen could not be confirmed.

[0373] [Table 23]

[0374]

[0375]

[0376] <Experimental Example 5> Evaluation of in vitro properties of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)

[0377] <5-1> Evaluation of T cell activity of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)

[0378] Using the same method as in Example 12-1 above, the T cell activity of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was evaluated by measuring the concentration of IL-2 (which is expressed in Jurkat cells) by allowing the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 above to hinder the interaction between CHO-K1 cells (which express PD-L1) and Jurkat cells (which express PD-1).

[0379] Results, such as Figure 8 As shown, the efficacy of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was 1.47 nM (IC50).

[0380] <5-2> Evaluation of the activation of CD4+ T cells in mixed lymphocyte reactions by humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)

[0381] Mixed lymphocyte reaction (MLR) assay was performed in the same manner as described in Example 2-2; the activation of CD4+T cells by purified humanized anti-PD-L1xCD47 in Experimental Example 4-1 was evaluated by mixed lymphocyte reaction of trivalent HCAb (hPPC Nb-IgG4). As a comparative example, humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4), humanized anti-CD47 HCAb (hCD47 Nb-IgG4) prepared in the above-mentioned Experimental Example 4-1 and a mixture thereof (1:1 mixture) were used.

[0382] Results, such as Fig.9A and 9B As shown, humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) was expressed in CD4 + The efficacy in IL-2 and IFN-γ production of T cells was 0.08 nM and 0.18 nM (Ec50), respectively. In addition, it was confirmed that it showed equal or higher efficacy compared with humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4), humanized anti-CD47 HCAb (hCD47 Nb-IgG4) and their combination administration.

[0383] <5-3> Evaluation of the activation of macrophage phagocytosis by humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)

[0384] Using the same method as in Example 12-3 above, THP-1 cells macrophaged by PMA were used to evaluate the level of phagocytosis of macrophages by selective binding of 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.

[0385] Results, such as Fig. 10A As shown, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)-treated group exhibited significantly superior phagocytosis than the control group.

[0386] In addition, using the same method as in Example 12-4 above, using macrophages differentiated from monocytes isolated from human PBMC, the phagocytic level of macrophages was evaluated by selective binding of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 to prey cells (Raji cells) expressing PD-L1 and CD47 antigens. As a comparative example, the humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4), humanized anti-CD47 HCAb (hCD47 Nb-IgG4) prepared in Experimental Example 4-1 above, and a mixture thereof (1:1 mixture) were used.

[0387] Results, such as Fig. 10B As shown, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) treatment group effectively induced phagocytosis at a low concentration (0.005 pM) compared with the humanized anti-PD-L1 bivalent HCAb (hPP Nb-IgG4) single treatment group (0.522 pM), the humanized anti-CD47 HCAb (hCD47Nb-IgG4) single treatment group (0.181 pM), and the combination administration group (0.159 pM).

[0388] <Experimental Example 6> Evaluation of the in vitro safety of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)

[0389] <6-1> Evaluation of human RBC binding of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)

[0390] The human RBC binding ability of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in the above Experimental Example 4-1 was evaluated by the same method as in the above Example 13-1.

[0391] Results, such as Fig.11 As shown, unlike the CD47 monoclonal antibody used as a positive control group, no RBC binding was observed with the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) at a concentration of 3 uM.

[0392] <6-2> Evaluation of the hemagglutination reaction of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)

[0393] The hemagglutination reaction of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in the above Experimental Example 4-1 was evaluated by the same method as in the above Example 13-2.

[0394] Results, such as Fig.12 As shown, unlike the CD47 monoclonal antibody used as the positive control group, no hemagglutination reaction was observed with the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) at a concentration of 3 uM.

[0395] <Experimental Example 7> Evaluation of the in vivo effectiveness of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4)

[0396] Using the same method as in Example 14, the anti-tumor effect of the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) purified in Experimental Example 4-1 was confirmed in a C57BL / 6 mouse model in which tumors were formed by injection of tumor cells (B16F10 cells) expressing PD-L1 antigen and CD47 antigen, a humanized NSG mouse model in which tumors were formed by injection of tumor cells (Raji_PD-L1 cells) expressing PD-L1 antigen and CD47 antigen, and a humanized NSG mouse model in which tumors were formed by injection of the human breast cancer cell line MDA-MB-231.

[0397] Results, such as Fig.13 As shown, in the tumor-forming C57BL / 6 mouse model, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) exhibited an anti-tumor effect of approximately 73.61% compared to the negative control group (Isotype).

[0398] In addition, if Fig.14 As shown, in the humanized NSG mouse model of tumor formation, the humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) exhibited an anti-tumor effect of approximately 73.41% compared with the negative control group (Isotype).

[0399] In addition, if Fig.15A and Fig. 15BAs shown, in the humanized NSG mouse model with tumor formation, humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) showed an anti-tumor effect of about 81.98% compared with the negative control group (Isotype). In addition, the expected survival rate of the group administered with 10 mpk of humanized anti-PD-L1×CD47 trivalent HCAb (hPPC Nb-IgG4) showed a significant difference compared with the negative control group.

[0400] Industrial Applicability

[0401] The humanized single-domain antibody bispecifically binding to the immune checkpoint proteins PD-L1 and CD47 according to the present invention exhibits efficacy in vitro and in vivo, and thus can be effectively used as an immune checkpoint protein in immune anti-cancer therapy.

Claims

1. A bispecific antibody that bispecifically binds to PD-L1 and CD47, comprising: a first humanized single domain antibody (first hsdAb) or an antigen-binding fragment thereof that specifically binds to programmed death-ligand 1 (PD-L1); as well as A second humanized single domain antibody (second hsdAb) or an antigen-binding fragment thereof that specifically binds to cluster of differentiation 47 (CD47).

2. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 1, wherein: The first hsdAb or antigen-binding fragment thereof comprises, CDR1 consisting of the amino acid sequence shown in SEQ ID NO: 2; CDR2 consisting of the amino acid sequence shown in SEQ ID NO: 3; and CDR3 consisting of the amino acid sequence shown in SEQ ID NO:

4.

3. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 1, wherein: The second hsdAb or antigen-binding fragment thereof comprises, CDR1 consisting of the amino acid sequence shown in SEQ ID NO:9; CDR2 consisting of the amino acid sequence shown in SEQ ID NO: 10; and CDR3 consisting of the amino acid sequence shown in SEQ ID NO:

11.

4. The bispecific antibody that bispecifically binds 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 to the sequence of SEQ ID NO:

16.

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

17.

6. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 1, wherein: The first hsdAb or antigen-binding fragment thereof or the second hsdAb or 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 bispecifically binds to PD-L1 and CD47 according to claim 6, wherein: The first hsdAb or antigen-binding fragment thereof comprises the following heavy chain variable domain framework sequence: (1) FR1 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 19 and 23; (2) FR2 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 20 and 24; (3) FR3 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 21 and 25; and (4) FR4 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 22 and 26.

8. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 7, wherein: The first hsdAb or antigen-binding fragment thereof comprises the following heavy chain variable domain framework sequence: FR1 consisting of the amino acid sequence shown in SEQ ID NO: 19; FR2 consisting of the amino acid sequence shown in SEQ ID NO: 20; FR3 consisting of the amino acid sequence shown in SEQ ID NO: 21; and FR4 consists of the amino acid sequence shown in SEQ ID NO:

22.

9. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 6, wherein: The second hsdAb or antigen-binding fragment thereof comprises the following heavy chain variable domain framework sequence: (1) FR1 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 19 and 23; (2) FR2 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 20 and 24; (3) FR3 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 21 and 25; and (4) FR4 consisting of the amino acid sequence shown in any one of SEQ ID NOs: 22 and 26.

10. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 9, wherein: The second hsdAb or antigen-binding fragment thereof comprises the following heavy chain variable domain framework sequence: FR1 consisting of the amino acid sequence shown in SEQ ID NO: 23; FR2 consisting of the amino acid sequence shown in SEQ ID NO: 24; FR3 consisting of the amino acid sequence shown in SEQ ID NO: 25; and FR4 consists of the amino acid sequence shown in SEQ ID NO:

26.

11. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 1, wherein: The first hsdAb consists of the amino acid sequence shown in SEQ ID NO:

16.

12. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 7, wherein: The second hsdAb consists of the amino acid sequence shown in SEQ ID NO:

17.

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

14. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 13, wherein: The first hsdAb and the second hsdAb fused to each other via a peptide linker consist of the amino acid sequence shown in SEQ ID NO:

15.

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

16. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 15, wherein: The heavy chain-only antibody (HCAb) is monomeric or multimeric.

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

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

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

18.

20. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 15, comprising: At least one amino acid substitution.

21. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 20, wherein: At least one amino acid substitution is a conservative substitution.

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

23. The bispecific antibody that bispecifically binds to PD-L1 and CD47 according to claim 1, wherein: It is conjugated to an immunomodulator, cytokine, cytotoxic drug, chemotherapeutic agent, diagnostic agent, antiviral agent, antimicrobial agent or drug.

24. An antibody conjugate comprising: The bispecific antibody bispecifically binding to PD-L1 and CD47 according to claim 1, The bispecific antibody bispecifically binding to PD-L1 and CD47 is conjugated to an immunomodulator, a cytokine, a cytotoxic drug, a chemotherapeutic agent, a diagnostic agent, an antiviral agent, an antimicrobial agent, or a drug.

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

26. An expression vector comprising: The nucleic acid molecule of claim 25. A host cell transformed by the expression vector according to claim 26.

28. A method for producing a bispecific antibody that bispecifically binds to PD-L1 and CD47, comprising: (a) culturing the host cell according to claim 26 under conditions allowing the bispecific antibody to be expressed; as well as (b) Recovering the expressed bispecific antibody.

29. A pharmaceutical composition for preventing or treating cancer, comprising the bispecific antibody bispecifically binding to PD-L1 and CD47 according to claim 1 or the antibody conjugate according to claim 24 as an active ingredient.

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, colon cancer, head and neck cancer, bladder cancer, renal cell carcinoma, gastric cancer, breast cancer, metastatic cancer, prostate cancer, pancreatic cancer, non-Hodgkin's lymphoma, Hodgkin's lymphoma, multiple myeloma, leukemia, lymphoma, myelodysplastic syndrome, acute lymphocytic leukemia, acute myeloid leukemia, chronic lymphocytic leukemia, chronic myeloid leukemia, solitary myeloma and aplastic anemia.

31. The pharmaceutical composition according to claim 29, further comprising: A pharmaceutically acceptable carrier.

32. A method for preventing or treating cancer, comprising: The bispecific antibody bispecifically binding to PD-L1 and CD47 according to claim 1 or the antibody conjugate according to claim 24 is administered to an individual.

33. Use of the bispecific antibody bispecifically binding to PD-L1 and CD47 according to claim 1 or the antibody conjugate according to claim 24 as a pharmaceutical composition for preventing or treating cancer.

34. Use of the bispecific antibody bispecifically binding to PD-L1 and CD47 according to claim 1 or the antibody conjugate according to claim 24 for preparing a pharmaceutical composition for preventing or treating cancer.

Images