Inhibitory anti-CD93 antibody

Abstract

This disclosure relates to agents such as binding proteins that bind to vascular CD93, wherein the binding of such agents to CD93 results in one or more of the following: i) inhibition of perivascular tumor cell migration, ii) inhibition of tumor cell invasion, and iii) inhibition of tumor cell proliferation. The agents or binding proteins, nucleic acid molecules, or pharmaceutical compositions thereof may be used in medical treatments such as cancer therapy. Binding proteins may also be used in cells that are engineered to express chimeric antigen receptors having the binding proteins of this disclosure as antigen-binding domains.

Description

[Technical Field]

[0001] This disclosure relates to a binding agent, which is a binding protein or comprises a binding protein that binds to human vascular CD93. In some embodiments, the binding protein constitutes a CD93-binding antibody or a fragment thereof. In some embodiments, the drug is a nucleic acid molecule encoding the binding agent, which may be in the form of mRNA inserted into a gene vector in a transduced host cell, thereby expressing the binding protein. In some embodiments, the drug, binding protein, antibody, or a pharmaceutical composition thereof is used for medical treatment, such as cancer therapy. In other embodiments, the binding protein is used to manipulate cells to express a chimeric antigen receptor having the binding protein of this disclosure as an antigen-binding domain. [Background technology]

[0002] Cancer is one of the most common fatal diseases and still accounts for a significant number of deaths each year despite recent advances in diagnosis and treatment. Gliomas are tumors that originate from glial cells in the brain or spinal cord, and they make up about 30% of all brain tumors and central nervous system tumors, and 80% of all malignant brain tumors. Gliomas are highly invasive tumors with a high rate of postoperative tumor recurrence, and complete surgical resection of the tumor is not possible due to their invasiveness. Therefore, at present, there is no curative treatment for patients with gliomas. Thus, novel and enhanced treatments for cancers such as gliomas are needed. [Overview of the Initiative]

[0003] The object of this disclosure is to provide novel and enhanced binding agents that can be used in medical treatment. This object is obtained by agents or binding proteins that specifically bind to vascular CD93.

[0004] In some embodiments, there are drugs comprising a binding portion that specifically binds to vascular differentiation antigen group 93, CD93, wherein the binding of the drug to CD93 results in one or more of the following: i) inhibition of perivascular tumor cell migration, ii) inhibition of tumor cell invasion, and iii) inhibition of tumor cell proliferation.

[0005] In some embodiments, the drug or its binding portion is a binding protein comprising an antibody binding domain, wherein the binding domain comprises a heavy chain variable domain (VH) and a light chain variable domain (VL), each comprising three complementarity-determining regions (CDRs), and the amino acid sequence of the CDR is selected from the group comprising VHCDR1 as defined by SEQ ID NO: 1, VHCDR2 as defined by SEQ ID NO: 2, VHCDR3 as defined by SEQ ID NO: 3, VLCDR1 as defined by SEQ ID NO: 4, VLCDR2 as defined by AAS, VLCDR3 as defined by SEQ ID NO: 5, and CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more, and the binding protein has a KD < 21 nM.

[0006] In some embodiments, the CDRs are individually selected from the group including VHCDR1 selected from sequence numbers 8-10, VHCDR2 selected from sequence numbers 11-14, VHCDR3 selected from sequence numbers 15-18, VLCDR1 defined by sequence number 4, VLCDR2 and VLCDR3 defined by AAS selected from sequence numbers 19-21, and CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more.

[0007] In some embodiments, the VH sequence includes an amino acid sequence selected from the group consisting of SEQ ID NOs: 22-25, and sequences having 80% or more identity thereto, such as 85%, 90%, 95%, or more; the VL sequence includes an amino acid sequence selected from the group consisting of SEQ ID NOs: 26-29, and sequences having 80% or more identity thereto, such as 85%, 90%, 95%, or more; and the CDR sequence contains no mutations in the amino acid sequence, or the sequence variation of the CDR amino acid sequence is at most 5%, such as 4%, 3%, 2%, 1%, or less.

[0008] In some aspects, the binding protein is an antigen-binding fragment selected from the group consisting of monoclonal antibodies such as IgG1 LALA antibodies, or Fv fragments such as scFv fragments, Fab-like fragments such as Fab or F(ab’)2 fragments, disulfide bond fragments, and domain antibodies. Typically, the binding protein is of human or human origin.

[0009] In some aspects, engineered cells are provided. The cells can be engineered to express a chimeric antigen receptor (CAR), which includes an antigen-binding domain, a transmembrane domain connected to the antigen-binding domain by a hinge region, and optionally, an intracellular domain connected to one or more co-stimulatory domains, and the antigen-binding domain includes the scFv fragment of the binding protein of the present disclosure.

[0010] According to some aspects, a nucleic acid molecule encoding a drug or a binding molecule, or the present disclosure is provided. In some aspects, a vector containing the nucleic acid molecule is provided.

[0011] According to some aspects, a pharmaceutical composition is provided. The pharmaceutical composition includes a drug, a binding protein, a nucleic acid molecule, or an engineered cell as described above, and a pharmaceutically acceptable carrier or excipient.

[0012] The above-mentioned binders, binding proteins, nucleic acid molecules, manipulated cells, or pharmaceutical compositions may be used for therapeutic purposes. In some embodiments, the therapy is a cancer therapy, such as glioma therapy. In some embodiments, the cancer is glioblastoma.

[0013] Other purposes and advantages will become apparent to those skilled in the art from the following detailed description, which proceeds with reference to the following exemplary drawings and accompanying claims.

[0014] With reference to the accompanying drawings, the above and additional objectives, features, and advantages of the concept of the present invention will be better understood through the following illustrative and non-limiting detailed description of different embodiments of the concept of the present invention. [Brief explanation of the drawing]

[0015] [Figure 1] This paper illustrates the effects of common CD93 conjugates on tumor cells invading perivascular spaces. [Figure 2] This study demonstrates enhanced CD93 expression in glioblastoma patients. CD93 mRNA expression in patients with grade II, grade III, and grade IV glioblastoma. RNASeq data downloaded from the publicly available database CGGA (http: / / gliovis.bioinfo.cnio.es / ). ****p<0.0001; *p=0.026. One-way ANOVA and Tukey's multiple comparison test. [Figure 3] This study demonstrates that CD93 knockout is associated with reduced tumor growth and improved survival. A) Quantification of the GL261 region showing a significant reduction in tumor growth in CD93 knockout mice (CD93- / -) compared to the wild-type group. *p<0.05, unpaired t-test. B) Survival curves of mice injected intracranially with GL261 glioma cells showing improved survival in CD93- / - mice compared to the wild-type (WT) group. *p<0.05, Gehan-Breslow-Wilcoxon test. Langenkamp E. et al Cancer Res. 2015 Nov 1;75(21):4504-16. [Figure 4] This study demonstrates that CD93 knockout is associated with increased expression of the leukocyte adhesion molecule VCAM1 in tumor vessels and increased abundance of cytotoxic T cells in GL261 tumors, and is linked to improved survival after αPD1 blockade immunotherapy. A) Immunofluorescence staining showing increased VCAM1 expression in glioma tumor vessels of CD93-deficient mice. B) Quantitative graph showing a significant increase in VCAM1 in tumor vessels in CD93- / -GL261 tumors compared to the wild-type group. ****p<0.0001, unpaired two-sided t-test. C) Immunofluorescence staining showing increased infiltration of CD8+ cytotoxic T cells in gliomas of CD93-deficient mice. D) Quantitative graph showing a significant increase in abundance of CD8+ T cells in CD93- / -GL261 tumors compared to the wild-type group. **p<0.01, unpaired two-sided t-test. E) Survival curves of mice injected intracranially with GL261 glioma cells, showing improved survival in CD93- / - mice compared to wild-type mice after αPD1 blockade immunotherapy. αPD1 (αPD1 antibody), Iso (isotype control IgG). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, Gehan-Breslow-Wilcoxon test. [Figure 5]This shows that CD93 is required for fibronectin fibril formation. A) Immunofluorescence staining showing the effect of CD93 downregulation (siCD93) on fibronectin fibril formation in human cutaneous blood endothelial cells (HDBECs) compared to control condition (siCtrl). Actin is visualized by phalloidin staining. B) Immunofluorescence staining of mouse brain sections showing the effect of CD93 deletion (CD93- / -) on fibronectin in a GL261 glioma model. The dotted line indicates the tumor area. Nuclei are visualized by Hoechst staining. C) Quantification graph showing a significant reduction in fibronectin deposition in CD93- / - GL261 tumors compared to the wild-type group. ****p<0.0001, unpaired t-test. D) Immunofluorescence image showing the disruption of vascular-associated fibronectin in CD93 knockout GL261 tumors. Blood vessels are visualized by CD31 staining. Lugano R. et al J Clin Invest.2018 Aug 1;128(8):3280-3297. [Figure 6] This shows that CD93 is expressed in blood vessels co-opted by tumor cells. A) Immunofluorescence image of a GL261 tumor in wild-type mice, showing CD93 expression in blood vessels co-opted by tumor cells at the tumor periphery (arrowhead). The dotted line indicates the tumor boundary. Scale bar 40 μm. B) Immunofluorescence image of CD93 expression (arrowhead) in blood vessels co-opted by tumor cells in an ex vivo mouse brain slice model co-cultured with GL261 tumor cells. The dotted line indicates the tumor boundary. Scale bar 40 μm. [Figure 7]This shows glioma cells migrating along fibronectin filaments. A) Immunofluorescence image of GL261 mouse brain endothelial cells (EC) co-culture, showing GL261 cells migrating along fibronectin filaments produced by endothelial cells (arrowhead). Scale bar: 100 μm. B) Immunofluorescence image of a mouse brain slice model, showing GL261 cells migrating along CD31-positive vessels expressing fibronectin (arrowhead). Scale bar: 50 μm. C) Immunofluorescence image of a mouse brain tumor, showing GL261 cells migrating along fibronectin-expressing vessels at the tumor margin (arrowhead). Scale bar: 100 μm. [Figure 8] This study demonstrates that CD93 deficiency is associated with reduced tumor cell invasion. A) Immunofluorescence image showing inhibition of GL261 cell diffusion when co-cultured in vitro on a monolayer of CD93-deficient (CD93- / -) brain endothelial cells compared to wild-type cells. Scale bar 200 μm. B) Quantification of the region of the endothelial cell monolayer covered with GL261 cells, showing a significant reduction in GL261 diffusion when co-cultured on a monolayer of CD93- / - endothelial cells. **p<0.01, unpaired t-test. C) Immunofluorescence image showing reduced invasion of GL261 cells along blood vessels when co-cultured as spheroids on brain slices of wild-type or CD93- / - mice. Scale bar 50 μm. D) Quantification of the number of GL261 sprouts along blood vessels in wild-type and CD93- / - brain slices. E) Invasion distance of GL261 cells along blood vessels in wild-type and CD93- / - brain slices. *p<0.05, **p<0.01. Unpaired t-test. F) H&E staining of GL261 tumor margins in wild-type and CD93- / - mice showing inhibition of tumor invasion in the CD93- / - group. Scale bar 100 μm. G) Quantification of invasive tumor area in GL261 gliomas from wild-type and CD93- / - mice. **p<0.01. Unpaired t-test. [Figure 9]This study demonstrates that CD93 deficiency reduces tumor cell proliferation. A) Proliferation assay of U87 human glioma cells cultured in conditional media derived from control (Mock and siCtrl) or CD93-downregulated (siCD93_1 and siCD93_2) human endothelial cells (HDBEC), showing that the absence of soluble CD93 in the supernatant reduces U87 proliferation. **p<0.01 Two-way ANOVA with Dunnet multiple comparison test. B) Immunofluorescence images of the proliferation marker Ki67 in GL261 spheroids cultured on wild-type or CD93- / - brain slices. The images show reduced Ki67 signaling in GL261 spheroids cultured on CD93- / - brain slices compared to wild-type brain slices. Scale bar: 100 μm. C) Quantification of Ki67-positive regions normalized by GL261 spheroid regions. The graph shows a significant reduction in Ki67-positive GL261 cells in the CD93- / - group. **p<0.01, independent t-test.** [Figure 10] Surface plasmon resonance sensograms of selected αCD93 antibodies are shown. Binding to AD169-8, 85, 309, and 324 was measured by capturing the antibodies on a CM5 series S tip with an immobilized α-human Fab antibody mixture. Single injections of 50 nM analytes (hCD93, hCD93lectin, and mCD93) were performed. After the dissociation phase, the surface was regenerated with 10 mM glycine-HCl, pH 2.1. The SPR single-cycle kinetic value is the nanomolar dissociation rate constant (KD). [Figure 11] This study demonstrates that a CD93 blocking antibody specifically binds to human endothelial cells expressing CD93. Human endothelial cells (HDBEC, expressing CD93) and human glioma cell line U87 (not expressing CD93) were incubated with αCD93 antibodies (human-reactive and human / mouse cross-reactive). The graph shows the percentage of cells positive for the α-kappa secondary antibody used to detect the αCD93 antibody, as determined by FACS analysis. [Figure 12]This study demonstrates that a CD93 blocking antibody specifically binds to mouse endothelial cells expressing CD93. Mouse endothelial cells (MS1, expressing CD93) and mouse GL261 glioma cells (not expressing CD93) were incubated with αCD93 antibody (human / mouse cross-reactive). The graph shows the percentage of cells labeled with α-kappa secondary antibody, which is used to detect the CD93 antibody by FACS analysis. [Figure 13] This study demonstrates that CD93 blocking antibodies reduce fibronectin fibril formation in human endothelial cells in vitro. A) Immunofluorescence staining of fibronectin in human endothelial cells (HDBECs) after 24 hours of treatment with 10 μg / ml αCD93 antibody or control IgG. Scale bar 20 μm. B) Quantification of fibronectin-covered regions normalized to total cell number. The graph shows the reduction of fibronectin regions in cells treated with αCD93 antibody compared to control IgG. [Figure 14] This study demonstrates that a CD93 blocking antibody reduces fibronectin fibril formation in mouse endothelial cells in vitro. A) Immunofluorescence staining of fibronectin in mouse endothelial cells MS1 after 24 hours of treatment with 10 μg / ml αCD93 antibody or control IgG. Scale bar 20 μm. B) Quantification of fibronectin-covered regions normalized to total cell number. The graph shows the reduction of fibronectin regions in cells treated with αCD93 antibody compared to control IgG. [Figure 15] This study demonstrates that a CD93 blocking antibody reduces vascular-associated fibronectin in mouse brain slices. Immunofluorescence staining of fibronectin in mouse brain slices treated with αCD93 antibody or control IgG (10 μg / ml over 24 hours). The image shows a significant reduction in fibronectin-positive signals (arrowheads) within CD31-positive vessels in brain slices treated with control IgG, and in brain slices treated with αCD93 antibody. The dotted line indicates the GL261 tumor region. Scale bar 100 μm. [Figure 16]This demonstrates that CD93 blocking antibodies disrupt human endothelial cell-to-cell junctions in vitro. A) Immunofluorescence staining of actin in HDBEC monolayer after 8 hours of treatment with 10 μg / ml αCD93 antibody or control IgG. Intercellular gaps are shown in white. Scale bar 20 μm. B) Quantification of gap area normalized by total cell number. The graph shows the increased intercellular gap area after αCD93 antibody treatment compared to control IgG. [Figure 17] This demonstrates that a CD93 blocking antibody disrupts mouse endothelial cell-cell junctions in vitro. A) Immunofluorescence staining of actin showing the MS1 monolayer after 8 hours of treatment with 10 μg / ml αCD93 antibody or control IgG. Intercellular gaps are shown in white. Scale bar 20 μm. B) Quantification of gap area normalized by total cell number. The graph shows the increase in intercellular gap area after αCD93 antibody treatment compared to control IgG. [Figure 18] This study demonstrates that a CD93 blocking antibody reduces human endothelial cell migration in vitro. A wound healing assay shows significantly reduced migration in human endothelial cells (HDBECs) treated with 10 μg / ml αCD93 antibody compared to control IgG-treated cells. [Figure 19] This study demonstrates that CD93 blocking antibodies do not induce cell apoptosis in vitro. A) Analysis of apoptosis marker cleavage caspase 3 (cCasp3) on human endothelial cells (HDBEC) treated with 10 μg / ml αCD93 antibody or control IgG for 24 hours. B) Analysis of apoptosis marker cleavage caspase 3 (cCasp3) on mouse endothelial cells (MS1) treated with 10 μg / ml αCD93 antibody or control IgG for 24 hours. [Figure 20]This study demonstrates that a CD93 blocking antibody inhibits the migration of GL261 cells along blood vessels in mouse brain slices. A) Immunofluorescence staining of GL261 spheroids cultured on mouse brain slices for 48 hours in the presence of 10 μg / ml αCD93 antibody or control IgG. Arrowheads indicate tumor cells hijacking blood vessels. The image shows a significant reduction in tumor-hijacked blood vessels in brain slices treated with αCD93 antibody compared to control IgG. B) Quantification of the number of tumor-hijacked blood vessels at 100 μm of the GL261 spheroid boundary. [Figure 21] This study demonstrates that a CD93 blocking antibody inhibits the migration of U3013 cells along blood vessels in human brain slices. Immunofluorescence staining of U3013 human glioma cells cultured on human brain slices for 48 hours in the presence of 10 μg / ml αCD93 antibody or control IgG. Arrowheads in the control IgG image indicate elongated tumor cells hijacking CD31-positive blood vessels. Arrowheads in the αCD93 antibody-treated slices indicate aggregates of tumor cells exhibiting a non-migrating phenotype. [Figure 22] This graph shows the in vivo distribution of αCD93 antibody injected intravenously. GL261-carrying mice (A, B, C) and healthy mice (D) were intravenously injected with 5 mg / Kg of αCD93 antibody or control IgG 24 hours prior to organ collection. αCD93 antibody was detected using an anti-kappa AlexaFluor488 conjugated antibody. Blood vessels were visualized by CD31 staining. αCD93 antibody signaling was evaluated in the tumor core region (A), tumor boundary (B), contralateral hemisphere of the tumor (CLH) (C), and healthy mice (D). In (B), the dotted line indicates the tumor margin, and T indicates the tumor region. Arrowheads indicate blood vessels positive for anti-kappa antibody. Scale bar: 50 μm. (E) Quantified graph showing the percentage of anti-kappa signaling against CD31-positive areas in the brain, kidney, liver, and spleen of GL261-carrying and healthy mice. [Figure 23]This study demonstrates that repeated systemic administration of a CD93-blocking antibody to GL261-carrying mice reduces perivascular invasion by glioma cells. A) Timeline of in vivo tumor studies (iv, intravenous). B) Representative immunofluorescence staining of GL261 tumor boundaries in mice treated with αCD93 antibody AD169-85 or AD169-309 or control IgG. Nuclei are visualized by Hoechst staining, blood vessels by CD31, and tumor cells GL261-GFP are visualized by GFP signaling. Dotted lines indicate the anterior tumor invasion, while continuous lines indicate the tumor mass margin. Scale bar: 50 μm. C) Quantification graphs show tumor invasion regions corresponding to the area between the continuous line (tumor mass margin) and the dotted line (tumor invasion boundary). Data values ​​are in arbitrary units (AU). *p<0.05, **p<0.01, one-way ANOVA with Dunnett multiple comparison test. [Figure 24] This study demonstrates that repeated intravenous injection of CD93-blocking antibodies inhibits vascular-associated fibronectin in GL261-carrying mice. A) Representative immunofluorescence staining of GL261 tumors isolated from mice treated with αCD93 antibodies AD169-85 or AD169-309 or control IgG. The image shows nuclei visualized by Hoechst staining, fibronectin, and blood vessels visualized by CD31. Scale bar: 50 μm. B) The quantification graph shows the percentage of fibronectin-covered area relative to the total CD31 area in the field of view. *p<0.05, ***p<0.001. One-way ANOVA using Dunnett multiple comparison test. [Figure 25]This study demonstrates that treatment of GL261-possessing mice with the αCD93 blocking antibody AD169-309 increases VCAM1 expression in tumor vascular tissue, promotes intratumoral invasion of CD8 and CD3-positive T cells, and enhances survival when combined with αPD1 immunotherapy. A) Representative immunofluorescence staining of tumor vascular tissue (visualized by CD31) and VCAM1 in mice treated with control IgG or αCD93 antibodies AD169-85 or AD169-309. Scale bar 50 μm. B) Quantification graph shows the percentage of VCAM1 coverage relative to the total CD31 area. C) Representative immunofluorescence staining of CD8-positive T cells in the control IgG, AD169-85, and AD169-309 groups. Nuclei are visualized by Hoechst staining. Scale bar 50 μm. D) Quantification graph shows the percentage of CD8 coverage relative to the total nuclear area. E) Immunofluorescence image showing CD3-positive T cells. Scale bar 20 μm. F) Quantification graphs show the percentage of CD3 coverage area within the visual field. *p<0.05, ns = not significant. One-way ANOVA using Dunnet's multiple comparison test. G) Survival curves of mice injected intracranially with GL261 glioma cells, showing improved survival in AD169-309 treated mice compared to a control IgG group after αPD1 blockade immunotherapy (AD169-309 + αPD1 vs. control IgG + αPD1). Rat IgG2 was used as an isotype IgG control for immunotherapy. *p<0.05, ***p<0.005. Log-Rank Mantel-Cox test. [Figure 26]This study demonstrates that treatment of GL261-carrying mice with the αCD93 blocking antibody AD169-85 reduces VE-cadherin levels in tumor blood vessels and increases endogenous fibrinogen leakage. A) Representative immunofluorescence staining of tumor vascular system (visualized by CD31) and VE-cadherin in mice treated with control IgG or αCD93 antibody AD169-85 or AD169-309. Scale bar 20 μm. B) Quantification graph showing the percentage of VE-cadherin-covered area relative to total CD31 area. C) Immunofluorescence staining showing signals for total endogenous fibrinogen (leaked into blood vessels and surrounding tissue) and leaked fibrinogen (only fibrinogen signals detected in surrounding tissue, excluding signals overlapping with the vascular region). Blood vessels are visualized by CD31. Scale bar 25 μm. D) Quantification graph showing the percentage of leaked fibrinogen relative to the total tissue area. *p<0.05, **p<0.01, One-way ANOVA with Dunnett multiple comparison test. [Figure 27] This shows the hCD93 construct used to determine the binding region of the αCD93 antibody.

