CN116496404A

CN116496404A - Bispecific antibody targeting CD47 and PD-L1 and application thereof

Info

Publication number: CN116496404A
Application number: CN202310403258.5A
Authority: CN
Inventors: 白先宏; 孔毅荣; 李先钟; 刘叶; 张清俊; 高永红; 侯小双
Original assignee: Individual
Current assignee: Individual
Priority date: 2023-04-14
Filing date: 2023-04-14
Publication date: 2023-07-28

Abstract

The invention relates to the technical field of antibodies and the technical field of immunity, in particular to a bispecific antibody specifically bound to both CD47 and PD-L1 in a targeting manner, a coding nucleic acid, a composition and application thereof in diagnosis and treatment, and the anti-CD 47/anti-PD-L1 antibody has high affinity to CD47 and PD-L1 double-positive tumor cells, has obvious anti-tumor activity, does not have obvious erythrocytotoxicity, and can be used for preparing anti-tumor medicines.

Description

Bispecific antibody targeting CD47 and PD-L1 and application thereof

Technical Field

The present invention relates to the biomedical field, in particular to the field of antibody technology and the field of immunotechnology, more particularly to bispecific antibodies targeting specific binding to both CD47 and PD-L1, their encoding nucleic acids, compositions and their use for diagnosis and therapy.

Background

The CD47 protein is also called as Integrin Associated Protein (IAP), belongs to a five-time transmembrane glycoprotein of IgG superfamily, and is widely expressed in different tissue cells.

1. Mechanism of action of antibodies targeting CD47-SIRPa in anti-tumor therapy

Tumor cell evasion immune recognition and clearance is dependent on a variety of processes including induction of immunosuppressive tumor microenvironment and reduction of tumor cell immunogenicity. One key mechanism of tumor cell immune escape is through the overexpression of the immunosuppressive signaling molecule CD47.CD47 is widely regarded as a "do not eat me" signal that helps to maintain the immunological tolerance of non-malignant cells under physiological conditions, but this molecule can help cancer cells of different cancer types survive. Cancer cells utilize the "don't eat me" function of CD47, expressing higher levels of CD47 on the surface than non-malignant cells; CD47 can bind to the ligand TSP-1 or sirpa, regulating diverse cellular functions including cell migration, adhesion, apoptosis, axon extension, cytokine production and T cell activation. Sirpa is a transmembrane protein containing a typical Immunoreceptor Tyrosine Inhibitory Motif (ITIM), and is mainly expressed on the surface of hematopoietic cell membranes of the myeloid lineage, such as macrophages, dendritic cells, and the like. After the CD47 is combined with SIRPalpha, the phosphorylation of ITIMs is caused, so that SHP-1/SHP-2 is recruited, the accumulation of myosin IIA in phagocytic synapses is further inhibited, and the phagocytic function of phagocytes is finally inhibited. In many cancer types, CD47, which binds to signal-regulating protein α (sirpa), initiates an inhibitory signaling pathway, leading to malignant cells escaping phagocytosis by macrophages. Therapies that inhibit CD47 signaling in cancer cells may promote phagocytosis of tumor cells by macrophages, thereby limiting tumor growth. Numerous studies have shown that CD47 is overexpressed in different types of tumors, including myeloma, leiomyosarcoma, acute lymphoblastic leukemia, non-hodgkin's lymphoma, breast cancer, osteosarcoma, head and neck squamous cell carcinoma. High expression levels of CD47 are associated with therapeutic response and prognosis for cancer exacerbations. Therapies that inhibit CD47 signaling in cancer cells may promote phagocytosis of tumor cells by macrophages, thereby limiting tumor growth, which provides a viable immune target for anti-tumor therapy.

Since over-expression of CD47 on the surface of tumor cells can help these cells evade immune cell monitoring and clearance, CD47 is an attractive target for the development of new anti-tumor drugs, where anti-CD 47 antibodies can block CD47-sirpa inhibition signals and promote phagocytosis of tumor cells by macrophages, and can exert their anti-tumor effects through a variety of different mechanisms. Given the broad expression of CD47 and the ability of CD47-sirpa signaling to enable malignant cells to evade macrophage-mediated phagocytosis, inhibition of the CD47-sirpa signaling axis is a promising cancer treatment strategy.

Clinical progression of targeted CD47-SIRPa therapy

Currently, some CD 47-specific targeting antibodies or drugs have entered clinical trials, including Hu5F9-G4 (Forty-Seven), CC-90002 (Celgene), TTI-621 (Trillium), ALX148 (Alexo Therapeutics), SRF231 (Surface Oncology), SHR-1603 (Henry), IBI188 (Innovent Biologics), and the like.

2. Correlation of PD-L1 and CD47

PD-L1 and CD47 both exhibit high expression in tumor cells and can be synchronously regulated by MYC. MYC expressed by tumor cells can regulate tumor microenvironment by acting on congenital and acquired immune cells and cytokines, and activating MYC can up-regulate expression of CD47 and PD-L1 (MYC can directly act on promoters of CD47 and PD-L1 so as to regulate expression level of mRNA and protein thereof), thereby leading to immunosuppression and tumor proliferation

3. Problems of the prior art

CD47 is also expressed on healthy cells, such as Red Blood Cells (RBCs) and platelets in peripheral blood, CD47 drugs inevitably traumatize human erythrocytes while mobilizing macrophages to phagocytose tumor cells, and blocking CD 47-sirpa interactions by monoclonal antibodies has been shown to deplete those cells, in some cases causing dose limiting toxicity, thus, the primary CD47 drug clinically presents serious side effects including hemolytic anemia, CD47 mab therapy suffers from a range of biosafety problems.

Although anti-PD-L1 immunotherapy, which blocks the binding of PD-L1 on tumor or antigen presenting cells to PD-1 on T cells, has efficacy in patients with multiple types of cancer (e.g., NSCLC, RCC, HCC, HNSCC, lymphoma, merkel cell carcinoma (Merkel cellcarcinoma)), many patients remain unresponsive to treatment even in such anti-PD-L1 sensitive tumor types. In addition to the tumor types listed above, many other tumor types with low sensitivity to PD- (L) 1 are also enriched for CD47. In view of the various limitations of current anti-CD 47 antibodies and anti-PD-L1 antibody therapeutics, there is a need for improved bispecific molecules targeting CD47 and PD-L1, which can both stimulate macrophages to produce phagocytosis and block PD-1 signaling pathways, both mechanisms work together to enhance the killing power of tumor cells while reducing the possibility of binding to CD47 expressed on erythrocytes, thus new CD47/PD-L1 bispecific antibody products that reduce the toxicity of CD47 drugs become a new product in the field of tumor immunotherapy and its promising development. The dual antibody class of CD47 and other immune checkpoint targets is another important CD47 targeting drug improvement direction in addition to improving the erythrocyte binding specificity of CD47. There is still a need in the art of cancer therapy for CD47/PD-L1 bispecific antibodies with ever-improving properties.

Disclosure of Invention

The technical problem to be solved by the present invention is to overcome various limitations of current anti-CD 47 antibodies and anti-PD-L1 antibody monotherapy agents, and clinical drawbacks of binding CD47 on erythrocytes to cause erythrocyte hemolysis and blood coagulation in case of high clinical concentration administration of CD47 antibodies.

The invention provides a bispecific antibody or antigen binding fragment which has a novel structure, can simultaneously target and bind CD47 and PD-L1, has good specific binding capacity of CD47 and PD-L1 targets, can block the binding capacity of CD47 and SIRPa and the binding capacity of PD-L1 and PD-1, has stronger affinity with PD-L1 than that of CD47, can selectively bind with tumor cells but not human erythrocytes, cannot cause human erythrocytes to agglutinate, and has less side effects, and application thereof. In addition, the invention adopts a symmetrical IgG-like bispecific antibody structure containing a 'stem-in-mortar' (Knobs-into-holes) structure, realizes the correct assembly of different heavy chains and light chains, increases the yield of the bispecific antibody, is easy to stably express and purify in cultured cells in vitro, and does not need complex production process.

A first aspect of the invention provides a bispecific antibody or antigen binding fragment comprising a binding arm a specific for CD47 and a binding arm B specific for PD-L1, wherein the binding arm a comprises a light chain a (LA), a heavy chain a (HA) and a linker 1 (X1) connecting the light chain a (LA), heavy chain a (HA) derived from an anti-CD 47 antibody or antigen binding fragment and the binding arm B comprises a light chain B (LB), a heavy chain B (HB) derived from an anti-PD-L1 antibody or antigen binding fragment and a linker 2 (X2) connecting the light chain B (LB), heavy chain B (HB), wherein the heavy chain a (HA) and the heavy chain B (HB) interact to form a FIH structure, the binding arm a and the binding arm B forming a heterodimer by the KIH structure. In some embodiments, in the binding arm a, the heavy chain a (HA) comprises a heavy chain variable region a (VHA) and the light chain a comprises a light chain variable region a (VLA); wherein, the liquid crystal display device comprises a liquid crystal display device,

(1) The VHA comprises or is selected from (A1), (A2) and (A3):

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs 22 to 28;

(A2) An amino acid sequence obtained by substituting, deleting or adding one or more amino acids in the amino acid sequence shown in (B1) and functionally identical or similar to the amino acid sequence shown in (B1); and

(A3) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (A1).

In some embodiments, the binding arm a, the heavy chain a comprises a heavy chain variable region a (VHA) and the light chain a comprises a light chain variable region a (VLA), wherein,

(2) The VLA comprises or is selected from (A4), (A5) and (A6)

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 57 to 63;

(A5) An amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids to the amino acid sequence shown in (A4) and has the same or similar function as the amino acid sequence shown in (B4); and

(A6) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (A4).

In some embodiments, a bispecific antibody according to the invention comprises a binding arm a specific for CD47 and a binding arm B specific for PD-L1, the anti-binding arm a comprising a heavy chain variable region a and a light chain variable region a, wherein:

In some embodiments, the PD-L1 specific binding arm B comprises a heavy chain variable region B (VHB) and a light chain variable region B (VLB); wherein the VHB comprises or is selected from:

(B1) An amino acid sequence as shown in SEQ ID NO. 85;

(B2) An amino acid sequence obtained by substituting, deleting or adding one or more amino acids in the amino acid sequence shown in (B1) and functionally identical or similar to the amino acid sequence shown in (B1); and

(B3) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (B1); and/or

The VLB comprises or is selected from:

(B4) An amino acid sequence as shown in SEQ ID NO. 86;

(B5) An amino acid sequence obtained by substituting, deleting or adding one or more amino acids in the amino acid sequence shown in (B4) and functionally identical or similar to the amino acid sequence shown in (B4); and

(B6) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (B4).

In some embodiments, the bispecific antibody comprises a CD47 specific binding arm a comprising a heavy chain variable region a (VHA) and a light chain variable region a (VLA), and a PD-L1 specific binding arm B, wherein:

the VHA comprises an amino acid sequence shown as SEQ ID NO. 22;

The VLA comprises an amino acid sequence as set forth in any one of SEQ ID NOs 57;

the PD-L1 specific binding arm B comprises a heavy chain variable region B (VHB) and a light chain variable region B (VLB); wherein the VHB comprises an amino acid sequence shown as SEQ ID NO. 85; the VLB comprises the amino acid sequence as shown in SEQ ID NO. 86.

In some embodiments, the bispecific antibody, the heavy chain A (HA) comprises an amino acid sequence as set forth in any one of SEQ ID NOS: 29-35, or an amino acid sequence having at least 90% homology with an amino acid sequence set forth in any one of SEQ ID NOS: 29-35; the light chain A (LA) comprises an amino acid sequence as set forth in any one of SEQ ID NOs 64-70, or an amino acid sequence that is at least 90% homologous to an amino acid sequence set forth in any one of SEQ ID NOs 64-70; the heavy chain B (HB) comprises an amino acid sequence shown as SEQ ID NO. 87 or an amino acid sequence having at least 90% homology with the amino acid sequence shown as SEQ ID NO. 87; the light chain B (LB) comprises an amino acid sequence shown as SEQ ID NO. 88 or an amino acid sequence with at least 80% homology with the amino acid sequence shown as SEQ ID NO. 154.

In one embodiment, the CD 47-specific binding arm a comprises SEQ ID NO:29 and a light chain of the amino acid sequence shown in SEQ ID NO. 64; and/or

The PD-L1 specifically binds to arm B comprising SEQ ID NO:87 and the heavy chain of the amino acid sequence shown in SEQ ID NO:88, and a light chain of the amino acid sequence shown in seq id no.

In some embodiments, wherein the heavy chain A comprises a heavy chain variable region A (VHA) and the light chain A comprises a light chain variable region A (VLA),

wherein the heavy chain variable region a (VHA) comprises one or more amino acid sequences of (a) - (d):

(a) A VHA CDR1 comprising an amino acid sequence as set forth in any one of SEQ ID NOs 1 to 7 or having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in any one of SEQ ID NOs 1 to 7;

(b) A VHA CDR2 comprising an amino acid sequence as set forth in any one of SEQ ID NOS.8-14 or having one or more conservative amino acid substitutions in comparison to the amino acid sequence set forth in any one of SEQ ID NOS.8-14;

(c) VHA CDR3 comprising an amino acid sequence as set forth in SEQ ID NOS.15-21 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NOS.15-21;

(d) An amino acid sequence having at least 85% sequence identity to an amino acid sequence set forth in any one of (a) - (c); and/or

The light chain variable region a (VLA) comprises one or more of the amino acid sequences of (e) - (h):

(e) VLA CDR1 comprising an amino acid sequence as set forth in any one of SEQ ID NOs 36 to 42;

(f) VLA CDR2 comprising an amino acid sequence as depicted in ATS;

(g) VLA CDR3 comprising an amino acid sequence as set forth in any one of SEQ ID NOs 50 to 56;

(h) An amino acid sequence having at least 85% sequence identity to an amino acid sequence set forth in any one of (e) - (g).

In some embodiments, the binding arm a comprises a heavy chain variable region a (VHA) comprising VHA CDR1, VHA CDR2, VHA CDR3; wherein VHA CDR1 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 1 to 7; VHA CDR2 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 8 to 14; VHA CDR3 comprises the amino acid sequence shown in SEQ ID NO 15-21.

In some embodiments, the binding arm a comprises a light chain variable region a (VLA) comprising VLA CDR1, VLA CDR2, and VLA CDR3; wherein VLA CDR1 comprises an amino acid sequence as set forth in any one of SEQ ID NOS: 36-42; VLA CDR2 comprises the amino acid sequence as shown by ATS; VLA CDR3 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 50 to 56.

In some embodiments, the binding arm a comprises VHA and VLA; the VHA comprises a VHA CDR1, a VHA CDR2, a VHA CDR3; the VLA comprises VLA CDR1, VLA CDR2 and VLA CDR3. Wherein VHA CDR1 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 1 to 7; VHA CDR2 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 8 to 14; VHA CDR3 comprises the amino acid sequence shown in SEQ ID NO 15-21; VLA CDR1 comprising an amino acid sequence as set forth in any one of SEQ ID NOS: 36-42; VLA CDR2 comprises the amino acid sequence as shown by ATS; VLA CDR3 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 50 to 56.

In one embodiment, the bispecific antibody of the present invention, wherein the binding arm B that specifically binds PD-L1 comprises a heavy chain variable region B (VHB) and a light chain variable region B (VLB),

wherein the VHB comprises one or more CDRs of amino acid sequences (i) - (l):

(i) A VHB CDR1 comprising the amino acid sequence as set forth in SEQ ID NO:80 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID No. 80;

(j) A VHB CDR2 comprising an amino acid sequence as set forth in SEQ ID NO. 81 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 81;

(k) A VHB CDR3 comprising an amino acid sequence as set forth in SEQ ID NO. 82 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 82;

(l) An amino acid sequence having at least 85% sequence identity to an amino acid sequence set forth in any one of (i) - (k); and/or

In one embodiment, the bispecific antibody of the present invention, the binding arm B that specifically binds PD-L1 comprises a heavy chain variable region VHB and a light chain variable region VLB, wherein the VLB comprises one or more of the CDRs amino acid sequences (m) - (p):

(m) a VLB CDR1 comprising the amino acid sequence as shown in SEQ ID NO. 83 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 83;

(n) VLB CDR2 comprising an amino acid sequence as depicted in SAS;

(o) a VLB CDR3 comprising the amino acid sequence as set forth in SEQ ID NO. 84 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 84;

(p) an amino acid sequence having at least 85% sequence identity to an amino acid sequence set forth in any one of (m) - (o).

In some embodiments, the binding arm B comprises the VHBCDR1, VHBCDR2, VHB CDR3, VLB CDR1, VLB CDR2, and VLB CDR3.

In one embodiment, the bispecific antibody or antigen binding fragment of the invention, wherein the binding arm A further comprises a heavy chain constant region A (CHA) comprising an amino acid sequence as shown in SEQ ID NO. 75 or an amino acid sequence having not less than 80%, or not less than 85%, or not less than 90%, or not less than 95%, or not less than 98%, or not less than 99.5%, or not less than 99.8% sequence identity, and a light chain constant region A (CLA) comprising an amino acid sequence as shown in SEQ ID NO. 73 or an amino acid sequence having not less than 80%, or not less than 85%, or not less than 90%, or not less than 95%, or not less than 98%, or not less than 99.5%, or not less than 99.8% sequence identity, and wherein the binding arm A further comprises an amino acid sequence as shown in SEQ ID NO. 73 or an amino acid sequence having not less than 80%, or not less than 85%, or not less than 90%, or not less than 98%, or not less than 90%, or not less than 98%, or not less than 90% sequence identity to the amino acid sequence shown in SEQ ID NO. 73, or not less than 99.5%, or not less than 99.8%, of the light chain constant region A (CLA) comprises an amino acid sequence as shown in SEQ ID NO. 75% or an amino acid sequence having not less than 80%, or not less than 90%, or not less than 95%, or not less than 98%, or not less than 99.5%, or not less than 99.8% of the amino acid sequence of the sequence identity.

In one embodiment, the bispecific antibody or antigen-binding fragment of the present invention, the binding arm A further comprises a heavy chain constant region A (CHA) as shown in SEQ ID NO. 75 and a light chain constant region (CLA) as shown in SEQ ID NO. 73, and the binding arm B comprises a heavy chain constant region B (CHB) as shown in SEQ ID NO. 90 and a light chain constant region B (CLB) as shown in SEQ ID NO. 89, respectively.

In one aspect, the invention provides a bispecific antibody or antigen-binding fragment comprising a binding arm a specific for CD47 and a binding arm B specific for PD-L1, wherein the binding arm a comprises a heavy chain variable region a (VHA) and a light chain variable region (VLA) and a linker 1 (X1) located between the heavy chain variable region a (VHA) and the light chain variable region (VLA), the binding arm B comprising a heavy chain variable region B (VHB) and a light chain variable region B (VLB) and a linker 2 (X2) located between the heavy chain variable region B (VHB) and the light chain variable region B (VLB); wherein the heavy chain variable region a (VHA) and the light chain variable region (VLA) pair to form a CD47 antigen binding site and the heavy chain variable region B (VHB) and the light chain variable region B (VLB) pair to form a PD-L1 antigen binding site.

In one aspect, the invention provides a bispecific antibody or antigen-binding fragment comprising a binding arm a that specifically binds CD47 and a binding arm B that specifically binds PD-L1; wherein, the liquid crystal display device comprises a liquid crystal display device,

The binding arm A comprises a scFab (A) comprising a light chain variable region A (VLA), a light chain constant region A (CLA), a heavy chain variable region A (VHA) and an Fc (A) derived from an anti-CD 47 antibodyFirst heavy chain constant region A(CH 1A), and a linker 1 (X1) connecting the light chain constant region a (CLA) and the heavy chain variable region a (VHA), the scFab (a) comprising an amino acid sequence as shown in SEQ ID No. 77, or an amino acid sequence having NO less than 80% sequence identity to the amino acid sequence shown in SEQ ID No. 77; and/or the number of the groups of groups,

the binding arm B comprises scFab (B) and Fc (B), the scFab (B) comprises a light chain variable region B (VLB), a light chain constant region B (CLB) and a heavy chain variable region of an anti-PD-L antibodyVariable region B (VHB)First heavy chain constant region B(CH 1B), and a linker 2 (X2) connecting the light chain constant region B (CLB) and the heavy chain variable region B (VHB), the scFab (B) comprising an amino acid sequence as shown in SEQ ID No. 93, or an amino acid sequence having NO less than 80% sequence identity to the amino acid sequence shown in SEQ ID No. 93.

In one embodiment, the scFab (a) and Fc (a) are fused to a single chain peptide chain a that specifically binds CD 47; the scFab (B) and Fc (B) are fused to a single chain peptide chain B that specifically binds PD-L1.

Wherein either the heavy chain A or the heavy chain B is subjected to amino acid modification to form a knob structure, the other is subjected to amino acid modification to form a hole structure, a KIH structure is formed between the knob and the hole by the CD47 specific binding arm A and the PD-L1 specific binding arm B, and a heterodimer is formed between the knob and the hole by the CD47 specific binding arm A and the PD-L1 specific binding arm B;

in one embodiment, wherein the Fc (a) comprises the amino acid sequence set forth in SEQ ID NO:78 or an amino acid sequence having not less than 80% sequence identity to the amino acid sequence set forth in SEQ ID No. 78, said Fc (B) comprising the amino acid sequence set forth in SEQ ID NO:94 or an amino acid sequence having not less than 80% sequence identity to the amino acid sequence shown in SEQ ID NO. 94.

In one embodiment, the Fc (a) is amino acid modified to form a hole structure, the Fc (B) is amino acid modified to form a hole structure, and the formation of KIH between the knob and the hole promotes heterodimer formation between the binding arm a and the binding arm B.

In one embodiment, the Fc (B) is amino acid modified to form a knob structure, the Fc (a) is amino acid modified to form a hole structure, and a KIH structure formed between the knob and the hole promotes heterodimer formation between the binding arm a and the binding arm B.

In one embodiment, the CD47 binding arm comprises a heavy chain constant region CHA as set forth in SEQ ID NO. 90, said CHA comprising an Fc (A) as set forth in SEQ ID NO. 78, said Fc (A) comprising an amino acid modification forming a hole structure in the third heavy chain constant region A (CH 3A).

In one embodiment, the PD-L1 binding arm comprises the heavy chain constant region CHB shown in SEQ ID NO. 90, said CHB comprising the Fc (B) shown in SEQ ID NO. 94, said Fc (B) comprising an amino acid modification forming a knob structure in the third heavy chain constant region B (CH 3B).

In one embodiment, the hole-forming amino acid modification and the knob-forming amino acid modification include one, two or more of S354C and T366W and Y349C, T366S, L A and Y407V.

In one embodiment, the hole-forming amino acid modifications include S354C and T366W; the knob-forming amino acid modifications include one, two or more of Y349C, T366S, L368A and Y407V.

In one aspect, the invention provides a bispecific antibody or antigen binding fragment comprising a binding arm a that specifically binds CD47 and a binding arm B that specifically binds PD-L1, wherein the binding arm a that specifically binds CD47 comprises a light chain variable region a (VLA) derived from an anti-CD 47 antibody or variant thereof, a light chain constant region a (CLA) derived from an anti-CD 47 antibody or variant thereof, a heavy chain variable region a (VHA) derived from an anti-CD 47 antibody or variant thereof, and a heavy chain constant region a (CHA) derived from an anti-CD 47 antibody or variant thereof; the antigen binding arm that specifically binds CD47 is N-terminal to C-terminal comprising the structure shown by the formula VLA-CLA-X1-VHA-CHA,

Wherein the "-" is a peptide bond,

x1 is a joint structure; the X1 comprises an amino acid sequence shown as SEQ ID NO. 74;

the binding arm A which specifically binds CD47 comprises the amino acid sequence shown as SEQ ID NO:79 or comprises an amino acid sequence having more than 80%, preferably more than 90%, more preferably more than 95%, more preferably more than 98.5% or most preferably more than 99.5% sequence homology with the amino acid sequence shown as SEQ ID NO: 79; the CLA comprises the amino acid sequence as shown in SEQ ID NO. 73, or comprises an amino sequence having a sequence homology of more than 80%, preferably more than 90%, more preferably more than 95%, more preferably more than 98.5% and most preferably more than 99.5% with the amino acid sequence shown in SEQ ID NO. 73; the CHA comprises an amino acid sequence as set forth in SEQ ID No. 75, or comprises an amino acid sequence having greater than 80%, preferably greater than 90%, more preferably greater than 95%, more preferably greater than 98.5%, or most preferably greater than 99.5% sequence homology with the amino acid sequence set forth in SEQ ID No. 75;

and/or

The binding arm B that specifically binds PD-L1 comprises a light chain variable region B (VLB) derived from an anti-PD-L1 antibody or variant thereof, a light chain constant region B (VLB) derived from an anti-PD-L1 antibody or variant thereof, a heavy chain variable region B (VHB) derived from an anti-PD-L1 antibody or variant thereof, and a heavy chain constant region B (CHB) derived from an anti-PD-L1 antibody or variant thereof; the antigen binding arm B which specifically binds to PD-L1 is a structure from N end to C end and comprises VLB-CLB-X2-VHB-CHB,

Wherein "-" is a peptide bond and X2 is a linker structure; x2 is shown as SEQ ID NO. 91;

the PD-L1 targeting antigen binding arm B comprises an amino acid sequence as shown in SEQ ID NO. 95 or an amino acid sequence having more than 80%, preferably more than 90%, more preferably more than 95%, more preferably more than 98.5% or most preferably more than 99.5% sequence homology with the amino acid sequence shown in SEQ ID NO. 95;

wherein the CLB comprises an amino acid sequence as set forth in SEQ ID No. 89 or comprises an amino acid sequence having greater than 80%, preferably greater than 90%, more preferably greater than 95%, more preferably greater than 98.5% or preferably greater than 99.5% sequence homology with the amino acid sequence set forth in SEQ ID No. 89;

the CHB comprises the amino acid sequence shown as SEQ ID No. 90 or comprises an amino sequence having greater than 80%, preferably greater than 90%, more preferably greater than 95%, more preferably greater than 98.5% or most preferably greater than 99.5% sequence homology with the amino acid sequence shown as SEQ ID No. 90.

In some embodiments, in a bispecific antibody or antigen binding fragment of the invention, the light chain constant region a (CLA) of the anti-CD 47 antibody and the light chain constant region B (CLB) of the anti-PD-L1 antibody are optionally kappa chain or lambda chain constant regions, and the CHA and the CHB are optionally IgG1 or IgG4 constant regions. In some embodiments, the anti-CD 47 and PD-L1 bispecific antibodies of the invention comprise a light chain constant region of murine or human kappa, lambda chains or variants thereof. In some embodiments, the anti-CD 47 and PD-L1 bispecific antibodies of the invention comprise a heavy chain constant region comprising murine or human IgG1, igG2, igG3 or IgG4 or variants thereof.

In some preferred embodiments, the anti-CD 47 and PD-L1 bispecific antibodies of the present invention are humanized antibodies, wherein the heavy chain comprises a heavy chain constant region of a human IgG1, igG2, igG3, igG4 or variant thereof, and the light chain comprises a light chain constant region of a human kappa, lambda chain or variant thereof.

In some exemplary embodiments, the light chain constant region CLA of the anti-CD 47 antibody and CLB of the anti-PD-L1 antibody are optionally human antibody kappa chain or lambda chain constant regions, and the heavy chain constant region CHA and the CHB are optionally human antibody IgG1 or IgG4 constant regions.

In some preferred embodiments, the anti-CD 47 and PD-L1 bispecific antibodies of the invention are humanized antibodies, wherein the heavy chain comprises a heavy chain constant region of an optional human IgG1, igG2, igG3, igG4 or variant thereof, and the light chain comprises a light chain constant region of an optional human kappa, lambda chain or variant thereof.

In an exemplary embodiment, the anti-CD 47 and PD-L1 bispecific antibodies of the invention are humanized antibodies, wherein the heavy chain comprises an optional human IgG1 heavy chain constant region and the light chain comprises a light chain constant region of an optional human kappa or variant thereof.

In an exemplary embodiment, the anti-CD 47 and PD-L1 bispecific antibodies of the invention are humanized antibodies, wherein the heavy chain comprises an optional human IgG1 heavy chain constant region and the light chain comprises a light chain constant region of an optional human λ or variant thereof.

In some embodiments, the antigen binding fragment is selected from a Fab, fab ' -SH, fv, scFv, or (Fab ') 2 fragment, or the antigen binding fragments are each selected from Fab and (Fab ') 2, scFab.

In yet another aspect, the invention provides an isolated polynucleotide encoding a humanized CD47 antibody, bispecific antibody or antigen binding fragment of the preceding aspect, comprising: (1) a polynucleotide encoding a heavy or light chain that binds arm a; (2) a polynucleotide encoding a heavy or light chain that binds arm B; (3) A polynucleotide encoding a heavy chain variable region a (VHA) or a light chain variable region a (VLA) of binding arm a; or (4) a polynucleotide encoding a seed heavy chain variable region B (VHB) or a light chain variable region B (VLB) that binds to arm B; (5) a polynucleotide encoding a knob structure; (6) a polynucleotide encoding a hole structure.

In some embodiments, the invention provides an isolated polynucleotide encoding a humanized CD47 antibody, bispecific antibody or antigen binding fragment of the foregoing aspects, comprising: (1) encodes an amino acid sequence as set forth in SEQ ID NO: a nucleotide sequence of the light chain of binding arm a of any one of claims 64-70 or a variant having the same function with at least 90% sequence homology thereto; (2) A nucleotide sequence encoding the heavy chain of binding arm a as set forth in any one of SEQ ID NOs 29-35 or a variant having the same function with at least 90% sequence homology thereto; (3) encodes an amino acid sequence as set forth in SEQ ID NO:79 or a variant having at least 90% sequence homology thereto: (4) encodes an amino acid sequence as set forth in SEQ ID NO:95 or a variant having at least 90% sequence homology thereto; (5) A nucleotide sequence encoding scFab (A) as shown in SEQ ID NO. 77 or a variant thereof having at least 90% sequence homology thereto; (6) encodes an amino acid sequence as set forth in SEQ ID NO:93, a nucleotide sequence of scFab (B); (7) encodes an amino acid sequence as set forth in SEQ I NO:57-63 or a variant having at least 90% sequence homology thereto; (8) encodes an amino acid sequence as set forth in SEQ I NO:22-28 or a variant having the same function that is at least 90% sequence homology thereto;

Or (b)

The polynucleotide comprises: (1) A nucleotide sequence encoding a light chain a (LA) of binding arm a as set forth in SEQ ID No. 29 or a variant having the same function with at least 90% sequence homology thereto, (2) a nucleotide sequence encoding a heavy chain a (HA) of binding arm a as set forth in SEQ ID No. 64 or a variant having the same function with at least 90% sequence homology thereto; (3) encodes an amino acid sequence as set forth in SEQ ID NO:88 or a variant having at least 90% sequence homology thereto, (4) a nucleotide sequence encoding a light chain B (LB) of binding arm B as set forth in SEQ ID NO:87 or a variant having the same function that binds to heavy chain B (HB) of arm B or at least 90% sequence homology thereto; (3) the sequence shown in SEQ ID NO:106 encodes a polypeptide as set forth in SEQ ID NO:179 or a variant having at least 90% sequence homology thereto: (4) the sequence shown in SEQ ID NO:107 encodes a polypeptide as set forth in SEQ ID NO:95 or a variant having at least 90% sequence homology thereto; (5) the sequence shown in SEQ ID NO:77 or a variant having at least 90% sequence homology thereto; (6) encodes an amino acid sequence as set forth in SEQ ID NO:93 or a variant having at least 90% sequence homology thereto; (7) the sequence shown in SEQ ID NO:96 encodes the sequence set forth in SEQ I NO:57 or a variant having at least 90% sequence homology thereto; (8) the sequence shown in SEQ ID NO:99 or a sequence having homology of not less than 90% thereto, which encodes a polypeptide as set forth in SEQ I NO:22 or a variant having the same function which encodes a heavy chain variable region a (VHA) or at least 90% sequence homology thereto.