[0016] The drawings are not necessarily to scale and generally show only the parts necessary to illustrate the concept of the present invention, with other parts omitted or merely suggested. [Modes for carrying out the invention]

[0017] This disclosure relates to novel drugs, binding proteins, or nucleic acid molecules encoding such drugs / binding proteins, including monoclonal antibodies or their antigen-binding fragments, that selectively bind to differentiation antigen group 93 (CD93) and inhibit the migration and entry of perivascular tumor cells. The drugs, binding proteins, and antibodies may be used in medical treatments such as cancer therapy. Binding proteins may also be used in cells engineered to express chimeric antigen receptors having the binding proteins of this disclosure as antigen-binding domains.

[0018] The purpose of this disclosure is to provide novel and enhanced CD93-specific conjugates that also affect / impair the ability of tumor cells to migrate along blood vessels within the vascular system, invade normal brain tissue, and proliferate, which may be used in therapy and cell engineering.

[0019] Aspects of this disclosure will be described in full below with reference to the accompanying drawings. However, the agents, binding proteins, nucleic acids, and methods disclosed herein can be realized in many different forms and should not be construed as being limited to the aspects described herein. Similar figures in the drawings refer to similar elements throughout.

[0020] The terms used herein are intended solely to describe specific aspects of the disclosure and are not intended to limit the disclosure. Where used herein, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0021] In some embodiments, the non-limiting term “drug” is used herein to refer to an entity that includes or codes for a binding site, where the binding site binds to CD93. Binding of a drug to CD93 typically provides an antagonistic effect, which can prevent functional activation of CD93 by its native ligand, such as by preventing the binding of the native ligand to CD93. The drug binding site may be a binding protein, such as an antibody, or a binding fragment thereof.

[0022] In some embodiments, the non-limiting term “binding protein” is used. The term “binding protein” is used herein to refer to a binding protein that includes an antibody binding domain (i.e., a binding domain obtained from, derived from, or based on the antibody binding domain). Thus, a binding protein is an antibody-based or antibody-like molecule that includes an antibody binding site, or a binding site derived from an antibody. Therefore, it is an immunological conjugate.

[0023] In some embodiments, the non-limiting terms “antibody” or “its antigen-binding fragment” are used. The term “antibody” is used herein in its broadest sense, including both monoclonal and polyclonal antibodies. As is well known, an antibody is an immunoglobulin molecule that can specifically bind to a target (antigen), such as a protein, carbohydrate, polynucleotide, lipid, polypeptide, or other, via at least one antigen-recognition site located in the variable region of the immunoglobulin molecule. As used herein, the terms “antibody” or “its antigen-binding fragment” encompass not only full-length or intact polyclonal or monoclonal antibodies, but also any other modified configurations of immunoglobulin molecules containing antigen-recognition sites of the desired specificity, including their antigen-binding fragments, e.g., Fab, Fab', F(ab')2, Fab3, Fv, and their variants, fusion proteins containing one or more antibody moieties, humanized antibodies, chimeric antibodies, minibodies, diabodies, triabodies, tetrabodies, linear antibodies, single-chain antibodies, multispecific antibodies (e.g., bispecific antibodies), as well as glycosylated variants of antibodies, amino acid sequence variants of antibodies, and covalently modified antibodies.

[0024] As is known to those skilled in the art, an antibody is a protein containing four polypeptide chains: two heavy chains and two light chains. Typically, the heavy chains are identical to each other, and the light chains are identical to each other. The light chains are shorter (and therefore lighter) than the heavy chains. The heavy chains contain four or five domains, with a variable (VH) domain located at the N-terminus, followed by three or four constant domains (each being CH1, CH2, CH3, and CH4, if present, from the N-terminus to the C-terminus). The light chains contain two domains, with a variable (VL) domain located at the N-terminus and a constant (CL) domain located at the C-terminus. In the heavy chains, an unstructured hinge region is located between the CH1 and CH2 domains. The two heavy chains of an antibody are joined by disulfide bonds formed between cysteine ​​residues present in the hinge region, and each heavy chain is joined to one light chain by disulfide bonds between cysteine ​​residues present in the CH1 and CL domains, respectively. In mammals, two types of light chains are produced, known as lambda (λ) and kappa (κ). In the case of kappa light chains, the variable and constant domains are V, respectively. K and C KThis can be referred to as a domain. Whether a light chain is a λ or κ light chain is determined by its constant region, the constant regions of λ and κ light chains are different but the same for all light chains of the same type in any given species. Depending on the amino acid sequence of the constant domain of its heavy chain, antibodies are assigned to different classes. There are six major classes of antibodies: IgA, IgD, IgE, IgG, IgM, and IgY, some of which can be further divided into subclasses, e.g., IgG1, IgG2, IgG3, IgG4, IgGA1, and IgGA2. As used herein, the term “full-length antibody” refers to any class of antibody, such as IgD, IgE, IgG, IgA, IgM, or IgY (or any of their subclasses). The term “antigen-binding fragment” refers to a portion or region of an antibody molecule or its derivative that holds all or a significant portion of the antigen binding of the corresponding full-length antibody. In some embodiments, the antibody heavy chain may consist of VH+CH1+hinge+CH2+CH3 and a light chain VL+CL. In preferred embodiments, the antibody has an IgG1 LALA format, where CH1 is defined by SEQ ID NO: 30, CH2 is defined by SEQ ID NO: 31, CH3 is defined by SEQ ID NO: 32, CL is defined by SEQ ID NO: 33, and hinge is defined by SEQ ID NO: 34.

[0025] As briefly mentioned above, examples of antigen-binding fragments include: (1) Fab fragments, which are monovalent fragments having VL-CL chains and VH-CH chains; (2) Fab' fragments, which are Fab fragments having a heavy chain hinge region; (3) F(ab')2 fragments, which are dimers of Fab' fragments joined by a heavy chain hinge region, for example, linked by disulfide crosslinks in the hinge region; (4) Fc fragments; (5) Fv fragments, which are minimal antibody fragments having VL and VH domains of a single arm of the antibody; and (6) scFv V H and V L A single-stranded Fv(scFv) fragment is a single polypeptide chain in which the domains are linked by a peptide linker, (7) two VH domains are associated through two VH domains via disulfide crosslinks. HThe domain antibodies include, but are not limited to, (scFv)2, which contains a domain and two VL domains, and (8) a single variable domain (VH or VL) polypeptide that specifically binds to an antigen. Antigen-binding fragments can be prepared by routine methods. For example, F(ab')2 fragments can be produced by pepsin digestion of a full-length antibody molecule, and Fab fragments can be produced by reducing the disulfide crosslinks of F(ab')2 fragments. Alternatively, fragments can be prepared by recombinant techniques by expressing heavy chain and light chain fragments in a suitable host cell (e.g., E. coli, yeast, mammalian, plant, or insect cell) and assembling them to form the desired antigen-binding fragment either in vivo or in vitro. Single-chain antibodies can be prepared by recombinant techniques by linking a nucleotide sequence encoding a heavy chain variable region with a nucleotide sequence encoding a light chain variable region. For example, a plastic linker can be incorporated between the two variable regions. Thus, the general terms “binding protein” or “antibody” are used throughout this specification. These terms are used in their broadest sense and therefore also incorporate all variants and fragments described above and below. In some embodiments, the binding protein is a monoclonal antibody or an antigen-binding fragment selected from the group consisting of Fv fragments (e.g., single-chain Fv and disulfide-linked Fv), Fab-like fragments (e.g., Fab fragment, Fab' fragment, and F(ab)2 fragment), and domain antibodies (e.g., single VH variable domain or VL variable domain).

[0026] Therefore, the constant region of the heavy chain is the same for all antibodies of any given isotype in a species, but differs between isotypes. The specificity of an antibody is determined by the sequence of its variable region. The sequence of the variable region varies among antibodies of the same type in any given individual. In particular, both the light and heavy chains of an antibody contain three hypervariable complementarity-determining regions (CDRs). In a light-heavy chain pair, the CDRs of the two chains form the antigen-binding site. The CDR sequence determines the specificity of the antibody. The light-heavy chain variable region pair containing the (antigen)-binding site is known as the (antigen)-binding domain. The three CDRs of the heavy chain are known as VHCDR1, VHCDR2, and VHCDR3 from N-terminus to C-terminus, and the three CDRs of the light chain are known as VLCDR1, VLCDR2, and VLCDR3 from N-terminus to C-terminus.

[0027] In antibodies, as described above, the CDR sequence is located in the variable domains of the heavy and light chains. The CDR sequence is located within a polypeptide framework that appropriately positions the CDR for antigen binding. Thus, the remainder of the variable domain (i.e., the portion of the variable domain sequence that does not form part of any one of the CDRs) constitutes the framework region. The N-terminus of the mature variable domain forms framework region 1 (FR1), the polypeptide sequence between CDR1 and CDR2 forms FR2, the polypeptide sequence between CDR2 and CDR3 forms FR3, and the polypeptide sequence that links CDR3 to the constant domain forms FR4. In the binding protein of the present invention, the variable domain framework region may have any suitable amino acid sequence so that the binding protein binds to CD93 via its CDR.

[0028] If the binding protein is an antibody, the antibody can be of any isotype and subtype. Thus, it can be an IgA, IgD, IgE, IgG, or IgM antibody. The heavy chain constant domains corresponding to different isotypes of immunoglobulin are denoted as α, δ, ε, γ, and μ, respectively. The subunit structures and three-dimensional structures of different isotypes of immunoglobulin are well known. Preferably, the antibody is an IgG antibody. As mentioned above, there are four subtypes of IgG antibodies: IgG1, IgG2, IgG3, and IgG4. The IgG αCD93 antibody of the present invention can be of any IgG subtype, i.e., it can be an IgG1, IgG2, IgG3, or IgG4 antibody. In a preferred embodiment, the antibody is an IgG1 antibody, such as an IgG1 LALA antibody. In IgG1 LALA, leucine (L) at amino acid positions 234 and 235 in the Fc region is substituted with alanine (A). LALA mutations eliminate Fc-mediated binding to the Fcγ receptor on immune cells, which impairs effector function. By eliminating this binding, immune responses are avoided, i.e., immunogenicity mediated by Fc effector function such as ADCC and CDC, as well as the risk of biopharmaceuticals causing undesirable off-target and on-target side effects.

[0029] Alternatively, the binding protein may be an antibody-binding fragment (i.e., an antibody fragment), that is, a fragment that retains the ability of the antibody to specifically bind to CD93. Such fragments are well known, and examples include Fab', Fab, F(ab')2, Fv, Fd, or dAb fragments, which can be prepared according to techniques well known in the art.

[0030] A Fab fragment consists of the antigen-binding domain of an antibody; that is, an individual antibody may contain two Fab fragments, each Fab fragment consisting of a light chain and the N-terminal portion of the heavy chain to which it is bound. Therefore, a Fab fragment consists of the entire light chain and the N-terminal portion of the heavy chain to which it is bound. H and C HIt contains 1 domain. A Fab fragment can be obtained by digesting an antibody with papain.

[0031] An F(ab’)2 fragment consists of the hinge region of the heavy chain that contains a disulfide bond linking the two heavy chains together in addition to the two Fab fragments of the antibody. In other words, an F(ab’)2 fragment can be regarded as two covalently linked Fab fragments. An F(ab’)2 fragment can be obtained by digesting an antibody with pepsin. Reduction of the F(ab’)2 fragment yields two Fab’ fragments, which can be regarded as Fab fragments containing additional sulfhydryl groups that may be useful for conjugation of the fragment to other molecules.

[0032] Alternatively, the binding protein can be a synthetic or artificial construct, i.e., an antibody-like molecule that contains a binding domain but has been genetically engineered or artificially constructed. This includes chimeric or CDR grafted antibodies, as well as single-chain antibodies, and other constructs such as scFv, dsFv, ds-scFv, dimers, minibodies, diabodies, single domain antibodies (DAB), TandAbs dimers, and V H H and other heavy chain antibodies such as. In certain embodiments, the artificial construct is a single-chain variable fragment (scFv). An scFv is a fusion protein in which a single polypeptide contains both the V H and V L domains of an antibody. An scFv fragment generally contains a peptide linker that covalently joins the V H and V L regions, which contributes to the stability of the molecule. The linker can contain, for example, 1, 2, 3, or 4 amino acids, 5, 10, or 15 amino acids, or other intermediate numbers in the convenient range of 1 - 20 such as 1 - 20 amino acids. The peptide linker can be formed from any generally convenient amino acid residues such as glycine and / or serine as well known to those skilled in the art. However, the presence of a linker is not essential, and the V L domain is linked to the V HIt can be linked to a domain. scFv is typically V, from the N-terminus to the C-terminus, via the linker sequence. L V connected to the region H This includes the region. The preparation of scFv molecules is well known in the art.

[0033] The binding domain of an antibody consists of a light chain variable domain and a heavy chain variable domain (classical bivalent antibodies have two binding domains). Therefore, the binding protein may be a native antibody or a fragment thereof, or an artificial or synthetic antibody, or an antibody construct, or a derivative (e.g., a single-chain antibody, as further discussed below). In summary, the binding protein of the present invention comprises an antibody binding domain, which comprises a light chain variable domain and a heavy chain variable domain.

[0034] As used herein, the term "capable of binding to X," where X is the antigen, refers to the properties of an antibody or its binding fragment that can be tested, for example, by ELISA, by the use of surface plasmon resonance (SPR) technology, by the use of kinetic exclusion assay (KinExA®), or by biolayer interferometry (BLI). Those skilled in the art will be familiar with such methods and others.

[0035] The term "antagonism," as well as "antagonist" or "antagonistic antibody," refers to a drug or monoclonal antibody that blocks the effector response (antagonistic effect) by binding to a receptor. Antagonistic antibodies mediate antagonist activity by blocking or inhibiting the binding of the native ligand to the receptor.

[0036] The term "specificity," sometimes referred to as "selectivity" of a drug, drug binding site, or binding protein, refers to a drug, binding site, or binding protein that binds to a target with high affinity but typically does not bind to other antigens. A selective or specific binding agent / site / protein / antibody does not cross-react with other targets other than the intended antigen, or cross-reacts to a low degree. Therefore, by "specifically" binding, it means, for example, that a binding protein binds to its target (i.e., CD93) in a manner that can be distinguished from binding to a non-target molecule, and more specifically, that the binding protein binds to its target (CD93) with a higher binding affinity than it binds to other molecules. That is, the binding protein does not bind to other, non-target molecules, or does not bind to a considerable or significant degree, or binds to such other molecules with a lower affinity than it binds to CD93. A binding protein that "specifically binds" to CD93 may alternatively be referred to as "counteracting" or "recognizing" CD93. In other words, CD93 is the antigen of the binding protein of the present invention, and the binding protein is an "antigen-binding protein" in the sense that it binds to CD93 as its antigen.

[0037] As used herein, “therapy” means the treatment of any medical condition. Such therapy may be prophylactic (i.e., preventive), curative (or a procedure intended to cure), or palliative (i.e., a procedure designed simply to limit, alleviate, or improve the symptoms of a condition). Thus, as used herein, “therapy” or “treating” a disorder such as cancer / cancerous tumors means preventing or improving a particular disorder or medical condition, or curing it. In the case of cancer and tumors, therapy may reduce or eliminate the current tumor, or they may halt or prevent further spread of the tumor. A sufficient amount to achieve this is defined as a “therapeutic effective dose.” The effective dose for a given purpose depends on the disease or condition being treated, its severity, and the size / weight and general condition of the subject. Thus, the agents or binding proteins described herein may be used to treat / therapy any condition in which a target antigen is expressed / overexpressed in a subject such as tumor blood vessels to improve the condition, and may be administered systemically or topically, and by any preferred method known in the art. The subjects as defined herein include any mammal, such as livestock such as cattle, horses, sheep, pigs, or goats; pet animals such as rabbits, cats, or dogs; or primates such as monkeys, chimpanzees, gorillas, or humans. Most preferably, the subject is human.

[0038] The term "immunotherapy" refers to any type of treatment intended to stimulate an immune response against a tumor, i.e., to target the immune system. This may include, for example, checkpoint blocker antibodies, vaccines, CAR-T cells, and AAV vectors. While tumor-targeting antibodies are considered a type of immunotherapy, immunotherapy can also refer to the more general use of checkpoint blocker antibodies or similar forms to stimulate an immune response.

[0039] Preventive treatment may include preventing a condition or delaying its onset or development. For example, binding proteins may be used to prevent, delay, or reduce the development or recurrence of cancer by influencing tumor growth, tumor cell migration, and tumor cell invasion into the perivascular system, or, for example, to prevent or reduce metastasis.

[0040] When this agent or binding protein binds to target CD93, it may inhibit the ability of tumor cells to maintain their blood supply, for example, by mitigating vascular hijacking in the target. When this agent or binding protein binds to CD93, it may inhibit "tumor cell invasion." "Invasion" refers to the (perivascular) migration of tumor cells along the visible side of the vascular system. This is illustrated in Figure 1, which shows the effect of common CD93 conjugates on perivascular invasion of tumor cells. Above, an untreated invasive tumor is shown, where tumor cell migration is also facilitated by the interaction of tumor cells with proteins in the extracellular matrix, such as fibronectin. Below, a tumor treated with an αCD93 antibody is shown, where tumor cell invasion in the vascular system near the tumor cells is inhibited. When this agent or binding protein binds to CD93, it may inhibit tumor cell migration. Glioma tumor cells have been shown to migrate along blood vessels, which is beneficial for their diffusion and blood supply. Therefore, "with tumor cell migration" means the migration of tumor cells along the visible side of blood vessels in the vascular system near the tumor. Therefore, this movement is mitigated when the drug / binding protein is administered. The drug or binding protein may also inhibit tumor cell proliferation, i.e., the proliferation of tumor cells by division, when bound to CD93. The proliferation of any tumor cells that grow by using angiography / vascular hijacking (instead of new angiography / angiogenesis) is affected.

[0041] This disclosure also includes nucleic acids encoding drugs or binding proteins, and therefore refers to nucleic acid molecules that, when translated, produce a drug, a drug-binding portion or binding protein, or a portion thereof such as an antibody-binding domain. Nucleic acid molecules may be in the form of mRNA expressed in mammalian cells and may be inserted into mammalian cells using the CRISPR / Cas method. A “vector,” such as an expression vector, may include a nucleic acid molecule, and a vector may be a plasmid or a viral vector. Viral vectors are tools commonly used by molecular biologists to deliver genetic material to cells. This process can be carried out in vivo or in cell culture. The delivery of genes or other genetic material by a vector is called transduction, and the infected cell is described as a transduced cell. In some embodiments, the viral vector is contained in a “host cell,” which is a cell capable of translating the nucleic acid molecule contained in the vector.

[0042] All solid tumors require vascular supply to progress. While the ability to induce angiogenesis (new blood vessel growth) is considered important for this purpose, it has also been shown that tumors can grow by hijacking existing blood vessels in surrounding non-malignant tissue, a process called vascular hijacking. Therefore, vascular hijacking is a crucial mechanism of tumor angiogenesis that can influence disease progression, metastasis, and response to treatment. Thus, methods to inhibit this process could lead to effective treatment of invasive cancers that utilize this process.

[0043] Differentiation antigen group 93 (CD93) is a type C lectin transmembrane receptor that plays a role not only in cell-cell adhesion processes but also in host defense. CD93 is expressed by a wide variety of cells, including platelets, monocytes, microglia, and endothelial cells. CD93 is also expressed in neutrophils, activated macrophages, B cell precursors, and subsets of dendritic cells and natural killer cells in the immune system. CD93 is involved in endothelial cell-cell adhesion, cell diffusion, cell migration, cell polarization, and tubular morphogenesis. Recently, it has been found that CD93 can regulate endothelial cell dynamics through its interaction with the extracellular matrix glycoprotein MMRN2.

[0044] CD93 is upregulated in tumor vessels of glioblastoma patients and has been found to be important in glioma development. CD93 influences the ability of endothelial cells to regulate their cytoskeleton and is important for angiogenesis. CD93 is highly expressed in glioma vessels, and high levels of vascular CD93 are associated with decreased patient survival. CD93 knockout mice with gliomas exhibit smaller tumor size and improved survival. It has been shown that it is possible to block angiogenesis in vitro either by knockdown of CD93 using siRNA or shRNA, or through treatment with mouse antibody 4E1, a neutralizing antibody that disrupts the CD93-MMRN2 complex (M. Orlandini et al. Oncotarget 2014 May 15;5(9):2750-60). It has been further shown that CD93 organizes fibronectin in the extracellular matrix by interacting with IGFBP7 (US2022 / 0235136) and MMRN2, and by regulating the activity of α5β1 integrin. Therefore, targeting the activity of MMRN2 (WO2018 / 020222) and α5β1 integrin was the objective of this field.

[0045] Despite these advances, effective therapies for gliomas are still lacking. Because gliomas are difficult to treat, partly due to their highly invasive nature, it was hypothesized that drugs affecting their invasiveness would yield therapeutic effects. Perivascular invasion is one of the pathways used by tumor cells to spread within the brain, and the extracellular matrix, including fibronectin, is essential for tumor cell migration and invasion. It has been shown that the CD93-MMRN2 complex promotes the formation of perivascular fibronectin (FN) fibers, and that CD93 knockout disrupts perivascular FN fibers. Surprisingly, current data from the inventors indicate that CD93 knockout or downmodulation inhibits tumor invasion and tumor cell proliferation, and that the inhibition of tumor invasion in response to CD93 deficiency is associated with a reduction in perivascular fibronectin fibers. Initially, it was hypothesized that CD93 plays a role in regulating the invasion of glioma cells; while mouse glioma cells can migrate along blood vessels in wild-type mice, CD93 knockout inhibits perivascular fibronectin fibers. - / - In mice, it was found that CD93 did not migrate, leading to the interpretation that antibodies or peptides inhibiting α5β1-integrin could reduce fibronectin fibril formation and glioma invasion. However, even though CD93 is required for perivascular invasion by tumor cells, it appears to have a role independent of the proposed complex with α5β1-integrin.