In a further aspect, the invention provides an expression vector comprising a polynucleotide encoding an antibody or antigen-binding fragment according to any one of the preceding aspects; (7) A nucleotide sequence of a nucleotide sequence shown as SEQ ID NO. 101 or having homology of not less than 90% thereto encoding an amino acid sequence shown as SEQ ID NO. 86 or a light chain variable region B (VHB) or a variant having at least 90% sequence homology thereto; (8) A nucleotide sequence encoding a heavy chain variable region B (VHB) as shown in SEQ ID NO. 85 or a variant having the same function with at least 90% sequence homology thereto as shown in SEQ ID NO. 104 or a nucleotide sequence having not less than 90% homology thereto. .

In a further aspect, the invention provides a recombinant production cell comprising a polynucleotide encoding an antibody or antigen-binding fragment as described herein before, or an expression vector as described herein before, for producing a bispecific antibody according to the invention.

In yet another aspect, the invention provides a method of producing a bispecific antibody or antigen-binding fragment comprising specifically binding CD47 and specifically binding PD-L1 as described herein before, comprising:

culturing the host cell according to claim under conditions that cause the production of the bispecific antibody described above, and purifying the bispecific antibody produced.

In yet another aspect, the invention provides a composition comprising a bispecific antibody or antigen binding fragment, polynucleotide, expression vector or cell according to the preceding aspects of the invention, and a pharmaceutically acceptable carrier.

In a further aspect, the invention provides the use of a bispecific antibody or antigen binding fragment, a polynucleotide, an expression vector, a cell and a composition as described in any of the preceding aspects for the preparation of a medicament for the treatment of a disease associated with simultaneous overexpression of CD47 and PD-L1.

In some embodiments, the disease associated with simultaneous overexpression of CD47 and PD-L1 comprises or is selected from autoimmune diseases, acute and chronic inflammatory diseases, infectious diseases, cancer.

In a further aspect, the present invention provides a method of treating a disease, comprising administering to a patient in need of treatment an effective amount of a bispecific antibody or antigen binding fragment as defined above, a polynucleotide as defined above, or an expression vector as defined above, a cell as defined above, or a composition as defined above, wherein, preferably, the disease is a disease associated with simultaneous overexpression of CD47 and PD-L1

In a further aspect, the invention relates to a kit or article of manufacture comprising a bispecific antibody or antigen binding fragment as described above, a polynucleotide as described above or an expression vector as described above.

In yet another aspect, the invention provides bispecific antibodies or antigen-binding fragments that bind to CD47 and PD-L1 antigens on the surface of tumor cells, and uses thereof. The antibodies or antigen binding fragments of the invention may be used in the treatment or prevention of various diseases, such as autoimmune diseases, acute and chronic inflammatory diseases, infectious diseases, cancer, and in the diagnosis and prognosis of related diseases.

In some embodiments, provided herein is a method of inhibiting or reducing cancer cell growth in an individual, the method comprising administering to the individual an effective amount of an antibody, bispecific antibody, or pharmaceutical composition provided herein.

In some embodiments, bispecific antibodies provided herein have a synergistic effect compared to the same amount administered between Ab-CD47 and PD-L1 mab combination.

In some embodiments, an antibody, bispecific antibody, or pharmaceutical composition provided herein is used in a method of treatment provided herein. In some embodiments, the antibodies, bispecific antibodies, or pharmaceutical compositions provided herein are for use in treating cancer and/or enhancing an immune response in an individual.

In some embodiments, the antibodies, bispecific antibodies, or pharmaceutical compositions provided herein are used in the manufacture of a medicament for use in the methods of treatment provided herein. In some embodiments, the agent is used to treat cancer in an individual and/or enhance an immune response in an individual.

In some embodiments, the disease is a disorder associated with overexpression of PD-L1 and CD 47. In some embodiments, the disease is an autoimmune disease, acute and chronic inflammatory disease, infectious disease (e.g., chronic infectious disease or sepsis), cancer.

In some embodiments, the disease is a disorder associated with PD-L1 and CD47 including, but not limited to, various hematological disorders and solid tumors, such as in one embodiment, the cancer is a gastrointestinal cancer, such as colon cancer.

Wherein the tumor condition comprises at least one of a hematological tumor, a solid tumor condition, the hematological tumor being at least one of acute myelogenous leukemia, chronic myelogenous leukemia, acute lymphoblastic leukemia, chronic lymphoblastic leukemia, adult T-cell leukemia, multiple myeloma, mixed leukemia, non-hodgkin's lymphoma; the solid tumor is at least one of lymphoma, breast cancer, head and neck cancer, gastric cancer, lung cancer, esophageal cancer, intestinal cancer, ovarian cancer, cervical cancer, liver cancer, renal cancer, pancreatic cancer, bladder cancer, colorectal cancer, glioma and melanoma.

Advantageous effects

The present invention provides humanized CD47 antibodies that bind specifically to CD47 simultaneously and bispecific antibodies that bind CD47 and PD-L1 simultaneously and antigen binding fragments, variants thereof. The bispecific antibody molecule produces the following advantageous functions:

(1) Based on the mechanism of the blocking of the double immune checkpoints of CD47 and PD-L1, the CD47 and the PD-L1 can be targeted and combined with high affinity;

(2) The bispecific antibody can combine human CD47 protein and simultaneously block the combination of CD47 and SIRPalpha, and can combine human PD-L1 and simultaneously block the combination of PD-L1 and PD1, thereby improving the inhibition effect on tumors;

(3) Has an antibody-mediated ADCC effect of not less than that of anti-CD 47 mab and anti-PD-L1 mab;

(4) The bispecific antibody has a synergistic anti-tumor effect in inhibiting the proliferation of tumor cells, and has a better synergistic anti-tumor therapeutic effect than that of single anti-CD 47 monoclonal antibody and PD-L monoclonal antibody and combined drug, so that the therapeutic effect of tumor simultaneously expressing CD47 and PD-L1 can be improved;

(5) According to the preparation process of the bispecific antibody, a plurality of pairs of disulfide bonds are formed between KIH and CH1 and CH2 of two binding arms, so that mismatching of two targeted light chains and heavy chains is effectively prevented, and the yield of the target CD47/PD-L1 bispecific antibody can be effectively improved;

(6) The bispecific antibody capable of simultaneously and targeting to combine with CD47 and PD-L1 has lower economic cost and better oral compliance of patients in anti-tumor treatment than the monoclonal antibody combined drug using a single target, and compared with the CD47 monoclonal antibody, the bispecific antibody can avoid the side effects of combining with erythrocytes and causing erythrocyte aggregation, and has better safety;

(7) The humanized CD47 antibody and the bispecific antibody capable of simultaneously and targeting the combination of CD47 and PD-L1 have species cross-binding activity with the cynomolgus monkey CD47 protein, have no species cross-binding activity with the rat CD47 protein and the mouse CD47 protein, and have species cross-binding activity with the rat, mouse and cynomolgus monkey PD-L1 protein.

(8) Obviously inhibit the growth of tumor, has no obvious erythrocyte toxicity, greatly reduces the immunogenicity of humanized antibody, effectively eliminates the rejection reaction of human immune system to exogenous monoclonal antibody, can be applied to the preparation of medicaments for treating various tumor diseases, and has wide market prospect.

Drawings

FIG. 1 example 1.1CD47 murine anti-binding CD47-His protein affinity assay results

FIG. 2 analysis of the data from the ELISA assay for determining CD47 humanized antibody binding to CD47 in example 1.3.1ELISA (logarithmic antibody concentration (ng/ml) on the abscissa and average antibody determination OD450 on the ordinate).

FIG. 3 analytical curve for determination of CD47 humanized antibody blocking CD47 binding to SIRP-alpha in example 1.3.2ELISA (logarithmic value of antibody concentration (ng/ml) on the abscissa and average value of antibody determination OD450 on the ordinate).

FIG. 4 shows the structure of PD-L1/CD47 of the invention. FIG. 5, binding dissociation curve fitting results of binding kinetics assays for the CD47/PDL1 bispecific antibody BsAb-463 against recombinant human CD47-His affinity were determined in example 3.1ForteBio's Octet System.

FIG. 6, binding dissociation curve fitting results of binding kinetics assays for determining the affinity of the CD47/PDL1 bispecific antibody BsAb-463 to hPD-L1-His protein are determined in example 3.2ForteBio's Octet System.

FIG. 7-analysis of binding of CD47/PD-L1 bispecific antibody BsAb-463 to CHO cells overexpressing hPD-L1 by FACS method of example 4.1.

FIG. 8A block activity assay for blocking binding of PD-L1 to PD1 on a CD47/PDL1 bispecific antibody BsAb-463 block CHO-K1-PD-L1 cell membrane by the FACS method of example 4.2.

FIG. 9, analytical curve for determining binding activity of the CD47/PDL1 bispecific antibody BsAb-463 to Raji cell surface CD47 by FACS method of example 5.1.

FIG. 10 analytical curve for the FACS determination of the blocking activity of CD47/PDL1 bispecific antibody to block the binding of CD47 to SIRPa on Raji cells, example 5.2.

FIG. 11 is an analytical curve of ELISA assay for the cross-binding activity of different antibodies to Cyno-CD 47.

FIG. 12 is an analytical curve of ELISA assay for the cross-binding activity of different antibodies to Mus-CD 47.

FIG. 13 is an analytical curve of ELISA assay for cross-binding activity of different antibodies to Rat-CD 47.

FIG. 14-ELISA-assay curves for detecting the cross-binding activity of different antibodies to Cyno-PD-L1.

FIG. 15 assay curves for ELISA detection of cross-binding activity of different antibodies to Mus-PD-L1.

FIG. 16 ELISA-binding assay for Cross-binding Activity of different antibodies to Rat-PD-L1

FIG. 17 shows the mean fluorescence intensity analysis curve of binding activity of different antibodies to Raji-hPDL1-GFP biscationic cells by FACS method.

FIG. 18A Mean Fluorescence Intensity (MFI) analysis curve for the determination of the binding activity of different antibodies to erythrocytes by the FACS method of example 8.

FIG. 19 detection of human hemagglutination by different antibodies by FACS method of example 9.

FIG. 20 shows the binding Mean Fluorescence Intensity (MFI) analysis of the binding activity of different antibodies to tumor cells in a mixture of Raji-hPDL1-GFP tumor cells and human erythrocytes.

FIG. 21 average fluorescence intensity of binding of different antibodies of example 10 to erythrocytes in Raji-hPDL1-GFP tumor cells and human erythrocytes mixed cells.

Fig. 22 shows the results of the analysis of ADCC effect by the diabodies.

FIG. 23 shows the ADCC effect of BsAb-463 antibody on Raji-PD-L1 cells and HCC827 cells.

FIG. 24 inhibition of the Raji/PD-L1 cytoma model by BsAb-463 antibody following subcutaneous transplantation of CB17/SCID mice.

Detailed Description

Provided herein are murine chimeric anti-CD 47 monoclonal antibodies that specifically bind to CD47, human anti-CD 47 monoclonal antibodies, and bispecific antibodies that specifically bind to CD47 and PD-L1. Related nucleic acids, compositions, and methods of making and using antibodies are also provided herein.

The practice of the present invention will employ, unless otherwise indicated, conventional techniques of molecular biology (including recombinant techniques), microbiology, cell biology, biochemistry and immunology, which are well within the skill of the art. Such techniques are fully explained in documents such as: molecular Cloning: a Laboratory Manual, second edition (Sambrook et al 1989) Cold Spring Harbor Press; oligonucleotide Synthesis (M.J.Gait et al 1984); methods in Molecular Biology, humana Press; cell Biology: ALabatosryNotebook (J.E.Cellis 1998) Academic Press; and, where applicable, in subsequent versions of the above references and corresponding websites.

The invention will now be described in detail by reference to the following definitions and examples. All patents and publications, including all sequences disclosed within such patents and publications, referred to herein are expressly incorporated by reference.

Unless otherwise indicated, each term below shall have the meaning described below. Unless otherwise indicated herein, numbering of amino acid residues in the constant region is according to the EU numbering system as described in Kabat et al, sequences of Proteins of Immunological Interes, 5 th edition, public Health Service, national Institutes ofHealth, bethesda, MD, 1991.

Definition of the definition

It should be noted that the term "an" entity refers to one or more of the entity, e.g. "an antibody" should be understood as one or more antibodies, and thus the terms "one" (or "one"), "one or more" and "at least one" can be used interchangeably herein.

"about" or "approximately" refers to a conventional error range of corresponding numerical values as readily known to one of ordinary skill in the relevant art. In some embodiments, references herein to "about" or "approximately" refer to the values recited and ranges thereof of + -10%, + -5%, + -1%, or + -0.1%.

The term "polypeptide" is intended to encompass both the singular and the plural of "polypeptides" and refers to a molecule composed of amino acid monomers linearly linked by amide bonds (also referred to as peptide bonds). The term "polypeptide" refers to any single chain or multiple chains of two or more amino acids, and does not refer to a particular length of product. Thus, the definition of "polypeptide" includes peptides, dipeptides, tripeptides, oligopeptides, "proteins", "amino acid chains" or any other term used to refer to two or more amino acid chains, and the term "polypeptide" may be used in place of, or in addition to, any of the terms described above. The term "polypeptide" is also intended to refer to products of modification of the polypeptide after expression, including but not limited to glycosylation, acetylation, phosphorylation, amidation, derivatization by known protecting/blocking groups, proteolytic cleavage, or non-naturally occurring amino acid modification. The polypeptide may be derived from a natural biological source or produced by recombinant techniques, but it need not be translated from the specified nucleic acid sequence, and it may be produced in any manner including chemical synthesis.

"amino acid" refers to an organic compound containing both amino and carboxyl groups, such as an alpha-amino acid, which may be encoded by a nucleic acid directly or in precursor form. A single amino acid is encoded by a nucleic acid consisting of three nucleotides, a so-called codon or base triplet. Each amino acid is encoded by at least one codon. The same amino acid is encoded by different codons called "degeneracy of the genetic code". Amino acids include natural amino acids and unnatural amino acids. Natural amino acids include alanine (three letter code: ala, one letter code: a), arginine (Arg, R), asparagine (Asn, N), aspartic acid (Asp, D), cysteine (Cys, C), glutamine (gin, Q), glutamic acid (Glu, E), glycine (Gly, G), histidine (His, H), isoleucine (Ile, I), leucine (Leu, L), lysine (Lys, K), methionine (Met, M), phenylalanine (Phe, F), proline (Pro, P), serine (Ser, S), threonine (Thr, T), tryptophan (Trp, W), tyrosine (Tyr, Y), and valine (Val, V).

"conservative amino acid substitution" refers to the substitution of one amino acid residue with another amino acid residue that contains a side chain (R group) that is similar in chemical properties (e.g., charge or hydrophobicity). In general, conservative amino acid substitutions will not substantially alter the functional properties of the protein. Examples of classes of amino acids containing chemically similar side chains include: 1) Aliphatic side chain: glycine, alanine, valine, leucine and isoleucine; 2) Aliphatic hydroxyl side chains: serine and threonine; 3) Amide-containing side chains: asparagine and glutamine; 4) Aromatic side chain: phenylalanine, tyrosine, and tryptophan; 5) Basic side chain: lysine, arginine, and histidine; 6) Acidic side chain: aspartic acid and glutamic acid.

The term "isolated" as used herein with respect to cells, nucleic acids, polypeptides, antibodies, etc., e.g., "isolated" DNA, RNA, polypeptides, antibodies, etc., refers to molecules that are separated by one or more of the other components of the cell's natural environment, such as DNA or RNA, respectively. The term "isolated" as used herein also refers to nucleic acids or peptides that are substantially free of cellular material, viral material, or cell culture medium when produced by recombinant DNA techniques, or chemical precursors or other chemicals when chemically synthesized. Furthermore, "isolated nucleic acid" is intended to include nucleic acid fragments that do not exist in a natural state and do not exist in a natural state. The term "isolated" is also used herein to refer to cells or polypeptides isolated from other cellular proteins or tissues. An isolated polypeptide is intended to include both purified and recombinant polypeptides. Isolated polypeptides, antibodies, and the like are typically prepared by at least one purification step. In some embodiments, the purity of the isolated nucleic acid, polypeptide, antibody, etc., is at least about 50%, about 60%, about 70%, about 80%, about 90%, about 95%, about 99%, or a range between any two of these values (inclusive) or any value therein.

The term "recombinant" refers to a polypeptide or polynucleotide, meaning a form of the polypeptide or polynucleotide that does not exist in nature, and non-limiting examples can be combined to produce a polynucleotide or polypeptide that does not normally exist.

"homology" or "identity" or "similarity" refers to sequence similarity between two peptides or between two nucleic acid molecules. Homology can be determined by comparing the positions in each sequence that can be aligned. When a position in the compared sequences is occupied by the same base or amino acid, then the molecules are homologous at that position. The degree of homology between sequences is a function of the number of matched or homologous positions shared by the sequences.

"at least 80% identical" is about 80% identical, about 81% identical, about 82% identical, about 83% identical, about 85% identical, about 86% identical, about 87% identical, about 88% identical, about 90% identical, about 91% identical, about 92% identical, about 94% identical, about 95% identical, about 98% identical, about 99% identical, or a range between any two of these values (inclusive) or any value therein.

"at least 90% identical" is about 90% identical, about 91% identical, about 92% identical, about 93% identical, about 95% identical, about 96% identical, about 97% identical, about 98% identical, about 99% identical, or a range between any two of these values (inclusive of the endpoints) or any value therein.

A polynucleotide or polynucleotide sequence (or polypeptide or antibody sequence) has a certain percentage (e.g., 90%, 95%, 98%, or 99%) of "identity or sequence identity" with another sequence, meaning that when sequences are aligned, the percentage of bases (or amino acids) in the two sequences that are compared are identical. The alignment and percent identity or sequence identity may be determined using visual inspection or software programs known in the art, such as the software program described in Ausubel et al eds. (2007) in CurrentProtocols in Molecular Biology. Preferably, the alignment is performed using default parameters. One such alignment program is BLAST using default parameters, such as BLASTN and BLASTP, both of which use the following default parameters: genetics code = standard; filter = none; strand = both; cutoff = 60; expect=10; matrix = BLOSUM62; descriptive = 50sequences; sortby=highscore; databases = non-redundants; genbank+embl+ddbj+pdb+genbank cdstransplation+swissprotein+spldate+pir. Biologically equivalent polynucleotides are those that have the indicated percent identity and encode polypeptides having the same or similar biological activity.

A polynucleotide consists of a specific sequence of four nucleotide bases: adenine (A), cytosine (C), guanine (G), thymine (T), or thymine to uracil (U) when the polynucleotide is RNA. A "polynucleotide sequence" may be represented by the letters of a polynucleotide molecule. The alphabetical representation may be entered into a database in a computer with a central processing unit and used for bioinformatic applications, such as for functional genomics and homology searches.

The terms "polynucleotide" and "oligonucleotide" are used interchangeably to refer to polymeric forms of nucleotides of any length, whether deoxyribonucleotides or ribonucleotides or analogs thereof. The polynucleotide may have any three-dimensional structure and may perform any function, known or unknown. The following are examples of non-limiting polynucleotides: genes or gene fragments (e.g., probes, primers, ESTs, or SAGE tags), exons, introns, messenger RNA (mRNA), transfer RNA, ribosomal RNA, ribozymes, cDNA, dsRNA, siRNA, miRNA, recombinant polynucleotides, branched polynucleotides, plasmids, vectors, isolated DNA of any sequence, isolated RNA of any sequence, nucleic acid probes, and primers. Polynucleotides may comprise modified nucleotides, such as methylated nucleotides and nucleotide analogs. If present, structural modification of the nucleotide may be performed before or after assembly of the polynucleotide. The sequence of nucleotides may be interrupted by non-nucleotide components. The polynucleotide may be further modified after polymerization, for example by conjugation with a labeling component. This term also refers to double-stranded and single-stranded molecules. Unless otherwise indicated or required, embodiments of any polynucleotide of the present disclosure include a double stranded form and each of two complementary single stranded forms known or predicted to constitute the double stranded form.

The term "encoding" when applied to a polynucleotide refers to a polynucleotide referred to as "encoding" a polypeptide, which polypeptide and/or fragment thereof may be produced by transcription and/or translation in its natural state or when manipulated by methods well known to those skilled in the art.

As used herein, the term "operably connected" refers to a context in which the components being described are in a relationship that permits them to function in their intended manner. For example, a control sequence that is "operably linked" to a coding sequence is joined in a manner such that expression of the coding sequence is achieved under conditions suitable for or compatible with the control sequence. In general, "operably linked" refers to the joining of DNA sequences that are contiguous, and in the case of a secretory leader, contiguous and in reading phase. However, the enhancers do not have to be contiguous. Ligation is achieved by ligation at convenient restriction sites. If such sites are not present, the synthetic oligonucleotides are attached to the monkshood or adaptor according to conventional practice.

As used herein, "vector" refers to a construct that is capable of being delivered in a host cell, and preferably expresses one or more related genes or sequences. Examples of vectors include, but are not limited to, viral vectors, naked DNA or RNA expression vectors, plasmids, cosmids or phage vectors, DNA or RNA expression vectors associated with cationic condensing agents, DNA or RNA expression vectors encapsulated in liposomes, and certain eukaryotic cells, such as producer cells.

As used herein, the term "expression control sequence" or "control sequence" refers to a polynucleotide sequence necessary to effect expression of a coding sequence to which it is ligated. The nature of such control sequences varies depending on the host organism. For example, in prokaryotes, such control sequences typically include promoters, ribosome binding sites, and terminators, and in some cases, enhancers. Thus, the term "control sequence" is intended to include at least all components whose presence is necessary for expression, and may also include additional components whose presence is advantageous, such as a leader sequence.

"host cells" include individual cells or cell cultures that may be or have become recipients for one or more vectors incorporating the polynucleotide insert. Host cells include the progeny of a single host cell, and the progeny may not necessarily be completely identical (in morphology or genomic DNA complement) to the original parent cell due to natural, accidental, or deliberate mutation. Host cells include cells transfected in vivo with one or more polynucleotides of the invention.

As used herein, "mammalian cells" includes reference to cells derived from mammals, including humans, rats, mice, hamsters, guinea pigs, chimpanzees, or macaques. The cells may be cultured in vivo or in vitro.

As used herein, the term "purified product" refers to a preparation of the product that has been separated from cellular components to which the product is normally bound and/or from other types of cells that may be present in the relevant sample.

The terms "polypeptide", "oligopeptide", "peptide" and "protein" are used interchangeably herein to refer to amino acid chains of any length. For example, the chain may be relatively short (e.g., 10 to 100 amino acids) or longer. The chain may be a straight or branched chain, which may comprise modified amino acids, and/or may be interspersed with non-amino acids. The term also encompasses amino acid chains that have been modified naturally or by intervention; for example disulfide bond formation, glycosylation, lipidation, acetylation, phosphorylation or any other manipulation or modification, such as binding to a labeling component. Also included within the definition are, for example, one or more analogs (including, for example, unnatural amino acids, etc.) that contain an amino acid, as well as other modified polypeptides known in the art. It is understood that the polypeptide may appear as a single chain or a binding chain.

An "antibody" or "antigen-binding fragment" refers to a polypeptide or complex of polypeptides that specifically recognizes and binds an antigen. The antibody may be an intact antibody, any antigen-binding fragment thereof, or a single chain thereof. The term "antibody" thus includes any protein or peptide comprising at least a portion of an immunoglobulin molecule having biological activity for binding to an antigen in a molecule. Antibodies and antigen binding fragments include, but are not limited to, complementarity Determining Regions (CDRs), heavy chain variable regions (VH), light chain variable regions (VL), heavy chain constant regions (CH), light chain constant regions (CL), framework Regions (FR) or any portion thereof, or at least a portion of a binding protein, of a heavy chain or light chain or ligand binding portion thereof. CDR regions include the CDR regions of the light chain (VL CDR 1-3) and the CDR regions of the heavy chain (VH CDR 1-3). The antibody or antigen binding fragment of the invention is a bispecific antibody comprising an antigen arm A peptide chain that specifically binds CD47 and an antigen arm B peptide chain that specifically binds PD-L1. In some embodiments, the antigen arm a peptide chain polypeptide chain comprises the structure VLA-CLA-X2-VHA-CHA and the second polypeptide chain comprises the structures VB-CLB-X2-VHB, CHB, wherein the linkers X1, X2 are comprised between CLB and VHB between CLA and CHA, respectively.

The term "antibody fragment" or "antigen-binding fragment" refers to a portion of an antibody, and the composition of an antibody fragment of the invention may be similar to F (ab ') 2, F (ab) 2, fab', fab, fv, scFv, etc., in monospecific antibody fragments. Regardless of its structure, the antibody fragment binds to the same antigen that is recognized by the intact antibody. The term "antibody fragment" includes aptamers, stereoisomers, and diabodies. The term "antigen binding fragment" also includes any synthetic or genetically engineered protein that functions as an antibody by binding to a specific antigen to form a complex.

"Single chain variable fragment" or "scFv" refers to a fusion protein of the variable regions of the heavy (VH) and light (VL) chains of an immunoglobulin. In some aspects, these regions are linked to short-linked peptides of 10 to about 25 amino acids. The linker may be glycine-rich to increase flexibility, serine-or threonine-rich to increase solubility, and may link the N-terminus of VH and the C-terminus of VL, or vice versa. Although the protein has the constant region removed and a linker introduced, it retains the original immunoglobulin specificity. ScFv molecules are generally known in the art and are described, for example, in U.S. Pat. No. 5,892,019.

Single chain Fab live "scFab" refers to the variable region of the light chain (VL), the light chain constant region (CL), the heavy chain variable region (VH) and the heavy chain constant region (CH) of an immunoglobulin, wherein the light chain constant region (CL) and the heavy chain variable region (VH) are linked by a short linker peptide of 10 to about 25 amino acids; the linker may be glycine-rich to increase flexibility, serine-or threonine-rich to increase solubility, and may link the N-terminus of VH with the C-terminus of CL. For example, the binding arm a of the invention that targets binding to anti-CD 47 comprises ScFab (a) and the binding arm B of anti-PD-L1 comprises ScFab (B), respectively.

The term "antibody" includes a wide variety of polypeptides that can be biochemically distinguished. Those skilled in the art will appreciate that the heavy chain classes include gamma, mu, alpha, delta or epsilon (γ, μ, α, δ, ε), some of which are also subclasses (e.g., γ1- γ4). The nature of this chain determines the "class" of antibody as IgG, igM, igA, igG or IgE, respectively. Immunoglobulin subclasses (isotypes), e.g., igG1, igG2, igG3, igG4, etc., have been well characterized and the functional specificities conferred are also known. All immunoglobulin classes are within the scope of the present disclosure. In some embodiments, the immunoglobulin molecule is an IgG class. The four chains are linked by disulfide bonds in a "Y" configuration, wherein the light chain starts at the "Y" mouth and continues through the variable region surrounding the heavy chain.

Antibodies, antigen binding fragments or derivatives disclosed herein include, but are not limited to, polyclonal, monoclonal, multispecific, fully human, humanized, primatized, chimeric antibodies, single chain antibodies, epitope-binding fragments (e.g., fab-like, fab '-like, and F (ab') 2-like), single chain-like Fvs (scFv).

Light chains can be classified as kappa (kappa) or lambda (lambda). Each heavy chain may be associated with a kappa or lambda light chain. In general, when immunoglobulins are produced by hybridomas, B cells or genetically engineered host cells, the light and heavy chains thereof are bound by covalent bonds, and the "tail" portions of the two heavy chains are bound by covalent disulfide bonds or non-covalent bonds. In the heavy chain, the amino acid sequence extends from the N-terminus of the forked end of the Y-configuration to the C-terminus of the bottom of each chain. Immunoglobulin kappa light chain variable region vkappa; the immunoglobulin lambda light chain variable region is vlambda.

Both the light and heavy chains are divided into regions of structural and functional homology. The terms "constant" and "variable" are used in terms of function. The variable regions of the light (VL) and heavy (VH) chain portions determine antigen recognition and specificity. The constant regions of the light and heavy chains confer important biological properties such as secretion, transplacental movement, fc receptor binding, complement fixation, and the like. Conventionally, the numbering of constant regions increases as they become farther from the antigen binding site or amino terminus of an antibody. The N-terminal portion is a variable region and the C-terminal portion is a constant region; the CH3 and CL domains comprise the carboxy-terminus of the heavy and light chains, respectively.

In naturally occurring antibodies, the six "complementarity determining regions" or "CDRs" present in each antigen binding domain are short, non-contiguous amino acid sequences that form the antigen binding domain that specifically bind to an antigen, provided that the antibody assumes its three-dimensional configuration in an aqueous environment. The remaining other amino acids in the antigen binding domain, known as the "framework" region, exhibit less intermolecular variability. The framework regions largely adopt a β -sheet conformation, with the CDRs forming a loop structure linked thereto, or in some cases forming part of a β -sheet structure. Thus, the framework regions form scaffolds to position the CDRs in the correct orientation by non-covalent interactions between the chains. An antigen binding domain with CDRs at specific positions forms a surface complementary to an epitope on an antigen that facilitates non-covalent binding of the antibody to its epitope. For a given heavy or light chain variable region, one of ordinary skill in the art can identify amino acids comprising CDRs and framework regions by known methods (see Kabat, e., et al, U.S. device of Health and Human Services, sequences ofProteins of Immunological Interest, (1983) and Chothia and Lesk, j. Mol. Biol.,196:901-917 (1987)).

The boundaries of CDRs of variable regions of the same antibody may differ according to different assignment systems. Thus, where reference is made to defining an antibody with a particular CDR sequence as defined herein, the scope of said antibody also encompasses such antibodies: the variable region sequence comprises the CDR sequences of the present invention, but the claimed CDR boundaries differ from the specific CDR boundaries defined by the present invention due to the application of different schemes. CDRs defined according to Kabat and Chothia include overlapping or subsets of amino acid residues when compared to each other. Nevertheless, it is within the scope of the invention to apply either definition to refer to the CDRs of an antibody or variant thereof.

Kabat et al also define a numbering system for variable region sequences suitable for use with any antibody. The "Kabat numbering" system can be applied to any variable region sequence by one of ordinary skill in the art independent of other experimental data than the sequence itself. "Kabat numbering" refers to the numbering system set forth by Kabat et al, U.S. Dept. Of Health and Human Services at "Sequence of Proteins of Immunological Interest" (1983). Antibodies may also use EU or Chothia numbering systems.

The antibodies disclosed herein can be derived from any animal, including birds and mammals. Preferably, the antibody is of human, murine, donkey, rabbit, goat, camel, llama, horse or chicken origin. In some embodiments, the variable region is murine. In another embodiment, the variable region may be humanized.

The heavy chain constant region comprises at least one of a CH1 domain, a hinge (e.g., upper, middle, and/or lower hinge region) domain, a CH2 domain, a CH3 domain, or a variant or fragment. The heavy chain constant regions of antibodies may be derived from different immunoglobulin molecules. For example, the heavy chain constant region of a polypeptide may include a CH1 domain derived from an IgG1 molecule and a hinge region derived from an IgG3 molecule. In another embodiment, the heavy chain constant region may include a hinge region derived in part from an IgG1 molecule and in part from an IgG3 molecule. In another embodiment, a portion of the heavy chain may include a chimeric hinge region derived in part from an IgG1 molecule and in part from an IgG4 molecule.

"light chain constant region" includes a portion of the amino acid sequence from an antibody light chain. Preferably, the light chain constant region comprises at least one of a constant kappa domain or a constant lambda domain.

The "VH domain" includes the amino-terminal variable domain of an immunoglobulin heavy chain, and the "CH1 domain" includes the first constant region of an immunoglobulin heavy chain. The CH2 domain is not tightly paired with other domains, but rather two N-linked branched carbohydrate chains are inserted between the two CH2 domains of the intact native IgG molecule. The CH3 domain extends from the CH2 domain to the C-terminus of the IgG molecule, approximately comprising 108 residues. The "hinge region" includes a portion of the heavy chain region connecting the CH1 domain and the CH2 domain. The hinge region comprises about 25 residues and is flexible, thereby enabling independent movement of the two N-terminal antigen binding regions. The hinge region can be subdivided into three distinct domains: upper, middle and lower hinge domains (rouxet al, j.immunol 161:4083 (1998)).

As used herein, the term "Fc chain" refers to the C-terminal region of an antibody heavy chain and comprises two to three constant domains depending on isotype. As used herein, an Fc chain may comprise a native or variant Fc sequence. Unless otherwise specified herein, numbering of amino acid residues in the Fc chain or constant region is according to the EU numbering system, also known as the EU index, as described in Kabat et al, sequences ofProteins of Immunological Interest, 5 th edition, public HealthService, nationalInstitutes of Health, bethesda, MD, 1991.

As used herein, the term "Fc domain" refers to a region of an antibody comprising two Fc chains. For example, in a standard IgG format, an antibody has two heavy chains, both of which have Fc chains. Collectively, the two Fc chains are referred to herein as "Fc domains.

A "wild-type Fc chain" comprises an amino acid sequence that is identical to the amino acid sequence of an Fc chain found in nature. "wild-type" human IgG Fc means the sequence of amino acids naturally occurring in the human population. One or more changes may be made to the wild-type sequence and remain within the scope of the invention.

"variant Fc chains or variants of Fc" comprise amino acid sequences which differ from the amino acid sequence of a wild-type Fc chain by at least one amino acid modification, but retain at least one effector function of the wild-type Fc chain. In some embodiments, the variant Fc-chain has at least one amino acid substitution in the wild-type Fc-chain, such as from about one to about ten amino acid substitutions, and preferably from about one to about five amino acid substitutions, as compared to the wild-type Fc-chain. Variant Fc chains herein will preferably have at least about 80% sequence identity with a wild-type Fc chain, and most preferably at least about 90% sequence identity therewith, more preferably at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99% sequence identity therewith. In some embodiments, the Fc chain comprises part or all of the wild-type hinge region (typically at the N-terminus thereof). In some embodiments, the Fc polypeptide does not comprise a functional or wild-type hinge region.

As used herein, the term "hinge region" includes the meaning known in the art, which is described, for example, in Janeway et al, immunoBiology: the immune system in health and disease, elsevier Science limited, NY (4 th edition, 1999); bloom et al, protein Science,6:407-415,1997; and Humphreys et al, J.Immunol. Methods,209:193-202,1997.

As used herein, "antigen of interest", "tumor antigen" or "tumor-specific antigen" refers to an antigenic determinant presented on the cell surface of a cell of interest, e.g., in a tumor, such as a cancer cell or tumor stroma.

As used herein, "isolated antibody" refers to an antibody that is identified and isolated and/or recovered from at least one component of its natural environment. For example, an antibody isolated or removed from at least one component of an organism, or from a tissue or cell in which the antibody naturally occurs or naturally occurs, is an "isolated antibody" for purposes of the present invention. Isolated antibodies also include in situ antibodies within recombinant cells. An isolated antibody is an antibody that has been subjected to at least one purification or isolation step. According to certain embodiments, the isolated antibody may be substantially

As used herein, the term "linker" refers to an amino acid sequence of two or more amino acids in length. The linker may consist of neutral polar or nonpolar amino acids. The linker may be, for example, 2 to 100 amino acids in length, such as between 2 and 50 amino acids in length, for example 3, 5, 10, 15, 20, 25, 30, 35, 40, 45 or 50 amino acids in length. The linker may be "cleavable", e.g. by self-cleavage or enzymatic cleavage or chemical cleavage. Cleavage sites in amino acid sequences and enzymes and chemicals that cleave at such sites are well known in the art and are also described herein

"disulfide" refers to a covalent bond formed between two sulfur atoms. The thiol group of cysteine may form a disulfide bond or bridge with the second thiol group. In most naturally occurring IgG molecules, the CH1 and CL regions are linked by disulfide bonds. "light chain-heavy chain pair" refers to the collection of light and heavy chains that can form dimers through disulfide bonds between the CL domain of the light chain and the CH1 domain of the heavy chain.

As used herein, the terms "ligate," "fused," and "fusion" are used interchangeably to refer to joining two or more components or components together by any means, including chemical bonding or recombination means.

As used herein, the term "covalently linked" means that the particular moieties are covalently bonded to each other directly or indirectly via one or more intervening moieties, such as a linking peptide or moiety.

As used herein, the terms "genetically fused" and "genetically fused" refer to the co-linear, covalent linkage or joining of gene expression of two or more proteins, polypeptides, or fragments thereof via their individual peptide backbones via a single polynucleotide molecule encoding those proteins, polypeptides, or fragments. Such gene fusion results in expression of a single contiguous gene sequence.

As used herein, the term "modification" refers to amino acid substitutions, insertions, and/or deletions in a polypeptide sequence; a change in the moiety chemically attached to the protein; or modification of the function of a protein, such as an antibody. For example, the modification may be an altered antibody function or an altered carbohydrate structure attached to the protein. As used herein, "amino acid modification" refers to a mutation (substitution), insertion (addition), or deletion of one or more amino acid residues in an antibody. The term "amino acid mutation" indicates substitution of at least one existing amino acid residue with another, different amino acid residue (e.g., replacement of an amino acid residue). The term "amino acid deletion" indicates the removal of at least one amino acid residue at a predetermined position in the amino acid sequence. For example, mutation L234A indicates that the amino acid residue lysine at position 234 in the Fc region of the antibody is substituted with the amino acid residue alanine (lysine is substituted with alanine) (numbering according to the EU index numbering system).

"humanized" antibody refers to chimeric antibodies that comprise amino acid residues from a non-human variable region and amino acid residues from a human framework region. Preferably, the humanized antibody is a human immunoglobulin (recipient antibody) in which residues from the Complementarity Determining Regions (CDRs) of the recipient are replaced by residues from CDRs of a non-human species (donor antibody) such as mouse, rat or rabbit having the desired specificity, affinity and capacity. The humanized antibody may comprise residues as follows: neither in the recipient antibody nor in the introduced CDR or framework sequences, but are included to further improve and optimize antibody performance.

A "human antibody" is an antibody having the following amino acid sequence: corresponding to the amino acid sequence of an antibody produced by a human and/or produced using any of the techniques for producing a human antibody as disclosed herein. This definition of human antibodies specifically excludes humanized antibodies that comprise non-human antigen binding residues.

"chimeric antibody" refers to any antibody whose variable region is obtained or derived from a first species, while its constant region (which may be intact, partial or modified) is derived from a second species. In certain embodiments, the variable region is from a non-human source (e.g., mouse or primate) and the constant region is from a human source.

With respect to bispecific antibodies of the invention, such as antibodies, fragments or derivatives thereof, "biologically active", "bioactivity" and "biological characteristics" are meant to have the ability to bind to biological molecules unless otherwise specified.

"specific binding" generally refers to the complementary binding of an antibody or antigen binding fragment to a particular antigen through its antigen binding domain to an epitope to form a relatively stable complex. "specificity" may be expressed in terms of the relative affinity of an antibody or antigen binding fragment to bind to a particular antigen or epitope. For example, if antibody "a" has a greater relative affinity for the same antigen than antibody "B", antibody "a" may be considered to have a higher specificity for that antigen than antibody "B". Specific binding can be described by equilibrium dissociation constants (KD), a smaller KD meaning a tighter binding.

Antibodies that specifically bind to an antigen may bind to other peptides or polypeptides with lower affinity, as determined by assays known in the art, such as immunoassays, biacore, or other assays. Preferably, antibodies that specifically bind to the antigen do not cross-react with other proteins.

When used in reference to the interaction of an antibody with a protein or peptide, the term "non-specific binding" or "background binding" refers to an interaction that is not dependent on the presence of a particular structure (i.e., an antibody generally binds to a protein, rather than a particular structure, such as an epitope).

As used herein, the term "kon" or "ka" refers to the rate constant at which an antibody associates with an antigen.

As used herein, the term "koff" or "kd" refers to the rate constant of dissociation of an antibody from an antibody/antigen complex.

As used herein, the term "KD" refers to the equilibrium dissociation constant of an antibody-antigen interaction.

Biosensors based on surface plasmon resonance can be used, for example, to characterize analyte/ligand interactions under conditions where the analyte is monovalent with respect to binding to ligands immobilized on the sensor surface at low capacities via capture reagents, to determine association and dissociation rate constants kon and koff, respectively, to determine KD and other ratios. The analysis may be performed, for example, using a kinetic titration method as described in Karlsson et al, anal. Biochem349,136-147,2006, or using a multi-cycle kinetic analysis. The sensor chip, capture reagent and assay buffer are selected for a given assay to allow stable capture of the ligand onto the sensor surface, minimize non-specific binding of the analyte to the surface, and produce a response

The proposal in Myszka, J.mol. Recognit12,279-284,1999 is suitable for analyte binding reactions for kinetic analysis. The analyte binding reactions per analyte/ligand interaction were double referenced and fitted to a 1:1 Langmuir "mass transfer limited model (mass transport limited model)", where ka, kd and Rmax are global parameters as described in Myszka and Morton et al, biophys. Chem 64,127-137 (1997). The equilibrium dissociation constant KD is derived from the ratio of kinetic rate constants, kd=koff/kon. Such assays are preferably performed at 25℃or 37 ℃. In general, the rate constants (kon/ka and koff/kd) and equilibrium dissociation constants are measured using whole antibodies and monomers (e.g., CD47 protein or PD-L1 protein).

As used herein, the term "binding affinity" generally refers to the strength of the sum of non-covalent interactions between a single binding site of a molecule (e.g., an antibody) and its binding partner (e.g., an antigen). As used herein, unless otherwise indicated, "binding affinity" refers to an inherent binding affinity that reflects a 1:1 interaction between members of a binding pair (e.g., an antibody and an antigen). The affinity of a molecule X for its partner Y can generally be expressed by the dissociation constant (KD). The affinity of a molecule X for its partner Y can generally be expressed by a dissociation constant (Kd). For example, the Kd may be about 200nM, 150nM, 100nM, 60nM, 50nM, 40nM, 30nM, 20nM, 10nM, 8nM, 6nM, 4nM, 2nM, 1nM or stronger. Affinity can be measured by common methods known in the art, including those described herein. Low affinity antibodies typically bind antigen slowly and tend to dissociate easily, while high affinity antibodies typically bind antigen faster and tend to remain bound for longer periods of time. A variety of methods of measuring binding affinity are known in the art, any of which may be used for the purposes of the present invention. In particular, the term "binding affinity" is intended to refer to the rate of dissociation of a particular antigen-antibody interaction. KD is the ratio of the dissociation rate (rate of dissociation) (also known as the "off-rate (koff)") to the association rate (association rate) (or "association rate (on-rate)"). Therefore, KD is equal to koff/kon and is expressed as molar concentration (M). Thus the smaller the KD the stronger the binding affinity. Thus, a KD of 1 μM indicates a weak binding affinity compared to a KD of 1 nM. The KD value of an antibody can be determined using well established methods in the art. One method of determining antibody KD is by using Surface Plasmon Resonance (SPR), typically using a biosensor system such as the BIACORE system. BIACORE kinetic analysis involves analysis of antigen binding and dissociation of a chip with immobilized molecules (e.g., molecules comprising epitope binding domains) on the surface. Another method for determining the KD of an antibody is by using biological layer interference measurements (Bio-Layer Interferometry), commonly using techniques (Octet QKe System, forteBio).

Methods for determining whether two molecules specifically bind are well known in the art and include, for example, equilibrium dialysis, surface plasmon resonance, biofilm layer optical interferometry, and the like. For example, an antibody that "specifically binds" antigen CD47 includes an antibody having an equilibrium dissociation constant KD for antigen CD47 of less than or equal to about 100nM, less than or equal to about 10nM, less than or equal to about 5nM, less than or equal to about 1nM, or less than or equal to about 0.5 nM; bispecific antibodies that "specifically bind" antigen CD47/PD-L1 include antibodies having an equilibrium dissociation constant KD of less than or equal to about 100nM, less than or equal to about 10nM, less than or equal to about 5nM, less than or equal to about 1nM or less than or equal to about 0.5nM and an equilibrium dissociation constant KD of less than or equal to about 100nM, less than or equal to about 10nM, less than or equal to about 5nM, less than or equal to about 1nM or less than or equal to about 0.5nM with antigen PD-L1.

As used herein with respect to antibodies, the term "compete" refers to antigen binding arm a or antigen binding fragment (or portion) thereof binds to an epitope in a manner sufficiently similar to the binding of antigen binding arm B or antigen binding portion thereof such that the result of antigen binding arm a binding to its cognate epitope in the presence of antigen binding arm B is detectably reduced as compared to the binding of antigen binding arm a in the absence of antigen binding arm B. This may be the case, but need not be the case, in the alternative that the binding of antigen binding arm B to its epitope in the presence of antigen binding arm a is also detectably reduced. That is, antigen-binding arm a may inhibit the binding of antigen-binding arm B to its epitope, whereas antigen-binding arm B does not inhibit the binding of antigen-binding arm a to its respective epitope. However, when each antibody detectably inhibits the binding of another antibody to its cognate epitope or ligand, whether to the same extent, greater or lesser, the antibodies are said to "cross-compete" with each other for binding to each of one or more of its epitopes. The present invention encompasses both competing and cross-competing antibodies. Regardless of the mechanism by which such competition or cross-competition occurs (e.g., steric hindrance, conformational change, or binding to a common epitope or portion thereof), those skilled in the art will appreciate based on the teachings provided herein that such competing and/or cross-competing antibodies are encompassed and applicable in the methods disclosed herein.

As used herein, "antibody-dependent cell-mediated cytotoxicity" or "ADCC" refers to a cell-mediated reaction in which nonspecific cytotoxic cells expressing Fc receptors (FcR), such as Natural Killer (NK) cells, neutrophils, and macrophages, recognize bound antibody on a cell of interest and subsequently cause lysis of the cell of interest. ADCC activity of the relevant molecules can be assessed using an in vitro ADCC assay, such as the assays described in U.S. Pat. nos. 5,500,362 or 5,821,337. Effector cells suitable for use in such assays include Peripheral Blood Mononuclear Cells (PBMCs) and NK cells. Alternatively or additionally, ADCC activity of the relevant molecules may be assessed in vivo, for example in an animal model such as that disclosed in Clynes et al, PNAS (USA), 95:652-656,1998.

As used herein, "complement dependent cytotoxicity" or "CDC" refers to the dissolution of a target in the presence of complement. The complement activation pathway is initiated by the binding of a first component of the complement system (C1 q) to a molecule (e.g., an antibody) that is complexed with a cognate antigen. To assess complement activation, CDC assays may be performed, such as, for example, gazzano-Santoro et al, j.immunol. Methods,202:163,1996.

"EC50", i.e., half maximal effect concentration (concentration for 50%of maximal effect,EC50), refers to the concentration that causes 50% of the maximal effect.

"bispecific" antibodies refer to antibodies having two antigen binding sites that may be different epitopes of the same antigen or different epitopes of different antigens. The term "effector functions" refers to those biological activities attributed to the Fc region of an immunoglobulin. Examples of immunoglobulin effector functions include: c1q binding and Complement Dependent Cytotoxicity (CDC), antibody dependent cell mediated cytotoxicity (ADCC), antibody Dependent Cellular Phagocytosis (ADCP), cytokine secretion, fc receptor binding, immune complex mediated antigen uptake by antigen presenting cells, down-regulation of cell surface receptors (e.g., B cell receptors), and B cell activation.

"treatment" refers to both therapeutic treatment and prophylactic or preventative measures, with the object of preventing, slowing, ameliorating or stopping an undesirable physiological change or disorder, such as the progression of a disease, including but not limited to, alleviation of symptoms, diminishment of extent of disease, stabilization (i.e., not worsening) of the disease state, delay or slowing of disease progression, amelioration, palliation or disappearance (whether partial or total), prolongation of life span expected when not receiving treatment, and the like. Patients in need of treatment include those already with the condition or disorder, those prone to the condition or disorder, or those in need of prophylaxis of the condition or disorder, for whom administration of the disclosed antibodies or pharmaceutical compositions for detection, diagnostic procedures, and/or treatment would be expected to benefit.

"patient" refers to any mammal in need of diagnosis, prevention, prognosis or treatment, including humans, dogs, cats, rabbits, mice, horses, cattle, and the like.

As used herein, the terms "polynucleotide," "nucleic acid," and "nucleotide sequence" include DNA molecules (e.g., cDNA or genomic DNA), RNA molecules (e.g., mRNA), combinations of DNA and RNA molecules, or hybrid DNA/RNA molecules and analogs of DNA or RNA molecules. Such analogs can be generated using, for example, nucleotide analogs, including, but not limited to, inosine or tritylated bases (tritylated bases). Such analogs may also include DNA or RNA molecules that include modified backbones that provide beneficial properties to the molecule, such as nuclease resistance or increased ability to cross cell membranes. The nucleic acid or nucleotide sequence may be single-stranded, double-stranded, may contain single-stranded and double-stranded portions, and may contain triple-stranded portions, but is preferably double-stranded DNA.

As used herein, the term "isolated" refers to a material that is removed from its original environment (e.g., the natural environment if it is naturally occurring). For example, a naturally occurring polynucleotide or polypeptide present in a living animal is not isolated, but the same polynucleotide or polypeptide that is isolated from some or all of the coexisting materials in the natural system is isolated. Such polynucleotides may be part of a vector and/or such polynucleotides or polypeptides may be a composition comprising the polynucleotides or polypeptides, such as a mixture, solution or suspension or comprising a portion of an isolated cell or cultured cell, and still be isolated, as the vector or composition is not part of its natural environment. Any of the molecules provided herein may be isolated.

As used herein, the term "pharmaceutically acceptable" refers to a product or compound that is approved (or approvable) by a regulatory agency of the federal or a state government or listed in the U.S. pharmacopeia (u.s.pharmacopeia) or other generally recognized pharmacopeia for use in animals, including humans.

As used herein, the term "pharmaceutically acceptable excipient, carrier or adjuvant" or "acceptable pharmaceutical carrier" refers to an excipient, carrier or adjuvant that can be administered to an individual with at least one antibody of the invention without disrupting the activity of the antibody. When administered with an antibody in a dose sufficient to deliver a therapeutic effect, the excipient, carrier or adjuvant should be non-toxic.

As used herein, the term "improvement" refers to a reduction or improvement in one or more symptoms as compared to the absence of administration of the antibody molecule of the invention. "ameliorating" also includes shortening or reducing the duration of symptoms.

As used herein, the terms "prevent", "prevention" and "prevention" refer to the prevention of recurrence or onset of one or more symptoms of a disorder in an individual as a result of administration of a prophylactic or therapeutic agent.

As used herein, "effective amount," "therapeutically sufficient amount," or "effective dose" refers to any amount of a therapeutic agent that, when administered to an individual in a single or multiple doses, is effective or sufficient to prevent, heal, ameliorate, treat or manage a disease, disorder, or side effect, or reduce the rate of progression of a disease or disorder, or prolong the cure, alleviate, or ameliorate the condition of an individual suffering from a disorder as described herein, beyond what would be expected in the absence of such treatment. The term also includes within its scope an amount effective to enhance normal physiological function. An effective amount may be considered in the context of administration of one or more therapeutic agents, and a single agent may be considered to be administered in an effective amount if, in combination with one or more other agents, the desired result is achieved or attained.

As used herein, "antibody that binds to CD47," "antibody that recognizes CD47," "anti-CD 47 antibody molecule," "antibody that specifically binds to CD47," "CD47 antibody," or an analog thereof comprises a molecule that contains at least one binding domain that specifically binds to CD 47. CD47 antibody molecules of the invention include antibodies that interact with CD47 or that recognize CD47, e.g., bind (e.g., specifically bind) to CD47, e.g., human CD47, mouse CD47, rat CD47, cynomolgus monkey CD 47. As used herein, "CD47" refers to a mammalian differentiation cluster 47 (CD 47) protein, preferably a human CD47 protein. The amino acid sequence of human CD47 and related information is provided, for example, under UniProtKB accession number A0A1TSG4, which is incorporated herein by reference for all purposes.

Typically, naturally occurring allelic variants of CD47 have an amino acid sequence at least 95%, 97% or 99% identical to the protein described in UniProtKB accession number A0A1TSG 4. The CD47 protein is characterized as a transmembrane protein that belongs to the immunoglobulin superfamily and interacts with various ligands, such as signal regulator protein α (sirpa), thrombospondin-1 (TSP-1), and membrane integrins. CD47 is overexpressed by various types of cancer cells. The CD47 ligand sirpa is expressed on various cells of the innate immune system, such as macrophages and Dendritic Cells (DCs). Binding of CD47 to sirpa inhibits the activity of sirpa expressing immune cells and thereby enables tumor cells to evade innate immune system monitoring.

Bispecific antibodies

With respect to bispecific antibodies of the invention, antibodies, fragments or derivatives thereof, "biologically active", "bioactivity" and "biological characteristics" are meant to have the ability to bind to biological molecules unless otherwise specified. Compared with monoclonal antibodies, the bispecific antibody has the remarkable advantages of stronger specificity, targeting tumor cells, reducing off-target toxicity and the like. Bispecific antibodies have the following therapeutic advantages over monoclonal antibodies by adding a specific antigen binding site: (1) The two antigen binding sites can be respectively combined with two targets indicated by one tumor cell or can be combined with the tumor cell and the immune cell, so that the T immune cell is gathered around the tumor cell, and the killing power to the tumor is enhanced; (2) Two different medium channels can be blocked simultaneously to play a unique or overlapped function, and a plurality of immune signal channels are mediated to enhance the cytotoxicity; (3) After binding of two different cell surface antigens, the binding specificity can be potentially increased, and side effects such as off-target can be reduced. Therefore, the bispecific antibody has wide application prospect in tumor immunotherapy and inflammation therapy.

Since in tumor immunity, therapeutic regimens targeting multiple targets can cooperate, which is advantageous for preventing immune escape of tumors, co-administration of antibodies targeting different targets in tumor immunity has also entered clinical trials. However, co-administration requires injection of two separate antibody products or a combined preparation of two different antibodies in a single injection. While two injections allow for flexibility in the amount and timing of administration, they cause patient inconvenience and pain. In addition, while combination formulations may provide some flexibility in terms of the amount of drug administered, it is often difficult to find formulation conditions that allow for the chemical and physical stability of the two antibodies in solution because the molecular characteristics of the two antibodies are different. Where co-administration and co-formulation of two different antibody therapies may add additional expense to the patient and/or payer, alternative immunotherapies for treating tumors are needed, and preferably such alternative immunotherapies involve bispecific antibodies.

As used herein, the term "CD47/PD-L1 bispecific antibody" refers to a molecule designed to specifically bind to CD47 and PD-L1.

As used herein, a binding arm a polypeptide that specifically binds CD47 and a binding arm B polypeptide that specifically binds PD-L1 are joined at an interface. In addition to the interface, binding arm a and binding arm B may comprise antigen-binding domain fragments such as scFab "binding domains" (e.g., antibody light chain variable region VL, light chain constant region CL, heavy chain variable region VH and first heavy chain constant region CH1 and linker structure X between CL and VH), and hinge region and Fc region constructed of second heavy chain constant region CH2 and third heavy chain constant region CH3, respectively.

Although any materials and methods similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred materials and methods are now described.

anti-CD 47 monoclonal antibodies that specifically bind to CD4 antibodies

In one aspect, the invention provides CD47 antibodies

In some embodiments, an antibody or antigen binding fragment of the invention that specifically binds CD47 can specifically bind CD47.

In some embodiments, the CD47 antibodies or antigen binding fragments of the invention can specifically bind to mammalian CD47. In some embodiments, CD47 is human CD47. In some embodiments, the antibody molecule binds to one or more extracellular domains of CD47.

In some embodiments, the anti-CD 47/anti-PD-L1 antibodies, or antigen-binding fragments thereof, according to the invention are independently murine, chimeric, humanized, or fully human antibodies. In an exemplary embodiment, the anti-CD 47 antibody or antigen-binding fragment thereof is a humanized CD47 antibody or a fully human CD47 antibody.

In one embodiment, an anti-CD 47 chimeric antibody is provided comprising a heavy chain variable region (CHI VH) comprising an amino acid sequence as set forth in SEQ ID No. 77 or an amino acid sequence having NO less than 85% homology with the amino acid sequence set forth in SEQ ID No. 77 and a light chain variable region (CHI VL) comprising an amino acid sequence as set forth in SEQ ID No. 93 or an amino acid sequence having NO less than 85% homology with the amino acid sequence set forth in SEQ ID No. 93.

In some embodiments, the invention provides chimeric anti-CD 47 antibodies comprising a heavy chain variable region (CHI VH) comprising CHI VH CDR1, CHI VH CDR2, and CHI VH CDR3, and a light chain variable region (CHI VL) comprising CHI VL CDR1, and CHI VL CDR3; CHI VH CDR1, CHI VH CDR2 and CHI VH CDR3 comprise the amino acid sequences shown as SEQ ID NO:43, 44 and 45 respectively, and CHI VL CDR1, CHI VL CDR1 and CHI VL CDR3 comprise the amino acid sequences shown as SEQ ID NO:46, ATS and SEQ ID NO.

In some embodiments, the humanized CD47 antibody comprises a heavy chain a (HA) comprising a heavy chain variable region a (VHA) and a light chain a (LA) comprising a light chain variable region a (VLA); wherein, the liquid crystal display device comprises a liquid crystal display device,

(1) The VHA comprises or is selected from:

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs 22 to 28;

(A2) An amino acid sequence obtained by substituting, deleting or adding one or more amino acids in the amino acid sequence shown in (A1) and functionally identical or similar to the amino acid sequence shown in (A1); and

(A3) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (A1); and/or

(2) The VLA comprises or is selected from

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 57 to 63;

(A5) An amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids to the amino acid sequence shown in (A4) and has the same or similar function as the amino acid sequence shown in (A4); and

In some embodiments, an antibody or antigen-binding fragment of the invention that specifically binds CD47 comprises a heavy chain variable region a (VHA) that specifically binds CD47, and a light chain variable region a (VLA). Wherein VHA comprises 3 CDRs (VHA CDR1, VHA CDR2, VHA CDR 3) and VLB comprises 3 CDRs (VLA CDR1, VLA CDR2, VLA CDR 3).

In one embodiment, an anti-CD 47 antibody is provided comprising (a) a heavy chain variable region (VHA) comprising the amino acid sequence of SEQ id no: 22. 23, 24, 25, 26, 27 or 28, a VHA complementarity determining region one (VHA CDR 1), a VHA complementarity determining region two (VHA CDR 2), and a VHA complementarity determining region three (VHA CDR 3) of a VHA sequence as set forth in any one of claims 23, 24, 25, 26, 27 or 28; and/or (b) a light chain variable region a (VLA) comprising the amino acid sequence as set forth in SEQ ID NO: 57. 58, 59, 60, 61, 62 or 63, a VLA complementarity determining region one (VLA CDR 1), a VLA complementarity determining region two (VLA CDR 2), and a VLA complementarity determining region three (VLA CDR 3) of the VLA sequence.

In one embodiment, the humanized CD47 antibody comprises a heavy chain variable region a (VHA) and a light chain variable region a (VLA) comprising or selected from the group consisting of:

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs.22;

(A2) An amino acid sequence obtained by substituting, deleting or adding one or more amino acids in the amino acid sequence shown in (A1) and functionally identical or similar to the amino acid sequence shown in (B1); and

(A3) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (A1);

and/or

The VLA comprises or is selected from:

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 57;

(A6) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (B4).