[0046] Furthermore, it has been found that neutralizing antibodies against CD93 can disrupt endothelial junctions and already formed fibronectin filaments. This data indicates that neutralizing or antagonistic antibodies against CD93 not only inhibit the active process of angiogenesis but also actually disrupt endothelial junctions, leading to a regression of the fibronectin network. In addition, CD93 - / -In mice, enhanced expression of leukocyte adhesion molecules was found in tumor blood vessels, and increased T cell abundance was observed in gliomas. Consistent with this, CD93 deficiency was associated with an improved response to αPD1 immunotherapy. Therefore, the novel data presented by the current inventors also suggest a previously unknown function of CD93 in regulating endothelial activation and adhesion molecule expression necessary for the recruitment of immune cells into the glioma tumor microenvironment. Thus, CD93 may also be a target for enhancing cancer immunotherapy.

[0047] The fact that antibodies against CD93 can disrupt already formed junctions and alter extracellular matrix remodeling is surprising. CD93 is upregulated during active angiogenesis, and its role in maintaining endothelial junctions is unexpected. Importantly, this suggests that neutralizing antibodies against CD93 may have a significant effect on vascular permeability and could be used in conjunction with other cancer drugs, such as temozolomide.

[0048] Therefore, it was concluded that drugs that target vascular CD93 and block cell entry and proliferation by inhibiting perivascular cell migration are beneficial for the treatment of cancer / tumors. Drugs that block the entry and proliferation of glioma cells by targeting vascular CD93 can be used to treat gliomas. Such drugs may be, for example, specific blocking antibodies that block the function of human endothelial CD93 and possess the above-mentioned properties.

[0049] Therefore, we generated 27 CD93 antibodies that target mAbs (human (Hu), mouse (Mu), or human / mouse (Hu / Mu)) (Table 1), successfully converted 26 of them to hIgG1 LALA (Table 2), and tested their ability to block tumor perivascular invasion. Surprisingly, based on in vitro and ex vivo assays, it was found that only four antibodies (three mouse (Ms) / human (Hu) and one Hu) exhibited the desired properties, as illustrated in Table 3. [Table 1]

[0050] Table 1 shows the results from scFv and Fab clone interaction studies. Binding studies of the most promising purified scFv and Fab clones to hCD93(H), hCD93lectin(L), mCD93(M), and unrelated control protein(C). For ELISA and HTRF (Homogeneous Time-Resolved Fluorescence), color coding is as follows: black: maximum or near-maximum binding signal, stripe: weak binding signal, white: no binding. SPR single-cycle kinetics are expressed as the dissociation rate constant (K) in nanomoles. D Flow cytometry (FACS) data show the percentage (%) of positive cells for different clones. Human endothelial cells (H) and mouse endothelial cells (M). Hu Neg (human U87 cells, negative control), M Neg (mouse GL261 cells, negative control).

[0051] As illustrated in Table 2 below, interaction studies of the converted αCD93 antibody were conducted to evaluate its binding properties. [Table 2]

[0052] Table 2 above shows the binding studies of the most promising purified scFv and Fab clones after conversion to hIgG1 LALA. Binding to hCD93(H), hCD93lectin(L), mCD93(M), and unrelated control protein(C) was measured by ELISA and SPR. For ELISA, the color coding is as follows: black: maximum or near-maximum binding signal, stripe: weak binding signal, white: no binding. SPR single-cycle kinetics are expressed as the nanomolar dissociation rate constant (K). DFlow cytometry (FACS) data show the percentage (%) of positive cells for different clones. Human endothelial cells (H) and mouse endothelial cells (M). Hu Neg (human U87 cells, negative control), M Neg (mouse GL261 cells, negative control).

[0053] The following Table 3 illustrates the study of the functional properties of αCD903 antibodies to determine which antibodies possess the desired functionality / characteristics / effects. [Table 3] [Table 4]

[0054] Twenty-six αCD93 antibodies (10 μg / ml) were tested for their functional effects using in vitro and ex vivo models. For the in vitro assay, human and mouse endothelial cells (ECs) were used. Intercellular gap formation (Gap), fibronectin deposition (FN), and cell migration (Migr.) were evaluated. For the ex vivo assay, co-cultures of glioma cells on human and mouse brain slice models were used. Perivascular invasion (PV invasion) of tumor cells and perivascular deposition of fibronectin (PV-FN deposition) were evaluated. The scoring criteria and symbols used to evaluate the effects of αCD93 antibodies in different assays are as follows: For "Gap," the increased multiplier change compared to the control isotype IgG was calculated: x (2-4x change), xx (4-8x change), xxx (>8x change). For "FN", the percentage of fibronectin reduction against control isotype IgG was measured: x (20-40% reduction), xx (40-60% reduction), xxx (>60% reduction). For EC transfer, the percentage of inhibition against isotype control IgG was measured: x (20-40% inhibition), xx (40-60% inhibition), xxx (>60% inhibition). For "PV entry" and "PV FN deposition": x (weak inhibition), xx (partial inhibition), xxx (strong inhibition). - (no effect), nd (not tested). Four antibodies with the desired effect(s) were identified, and these are referred to as AD169-85, AD169-309, AD169-324, and AD169-8, as illustrated at the top of Table 3. The K of these antibodies is shown in Table 2. D This range is 2 to 21 nM.

[0055] All four binding proteins / mAbs obtained (AD169-85, AD169-309, AD169-324, and AD169-8) were K DThese four binding proteins bind to vascular CD93 with high affinity, having a <21nM mass, and all exhibit inhibitory effects against tumor cell proliferation and tumor (glioma) cell invasion, resulting in inhibition of perivascular tumor cell migration, i.e., the migration of tumor cells along the vascular system. Therefore, administration of the four binding proteins to a subject, and their subsequent binding to vascular CD93, results in one or more of the following: i) inhibition of perivascular tumor cell migration, ii) inhibition of tumor cell invasion, and iii) inhibition of tumor cell proliferation. Thus, vascular hijacking, migration, and proliferation of tumor cells can be mitigated / inhibited using these binding molecules or the nucleic acid molecules encoding them. These effects mediated by the binding proteins result in increased survival in mice with cancer, as shown in the Examples section below. The structures of the four binding molecules (antibodies) also exhibit a common CDR motif, as defined below. They are also all of human origin, compared to mouse or humanized versions.

[0056] Therefore, a binding protein that specifically binds to vascular CD93 and inhibits perivascular tumor cell migration is provided herein. The binding protein comprises light chain and heavy chain variable domains. The light chain variable domain and the heavy chain variable domain each contain three CDRs, the light chain variable domains containing VLCDR1, VLCDR2, and VLCDR3, and the heavy chain variable domains containing VHCDR1, VHCDR2, and VHCDR3. The six CDRs have the following amino acid sequence: VHCDR1, defined by sequence number 1, VHCDR2, defined by Sequence ID 2, VHCDR3 as defined by Sequence ID 3, VLCDR1, defined by sequence number 4, VLCDR2 is defined as AAS; VLCDR3 as defined by sequence number 5. Here, VLCDR1 and VLCDR2 are identical in all four binding protein variants, while VHCDR1, VHCDR2, VHCDR3, and VLCDR3 include several variants, as shown in Table 4 below. In addition to CDRs by SEQ ID NOs: 1-5 and AAS, the present invention also includes CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or higher, which exhibit similar effects on tumor growth, cell entry, and tumor cell migration upon CD93 binding. [Table 5]

[0057] The sequences of the binding protein CDRs are shown in Table 4 with single-letter amino acid codes, and VHCDR1, VHCDR2, VHCDR3, and VHCDR3 contain sequence variations between the four binding protein variants, as indicated by X. In some cases, one of the CDR sequences contains fewer residues than the motif, and X is indicated as a "gap" (i.e., none) where there is no residue at that position. In VHCDR1, X5 can be G, S, or Y, X6 can be S or Y, X7 can be Y or S, and X8 can be S or G. In VHCDR2, X2 can be S or Y, X3 can be G, S, or Y, X4 can be S, Y, or G, and X7 can be S, G, or Y. In VHCDR3, X3 can be Y, S, or P; X4 can be G, D, or S; X5 can be W, Y, or G; X6 can be T or a gap; X7 can be Y or a gap; X8 can be P or a gap; X9 can be D, Y, or a gap; X 10 is Y, V, or G, and X 11 X can be I, L, or F. In VLCDR3, X3 can be S or R, X5 can be S or a gap, X6 can be Y, T, or F, and X8 can be P or Y.

[0058] In some embodiments, some CDRs are surrounded by specific amino acids called framework amino acids (faa), which are conserved with some minor mutations between different binding proteins. Thus, the binding proteins VHCDR1 and VHCDR2 may be located next to specific framework amino acids, and the CDR and framework amino acid sequences are, VHCDR1 and faa, defined by Sequence ID 6, As shown in Table 5 below, the selection is made from the group including VHCDR2 and faa as defined by Sequence ID No. 7. In addition to CDR+faa as defined by Sequence ID Nos. 6-7, the present invention also includes CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more, which also exhibit the same effect when bound to CD93. [Table 6]

[0059] In VHCDR1+faa, X 10 X1 can be S or Y, and the remaining unknown corresponds to the sequence in Table 4 above. In VHCDR2+faa, faa is inserted at the beginning and end, and the first one, X1, can be A or G, and the remaining unknown corresponds to the sequence in Table 4 above, but the numbering has been changed by the amino acid residue inserted at the beginning, so X3 corresponds to X2 in Table 4, and so on.

[0060] In some embodiments, four different binding proteins are presented, each having one of the individual CDR combinations shown in Tables 6-7 below, where Table 6 shows CDRH1, H2, and H3 for the four different binding proteins, and Table 7 shows CDRL1, L2, and L3. [Table 7] [Table 8]

[0061] In some embodiments, four different binding proteins are presented, each having a surrounding framework amino acid for the individual CDRs described above, as well as for VHCDR1 and VHCDR2, as shown in Table 8 below. [Table 9]

[0062] The binding proteins of the present invention can be synthesized by any method known in the art. Preferably, the binding proteins are synthesized using a protein expression system, such as a cell expression system using prokaryotic (e.g., bacterial) cells or eukaryotic (e.g., yeast, fungi, insects, or mammals) cells. An alternative protein expression system is a cell-free in vitro expression system in which the nucleotide sequence encoding the binding protein is transcribed into mRNA, and the mRNA is translated into a protein in vitro. Cell-free expression system kits are widely available and can be purchased, for example, from Thermo Fisher Scientific. Alternatively, the binding proteins can be chemically synthesized in a non-biological system. Liquid-phase synthesis or solid-phase synthesis can be used to form the binding proteins of the present invention or to generate polypeptides that may be contained within the binding proteins of the present invention.

[0063] Those skilled in the art can readily produce binding proteins using appropriate methodologies common in the art. In particular, binding proteins can be recombinantly expressed in mammalian cells such as CHO cells. Binding proteins synthesized in a protein expression system can be purified using standard techniques in the art; for example, they can be synthesized with affinity tags and purified by affinity chromatography. If the binding protein is an antibody, it can be purified using affinity chromatography with one or more antibody-binding proteins such as protein G, protein A, protein A / G, or protein L.

[0064] As described above, binding proteins are typically antibody-based or antibody-like molecules. Therefore, binding proteins can be native antibodies or fragments thereof, or artificial or synthetic antibodies, or antibody constructs or derivatives (e.g., single-chain antibodies). In preferred embodiments, the binding protein is a human protein (of human origin, compared to humanized ones), particularly a human monoclonal antibody, antibody fragment, or scFv. Human binding proteins may include the VH and VL regions, both of which are derived from human germline immunoglobulin sequences, as well as the human constant region, if the constant region is present in the protein. However, such proteins may contain amino acids not encoded by human germline Ig sequences, e.g., mutations introduced by random or site-directed mutagenesis.

[0065] As described above, the binding protein of the present invention includes an antibody binding domain, and the binding domain includes a heavy chain variable domain (or variable region) and a light chain variable domain. Therefore, in certain embodiments, the binding protein of the present invention is (i) A heavy chain variable domain (VH) containing (or consisting of) an amino acid sequence described in any one of sequence numbers 22-25, or a variant thereof, (ii) A light chain variable domain (VL) comprising (or consisting of) an amino acid sequence described in any one of Sequence IDs 26-29, or a variant thereof. The variants may define sequences that are 80% or more identical to them, such as 85%, 90%, 95%, or more, and the CDR sequence contains no mutations in the amino acid sequence, or the sequence mutations in the CDR amino acid sequence are at most 5%, such as 4%, 3%, 2%, 1%, or less. Thus, a variant defined as a sequence that is 80% or more identical to them, such as 85%, 90%, 95%, or more, is conditional on the variant's CDR sequence not being modified to account for the antibody variant defined by the VH or VL domain, and as a result the sequence mutations in the CDR amino acid sequence are at most 5%, such as 4%, 3%, 2%, 1%, or less, and as a result the sequence forming the CDR has at least 95% identity to it, such as 96%, 97%, 98%, 99%. Binding proteins having variants in the variable domain and / or constant domain sequences are functional variants with the activity described above (i.e., they specifically bind to CD93 and inhibit one or more of the proliferation, invasion, and migration of tumor cells). Variant sequences may be modified from the native sequence by substitution, insertion, and / or deletion of one or more amino acids.

[0066] Sequence identity can be assessed by any convenient method. However, computer programs that perform pairwise or multiple alignment of sequences are useful for determining the degree of sequence identity between sequences. For example, the EMBOSS needle or EMBOSS stretcher (both, Rice, P. et al., Trends Genet., 16, (6) pp276-277, 2000) can be used for pairwise sequence alignment, while Clustal Omega (Sievers F et al., Mol.Syst.Biol.7:539, 2011) or MUSCLE (Edgar, RC, Nucleic Acids Res.32(5):1792-1797, 2004) can be used for multiple sequence alignment, although any other suitable program may be used. Whether the alignment is pairwise or multiple, it must be performed globally (i.e., across the entire reference sequence), not locally. Sequence alignment and % identity calculation can be determined, for example, using standard Clustal Omega parameters: matrix Gonnet, gap start penalty 6, gap extension penalty 1. Alternatively, standard EMBOSS needle parameters: matrix BLOSUM62, gap start penalty 10, gap extension penalty 0.5 may be used. Alternatively, any other suitable parameters may be used.

[0067] In some embodiments, the binding protein is an antibody comprising a heavy chain and a light chain including variable and constant regions, the heavy chain comprising a VH domain and three constant domains, CH1, CH2, and CH3, CH1 and CH2 conjugated via a hinge region, and the light chain comprising a VL domain and a constant domain, CL. The binding protein may be a monoclonal antibody of an IgG1 isotype, such as an IgG1 LALA antibody, and the amino acid sequences of the domains and hinge regions are defined by SEQ ID NO: 30 for CH1, SEQ ID NO: 31 for CH2, SEQ ID NO: 32 for CH3, SEQ ID NO: 33 for CL, and SEQ ID NO: 34 for the hinge region.

[0068] The binding protein may be a monoclonal antibody or an antigen-binding fragment selected from the group consisting of an Fv fragment, a Fab-like fragment, a disulfide-binding fragment, and a domain antibody, where the Fv fragment may be an scFv fragment and the Fab-like fragment may be a Fab or F(ab')2 fragment.

[0069] The binding protein has been shown to bind to CD93 with high affinity and can therefore be used as a standalone cancer treatment as described above. The binding protein of this disclosure can also be used to design chimeric antigen receptors (CARs) for CD93 in order to obtain CD93-targeted CAR cells such as CAR-T cells. Antibodies can also be produced as a separate portion within the CAR to deliver T cells to the tumor, i.e., by designing cells engineered to express a CAR in which an antibody or antigen-binding domain is produced intracellularly and directed to direct it towards tumor blood vessels via CD93 binding.

[0070] These CD93-targeted CAR T cells can then be used therapeutically to bind to CD93-expressing cells within tumor blood vessels, eliminating surrounding malignant cells or other CD93-expressing cells such as cancer cells. For example, CAR T cells are created by isolating T cells from a subject and inserting the CAR gene into the T cells to create CAR T cells that express the CAR protein. CAR is a hybrid of a T cell and an antibody receptor, containing four distinct regions: an extracellular domain that recognizes an antigen (typically an scFv fragment of an antibody) connected to a transmembrane domain by a hinge (spacer); the transmembrane domain has a hydrophobic alpha-helix structure; and the transmembrane domain is connected to an endodomain (intracellular domain) that undergoes a conformational change after antigen recognition, which triggers downstream signaling pathways to induce an immune response. The endodomain may also contain one or more costimulatory domains to enhance antitumor activity. Accordingly, the present disclosure provides cells engineered to express a chimeric antigen receptor (CAR), the CAR comprising an antigen-binding domain, a transmembrane domain connected to the antigen-binding domain by a hinge region, and an intracellular domain optionally connected to one or more costimulatory domains, wherein the antigen-binding domain comprises an scFv fragment of a binding protein. The cells may be human cells, T cells, NK cells, or immune effector cells such as macrophages. Accordingly, aspects of the present disclosure include cells engineered to express a chimeric antigen receptor (CAR), the CAR comprising an antigen-binding domain, a transmembrane domain connected to the antigen-binding domain by a hinge region, and an intracellular domain optionally connected to one or more costimulatory domains, wherein the antigen-binding domain comprises a binding protein of the present disclosure, and the cells may be human cells. The cells may also be immune effector cells such as T cells, NK cells, or macrophages.

[0071] Further provided herein are nucleic acid molecules encoding the drugs or binding proteins of the Disclosure. A nucleic acid molecule may encode a binding protein containing an antibody-binding domain, the binding domain comprising a heavy-chain variable domain (VH) and a light-chain variable domain (VL) as provided herein. Thus, the drugs or binding proteins may be administered in protein form, or they may be administered in nucleic acid form so as to be expressed in the body of the subject. Therefore, any nucleotide sequence may be used to translate into any one of the drugs or binding proteins of the Disclosure. A nucleic acid molecule contains a nucleotide sequence in the form of mRNA expressed in mammalian cells. In some embodiments, the nucleotide sequence of the nucleic acid molecule is inserted into mammalian cells using the CRISPR / Cas method. The nucleic acid molecule may also be contained in a vector, such as an expression vector. The vector may be a plasmid or a viral vector. The vector may be transfected into isolated host cells for translation of the nucleic acid.

[0072] Furthermore, drugs that may contain a binding site specifically that binds to vascular CD93 are also provided herein. Binding of the drug to CD93 may provide an antagonistic effect, such as by preventing the functional activation of CD93 by its natural ligand. The drug, or its binding site, may be or encode a binding protein, the encoded binding protein binding to CD93.

[0073] This invention provides novel enhanced binding proteins that bind to human vascular CD93. The binding proteins of this invention have been evaluated for binding affinity and therapeutic efficacy, and it has been shown that they offer enhanced performance compared to other binding proteins in the art, as they provide efficacy or inhibited tumor invasion. As mentioned above, several CD93-binding antibodies have been produced, but only four exhibited the desired property of inhibiting glioma invasion. Therefore, it has been demonstrated that binding CD93 is not sufficient to mediate this effect, and that several other interactions, such as steric hindrance, may exist. Accordingly, this disclosure relates to agents such as binding proteins that specifically bind to CD93, where binding results in one or more of the following: i) inhibition of perivascular tumor cell migration, ii) inhibition of tumor cell invasion, and iii) inhibition of tumor cell proliferation. One important factor is that binding affects tumor cell migration without damaging / killing endothelial cells. Since vascular hijacking, in addition to angiogenesis, can play a significant role in tumor growth, inhibiting this mechanism offers new therapeutic possibilities.

[0074] This disclosure relates to a drug comprising a binding moiety that specifically binds to vascular CD93, wherein the drug or its binding moiety is a binding protein. Binding of the drug to CD93 may provide an antagonistic effect by preventing functional activation of CD93 by its native ligand. Since blockade of fibronectin fibril formation is observed, this is hypothesized to lead to altered interaction with MMRN2. The drug or binding protein may inhibit the ability of tumor cells to maintain their blood supply, for example, by mitigating vascular hijacking. Therefore, binding of the drug or binding protein to CD93 results in one or more of the following: i) inhibition of perivascular tumor cell migration, ii) inhibition of tumor cell invasion, and iii) inhibition of tumor cell proliferation.

[0075] One objective of this disclosure is to provide improved therapies, particularly novel enhanced tumor targeting agents for cancer therapy. The binding protein is introduced into the body by various means (such as injection or ingestion), localizes a specific antigen (CD93), binds to a receptor, and either remains on the cell surface or is internalized within the cell.

[0076] The binding proteins of this disclosure may be used to treat several cancers in which vascular hijacking is observed and / or associated with the expression / overexpression of the target antigen CD93, as described above and below. CD93 may be targeted to any type of cancer in which it is expressed in tumor blood vessels / vascular systems. Subjects suffering from brain cancer, particularly gliomas including glioblastoma, can greatly benefit from these treatments, as effective treatments are often unavailable. As mentioned above, gliomas are highly invasive tumors, and due to their invasiveness, complete surgical resection of the tumor is not possible. Therefore, the ability to reduce this invasiveness provides an efficient way to treat these tumors, as demonstrated in the following examples.

[0077] In further embodiments, the present invention provides a pharmaceutical composition comprising the agent or binding protein of the present invention as described above, or a nucleic acid molecule encoding the agent or binding protein as described above, or engineered CAR cells as described above. In addition, the pharmaceutical composition also comprises at least one pharmaceutically acceptable carrier or excipient. As used herein, “pharmaceutically acceptable carrier or excipient” includes any and all physiologically compatible solvents, dispersion media, coatings, antimicrobial and antifungal agents, isotonic agents, and absorption retarders.

[0078] Preferably, the carrier or excipient is suitable for parenteral administration, such as intradermal, intravenous, intramuscular, or subcutaneous administration (e.g., by injection or infusion). Depending on the route of administration, the binding protein or conjugated binding protein may be coated with a material to protect them from the action of acids and other natural conditions that may inactivate or denature them. Preferred pharmaceutically acceptable carriers include aqueous carriers or diluents. Examples of suitable aqueous carriers that can be used in pharmaceutical compositions, kits, and products include water, buffer water, and physiological saline. Examples of other carriers include ethanol, polyols (such as glycerol, propylene glycol, polyethylene glycol), and suitable mixtures thereof, vegetable oils, and injectable organic esters such as ethyl oleate. Adequate fluidity can be maintained, for example, by the use of coating materials such as lecithin, maintaining the required particle size in the case of dispersions, and by the use of surfactants. In many cases, it is preferable to include isotonic agents, such as sugars, polyhydric alcohols such as mannitol and sorbitol, and sodium chloride.

[0079] Accordingly, this disclosure relates to the agents, binding proteins, cells, nucleic acid molecules, or pharmaceutical compositions described herein for use in therapies such as cancer therapy. The agents, binding proteins, cells, or pharmaceutical compositions thereof may be used for ii) inhibiting perivascular tumor cell migration, ii) inhibiting tumor cell entry, and / or iii) inhibiting tumor cell proliferation, and / or for use in combination with one or more additional therapeutic regimens, the additional therapeutic regimens being selected from i) the use of antitumor agents, i.e., agents known to those skilled in the art to be able to treat cancer, ii) chemotherapy, iii) immunotherapy including tumor-targeted antibodies, iv) irradiation, and v) the use of pathway blockers for treating cancer, the pathway blockers may include tyrosine kinase inhibitors or neutralizing antibodies against growth factors / cytokines or their receptors.

[0080] Accordingly, methods for treating a subject in need of treatment, comprising administering a therapeutically effective amount of a drug, binding protein, cell, or nucleic acid molecule, as described herein, are described herein. Also provided are the use of drugs, binding proteins, cells, nucleic acid molecules, or pharmaceutical compositions for the treatment or prevention of cancer, preferably glioblastoma.

[0081] Drugs, binding proteins, nucleic acid molecules, cells, or pharmaceutical compositions thereof may be administered via one or more routes of administration using one or more of the various methods known in the art. As will be understood by those skilled in the art, the route and / or method of administration will vary depending on the desired outcome. Preferred routes of administration include, for example, by injection or infusion, directly to the site of the tumor, intravenous, intramuscular, intradermal, intraperitoneal, subcutaneous, spinal, or other parenteral administration routes. As used herein, the term “parenteral administration” means a mode of administration other than enteral and topical administration, usually by injection. Alternatively, non-parenteral routes such as topical, epidermal, or mucosal administration routes may be used. Topical administration is preferred, including peritumoral, peritumoral, intratumoral, intralesional, perilesional, intracavitary injection, intravesical administration, and inhalation. However, antigen-binding proteins, conjugates, or manipulated cells may also be administered systemically.

[0082] The preferred dosage of the specific binding protein, conjugated binding protein, or pharmaceutical composition of the present invention can be determined by those skilled in the art. The actual dosage levels of the active ingredient and product in the pharmaceutical composition of the present invention may be varied to obtain an amount of the active ingredient that is effective in achieving a desired therapeutic response for a particular target, i.e., a patient, and is not harmful to the patient. The selected dosage level will depend on a variety of pharmacokinetic factors, including the activity of the particular protein / conjugate used, the route of administration, the time of administration, the rate of protein excretion, the duration of treatment, other drugs, compounds, and / or materials used in combination with the particular composition used, the age, sex, weight, condition, general health status and prior medical history of the patient being treated, and similar factors well known in the pharmaceutical art.

[0083] The dosage may be administered at specific intervals, such as every other week, or at shorter intervals, such as a daily dose, or, if preferable, at higher doses. Appropriate intervals for different types of treatment will be obvious to those skilled in the art.

[0084] The drug regimen may be adjusted to provide the optimal desired response (e.g., therapeutic response). For example, a single bolus may be administered, or doses may be administered over time in several divided doses, or the dose may be proportionally reduced or increased as indicated by the urgency of the treatment situation. For ease of administration and uniformity of dosage, it is particularly advantageous to formulate parenteral compositions into unit dosage forms. As used herein, a unit dosage form refers to a physically distinct unit suitable as a unit dose for the subject being treated, each unit containing a predetermined amount of the active compound calculated to produce the desired therapeutic effect in relation to the required pharmaceutical carrier.

[0085] Drugs, binding proteins, or compositions may be administered in single or multiple doses. Multiple doses may be administered via the same or different routes and at the same or different locations. Alternatively, they may be administered as sustained-release formulations, in which case a lower frequency of administration is required. Dosage and frequency may vary depending on the half-life of the administered species in the patient and the desired duration of treatment. Dosage and frequency may also vary depending on whether the treatment is prophylactic or therapeutic. For prophylactic use, relatively low doses may be administered at relatively infrequent intervals over a long period. For therapeutic use, relatively high doses may be administered, for example, until the patient shows partial or complete improvement of the symptoms of the disease. In an exemplary drug regimen, the conjugated binding protein is administered to the subject in cycles repeated 2 to 10 times, once weekly, once every two weeks, or once every three weeks.

[0086] Accordingly, the present disclosure enables the treatment of tumor invasion and CD93 expression-related disorders, such as cancer, by administering the binding protein of the present invention. Exemplary embodiments of the present disclosure are disclosed in the drawings and specification. However, many variations and modifications can be made to these embodiments without substantially departing from the principles of the present disclosure. Therefore, the present disclosure should be considered illustrative rather than restrictive, and not limited to the specific embodiments discussed above. Accordingly, specific terms are used, but only in a general and descriptive sense, and not for restrictive purposes.

[0087] The description of the exemplary embodiments provided herein is presented for illustrative purposes only. The description is not intended to be comprehensive or to limit the exemplary embodiments to the exact forms disclosed, and modifications and variations may be possible in light of the teachings above or obtained from the practice of various alternatives to the provided embodiments. The examples considered herein are selected and described to illustrate the principles and properties of various exemplary embodiments and their practical applications, enabling those skilled in the art to utilize the exemplary embodiments in various modes and with various modifications to suit a particular intended use. The features of the embodiments described herein may be combined in all possible combinations of methods, products, and systems. It should be understood that the exemplary embodiments presented herein may be practiced in any combination of each other. It should also be noted that the word “includes” does not necessarily exclude the existence of other elements or steps other than those enumerated, and the words “a” or “an” preceding an element do not exclude the existence of multiple such elements. Furthermore, it should be noted that any reference numerals do not limit the scope of the claims, and exemplary embodiments may be realized in the broadest sense of the claims. [Examples]

[0088] In short, a CD93-binding antibody was selected. The antibody production and validation workflow began with phage display selection, followed by cloning and expression of the conjugate. Binding screening was performed by ELISA, followed by sequencing of hits (154 sequences of unique clones). A secondary ELISA was performed to validate binding (clone selection based on their binding to hCD93, the hCD93 lectin domain, and mCD93), followed by an HTRF assay (uniform time-resolved FRET). Affinity screening using SPR was performed on 30 scFv clones and 14 Fab clones (44 clones in total), followed by small-scale protein purification (Kingfisher Flex). Kinetic measurements were performed using SPR, followed by FACS studies of binding to human and mouse endothelial cells for 27 selected clones. Conversion to hIgG1 LALA was performed, followed by in vitro functional assays in human and mouse endothelial cells (15 hIgG were selected). Ex vivo functional assays were performed on human and mouse brain slices, and four selected hIgGs (three cross-reactive with mouse and one binding only to human CD93) were chosen to exhibit the desired functionality. Three mouse-cross-reactive hIgGs were converted within the mouse skeleton, and in vivo functional assays were performed in mice.

[0089] Example 1: CD93 expression in glioma patients The clinical relevance of CD93 in gliomas was assessed by analyzing CD93 expression in patients with low-grade glioma (grade II), high-grade glioma (grade III), and glioblastoma (grade IV) from the publicly available CGGA database.

[0090] method: CD93 RNA expression in human glioma samples was analyzed using the CGGA (Chinese Glioma Genome Atlas) RNASeq glioma dataset (Zhao Z et al. Genomics Proteomics Bioinformatics. 2021 Feb;19(1):1-12.) downloaded from the GlioVis data portal (http: / / gliovis.bioinfo.cnio.es). Glioma samples were graded according to the 2007 WHO classification. Each dot in the graph represents the CD93 mRNA level in a tumor obtained from one individual patient (a total of 651 patients). The p-value was determined using one-way ANOVA followed by Tukey's multiple comparison test. P<0.05 was considered statistically significant.

[0091] result: The graph in Figure 2 shows increased CD93 expression in patients diagnosed with high-grade glioma (grade III) and glioblastoma (grade IV) compared to patients with low-grade glioma (grade II).

[0092] Conclusion: These data indicate that CD93 is highly expressed in glioma patients, and that its expression increases with disease severity. In fact, CD93 mRNA levels are significantly higher in tumor samples from patients with the most aggressive forms of glioma (grade IV or glioblastoma) compared to patients with grade III or grade II glioma.

[0093] Example 2: Tumor growth and survival in CD93-deficient mice The role of CD93 in glioma progression was evaluated in CD93 knockout and wild-type mice orthotopically injected with the mouse glioma model GL261. Tumor growth and survival analyses were performed.

[0094] material and method: mouse: CD93 (CD93 - / -We bred C57BL / 6 mice with overall knockout of ) and wild-type (WT) littermates in-house. C57BL / 6WT mice were purchased from Taconic M&B.

[0095] Tumor cell culture: GL261 mouse glioma cells were cultured in DMEM (Life Technologies) supplemented with 10% FBS (Sigma-Aldrich) at 37°C in a humidified chamber with 5% CO2 / 95% air.

[0096] In vivo tumor model: GL261 cells (20,000 cells in 2 μl of Dulbecco PBS) were mixed with CD93 - / - The tumor was injected orthotopically into the brains of wild-type mice. Tumor growth was assessed at the study endpoint (23 days after tumor injection) by analyzing tumor areas visualized by nuclear staining in at least six tissue sections (80 μm thick sections) throughout the brain. The p-value was determined using an unpaired t-test. P < 0.05 was considered statistically significant.

[0097] In survival studies, mice were sacrificed when their body weight decreased by more than 10% or when they exhibited symptoms. The p-value was determined using the Gehan-Breslow-Wilcoxon test. A p-value of <0.05 was considered statistically significant.

[0098] result: The graph in Figure 3A shows CD93 deletion (CD93 - / - This shows the quantification of the GL261 tumor area in mice and wild-type mice, demonstrating a significant reduction in tumor growth in CD93-deficient mice compared to the wild-type group. In line with this observation, Figure 3B shows the CD93-deficient area compared to wild-type (WT) mice. - / - This shows improved survival in mice.

[0099] Conclusion: These data demonstrate that CD93 plays a crucial role in tumor progression in mouse glioma models. In fact, CD93 deficiency results in reduced tumor growth and improved survival in the GL261 model.

[0100] Example 3: Endothelial activation and T cell mobilization in glioma-carrying CD93-deficient mice, and survival in glioma-carrying CD93-deficient mice after αPD1 blockade immunotherapy. The role of CD93 in endothelial activation and T cell infiltration during glioma progression, as well as in the response to αPD1-blocking immunotherapy, was evaluated in CD93 knockout and wild-type mice orthotopically injected with the mouse glioma model GL261. Survival analysis was performed.

[0101] material and method: mouse: CD93 (CD93 - / - We bred C57BL / 6 mice with overall knockout of ) and wild-type (WT) littermates in-house. C57BL / 6WT mice were purchased from Taconic M&B.

[0102] Tumor cell culture: GL261 mouse glioma cells were cultured in DMEM (Life Technologies) supplemented with 10% FBS (Sigma-Aldrich) in a humidified chamber at 37°C and under 5% CO2 / 95% air.

[0103] In vivo tumor model: GL261 cells (20,000 cells in 2 μl of Dulbecco PBS) were mixed with CD93 - / -The αPD1 antibody was injected orthotopically into the brains of wild-type mice. Mice were treated three times with αPD1 antibody (clone: ​​RMP1-14, catalog number BE0146, BioXCell, USA) (200 μg / dose every three days, starting 10 days after tumor injection). Mice were sacrificed when they lost more than 20% of their body weight or when they showed symptoms. Survival analysis was performed using the Gehan-Breslow-Wilcoxon test. P<0.05 was considered statistically significant.

[0104] Immunofluorescence staining: Tumor tissue was stained with anti-CD31 (2H8, Thermo Fisher Scientific, MA3105), anti-VCAM1 (R&D System, AF643), and anti-CD8 (Abcam, ab237723). Sections were washed and incubated with a specific Alexa Fluor conjugate secondary antibody (Invitrogen). Nuclei were visualized using Hoechst (Life Technologies) and images were acquired using a confocal microscope (Leica SP8).

[0105] result: Figures 4A and 4B show increased VCAM1 expression in tumor blood vessels in CD93-deficient mice. Figures 4C and 4D show increased VCAM1 expression in tumor blood vessels in CD93-deficient mice. - / - This shows increased infiltration of T cells (CD8-positive cells) in the tumor. Figure 4E shows improved survival in CD93-deficient mice after αPD1 blockade immunotherapy compared to wild-type mice.

[0106] Conclusion: These data suggest that CD93 expression is associated with limiting endothelial activation, reducing T cell infiltration in gliomas, and a worse response to immunotherapy. Indeed, CD93 deficiency results in increased endothelial activation and increased abundance of cytotoxic CD8 T cells in GL261 tumors. In addition, survival after αPD1-antibody therapy in GL261 models is improved in CD93-deficient mice.

[0107] Example 4: The role of CD93 in the extracellular matrix of tissues The extracellular matrix plays a crucial role in tumor progression and spread. Here, we analyzed the role of CD93 in regulating the organization of the extracellular matrix protein fibronectin into fibrous structures, both in vitro on cultured human endothelial cells and in vivo using a GL261 glioma model.

[0108] material and method: mouse: CD93 (CD93 - / - We bred C57BL / 6 mice with overall knockout of ) and wild-type (WT) littermates in-house. C57BL / 6WT mice were purchased from Taconic M&B.

[0109] Cell culture: Human dermal hematopoietic endothelial cells (HDBECs) (PromoCell) were cultured in gelatin-coated culture dishes in endothelial cell basal medium (PromoCell, EBM-MV2) containing a complete supplement. GL261 mouse glioma cells were cultured in DMEM (Life Technologies) supplemented with 10% FBS (Sigma-Aldrich). Both cell types were maintained in a humidified chamber at 37°C and 5% CO2 / 95% air.

[0110] In vivo tumor model: GL261 cells (20,000 cells in 2 μl of Dulbecco PBS) were mixed with CD93 - / - The drug was also injected orthotopically into the brains of wild-type mice. Tumor tissue was collected 23 days after tumor transplantation and vibratome-sectioned for further analysis.

[0111] siRNA transfection: HDBEC cells were incubated for 4–6 hours with a 2nM scrambled control siRNA or siRNA specifically targeting CD93 (Hs_CD93_1; FlexiTube, Qiagen) in a mixture of 20% Opti-MEM (Life Technologies) in endothelial cell medium supplemented with 30 μl / ml Lipofectamine RNAiMAX (Life Technologies), followed by replacement of the medium with fresh medium. Experiments were performed two days after siRNA transfection.

[0112] Immunofluorescence staining: siRNA-transfected HDBEC growths in 8-well chamber slides were fixed with 4% PFA and incubated with anti-fibronectin (Abcam, ab2413). Cells were washed and incubated with Alexa Fluor-conjugated secondary antibody (Invitrogen). The nucleus and cytoskeleton were visualized with phalloidin-labeled Hoechst and Alexa Fluor-647, respectively (all from Life Technologies). Immunofluorescence staining of tumor tissue was performed on vibratome sections of mouse GL261 tumors. Tumor tissue was stained with anti-CD31 (2H8, Thermo Fisher Scientific, MA3105) and anti-fibronectin (Abcam, ab2413). Sections were washed and incubated with specific Alexa Fluor-conjugated secondary antibody (Invitrogen). Nuclei were visualized with Hoechst (Life Technologies). Cells and tumor tissue were analyzed under a confocal microscope (Leica SP8).

[0113] result: Downregulation of CD93 in human endothelial cells inhibited fibronectin fibril formation in vitro. As shown in Figure 5A, control endothelial cells (siCtrl) formed a dense fibronectin matrix organized into a fibril network. However, downregulation of CD93 (siCD93) was associated with the disruption of fibronectin filaments compared to control conditions.

[0114] Similarly, immunohistochemical staining of GL261 tumors compared to tumors from wild-type mice showed CD93 - / - We revealed a significant reduction in fibronectin deposition in mice (quantification in Figures 5B and 5C). In addition, as shown by the high-magnification image in Figure 5D, fibronectin deposition in wild-type tumors is mainly associated with the vascular system (CD31 positive), suggesting that the fibronectin matrix is ​​primarily deposited and organized in a fibril structure by endothelial cells in GL261 tumors.

[0115] Conclusion: Here, we demonstrated that CD93 plays a crucial role in the deposition and organization of the extracellular matrix protein fibronectin in both in vitro human endothelial cells and in vivo GL261 vascular systems.

[0116] In Example 5, CD93 is expressed in blood vessels hijacked by cancer cells in the pre-tumor invasion area. Malignant gliomas are highly invasive cancers. Migration along blood vessels is one of the strategies used by glioma cells to invade the brain parenchyma.

[0117] We previously reported that CD93 is highly expressed in glioma blood vessels within the tumor core (Langenkamp E. et al Cancer Res. 2015 Nov 1;75(21):4504-16), and that CD93 regulates the deposition and organization of endothelial-associated fibronectin, which is important for tumor cell invasion (Lugano R. et al J Clin Invest. 2018 Aug 1;128(8):3280-3297).

[0118] Here, we analyzed CD93 expression in blood vessels localized to the anterior region of tumor invasion in both in vivo and ex vivo models of mouse brain slices.

[0119] material and method: mouse: I purchased C57BL / 6 wild-type mice from Taconic M&B.

[0120] Tumor cell culture: GL261-GFP mouse glioma cells were cultured in DMEM (Life Technologies) supplemented with 10% FBS (Sigma-Aldrich) in a humidified chamber at 37°C and under 5% CO2 / 95% air.

[0121] In vivo tumor model: GL261 cells expressing GFP (20,000 cells in 2 μl of Dulbecco PBS) were orthotopically injected into the brains of wild-type mice. Tumor tissue was collected 23 days after tumor transplantation and vibratome-sectioned for further immunofluorescence staining.

[0122] Brain slice model: Brains from wild-type mice were newly dissected, embedded in low-melting-point agarose (Invitrogen), and vibratome-sectioned into 300 μm thick slices in HBSS complete buffer (1 × HBSS, 2.5 mM HEPES, 30 mM D-glucose, 1 mM CaCl2, 1 mM MgSO4, 3 mM NaHCO3, all from Sigma-Aldrich). The slices were then cultured on 8 μm membrane 12-well culture inserts (vWR; 7342736P) in DMEM medium supplemented with L-glutamine, 5% FBS, and 25% HBSS complete buffer in the presence of antibiotics. GL261-GFP cells (10,000 cells resuspended in 2 μl of DMEM medium) were injected into the brain slices using a 10 μl Hamilton syringe and co-cultured for 48 hours at 37°C and 5% CO2 / 95% air in a humidified chamber. After this period, brain slices were fixed with 4% PFA at room temperature for 1.5 hours and stained as described below.

[0123] Immunofluorescence staining: Tumor tissue and brain slices were stained with anti-CD31 (2H8, Thermo Fisher Scientific, MA3105) and anti-CD93 (R&D System, AF1696). Sections were washed and incubated with a specific Alexa Fluor conjugate secondary antibody (Invitrogen). Nuclei were visualized by Hoechst (Life Technologies), and the GL261-GFP signal was detected at a wavelength of 488 nm by confocal microscopy (Leica SP8).

[0124] result: To investigate the role of CD93 in tumor invasion, we analyzed CD93 expression in blood vessels hijacked by tumor cells in the tumor invasion area using immunofluorescence staining. Interestingly, we found that CD93 was highly expressed in tumor boundary blood vessels in GL261 tumor-bearing mice. As shown in Figure 6A (highlighted with arrowheads), a strong CD93 signal was detected in blood vessels (CD31-positive) hijacked by tumor cells (GL261) at the tumor margin, indicated by the dotted line. Similarly, immunofluorescence staining in mouse brain slices co-cultured with GL261 cells showed high CD93 expression in tumor cell-associated blood vessels in the anterior tumor invasion area (arrowheads in Figure 6B).

[0125] Conclusion: These data indicate that CD93 expression is enhanced not only in the tumor core vessels (Langenkamp E. et al. Cancer Res. 2015 Nov. 1; 75(21):4504-16), as previously shown, but also in vessels hijacked by tumor cells localized at the tumor invasion boundary.

[0126] Example 6: Glioma cells migrate along fibronectin fibers. Fibronectin is a major component of the extracellular matrix and is essential for tumor progression and tumor cell invasion. Here, we evaluated whether GL261 cells migrate along endothelial-associated fibronectin filaments in in vitro, ex vivo, and in vivo models.

[0127] material and method: Tumor cell culture: GL261-GFP cells were cultured as described in Example 5.

[0128] GL261 Spheroid: For spheroid formation, GL261-GFP cells were enzymatically isolated, counted, and placed in 24-well low-adhesion plates (Sarstedt) at a density of 80,000 cells per well. The cells were cultured in serum-free DMEM in the presence of B-27 supplement (Gibco) and growth factors (20 ng / ml FGF and 20 ng / ml EGF, all from Peprotech). After 24 hours of culture, spheroids of similar size were seeded onto a monolayer formed by wild-type mouse brain endothelial cells and co-cultured for 48 hours in a humidified chamber at 37°C and 5% CO2 / 95% air.