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs 23;

and/or

The VLA comprises or is selected from:

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 58;

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs 24;

and/or

The VLA comprises or is selected from:

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 59;

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs 25;

and/or

The VLA comprises or is selected from:

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 60;

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs 26;

and/or

The VLA comprises or is selected from:

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 61;

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs 27;

And/or

The VLA comprises or is selected from:

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs.62;

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs 28;

and/or

The VLA comprises or is selected from:

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 63;

In another embodiment, an anti-CD 47 antibody is provided having any of the VHA and/or any of the VLA sequences as listed in table 1 and having any of the CDRs listed in table 2. Table 1 shows exemplary anti-CD 47 VHA and VLA sequences. CDR sequences are Table 2

Table 1: variable region VHA and VLA sequences of exemplary anti-CD 47 antibodies

/>

In some embodiments, the anti-CD 47 antibodies provided herein comprise VHA and VLA, wherein VHA comprises the amino acid sequence of SEQ ID NO:22 and the VLA comprises the sequence of SEQ ID NO: 57. In one embodiment, the anti-CD 47 antibodies provided herein comprise VHA and VLA, wherein VHA comprises the amino acid sequence of SEQ ID NO:23 and VLA comprises the sequence of SEQ ID NO: 58. In one embodiment, the anti-CD 47 antibodies provided herein comprise VHA and VLA, wherein VHA comprises the amino acid sequence of SEQ ID NO:24 and VLA comprises the sequence of SEQ ID NO: 59. In one embodiment, the anti-CD 47 antibodies provided herein comprise VHA and VLA, wherein VHA comprises the amino acid sequence of SEQ ID NO:25 and VAL comprises the sequence of SEQ ID NO: 60. In one embodiment, the anti-CD 47 antibodies provided herein comprise VHA and VLA, wherein VHA comprises the amino acid sequence of SEQ ID NO:26 and VAL comprises the sequence of SEQ ID NO: 61. In one embodiment, the anti-CD 47 antibodies provided herein comprise, for example, VHA and VLA, wherein VHA comprises the amino acid sequence of SEQ ID NO:27 and VAL comprises the sequence of SEQ ID NO: 62.

In one embodiment, the anti-CD 47 antibodies provided herein comprise VHA and VLA, wherein VHA comprises the amino acid sequence of SEQ ID NO:28 and VAL comprises the sequence of SEQ ID NO: 63.

In some embodiments, provided herein are antibodies that bind to CD47 and/or PD-L1 and compete with the antibodies described herein for binding to the respective antigen, e.g., humanized Hu06-11, hu06-12, hu06-14_, hu06-16, hu06-15, hu06-32 parent, or Hu06-31 compete for binding to CD47

In some embodiments, provided herein is sequence information for the heavy and light chains that are part of the anti-CD 47 parent antibodies of the bispecific antibodies provided herein.

An exemplary embodiment is a murine anti-CD 47 antibody BT007-06-CHI, the heavy chain BT007-06-CHI H of which is shown in SEQ ID NO:71, and the light chain BT007-06-CHI L of which is shown in SEQ ID NO: 72.

An exemplary embodiment is a humanized anti-CD 47 parent antibody Hu06-31, heavy chain Hu06-11H shown in SEQ ID NO. 29, and light chain Hu06-31L shown in SEQ ID NO. 64;

an exemplary embodiment is a humanized anti-CD 47 parent antibody Hu06-12, heavy chain Hu06-1 H1 of which is shown as SEQ ID NO. 30 and light chain Hu 06-12L of which is shown as SEQ ID NO. 65;

an exemplary embodiment is a humanized anti-CD 47 parent antibody Hu06-14, heavy chain Hu 06-14H as shown in SEQ ID NO. 31 and light chain Hu 06-14L as shown in SEQ ID NO. 66;

An exemplary embodiment is a humanized anti-CD 47 parent antibody having heavy chain Hu 06-16H as set forth in SEQ ID NO. 32 and light chain Hu 06-16L (as set forth in SEQ ID NO. 67;

an exemplary embodiment is a humanized anti-CD 47 parent antibody having heavy chain Hu 06-15H as set forth in SEQ ID NO. 33 and light chain Hu 06-15L as set forth in SEQ ID NO. 68.

An exemplary embodiment is an anti-CD 47 parent antibody, the heavy chain Hu 06-15H of which is shown as SEQ ID NO. 34 and the light chain Hu 06-15L of which is shown as SEQ ID NO. 69.

An exemplary embodiment is an anti-CD 47 parent antibody, the heavy chain Hu 06-11H of which is shown in SEQ ID NO. 35 and the light chain Hu 06-11L of which is shown in SEQ ID NO. 66.

The invention also provides CDR portions of antibodies directed against CD 47.

In one embodiment, the invention provides a humanized anti-CD 47 antibody having any one of the VHA CDRs sequences and/or any one of the VLA CDRs sequences as set forth in table 2. In one embodiment, the invention provides an antibody that specifically binds to CD47, wherein the antibody comprises: (a) a VHA comprising (i) a sequence comprising SEQ ID NO: 1. 2, 3, 4, 5, 6 or 7; (ii) a polypeptide comprising SEQ ID NO: 8. 9, 10, 11, 12, 13 or 14; and (iii) a polypeptide comprising SEQ ID NO: 15. 16, 17, 18, 19, 20 or 21; and/or (b) a VLA comprising (i) a nucleic acid comprising SEQ ID NO: 36. 37, 38, 39, 40, 41, 42 or 43; (ii) VLA CDR2 comprising the amino acid sequence ATS; and (iii) a polypeptide comprising SEQ ID NO: 50. 51, 52, 53, 54, 55 or 56.

In some embodiments, provided herein are humanized anti-CD 47 antibodies comprising a heavy chain variable region a (VHA) and a light chain variable region a (VLA), wherein VHA comprises the amino acid sequence of SEQ ID NO: 22. 23, 24, 25, 26, 27, or 28, or a variant thereof having 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions in the sequence; and/or wherein VLB comprises SEQ ID NO: 57. 58, 59, 60, 61, 62 or 63, or a variant thereof having 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 amino acid substitutions in the sequence. Optionally, the substitution is a conservative amino acid substitution. Optionally, the substitution is not in the CDR region.

The present invention also encompasses scFv fragments of the anti-CD 47 antibodies provided herein. The invention also encompasses scFab fragments of the anti-CD 47 antibodies provided herein. Single-chain variable region fragments are made, for example, but not limited to, by fusing light and/or heavy chains using short connecting peptide genes, or light and/or heavy chain variable regions (Bird et al, science242:423-426, 1988). Other forms of single chain antibodies are also contemplated, such as bifunctional antibodies or minibodies.

In some embodiments, the anti-CD 47 antibodies provided herein are monoclonal antibodies. Optionally, the anti-CD 47 antibody is a human antibody or a humanized antibody.

Table 2 exemplary anti-CD 47 CDR sequence numbers

In some embodiments, the present invention provides a murine chimeric antibody antigen CD47-His with an affinity KD of between 0.096 and 5.01 x 10 ^-8 M, or between 0.11 and 4.88X 10 ^-8 M, or between 0.22 and 4.58X 10 ^-8 M, or between 0.42 and 4.32 x 10 ^-8 M, or between 0.60 and 4.02 x 10 ^-8 M, or between 0.80 and 3.80X 10 ^-8 M, or between 0.92 and 3.60X 10 ^-8 M，0.98-3.40ⅹ10 ^-8 M, or between 1.10 and 3.12 x 10 ^-8 M, or between 1.30 and 3.00 x 10 ^-8 M, or between 1.50 and 2.80X 10 ^-8 M, or between 1.70 and 2.60X 10 ^-8 M, or between 1.85 and 2.45X 10 ^-8 M, or between 1.96 and 2.30X 10 ^-8 M，2.02-2.22ⅹ10 ^-8 M, or between 2.08 and 2.18X 10 ^-8 M. In some embodiments, the humanized CD47 antibodies of the invention bind CD47 with high affinity. In some embodiments, the IC50 assay for binding activity of a humanized CD47 antibody of the invention to CD47 is between 1.0 and 55.0ng/mL, or between 2.0 and 50.0ng/mL, or between 3.0 and 45.0ng/mL, or between 4.0 and 40.0ng/mL, or between 5.0 and 35.0ng/mL, or between 5.5 and 30.0ng/mL, or between 6.0 and 28.0ng/mL, or between 6.5 and 25.0ng/mL, or between 7.0 and 24.0ng/mL, or between 7.5 and 22.0ng/mL, or between 8.0 and 22.0ng/mL, or between 8.5 and 20.0ng/mL, or between 9.0 and 18.0ng/mL, or between 9.5 and 15.0ng/mL, or between 9.8 and 13.0ng/mL, or between 10.0 and 10.8ng or between 10.8 and 10.0 ng/mL.

In some embodiments, the humanized CD47 antibodies of the invention can block SIRP-a binding to tumor cell surface CD47 with high affinity. In some embodiments, the ELISA assay determines that the humanized CD47 antibodies of the invention have an activity of blocking SIRP-alpha binding to tumor cell surface CD47 of 300-5000ng/mL, or 400-4000ng/mL, or 500-3500ng/mL, or 600-3000ng/mL, or 700-2800ng/mL, or 700-2500ng/mL, or 700-2000ng/mL, or 700-1500ng/mL, or 700-1000ng/mL.

In some embodiments, the humanized CD47 antibodies of the invention can block SIRP-a binding to tumor cell surface CD47 with high affinity. In some embodiments, the humanized CD47 antibodies of the invention block the binding of SIRP-alpha to CD47 on the surface of Raji-CD47 cells.

In some embodiments, an ELISA assay determines the IC50 assay for blocking activity of a humanized CD47 antibody of the invention to block SIRP-alpha binding to CD47 on the surface of Raji-CD47 cells, or between 1 and 800ng/mL, or between 1 and 700ng/mL, or between 2 and 650ng/mL, or between 3 and 600ng/mL, or between 4 and 550ng/mL, or between 5 and 500ng/mL, or between 10 and 450ng/mL, or between 10 and 400ng/mL, or between 10 and 350ng/mL, or between 20 and 300ng/mL, or between 30 and 280ng/mL, or between 50 and 260ng/mL, or between 70 and 220ng/mL, or between 100 and 180ng/mL, or between 130 and 150ng/mL.

(II) bispecific antibodies that specifically bind CD47 and PD-L1

In one aspect, provided herein are CD47/PD-L1 bispecific antibodies.

In some embodiments, the CD47/PD-L1 bispecific antibody comprises at least an antigen-binding arm a derived from a CD47 antibody that specifically binds to CD47 and an antigen-binding arm B derived from a PD-L1 antibody that specifically binds to PD-L1. In some embodiments, antigen binding arm a comprises a heavy chain variable region a (VHA) and antigen binding arm B comprises a heavy chain variable region B (VHB). In some embodiments, antigen binding arm a comprises heavy chain variable region a (VHA (and light chain variable region a (VLA)) and antigen arm B comprises heavy chain variable region B (VHB) and light chain variable region B (VLB).

In some embodiments, the bispecific antibodies that specifically bind CD47 and PD-L1 according to the invention, i.e., the bizarre CD47/PD-L1 antibodies, the anti-CD 47 antibodies or antigen-binding fragments thereof and the anti-PD-L1 antibodies or antigen-binding fragments thereof are each independently murine, chimeric, humanized or fully human antibodies. In an exemplary embodiment, the anti-CD 47/anti-PD-L1 antibodies of the invention, the anti-CD 47 antibodies or antigen-binding fragments thereof, and the anti-PD-L1 antibodies or antigen-binding fragments thereof, are humanized antibodies or fully human antibodies.

The CD47/PD-L1 bispecific antibodies provided herein may contain any of the humanized anti-CD 47 antibodies provided herein (e.g., as antigen binding arm a) and an anti-PD-L1 antibody to Atezolizumab (e.g., as antigen binding arm B)

In one aspect, the invention provides a bispecific antibody comprising a binding arm a that specifically binds CD47 and a binding arm B that specifically binds PD-L1, wherein the binding arm a comprises a light chain a (LA) derived from an anti-CD 47 antibody or antigen binding fragment, a heavy chain a (HA), and a linker 1 (X1) connecting between the light chain a and the heavy chain a, and the binding arm B comprises a light chain B (LB), a heavy chain B (HB), and a linker 2 (X2) connecting between the light chain B and the heavy chain B, derived from an anti-PD-L1 antibody or antigen binding fragment, wherein the heavy chain a and the heavy chain B interact to form a FIH structure, the binding arm a and the binding arm B forming a heterodimer by the KIH structure.

In some embodiments, a CD47/PD-L1 bispecific antibody is provided comprising an antigen binding arm a that specifically binds to CD47 and an antigen binding arm B that specifically binds to PD-L1, wherein the bispecific antibody comprises VHA and VLA of any one of the anti-CD 47 antibodies as shown in table 3 or the CDRs thereof of table 4, and wherein the bispecific antibody comprises VHB and VLB of any one of the anti-PD-L1 antibodies as shown in table 3 or the CDRs thereof of table 4.

Table 3: CD47 binding arm and VH and VL sequences of anti-PD-L1 binding arm

Table 4: CD47 binding arm and CDR of anti-PD-L1 binding arm

/>

The CD47/PD-L1 bispecific antibodies provided herein can contain any of the anti-CD 47 antibodies provided herein (e.g., as antigen-binding arm a) and any of the anti-PD-L1 antibodies of Atezolizumab (e.g., as antigen-binding arm B)

In some embodiments, the invention provides a CD47/PD-L1 bispecific antibody comprising an antigen-binding arm a that specifically binds to CD47 and an antigen-binding arm B that specifically binds to PD-L1, wherein the antigen-binding arm a comprises VHA and/or VLA, wherein: (a) VHA comprises SEQ ID NO:22 and (b) VLA comprises SEQ ID NO:57; and antigen binding arm B comprises VHB and/or VLB, wherein: (a) VHB comprises SEQ ID NO:85; and (b) VLB comprises SEQ ID NO:86.

in some embodiments, a CD47/PD-L1 bispecific antibody is provided comprising an antigen-binding arm a that specifically binds to CD47 and an antigen-binding arm B that specifically binds to PD-L1, wherein the antigen-binding arm a comprises VHA and/or VLA, wherein: (a) VHA comprises SEQ ID NO:23 and (b) VLA comprises SEQ ID NO:58; and antigen binding arm B comprises VHB and/or VLB, wherein: (a) VH comprises SEQ ID NO:86; and (b) VL comprises SEQ ID NO:86.

Functional characteristics

1) CD47 binding moieties in bispecific antibodies

In some embodiments, the bispecific antibodies or antigen-binding fragments of the invention that specifically bind CD47 and PD-L1 can specifically bind CD47. In some embodiments, the bispecific antibodies or antigen-binding fragments of the invention can specifically bind to mammalian CD47. In some embodiments, CD47 is human CD47. In some embodiments, the antibody molecule binds to one or more extracellular domains of CD47.

The bispecific antibody of the present invention that specifically binds CD47 and PD-L1 contains VHA and VLA of any one of the anti-CD 47 antibodies as shown in table 3 or VHA CDRs and VLA CDRs of table 4, and wherein the bispecific antibody contains VHB and VLB in the anti-PD-L1 antibody as shown in table 5 or VHB CDRs and VLB CDRs of table 4.

Table 3: CD47 binding arm and VH and VL sequences of anti-PD-L1 binding arm

/>

Table 4: CD47 binding arm and CDR of anti-PD-L1 binding arm

/>

In some embodiments, the bispecific antibodies or antigen binding fragments of the invention have one or more of the following properties:

(a) The bispecific antibodies or antigen-binding fragments of the invention bind CD47 with high affinity.

In some embodiments, the bispecific antibodies or antigen-binding fragments of the invention that specifically bind CD47 and PD-L1 bind recombinant human CD47-His with high affinity, e.g., with CD47 with the following equilibrium dissociation constants (KD), the CD47/PD-L1 bispecific antibodies of the invention have a KD of 0.01-8 x 10 ^-8 M, or between 0.01 and 7 x 10 _-8 M, or between 0.01 and 6.5 x 10 ^-8 M, or between 0.02 and 5.8 x 10 ^-8 M, or between 0.08 and 5.2X 10 ^-8 M, or between 0.1 and 4.8 x 10 ^-8 M, or between 0.2 and 4.5 x 10 ^-8 M, or between 0.5 and 4.0 x 10 ^-8 M, or between 0.8 and 3.8 x 10 ^-8 M, or between 1.0 and 3.5X 10 ^-8 M, or between 1.2 and 3.3 x 10- ⁸ M, or between 1.5 and 3.0 x 10 ^-8 M, or between 1.8 and 2.8 x 10 ^-8 M, or between 2.0 and 2.5X 10 ^-8 M. In some embodiments, the CD47/PD-L1 bispecific antibody of the invention binds with affinity to recombinant human PD-L1-His with a KD value of 4-10 x 10 ^-10 M, or between 4 and 10 x 10 ^-10 M, or between 5 and 10 x 10 and 10M, or between 6 and 10 x 10 and 10M, or between 6.2 and 8 x 10 ^-10 M, or between 6.3 and 7 x 10 ^-10 M, or between 6.3 and 6.5X 10 ^-10 M. In an exemplary embodiment, the CD47/PD-L1 bispecific antibody of the invention has an affinity KD value of between 6.3 and 6.5 x 10 for recombinant human CD47-His ^-10 M。

The CD47/PD-L1 bispecific antibodies of the invention bind with high affinity to hCD47 overexpressing tumor cells. In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention bind to Raji cells that overexpress hCD 47. In some embodiments, the EC50 value of FACS-binding activity assay for CD47 binding of a CD47/PD-L1 bispecific antibody of the invention to a Raji cell overexpressing hCD47 is between 1000 and 2000ng/Ml, or between 1000 and 1500ng/Ml, or between 0.10 and 15000ng/Ml, or between 0.01 and 12000ng/Ml, between 0.02 and 10000ng/m, L, or between 0.05 and 8000ng/Ml, or between 0.1 and 6000ng/Ml, or between 0.5 and 5000ng/Ml, or between 1 and 4500ng/Ml, or between 5 and 4000ng/Ml, or between 10 and 3500ng/Ml, or between 50 and 3000ng/Ml, or between 100 and 2600ng/Ml, or between 200 and 2300ng/Ml, or between 500 and 2000ng/Ml, or between 800 and 1800ng/Ml, or between 1000 and 1600ng/Ml. (d) In some embodiments, the bispecific antibody or antigen binding fragment of the invention is a blocking antibody that blocks the binding of CD47 to sirpa. In some embodiments, a CD47/PD-L1 bispecific antibody of the invention can block CD47 binding to sirpa on Raji cells. In some embodiments, a CD47/PD-L1 bispecific antibody of the invention blocks FACS-blocking Activity assay for binding of CD47 to SIRPalpha on Raji cells has an EC50 value of between 0.05 and 28000ng/mL, or between 0.08 and 26000ng/mL, or between 0.1 and 25000ng/mL, or between

0.5-20000ng/mL, or between 1-15000ng/mL, or between 5-13000ng/mL, or between 10-12000ng/mL, or between 20-10000ng/mL, or between 50-8000ng/mL, or between 100-6000ng/mL, or between 200-5000ng/mL, or between 500-4000ng/mL, or between 800-3800ng/mL, or between 1100-3500ng/mL, or between 1300-3000ng/mL. In some embodiments, a CD47/PD-L1 bispecific antibody of the invention blocks FACS-blocking activity of binding of CD47 to sirpa on Raji cells with EC50 values between 1600-2500ng/mL.

In an exemplary embodiment, a CD47/PD-L1 bispecific antibody of the invention blocks FACS-blocking activity assay for binding of CD47 to SIRPalpha on Raji cells resulting in an EC50 value of 1800-2200ng/mL.

In some embodiments, a bispecific antibody or antigen-binding fragment of the invention comprises a heavy chain variable region a (VHA) that specifically binds CD47, and a light chain variable region a (VLA). Wherein the VHA comprises 3 CDRs (VHA CDR1, VHb CDR2, VHbCDR 3) and the VLA comprises 3 CDRs (VLA CDR1, VLA CDR2, VLA CDR 3).

In some embodiments, VHA CDR1 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 1-7; VHA CDR2 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 8 to 14; VHA CDR3 comprises an amino acid sequence as set forth in any one of SEQ ID NOS.15-21. In some embodiments, VHb CDR1 comprises the amino acid sequence shown as SEQ ID NO. 1; VHb CDR2 comprises the amino acid sequence shown in SEQ ID NO. 8; VHb CDR3 comprises the amino acid sequence shown as SEQ ID NO. 15.

In some embodiments, VLA CDR1 comprises an amino acid sequence as set forth in any one of SEQ ID NOS: 36-42; VLA CDR2 comprises the amino acid sequence as shown by ATS; VLA CDR3 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 50 to 56. In some embodiments, VLA CDR1 comprises the amino acid sequence as shown in SEQ ID NO. 36; VLA CDR2 comprises the amino acid sequence shown as SEQ ID NO. 43; VLA CDR3 comprises the amino acid sequence as shown in SEQ ID NO: 50.

In some embodiments, VHA CDR1 comprises the amino acid sequence shown as SEQ ID NO. 2; VHA CDR2 comprises the amino acid sequence shown as SEQ ID NO 9; VHA CDR3 comprises the amino acid sequence shown as SEQ ID NO. 16; VLA CDR1 comprises the amino acid sequence shown as SEQ ID NO. 37; VLA CDR2 comprises the amino acid sequence as shown by ATS; VLA CDR3 comprises the amino acid sequence as shown in SEQ ID NO: 51.

In some embodiments, the VHA comprises an amino acid sequence as set forth in any one of SEQ ID NOs 22-28; in some embodiments, VLA comprises the amino acid sequence as set forth in any one of SEQ ID NOs 57-63.

In some embodiments, the bispecific antibodies or antigen-binding fragments of the invention comprise a variable region A consisting of a VHA having the amino acid sequence shown in SEQ ID NO. 22 and a VLA having the amino acid sequence shown in SEQ ID NO. 57.

2) Part of bispecific antibody binding to PD-L1

In some embodiments, the bispecific antibodies or antigen-binding fragments of the invention can specifically bind PD-L1. In some embodiments, the bispecific antibodies or antigen-binding fragments of the invention can specifically bind to mammalian PD-L1. In some embodiments, PD-L1 is human PD-L1. In some embodiments, the antibody molecule binds to one or more extracellular domains of PD-L1.

In some embodiments, the antibody binding affinity is determined using surface plasmon resonance techniques (e.g., biacore affinity measurement).

(a) The bispecific antibodies or antigen-binding fragments of the invention bind to PD-L1 (e.g., recombinant human PD-L1-His) with high affinity, e.g., in some embodiments, the CD47/PD-L1 bispecific antibodies of the invention bind to recombinant human PD-L1-His with high affinity to PD-L1 with an equilibrium dissociation constant (KD) of 0.01 to 600 x 10, in one embodiment, the CD47/PD-L1 bispecific antibodies of the invention bind to PD-L1 with an equilibrium dissociation constant (KD) ^-10 M, or between 0.03 and 300 x 10 ^-10 M, or between 0.06 and 100 x 10 ^-10 M, or between 0.5 and 50 x 10 ^-10 M, or between 1.0 and 30 x 10 ^-10 M, or between 1.2 and 20 x 10 ^-10M Or between 1.5 and 10 x 10 ^-10 M, or between 2 and 10 x 10- ¹⁰ M, or between 3 and 10 x 10 ^-10 M, or between 4 and 10 x 10 ^- ¹⁰ M, or between 4 and 10 x 10 ^-10 M, or between 5 and 10 x 10 ^-10 M, or between 6 and 10 x 10 ^-10 M, or between 6.2 and 8 x 10- ¹⁰ M, or between 6.3 and 7 x 10 ^-10 M, or between 6.3 and 6.5X 10 ^-10 M。

In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention bind PD-L1 with a high affinity binding KD value of between 4 and 10 x 10 ^-10 M, or between 4 and 10 x 10 ^-10 M, or between 5 and 10 x 10 and 10M, or between 6 and 10 x 10 and 10M, or between 6.2 and 8 x 10 ^-10 M, or between 6.3 and 7 x 10 ^-10 M, or between 6.3 and 6.5X 10 ^-10 M。

(b) The bispecific antibodies or antigen-binding fragments of the invention bind with high affinity to cells expressing human PD-L1. In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention can bind with high affinity to CHO cells overexpressing hPD-L1. In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention have an EC50 value of 40 to 40000ng/mL, or 60 to 30000ng/mL, or an EC50 value, or 100 to 10000ng/mL, or 200 to 5000ng/mL, as determined by FACS for binding activity to CHO cells overexpressing hPD-L1. In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention have an EC50 value of 400 to 4000ng/mL, or 500 to 2000ng/mL, or 500 to 1500ng/mL, or 500 to 1000ng/mL, as determined by FACS for binding activity to CHO cells overexpressing hPD-L1. In an exemplary embodiment, the EC50 value of FACS assay for binding activity of a CD47/PD-L1 bispecific antibody of the invention to CHO cells overexpressing hPD-L1 is between 800 and 900ng/mL.

(c) In some embodiments, the bispecific antibodies or antigen binding fragments of the invention block the relevant activity of PD-L1/PD-1. In some embodiments, the relevant activity of PD-L1 is the binding of PD-L1 to PD-1.

In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention can block binding of PD-L1 to PD1 on CHO-K1-PD-L1 cell membranes with high affinity. In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention have an EC50 value of 50-10000ng/mL, or 100-8000ng/mL, or 200-7000ng/mL, or 300-6500ng/mL, or 500-6000ng/mL, or 1000-5000ng/mL, or 1200-4000ng/mL, or 1500-3600ng/mL, or 2000-3000ng/mL, or 2500-3000ng/mL, as determined by the blocking activity of a FACS blocking binding of PD-L1 to PD1 on a CHO-K1-PD-L1 cell membrane. In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention block FACS blocking activity of PD-L1 binding to PD1 on CHO-K1-PD-L1 cell membranes with EC50 values between 2500 and 3000ng/mL.

(d) The bispecific antibodies or antigen binding fragments of the invention are capable of inducing antibody-dependent cell-mediated cytotoxicity (ADCC).

(e) Bispecific antibodies or antigen binding fragments of the invention inhibit one or more activities of PD-L1, e.g., resulting in one or more of the following: increased tumor infiltrating lymphocytes, increased T cell receptor mediated proliferation, or decreased immune evasion by cancer cells. In some embodiments, the bispecific antibody or antigen binding fragment is capable of inhibiting proliferation of a tumor, which is tumor immune escape. In some embodiments, the tumor is a gastrointestinal tumor (e.g., cancer), such as colon cancer.

In some embodiments, the bispecific antibodies or antigen-binding fragments of the invention comprise a heavy chain variable region a (VHA), and a light chain variable region a (VLA) that specifically bind PD-L1. Wherein the VHA comprises 3 CDRs (VHA CDR1, VHA CDR2, VHA CDR 3) and the VLA comprises 3 CDRs (VLA CDR1, VLA CDR2, VLA CDR 3). In some embodiments, the CDRs of VHB employ heavy chain CDRs in Atezolizumab. In some embodiments, the VHB-derived antibody is selected from the heavy chain variable region of Atezolizumab. In some embodiments, the CDRs of the VLB are selected from the group consisting of light chain CDRs in Atezolizumab. In some embodiments, VLB employs a light chain variable region in Atezolizumab.

In some embodiments, VHB CDR1 comprises an amino acid sequence as shown in NO. 80; VHB CDR2 comprises the amino acid sequence shown as SEQ ID NO. 81; VHB CDR3 comprises the amino acid sequence shown as SEQ ID NO. 82. In some embodiments, VLB CDR1 comprises an amino acid sequence as shown in NO. 80; VLB CDR2 comprising the amino acid sequence shown as SEQ ID NO. 81; VLB CDR3 comprises the amino acid sequence as shown in SEQ ID NO. 82. In some embodiments, VLB CDR1 comprises the amino acid sequence as shown in SEQ ID NO. 148; VLB CDR2 comprises an amino acid sequence as shown in SAS; VLB CDR3 comprises the amino acid sequence as shown in SEQ ID NO: 84.

In some embodiments, VHB comprises the amino acid sequence shown as SEQ ID NO. 85; in some embodiments, the VLB comprises the amino acid sequence shown as SEQ ID NO. 86.

3) anti-PD-L1/CD 47 bispecific antibodies

In the innate immune system, monocytes, macrophages and dendritic cells act as Antigen Presenting Cells (APCs) by phagocytosis. While the ability of APCs to engulf tumor cells by phagocytosis is an integral bridge linking innate and adaptive immunity. Reduced T cell activation results indirectly from reduced uptake of tumor cells by APCs, see, e.g., avice MN et al, rotor of CD47 in the induction of human naive T cell anergy, journal ofImmunology,167 (5): 2459-2468 (2001). It has been demonstrated that both PD-L1 and CD47 proteins are expressed under the control of the transcription factor MYC and are simultaneously overexpressed on tumor cells, see, e.g., stephanie C et al MYC regulatesthe antitumor immune response through CD and PD-L1, science,352 (6282): 227-31 (2016). Tumor cells evade the monitoring of the innate immune system through the PD-L1/PD1 and CD47/SIRP alpha channels, and obtain adaptive immune tolerance.

In some embodiments, the invention provides anti-PD-L1/CD 47 bispecific antibodies or antigen-binding fragments that can specifically bind to PD-L1 and CD47. In some embodiments, the antibodies or fragments thereof of the invention bind to mammalian PD-L1 and CD47, such as human PD-L1 and CD47. For example, the antibody molecule specifically binds to an epitope (e.g., a linear or conformational epitope) on PD-L1 and CD47. In some embodiments, the antibody molecule binds to one or more extracellular domains of PD-L1 and CD47.

In some embodiments, the anti-PD-L1/CD 47 bispecific antibodies or antigen-binding fragments of the invention have one or more of the following properties:

(a) In some embodiments, the anti-PD-L1/CD 47 bispecific antibodies of the invention are capable of binding to both PD-L1 and CD47 and maintain the affinity constant of the parent antibody, thereby being capable of blocking the PD1/PD-L1 signaling pathway and blocking the sirpa/CD 47 signaling pathway;

(b) In some embodiments, the anti-PD-L1/CD 47 bispecific antibodies of the invention are capable of binding with ultra high affinity (e.g., kd=1-10 x 10 ^-10 M) binds to PD-L1 with high affinity (e.g., kd=1-5×10 at a concentration of 5 μg/mL) ^-9 M) binds to CD 47; selective binding of the anti-PD-L1/CD 47 bispecific antibodies of the invention to tumor cells is promoted by specific binding to PD-L1 on tumor cells Avoiding binding to CD47 expressed in many normal tissues and reducing side effects; by significantly expanding the effective dose range of the anti-PD-L1/CD 47 bispecific antibodies of the invention when the affinity to CD47 is much lower than to PD-L1. In some embodiments, in a CD47/PD-L1 bispecific antibody provided herein that has a greater affinity for PD-L1 than CD47, the anti-PD-L1 binding moiety of the antibody has an affinity for PD-L1 of at least 2× (i.e., 2 fold), 3×, 5× 0, 10× 1, 20×, 50×, 100×, 200×, 500×, 1000×, 2000×, or 5000× greater than the affinity of the anti-CD 47 binding moiety of the antibody for CD 47.