[0129] Isolation of mouse brain endothelial cells: Endothelial cells were newly isolated from 12-week-old wild-type mice as previously described (Lugano R. et al J Clin Invest. 2018 Aug 1;128(8):3280-3297) and cultured on 8-well chamber slides until confluence. GL261-GFP spheroids were seeded onto a monolayer of endothelial cells and co-cultured for 24 hours in a humidified chamber at 37°C and 5% CO2 / 95% air.

[0130] Brain slice model: Mouse brain slice cultures and GL261-GFP co-cultures were performed as described in Example 5.

[0131] In vivo tumor model: Tumor tissue was obtained from wild-type mice carrying GL261-GFP tumors, as described in Example 5.

[0132] Immunofluorescence staining: Immunofluorescence staining for fibronectin in cultured endothelial cells, brain slices, and tumor tissue was performed as described in Example 5.

[0133] result: To investigate whether GL261 glioma cells interact with the fibronectin matrix produced by endothelial cells, GL261-GFP spheroids were first co-cultured on a monolayer of mouse brain endothelial cells isolated from wild-type mice. As shown in Figure 7A, endothelial cells that grew to confluence produced a thick fibronectin matrix organized as fibril structures (see fibronectin fiber image in Figure 7A). Interestingly, when GL261 spheroids were seeded onto endothelial cells, they migrated out of the spheroids and invaded the endothelial monolayer along the fibronectin fibers produced by the endothelial cells (arrowheads in Figure 7A). In line with this observation, GL261 cells co-cultured on mouse brain slices derived from wild-type mice interacted with and migrated along CD31-positive blood vessels expressing fibronectin (arrowheads in Figure 7B). Similarly, GL261 tumor cells that grow sympatrically in the brain of wild-type mice invade the brain parenchyma by migrating along CD31-positive blood vessels that express fibronectin (arrowheads in Figure 7C).

[0134] Conclusion: These data indicate that GL261 glioma cells migrate along the endothelium-associated fibronectin matrix.

[0135] Example 7: CD93 deficiency is associated with reduced tumor cell invasion. The effect of tumor vascular CD93 expression on glioma cell invasion was evaluated using a CD93 knockout mouse model. CD93-dependent tumor invasion was analyzed through in vitro, ex vivo, and in vivo tumor models.

[0136] material and method: Tumor cell culture: GL261-GFP cells were cultured as described in Example 5.

[0137] Isolation of mouse brain endothelial cells: Mouse brain endothelial cells were subjected to wild-type and CD93 as described in Example 6. - / - It was isolated from mice.

[0138] Brain slice model: Wild type and CD93 - / - Mouse brain slices derived from mice were obtained as described in Example 5.

[0139] GL261 Spheroid: GL261-GFP spheroids were obtained as described in Example 6. GL261 spheroids of similar size were obtained in wild-type and CD93. - / - The cells were seeded into brain slices and co-cultured for 48 hours.

[0140] In vivo tumor model: The tumor tissue, as described in Example 5, consists of wild-type and CD93 grafts containing GL261 tumors. - / - The tumor tissue was obtained from mice. GL261 tumor sections were counter-stained with hematoxylin and eosin (H&E) to visualize the tumor boundaries and quantify the tumor invasion area.

[0141] Imaging analysis: Image analysis was performed using ImageJ software. GL261 cell invasion into the mouse brain endothelial monolayer was quantified by measuring the area covered by tumor cells, normalized by the area occupied by the endothelial cell monolayer.

[0142] The ability of GL261 cells to invade mouse brain slices was determined by quantifying the number of vascular-associated GL261 sprouts in each spheroid and the invasion distance calculated from the spheroid body.

[0143] In vivo GL261 invasion was evaluated by quantifying the region of brain parenchyma invaded by tumor cells adjacent to the tumor bulk, which were identified by staining as densely packed cellular areas. The tumor invasion region was normalized by the area surrounding the tumor bulk.

[0144] result: To evaluate the role of CD93 in glioma invasion, first, wild-type or CD93 - / - The ability of GL261 cells to migrate on a monolayer of mouse brain endothelial cells isolated from mice was analyzed. As shown in the quantification graphs in Figures 8A and 8B, CD93 - / - GL261 cells co-cultured with endothelial cells show a significant reduction in invasiveness compared to GL261 cells co-cultured with endothelial cells derived from wild-type mice. Similar results were obtained with CD93. - / - This was obtained when GL261 spheroids were cultured on mouse brain slices obtained from mice. In fact, as shown in Figure 8C and the quantification graphs in Figures 8D and 8E, tumor cell invasion along blood vessels was compared with tumor cell migration observed in brain slices derived from wild-type mice. - / - It was significantly reduced in brain slices.

[0145] This observation was made between the wild type and CD93. - / - This was further confirmed in vivo by analyzing the tumor invasion area of ​​GL261 tumors grown in mice. Interestingly, tumors in wild-type mice showed irregular invasive tumor boundaries, while CD93 - / - The mouse tumors were characterized by linear tumor margins with a significantly reduced tumor invasion area (quantification graphs in Figure 8F and Figure 8G).

[0146] Conclusion: These results highlight the crucial role of CD93 in promoting perivascular invasion of glioma cells. Indeed, CD93 deficiency significantly reduces GL261 cell invasion along blood vessels in the brain parenchyma. This observation suggests CD93 as a potential therapeutic target for inhibiting glioma cell diffusion.

[0147] Example 8: CD93 deficiency reduces tumor cell proliferation. The proliferation of human glioma cell line U87 and mouse GL261 glioma cell line was evaluated through in vitro and ex vivo co-culture experiments in the presence or absence of CD93.

[0148] material and method: Cell culture: Human dermal blood endothelial cells (HDBECs) and GL261 cells were cultured as described in Example 4.

[0149] U87 human glioma cells were cultured in MEM (Gibco) supplemented with 10% FBS.

[0150] siRNA transfection: CD93 downward adjustment in HDBEC was performed as described in Example 4.

[0151] Growth assay: An in vitro proliferation assay was performed using U87 cells. Briefly, U87 cells were placed in a 24-well plate at a density of 30,000 cells per well. After adhesion, the cells were treated with conditional media derived from HDBEC transfected with either control (Mock and siCtrl) or CD93 siRNA (siCD93_1 and siCD93_5). The conditional media (EBM medium supplemented with 1% FBS) was collected after 24 hours of incubation, centrifuged at 1200 rpm to remove cell debris, and transferred to the U87 cells. Tumor cells were incubated in the conditional media for 4 days. Cell proliferation was assessed by manual cell counting on days 2, 3, and 4 after seeding.

[0152] Mouse brain slices and GL261 spheroids: Wild type and CD93 - / - Mouse brain slices were obtained as described in Example 5. Co-culturing with GL261 spheroids was carried out as described in Example 6.

[0153] Immunofluorescence staining: Brain slices were stained with anti-Ki67 (Abcam, ab16667), washed, and incubated with a specific Alexa Fluor-568 conjugated secondary antibody (Invitrogen). Nuclei were visualized by Hoechst (Life Technologies). Ki67 expression (Leica SP8) was analyzed under confocal microscopy.

[0154] result: Membrane-bound CD93 can be cleaved by metalloproteinases and released as a soluble form. To investigate whether soluble CD93 released from endothelial cells regulates tumor cell proliferation, human glioma U87 cells were incubated in a condition medium derived from human endothelial cell controls containing soluble CD93, and in a condition medium derived from CD93 siRNA-silencing endothelial cells containing significantly reduced amounts of soluble CD93 (Figure 9A). Surprisingly, analysis of U83 proliferation monitored for 4 days revealed that U87 cells incubated in condition medium derived from CD93-silencing endothelial cells (siCD93_1 and siCD93_5) proliferated significantly less than U87 treated in condition medium derived from control endothelial cells (Mock and siCtrl). In line with this, CD93 - / - Immunofluorescence staining with the proliferation marker Ki67 in GL261 spheroids cultured on brain slices showed a significant reduction in Ki67-positive cells compared to Ki67 signaling detected in spheroids co-cultured with wild-type brain slices (Figures 9B and 9C).

[0155] Conclusion: These data demonstrate that CD93 deficiency inhibits tumor cell proliferation in both in vitro and ex vivo brain slice models.

[0156] Example 9: Phage display selection for human and / or mouse CD93 using scFv and Fab antibody libraries Phage display selection was performed to enable the isolation of scFv and Fab fragments with specificity for human and / or mouse CD93.

[0157] material and method Phage display selection Biopanning was performed using four selective enrichment rounds with two in-house constructed human synthetic scFv phage libraries, SciLifeLib1 and Fab phage library, SciLifeLib3 (SciLifeLab, Stockholm, Sweden). SciLifeLib1 is a naive human synthetic scFv library, similar in design and construction to a previously reported one (Saell, et al., Protein Eng Des Sel (2016) 29:427-437). Briefly, human germline genes IGHV3-23 and IGKV1-39 were used as the library backbone, and Kunkel mutagenesis was used to introduce diversity into four of the six complementarity-determining regions (CDRs), namely CDR-H1, CDR-H2, CDR-H3, and CDR-L3. SciLifeLib3 was constructed in a Fab format using a similar method. Selection was performed using streptavidin-coated magnetic beads (Dynabeads M-280, ThermoFisher Scientific, #11206D) and biotinylated antigens. Three Avi-tagged antigen constructs were used: human or mouse CD93 extracellular domains (amino acids 1-580) designated as hCD93-avi and mCD93-avi, respectively, and an avi-tagged human CD93 lectin domain (amino acids 1-258) designated as hCD93lectin-avi. The Avi tag enables site-specific biotin incorporation. For each of the two phage libraries, a sequential selection track was included using either hCD93-avi, mCD93-avi, or hCD93lectin-avi as bait, respectively. In additional tracks, the antigen was alternated between hCD93-avi and mCD93-avi between different rounds to preferentially select interspecies-reactive scFv and Fab clones. The selective pressure was increased by gradually decreasing the antigen amount (Round 1: 200 nM, Round 2: 50 nM, Round 3: 10 nM, Round 4: 2 nM) and by increasing the number and intensity of washes between different rounds.To remove nonspecific or streptavidin-binding agents, preliminary selection was performed before rounds 1 and 2 by incubating phage stocks against empty streptavidin-coated magnetic beads. Additionally, 1% bovine serum albumin (BSA) was included throughout the selection procedure as a blocker. Elution of antigen-bound phages was performed using the trypsin aprotinin method. The entire selection process, excluding the phage target protein incubation step, was automated and performed using a Kingfisher Flex robot. The recovered phages were grown in Top10F'E.coli overnight on agar plates at 37°C (rounds 1 and 2) or overnight in solution at 30°C (rounds 3 and 4). Phage stocks were prepared by infection with excess M13K07 helper phage (New England Biolabs, #N0315S) and scFv expression induced by the addition of IPTG. The cultures were precipitated overnight in PEG / NaCl, resuspended in selection buffer, and used for subsequent selection rounds.

[0158] Recloning and expression of scFv and Fab To enable the production of soluble scFv and Fab, phagemide DNA was isolated from the third and fourth rounds of each selected track. In the pool, the genes encoding the scFv fragment were digested with restriction enzymes and subcloned into screening vectors, providing a signal for scFv secretion along with a triple-flag tag and a hexahistidine (His) tag at the C-terminus. Similarly, the genes encoding the Fab fragment were digested with restriction enzymes and subcloned into vectors containing a signal for Fab secretion along with a C-terminal hexahistidine (His) tag. The constructs were then transformed into TOP10 E. coli. Single colonies were picked, cultured, and IPTG induced for soluble scFv expression in a 96-well format. In total, 752 clones, 376 scFv and 376 Fab present in the bacterial supernatant were prepared for primary ELISA screening.

[0159] ELISA screening The antigens hm CD93, ms CD93, and hm lectin domain were coated onto 384-well ELISA plates at 1 μg / ml via streptavidin. scFv and Fab clones (PBS supplemented with 0.5% BSA + 0.05% Tween 20) present in the bacterial supernatant diluted 1:5 in block buffer were added and conjugated to the coated proteins. Binding detection was enabled via HRP-conjugated αFlag M2 antibody (Sigma-Aldrich #A8592), followed by incubation with TMB ELISA substrate (ThermoFisher Scientific #34029). Colorimetric signal generation was stopped by adding 1 M sulfuric acid, and the plates were read at 450 nm. All samples were assayed in duplicate, and the average Abs450 nm was calculated from these values. The average Abs450 nm of the blank wells (with block buffer added instead of scFv or Fab clones) was then subtracted.

[0160] DNA sequencing 537 positive scFv or Fab clones showing binding to CD93 were sent to GATC Biotech (Ebersberg, Germany) for Sanger DNA sequencing.

[0161] result Using the phage libraries SciLifeLib1_scFv and SciLifeLib3_Fab, a total of eight selection tracks were run in parallel on human CD93, human lectin domains, and mouse CD93. After recloning the scFv and Fab clones into soluble expression vectors, 94 clones (colonies) were picked per track from selection rounds 3 and 4, for a total of 752 clones that were picked and cultured.

[0162] ELISA screening identified 537 potential hits with four distinct binding properties, each containing a clone that binds to either 1) hCD93-avi, 2) hCD93-avi and hCD93lectin-avi, 3) hCD93-avi and mCD93-avi, or 4) mCD93-avi. The majority of the hits showed binding to hCD93-avi and hCD93lectin-avi.

[0163] DNA sequencing of 537 clone hits identified 154 sequence-specific clones, 117 scFvs, and 37 Fab clones.

[0164] conclusion scFv and Fab clones that bind to human and / or mouse CD93 were successfully isolated by phage display selection. Following initial ELISA screening of 752 clones and DNA sequencing of 537 positive ELISA hits, a total of 154 sequence-specific clones (117 scFv clones and 37 Fab clones) were identified.

[0165] Example 10: Evaluate the binding characteristics of 154 sequence-specific clones. 154 sequence-specific scFv and Fab clones from Example 9 were selected for further characterization, enabling ranking of different clones in a kinetic screening-based method, by secondary ELISA (enzyme-linked immunosorbent assay), HTRF assay (uniform time-resolved fluorescence), and surface plasmon resonance (SPR). The most promising clones were tested by flow cytometry (FACS) to assess their binding to endothelial cells.

[0166] material and method ELISA hCD93, mCD93, hCD93lectin, and unrelated proteins (negative control) were coated in 384-well ELISA plates at 1 μg / ml via streptavidin. Hm CD93-Fc (RnD Systems #2379-CD) was directly coated in PBS at 1 μg / ml overnight at 4°C. scFv and Fab clones present in the bacterial supernatant were diluted 1:10 in block buffer (PBS supplemented with 0.5% BSA + 0.05% Tween 20) and conjugated to the coated proteins. Binding detection was enabled via HRP-conjugated α-FLAG M2 antibody (Sigma-Aldrich #A8592) for scFv clones or HRP-conjugated α-hm KAPPA antibody (Southern Biotech #9230) for Fab clones, followed by incubation with TMB ELISA substrate (ThermoFisher Scientific #34029). The colorimetric signal generation was stopped by adding 1M sulfuric acid, and the plate was analyzed at Abs450nm.

[0167] All samples were assayed in duplicate, and the average Abs450nm was calculated from these results. The mead abs450nm of the blank wells (with block buffer added instead of scFv or Fab clones) was then subtracted.

[0168] HTRF Along with unrelated scFv and unrelated Fab (negative control), scFv and Fab clones were diluted 1:5 in assay buffer (PBS supplemented with 0.1% BSA) and conjugated to hCD93, mCD93, unrelated proteins, and unrelated peptides, respectively, diluted to 200 nM in assay buffer. Binding detection was enabled through the donor molecule terbium-conjugated α-FLAG antibody (Cisbio#611FG2TL) for scFv clones or europium-conjugated α-KAPPA antibody (Cisbio#61KAPKAA) for Fab clones, in combination with the acceptor molecule streptavidin-conjugated XL665 (Cisbio#610SAXL). After incubation of plates in the dark at room temperature for 2 hours, analysis was performed using an Envision spectrometer (Perkin Elmer) at 615 nm (background / noise signal) and 665 nm (binding signal). The delta-R value for each sample was obtained by dividing the 665nm value by the 615nm value and multiplying by 10,000.

[0169] All samples were assayed twice, the mean values ​​were taken, and the mean values ​​of the blank wells (where assay buffer was added instead of clones) were subtracted.

[0170] kinetic screening Kinetic screening was performed using a BIAcore T200 instrument (Cytiva). αFLAG M2 antibody (Merck#F1804) or α-hm KAPPA antibody (Cytiva#28958325), each acting as a capture ligand for scFv or Fab clones, was immobilized on all four surfaces of CM5 series S sensor tips using EDC / NHS amine coupling chemistry, as recommended by the manufacturer. All experiments were performed at 25°C in electrophoresis buffer (HBS supplemented with 0.05% Tween 20, pH 7.5).

[0171] scFv and Fab clones present in the bacterial supernatant were injected, respectively, and captured on the chip surface via antibody capture ligands. Subsequently, single antigen injections of 50 nM hCD93, mCD93, and hCD93lectin were performed. After the dissociation phase, the surface was regenerated with 10 mM glycine-HCl at pH 2.1.

[0172] For each clone, the obtained sensorgram was subtracted from the reference surface and from a blank run in which the same clone was captured and injected with electrophoresis buffer as the antigen. The double-referenced subtracted data could be fitted to a 1:1 Langmuir binding model using BIAeval software (Cytiva) to extract the apparent kinetic constants.

[0173] Flow cytometry: Human endothelial cells (HDBEC) were cultured as described in Example 4. U87 cells were cultured as described in Example 8. Mouse endothelial cells (MS1) were cultured in DMEM (Gibco) supplemented with 10% FBS. GL261 cells were cultured as described in Example 5.

[0174] Growing cells in cell culture plates were isolated using an acutase dissociation buffer (ThermoFisher Scientific). Cells were incubated at 4°C for 20 minutes with different clones; specifically, HDBEC and negative control U87 cells were incubated with 150 ng / ml of human-specific and human-mouse cross-reactive clones, and MS1 and negative control GL261 cells were incubated with 300 ng / ml of mouse-specific and human-mouse cross-reactive clones. Clone concentrations were determined based on CD93 expression in the endothelial cell cultures used in the assay. After this incubation period, cells were washed and stained with the secondary antibody anti-human kappa PE (SouthernBiotech #9230-09) to detect scFv and Fab clones. Samples were run on a CytoFLEX LX (Beckman Coulter), and data were analyzed using FlowJo version 10.5.3 (FlowJo LLC). The percentage of positive cells for each clone was evaluated.

[0175] result The binding of 154 scFv and Fab clones to hCD93, mCD93, and hCD93lectin was measured by ELISA, HTRF, SPR kinetic screening, and cell binding (FACS). The results for the most promising clones are summarized in Table 1 above. The results of the methods correlated well. Based on binding specificity, scFv clones could be divided into four main groups: 1) clones that bind to hCD93, 2) clones that bind to both hCD93 and hCD93lectin, 3) clones that bind to both hCD93 and mCD93, and 4) clones that bind to mCD93. Based on binding specificity, Fab clones could be divided into three main groups: 1) clones that bind to both hCD93 and hCD93lectin, 2) clones that bind to both hCD93 and mCD93, and 3) clones that bind to mCD93. No significant background binding to unrelated proteins or streptavidin was detected for any of the clones. The most promising clones exhibit apparent affinity in the sub- to low nanomolar range for their antigens or multiple antigens, K D app This showed (because only the concentration of a single antigen was analyzed, the apparent affinity of the abbreviation, K D app (These are used). Subsequently, the most promising clones were tested for their specific binding to CD93 expressed by endothelial cells. For this purpose, human and / or mouse endothelial cells were exposed to each clone, and binding was evaluated by flow cytometry. U87 and GL261 cells were used as negative controls because they do not express CD93. As shown in Table 1, all tested clones bound to endothelial cells with high affinity and specificity. In fact, the percentage of positive cells for each clone varied from 74.6% to 99.8%, while the percentage of positive cells in the negative control was less than 4.7%.

[0176] Based on the results shown in Table 1, 26 clones (16 scFv and 10 Fab clones) were selected for conversion to full-length human IgG1 LALA.

[0177] conclusion Secondary ELISA, HTRF assay, and kinetic screening by SPR were performed on 154 sequence-specific scFv and Fab clones. Specific binding of the most promising clones was tested on endothelial cells by flow cytometry, demonstrating high affinity and specificity. The combined data facilitated further ranking of clones and enabled the selection of 27 most promising clones (17 scFv clones and 10 Fab clones) for small-scale protein purification and further analysis.

[0178] Example 11: Conversion of 26 clones to human IgG1 LALA and further evaluation of binding to CD93. Twenty-six of the most promising scFv and Fab clones from Example 9 were successfully converted to human IgG1 LALA. The selection of these clones was based on their performance in biophysical, biochemical, and cell-binding assays, as well as their reactivity across different antigen constructs.

[0179] material and method cloning The PCR-amplified VH and VL regions of 26 scFv and Fab clones to be converted were inserted into the in-house constructed vector pHAT-hIgG1-LALA using the InFusion HD Plus cloning kit (Clontech #638909). In addition, scFv HuF11 (from patent WO2020 / 180706) and two isotype controls were also converted to hIgG1 LALA.

[0180] The cloning reaction transformed E. coli Stellar cells, followed by colony PCR screening. Two clones per construct were then sent for DNA sequencing, and the correct DNA sequence of each antibody clone was verified.

[0181] Expression and Purification Plasmid was isolated for transient transfection. Expi293F® cells (Thermo Scientific A14527) were cultured at 37°C, 105 rpm, 70% rH, and 7% CO2 in Expi293® expression medium (Thermo Fisher A1435101). For transfection, cells were placed in medium in approximately 2.5 x 10⁻⁶ units. 6 The DNA was diluted to cells / ml. Sterile filtered DNA was mixed with Opti-MEM (Thermo Scientific 31985062) and FectoPRO DNA transfection reagent (Polyplus 116-010), incubated for 10–60 minutes, and then added to cells. FectoPRO Booster (Polyplus 116-010) was added to 125 ml of culture and expressed for 5 days. At harvest, the culture was centrifuged at 3000 rcf. The supernatant was filtered using a Steriflip-GP 0.22 μm polyethersulfone gamma irradiation filter unit (Sigma SCGP00525) and stored at 4°C until purified.

[0182] Purification was performed using an AKTA Pure25 Protein Purification system (Cytiva) in a refrigerated cabinet. Samples were packed using an autosampler (Teledyne Cetac ASX-280). A Fibro HiTrap PrismA column (Cytiva 17549855) was equilibrated with Dulbecco phosphate-buffered saline (PBS; Sigma D8537-500ML), packed with samples, and washed with 10 column volumes (CV) of PBS. The bound proteins were eluted with 6 CV of 0.1M glycine-HCl buffer (Polysciences Inc 24074-500). The eluted antibodies were collected in a 2 ml loop using the peak acquisition function of Unicorn software. The antibodies were then packed from the sample loop onto a HiTrap desalted column (Cytiva 17140801) equilibrated with PBS, buffer-exchanged proteins were collected, and the column was re-equalised for the next sample. Antibodies were randomly selected for endotoxin testing (Endosafe Portable Endotoxin System, Charles River). All purified antibodies were analyzed by SDS-PAGE and size exclusion chromatography (SEC) using HLPC (Agilent 1200 system) and Agilent Bio-3 columns.

[0183] ELISA HCD93, mCD93, hCD93lectin, and isotype control antigens were coated into 384-well ELISA plates at 1 μg / ml via streptavidin. Purified antibodies were diluted in 1 μg / ml block buffer (PBS supplemented with 0.5% BSA + 0.05% Tween20) and conjugated to the coated proteins. α-hm CD93 ab (BD Bioscience #552954) and α-ms CD93 ab (RnD Systems #MAB1696) were also included as positive assay controls. Binding detection was enabled via positive control antibodies: HRP-conjugated α-hm KAPPA antibody (Southern Biotech #9230), HRP-α-mouse KAPPA chain ab (Southern Biotech #1050), and HRP-α-rat IgG ab (Thermo Scientific #31470), followed by incubation with TMB ELISA substrate (Thermo Scientific #34029). Colorimetric signal generation was stopped by adding 1 M sulfuric acid, and the plates were analyzed at Abs 450 nm.

[0184] SPR Kinetic screening was performed using a capture-based method with a BIAcore T200 instrument (Cytiva). All experiments were conducted at 25°C in electrophoresis buffer (HBS supplemented with 0.05% Tween20, pH 7.5).

[0185] An α-hm KAPPA antibody (Cytiva #28958325), functioning as a capture ligand, was immobilized on all four surfaces of CM5 series S sensor tips using EDC / NHS amine coupling chemistry, as recommended by the manufacturer. Purified hIgG1 LALA clones diluted in electrophoresis buffer were injected into each surface and captured via the antibody capture ligand, followed by single antigen injection of 50 nM hCD93, mCD93, and hCD93lectin, respectively. The tip surfaces were regenerated with 10 mM glycine-HCl at pH 2.1.

[0186] For each clone, the obtained sensorgram was subtracted from the reference surface and from a blank run in which the same clone was captured and injected with electrophoresis buffer as the antigen. The double-referenced subtracted data could be fitted to a 1:1 Langmuir binding model using BIAeval software (Cytiva) to extract the apparent kinetic constants.

[0187] BVP ELISA Baculovirus particles (BVP) were precipitated from the supernatant of baculovirus-infected Sf9 cells using a standard PEG / NaCl precipitation protocol. Briefly, 4% PEG8000 and 0.5M NaCl were added to the supernatant, incubated on ice for 30 minutes, and then the precipitated BVP was pelleted by centrifugation. The pellet was resuspended in PBS and centrifuged again to remove cell debris.

[0188] Baculovirus particles, diluted 1:20 in PBS, were directly immobilized overnight in a 384-well microtiter at 4°C. αCD93 antibody and reference IgG (freticumab, panitumab, trastuzumab, bevacizumab, durigotuzumab, and randilumab; absolute antibodies) were diluted to final concentrations of 100 nM and 20 nM in blocking buffer (PBS + 0.5% BSA) and added to the wells. Detection of bound IgG was performed using horseradish peroxidase (HRP) conjugated anti-human IgG kappa antibody. After incubation with the secondary detection antibody, signal generation was initiated using Chromogen Ultra TMB-ELISA (Thermo Scientific), and the reaction was stopped by adding 1 M sulfuric acid. Absorbance was measured at 450 nm.

[0189] Off-target binding by ELISA A panel of antigens (cardiolipin from bovine heart, Sigma Aldrich; keyhole limpet hemocyanin, Sigma Aldrich; LPS from E. coli O111:B4, InvivoGen; ssDNA from calf thymus, Sigma Aldrich; dsDNA from calf thymus, Sigma Aldrich; human insulin, Sigma Aldrich) were immobilized in PBS at different concentrations on 384-well microtiter plates by direct immobilization overnight at 4°C. Reference and corresponding antigens for the AD169 antibody (IL20, Sino Biological (untagged); biotinylated avi-tagged human Her2, Acro Biosystems; biotinylated avi-tagged human VEGF165, Acro Biosystems; biotinylated human ErbB3, Acro Biosystems; biotinylated human GM-CSF, Abcam; biotinylated avi-tagged SARS CoV-2 S1 RBD, in-house produced; biotinylated avi-tagged hCD93, in-house produced) were immobilized on microtiter plates by direct fixation overnight at 4°C (for all non-biotinylated antigens) or via immobilized streptavidin at 1-2 μg / ml in PBS (for all biotinylated antigens). IgG was added and the following steps were performed as described for the BVP ELISA.

[0190] result Cloning, expression, and purification The genes encoding VH and VL from 26 clones, reference HuF11, and two isotype controls were successfully transferred to vectors encoding the human IgG1 LALA subclass, as confirmed by DNA sequencing.

[0191] Following clonal production in Expi293F cells, protein A purification, and buffer exchange to PBS, clone 39 was removed from the set due to its low yield.

[0192] ELISA The binding patterns of the clones correlated with previously performed ELISAs of clones purified in scFv and Fab formats, although clones AD169-33, -85, -108, and -334 showed some binding to additional CD93 antigens, albeit with lower binding signals than expected for their CD93 antigens (Table 2).

[0193] Positive control antibodies were found to bind only to their expected antigens. Isotype controls showed binding only to their antigens, the SARS-CoV-2 S1 protein. However, the isotype control MO176-301 demonstrated increased background binding to mCD93.

[0194] SPR The converted hIgG1 LALA clones were subjected to kinetic screening, and their binding to hCD93, hCD93lectin, and mCD93 was evaluated using a single antigen concentration of 50 nM, respectively. SPR binding data correlated well with ELISA data, with the exception of clone AD169-33, which showed binding only to hmCD93 and not to mCD93, as in ELISA.

[0195] The most promising clones have apparent affinity in the sub- to low nanomolar range, K D app This was shown (Table 2). Because only the concentration of a single antigen was analyzed, the apparent affinity of the abbreviation, K D app This is used here. The SPR sensorgrams of the four most promising antibodies are shown in Figure 10.

[0196] Off-target coupling evaluation hIgG1 LALA clones were assayed for nonspecific binding by measuring their binding to BVP and a panel of various biomolecules. Low signals were observed in the BVP binding assay for most αCD93 antibodies (Table 2). However, clones 41 and 334 showed moderate levels of binding to BVP. None of the clones were classified as high in terms of BVP binding. The binding patterns of the reference antibodies were as expected—rengirumab and durigotuzumab gave moderate to high signals, while others remained low (data not shown).

[0197] In ELISA measuring binding to off-target panels, most αCD93 antibodies yielded low-intensity signals (Table 2). Clone 307 gave a moderate signal, and high binding to various off-target antigens was recorded for clone 334.

[0198] conclusion Following the conversion of scFv and Fab fragments to hIgG1 LALA, all conjugates retained their binding to the antigen as determined by ELISA, SPR, and FACS experiments. Clones 41, 307, and 334 were removed due to nonspecific binding behavior.

[0199] Example 12: Selection of αCD93 IgG based on their performance in functional assays. The selection of the most promising αCD93 antibodies was carried out by testing their functional effects in human and mouse endothelial cells (in vitro assay) as well as in mouse and human brain slice-tumor cell co-cultures (ex vivo assay).

[0200] result The blocking effects of CD93 antibodies were tested in in vitro and ex vivo functional assays. Results from all functional assays performed with 26 selected CD93 antibodies are summarized in Table 3 above. Experiments showing positive results in functional assays using four selected antibodies are presented as Examples 12-23, which include the materials and methods for each assay. Specifically, CD93 antibodies were tested for their ability to induce disruption of endothelial cell-cell junctions (intercellular gap formation), inhibit fibronectin fibril formation on endothelial cells, and hinder endothelial cell migration. These functional effects have been previously demonstrated by knocking down CD93 in endothelial cells using siRNA. In addition, using ex vivo mouse and human brain slice models, the blocking effects of αCD93 antibodies against perivascular invasion by tumor cells and deposition of vascular-associated fibronectin, which were previously found to be inhibited in CD93 knockout mouse brain slice Examples 6 and 7, were evaluated.

[0201] Interestingly, as shown in Table 3, different αCD93 antibodies exhibited varying magnitudes of effect on the functional assays tested, ranging from no effect to strong effect (Table 3). These results allowed us to score 26 αCD93 antibodies based on their performance in blocking CD93 function in complex environments and select the most promising antibody. As shown in Table 3, the human-mouse cross-reactive αCD93 antibodies AD169-85, AD169-309, AD169-324, and the human-specific αCD93 antibody AD169-8 showed effect in all functional assays.

[0202] conclusion Four αCD93 antibodies—AD169-8, AD169-85, AD169-309, and AD169-324—were selected based on their ability to block CD93 function.

[0203] Example 13: Evaluation of the binding of selected αCD93 IgG to human endothelial cells. The specificity of four selected αCD93 IgGs was tested in vitro by comparing antibody binding signals in human endothelial cells expressing CD93 with those in control cells not expressing CD93.

[0204] Materials and Methods: Cell Culture: Human dermal microvascular endothelial cells (HDBECs) were cultured as described in Example 4. U87 cells were cultured as described in Example 8.

[0205] Flow Cytometry: HDBEC cell growth in cell culture plates was detached by using Accutase dissociation buffer (ThermoFisher Scientific). Cells were incubated at 4°C for 20 minutes with different concentrations of αCD93 IgG1 or control IgG1 (0.4 μg / ml, 2 μg / ml, 10 μg / ml, and 50 μg / ml). After this incubation time, cells were washed and stained with secondary antibody anti-human kappa PE (SouthernBiotech #9230-09) to detect αCD93 IgG or control IgG. Samples were run on a CytoFLEX LX (Beckman Coulter) and data were analyzed using FlowJo version 10.5.3 (FlowJo LLC).

[0206] Human-specific and human- and mouse-cross-reactive antibodies AD169-8, AD169-85, AD169-309, and AD169-324 were tested in HDBEC and U87 cells.

[0207] Results: The graph in Figure 11 shows the percentage of positive cells for each of the αCD93 antibodies tested. All antibodies specifically bound to human endothelial cells expressing CD93, showing an increased percentage of positive cells in samples incubated with higher antibody concentrations. Importantly, undetectable or significantly low signals were found in U87 cells not expressing CD93 and in control IgG-treated cells.

[0208] Conclusion: This data shows that all tested αCD93 antibodies had high specificity for CD93 expressed by human endothelial cells in vitro.

[0209] Example 14. Evaluation of binding of selected αCD93 IgG to mouse endothelial cells In Example 13, specific binding of selected αCD93 IgG antibodies was tested by flow cytometry in human endothelial cells in vitro. In this example, specific binding of human and mouse cross-reactive αCD93 IgG was tested in mouse endothelial cells MS1 expressing CD93 and GL261 cells not expressing CD93.

[0210] Materials and methods: Cell culture: Mouse endothelial cells MS1 were cultured in DMEM (Gibco) supplemented with 10% FBS. GL261 cells were cultured as described in Example 5.

[0211] Flow cytometry: Specific binding of human and mouse cross-reactive antibodies AD169-85, AD169-309, and AD169-324 was tested in MS1 cells and GL261 cells as described in Example 13.

[0212] Results: The graph in Figure 12 shows the percentage of positive cells for each of the tested αCD93 antibodies. All antibodies specifically bound to mouse endothelial cells expressing CD93 and showed a high percentage of positive cells even in samples incubated with lower concentrations of αCD93 IgG (compare the bar corresponding to 0.2 μg / ml for each antibody to the bar corresponding to 50 μg / ml). Importantly, undetectable or significantly low signals were found in GL261 cells not expressing CD93 and in control IgG-treated cells.

[0213] Conclusion: This data demonstrates that all αCD93 IgG1 strains tested exhibit high specificity for CD93 expressed in mouse endothelial cells in vitro.

[0214] Example 15: Evaluation of the ability of αCD93 IgG to block fibronectin fibril formation in human endothelial cells. As shown in Example 4, CD93 deficiency resulted in the disruption of extracellular matrix fibronectin in endothelial cells. Here, we tested the ability of αCD93 IgG to block CD93 function by analyzing fibronectin deposition on human endothelial cells.

[0215] material and method Cell culture: Human endothelial cells were cultured as described in Example 4. Confluent HDBEC cells were incubated for 24 hours with 10 μg / ml of αCD93 IgG AD169-8, AD169-85, AD169-309, and AD169-32 or control IgG.

[0216] Immunofluorescence staining: HDBEC cells were immunofluorescently stained for fibronectin and cytoskeleton as described in Example 4. The cells were analyzed under a fluorescence microscope (DMi8, Leica).

[0217] result The immunofluorescence image in Figure 13A shows high-density fibronectin matrix deposition in endothelial cells treated with control IgG. However, treatment with αCD93 IgG shows a reduction in fibronectin deposition on the endothelial cell monolayer. The graph in Figure 13B shows the reduction in fibronectin regions in cells treated with αCD93 antibody compared to control IgG.

[0218] conclusion This data demonstrates that selected CD93 blocking antibodies reduce fibronectin fibril formation in human endothelial cells in vitro.

[0219] Example 16. Evaluation of the blocking effect of αCD93 IgG on fibronectin fibril formation in mouse endothelial cells in vitro Similar to Example 15, the ability of αCD93 IgG to block CD93 function was tested by analyzing fibronectin deposition on mouse endothelial cells.

[0220] Materials and methods Cell culture: Mouse endothelial cells MS1 were cultured as described in Example 14. Confluent MS1 was incubated with 10 μg / ml of αCD93 IgG AD169-85, AD169-309, and AD169-324 or control IgG for 24 hours.

[0221] Immunofluorescence staining: MS1 was immunofluorescently stained for fibronectin and cytoskeleton as described in Example 4. Cells were analyzed under a fluorescence microscope (DMi8, Leica).

[0222] Results Similar to the human endothelial cells in Example 15, the immunofluorescence images of MS1 cells showed a high-density fibronectin matrix deposition in endothelial cells treated with control IgG (fibronectin signal in control IgG, Figure 14A). In particular, treatment with any of the three αCD93 IgGs resulted in a significant reduction in fibronectin deposition on the endothelial cell monolayer compared to control IgG (Figure 14A and Figure 14B).

[0223] Conclusion This data indicates that the selected CD93 blocking antibodies reduce fibronectin fibril formation in mouse endothelial cells in vitro.

[0224] Example 17. Evaluation of the blocking effect of αCD93 IgG on fibronectin fibril formation in a mouse brain slice model As shown in Example 16, treatment with αCD93 IgG resulted in the disruption of extracellular matrix fibronectin in mouse endothelial cells in vitro. Here, we evaluate the ability of the αCD93 antibody to inhibit vascular-associated fibronectin in mouse brain slices co-cultured with GL261 cells.

[0225] material and method Cell culture: GL261 cells were cultured as shown in Example 2.

[0226] Brain slice model: Brain slices were obtained from wild-type mice as described in Example 5. GL261 cells were seeded onto the tissue by adding 10 μg / ml of αCD93 antibodies AD169-85, AD169-309, and AD169-324 to the brain slice medium 2 hours prior to seeding. After 24 hours, the brain slice medium containing αCD93 antibodies was replaced with fresh medium supplemented with 10 μg / ml of αCD93 antibodies. The brain slices were cultured for a further 24 hours, for a total of 48 hours.

[0227] Immunofluorescence staining: Immunofluorescence staining for fibronectin and the vascular marker CD31 was performed as described in Example 4.

[0228] result To investigate whether αCD93 IgG blocks CD93 function and inhibits fibronectin fibril formation, GL261 cells were co-cultured on brain slice tissue obtained from wild-type mice, and fibronectin deposition was evaluated by immunofluorescence staining after 48 hours of treatment with αCD93 antibody or control IgG. As shown in Figure 15, control IgG-treated brain slices show a high-density fibronectin matrix in the GL261 tumor region (defined by the dotted line) and in the vessels anterior to tumor invasion (Figure 15, arrowheads in the control IgG image). In contrast, brain slices treated with αCD93 antibodies AD169-85, AD169-309, and AD169-324 show a significant reduction in fibronectin signaling in both the region covered by GL261 cells and the vessels anterior to invasion (Figure 15, arrowheads in the AD169-85, AD169-309, and AD169-324 images).

[0229] conclusion Human / mouse cross-reactive αCD93 IgG, AD169-85, AD169-309, and AD169-324 neutralize CD93 function and inhibit vascular-associated fibronectin in mouse brain slices.

[0230] Example 18: Evaluation of the effect of αCD93 IgG on the integrity of human endothelial cell-to-cell junctions in vitro. CD93 deficiency is associated with disruption of endothelial cell-to-cell junctions in vitro (Langenkamp E. et al Cancer Res. 2015 Nov 1;75(21):4504-16). Here, we investigate whether treatment with αCD93 IgG replicates the junctional inhibition observed in CD93-deficient endothelial cells.

[0231] material and method Cell culture: Human endothelial cells (HDBECs) were cultured as described in Example 3. Confluent HDBECs were incubated for 24 hours with 10 μg / ml of αCD93 IgG AD169-8, AD169-85, AD169-309, and AD169-324 or control IgG.

[0232] Immunofluorescence staining: HDBEC cells were immunofluorescently stained with phalloidin-555 to detect the actin cytoskeleton, as described in Example 3. The cells were analyzed under a fluorescence microscope (DMi8, Leica).

[0233] Endothelial cell junction analysis: Disruption of cell junctions was assessed by measuring the formation of gaps between adjacent endothelial cells after treatment with αCD93 IgG or control IgG. Gap quantification was analyzed by thresholding cell-free areas based on actin staining. Gap areas were normalized by the number of cells in the field of view.

[0234] result To investigate whether αCD93 antibodies affect endothelial cell junctions, HDBEC cells were treated with AD169-8, AD169-85, AD169-309, and AD169-324 αCD93 antibodies, or with control IgG, and the cell monolayer was analyzed under a fluorescence microscope. As shown in Figure 16A, treatment with αCD93 antibodies induced disruption of endothelial cell-to-cell junctions, indicated by the appearance of gaps between adjacent cells. In contrast, HDBEC cells treated with control IgG showed intact endothelial monolayers with minimal gap formation (see the threshold gap region visualized in white in the "Gap" panel of Figure 16A). The increased gap formation in cells treated with αCD93 antibodies is represented by the quantification graph in Figure 16B.

[0235] conclusion These data demonstrate that CD93 blocking antibodies disrupt human endothelial cell-to-cell junctions in vitro, replicating the effects observed with silenced CD93 expression as described above (Langenkamp E. et al Cancer Res. 2015 Nov 1;75(21):4504-16). Furthermore, these data show that selected αCD93 antibodies exert blocking activity against human endothelial CD93.

[0236] Example 19: Evaluation of the effect of αCD93 IgG on the integrity of mouse endothelial cell-to-cell junctions in vitro. Similar to Example 18, this study investigates whether treatment with αCD93 IgG disrupts endothelial cell junctions in mouse endothelial cells in vitro.

[0237] material and method Cell culture: Mouse endothelial cells (MS1) were cultured as described in Example 14. Confluent MS1 cells were incubated for 24 hours with 10 μg / ml human / mouse cross-reactive αCD93 IgG AD169-85, AD169-309, and AD169-324 or control IgG.

[0238] Immunofluorescence staining: MS1 cells were immunofluorescently stained with phalloidin-555 to detect the actin cytoskeleton, as described in Example 4. The cells were then analyzed under a fluorescence microscope (DMi8, Leica).

[0239] Endothelial cell junction analysis: The disruption of cell junctions in MS1 ​​cells was evaluated as described in Example 18.

[0240] result Similar to the results observed with HDBEC (Example 18), MS1 cells treated with αCD93 antibodies AD169-85, AD169-309, and AD169-324 showed a damaged endothelial monolayer with gaps forming between adjacent cells. In contrast, an intact monolayer without gaps was observed in cells treated with control IgG (Figure 17A). The increased gap formation in cells treated with αCD93 antibodies is shown by the quantification of gap areas compared with control IgG-treated cells in Figure 17B.

[0241] conclusion These data demonstrate that CD93 blocking antibodies disrupt mouse endothelial cell-cell junctions. Furthermore, these results indicate that selected human / mouse cross-reactive αCD93 antibodies exhibited similar effects in human and mouse endothelial cells.

[0242] Example 20: Evaluation of the effect of αCD93 IgG on human endothelial cell migration. Downregulation of CD93 inhibits human endothelial cell migration in vitro (Langenkamp E. et al Cancer Res. 2015 Nov 1;75(21):4504-16). Here, we investigate whether blocking CD93 function on human endothelial cells using αCD93 IgG replicates the inhibition of cell migration observed in CD93-downregulated cells.

[0243] material and method Cell culture: Human endothelial cells (HDBEC) were cultured as described in Example 4.