In some embodiments, a CD47/PD-L1 bispecific antibody of the invention can bind to a biscationic tumor cell that expresses both CD47 and PD-L1. In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention have an EC50 value of between 0.01 and 4000ng/mL, or between 0.02 and 3000ng/mL, or between 0.05 and 2500ng/mL, or between 0.1 and 1800ng/mL, or between 0.8 and 1300ng/mL, or between 1 and 1000ng/mL, or between 2 and 800ng/mL, or between 5 and 500ng/mL, or between 10 and 300ng/mL, or between 20 and 200ng/mL, or between 50 and 180ng/mL, or between 80 and 150ng/mL, as determined by FACS-binding activity associated with a dual-positive tumor cell expressing both CD47 and PD-L1.

In an exemplary embodiment, the EC50 value of FACS-binding activity assay for binding of a CD47/PD-L1 bispecific antibody of the invention to a biscationic tumor cell expressing both CD47 and PD-L1 is between 100 and 130ng/mL.

In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention bind to erythrocytes in a dose-response. In some embodiments, the CD47/PD-L1 bispecific antibody of the invention is at a concentration of 0.0001 to 100 μg/mL, or 0.0005 to 800 μg/mL, or 0.001 to 40 μg/mL, or 0.05 to 20 μg/mL, or 0.01 to 10 μg/mL, or 0.05 to 6 μg/mL, or 0.1 to 3 μg/mL, or 0.5 to 2 μg/mL, or 1 to 1.5 μg/mL, in combination with erythrocytes as a dose effect. In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention have an average fluorescence intensity (MFI) assay of between 0.02 and 20000ng/mL, or between 0.05 and 20000ng/mL, or between 0.1 and 18000ng/mL, or between 0.2 and 16000ng/mL, or between 0.5 and 15000ng/mL, or between 5 and 12000ng/mL, or between 10 and 10000ng/mL, or between 20 and 9000ng/mL, or between 50 and 8500ng/mL, or between 100 and 8000ng/mL, or between 200 and 7500ng/mL, or between 500 and 7000ng/mL, or between 1000 and 6000ng/mL, or between 1200 and 5000ng/mL, or between 1500 and 4000ng/mL, or between 1800 and 3000ng/mL, or between 1800 and 2500ng/mL.

In an exemplary embodiment, the CD47/PD-L1 bispecific antibodies of the invention bind to erythrocytes in a dose-response EC50 assay ranging from 2000 to 2500ng/mL. In an exemplary embodiment, the CD47/PD-L1 mab of the present invention binds to erythrocytes less than humanized CD47 mab by about 8-60 fold, or about 10-50 fold, or about 15-45 fold, or about 21-40 fold, or about 26-35 fold.

In some embodiments, the humanized CD47 antibodies of the invention have selective binding activity to tumor cells in a mixture of tumor cells and erythrocytes.

In an exemplary embodiment, the CD47/PD-L1 bispecific antibody of the invention has a concentration of 20 μg/mL to 0.001 μg/mL in a dose-response to binding to erythrocytes.

In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention do not cause hemolysis and coagulation of erythrocytes at a concentration ranging from 1000 to 0.05 μg/mL, or from 800 to 0.02 μg/mL, or from 400 μg/mL to 0.4 μg/mL, or from 100 μg/mL to 1 μg/mL, or from 25 μg/mL to 6.25 μg/mL, or from 8 μg/mL to 45 μg/mL.

In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention can selectively bind to tumor cells in a mixture of tumor cells and erythrocytes. In some embodiments, the CD47/PD-L1 bispecific antibody of the invention has an EC50 value of 40-50ng/mL for bound tumor cells and an EC50 of 5000-6000ng/mL for bound erythrocytes when the concentration is 0.0001-100 μg/mL, or 0.0005-80 μg/mL, or 0.001-40 μg/mL, or 0.05-20 μg/mL, or 0.01-10 μg/mL, or 0.05-6 μg/mL, or 0.1-3 μg/mL, or 0.5-2 μg/mL, or 1-1.5 μg/mL. The CD47/PD-L1 bispecific antibody of the invention has selective binding activity to tumor cells in a mixture of tumor cells and erythrocytes.

In some embodiments, the fluorescence intensity of binding of the CD47/PD-L1 bispecific antibodies of the invention to tumor cells in a tumor cell and red blood cell mixture is measured using a flow cytometer as a dose-response curve. The EC50 value is between 0.05 and 3000ng/mL, or between 0.05 and 2000ng/mL, or between 0.05 and 1000ng/mL, or between 0.1 and 600ng/mL, or between 0.2 and 400ng/mL, or between 0.5 and 300ng/mL, or between 1 and 280ng/mL, or between 2 and 240ng/mL, or between 5 and 200ng/mL, or between 10 and 150ng/mL, or between 30 and 120ng/mL, or between 40 and 100ng/mL, or between 40 and 50ng/mL. In some embodiments, the CD47/PD-L1 bispecific antibodies of the invention are assayed with a flow cytometer for the fluorescence intensity of PE/FITC/PB450 binding to erythrocytes in a tumor cell and erythrocyte mixture, EC50, or in the range of 600-13500ng/mL, or in the range of 800-12500ng/mL, or in the range of 1000-11500ng/mL, or in the range of 1500-10500ng/mL, or in the range of 2000-9500ng/mL, or in the range of 3000-7500ng/mL, or in the range of 3500-7000ng/mL, or in the range of 4000-6500ng/mL, or in the range of 4500-6000ng/mL, or in the range of 5000-5500ng/mL. The diabodies of the present invention have binding activity that preferentially binds tumor cells.

In some embodiments, the CD47/PD-L1 bispecific antibody and the humanized CD47 monoclonal antibody have ADCP effects at a concentration of not less than 8ng/Ml, not less than 6.5ng/Ml, not less than 5.5ng/Ml, not less than 5ng/Ml, not less than 4.5ng/Ml, not less than 4.0ng/Ml, not less than 3.0ng/Ml, not less than 2.0ng/Ml, and can induce macrophage lines to phagocytize tumor cells, and the double antibody has better phagocytic effects on tumor cells of CD47/PD-L1 double positive than the humanized monoclonal antibody.

In some embodiments, the CD47/PD-L1 bispecific antibody and the humanized CD47 monoclonal antibody have ADCP effects at 0.001-8ng/mL,0.003-6.5ng/mL,0.005-5.5ng/mL,0.1-5ng/mL,0.2-4.5ng/mL,0.4-4.0ng/mL,0.8-3.0ng/mL,1.0-2.0ng/mL and 1.5-1.8ng/mL, can induce macrophage lines to phagocytize tumor cells, and the double antibody has better phagocytic effects on tumor cells of CD47/PD-L1 double positive than the humanized monoclonal antibody.

In some embodiments, both the CD47/PD-L1 bispecific antibodies and the humanized CD47 monoclonal antibodies of the invention have Antibody Dependent Cellular Cytotoxicity (ADCC). In some embodiments, both the CD47/PD-L1 bispecific antibody and the humanized CD47 mab pair of the invention have Antibody Dependent Cellular Cytotoxicity (ADCC) against Raji-PD-L1 cells.

In some embodiments, the CD47/PD-L1 bispecific antibodies and humanized CD47 monoclonal antibodies zai Raji-PD-L1 cells of the invention can mediate Antibody Dependent Cellular Cytotoxicity (ADCC). In some embodiments the CD47/PD-L1 bispecific antibody can mediate Antibody Dependent Cellular Cytotoxicity (ADCC) at a concentration of 0.001-2.5 μg/ml,0.005-2.5 μg/ml, or 0.01-2.2 μg/ml, or 0.02-1.8 μg/ml, or 0.04-1.5 μg/ml, or 0.06-1.2 μg/ml, or 0.08-1.0 μg/ml, or 0.1-0.5 μg/ml, or 0.2-0.4 μg/ml at Raji-PD-L1 cells.

In some embodiments CD47/PD-L1 bispecific antibodies can mediate Antibody Dependent Cellular Cytotoxicity (ADCC) at Raji-PD-L1 at a concentration of 0.008-0.2 μg/ml. Some embodiments of the CD47/PD-L1 bispecific antibody may mediate Antibody Dependent Cellular Cytotoxicity (ADCC) at HCC827 at a concentration of 0.04-1.0 μg/ml.

In some embodiments the CD47/PD-L1 bispecific antibody may mediate Antibody Dependent Cellular Cytotoxicity (ADCC) at HCC827 at a concentration of 0.001-2. Mu.g/ml, 0.001-1.8. Mu.g/ml, or 0.001-1.5. Mu.g/ml, or 0.001-1. Mu.g/ml, or 0.001-0.5. Mu.g/ml, or 0.001-0.2. Mu.g/ml, or 0.001-0.1. Mu.g/ml, or 0.002-0.06. Mu.g/ml, 0.005-0.04. Mu.g/m, 0.01-0.02. Mu.g/ml.

In some embodiments, an anti-PD-L1/CD 47 bispecific antibody of the invention comprises a binding arm a that specifically binds CD47 and a binding arm B that specifically binds PD-L1; wherein the binding arm a comprises a scFab (a) comprising a light chain variable region a (VLA), a light chain constant region a (CLA), a heavy chain variable region a (VHA), and a first heavy chain constant region a (CH 1A) derived from an anti-CD 47 antibody, and a linker 1 (X1) connecting the light chain constant region a (CLA) and the heavy chain variable region a (VHA), the scFab (a) comprising an amino acid sequence as shown in SEQ ID NO:168 or an amino acid sequence having NO less than 80% sequence identity to the amino acid sequence shown in SEQ ID NO: 168; and/or the number of the groups of groups,

The binding arm B comprises a scFab (B) comprising a light chain variable region B (VLB), a light chain constant region B (CLB), a heavy chain variable region B (VHB), and a first heavy chain constant region B (CH 1B) of an anti-PD-L antibody, and a linker 2 (X2) connecting the light chain constant region B (CLB) and the heavy chain variable region B (VHB), and an Fc (B), the scFab (B) comprising an amino acid sequence as shown in SEQ ID NO:93, or an amino acid sequence having NO less than 80% sequence identity to the amino acid sequence shown in SEQ ID NO: 93.

In one embodiment, the scFab (a) and Fc (a) are fused to a single chain peptide chain a that specifically binds CD 47; the scFab (B) and Fc (B) are fused to a single-chain peptide chain B that specifically binds PD-L1; wherein the CD 47-specific binding arm a and PD-L1-specific binding arm B form heterodimers via the KIH structure formed by the Fc (a) and the Fc (B); wherein the Fc (A) comprises an amino acid sequence shown as SEQ ID NO:78 and the Fc (B) comprises an amino acid sequence shown as SEQ ID NO: 94.

In some embodiments, the anti-PD-L1/CD 47 bispecific antibodies of the invention are designed to stabilize the amino acid residues of the Fc region of the light and heavy chains for proper coupling or pairing. In some embodiments, the bispecific antibodies of the invention each comprise Y349C and S354C or S354C and Y349C, respectively, in the Fc domain; the bispecific antibodies of the invention each comprise a bulge ("knob") or a cavity ("hole") in the Fc domain, respectively, and the bulge or cavity in the Fc domain of one targeting polypeptide chain may be placed in the cavity or bulge, respectively, in the Fc domain of the other targeting polypeptide chain, whereby the former and latter targeting polypeptide chains form a stable association with each other "knob-in-hole". Illustratively, the polypeptide chain targeting binding arm B of PD-L1 comprises a Knob projection structure resulting from the S354 mutation and the T366W mutation, and the polypeptide chain targeting binding arm a of CD47 comprises a hole structure resulting from the S349 mutation and the T366S, L368A and Y407V mutations; wherein the binding arm B polypeptide chain comprises the structure VHB-CHB, specifically recognizes and binds PD-L1; the binding arm A polypeptide chain comprises the structure VHA-CHA, and can specifically recognize and bind CD47.

In some embodiments, the Fc region of a bispecific antibody of the invention comprises an amino acid modification of the binding affinity for an Fc receptor. In some embodiments, the Fc receptor is an fcγ receptor, particularly a human fcγ receptor. In one embodiment, the Fc receptor is an activating Fc receptor. In some embodiments, the modification reduces effector function of the bispecific antibodies of the invention. In some embodiments, the effector function is antibody-dependent cell-mediated cytotoxicity (ADCC). In some embodiments, the modification is within the immunoglobulin molecule Fc region, particularly within the CH2 region thereof. In some embodiments, the immunoglobulin molecule comprises an amino acid substitution at position 297 (Eu numbering) of the immunoglobulin heavy chain. In a specific embodiment, the amino acid substitution is N297A (see, e.g., J.Lund et al, oligosaccharide-protein interactions in IgG can modulate recognition by Fcgamma receptors, FASEB.J.9,115-119 (1995)).

In some embodiments, optional amino acid substitutions in the Fc region are shown in table 5 below:

TABLE 5 Fc (A) and Fc (B) amino acid substitutions and combinations of the KIH Structure of the CD47/PD-L1 bispecific antibody of the invention

Raised structures (Knob) and pore structures (Hole) can be prepared by altering the nucleotide sequence encoding a polypeptide

4) Configuration and fragment connection mode of CD47/PD-L1 double antibody

The CD47/PD-L1 bispecific antibody of the invention may be an IgG-scFv, triomab, knob-to-mortar structure with common light chain (kih) IgG, cross-smab, o-fabg, dual variable domain immunoglobulin (DVD-Ig), 2 in 1-IgG, scFv2-Fc, dual nanobody, bispecific T cell cement (BiTE), tan dabs, dual affinity re-targeting (DART) antibody, DART-Fc, scFv-Human Serum Albumin (HSA) -scFv, docking and locking (DNL) Fab3, minibody, scFv-Fc, scFv-zipper, scFv, fab, fab2 (dual specificity), fab3 (tri-specificity), scFab, bis-scFv (dual specificity), sdAb (VH/VHH), tetrabodibody, tri-antibody, diabody, camel Ig, igNAR, igG, dual specificity construct comprising a knob structure, dual specificity construct comprising a dual antibody, tetravalent multispecific antibody, tetravalent construct, tetravalent Dual Variable Domain (DVD) construct, tetravalent to IgG, tetravalent construct, tetravalent to IgG, or any combination thereof.

In an exemplary embodiment, a preferred form of the CD47/PD-L1 bispecific antibody of the invention is a KIH or "knob" arrangement in which the CH3 domain of the heavy chain of each binding arm antibody is mutated to allow heterodimerization between the heavy chain from the anti-CD 47 antibody and the heavy chain from the anti-PD-L1 antibody. Each heavy chain associates with its corresponding light chain to form one complete CD47 binding domain and one complete PD-L1 binding domain. The heavy chain CH1 region may be modified to facilitate association with the correct light chain. Bispecific antibodies in the form of KIH are well known in the art. See, e.g., ridgway et al, 1996; proteinEng9:617-621, the disclosure of which is incorporated herein by reference.

In one non-limiting embodiment, the bispecific antibody formats of the invention comprise a knob-to-hole (KiH) Fc region heterodimerization of a tandem Fab heavy chain with a truncated Fc chain, which enables simultaneous monovalent binding of two different antigens.

In one aspect, the invention provides a bispecific antibody or antigen-binding fragment comprising a binding arm a specific for CD47 and a binding arm B specific for PD-L1, wherein the binding arm a comprises a heavy chain variable region a (VHA) and a light chain variable region (VLA) and a linker 1 (X1) located between the heavy chain variable region a (VHA) and the light chain variable region (VLA), the binding arm B comprising a light chain variable region B (VHB) and a light chain variable region B (VLB) and a linker 2 (X2) located between the heavy chain variable region a (VHA) and the light chain variable region (VLA); wherein the heavy chain variable region a and the light chain variable region a pair to form a CD47 antigen binding site, and the heavy chain variable region B and the light chain variable region B pair to form a PD-L1 antigen binding site.

In one aspect, the invention provides a bispecific antibody or antigen-binding fragment comprising a binding arm a that specifically binds CD47 and a binding arm B that specifically binds PD-L1, wherein the binding arm a that specifically binds CD47 comprises an anti-CD 47 antibody light chain variable region VLA, an anti-CD 47 antibody light chain constant region CLA, an anti-CD 47 antibody heavy chain variable region VHA, and an anti-CD 47 antibody heavy chain constant region CHA; the CD 47-targeting antigen binding arm is of the structure shown in the N-terminal to C-terminal comprising the formula VLA-CLA-X1-VHA-CHA,

Wherein the "-" is a peptide bond,

the binding arm A which specifically binds CD47 comprises the amino acid sequence shown as SEQ ID NO:79 or comprises an amino acid sequence having more than 85%, preferably more than 90%, more preferably more than 95%, more preferably more than 98.5% or more preferably more than 99.5% sequence homology with the amino acid sequence shown as SEQ ID NO: 79;

and/or

The binding arm B which specifically binds to PD-L1 comprises an anti-PD-L1 antibody light chain variable region VLB, an anti-PD-L1 antibody light chain constant region VLB, an anti-PD-L1 antibody heavy chain variable region VHB and an anti-PD-L1 antibody heavy chain constant region CHB; the PD-L1 targeting antigen-binding arm is N-terminal to C-terminal comprising the structure shown by the formula VLB-CLB-X2-VHB-CHB,

the PD-L1 targeting antigen binding arm B comprises an amino acid sequence as shown in SEQ ID NO:95 or an amino acid sequence having a sequence homology of more than 85%, preferably more than 90%, more preferably more than 95%, more preferably more than 98.5% or more preferably more than 99.5% with the amino acid sequence shown in SEQ ID NO: 95.

In some embodiments, the CD 47-and PD-L1 bispecific antibodies of the invention target the binding arm of CD47 and the binding arm of PD-L1 are linked by a linker structure. The linker structure is preferably (G4S) n, which is preferably an integer between 0 and 10, more preferably 1, 2, 3 or 4, such as GGGGSGGGGSGGGGSGGGGSGGGGSGGGGS (SEQ ID NO: 196), GGGGSGGGGSGGGGSGGGGSGGGGSGGGGS (SEQ ID NO: 198), GGGGSGGGGSGGGGSGGGGSGGGGSGGGGS (SEQ ID NO: 198), and GGGGSGGGGSGGGGSGGGGSGGGGSGGGGS (SEQ ID NO: 199). Preferably, the linker is (G4S) 3. For example, the polypeptide linker may be a short connecting peptide of between about 10 and about 25 amino acids. The linker is typically glycine rich to enhance flexibility, and serine or threonine to enhance solubility, and may link the N-terminus of VH to the C-terminus of VL, and vice versa. In some embodiments the linker X1 or X2 is selected from the group consisting of a linker peptide or peptide linker as set forth in any one of SEQ ID NOs 74 or 91.

In some embodiments, the peptide linkers each independently have the structure of L1- (GGGGS) n-L2, wherein,

l1 is a bond A, GS, GGS or GGGS or (G4S) n, n is 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10, L2 is a bond G, GG, GGG or GGGGGG, and the peptide linker is not a bond. In some embodiments, the peptide linker is 3-15 amino acid residues in length. In some embodiments, the peptide linking structures each independently have the structure of (G4S) n, where n is 1, 2, or 3. In some embodiments, the peptide linking structure is GGGGS or GGGGSGGGGS.

In some embodiments, the binding arm a comprises a light chain variable region a (VLA) comprising VLA CDR1, VLA CDR2, and VLA CDR3; wherein VLA CDR1 comprises an amino acid sequence as set forth in any one of SEQ ID NOS: 36-42; VLA CDR2 comprising an amino acid sequence as set forth in any one of SEQ ID NOs 43 to 49; VLA CDR3 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 50 to 56.

In some embodiments, the binding arm a comprises VHA and VLA; the VHA comprises a VHA CDR1, a VHA CDR2, a VHA CDR3; the VLA comprises VLA CDR1, VLA CDR2 and VLA CDR3. Wherein VHA CDR1 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 1 to 7; VHA CDR2 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 8 to 14; VHA CDR3 comprises the amino acid sequence shown in SEQ ID NO 15-21; VLA CDR1 comprising an amino acid sequence as set forth in any one of SEQ ID NOS: 36-42; VLA CDR2 comprising an amino acid sequence as set forth in any one of SEQ ID NOs 43 to 49; VLA CDR3 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 50 to 56.

In some embodiments, the VHA CDR1 comprises an amino acid sequence as set forth in SEQ ID NO. 1 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 1; the VHA CDR2 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 8-14, or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in any one of SEQ ID NOs 8-14; the VHA CDR3 comprises an amino acid sequence as shown in SEQ ID NO. 15 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 15.

In one embodiment, the heavy chain a comprises a heavy chain variable region a, and the light chain a light chain variable region a, wherein:

(1) The heavy chain variable region a comprises a VHA CDR1, a VHA CDR2, and a VHA CDR3 selected from the group consisting of:

(a1) Amino acid sequences as shown in SEQ ID NO. 1, 8 and 15;

(a2) Amino acid sequences as shown in SEQ ID NO. 2, 9 and 16;

(a3) Amino acid sequences as shown in SEQ ID NO 3, 10 and 17; and

(a4) Amino acid sequences as shown in SEQ ID NO. 4, 11 and 18;

(a5) Amino acid sequences as shown in SEQ ID NO. 5, 12 and 19;

(a6) Amino acid sequences as shown in SEQ ID NO. 6, 13 and 20;

(a7) Amino acid sequences as shown in SEQ ID NO. 7, 14 and 21;

(a8) Has at least 85% amino acid sequence as shown in (a 1), (a 2), (a 3), (a 4), (a 5), (a 6) or (a 7)

Amino acid sequence of sequence identity; and

(2) The light chain variable region a comprises a VLACDR1, a VLACDR2, and a VLACDR3 selected from the group consisting of:

(a9) Amino acid sequences shown as SEQ ID NO. 36, ATS and SEQ ID NO. 50;

(a10) The amino acid sequences shown as SEQ ID NO. 37, ATS and SEQ ID NO. 51;

(a11) Amino acid sequences shown as SEQ ID NO. 38, ATS and SEQ ID NO. 52; and

(a12) Amino acid sequences shown as SEQ ID NO. 39, ATS and SEQ ID NO. 53;

(a13) Amino acid sequences shown as SEQ ID NO. 40, ATS and SEQ ID NO. 54;

(a14) Amino acid sequences shown as SEQ ID NO. 41, ATS and SEQ ID NO. 55;

(a15) Amino acid sequences as shown in SEQ ID NO. 42, ATS and SEQ ID NO. 56;

(a16) An amino acid sequence having at least 85% sequence identity to the amino acid sequence shown in (a 9), (a 10), (a 11), (a 12), (a 13), (a 14) or (a 15).

In some embodiments, a bispecific antibody or antigen-binding fragment thereof according to the invention, wherein the anti-binding arm a comprises VHA CDR1, VHA CDR2 and VHA CDR3 and VLA CDR1, VLA CDR2 and VLA CDR3, wherein:

The VHA CDR1, VHA CDR2 and VHA CDR3 are SEQ ID NO. 1, 8 and 15 or an amino acid sequence having at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO. 1, 8 and 15, respectively, and the VLA CDR1, VLA CDR2 and VLA CDR3 are SEQ ID NO. 36, ATS and an amino acid sequence having at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO. 36, ATS and SEQ ID NO. respectively; or (b)

The VHA CDR1, VHA CDR2 and VHA CDR3 are SEQ ID NO. 2, 9 and 16 or have an amino acid sequence with at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO. 2, 9 and 16, respectively, and the VLA CDR1, VLA CDR2 and VLA CDR3 are SEQ ID NO. 37, ATS and SEQ ID NO. 51 or have an amino acid sequence with at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO. 37, ATS and SEQ ID NO. 51, respectively; or (b)

The VHA CDR1, VHA CDR2 and VHA CDR3 are SEQ ID NO 3, 10 and 17 or an amino acid sequence having at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO 3, 10 and 17, respectively, and the VLA CDR1, VLA CDR2 and VLA CDR3 are SEQ ID NO 38, ATS and SEQ ID NO 52 or an amino acid sequence having at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO 38, ATS and SEQ ID NO 52, respectively; or (b)

The VHA CDR1, VHA CDR2 and VHA CDR3 are SEQ ID NO 4, 11 and 18 or have at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO 4, 11 and 18, respectively, and the VLA CDR1, VLA CDR2 and VLA CDR3 are SEQ ID NO 39, ATS and SEQ ID NO 53 or have at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO 39, ATS and SEQ ID NO 53, respectively; or (b)

The VHA CDR1, VHA CDR2 and VHA CDR3 are SEQ ID NO 5, 12 and 19 or an amino acid sequence having at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO 5, 12 and 19, respectively, and the VLA CDR1, VLA CDR2 and VLA CDR3 are SEQ ID NO 40, ATS and SEQ ID NO 54 or an amino acid sequence having at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO 40, ATS and SEQ ID NO 54, respectively, or

The VHA CDR1, VHA CDR2 and VHA CDR3 are SEQ ID NO. 6, 13 and 20 or have an amino acid sequence with at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO. 6, 13 and 20, respectively, and the VLA CDR1, VLA CDR2 and VLA CDR3 are SEQ ID NO. 41, ATS and SEQ ID NO. 55 or have an amino acid sequence with at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO. 41, ATS and SEQ ID NO. 55, respectively, or

The VHA CDR1, VHA CDR2 and VHA CDR3 are SEQ ID NO 7, 14 and 21 or an amino acid sequence having at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO 7, 14 and 21, respectively, and the VLA CDR1, VLA CDR2 and VLA CDR3 are SEQ ID NO 42, ATS and SEQ ID NO 56 or an amino acid sequence having at least 85% sequence identity to the amino acid sequences shown in SEQ ID NO 42, ATS and SEQ ID NO 56, respectively.

In an exemplary embodiment, the VHA CDR1 comprises an amino acid sequence as set forth in SEQ ID NO. 1 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 1; the VHA CDR2 comprises an amino acid sequence as shown in SEQ ID NO. 8 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 8; the VHA CDR3 comprises an amino acid sequence as shown in SEQ ID NO. 15 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 15.

In some embodiments, the VLA CDR1 comprises an amino acid sequence as set forth in SEQ ID NO. 36 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 36; the VLA CDR2 comprises an amino acid sequence as shown in ATS and the VLA CDR3 comprises an amino acid sequence as shown in SEQ ID NO:50 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO: 50.

In an exemplary embodiment, the VHA CDR1 comprises an amino acid sequence as set forth in SEQ ID NO. 2 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 2; the VHA CDR2 comprises an amino acid sequence as shown in SEQ ID NO. 9 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 9; the VHA CDR3 comprises an amino acid sequence as shown in SEQ ID NO. 16 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 16.

In some embodiments, the VLA CDR1 comprises an amino acid sequence as set forth in SEQ ID NO. 37 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 37; the VLA CDR2 comprises an amino acid sequence as shown in ATS and SEQ ID NO, or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in ATS and SEQ ID NO; the VLA CDR3 comprises an amino acid sequence as shown in SEQ ID NO. 51 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 51

In one exemplary embodiment, the VHA CDR1 comprises an amino acid sequence as set forth in SEQ ID NO. 1 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 1; the VHA CDR2 comprises an amino acid sequence as shown in SEQ ID NO. 8 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 8; the VHA CDR3 comprises an amino acid sequence as shown in SEQ ID NO. 15 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 15; the VLA CDR1 comprises an amino acid sequence as shown in SEQ ID NO. 36 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 36; the VLA CDR2 comprises an amino acid sequence as shown by ATS, or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown by ATS; the VLA CDR3 comprises an amino acid sequence as set forth in SEQ ID NO. 50 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 50.

In one embodiment, the bispecific antibody of the present invention, the binding arm B of PD-L1 comprises a heavy chain variable region VHB and a light chain variable region VLB,

wherein the VHB comprises one or more CDRs of amino acid sequences (i) - (n):

(m) a VLB CDR1 comprising the amino acid sequence as shown in SEQ ID NO. 148 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 83;

(n) a VLB CDR2 comprising an amino acid sequence as shown in SAS or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SAS;

In some embodiments, the binding arm B comprises a heavy chain variable region B (VHB) comprising VHB CDR1, VHB CDR2, VHB CDR3; wherein VHB CDR1 comprises a sequence as set forth in SEQ ID NO: 80: 80, and a sequence of amino acids shown in seq id no; VHB CDR2 comprises the amino acid sequence shown as SEQ ID NO. 81; VHB CDR3 comprises the amino acid sequence shown as SEQ ID NO. 82.

In some embodiments, the binding arm B comprises a light chain variable region B (VLb) comprising VLb CDR1, VLb CDR2, and VLb CDR3; wherein VLB CDR1 comprises the amino acid sequence as shown in SEQ ID NO. 83; VLB CDR2 comprises an amino acid sequence as shown in SAS; VLB CDR3 comprises the amino acid sequence as shown in SEQ ID NO: 84.

In some embodiments, in the binding arm a, the heavy chain a comprises a heavy chain variable region a (VHA) and the light chain a comprises a light chain variable region a (VLA); wherein, the liquid crystal display device comprises a liquid crystal display device,

(1) The VHA comprises or is selected from (A1), (A2) and (A3):

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs 22 to 28;

(A3) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (A1). In some embodiments, the binding arm a, the heavy chain a comprises a heavy chain variable region a (VHA) and the light chain a comprises a light chain variable region a (VLA), wherein,

(2) The VLA comprises or is selected from (A4), (A5) and (A6)

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 57 to 63;

(A6) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (4).

in some embodiments, the source of the variable region a anti-CD 47 antibody or antigen-binding fragment thereof is a humanized antibody or antigen-binding fragment thereof.

In some embodiments, it comprises a heavy chain variable region a and a light chain variable region a, wherein

(1) The heavy chain variable region a (VHA) comprises an amino acid sequence selected from any one of (A7) to (A9):

(A7) The amino acid sequence is shown as any one of SEQ ID NO. 22, SEQ ID NO. 23, SEQ ID NO. 24, SEQ ID NO. 25, SEQ ID NO. 26, SEQ ID NO. 27 and SEQ ID NO. 28;

(A8) An amino acid sequence obtained by substituting, deleting or adding one or more amino acids to the amino acid sequence shown in (A7) and functionally identical or similar to the amino acid sequence shown in (A7); and

(A9) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (A7); and

(2) The amino acid sequence of the light chain variable region a (VLA) comprises an amino acid sequence selected from any one of (a 10) - (a 12):

(A10) Amino acid sequences as shown in SEQ ID NO. 57, SEQ ID NO. 58, SEQ ID NO. 59, SEQ ID NO. 60, SEQ ID NO. 62, and SEQ ID NO. 63;

(A11) An amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids to the amino acid sequence shown in (A10) and has the same or similar function as the amino acid sequence shown in (A10); and

(A12) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (a 10).