[0244] Wound healing assay: Cell migration was investigated using a wound healing assay. HDBEC cells were cultured in IncuCyte ImageLock 96-well plates (Essen Bioscience) until confluence. Cells were treated with 10 μg / ml αCD93 IgG AD169-8, AD169-85, AD169-309, and AD169-324 or control IgG, and linear scratches were performed using IncuCyte WoundMaker (Essen Bioscience) after 2 hours. Scratch images were automatically captured every hour by IncuCyte Scan (IncuCyte ZOOM live cell analysis system, Essen Bioscience). Images were then processed using IncuCyte Cell Migration Analysis Software by defining the scratch mask and cell confluence mask. Cell migration was determined by the cell's ability to close the scratch over a 24-hour period.

[0245] result Treatment with αCD93 antibody significantly inhibits the migration of human endothelial cells. In fact, as shown in Figure 18, 24 hours after scratch creation, endothelial cells treated with control IgG showed approximately 90% wound closure, while in contrast, endothelial cells treated with αCD93 antibody showed only 45%–60% wound closure.

[0246] conclusion This data demonstrates that αCD93 antibodies AD169-8, AD169-85, AD169-309, and AD169-324 inhibit human endothelial cell migration in vitro.

[0247] Example 21: Evaluation of the potential cytotoxic effects of αCD93 IgG The potential cytotoxic effect on endothelial cells during treatment with αCD93 antibody was evaluated by analyzing apoptosis.

[0248] material and method Cell culture: Human endothelial cells (HDBEC) were cultured as described in Example 4. Mouse endothelial cells (MS1) were cultured as described in Example 14.

[0249] Apoptosis assay: Apoptosis was analyzed in endothelial cells by detecting the apoptosis marker cleaved caspase-3 (cCasp3) using immunofluorescence staining. HDBEC and MS1 cells were treated for 24 hours with 10 μg / ml αCD93 IgGLALA antibodies AD169-8, AD169-85, AD169-309, and AD169-324 or control IgG. Cells were then fixed with 4% PFA and stained with cleaved caspase-3 (Cell Signaling Technology, #9664), followed by a specific AlexaFluor-568 secondary antibody. Actin and nuclei were detected by phalloidin-647 and Hoechst staining. Images were acquired under a fluorescence microscope (DMi8, Leica). Treatment with 500 ng / ml TNFα was used as a positive control for cellular apoptosis. Cleaved caspase-3-positive cells were normalized by the total number of cells in each field of view.

[0250] result As shown in Figure 19, αCD93 IgG treatment did not induce apoptosis in human or mouse endothelial cells 24 hours after treatment (Figures 19A and 19B, respectively). In fact, less than 1% of the human and mouse endothelial cells analyzed were positive for the apoptosis marker cleavage caspase-3 (cCasp3).

[0251] conclusion This data demonstrates that the αCD93 antibodies AD169-8, AD169-85, AD169-309, and AD169-324 do not induce apoptosis in human or mouse endothelial cells in vitro.

[0252] Example 22: Evaluation of the effect of αCD93 antibody on perivascular invasion of tumor cells on mouse brain slices. In Example 7, we demonstrated that CD93 deficiency inhibits perivascular invasion of glioma cells in a mouse brain slice model. Here, we evaluated whether targeting CD93 using block antibodies AD169-85, AD169-309, and AD169-324 inhibits GL261 perivascular invasion and thereby reproduces the CD93 knockout phenotype.

[0253] material and method Cell culture: GL261 cells were cultured as described in Example 5.

[0254] GL261 Spheroid: GL261 spheroids were obtained as described in Example 6.

[0255] Brain slice models and antibody treatment: Mouse brain slices derived from wild-type mice were obtained as described in Example 5. Treatments AD169-85, AD169-309, and AD169-324 were performed on the brain slices as described in Example 17.

[0256] Immunofluorescence staining: Immunofluorescence staining of CD31-positive blood vessels and detection of GFP signals from GL261 cells were performed as described in Example 5.

[0257] result Treatment with CD93-blocking antibodies significantly inhibited perivascular invasion of GL261 cells on mouse brain slices. As shown in the immunofluorescence image in Figure 20A, GL261 cells seeded in control IgG-treated brain slices invaded brain tissue by hijacking blood vessels (arrowheads in the control IgG image, Figure 20A). In contrast, significantly fewer GL261 vascular-associated sprouts were observed in brain slices treated with αCD93 antibodies AD169-85, AD169-309, and AD169-324 (arrowheads in the AD169-85, AD169-309, and AD169-324 images, Figure 20A). Quantification of GL261 hijacking vessels (indicated by arrowheads) adjacent to GL261 spheroid bodies (indicated by dotted lines) showed statistically significant inhibition of tumor cell invasion in brain slices treated with AD169-85, AD169-309, and AD169-324 antibodies compared to control IgG-treated samples (Figure 20B).

[0258] conclusion This data demonstrates that the CD93 blocking antibody suppresses the migration of GL261 cells along blood vessels in mouse brain slices, replicating the inhibitory effect of CD93 deficiency shown in Example 7.

[0259] Example 23: Evaluation of the effect of αCD93 IgG on perivascular invasion of tumor cells on human brain slices. In Example 22, we demonstrated that αCD93 antibody treatment inhibited perivascular invasion of mouse glioma cells in a mouse brain slice model. Here, we evaluate whether αCD93 antibody blocks perivascular invasion of human glioma cells co-cultured on human brain slices derived from surgically resected brain tissue of glioblastoma patients.

[0260] material and method Cell culture: U3013 human glioma cells were obtained from the Human Glioblastoma Cell Culture (HGCC) Biobank (https: / / www.hgcc.se / ). U3013 cells were cultured in laminin-coated Primaria 60mm dishes (Corning, 353802) in Neurobasal medium supplemented with B27 1X (Life Technologies 12587-010), N2 1X (Life Technologies 17502-048), 10 μg / ml human FGFBasic and human EGF (Peprotech), and DMEM / F12 Glutamax in a 1:1 ratio (Life Technologies). If confluent, U3013 cells were stained with 6 μM CellTracker Orange CMRA (ThermoFisher Scientific, C34551) according to the manufacturer's instructions.

[0261] Human brain slices: Tissue newly excised from the periphery of tumors in glioblastoma patients undergoing brain surgery was vibratome-sectioned into 300 μm thick slices in Hibernate-A medium (A1247501, Gibco) supplemented with 13 mM D-glucose (Sigma-Aldrich), 30 mM N-methyl-D-GLucamin (Sigma-Aldrich), and 1 mM GlutaMAX (Gibco). Brain slices were cultured in 12-well culture inserts (vWR, 7342736P) in Neurobasal A-Medium (10888022, Gibco) supplemented with 2% serum-free B-27 (50×) (17504001, Gibco), 13 mM D-glucose (Sigma-Aldrich), 1 mM MgSO4 (Sigma-Aldrich), 15 mM Hepes (Sigma-Aldrich), and 2 mM GlutaMAX (lot number 1978435, Gibco), in the presence of antibiotics. 10 μg / ml of αCD93 antibodies AD169-8, AD169-85, AD169-309, and AD169-324, or control IgG, were added to the brain slice medium. Two hours after incubation with the antibody, U3013 cells (10,000 cells resuspended in 1 μl of complete medium) were injected into brain slices using a 10 μl Hamilton syringe. The brain slice medium containing αCD93 antibody was replaced after 24 hours with fresh medium supplemented with 10 μg / ml of αCD93 antibody. The brain slices were incubated in a humidified chamber at 37°C and 5% CO2 / 95% air for a further 24 hours (total 48 hours). After this period, the brain slices were fixed with 4% PFA at room temperature for 1.5 hours and stained as described below.

[0262] Immunofluorescence staining: Human brain slices were stained with monoclonal mouse α-human CD31 (MO823 Dako). The sections were washed and incubated with a specific mouse Alexa Fluor488 conjugated secondary antibody (Invitrogen). Nuclei were visualized using Hoechst (Life Technologies), and the U3013 fluorescence signal was detected by a 555 nm laser. Brain slices were analyzed under a confocal microscope (Leica SP8).

[0263] result U3013 glioma cells co-cultured on human brain slices showed perivascular invasion in the presence of control IgG treatment. Indeed, as shown in the fluorescence image (Figure 21, control IgG), U3013 cells aligned along CD31-positive vessels and exhibited an elongated shape indicating a migratory phenotype (arrowhead in the U3013 control IgG image). In contrast, U3013 cells co-cultured on human brain slices treated with αCD93 antibodies AD169-8, AD169-85, AD169-309, and AD169-324 showed a shift in cell morphology from the elongated migratory shape observed in brain slices treated with control IgG to a rounder, less invasive phenotype. Indeed, U3013 perivascular invasion was significantly inhibited in the presence of all four αCD93 antibodies, as indicated by the lack of cells extending along CD31-positive vessels. Treatment with AD169-8 and AD169-309 antibodies induced aggregation of tumor cells (arrowheads in AD169-8 and AD169-309 images, Figure 21) and inhibited interaction with the vascular system. Conversely, treatment with AD169-85 and AD169-324 induced a change to a round, single-cell shape of U3013 morphology, resulting in a less invasive phenotype (arrowheads in AD169-85 and AD169-309 images, Figure 21).

[0264] conclusion This data shows that all four CD93 blocking antibodies significantly inhibit the perivascular invasion of tumor cells during co-culture in human brain slices derived from glioblastoma patients.

[0265] Example 24: In vivo distribution of αCD93 antibody The in vivo distribution of selected human-mouse cross-reactive αCD93 IgG (AD169-85, AD169-309, and AD169-324) was evaluated in healthy and tumor-carrying mice after intravenous administration.

[0266] material and method In vivo research: GL261 cells (20,000 cells in 2 μl of Dulbecco's PBS) were orthotopically injected into the brains of C57BL / 6 mice. Tumor tissue was collected 23 days post-tumor transplantation, and the brain, kidneys, liver, and spleen were harvested and frozen-sectioned for immunofluorescence staining. 5 mg / kg of αCD93 antibodies (AD169-85, AD169-309, and AD169-324), as well as control IgG, were injected into the tail vein 24 hours prior to sacrifice of the mice and collection of organs.

[0267] Immunofluorescence staining and imaging: Frozen-sectioned tissues were immunofluorescently stained for anti-kappa-AF488 (Southern Biotech, REF 2060-30) to detect αCD93 antibody and control IgG, and for CD31 (2H8, ThermoFisher Scientific, MA3105) to visualize blood vessels. Images were acquired under a confocal microscope (Leica SP8).

[0268] result Biodistribution analysis of intravenously injected αCD93 antibodies AD169-85, AD169-309, and AD169-324 showed that all three antibodies reached the tumor site after 24 hours of circulation and were able to bind to tumor-associated blood vessels with high affinity (Figure 22). Indeed, as shown in Figure 22A, a positive signal for αCD93 antibody (anti-kappa) was found in the GL261 tumor core co-localized with the vascular marker CD31. In contrast, a negligibly small amount of anti-kappa signal was detected in the tumor core of mice injected with control IgG (Figure 22A). Similarly, analysis of the anterior tumor invasion (dotted line in Figure 22B) showed a strong positive signal for αCD93 antibody in the blood vessels at the tumor boundary (arrowhead in Figure 22B). Importantly, minimal αCD93 antibody signaling was detected in the contralateral hemisphere vessels of the tumor (tumor CLH, Figure 22C) or in brain tissue from healthy mice (Figure 22D), suggesting that αCD93 antibodies target tumor vessels with high affinity. Quantification of the percentage of anti-kappa signaling against the vascular marker CD31 in the brain, as well as the kidneys, liver, and spleen, is shown in the graph in Figure 22E. Interestingly, signals from all three tested antibodies covered 43–85% of the total vascular area in the tumor core and tumor borders. In contrast, signals from all three tested antibodies covered less than 12.8% of the vessels in the contralateral hemisphere of the tumor or in healthy mice. In addition, minimal αCD93 antibody signaling was detected in the kidneys, liver, and spleen of GL261-possessing mice (graph in Figure 22E).

[0269] conclusion These data demonstrate that all three human-mouse cross-reactive antibodies against CD93 can cross the impaired blood-brain barrier at tumor sites and bind to tumor blood vessels with high affinity and specificity. In addition, these results show minimal binding of the antibodies to brain tissue far from tumor sites, as well as minimal binding to tissues including the brain, kidneys, liver, and spleen of healthy individuals.

[0270] Example 25: In vivo evaluation of the effect of αCD93 antibody on perivascular invasion by tumor cells. In Example 24, it was demonstrated that αCD93 antibody administered intravenously to tumor-bearing mice specifically bound to tumor blood vessels. In addition, Examples 22 and 23 showed that αCD93 antibody inhibited perivascular invasion of glioma cells in ex vivo models of mouse and human brain slices.

[0271] In this example, we evaluate whether intravenous administration of αCD93 antibody converted to the mouse IgG1 skeleton blocks perivascular invasion of glioma cells in tumor-carrying mice.

[0272] material and method In vivo research: GL261-GFP cells (20,000 cells in 2 μl of Dulbecco PBS) were orthotopically injected into the brains of C57BL / 6 mice (10 mice / group). Seven days after tumor injection, 20 mg / kg of αCD93 antibodies AD169-85 and AD169-309 converted to mouse IgG1 scaffolds, or control IgG, were injected into the tail vein. Antibody administration was repeated on days 11, 15, and 19 after tumor injection, for a total of four intravenous injections. Tumor tissue was collected 23 days after tumor transplantation, the brain was collected, vibratome sectioned, and immunofluorescence staining and tumor boundary analysis were performed (see experimental timeline in Figure 23A).

[0273] Immunofluorescence staining and imaging: Brain tissue was immunofluorescently stained with CD31 (2H8, ThermoFisher Scientific, MA3105) to visualize blood vessels. Nuclei were visualized using Hoechst (Life Technologies), and the GL261-GFP fluorescence signal was detected by a 488 nm laser. Images were acquired under a confocal microscope (Leica SP8).

[0274] result Repeated intravenous administration of αCD93 antibodies AD169-85 or AD169-309 in GL261-carrying mice (Figure 23A) resulted in a significant reduction in perivascular invasion of GL261 cells. As shown in Figure 23B, the tumor invasion area, indicated by the dotted line, was reduced in mice treated with αCD93 antibodies AD169-85 and AD169-309 compared to the invasion area observed at the tumor boundary in the control IgG group. The quantitative analysis of the invasion area at the tumor boundary is shown in the graph in Figure 23C.

[0275] conclusion This data demonstrates that systemic administration of AD169-85 or AD169-309 αCD93 antibodies significantly inhibits perivascular invasion of tumor cells in mice carrying GL261 glioma.

[0276] Example 26: In vivo evaluation of the inhibitory effect of αCD93 antibody on fibronectin fibril formation in GL261-possessing mice. Example 17 demonstrated that the αCD93 antibody inhibits the formation and generation of vascular-associated fibronectin fibrils in mouse brain slices co-cultured with GL261 cells.

[0277] In this example, we evaluate whether in vivo administration of αCD93 antibody converted to a mouse skeleton affects fibronectin fibril formation in GL261-possessing mice.

[0278] material and method In vivo research: The in vivo study was designed as described in Example 25.

[0279] Immunofluorescence staining and imaging: Tumor tissue was stained with anti-CD31 (2H8, Thermo Fisher Scientific, MA3105) to visualize blood vessels, and fibronectin was stained with anti-fibronectin (Abcam, ab2413). The nuclei were visualized using Hoechst (Life Technologies). Images were acquired under a confocal microscope (Leica SP8).

[0280] result Intravenous administration of αCD93 antibodies AD169-85 or AD169-309 in GL261-possessing mice showed that both antibodies significantly reduced vascularized fibronectin deposition in tumor areas compared to levels observed in control IgG-treated mice (quantification graphs in Figures 24A and 24B).

[0281] conclusion This data demonstrates that systemic administration of AD169-85 or AD169-309 αCD93 antibodies significantly inhibits fibronectin fibril formation in mice carrying GL261 glioma.

[0282] Example 27: Evaluation of the effects of αCD93 antibody treatment on endothelial activation, T cell recruitment, and response to αPD1 blockade immunotherapy in GL261-possessing mice. Example 3 demonstrated that CD93 deficiency induces endothelial activation and T cell recruitment in glioma-carrying mice. This example evaluates whether systemic administration of αCD93 antibody in GL261-carrying mice reproduces this phenotype.

[0283] material and method In vivo research: An in vivo study aimed at analyzing VCAM1 levels and the abundance of immune cells (CD8 and CD3-positive T cells) was designed as described in Example 25.

[0284] For immunotherapy treatment, survival studies were conducted. As described in Example 25, mice treated with control IgG or AD169-309 were further treated three times with intraperitoneal injection of αPD1 antibody (clone: ​​RMP1-14, catalog number BE0146, BioXCell, USA) or rat-IgG2 (used as an isotype control) (200 μg / dose on days 9, 13, and 17 after tumor injection). Mice were sacrificed when they lost more than 20% of their body weight or showed symptoms.

[0285] Immunofluorescence staining and imaging: Tumor tissue was stained with anti-CD31 (2H8, Thermo Fisher Scientific, MA3105), anti-VCAM1 (R&D System, AF643), anti-CD8 (BD BioSciences, 550281), and anti-CD3 (BD ​​BioSciences, 557869) to visualize blood vessels. Nuclei were visualized using Hoechst (Life Technologies). Images were acquired under a confocal microscope (Leica SP8).

[0286] result Analysis of tumor tissue after systemic delivery of αCD93 antibody in GL261-carrying mice showed that mice receiving AD169-309 antibody exhibited increased expression of the vascular adhesion molecule VCAM1 in tumor vessels compared to VCAM1 levels observed in groups treated with AD169-85 or control IgG (quantification graphs in Figures 25A and 25B). The increased endothelial activation observed in GL261 tumors treated with AD169-309 αCD93 antibody was accompanied by an increased abundance of T cells. Indeed, as shown in Figures 25C-D and 25E-F, the presence of CD8-positive and CD3-positive T cells was increased in the AD169-309-treated group compared to the AD169-85-treated group or the control IgG group. Furthermore, as shown in the survival graph in Figure 25G, combination therapy with the CD93 blocking antibody AD169-309 and αPD1 blocking immunotherapy shows a significant improvement in survival compared to control IgG mice (control IgG-αPD1) that received αPD1 immunotherapy.

[0287] conclusion These data indicate that systemic treatment with the αCD93 antibody AD169-309 in GL261 tumor-carrying mice is associated with a marked induction of tumor endothelial activation and promotion of cytotoxic T cell infiltration, as indicated by increased VCAM1 expression in tumor blood vessels. In addition, when administered in combination with αPD1 immunotherapy, the CD93 blocking antibody AD169-309 extends mouse survival, demonstrating that AD169-309 improves the efficacy of immunotherapy.

[0288] Example 28: Evaluation of the effect of αCD93 treatment on vascular permeability in mice carrying GL261 tumors. Examples 18 and 19 demonstrated that in vitro treatment of human and mouse endothelial cells with αCD93 antibodies disrupts endothelial cell-to-cell junctions, suggesting a potential effect on vascular permeability. Therefore, here we evaluate whether systemic administration of αCD93 IgG in GL261 tumor-carrying mice increases vascular permeability by loosening endothelial cell-to-cell junctions.

[0289] material and method In vivo research: The in vivo study was designed as described in Example 25.

[0290] Immunofluorescence staining and imaging: Tumor tissue was stained with anti-CD31 (2H8, Thermo Fisher Scientific, MA3105), anti-VE-cadherin (BD BioSciences, 555289), and anti-fibrinogen (Dako, A0080) to visualize blood vessels. Nuclei were visualized using Hoechst (Life Technologies). Images were acquired under a confocal microscope (Leica SP8).

[0291] result The effect of in vivo treatment with αCD93 antibody on endothelial junction integrity was evaluated in GL261 tumors by immunofluorescence staining, and endothelial junction molecule VE-cadherin was visualized. As shown in Figure 26A and the quantification graph in Figure 26B, AD169-85 treatment significantly reduced VE-cadherin signaling in tumor vessels. In contrast, tumors treated with AD169-309 showed no effect on VE-cadherin levels compared to the control IgG group (Figures 26A-B).

[0292] In line with these results, mice treated with AD169-85 showed increased extravasation of endogenous fibrinogen compared to the control IgG group (Figure 26C-D). In contrast, mice treated with AD169-309 showed no increase in fibrinogen leakage compared to the control group (Figure 26C-D).

[0293] conclusion These data indicate that treatment with the αCD93 antibody AD169-85 affects endothelial junction integrity and promotes vascular permeability in GL261-possessing mice. This observation suggests that AD169-85 treatment may facilitate the delivery of several antitumor drugs by increasing vascular permeability.

[0294] Example 29: Epitope binning of hIgG1 LALA antibody, and characterization of their interactions with CD93 and interaction partners. CD93 blocking antibodies AD169-8, AD169-85, AD169-309, and AD169-324 were binned against each other and against previously identified interaction partners of CD93. Binding was also tested against different cleavage constructs of CD93 to determine their approximate binding sites.

[0295] material and method Cutting and production of hCD93 constructs The constructs used in this example are shown in Figure 27. The hCD93 and hCD93 lectin constructs were ordered as synthetic genes containing 'a' from GeneArt (Thermo Scientific). hCD93 (amino acids 1-580) has a C-terminal aviHis tag, while hCD93lectin (amino acids 1-258) has an N-terminal aviHis tag followed by a TEV protease site. Constructs hCD93_470, hCD93_344, hCD93EGF, and hEGF1-2 were produced from hCD93 by PCR and all have a C-terminal aviHis tag.

[0296] All hCD93 cleavage constructs possessed the gp67 signaling sequence and were expressed as secreted proteins in baculovirus-infected insect cells. The virus was produced using the flashBac system (Oxford Expression Technologies). Recombinant viruses were harvested after approximately 5 days. The virus was then amplified by infecting Sf9 cells with the first-generation virus. The virus was harvested after 3-4 days, and amplification was repeated. The third-generation virus was used for expression. For hCD93 expression, Sf9 cells were expanded to 1.5-2 million cells / ml in 1-2 L of water. Virus and biotin were added to the culture medium, and the medium containing secreted hCD93 was harvested after 92-96 hours. The protein was captured from the medium by immobilized metal affinity chromatography (IMAC) in 50 mM Tris pH 7.8, 300 mM NaCl, and 20 mM imidazole, and eluted with 250 mM imidazole. The proteins were further purified by either size exclusion chromatography in TBS + 5% glycerol, or by capture in a streptavidin mutein matrix (Roche) and elution with biotin. The mass and identity of the constructs were confirmed by mass spectrometry.

[0297] ELISA To test the binding of IgG to different biotinylated CD93 constructs (Figure 27) by ELISA, streptavidin (SA, 1 mg / ml diluted in PBS) was directly immobilized on a 384-well microtiter plate overnight at 4°C. The plate was washed four times between each incubation step using an automated plate washer with PBS + 0.05% Tween20. Biotinylated CD93 constructs (1 μg / ml, Table 1) were captured on the SA coating at room temperature for 1 hour. The microtiter plate was blocked at room temperature for 1 hour with blocking buffer (PBS + 0.5% BSA + 0.05% Tween20). IgG antibody, diluted to 1 μg / ml in blocking buffer, was captured on the immobilized antigen at room temperature for 1 hour. Binding IgG was detected using wasabi peroxidase (HRP) conjugated anti-human IgG kappa antibody (Southern Biotech, #9230-05), followed by incubation with chromogen Ultra TMB-ELISA (Thermo Scientific). Signal generation was stopped by adding 1M sulfuric acid, and absorbance was measured at 450 nm.

[0298] SPR The following reference antibodies were used: R3 (BD ​​Bioscience), R139 (BD Bioscience), MM01 (Sino Biological), and MM02 (Sino Biological). The reported CD93 interaction partner IGFBP7 was purchased from Acro Biosystems. Fragments of MMRN2, which have been reported to interact with CD93 (amino acids 495-674), were produced in-house, expressed in E. coli with a C-terminal His tag, and purified by IMAC in 50 mM Tris pH 7.8, 300 mM NaCl, and 20 mM imidazole (eluted with 250 mM imidazole). This was further purified by size exclusion chromatography on a Superdex200 column (Cytiva) in 50 mM Tris pH 7.8 and 150 mM NaCl. Binding to hCD93 was confirmed by SPR and ELISA, and mass and identity were determined by mass spectrometry.

[0299] The sensor chips were prepared according to the manufacturer's recommendations. The α-human Fab antibody mixture was diluted to 50 μg / ml (1:20) in immobilization buffer and immobilized on all eight surfaces of the CM5 Series S chip, including all eight reference planes. Approximately 7,000 RU was immobilized per surface.

[0300] Each AD169 antibody was diluted to 5 μg / mL in electrophoresis buffer to obtain capture RU levels of approximately 400–500 RU. Once captured on the α-human Fab surface, the remaining unbound α-human Fab antibody was blocked by injection of 150 μg / mL of Herceptin. A single injection of 250 nM hCD93 pre-incubated with 50 nM AD169 antibody was performed, followed by regeneration in 10 mM glycine-HCl, pH 2.1. The binding or non-binding of the captured AD169 conjugate to CD93 pre-incubated with a second AD169 conjugate was compared to cycles injecting CD93 alone. Reference antibodies R3, R139, MM01, and MM02 were assayed only as pre-incubations with CD93 and were not captured on the chip surface (they were not compatible with the immobilized capture antibodies).

[0301] result ELISA for analyzing antibody binding to various cleaved CD93 protein variants The binding of 15 αCD93 antibodies to CD93 cleavage was determined by ELISA. The antibodies can be classified into three main groups: EGF1-2 conjugates, lectin-EGF1-2 conjugates, and lectin domain binding factors (Table 8). EGF1-2 and lectin-EGF1-2 conjugates also bind to mCD93, while lectin domain binding factors do not. AD169-33 binds only to hCD93, while AD169-108 and 334 primarily bind to mCD93. ELISA results were confirmed by SPR (data not shown). [Table 10]

[0302] Epitope binning of antibodies and interaction partners Epitope binning results were evaluated and clones were grouped into bins sharing similar binding patterns. The data suggested that clone AD169-8 did not compete with other conjugates for binding to CD93, while AD169-85, -309, and -324 competed with each other (Table 9). Reference antibodies R139 and MM02 did not compete with any of the AD169 clones for binding to CD93. AD169-8 competed with reference R3 for binding but not with MM01. [Table 11]

[0303] conclusion Binning experiments showed that clones AD169-85, -309, and -324 competed for binding to CD93, suggesting they bind to overlapping epitopes. AD169-8 did not compete with the other conjugates. Reference antibodies R139 and MM02 did not compete with any of the AD169 clones for binding to CD93. AD169-8 was the only clone that competed with reference antibody R3 for binding but not with MM01.

[0304] Itemized Embodiments 1. A drug containing a binding site that specifically binds to vascular differentiation antigen group 93 and CD93.

[0305] 2. The agent described in item 1, wherein the binding of the agent to CD93 results in one or more of the following: i) inhibition of perivascular tumor cell migration, ii) inhibition of tumor cell invasion, and iii) inhibition of tumor cell proliferation.

[0306] 3. The agent according to item 1 or 2, wherein the binding of the agent to CD93 provides an antagonistic effect by preventing the functional activation of CD93 by its natural ligand.

[0307] 4. The drug or its binding portion is a binding protein, as described in items 1 to 3.

[0308] 5. A binding protein that specifically binds to CD93 and includes an antibody binding domain, wherein the binding domain includes a heavy chain variable domain (VH) and a light chain variable domain (VL), each of which includes three complementarity-determining regions (CDRs), and the amino acid sequence of the CDRs is VHCDR1, defined by sequence number 1, VHCDR2, defined by Sequence ID 2, VHCDR3 as defined by Sequence ID 3, VLCDR1, defined by sequence number 4, VLCDR2 as defined by AAS, VLCDR3 as defined by sequence number 5, and selected from the group including CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more, wherein the binding protein is K D A binding protein with a mass of <21 nM.

[0309] 6. VHCDR1, VHCDR2, and VLCDR2 are located next to specific framework amino acids, and the CDR and framework amino acid (faa) sequences are VHCDR1 and faa, defined by Sequence ID 6, VHCDR2 and faa, defined by Sequence ID 7, The binding proteins described in item 5, selected from the group containing CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more.

[0310] 7. The CDR is VHCDR1 is selected from sequence numbers 8-10. VHCDR2 is selected from sequence numbers 11-14. VHCDR3 is selected from sequence numbers 15-18. VLCDR1, defined by sequence number 4, VLCDR2 as defined by AAS, VLCDR3 is selected from sequence numbers 19-21. and the binding proteins described in item 1, individually selected from the group containing CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more.

[0311] 8. The amino acid sequence of the CDR is i) VHCDR1, defined by sequence number 8, VHCDR2 as defined by sequence number 11, VHCDR3 as defined by sequence number 15, VLCDR1, defined by sequence number 4, VLCDR2 as defined by AAS, and A binding protein having VLCDR3 as defined by Sequence ID No. 19, ii) VHCDR1, defined by sequence number 8, VHCDR2, defined by sequence number 12, VHCDR3 as defined by sequence number 16, VLCDR1, defined by sequence number 4, VLCDR2 as defined by AAS, and A binding protein having VLCDR3 as defined by Sequence ID No. 20, iii) VHCDR1, defined by sequence number 9, VHCDR2, defined by sequence number 13, VHCDR3 as defined by sequence number 17, VLCDR1, defined by sequence number 4, VLCDR2 as defined by AAS, and A binding protein having VLCDR3 as defined by Sequence ID No. 20, iv) VHCDR1, defined by sequence number 10, VHCDR2, defined by sequence number 14, VHCDR3 as defined by sequence number 18, VLCDR1, defined by sequence number 4, VLCDR2 as defined by AAS, and A binding protein having VLCDR3 as defined by Sequence ID No. 21, The binding proteins described in item 5 or 7, selected from the group containing CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more.

[0312] 9. The sequence of the CDR containing the framework amino acid (faa) is i) VHCDR1 and faa, defined by sequence number 43, The binding protein having VHCDR2 and faa as defined by Sequence ID No. 46, ii) VHCDR1 and faa, defined by sequence number 43, The binding protein having VHCDR2 and faa as defined by Sequence ID No. 47, iii) VHCDR1 and faa, defined by sequence number 44, The binding protein having VHCDR2 and faa as defined by Sequence ID No. 48, iv) VHCDR1 and faa, defined by sequence number 45, The binding protein having VHCDR2 and faa as defined by Sequence ID No. 49, The binding proteins described in item 6, selected from the group containing CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more.

[0313] 10. The binding protein described in items 5 and 7-8, wherein the VH sequence includes an amino acid sequence selected from the group consisting of SEQ ID NOs. 22-25 and a sequence having 80% or more identity with it, such as 85%, 90%, 95%, or more, and the VL sequence includes an amino acid sequence selected from the group consisting of SEQ ID NOs. 26-29 and a sequence having 80% or more identity with it, such as 85%, 90%, 95%, or more.

[0314] 11. The binding protein according to item 10, wherein the CDR sequence does not contain any mutations in the amino acid sequence, or the sequence mutations in the CDR amino acid sequence are 4%, 3%, 2%, 1%, or less, up to a maximum of 5%.

[0315] 12. The amino acid sequences of VH and VL are, i) VH as defined by Sequence ID 22, and Binding proteins having VL as defined by Sequence ID No. 26, ii) VH as defined by Sequence ID 23, and Binding proteins having VL as defined by Sequence ID No. 27, iii) VH as defined by sequence number 24, and Binding proteins having VL as defined by Sequence ID No. 28, iv) VH as defined by Sequence ID 25, and Binding proteins having VL as defined by Sequence ID No. 29, The binding proteins described in item 10 or 11, selected from the group including sequences having 80% or more identity with them, such as 85%, 90%, 95%, or more.

[0316] 13. The binding protein according to items 5 to 12, wherein the binding protein is a monoclonal antibody or an antigen-binding fragment selected from the group consisting of an Fv fragment, a Fab-like fragment, a disulfide-binding fragment, and a domain antibody.

[0317] 14. The binding protein described in item 13, wherein the Fv fragment is an scFv fragment.

[0318] 15. The binding protein described in item 13, wherein the Fab-like fragment is a Fab or F(ab')2 fragment.

[0319] 16. The binding protein according to items 5-13 and 15, wherein the binding protein is a monoclonal antibody of an IgG1 isotype, such as an IgG1 LALA antibody.

[0320] 17. The binding protein comprises a heavy chain and a light chain, the heavy chain comprising a VH domain and three constant domains, CH1, CH2, and CH3, CH1 and CH2 being joined via a hinge region, the light chain comprising a VL domain and a constant domain, CL, and the amino acid sequences of the domains and hinge region are, CH1, defined by sequence number 30, CH2, defined by Sequence ID 31, CH3 as defined by Sequence ID No. 32, CL as defined by sequence number 33, Hinge region defined by sequence number 34, and the binding proteins described in items 13-16, defined as sequences having 80% or more identity with them, such as 85%, 90%, 95%, or more.

[0321] 18. Cells engineered to express a chimeric antigen receptor (CAR), wherein the CAR comprises an antigen-binding domain, a transmembrane domain connected to the antigen-binding domain by a hinge region, and optionally, an intracellular domain connected to one or more costimulatory domains, and the antigen-binding domain comprises a binding protein described in any one of items 5 to 17.

[0322] 19. The cells described in item 18, wherein the cells are human cells.

[0323] 20. The cells described in item 18 or 19, wherein the cells are immune effector cells such as T cells, NK cells, or macrophages.

[0324] 21. A nucleic acid molecule that codes for a drug or binding protein as described in items 1-17.

[0325] 22. The nucleic acid molecule according to item 21, wherein the nucleic acid molecule encodes a binding protein containing an antibody binding domain, and the binding domain includes a heavy chain variable domain (VH) and a light chain variable domain (VL).

[0326] 23. A nucleic acid molecule as described in item 22, wherein the VH coding sequence includes a nucleotide sequence selected from the group consisting of SEQ ID NOs. 35 to 38, and a sequence having 80% or more identity with it, such as 85%, 90%, 95%, or more, and the VL coding sequence includes a nucleotide sequence selected from the group consisting of SEQ ID NOs. 39 to 42, and a sequence having 80% or more identity with it, such as 85%, 90%, 95%, or more.

[0327] 24. The nucleotide sequences of VH and VL are, i) VH as defined by Sequence ID 35, and The binding protein encoded by VL as defined by Sequence ID No. 39, ii) VH as defined by sequence number 36, and The binding protein encoded by VL as defined by Sequence ID No. 40, iii) VH as defined by Sequence ID 37, and The binding protein encoded by VL as defined by Sequence ID No. 41, iv) VH as defined by sequence number 38, and The binding protein encoded by VL as defined by Sequence ID No. 42, Furthermore, nucleic acid molecules as described in item 23, selected from the group including sequences having 80% or more identity with them, such as 85%, 90%, 95%, or more.

[0328] 25. The nucleic acid molecule according to any one of items 21 to 24, wherein the nucleic acid molecule comprises a nucleotide sequence in the form of mRNA expressed in mammalian cells.

[0329] 26. A nucleic acid molecule according to any one of items 21-25, wherein the nucleotide sequence of the nucleic acid molecule is inserted into a mammalian cell using the CRISPR / Cas9 method.

[0330] 27. Vectors, such as expression vectors, containing nucleic acid molecules as described in items 21-24.

[0331] 28. The vector described in item 27, wherein the vector is a plasmid or a viral vector.

[0332] 29. Isolated host cells for the production of binding proteins, containing the vector described in item 27 or 28.

[0333] 30. A pharmaceutical composition comprising a drug described in any one of items 1 to 4, a binding protein described in any one of items 5 to 17, a cell described in any one of items 18 to 20, or a nucleic acid molecule described in any one of items 21 to 26, and a pharmaceutically acceptable carrier or excipient.

[0334] 31. A drug described in any one of items 1-4, a binding protein described in any one of items 5-17, a cell described in any one of items 18-20, a nucleic acid molecule described in any one of items 21-26, or a pharmaceutical composition described in item 30, for use in therapy.

[0335] 32. Drugs, binding proteins, cells, nucleic acid molecules, or pharmaceutical compositions described in item 31, for use in cancer therapy.

[0336] 33. The drugs, binding proteins, cells, nucleic acid molecules, or pharmaceutical compositions described in item 32, for use in brain cancer therapies, such as glioma therapy, including glioblastoma therapy.

[0337] 34. Any of the agents, binding proteins, cells, or pharmaceutical compositions described in any one of items 31-33 for use in inhibiting perivascular tumor cell migration, ii) inhibiting tumor cell invasion, and / or iii) inhibiting tumor cell proliferation.

[0338] 35. A drug, binding protein, cell, nucleic acid molecule, or pharmaceutical composition described in any one of items 31 to 35, for use in combination with one or more additional therapeutic regimens, wherein the additional therapeutic regimen is selected from i) antitumor agents, ii) chemotherapy, iii) immunotherapy including tumor-targeting antibodies, iv) irradiation, and v) tyrosine kinase inhibitors or neutralizing antibodies against growth factors / cytokines or their receptors, which are pathway blockers for treating cancer.

[0339] 36. A method for treating a subject in need of treatment, comprising administering a therapeutically effective amount of any one of items 1 to 4, any one of items 5 to 17, any one of items 18 to 20, any one of items 21 to 26, or any one of items 30, a drug described in any of items 30.

[0340] 37. Use of any one of the agents described in item 1 to 4, any one of the binding proteins described in item 5 to 17, any one of the cells described in item 18 to 20, or any nucleic acid molecule described in any one of the items 21 to 26, or any pharmaceutical composition described in item 30, for the treatment or prevention of cancer, preferably glioblastoma.

Claims

1. A drug comprising a binding portion that specifically binds to vascular differentiation antigen group 93 and CD93, wherein the binding of the drug to CD93 results in one or more of the following: i) inhibition of perivascular tumor cell migration, ii) inhibition of tumor cell invasion, and iii) inhibition of tumor cell proliferation.

2. The drug or its binding portion is a binding protein containing an antibody binding domain, wherein the binding domain comprises a heavy chain variable domain (VH) and a light chain variable domain (VL), each comprising three complementarity-determining regions (CDRs), and the amino acid sequence of the CDRs is, VHCDR1, defined by sequence number 1, VHCDR2, defined by Sequence ID No. 2, VHCDR3 as defined by Sequence ID 3, VLCDR1 as defined by sequence number 4, VLCDR2 as defined by AAS, VLCDR3 as defined by sequence number 5, The agent according to claim 1, wherein the binding protein is selected from the group consisting of CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more, and the binding protein has a KD < 21 nM.

3. The aforementioned CD-R VHCDR1 is selected from sequence numbers 8-10. VHCDR2 is selected from sequence numbers 11-14. VHCDR3 is selected from sequence numbers 15-18. VLCDR1 as defined by sequence number 4, VLCDR2 as defined by AAS, VLCDR3 is selected from sequence numbers 19 to 21. The agent according to claim 2, and individually selected from the group comprising CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more.

4. The amino acid sequence of the CDR is i) VHCDR1, defined by sequence number 8, VHCDR2 as defined by sequence number 11, VHCDR3 as defined by sequence number 15, VLCDR1 as defined by sequence number 4, VLCDR2 as defined by AAS, and A binding protein having VLCDR3 as defined by Sequence ID No. 19, ii) VHCDR1, defined by sequence number 8, VHCDR2, defined by sequence number 12, VHCDR3 as defined by sequence number 16, VLCDR1 as defined by sequence number 4, VLCDR2 as defined by AAS, and A binding protein having VLCDR3 as defined by Sequence ID No. 20, iii) VHCDR1 as defined by sequence number 9, VHCDR2, defined by sequence number 13, VHCDR3 as defined by sequence number 17, VLCDR1 as defined by sequence number 4, VLCDR2 as defined by AAS, and A binding protein having VLCDR3 as defined by Sequence ID No. 20, iv) VHCDR1, defined by sequence number 10, VHCDR2, defined by sequence number 14, VHCDR3 as defined by sequence number 18, VLCDR1 as defined by sequence number 4, VLCDR2 as defined by AAS, and A binding protein having VLCDR3 as defined by Sequence ID No. 21, The agent according to claim 2 or 3, further comprising a group of CDR sequences having 95% or more identity with them, such as 96%, 97%, 98%, 99%, or more.

5. The VH sequence includes an amino acid sequence selected from the group consisting of SEQ ID NOs. 22 to 25, and a sequence having 80% or more identity with it, such as 85%, 90%, 95%, or more; and the VL sequence includes an amino acid sequence selected from the group consisting of SEQ ID NOs. 26 to 29, and a sequence having 80% or more identity with it, such as 85%, 90%, 95%, or more. The agent according to any one of claims 2 to 4, wherein the CDR sequence does not contain any mutations in the amino acid sequence, or the sequence mutations in the CDR amino acid sequence are 4%, 3%, 2%, 1%, or less, and are at most 5%.

6. The amino acid sequences of VH and VL are, i) VH as defined by Sequence ID 22, and A binding protein having VL as defined by Sequence ID No. 26, ii) VH as defined by Sequence ID 23, and A binding protein having VL as defined by Sequence ID No. 27, iii) VH as defined by sequence number 24, and A binding protein having VL as defined by Sequence ID No. 28, iv) VH as defined by sequence number 25, and A binding protein having VL as defined by Sequence ID No. 29, The agent according to any one of claims 2 to 5, further selected from the group comprising sequences having 80% or more identity with them, such as 85%, 90%, 95%, or more.

7. The binding protein is a monoclonal antibody such as IgG1 LALA antibody, or an Fv fragment such as an scFv fragment, Fab or F(ab') ２ The drug according to any one of claims 2 to 6, wherein the antigen-binding fragment is selected from the group consisting of Fab-like fragments such as fragments, disulfide-bonded fragments, and domain antibodies.

8. A cell manipulated to express a chimeric antigen receptor (CAR), wherein the CAR comprises an antigen-binding domain, a transmembrane domain connected to the antigen-binding domain by a hinge region, and an intracellular domain optionally connected to one or more costimulatory domains, wherein the antigen-binding domain contains the drug according to any one of claims 2 to 7, or a cell manipulated to express a CAR, wherein the antigen-binding domain is produced intracellularly and directed toward tumor blood vessels, and the antigen-binding domain contains the drug according to any one of claims 2 to 7.

9. A nucleic acid molecule encoding the drug according to any one of claims 1 to 7.

10. The nucleic acid molecule according to claim 9, wherein the nucleic acid molecule encodes a binding protein containing an antibody binding domain, the binding domain comprises a heavy chain variable domain (VH) and a light chain variable domain (VL), the VH coding sequence comprises a nucleotide sequence selected from the group consisting of SEQ ID NOs. 35 to 38 and a sequence having 80% or more identity thereto, such as 85%, 90%, 95%, or more, and the VL coding sequence comprises a nucleotide sequence selected from the group consisting of SEQ ID NOs. 39 to 42 and a sequence having 80% or more identity thereto, such as 85%, 90%, 95%, or more.

11. A vector, such as an expression vector, comprising the nucleic acid molecule described in claim 9 or 10.

12. A pharmaceutical composition comprising a drug according to any one of claims 1 to 7, a cell according to claim 8, or a nucleic acid molecule according to claim 9 or 10, and a pharmaceutically acceptable carrier or excipient.

13. A drug according to any one of claims 1 to 7, a cell according to claim 8, a nucleic acid molecule according to claim 9 or 10, or a pharmaceutical composition according to claim 12, for use in therapy.

14. The agent, cell, nucleic acid molecule, or pharmaceutical composition according to claim 13, for use in cancer therapies such as brain cancer therapies, including glioma and glioblastoma therapies.

15. A drug, cell, nucleic acid molecule, or pharmaceutical composition according to claim 13 or 14 for use in combination with one or more additional therapeutic regimens, wherein the additional therapeutic regimen is selected from i) treatment with an antitumor agent, ii) chemotherapy, iii) immunotherapy comprising a tumor-targeting antibody, iv) irradiation, and v) the use of a pathway blocker for treating cancer.

Images