In one embodiment, a bispecific antibody of the invention comprises a binding arm a specific for CD47 and a binding arm B specific for PD-L1, said anti-binding arm a comprising a heavy chain variable region a and a light chain variable region a, wherein:

the amino acid sequence of the heavy chain variable region A (VHA) is SEQ ID NO. 22, the amino acid sequence obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 22 and having the same function as the SEQ ID NO. 22 or the amino acid sequence having at least 85% sequence identity with the SEQ ID NO. 22, and the amino acid sequence of the light chain variable region A (VLA) is SEQ ID NO. 57, the amino acid sequence obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 57 and having the same function as the SEQ ID NO. 57 or the amino acid sequence having at least 85% sequence identity with the SEQ ID NO. 57; or (b)

The amino acid sequence of the heavy chain variable region A is an amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in SEQ ID NO. 24 and has the same function as SEQ ID NO. 24 or an amino acid sequence which has at least 85 percent sequence identity with SEQ ID NO. 24, the amino acid sequence of the light chain variable region A is SEQ ID NO. 59, and the amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in SEQ ID NO. 59 and has the same function as SEQ ID NO. 59 or an amino acid sequence which has at least 85 percent sequence identity with SEQ ID NO. 52; or (b)

The amino acid sequence of the heavy chain variable region A is SEQ ID NO. 23, an amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 23 and has the same function as the SEQ ID NO. 23 or an amino acid sequence which has at least 85 percent sequence identity with the SEQ ID NO. 58, and the amino acid sequence of the light chain variable region A is SEQ ID NO. 58, an amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 58 and has the same function as the SEQ ID NO. 58 or an amino acid sequence which has at least 85 percent sequence identity with the SEQ ID NO. 26; or (b)

The amino acid sequence of the heavy chain variable region A is SEQ ID NO. 27, an amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 27 and has the same function as the SEQ ID NO. 27 or an amino acid sequence which has at least 85% sequence identity with the SEQ ID NO. 27, and the amino acid sequence of the light chain variable region A is SEQ ID NO. 55, an amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 55 and has the same function as the SEQ ID NO. 55 or an amino acid sequence which has at least 85% sequence identity with the SEQ ID NO. 55; or (b)

The amino acid sequence of the heavy chain variable region A is SEQ ID NO. 28, the amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 28 and has the same function as the SEQ ID NO. 28 or the amino acid sequence which has at least 85 percent of sequence identity with the SEQ ID NO. 28, and the amino acid sequence of the light chain variable region A is SEQ ID NO. 63, the amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 63 and has the same function as the SEQ ID NO. 63 or the amino acid sequence which has at least 85 percent of sequence identity with the SEQ ID NO. 63.

In an exemplary embodiment, the CD 47-specific binding arm A comprises a heavy chain variable region A (VHA) and a light chain variable region A (VLA), wherein,

1) The amino acid sequence of heavy chain variable region a comprises a sequence selected from the group consisting of:

(B1) An amino acid sequence as set forth in any one of SEQ ID NOs.22;

(B3) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (B1);

and/or

(2) The VLA comprises or is selected from:

(B4) An amino acid sequence as set forth in any one of SEQ ID NOs 57;

(B5) An amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids to the amino acid sequence shown in (B4) and has the same or similar function as the amino acid sequence shown in (B4); and

In some embodiments, provided herein are bispecific antibodies comprising a binding arm a that specifically binds CD47 and a binding arm B that specifically binds PD-L1, wherein the binding arm a comprises a heavy chain variable region a (VHA) comprising a sequence as set forth in SEQ ID NO: or 22, 23, 24, 25, 26, 27 or 28 (VHA CDR 1), VHA complementarity determining region 2 (VHA CDR 2), and VHA complementarity determining region 3 (VHA CDR 3). Optionally, the VHA comprises (i) a sequence comprising SEQ ID NO: 1. 2, 3, 4, 5, 6 or 7; (ii) a polypeptide comprising SEQ ID NO: 8. 9, 10, 11, 12, 13 or 14; and (iii) a polypeptide comprising SEQ ID NO: 15. 16, 17, 18, 19, 20 or 21. Optionally, the CD47 binding arm further comprises a light chain variable region a (VLA) comprising the amino acid sequence as set forth in SEQ ID NO: 57. 58, 59, 60, 61, 62 or 63, a VLA complementarity determining region 1 (VLA CDR 1), a VLA complementarity determining region 2 (VLA CDR 2), and a VLA complementarity determining region 3 (VLA CDR 3) of a VLA sequence as set forth in any one of claims 58, 59, 60, 61, 62 or 63. Optionally, the VLA comprises (i) a nucleic acid comprising SEQ ID NO: 36. 37, 39, 40, 41 or 42; (ii) VLA CDR2 comprising the amino acid sequence depicted by ATS; and (iii) a polypeptide comprising SEQ ID NO: 50. 51, 52, 53, 54, 55, or 56.

In some embodiments, the disclosure relates to an antibody that specifically binds to CD47, wherein the CD47 antibody comprises a heavy chain variable region a (VHA) and a light chain variable region a (VLA), the VHA comprising (i) a heavy chain variable region comprising SEQ ID NO: 1. 2 or 3, and a VHA CDR1 of the amino acid sequence shown in seq id no; (ii) a polypeptide comprising SEQ ID NO:8 or 9; and (iii) a polypeptide comprising SEQ ID NO:15, and a VHA CDR3 of the amino acid sequence depicted in seq id no. Optionally, the VLA comprises (i) a nucleic acid comprising SEQ ID NO:36 or 37, and a VLA CDR1 of the amino acid sequence shown in seq id no; (ii) VLA CDR2 comprising the amino acid sequence depicted by ATS; and (iii) a polypeptide comprising SEQ ID NO:50 or 51, and a VLA CDR3 of the amino acid sequence depicted in seq id no. Optionally, the VHA comprises SEQ ID NO: 22. 23, 24, 25, 26, 27 or 28. Optionally, VAL comprises SEQ ID NO: 57. 58, 59, 60, 61, 62 or 63. Optionally, the VHA comprises SEQ ID NO:22 or 23 and VLA comprises the amino acid sequence of SEQ ID NO: 57. Optionally, one or both of the following: (i) VHA comprises SEQ ID NO:22, wherein the variant sequence comprises an amino acid sequence that hybridizes to SEQ ID NO:22, compared to one, two, three, four, five, six, seven or eight amino acid substitutions; and (ii) the VLA comprises SEQ ID NO:57, wherein the variant sequence comprises an amino acid sequence that hybridizes to SEQ ID NO:57, compared to one, two, three, four, five, six, seven or eight amino acid substitutions. Optionally, the substitution is a conservative amino acid substitution. Optionally, the substitution is in a residue within a non-CDR region. Optionally, the substitution is in a residue within a CDR region.

(B1) An amino acid sequence as shown in SEQ ID NO. 85;

The VLB comprises or is selected from:

(B4) An amino acid sequence as shown in SEQ ID NO. 86;

In some embodiments, the bispecific antibody comprises a CD 47-specific binding arm a comprising a heavy chain variable region a (VHA) and a light chain variable region a (VLA) and a PD-L1-specific binding arm B comprising a heavy chain variable region B (VHB) and a light chain variable region B (VLB), wherein:

The VHA comprises an amino acid sequence shown as SEQ ID NO. 22;

the PD-L1 specific binding arm B comprises a heavy chain variable region B (VHB) and a light chain variable region B (VLB); wherein the VHB comprises an amino acid sequence shown as SEQ ID NO. 85;

the VLB comprises the amino acid sequence as shown in SEQ ID NO. 86.

(1) The amino acid sequence of the heavy chain variable region B comprises an amino acid sequence shown as SEQ ID NO. 85;

and

(2) The amino acid sequence of the light chain variable region B comprises the amino acid sequence shown as SEQ ID NO. 86.

In a specific embodiment, the CD 47-specific binding arm a comprises a sequence comprising SEQ ID NO:29 and a light chain of the amino acid sequence shown in SEQ ID NO. 64; and/or

The PD-L1 specifically binds to arm B comprising SEQ ID NO:87 and the heavy chain of the amino acid sequence of SEQ ID NO:88, and a light chain of the amino acid sequence of seq id no. Wherein the VHA comprises an amino acid sequence shown as SEQ ID NO. 22;

the HA comprises a heavy chain variable region A (VHA) comprising an amino acid sequence as set forth in any one of SEQ ID NOs.22;

the LA comprises a light chain variable region A (VLA) comprising an amino acid sequence as set forth in any one of SEQ ID NOs 57;

The HB comprises an amino acid sequence heavy chain variable region B (VHB) as shown in SEQ ID NO. 85;

the VLB comprises the light chain variable region B (VLB) of the amino acid sequence shown as SEQ ID NO. 86.

In some embodiments, provided herein is a CD47/PD-L1 bispecific antibody comprising an antigen-binding arm a that specifically binds to CD47 and an antigen-binding arm B that specifically binds to PD-L1, wherein antigen-binding arm a comprises VHA and VLA, wherein antigen-binding arm B comprises VHB and VLB, and wherein one, two, three, or all four of: (a) the VHA of antigen binding arm a comprises the amino acid sequence as set forth in SEQ ID NO: 22. 23, 24, 25, 26 or 27, a VHA CDR1, a VHA CDR2, and a VHA CDR3 of a VHA sequence shown in seq id no; (b) VLA of antigen binding arm a comprises the amino acid sequence as set forth in SEQ ID NO: 57. 58, 59, 60, 61, 62 or 63, a VLA CDR1, a VLA CDR2 and a VLA CDR3 of a VLA sequence shown in seq id no; (c) the VHB of antigen binding arm B comprises the amino acid sequence as set forth in SEQ ID NO:85, a VHB CDR1, a VHB CDR2, and a VHB CDR3 of a VHB sequence; and (d) the VLB of antigen binding arm B comprises the amino acid sequence as set forth in SEQ ID NO:86, VLB CDR1, VLB CDR2, and VLB CDR3 of the VLB sequence shown.

Optionally, one, two, three or all four of the following: (a) The heavy chain variable region VHA of antigen binding arm a comprises (i) a heavy chain variable region comprising SEQ ID NO: 1. 2, 3, 4, 5, 6 or 7; (ii) a polypeptide comprising SEQ ID NO: 8. 9, 10, 11, 12, 13 or 14; and (iii) a polypeptide comprising SEQ ID NO: 15. 16, 17, 18, 19, 20 or 21; (b) The variable region VLA of antigen binding arm a comprises (i) a polypeptide comprising SEQ ID NO: 36. 37, 38, 39, 40, 41, 42 or 43; (ii) VLA CDR2 comprising the amino acid sequence depicted by ATS; and (iii) a polypeptide comprising SEQ ID NO: 50. 51, 52, 53, 54, 55, or 56; (c) The variable region VHB of antigen binding arm B comprises (i) a sequence comprising SEQ ID NO:80, a VHB CDR1 of the amino acid sequence of seq id no; (ii) a polypeptide comprising SEQ ID NO:81, a VHB CDR2 of the amino acid sequence of seq id no; and (iii) a polypeptide comprising SEQ ID NO:82, a VHB CDR3 of the amino acid sequence of seq id no; and (d) the variable region VLB of antigen binding arm B comprises (i) a polypeptide comprising SEQ ID NO:83, a VLB CDR1 of the amino acid sequence of seq id no; (ii) VLB CDR2 comprising the amino acid sequence of ATS; and (iii) a polypeptide comprising SEQ ID NO:84, and VLB CDR3 of the amino acid sequence of seq id no.

Optionally, one, two, three or all four of the following: (a) The heavy chain variable region VHA of antigen binding arm a comprises (i) a heavy chain variable region comprising SEQ ID NO:1 or 2, and a VHA CDR1 of the amino acid sequence of 1 or 2; (ii) a polypeptide comprising SEQ ID NO:8 or 9, and a VHA CDR2 of the amino acid sequence of seq id no; and (iii) a polypeptide comprising SEQ ID NO:15, a VHA CDR3 of the amino acid sequence of seq id no; (b) The light chain variable region VLA of antigen binding arm a comprises (i) a light chain comprising SEQ ID NO:36, a VLA CDR1 of the amino acid sequence of seq id no; (ii) VLA CDR2 comprising the amino acid sequence of ATS; and (iii) a polypeptide comprising SEQ ID NO:50, a VLA CDR3 of the amino acid sequence of seq id no; (c) The heavy chain variable region VHB of antigen binding arm B comprises (i) a heavy chain variable region comprising SEQ ID NO:80, a VHB CDR1 of the amino acid sequence of seq id no; (ii) a polypeptide comprising SEQ ID NO:81, a VHB CDR2 of the amino acid sequence of seq id no; and (iii) a polypeptide comprising SEQ ID NO:82, a VHB CDR3 of the amino acid sequence of seq id no; and (d) the light chain variable region VLB of antigen binding arm B comprises (i) a light chain variable region comprising SEQ ID NO:83 VLB CDR1 of amino acid sequence; (ii) VLB CDR2 comprising the amino acid sequence of SAS; and (iii) a polypeptide comprising SEQ ID NO:84, and VLB CDR3 of the amino acid sequence of seq id no.

In some embodiments, a bispecific antibody according to the invention comprises a binding arm a specific for CD47 and a binding arm B specific for PD-L1, wherein the binding arm specific for anti-CD 47 comprises a heavy chain variable region a and a light chain variable region a, wherein:

(1) The amino acid sequence of the heavy chain variable region a is selected from the group consisting of:

(b1) The amino acid sequences shown in SEQ ID NO. 22, SEQ ID NO. 23, SEQ ID NO. 24, SEQ ID NO. 25, SEQ ID NO. 26, SEQ ID NO. 27, SEQ ID NO. 28;

(b2) An amino acid sequence obtained by substituting, deleting or adding one or more amino acids in the amino acid sequence shown in (b 1) and functionally identical or similar to the amino acid sequence shown in (b 1); and

(b3) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (b 1); and

(2) The amino acid sequence of the light chain variable region a is selected from the group consisting of:

(b4) Amino acid sequences as shown in SEQ ID NO. 57, SEQ ID NO. 58, SEQ ID NO. 59, SEQ ID NO. 60, SEQ ID NO. 62, and SEQ ID NO. 63;

(b5) An amino acid sequence obtained by substituting, deleting or adding one or more amino acids to the amino acid sequence shown in (b 4) and functionally identical or similar to the amino acid sequence shown in (b 4); and

(b6) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (b 4).

In some embodiments, an anti-CD 47/anti-PD-L1 bispecific antibody according to the invention comprises a binding arm a specific for CD47 and a binding arm B specific for PD-L1, wherein the anti-CD 47 specific binding arm comprises a heavy chain variable region a and a light chain variable region a, wherein:

The amino acid sequence of the heavy chain variable region A is SEQ ID NO. 22, the amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 22 and has the same function as the SEQ ID NO. 22 or the amino acid sequence which has at least 85 percent of sequence identity with the SEQ ID NO. 22, the amino acid sequence of the light chain variable region A is SEQ ID NO. 57, and the amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 57 and has the same function as the SEQ ID NO. 57 or the amino acid sequence which has at least 85 percent of sequence identity with the SEQ ID NO. 57; or (b)

The amino acid sequence of the heavy chain variable region A is SEQ ID NO. 22, an amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in SEQ ID NO. 24 and has the same function as SEQ ID NO. 24 or an amino acid sequence which has at least 85% sequence identity with SEQ ID NO. 24, and the amino acid sequence of the light chain variable region A is SEQ ID NO. 59, an amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in SEQ ID NO. 59 and has the same function as SEQ ID NO. 59 or an amino acid sequence which has at least 85% sequence identity with SEQ ID NO. 52; or (b)

In some embodiments, the bispecific antibody according to the invention, the anti-CD 47 antibody or antigen-binding fragment thereof is a humanized antibody or antigen-binding fragment thereof, the CD 47-specific binding arm a comprising a heavy chain variable region a and a light chain variable region a, wherein

The amino acid sequence of the heavy chain variable region A is SEQ ID NO. 22, the amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 22 and has the same function as the SEQ ID NO. 22 or the amino acid sequence which has at least 85 percent of sequence identity with the SEQ ID NO. 22, and the amino acid sequence of the light chain variable region A is SEQ ID NO. 57, the amino acid sequence which is obtained by substituting, deleting or adding one or more amino acids in the SEQ ID NO. 57 and has the same function as the SEQ ID NO. 57 or the amino acid sequence which has at least 85 percent of sequence identity with the SEQ ID NO. 57.

(c1) An amino acid sequence as shown in SEQ ID NO. 85;

(c2) An amino acid sequence obtained by substituting, deleting or adding one or more amino acids in the amino acid sequence shown in (c 1) and functionally identical or similar to the amino acid sequence shown in (c 1); and

(c3) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (c 1); and/or

The VLB comprises or is selected from:

(c4) An amino acid sequence as shown in SEQ ID NO. 86;

(c5) An amino acid sequence obtained by substituting, deleting or adding one or more amino acids to the amino acid sequence shown in (c 4) and functionally identical or similar to the amino acid sequence shown in (c 4); and

(c6) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (c 4).

In one embodiment, the VHA comprises the amino acid sequence shown as SEQ ID NO. 22; the VLA comprises an amino acid sequence comprising the sequence shown as SEQ ID NO. 57; the HB comprises an amino acid sequence shown as SEQ ID NO. 85; the VLB comprises the amino acid sequence as shown in SEQ ID NO. 86.

(1) The amino acid sequence of the heavy chain variable region a comprises:

an amino acid sequence as shown in SEQ ID NO. 22; and (2) the VLA comprises

SEQ ID NO. 57; and

the PD-L1 specific binding arm B comprises a heavy chain variable region B (VHB) and a light chain variable region B (VLB); wherein the VHB comprises:

an amino acid sequence as shown in SEQ ID NO. 85; and

the VLB comprises the amino acid sequence as shown in SEQ ID NO. 86.

Optionally, the VHA comprises SEQ ID NO: 22. 23, 24, 25, 26, 27 or 28. Optionally, VAL comprises SEQ ID NO: 57. 58, 59, 60, 61, 62 or 63. Optionally, the VHA comprises SEQ ID NO:22 or 23 and VA comprises the amino acid sequence of SEQ ID NO: 57. Optionally, one or both of the following: (i) VHA comprises SEQ ID NO:22, wherein the variant sequence comprises an amino acid sequence that hybridizes to SEQ ID NO:22, compared to one, two, three, four, five, six, seven or eight amino acid substitutions; and (ii) the VLA comprises SEQ ID NO:57, wherein the variant sequence comprises an amino acid sequence that hybridizes to SEQ ID NO:57, compared to one, two, three, four, five, six, seven or eight amino acid substitutions. Optionally, the substitution is a conservative amino acid substitution. Optionally, the substitution is in a residue that is not within a CDR region.

Optionally, the amino acid sequence of VLA of antigen binding arm a comprises the amino acid sequence as set forth in SEQ ID NO: 57. 58, 59, or the amino acid sequence shown in SEQ ID NO: 57. 58, 59, 60, 61, 62 or 63 comprise variants thereof with one, two, three, four, five, six, seven or eight amino acid substitutions. Optionally, the amino acid sequence of VLB of antigen binding arm B comprises the amino acid sequence as set forth in SEQ ID NO:86 or an amino acid sequence as set forth in SEQ ID NO:86 comprise variants thereof with one, two, three, four, five, six, seven or eight amino acid substitutions. Optionally, the substitution is a conservative amino acid substitution. Optionally, the substitution is in a residue that is not within a CDR region.

In some embodiments, provided herein is an isolated antibody that specifically binds to CD47 comprising: VHA comprising the amino acid sequence of SEQ ID NO:22, an amino acid sequence of seq id no; and VLA comprising SEQ ID NO: 57. In some embodiments, provided herein is an antibody that specifically binds to CD47, wherein the antibody comprises: VHA comprising the amino acid sequence of SEQ ID NO:23, an amino acid sequence of seq id no; and VLA comprising SEQ ID NO:58, and a sequence of amino acids.

In some embodiments, provided herein is an isolated antibody that specifically binds to CD47, wherein the antibody comprises a VHA comprising a VHA sequence of S (SEQ ID NO: 22) or a variant thereof, wherein the sequence is set forth in SEQ ID NO:22 having a variant amino acid at one or more of the positions; comprising (i) a sequence comprising SEQ ID NO:1 or 2, and a VHA CDR1 of the amino acid sequence of 1 or 2; (ii) a polypeptide comprising SEQ ID NO:8, a VHA CDR2 of the amino acid sequence of seq id no; and (iii) a polypeptide comprising SEQ ID NO:15 or 16, and a VLA CDR3 of the amino acid sequence of 15 or 16. . Optionally, the antibody further comprises a VLA comprising (i) a VLA sequence comprising (SEQ ID NO: 57) or a variant thereof, optionally the VLA comprises the amino acid sequence of SEQ ID NO:57 or 58, wherein said sequence is set forth in SEQ ID NO:57 has a variant amino acid at one or more of positions; which comprises SEQ ID NO:36 or 37, and a VLA CDR1 of the amino acid sequence of 36 or 37; (ii) VLA CDR2 comprising the amino acid sequence of ATS; and (iii) a polypeptide comprising SEQ ID NO:50 or 51, and a VLA CDR3 of the amino acid sequence of seq id no.

In some embodiments, the invention provides an isolated bispecific antibody comprising an antigen-binding arm a that specifically binds to CD47 and an antigen-binding arm B that specifically binds to PD-L1, wherein the bispecific antibody comprises: comprising the amino acid sequence as set forth in SEQ ID NO:22, a heavy chain variable region a (VHA) of an antigen binding arm a comprising an amino acid sequence as set forth in SEQ ID NO:57, a light chain variable region a (VLA) of antigen binding arm a comprising an amino acid sequence as set forth in SEQ ID NO:85, and a heavy chain variable region B (VHB) of an antigen binding arm B comprising an amino acid sequence as set forth in SEQ ID NO:86, and a light chain variable region B (VLB) of an antigen binding arm B of the amino acid sequence shown in seq id no. In some embodiments, provided herein is an isolated bispecific antibody comprising an antigen-binding arm a that specifically binds to CD47 and an antigen-binding arm B that specifically binds to PD-L1, wherein the antibody comprises: comprising the amino acid sequence as set forth in SEQ ID NO:23, a heavy chain variable region a (VHA) of an antigen binding arm a comprising an amino acid sequence as set forth in SEQ ID NO:57, a light chain variable region a (VLA) of antigen binding arm a comprising an amino acid sequence as set forth in SEQ ID NO:85, and a heavy chain variable region B (VHB) of an antigen binding arm B comprising an amino acid sequence as set forth in SEQ ID NO:86, and a light chain variable region B (VLB) of an antigen binding arm B of the amino acid sequence shown in seq id no.

In some embodiments, provided herein is an isolated bispecific antibody comprising an antigen-binding arm a that specifically binds to CD47 and an antigen-binding arm B that specifically binds to PD-L1, wherein antigen-binding arm a comprises a heavy chain variable region VHA and a light chain variable region VLA, wherein antigen-binding arm B comprises a heavy chain variable region VHB and a light chain variable region VLB, and wherein (a) the heavy chain variable region VHA of antigen-binding arm a comprises (i) a light chain variable region comprising SEQ ID NO: 1. 2 or 3, and a VHA CDR1 of the amino acid sequence of seq id no; (ii) a polypeptide comprising SEQ ID NO:8 or 9, and a VHA CDR2 of the amino acid sequence of seq id no; and (iii) a polypeptide comprising SEQ ID NO:15, a VHA CDR3 of the amino acid sequence of seq id no; and (B) the heavy chain variable region VHB of antigen binding arm B comprises (i) a heavy chain variable region comprising SEQ ID NO:80, a VHB CDR1 of the amino acid sequence of seq id no; (ii) a polypeptide comprising SEQ ID NO:81, a VHB CDR2 of the amino acid sequence of seq id no; and (iii) a polypeptide comprising SEQ ID NO:82, and a VHB CDR3 of the amino acid sequence of seq id no. Optionally, the heavy chain variable region VHA of antigen binding arm a comprises the amino acid sequence as set forth in SEQ ID NO:22 and the heavy chain variable region VHB of antigen binding arm B comprises the amino acid sequence set forth in SEQ ID NO: 85. Optionally, (a) the light chain variable region VLA of antigen binding arm a comprises (i) a light chain variable region comprising SEQ ID NO:36 or 37, and a VLA CDR1 of the amino acid sequence of 36 or 37; (ii) VLA CDR2 comprising the amino acid sequence of ATS; and (iii) a polypeptide comprising SEQ ID NO:50 or 51, and a VLA CDR3 of the amino acid sequence of seq id no; and (B) the light chain variable region VLB of antigen binding arm B comprises (i) a light chain variable region comprising SEQ ID NO:83, a VLB CDR1 of the amino acid sequence of seq id no; (ii) VLB CDR2 comprising the amino acid sequence of SAS; and (iii) a polypeptide comprising SEQ ID NO:84, and VLB CDR3 of the amino acid sequence of seq id no. Optionally, the light chain variable region VLA of antigen binding arm a comprises the amino acid sequence as set forth in SEQ ID NO:57 and the light chain variable region VLB of antigen binding arm B comprises the amino acid sequence as set forth in SEQ ID NO:86, and a sequence of amino acids shown in seq id no.

In some embodiments, the isolated bispecific antibody, wherein the CD 47-specific binding arm a comprises a heavy chain variable region a (VHA) and a light chain variable region a (VLA), the VHA comprising or being selected from:

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs.22;

and/or

The VLA comprises or is selected from:

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 57;

(A6) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (A4); and

the PD-L1 specific binding arm B comprises a heavy chain variable region B (VHB) and a light chain variable region B (VLB); wherein the VHB comprises or is selected from:

(c1) An amino acid sequence as shown in SEQ ID NO. 85;

The VLB comprises or is selected from:

(c4) An amino acid sequence as shown in SEQ ID NO. 86;

In one embodiment, the CD 47-specific binding arm A comprises a heavy chain variable region A (VHA) and a light chain variable region A (VLA), and the PD-L1-specific binding arm B comprises a heavy chain variable region B (VHB) and a light chain variable region (VLB), wherein the VHA comprises the amino acid sequence set forth in SEQ ID NO. 22; the VLA comprises an amino acid sequence as set forth in any one of SEQ ID NOs 57; the PD-L1 specific binding arm B comprises a heavy chain variable region B (VHB) and a light chain variable region B (VLB); wherein the VHB comprises an amino acid sequence shown as SEQ ID NO. 85; the VLB comprises the amino acid sequence as shown in SEQ ID NO. 86.

In some embodiments, provided herein is a bispecific antibody comprising an antigen-binding arm a that specifically binds to CD47 and an antigen-binding arm B that specifically binds to PD-L1, wherein antigen-binding arm a comprises VHA and VLA, wherein antigen-binding arm B comprises VHB and VLB, and the binding arm a, binding arm B comprises one, two, three, or all four of: (a) the VHA of antigen binding arm a comprises the amino acid sequence as set forth in SEQ ID NO: 22. 23, 24, 25, 26 or 27, a VHA CDR1, a VHA CDR2, and a VHA CDR3 of a VHA sequence shown in seq id no; (b) the antigen binding arm AVLA comprises the amino acid sequence as set forth in SEQ ID NO: 57. 58, 59, 60, 61, 62 or 63, a VLA CDR1, a VLA CDR2 and a VLA CDR3 of a VLA sequence shown in seq id no; (c) antigen binding arm BVHB comprises an amino acid sequence as set forth in SEQ ID NO:85, a VHB CDR1, a VHB CDR2, and a VHB CDR3 of a VHB sequence; and (d) the VLB of antigen binding arm B comprises the amino acid sequence as set forth in SEQ ID NO:86, VLB CDR1, VLB CDR2, and VLB CDR3 of the VLB sequence shown. Optionally, one, two, three or all four of the following: (a) The heavy chain variable region VHA of antigen binding arm a comprises (i) a heavy chain variable region comprising SEQ ID NO: 1. 2, 3, 4, 5, 6 or 7; (ii) a polypeptide comprising SEQ ID NO: 8. 9, 10, 11, 12, 13 or 14; and (iii) a polypeptide comprising SEQ ID NO: 15. 16, 17, 18, 19, 20 or 21; (b) The variable region VLA of antigen binding arm a comprises (i) a polypeptide comprising SEQ ID NO: 36. 37, 38, 39, 40, 41, 42 or 43; (ii) VLA CDR2 comprising the amino acid sequence of ATS; and (iii) a polypeptide comprising SEQ ID NO: 50. 51, 52, 53, 54, 55, or 56; (c) The variable region VHB of antigen binding arm B comprises (i) a sequence comprising SEQ ID NO:80, a VHB CDR1 of the amino acid sequence of seq id no; (ii) a polypeptide comprising SEQ ID NO:81, a VHB CDR2 of the amino acid sequence of seq id no; and (iii) a polypeptide comprising SEQ ID NO:82, a VHB CDR3 of the amino acid sequence of seq id no; and (d) the variable region VLB of antigen binding arm B comprises (i) a polypeptide comprising SEQ ID NO:83, a VLB CDR1 of the amino acid sequence of seq id no; (ii) VLB CDR2 comprising the amino acid sequence of SAS; and (iii) a polypeptide comprising SEQ ID NO:84, and VLB CDR3 of the amino acid sequence of seq id no. Optionally, one, two, three or all four of the following: (a) The heavy chain variable region VHA of antigen binding arm a comprises (i) a heavy chain variable region comprising SEQ ID NO:1 or 2, and a VHA CDR1 of the amino acid sequence of 1 or 2; (ii) a polypeptide comprising SEQ ID NO:8 or 9, and a VHA CDR2 of the amino acid sequence of seq id no; and (iii) a polypeptide comprising SEQ ID NO:15, a VHA CDR3 of the amino acid sequence of seq id no; (b) The light chain variable region VLA of antigen binding arm a comprises (i) a light chain comprising SEQ ID NO:36, a VLA CDR1 of the amino acid sequence of seq id no; (ii) VLA CDR2 comprising the amino acid sequence of ATS; and (iii) a polypeptide comprising SEQ ID NO:50, a VLA CDR3 of the amino acid sequence of seq id no; (c) The heavy chain variable region VHB of antigen binding arm B comprises (i) a heavy chain variable region comprising SEQ ID NO:80, a VHB CDR1 of the amino acid sequence of seq id no; (ii) a polypeptide comprising SEQ ID NO:81, a VHB CDR2 of the amino acid sequence of seq id no; and (iii) a polypeptide comprising SEQ ID NO:82, a VHB CDR3 of the amino acid sequence of seq id no; and (d) the light chain variable region VLB of antigen binding arm B comprises (i) a light chain variable region comprising SEQ ID NO:83 VLB CDR1 of amino acid sequence; (ii) VLB CDR2 comprising the amino acid sequence of SAS; and (iii) a polypeptide comprising SEQ ID NO:84, and VLB CDR3 of the amino acid sequence of seq id no.

In some embodiments, a bispecific antibody according to the invention comprises a CD47 specific binding arm a comprising a heavy chain a (HA) and a light chain a (LA) derived from a CD47 antibody and a PD-L1 specific binding arm B comprising a heavy chain B (HB) and a light chain B (LB) derived from a PD-L1 antibody, wherein the heavy chain a (HA) comprises an amino acid sequence as set forth in any one of SEQ ID NOs 29-35, or an amino acid sequence having at least 90% homology to an amino acid sequence set forth in any one of SEQ ID NOs 29-35; and/or the light chain A (LA) comprises an amino acid sequence as set forth in any one of SEQ ID NOS: 64-70, or an amino acid sequence having at least 80% homology with the amino acid sequence set forth in any one of SEQ ID NOS: 64-70.

In some embodiments, a bispecific antibody according to the invention, wherein the heavy chain B (HB) comprises the amino acid sequence shown in SEQ ID NO. 87, or an amino acid sequence having at least 80% homology with the amino acid sequence shown in SEQ ID NO. 87; the light chain B (LB) comprises an amino acid sequence shown as SEQ ID NO. 88 or an amino acid sequence with at least 80% homology with the amino acid sequence shown as SEQ ID NO. 154.

In one embodiment, the anti-CD 47 specific binding arm a comprises a sequence comprising SEQ ID NO:29 and the light chain of the amino acid sequence shown in SEQ ID NO. 64.

In one embodiment the PD-L1 specific binding arm B comprises a sequence comprising SEQ ID NO:87 and the heavy chain of the amino acid sequence of SEQ ID NO:88, and a light chain of the amino acid sequence of seq id no.

In one embodiment, the anti-CD 47 specific binding arm a comprises a sequence comprising SEQ ID NO:29 and the light chain of the amino acid sequence shown in SEQ ID No. 64, said PD-L1 specific binding arm B comprising a heavy chain comprising the amino acid sequence shown in SEQ ID NO:87 and the heavy chain of the amino acid sequence of SEQ ID NO:88, and a light chain of the amino acid sequence of seq id no.

(III) constant region

In one embodiment, the bispecific antibody or antigen binding fragment of the present invention, the binding arm A further comprises a heavy chain constant region A (CHA) as shown in SEQ ID NO. 75 and a light chain constant region (CLA) as shown in SEQ ID NO. 73, or an amino acid sequence having 90% -99.5% homology to the heavy chain constant region A (CHA) as shown in SEQ ID NO. 75 and the light chain constant region (CLA) as shown in SEQ ID NO. 73, the binding arm B comprises a heavy chain constant region B (CHB) as shown in SEQ ID NO. 90 and a light chain constant region B (CLB) as shown in SEQ ID NO. 89, or an amino acid sequence having 90% -99.5% homology to the heavy chain constant region B (CHB) as shown in SEQ ID NO. 90 and the light chain constant region (CLA) as shown in SEQ ID NO. 89, respectively.

In an exemplary BsAb-463 diabody embodiment, the heavy chain constant region CHA of the CD47 binding arm is of the IgG1 subtype (SEQ ID NO: 75). In some embodiments, the heavy chain constant region of the CD47 binding arm comprises one or more of the following amino acid mutations: N297A, Y349C, S354C, T366W, T366S, L a and Y407V.

In some embodiments, an anti-PD-L1/CD 47 bispecific antibody comprises a CHA as set forth in SEQ ID NO. 75 and a CHB as set forth in SEQ ID NO. 90; and CLA and CLB as shown in SEQ ID NOS: 73 and 89.

In some embodiments, the light chain constant region CLA of the anti-CD 47 antibody and CLB of the anti-PD-L1 antibody are optionally kappa chain or lambda chain constant regions, and the heavy chain constant region CHA and the CHB are optionally IgG1 or IgG4 constant regions. In some embodiments, the anti-CD 47 and PD-L1 bispecific antibodies of the invention comprise a light chain constant region of murine or human kappa, lambda chain or variant thereof, and/or further comprise a heavy chain constant region of murine or human IgG1, igG2, igG3 or IgG4 or variant thereof.

In some embodiments, the light chain constant region CLA of the anti-CD 47 antibody and CLB of the anti-PD-L1 antibody are optionally human antibody kappa or lambda chain constant regions, and the heavy chain constant region CHA and the CHB are optionally human antibody IgG1 or IgG4 constant regions.

In some preferred embodiments, the anti-CD 47 and PD-L1 bispecific antibodies of the invention are humanized antibodies, wherein the heavy chain comprises a heavy chain constant region of an optional human IgG1, igG2, igG3, igG4 or variant thereof, and the light chain comprises a light chain constant region of an optional human kappa, lambda chain or variant thereof. The variants are constant regions modified with one or more amino acids. In some embodiments, the modification refers to substitution of up to 1, 2, 3, 4, 5, or 6 amino acids.

In an exemplary embodiment, the anti-CD 47 and PD-L1 bispecific antibodies of the present invention are humanized antibodies, wherein the heavy chain comprises an optional human IgG1 heavy chain constant region and the light chain comprises a light chain constant region of an optional human kappa chain or variant thereof.

In an exemplary embodiment, the anti-CD 47 and PD-L1 bispecific antibodies of the present invention are humanized antibodies, wherein the heavy chain comprises an optional human IgG1 heavy chain constant region and the light chain comprises a light chain constant region of an optional human lambda chain or variant thereof.

In an exemplary embodiment BsAb-463, the anti-CD 47-L1/PD-L1 bispecific antibody comprises two different heavy chains and two different light chains, the CD47 targeting heavy chain A (HA) as shown in SEQ ID NO. 29, the light chain A (LA) as shown in SEQ ID NO. 64, the PD-L1 targeting heavy chain B (HB) as shown in SEQ ID NO. 87, and the light chain B (LB) as shown in SEQ ID NO. 88.

Exemplary constant regions and full length sequences are shown in table 6 below.

TABLE 6 constitutive fragments and amino acid sequences of the CD47 binding arm and the PD-L1 binding arm

/>

Constant regions can mediate a variety of effector functions. For example, the Fc region of an antibody may bind to cells expressing an Fc receptor (FcR). There are many Fc receptors specific for different classes of antibodies, including IgG (gamma receptor), igE (epsilon receptor), igA (alpha receptor), and IgM (mu receptor). Binding of antibodies to cell surface Fc receptors triggers a number of important and diverse biological responses including phagocytosis and destruction of antibody-coated particles, clearance of immune complexes, lysis of antibody-coated target cells by killer cells (known as antibody-dependent cellular cytotoxicity or ADCC), release of inflammatory mediators, and immunoglobulins.

It will also be appreciated by those of ordinary skill in the art that the disclosed antibodies or antigen binding fragment sequences may be substituted with amino acid sequences that differ from the naturally occurring amino acid sequences of the antibodies. For example, the substituted amino acid sequence may be similar to the starting sequence, such as having a certain proportion of identity to the starting sequence, such as it may be about 80%, about 85%, about 90%, about 95%, about 98% or about 99% identical to the starting sequence, or a range between any two of these values (inclusive) or any value therein.

In certain embodiments, the antibody comprises an amino acid sequence having one or more modifying groups. For example, the disclosed bispecific antibodies can be modified to add functional groups (e.g., PEG, drug, toxin, or tag).

The antibodies, antigen binding fragments, disclosed herein include modified derivatives, i.e., modified by covalent attachment of any type of molecule to the antibody, wherein the covalent attachment does not prevent binding of the antibody to the epitope. Including but not limited to the following examples, the antibodies may be glycosylated, acetylated, pegylated, phosphorylated, amidated, derivatized with known protecting/blocking groups, proteolytically cleaved, linked to cellular ligands or other proteins, and the like. Any of a number of chemical modifications may be made by the prior art, including but not limited to specific chemical cleavage, acetylation, formylation, metabolic synthesis of tunicamycin, and the like.

In some embodiments, the antibody may be conjugated to a therapeutic agent, prodrug, peptide, protein, enzyme, virus, lipid, biological response modifier, agent, or PEG.

The antibodies may be conjugated or fused to therapeutic agents, which may include detectable labels (e.g., radiolabels), immunomodulators, hormones, enzymes, oligonucleotides, photoactive therapeutic agents, diagnostic agents, cytotoxic agents for drugs or toxins, ultrasound enhancers, non-radioactive labels and combinations thereof, and other such agents known in the art.

Antibodies can be detectably labeled by coupling them to chemiluminescent compounds. The presence of the chemiluminescent-tagged antibody is then determined by detecting the luminescence that occurs during the chemical reaction. Examples of chemiluminescent labeling compounds include luminol, isoluminol, aromatic acridinium esters, imidazoles, acridinium salts and oxalic esters.

Method for preparing antibody and polynucleotide for encoding antibody

Also disclosed are polynucleotides or nucleic acid molecules encoding the antibodies, antigen binding fragments, and derivatives thereof of the invention. The polynucleotides disclosed herein may encode VHA, VHB, VLA, VLB, hole-containing CHA, knob-containing CHB, CLA, CLB, fc (A) and Fc (B) regions, heavy chain A, heavy chain B, light chain A and light chain B, and the like. The nucleic acids encoding the constituent fragments of CD47 binding arm A and PD-L1 binding arm B of an exemplary CD47/PD-L1 bispecific antibody are shown in Table 7.

TABLE 7 coding nucleotide sequences for the constitutive fragments of the CD47 binding arm A and the PD-L1 binding arm B of the exemplary CD47/PD-L1 bispecific antibody

/>

Methods of making antibodies are well known in the art and are described in the present invention. In certain embodiments, the antibodies, antigen binding fragments, and variable and constant regions encompassed by the disclosed antibodies are all of human origin. Fully human antibodies and antigen binding fragments can be prepared using techniques disclosed in the art and described herein. For example, fully human antibodies directed against a particular antigen can be prepared by administering the antigen to transgenic animals that have been modified to produce fully human antibodies in response to antigen challenge. The bispecific antibody disclosed by the invention is prepared by fusing the encoding nucleic acid fragments of a binding arm A specifically binding to CD47 and a binding arm B specifically binding to PD-L1 in one expression vector, recombinantly expressing two targeted binding arms, and combining the two targeted binding arms in a recombinant host cell to obtain the CD47/PD-L1 bispecific antibody disclosed by the invention.

In some embodiments, the antibodies produced do not elicit a detrimental immune response in the animal (e.g., human) to be treated. In one embodiment, the antibodies, antigen binding fragments, or derivatives of the present disclosure are modified to reduce their immunogenicity using art-recognized techniques. For example, the antibodies may be humanized, primatized, deimmunized or chimeric antibodies may be prepared. These types of antibodies are derived from non-human antibodies, typically murine or primate antibodies, which retain or substantially retain the antigen binding properties of the parent antibody but are less immunogenic in humans. This can be accomplished by a variety of methods, including (a) grafting the entire non-human variable region to a human constant region to produce a chimeric antibody; (b) Transplanting at least a portion of one or more non-human Complementarity Determining Regions (CDRs) into framework and constant regions of human origin, with or without the retention of critical framework residues; or (c) transplanting the entire non-human variable regions, but "hiding" them by replacing surface residues with human-like moieties. Typically the framework residues in the human framework region will be replaced with corresponding residues from the CDR donor antibody, such as residues capable of improving antigen binding. These framework substitutions can be identified by methods well known in the art, for example by modeling the interactions of CDRs and framework residues to identify framework residues that play an important role in antigen binding and by sequence alignment to identify aberrant framework residues at specific positions. (see U.S. Pat. No. 5,585,089; incorporated herein by reference in its entirety). Antibodies can be humanized using a variety of techniques well known in the art, such as CDR grafting (EP 239,400; wo 91/09967; U.S. Pat. nos. 5,225,539,5,530,101 and 5,585,089), repair or surface rearrangement (EP 592,106; EP519,596; and chain rearrangement (U.S. Pat. No. 5,565,332), the entire contents of which are incorporated herein by reference.

For example, the heavy and light chain variable region sequences from the starting antibody can be analyzed and a "map" of human T cell epitopes from each variable region generated, showing the positions of the epitopes relative to the Complementarity Determining Regions (CDRs) and other key residues within the sequence. Single T cell epitopes from the T cell epitope map were analyzed to identify alternative amino acid substitutions with lower risk of altering antibody activity. A series of alternative heavy chain variable region sequences and light chain variable region sequences comprising a combination of amino acid substitutions are designed and subsequently incorporated into a series of binding polypeptides. The genes for the complete heavy and light chains, comprising the modified variable and human constant regions, are then cloned into expression vectors, and the plasmids are subsequently transferred into cell lines to produce the complete antibodies. The antibodies are then compared using appropriate biochemical and biological experiments to identify the best antibodies.

The binding specificity of the bispecific antibodies or antigen binding fragments disclosed herein can be detected by in vitro assays, such as co-immunoprecipitation, radioimmunoassay (RIA), or enzyme-linked immunosorbent assay (ELISA).

Alternatively, scFab production in bispecific antibodies of the invention forms single chain units by bridging heavy and light chain fragments by amino acids, resulting in single chain fusion peptides.

For certain uses including the use of antibodies in humans and in vitro detection assays, chimeric, humanized or fully human antibodies may be used. Exemplary embodiments of the invention humanized CD47 antibodies and humanized PD-L1 antibodies are used as sources of binding arm sequences for the construction of the CD47 and PD-L1 bispecific antibodies of the invention.

Furthermore, one or more CDRs of an antibody of the invention can be inserted into a framework region, e.g., into a human framework region, using conventional recombinant DNA techniques to construct a humanized non-fully human antibody. The framework regions may be naturally occurring or consensus framework regions, preferably human framework regions (see Chothia et al, J. Mol. Biol.278:457-479 (1998), which lists a range of human framework regions). Some polynucleotides may encode antibodies that bind specifically to at least one epitope of an antigen of interest produced by the combination of framework regions and CDRs. One or more amino acid substitutions may be made within the framework region, and amino acid substitutions may be selected that improve binding of the antibody to its antigen. In addition, substitution or deletion of cysteine residues in one or more of the variable regions involved in interchain disulfide formation may be performed in this manner, thereby producing an antibody molecule lacking one or more interchain disulfide bonds. Other variations on polynucleotides within the skill of the art are also encompassed by the present invention.

The CD47 and PD-L1 bispecific antibodies of the invention can be prepared using conventional recombinant DNA techniques. Vectors, cell lines, etc. for antibody production can be selected, constructed and cultured using techniques well known to those skilled in the art. These techniques are described in various laboratory manuals and major publications, such as Recombinant DNA Technology for Production of ProteinTherapeutics in Cultured Mammalian Cells, D.L.Hacker, F.M.Wurm, in ReferenceModule in Life Sciences,2017, the entire contents of which, including the supplementary contents, are incorporated by reference in their entirety.

In some embodiments, DNA encoding an antibody may be synthesized according to conventional methods from the antibody amino acid sequences described herein, placed into an expression vector, and then transfected into a host cell, and the transfected host cell cultured in culture medium to produce the antibody.

Illustratively, in the methods of the invention for targeting human CD47 and human PD-L1 diabodies, the binding arm sequence that binds CD47 is derived from an anti-CD 47 antibody. The preparation method of the anti-CD 47 antibody is based on a humanized design of a murine chimeric antibody BT007-06-CHI, and the humanized targeting human CD47 and human PD-L1 protein double antibodies are obtained by exemplary humanized targeting human CD47 and human PD-L1 protein double antibodies, such as Hu06-11, hu06-12, hu06-14, hu06-16, hu06-15, hu06-32 and Hu06-31.

The humanized CD47 antibodies according to the examples are the heavy and light chain nucleotide sequences of Hu06-31, while the anti-PD-L1 antibody sequences are derived from the heavy and light chain amino acid sequences of the anti-PD-L1 proto-developed antibodies that have been marketed. In another embodiment, DNA encoding the desired monoclonal antibody may be isolated and sequenced using conventional procedures (e.g., using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of murine antibodies) to obtain nucleic acids encoding the constituent fragments of binding arm B as a source of such DNA.

In some exemplary embodiments, the bispecific antibodies of the invention are prepared by designing a nucleotide sequence comprising a targeting PD-L1 binding arm encoding a heavy and light chain of a point mutated anti-PD-L1 antibody that forms a Knob structure, constructed into an expression vector, such as PEE6.4, and a nucleotide sequence comprising a targeting CD47 binding arm encoding a heavy and light chain of a point mutated anti-CD 47 antibody that forms a hole structure, separately constructed into an expression vector, such as PEE12.4, resulting in a recombinant expression vector expressing anti-PD-L1 and anti-CD 47 single arm antibodies, respectively; and further performing enzyme digestion and connection on the PEE 6.4-binding arm A and the PEE 12.4-binding arm respectively comprising two coding nucleic acid fragments of the targeting binding arms to obtain an expression vector carrying the binding arms comprising CD47 and PD-L1, further obtaining a recombinant host cell, expressing the recombinant host cell, and purifying to obtain the CD47/PD-L1 bispecific antibody.

The first step separately designs and synthesizes Fc (A) comprising a light chain variable region A (VLA) encoding or encoding a light chain constant region A (CLA) encoding or encoding a heavy chain variable region A (VHA) encoding or encoding a heavy chain variable region A (CHA) encoding or encoding a heavy chain variable region A (CH 1A) encoding or encoding a heavy chain variable region A (Fc) encoding or encoding a heavy chain variable region A (Fc) encoding or encoding a heavy chain variable region A (CD 47) or variants thereof, or a CD 47-targeting binding arm single strand comprising a light chain and a CH3 amino acid modified heavy chain of an anti-CD 47 antibody, and a coding nucleotide sequence 2 comprising a binding arm single strand encoding an Fc (B) encoding a light chain variable region a (VLA) or a light chain constant region a (CLA) derived from an anti-PD-L1 antibody or a variant thereof, a heavy chain variable region a (VHA) derived from an anti-PD-L1 antibody or a variant thereof, and a heavy chain constant region a (CHA) derived from an anti-PD-L1 antibody or a variant thereof, a first heavy chain constant region (CH 1B) derived from an anti-PD-L1 antibody, a hinge region, an anti-PD-L1 antibody that can form a site-directed mutation of a knob structure, or a PD-L1-targeting binding arm single strand comprising a light chain of an anti-PD-L1 antibody and a CH3 amino acid modified anti-PD-L1 antibody;

In the second step, expression vectors 1 and 2 encoding nucleic acids 1 and 2, respectively, are constructed.

A third step of enzymatically cleaving the expression vectors 1 and 2 of the encoding nucleic acids 1 and 2 obtained in the first step to ligate the encoding nucleic acid 1 to a fragment carrying the encoding nucleic acid 2; recombinant expression vector 3 of two targeting binding arms of the double antibody which simultaneously expresses the specific binding CD47 and PD-L1 is constructed.

The fourth step is to ligate the recombinant expression vector 3 obtained in the third step to a mammalian expression vector to obtain a recombinant mammalian host cell, such as a CH0 cell.

The fifth step expresses the bispecific antibody of the present invention in a host cell culture condition that achieves multiple sets of disulfide bonds between the heavy chain CH2 of two targeted single chain peptide chains to prevent light chain mismatches.

Sixth, harvesting and purifying the cells to obtain the bispecific antibody containing the CD47 antibody, the PD-L1 antibody and the Fc mutant anti-CD 47/PD-L1.

The present disclosure also provides polynucleotide sequences comprising amino acid sequences encoding Knob and hole regions on the CD47 and PD-L1 bispecific antibody-constituting peptide chains that constitute the present disclosure that encode the sequence numbers listed in table 9.

In some embodiments, the expression antibody vector includes at least one promoter element, an antibody coding sequence, a transcription termination signal, and a polyA tail. Other elements include enhancers, kozak sequences, and donor and acceptor sites for RNA splicing flanking the insertion. Efficient transcription can be obtained by the early and late promoters of SV40, the long terminal repeats from retroviruses such as the early promoters of RSV, HTLV1, HIVI and cytomegalovirus, and other cellular promoters such as actin promoters may be used. Suitable expression vectors may include pIRES1neo, pRetro-Off, pRetro-On, PLXSN, or Plncx, pcDNA3.1 (+/-), pcDNA/Zeo (+/-), pcDNA3.1/Hygro (+/-), PSVL, PMSG, pRSVcat, pSV2dhfr, pBC12MI and pCS2, among others. Commonly used mammalian cells include 293 cells, cos1 cells, cos7 cells, CV1 cells, murine L cells, CHO cells, and the like.

In some embodiments, the insert should contain a selectable marker, common selectable markers including dihydrofolate reductase, glutamine synthetase, neomycin resistance, hygromycin resistance, and the like, to facilitate selection and isolation of transfected cells. The constructed plasmid is transfected into host cells without the genes, and the transfected cells grow in a large quantity after being cultured by a selective medium to generate target proteins to be obtained.

In addition, mutations may be introduced into the nucleotide sequences encoding the antibodies of the invention using standard techniques known to those skilled in the art, including, but not limited to, site-directed mutagenesis and PCR-mediated mutagenesis that result in amino acid substitutions. Variants (including derivatives) encode substitutions of less than 50 amino acids, substitutions of less than 40 amino acids, substitutions of less than 30 amino acids, substitutions of less than 25 amino acids, substitutions of less than 20 amino acids, substitutions of less than 15 amino acids, substitutions of less than 10 amino acids, substitutions of less than 5 amino acids, substitutions of less than 4 amino acids, substitutions of less than 3 amino acids, or substitutions of less than 2 amino acids relative to the original heavy chain variable region VH CDR1, VH CDR2, VH CDR3, and light chain variable region VL CDR1, VLCDR2, or VL CDR 3. Alternatively, mutations may be introduced randomly along all or part of the coding sequence, for example by saturation mutagenesis, and the resulting mutants may be screened for biological activity to identify mutants that retain activity.

In some embodiments, the substitutions described herein are conservative amino acid substitutions.

Pharmaceutical composition

The invention also provides a pharmaceutical composition. Such compositions comprise an effective amount of an antibody or antigen-binding fragment and a pharmaceutically acceptable carrier. In some embodiments, the pharmaceutical composition further comprises an anti-cancer agent (e.g., an immune checkpoint inhibitor).

In some embodiments, the term "pharmaceutically acceptable" refers to substances approved by a regulatory agency of the government or listed in the generally recognized pharmacopoeia for use in animals, and in particular for use in humans. In addition, a "pharmaceutically acceptable carrier" generally refers to any type of non-toxic solid, semi-solid, or liquid filler, diluent, encapsulating material, formulation aid, or the like.

The term "carrier" refers to a diluent, adjuvant, excipient, or carrier with which the active ingredient may be administered to a patient. Such pharmaceutical carriers can be sterile liquids, such as water and oils, including those of petroleum, animal or vegetable origin, such as peanut oil, soybean oil, mineral oil, sesame oil and the like. When the pharmaceutical composition is administered intravenously, water is the preferred carrier. Saline solutions and aqueous dextrose and glycerol solutions can also be employed as liquid carriers, particularly for injectable solutions. Suitable pharmaceutical excipients include starch, glucose, lactose, sucrose, gelatin, malt, rice, flour, chalk, silica gel, sodium stearate, glycerol monostearate, talc, sodium chloride, dried skim milk, glycerol, propylene, glycol, water, ethanol and the like. If desired, the composition may also contain minor amounts of wetting or emulsifying agents, or pH buffering agents, such as acetates, citrates or phosphates. Antibacterial agents such as benzyl alcohol or methyl parahydroxybenzoate, antioxidants such as ascorbic acid or sodium bisulphite, chelating agents such as ethylenediamine tetraacetic acid, and tonicity adjusting agents such as sodium chloride or dextrose are also contemplated. These compositions may take the form of solutions, suspensions, emulsions, tablets, pills, capsules, powders, sustained release formulations and the like. The composition may be formulated as a suppository with conventional binders and carriers such as triglycerides. Oral formulations may include standard carriers such as pharmaceutical grades of mannitol, lactose, starch, magnesium stearate, sodium saccharine, cellulose, magnesium carbonate, and the like. Examples of suitable pharmaceutical carriers are described in Remington' sPharmaceutical Sciences of e.w. martin, incorporated herein by reference. Such compositions will contain a clinically effective dose of the antibody or antigen-binding fragment, preferably in purified form, together with a suitable amount of carrier to provide a form of administration suitable for the patient. The formulation should be suitable for the mode of administration. The parent formulation may be packaged in ampules, disposable syringes or multiple dose vials made of glass or plastic.

In some embodiments, the composition is formulated according to conventional procedures into a pharmaceutical composition suitable for intravenous injection into the human body. Compositions for intravenous administration are typically solutions in sterile isotonic aqueous buffers. The composition may also contain a solubilizing agent and a local anesthetic such as lidocaine to relieve pain at the injection site. In general, the active ingredients are supplied individually or in admixture in unit dosage form, such as in the form of a dry lyophilized powder or dry concentrate, in a sealed container (e.g., ampoule or pouch) that is indicative of the amount of active agent. In the case of administration of the composition by infusion, the composition may be dispensed using an infusion bottle containing sterile water or saline of pharmaceutical grade. In the case of administering the composition by injection, an ampoule of sterile water for injection or saline may be used so that the active ingredients may be mixed prior to administration.

The compounds of the present invention may be formulated in neutral or salt form. Pharmaceutically acceptable salts include those derived from anions such as hydrochloric acid, phosphoric acid, acetic acid, oxalic acid, tartaric acid, and the like, and those derived from cations such as sodium, potassium, ammonium, calcium, ferric hydroxide, isopropylamine, triethylamine, 2-ethylamino ethanol, histidine, procaine, and the like.

Therapeutic method

The invention also provides methods of treatment and uses. In some embodiments, methods for treating or ameliorating various types of cancer, tumor, or infection-related diseases are provided, the methods comprising administering to a patient an effective dose of the bispecific antibody. In some embodiments, there is provided the use of the bispecific antibody for the treatment or amelioration of a cancer, tumor, or infection-related disease. In some embodiments, there is provided the use of the bispecific antibody in the manufacture of a medicament for treating or ameliorating a cancer, tumor, or infection-related disorder.

The present invention relates to methods of treating diseases associated with PD-L1 and/or CD47 as therapeutic targets; to methods that can ameliorate, slow, inhibit or prevent any disease or condition by eliminating, inhibiting or reducing the binding of PD-L1 to PD1 and/or the binding of CD47 to sirpa; to methods of treating cancer or a tumor in a subject, methods of alleviating a symptom of cancer or tumor in a subject, methods of avoiding recurrence of a tumor or cancer in a subject, comprising administering to a subject an effective amount of any anti-PD-L1/CD 47 bispecific antibody or fragment thereof described herein. Use of antibodies and pharmaceutical compositions

The antibodies, antigen binding fragments thereof, and pharmaceutical compositions comprising the same provided herein may be used as therapeutic agents for diagnosing, prognosing, monitoring, treating, alleviating and/or preventing diseases and disorders associated with aberrant PD-L1 and/or CD47 expression, activity and/or signaling in a subject. Monoclonal antibodies, bispecific antibodies and antigen-binding fragments thereof disclosed herein and pharmaceutical compositions comprising the same may be administered when diseases and conditions associated with aberrant PD-L1 or CD47 expression, activity and/or signaling are identified in a subject using standard methods.

In view of the fact that most CD47 antibodies disclosed in the prior art are capable of eliciting hemagglutination of human erythrocytes, there is still a great need to obtain new anti-CD 47 antibodies that are not only effective in promoting phagocytosis of macrophages but also do not lead to cell agglutination. The anti-PD-L1/CD 47 bispecific antibodies disclosed herein meet this need, not only effectively promote phagocytosis, but also do not result in significant hemagglutination, and more preferably do not significantly bind to human erythrocytes.

In some embodiments, the anti-CD 47/PD-L1 bispecific antibodies of the invention are capable of binding to PD-L1 with ultra high affinity, binding to CD47 with high or medium affinity; the selective binding of the anti-PD-L1/CD 47 bispecific antibody to tumor cells is promoted by the specific binding to PD-L1 on the tumor cells, so that the binding to CD47 expressed in a plurality of normal tissues is avoided, and side effects are reduced; by significantly expanding the effective dose range of the anti-PD-L1/CD 47 bispecific antibodies of the invention when the affinity to CD47 is much lower than to PD-L1.

In some embodiments, cancers treated and/or prevented with the antibodies of the invention include, but are not limited to, solid tumors, hematological cancers (e.g., leukemia, lymphoma, myeloma, e.g., multiple myeloma), and metastatic lesions. In one embodiment, the cancer is a solid tumor. Examples of solid tumors include malignant tumors, e.g., sarcomas and carcinomas of multiple organ systems, such as those that affect the lung, breast, ovary, lymphoid, gastrointestinal tract (e.g., colon), anus, genitalia, and genitourinary tract (e.g., kidney, bladder epithelium, bladder cells, prostate), pharynx, CNS (e.g., brain, neural, or glial cells), head and neck, skin (e.g., melanoma), nasopharynx (e.g., differentiated or undifferentiated metastatic or locally recurrent nasopharyngeal carcinoma), and pancreas. The cancer may be in early, intermediate or late stages or metastatic cancer. In one embodiment, the tumor is tumor immune escape. In some embodiments, the tumor is a gastrointestinal tumor (e.g., cancer), such as colon cancer.

The specific dosage and treatment regimen for any particular patient will depend on a variety of factors including the particular antibody or derivative used, the age and weight of the patient, the general health, sex and diet, and the time of administration, frequency of excretion, drug combination, and the severity of the particular disease being treated. These factors are judged by medical care personnel included within the scope of one of ordinary skill in the art. The dosage will also depend on the individual patient to be treated, the route of administration, the type of formulation, the nature of the compound used, the severity of the disease and the desired effect. The dosages used can be determined by pharmacological and pharmacokinetic principles well known in the art. In some embodiments, the effective dose ranges from about 0.01mg/kg to about 100mg/kg, which may be, for example, twice weekly (BIW) or to once monthly.

Methods of administration of the antibody or derivative include, but are not limited to, intradermal, intramuscular, intraperitoneal, intravenous, subcutaneous, nasal, epidural, and oral injection. The pharmaceutical compositions may be administered by any convenient route, for example by infusion or bolus injection, by absorption through epithelial or skin mucosa (e.g. oral mucosa, rectal and intestinal mucosa, etc.), and may be co-administered with other bioactive agents. Thus, pharmaceutical compositions containing the antibodies of the invention may be administered orally, rectally, parenterally, intracisternally, intravaginally, intraperitoneally, topically (e.g., by powders, ointments, drops, or transdermal patches), bucally, or by oral or nasal spray.

The term "parenteral" as used herein refers to modes of administration including intravenous, intramuscular, intraperitoneal, intrasternal, subcutaneous and intra-articular injection and infusion.

The mode of administration may be systemic or local. Furthermore, it may be desirable to introduce the antibodies of the invention into the central nervous system by any suitable route, including intraventricular and intrathecal injection; the intraventricular injection may be assisted by connecting the intraventricular catheter to, for example, a reservoir (which may be an Ommaya reservoir). Pulmonary administration is also possible, for example by using an inhaler or nebulizer, and the use of an aerosolized formulation.

The antibodies of the invention may be topically applied to an area in need of treatment; the following means may be used, but are not limited to: local infusion during surgery, for example local application in combination with a post-operative wound dressing, is achieved by injection, through a catheter, by means of suppositories or by means of implants which are porous, non-porous or gelatinous materials, including membranes (e.g. silicone rubber membranes) or fibres. Preferably, when administering the proteins (including antibodies) of the present invention, care must be taken to use materials that do not absorb the protein.

Methods for treating diseases are typically performed in an in vitro test comprising administering an antibody or derivative of the invention, then testing the desired therapeutic or prophylactic activity in vivo in an acceptable animal model, and finally administering to a human. Suitable animal models (including transgenic animals) are well known to those of ordinary skill in the art. For example, in vitro assays useful in demonstrating the therapeutic use of the antibodies, antigen binding fragments of the invention include the effect of the antibodies on cell lines or patient tissue samples. The effect of an antibody on a cell line and/or tissue sample can be detected using techniques known to those skilled in the art, such as those disclosed elsewhere in the present invention. In accordance with the present disclosure, in vitro assay experiments useful for determining whether to administer a specific antibody include in vitro cell culture experiments in which a patient tissue sample is cultured in culture and exposed to or otherwise administered a compound, and observing the effect of such compound on the tissue sample.

Various known delivery systems may be used to administer the antibodies or derivatives of the invention or polynucleotides encoding them, e.g., encapsulated in liposomes, microparticles, microcapsules, recombinant cells capable of expressing the compounds, receptor-mediated endocytosis (see, e.g., wu and Wu,1987, j. Biol. Chem. 262:4429-4432), construction of nucleic acids as part of retroviruses or other vectors, and the like.

Combination therapy

In some embodiments, the antibodies of the invention may be used in combination or in combination with other therapeutic or prophylactic regimens, including administration of one or more antibodies of the invention and one or more other therapeutic agents or methods. For combination therapy, the antibody may be administered simultaneously or separately with other therapeutic agents. When administered separately, the antibodies of the invention may be administered before or after administration of another other therapeutic agent.

In some embodiments, when administering a bispecific antibody of the invention to a patient, an antibody molecule or pharmaceutical composition or immunoconjugate disclosed herein may also be co-administered to the patient, along with one or more other therapies, such as therapeutic regimens and/or other therapeutic agents (e.g., chemotherapeutic agents, radiotherapeutic agents, or biomacromolecule drugs).

Such combination therapies encompass the administration of a combination (wherein two or more therapeutic agents are contained in the same formulation or separate formulations), and the administration of a separate, in which case the administration of an antibody of the invention may occur prior to, concurrent with, and/or subsequent to the administration of another therapy, e.g., a therapeutic regimen and/or therapeutic agent. The antibody molecule and/or other therapy, e.g., therapeutic agent or mode of treatment, may be administered during active disease or during remission or less active disease. The antibody molecule may be administered prior to, concurrently with, after, or during remission of the disease.

In some embodiments, the bispecific antibodies of the invention are administered in combination with a chemotherapeutic agent. In some embodiments, chemotherapeutic agents that may be administered with the anti-PD-L1/CD 47 bispecific antibodies of the invention include, but are not limited to, antibiotic derivatives (e.g., doxorubicin, bleomycin, daunorubicin, and actinomycin D), antiestrogens (e.g., tamoxifen), antimetabolites (e.g., fluorouracil, 5-FU, methotrexate, fluorouridine, interferon alpha-2 b, glutamic acid, mithramycin, mercaptopurine, and 6-thioguanine), cytotoxic agents (e.g., carmustine, BCNU, lomustine, CCNU, cytarabine, cyclophosphamide, estramustine, hydroxyurea, methylbenzyl hydrazine, mitomycin, busulfan, cisplatin, and vincristine sulfate), hormones (e.g., medroxyprogesterone, estramustine phosphate sodium, ethinyl estradiol, megestrol acetate, methyltestosterone, diethylstilbestrol diphosphate, chloroestrol, and lactone), nitrogen derivatives (e.g., melphalan, benzodiazepine, dichloromethyl butyrate, dichloromethyl-ethyl acetate, and dexamethasone, and other compounds (vinblastine), and other compounds (e), such as vinblastine sulfate, and other drugs, and combinations thereof.

In some embodiments, the antibodies of the invention are administered in combination with a cytokine. Cytokines that may be administered with the antibodies of the invention include, but are not limited to, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-10, IL-12, IL-13, IL-15, and the like.

In some embodiments, the bispecific antibodies of the invention are administered in combination with a biomacromolecule drug. Examples of biological macromolecular drugs include immunotherapeutic agents including, but not limited to, therapeutic antibodies suitable for treating patients. Some examples of therapeutic antibodies include Xin Tuozhu mab, aba Fu Shan-mab, adekamtumab, alfutuzumab, alemtumab, albemtumab, almomab, al Ma Tuo-mab, anatuzumab, alsimmomab, bavintuximab, bei Tuo-mab, bevacizumab, bivacizumab, blattitumomab, brentuzumab, comptuzumab, katuzumab Ma Suoshan-mab, cetuximab cetuximab, clarituximab, dactylumumab, qiao Ji tozumab, du Lige tozumab, du Xiji tozumab, delumomab, damuzumab, darstuzumab, halloysite, erlotuzumab, encyclocontemplated, ertuximab, etallic Ma Suoshan, etallic beadab, falexon beadab, fekorituximab Feichiometriol, franko, futuximab, ganitumumab, gituzumab, ji Luntuo Xitumumab, gryllitumumab, emamizumab, igofuzumab, iymtuzumab, indamitumumab, inootuzumab, emamizumab, ipimitumumab, irtuximab, la Bei Tuozhu Xitumumab, leishukamumab, lintuzumab, lo Wo Tuozhu Xitumumab, lu Katuo MXitumumab Marpatuzumab, martuzumab, mi Latuo bead mab, merozolomab, mi Tuom mab, monerituximab, nanatozumab, naltuzumab, inner Ji Tuom mab, ni Mo Tuozhu mab, nor Fei Tuoshan mab, oxcarbatuzumab, ofatuzumab, olatuzumab, onatuzumab, obroman, aogore Fu Shan mab, panitumumab, peng Tuom mab, pertuzumab, pratuzumab, lamomuzumab, raschituzumab, rituximab, lobutrab, sha Tuo mab, sibutruzumab, rituximab, sortiuximab, tacaruzumab, tagatouzumab, tetratuzumab, tertrastuzumab, terprarituximab, levetiuzumab, trastuzumab, tutututuzumab, you-rituximab, valtuzumab, vortiotuzumab, vomitobizumab, votutuzumab and za Lu Tuom mab and the like.

In some embodiments, the antibodies of the invention may be used with immune checkpoint inhibitors. In some embodiments, the antibodies of the invention are administered in combination with other therapeutic or prophylactic regimens, such as radiation therapy.

The invention is further illustrated by means of the following examples, which are not intended to limit the scope of the invention.

The experimental methods in the following embodiments, in which specific conditions are not specified, are according to conventional conditions or according to conditions recommended by the manufacturer of the raw materials or goods. Or biotechnological textbooks such as molecular cloning, laboratory manuals, cold spring harbor laboratories, contemporary molecular biology methods, cell biology and the like. The reagents of particular origin are not noted and are conventional reagents purchased commercially.

The reagents and materials used in the present invention are commercially available.

Cells and plasmids used in the bispecific antibody function test examples of the present invention can be obtained according to construction methods disclosed in conventional laboratory manuals in the art, such as Molecular cloning: a laboratory manual =molecular cloning: laboratory manual (third edition).

Examples

The technical solution of the present invention is further illustrated by the following specific examples, which do not represent limitations on the scope of the present invention. Some insubstantial modifications and adaptations of the invention based on the inventive concept by others remain within the scope of the invention.

Example 1 humanized engineering of anti-human CD47 murine antibody and humanized antibody evaluation

1.1 detection of affinity of the CD47 antibody BT007-06 to the antigen CD47-His based on Surface Plasmon Resonance (SPR) technique

The interaction of the CD47 antibody BT007-06 with the antigen CD47-His (ACRO, cat. No. CD 7-H5227) was measured using a biomacromolecule interactometer (brand: cytiva, model BiacoreX 100) as an SPR dependent biosensor. CD47 antibody BT007-06 was coupled to the surface of an activated Protein chip (GE Healthcare, lot BR-1005-30: 10285729) chip (coupling limited to channel number two). The CD47 antibody concentration was adjusted to 0.34 μg/mL, and CD47 antibody was specifically captured using the antibody of the Protein chip to obtain about 200 Response Units (RU). The analyte CD47-His protein (from 200nM to 0 nM) was then diluted with 1xHBS-EP+ (GE Healthcare, cytiva, BR-1006-69, lot BCBJ 6372V) for running buffer. And detecting the binding/dissociating capacity between the CD47 antibody BT007-06 and the CD47-His by sample injection under the working environment of 25 ℃ and the sample injection flow rate of 30 mu L/min. Meanwhile, channel number one served as a reference channel without coupling or capturing any protein. The signal detected by channel one will be used as reference data. And (3) naturally dissociating 600S between the CD47 antibody BT007-06 and the CD47-His, and regenerating the antigen-antibody complex on the surface of the eluting chip for 30 seconds by using glycine (10 mM, pH1.5) regeneration buffer solution after reaching a preset time, wherein the eluting target is the initial response value level after the baseline response value is subjected to the regression coupling. After the data of the sensorgram are obtained, the binding constant (ka) and the dissociation constant (KD) are analyzed by using Biacore X100 Evaluation software, an ideal binding dissociation curve is fitted, and the equilibrium dissociation constant KD (KD/ka) between the antibody antigens is obtained by calculation. The Biacore X100 Evaluation software analyzes the binding constant (ka) and the dissociation constant (KD), fits an ideal binding dissociation curve, calculates the equilibrium dissociation constant KD (KD/ka) between the obtained antibody antigens, and the determination result of the affinity of the CD47 mouse anti-binding CD47-His protein is shown in FIG. 1. FIG. 1 shows the results of affinity detection of the CD47 murine anti-binding CD47-His protein, table 8 shows the CDR sequences of the murine anti-BT 007-06-CHI, and Table 9 shows the variable region sequences of the murine anti-BT 007-06-CHI.

The affinity KD of the CD47 antibody BT007-06 to the antigen CD47-His is 2.12 x 10 ^-8 M。

1.2 humanization of anti-human CD47 murine anti-BT 007-06

The variable region sequence of the murine antibody BT007-06 was subjected to humanized design, and the bioinformatics analysis was used to determine that the heavy chain sequence of the murine antibody BT007-06 was most similar to IGHV1-2*06germline gene, which was 69.4%. The light chain sequence is the most similar to IGKV3D-7*01germline gene, and is 65.6%. The murine anti-JH 007-06 light and heavy chain variable region CDR sequences were grafted into IGHV 1-2X 06 and IGKV3D-7*01germline gene. The homologous structure with highest structural similarity with the variable region sequence of the mouse antibody BT007-06 is determined by BLAST analysis, a three-dimensional structure model is built after the variable region sequences of the antibody light and heavy chains are aligned, key amino acids are found through the three-dimensional structure to carry out back mutation, and finally 6 humanized heavy chain variable region sequences and 6 humanized light chain variable region sequences are obtained, and 7 humanized CD47 antibody names, CDRs, variable regions, light chains and heavy chains are formed by combination as shown in the following tables 10, 11 and 12:

table 8 example 1 humanized heavy chain variable region sequences and humanized light chain variable region sequences were assembled to form 7 humanized CD47 antibodies

/>

TABLE 9 CDR of humanized antibody obtained in example 1

/>

TABLE 10 light and heavy chains of humanized CD47 antibodies of example 1

/>

1.3CD47 humanized antibody Activity assay

The activities of CD47 humanized antibody and positive reference antibody hu-5F9 (supplied Shang Yi qiao shen, cat No. 68063-H001) were analyzed and compared by ELISA-binding and ELISA-blocking experiments.

1.3.1CD47 humanized antibody ELISA binding to CD47 protein Activity assay

CD47-His (ACRO, cat. No. CD 7-H5227) was diluted to 1. Mu.g/mL with a coating solution (0.2 mol/LNa2CO38mL,0.2mol/L NaHCO317mL, 75mL distilled water was added, pH was adjusted to 9.6), and the coated 50. Mu.L/well was added to wells of an ELISA plate (costar, cat. No. 3590) and incubated at 37℃for 2 hours. Discarding the liquid in the well by Wellwash ^TM Versa plate washer (thermo scientific, cat. No. 5165050) wash (PBST, PBS dilution with 0.05% Tween-20) 200. Mu.L/Kong Xi times. 200. Mu.L of blocking solution (PBST diluted 3% BSA) was added to each well overnight at 4 ℃. Removing the liquid in the wells, washing with a plate washer, and adding the washing solution (PBST, PBS diluent0.05% Tween-20) 200. Mu.l/Kong Xi times. 50 mu L of antibody (CD 47 humanized antibody and 5F 9) with initial concentration of 10 mu g/mL is added into each hole, continuous 5-time dilution is carried out, 7 concentration points are sequentially diluted in a gradient way, meanwhile, the antibody diluent is set as the concentration point of 0ug/mL of antibody, and incubation is carried out for 60min at 37 ℃; the liquid in the wells was discarded and the plate was washed 5 times with a plate washer. mu.L of secondary antibody Peroxidase AffiniPure Goat Anti-Human IgG (Jackson, cat. No. 109-035-170) (1:10000 dilution) was added to each well and incubated at 37℃for 60min; the liquid in the wells was discarded and the plate was washed 5 times with a plate washer. The color development solution (sigma, cat No. T4444) was added at 50. Mu.L/well and incubated at 37℃for 15min. 50. Mu.L/well 2mol/L H are added ₂ The reaction was terminated with SO4 and the OD450 value was read on a microplate reader (thermo scientific, model MultiskanFC). The results are shown in FIG. 2, and FIG. 2 is an analysis of the activity of the CD47 humanized antibody in binding to CD 47. The humanized CD47 antibodies Hu06-11, hu06-12, hu06-14, hu06-15, hu06-16, hu06-24, hu06-31 and Hu06-32 provided by the present discovery all have the binding activity with human CD47 protein, and the binding activity is equivalent to that of the reference antibody Hu-5F 9.

Analysis of binding Activity of 1.3.2CD47 humanized antibody ELISA blocking CD47 and SIRP-alpha

SIRP-alpha-His was diluted to 5. Mu.g/mL with a coating solution (0.2 mol/L Na2CO3 8mL,0.2mol/L NaHCO31 7mL was mixed, 75mL distilled water was added thereto, pH was adjusted to 9.6), and the coated 50. Mu.L/well was added to the wells of the microplate and incubated at 37℃for 2 hours. The wells were discarded and the plate was washed with 200. Mu.l/Kong Xi wash solution 5 times. 200. Mu.L of blocking solution was added to each well overnight at 4 ℃. The wells were discarded and the plate washer was used, washing 200. Mu.l// Kong Xi times. Adding 25 mu L of biotinylated human CD47 protein (Baiposis, product number CD7-H82A 3) (0.3 mu g/mL) +25 mu L of CD47 humanized antibody and 5F9 into each well, sequentially diluting 5 times at the concentration of 20 mu g/mL of the first well to obtain 7 concentration gradient antibodies, setting an antibody diluent as an antibody 0 concentration point, and incubating at 37 ℃ for 120min; the liquid in the wells was discarded, and washed 5 times with 200. Mu.l/time of the washing liquid using a plate washer. 50. Mu.L of SA-HRP-Secondary-Antibody (brand: abcam, cat#: ab 7403) (1:200 dilution) was added to each well and incubated at 37℃for 60min; the liquid in the wells was discarded and the plate was washed 5 times with a plate washer. The reaction was stopped by adding 50. Mu.L/well of chromogenic solution, incubating at 37℃for 15min, and adding 50. Mu.L/well of 2mol/L H SO4, and the OD450 value was read on a microplate reader (Thermoscientific, model MultiskanFC). As shown in FIG. 3 (analysis of CD47 humanized antibody blocking the binding activity of CD47 and SIRP-alpha), the humanized CD47 antibodies Hu06-11, hu06-12, hu06-16, hu06-24, hu06-31 and Hu06-32 provided by the present discovery all have the capacity of blocking the binding of SIRP-alpha protein and CD47 protein, wherein the IC50 value of the Hu06-31 antibody is slightly lower than the IC50 value of Hu-5F9 of a reference antibody, and the blocking activity and affinity of the humanized antibody are moderate by combining the EC50 detection data of the humanized antibody, so that the humanized antibody is suitable for constructing the CD47/PD-L1 double antibody.

Example 2 construction and expression of anti-CD 47/PD-L1 diabodies

2.1 construction of anti-PD-L1/CD 47 bispecific antibodies

In this example, 1 anti-PD-L1/CD 47 bispecific antibody was constructed and designated BsAb-463, the structural schematic diagram of which is shown in FIG. 4.

As can be seen from the structural schematic of FIG. 4, the anti-PD-L1/CD 47 bispecific antibody consists of 2 peptide chains. Peptide chain 1# is an amino acid sequence of a binding arm B which specifically binds to PD-L1, wherein the C-terminus of the VL of anti-PD-L1 is linked to the N-terminus of a kappa chain constant region derived from human, the C-terminus of the kappa chain constant region is followed by a Linker (SEQ ID NO: 3), the C-terminus of the Linker is followed by the N-terminus amino acid sequence of the VH of anti-PD-L1, the C-terminus of the VH of anti-PD-L1 is followed by the N-terminus of the human IgG1 constant region (SEQ ID NO: 7); peptide chain 2# is the amino acid sequence of binding arm A that specifically binds CD47, wherein the C-terminus of the VL (SEQ ID NO: 1) that is anti-CD 47 is linked to the N-terminus of a kappa chain constant region derived from human, followed by a Linker (SEQ ID NO: 3) that is followed by the N-terminus of the VH (SEQ ID NO: 2) that is anti-CD 47, followed by the C-terminus of the VH that is anti-PD-L1, followed by the N-terminus of the human IgG1 constant region (SEQ ID NO: 8). In this example, the PD-L1 antibody sequence uses the antibody sequence of the variable region of Abelizumab (Atezolizumab). Coding nucleic acid 1# of aggregate arm peptide chain 1 with SalI/NotI cleavage site added at two ends of the offshore engineering synthesis and specifically binding to CD47 and coding nucleic acid 2# of binding arm peptide chain 2 of PD-L1 with SalI/NotI cleavage site are entrusted. The amino acid sequences and coding nucleotide sequences of binding arm a and binding arm B are shown in tables 6 and 7.

SalI/NotI cleavage of the encoding nucleic acids of PEE6.4 and encoding nucleic acid 1# of the binding arm peptide chain 1, respectively constructing nucleotide sequences of two binding arm peptide chains 1# and 2# of the anti-PD-L1/CD 47 bispecific antibody into PEE6.4 (Feng Hui organism, cat# BR 229) and PEE12.4 expression vectors (Feng Hui organism, cat# BR 137) respectively to obtain anti-PD-L1 and anti-CD 47 single arm antibody plasmids, and then carrying out SalI/NotI cleavage of the two plasmids to obtain plasmids for co-expressing anti-PD-L1 and anti-CD 47, and constructing the two nucleotide sequences into the same expression vector.

Meanwhile, the Fc of the bispecific antibody is mutated, and the technology for preventing heavy chain mismatch invented by Gennetech team in the 90 th century is cited to carry out point mutation on CH3 of the antibody; serine T at CH3 position 366 of the heavy chain (Knob chain) of an antibody is mutated to tryptophan W; serine T at position 366 of the other heavy chain (Hole chain) is mutated to serine S, leucine L at position 368 is mutated to alanine a, and amino acid 407 is mutated from tyrosine Y to valine V (T366S, L368A, Y407V); two site mutations in S354C Y349C are to form disulfide bonds, increasing antibody stability. The anti-PD-L1/CD 47 bispecific antibody nucleotide sequence is constructed into an expression vector, expressed and purified in an Expi CHO-s cell to obtain the anti-PD-L1/CD 47 bispecific antibody.

Example 2.1CD47/PD-L1 bispecific antibody-binding arm-constituting fragment amino acid sequence and nucleotide sequence of coding nucleic acid

The amino acid sequences of the constituent fragments of the CD47/PD-L1 bispecific antibody-binding arms and the nucleotide sequences of the encoding nucleic acids are shown in Table 6 and Table 7, respectively, of the summary of the invention. Table 11 shows the parent antibody light and heavy chains binding to arm A

TABLE 6 amino acid sequences of the constituent fragments of the binding arms of CD47/PD-L1 bispecific antibodies

/>

TABLE 7 nucleotide sequence of nucleic acid encoding a fragment of a CD47/PD-L1 bispecific antibody binding arm

/>

2.2 preparation of anti-CD 47/PD-L1 double-antibody

The Expi CHO-s cells (available from Gibco under accession number a 29133) were sub-cultured in an Expi CHO-s cell culture broth (available from Gibco under accession number a 2910002). The cell density was adjusted to 4X 10 the day before transfection ⁶ Individual cells/ml. The culture of the Expi CHO-s cells was continued so that the cell density in the culture on the day of transfection was 6X 10 ⁶ Individual cells/ml. Taking 4% OptiPRO ^TM SFM complexing Medium (available from Gibco company under the trade designation 12309-019) was used as transfection buffer. To each ml of transfection buffer was added 0.8. Mu.g of plasmid DNA,mixing, and adding Expifectamine ^TM CHO reagent (available from Gibco company under the trade designation a 29130)), was mixed and the cationic transfection reagent/DNA mixture was gently poured into the Expi CHO-s cell suspension. Mix gently to 5% CO2 and incubate overnight at 37 ℃.

After overnight incubation to 18-22h, the flasks were supplemented with 24% of the post-transfection culture volume by ExpiCHO ^TM Auxiliary material (ExpiCHO from Gibco company ^TM Core component of ExpressionSystemkit (cat# A29130) and ExpiCHO at 0.6% of the volume of the culture after transfection ^TM Enhancer ((ExpiCHO from Gibco Co.) ^TM The components of the expresssystemkit (cat No. A29130) were gently mixed, and the culture was collected until the 10 th day or when the cell viability was not more than 70%. During the incubation, disulfide bonds are formed between heavy chain CH1 of the binding arm targeting CD47 and heavy chain CH1 of the binding arm targeting PD-L1, thereby preventing light chain mismatches.

The double antibody purification method comprises the following steps: the first step of capturing is carried out by using affinity chromatography (GE, product number 17547401), so that a part of process related impurities, HCP, residual DNA and the like can be effectively removed, and the method has a good endotoxin removal effect; and in the second step, CHT hydroxyapatite filler (Bio-Rad, product number: 1570040) is selected for fine purification, so that not only can a part of process related impurities be removed, but also the product related impurities can be well removed, and aggregation and degradation fragments in the double antibody can be well removed, and the double antibody with the purity more than 98% can be obtained through two-step purification.

Example 3 detection of affinity of BsAb-463 to antigen based on biological film layer interference technique (Bio layer interferometry, BLI)

3.1CD47/PDL1 bispecific antibody BsAb-463 binding kinetics detection against recombinant human CD47-His hcD47-His (ACRO, cat# CD 7-H5227) was diluted in a gradient from 100nM to 7 concentration spots 2-fold in sequence with Loading Buffer (1 XPBS, pH7.4, with 0.02% Tween-20,0.1% BSA) with zero concentration reference wells. The CD47/PD-L1 bispecific antibody BsAb-463 was diluted to 5. Mu.g/mL with a Loading Buffer and loaded onto the sensor (0.3 nm threshold, 180s binding). Set molecular interaction instrument ForteBio's Octet System (brand: sidoris, model: octet)R8) operating conditions: the temperature was 30℃and the vibration speed was 1000rpm. The coated AHC2 sensor (brand: sidoris, cat# 18-5142) was used in combination with the gradient diluted sample for 120s; the dissociation time was 300s. The on-board detection was performed with a ForteBio's Octet System (vendor ForteBio, model Octet R8). After obtaining the sensorgram data, the binding constant (ka) and the dissociation constant (KD) were analyzed by Octet BLI Analysis software, and an ideal binding dissociation curve was fitted, and the equilibrium dissociation constant KD (KD/ka) between the antibody antigens was calculated as shown in fig. 5. The affinity KD of the CD47/PD-L1 bispecific antibody to the antigen CD47-His is 2.320 x 10 ^-9 M。

FIG. 5 affinity of bispecific antibody BsAb-463 for binding CD47-His protein

Binding kinetics detection of 3.2CD47/PDL1 bispecific antibody BsAb-463 against recombinant human PD-L1-His

hPDL1-His (ACRO, cat. PD 1-H5229) was diluted from 100nM in a gradient from 100nM to 7 concentration spots by 2-fold dilution in sequence with zero concentration reference wells using Loading Buffer (1 XPBS, pH7.4, with 0.02% Tween-20,0.1% BSA). The CD47/PD-L1 bispecific antibody BsAb-463 was diluted to 5. Mu.g/mL with a Loading Buffer and loaded onto the sensor (0.3 nm threshold, 180s binding). Set up molecular interaction instrument ForteBio's oct System (brand: celecox, model: oct R8) operating conditions: the temperature was 30℃and the vibration speed was 1000rpm. The gradient diluted samples were combined using a coated AHC2 sensor (brand: sidoris, cat# 18-5142) for 120s; the dissociation time was 300s. The on-board detection was performed with a ForteBio's Octet System (brand: sidoris, model: octet R8). After obtaining the sensorgram data, the binding constant (ka) and the dissociation constant (KD) were analyzed by Octet BLI Analysis software, and an ideal binding dissociation curve was fitted, as shown in fig. 6, and the equilibrium dissociation constant KD (KD/ka) between the antibody antigens was calculated. The CD47/PD-L1 bispecific antibody has an affinity KD of 6.413X 10 for the antigen PDL1-His (ACRO, cat# PD 1-H5229) ^-10 M。

EXAMPLE 4CD47/PD-L1 bispecific antibody binding assay to CHO cells overexpressing hPD-L1

Binding of 4.1CD47/PD-L1 bispecific antibodies to CHO-K1-PD-L1 cell surface PD-L1

The pHR-PDL1-ECD-His plasmid DNA expressing PD-L1-ECD-His is transiently transfected into CHO-K1 cells (hamster ovary cell sub-strain, ATCC # CCL-61), and then hygromycin B,50ug/mL is screened by plasmid electrotransformation and dosing, so that the stable CHO-K1-PDL1 of PD-L1 over-expressing human PD-L1 membrane protein in the CHO-K1 cells is obtained. CHO-K1-PDL1 cells overexpressing human PDL1 were counted after digestion, resuspended in 96-well plates (3E 5 cells added per well) and washed with PBS (twice). A blank control group, an experimental group and a PDL1 monoclonal Antibody control group (prepared by commercial original manufacturers) are arranged on a 96-well plate, a CD47/PDL1 bispecific Antibody BsAb-463 (with the concentration of 20 mug/mL) is added into a hole of the experimental group, and is diluted to 7 concentration points sequentially by 5 times by adopting a gradient dilution method, meanwhile, the 96-well plate is placed in a blank control hole, the 96-well plate is incubated at 4 ℃ for 30mins, centrifuged (1000 rpm,2 mins) and washed by PBS after liquid in the plate hole is removed (200 mu L/time, 2 times), PE-anti-human-IgG-Fc-Antibody (Jackson, product number 109-116-098, 1:200) is added into each hole, 50 mu L is incubated at 4 ℃ for 30mins, centrifuged (1000 rpm,2 min) and then washed by PBS (200 mu L/time, 2 times), BC (200 mu L/PBS) is added into each hole, and the two PD-human-Fc-Antibody antibodies (Jackson) are combined with PD-1, and have the specific activity value of the PD-1, and the PD-1-human-monoclonal Antibody has the contrast value of the PD-1, and the contrast value of the PD-1-monoclonal Antibody is compared with the PD-1.

4.2CD47/PD-L1 bispecific antibodies block PD-L1 binding to PD1 on CHO-K1-PD-L1 cell membranes

CHO-K1-PD-L1 cells overexpressing human PD-L1 were counted after digestion, resuspended in 96-well plates (bisofil model: TCP 002096), 3E5 cells were added to each well and washed 2 times with PBS. A blank control group, an experimental group and a PD-L1 monoclonal antibody control group (the atilizumab produced by the commercial original grinding manufacturer) are arranged on a 96-well plate, a CD47/PD-L1 bispecific antibody BsAb-463 (the concentration is 50 mu g/mL) is added into the hole of the experimental group, and the holes are sequentially diluted to 9 concentration points by 2.5 times by adopting a gradient dilution method, and meanwhile, the blank control hole is arranged. Simultaneously adding 25 mu L of PD-1-mFc into a plate hole, and mixing at a concentration of 0.5 mu g/mL and a ratio of 1:1; incubating on an ice bin; taking 50 μl of each concentration, adding into cell pellet, re-suspending cells, mixing, placing 96-well plate, incubating at 4deg.C for 30min, centrifuging (1000 rpm,2 min), removing liquid in the plate hole, and washing with PBS (200 μl/time, 2 times); PE-anti-Mouse-IgG-Fc-second-Antibody (Jackson, cat. No. 109-116-098,1:200 dilution, 50. Mu.L) was added to each well, incubated at 4deg.C for 30min, centrifuged (1000 rpm,2 min), and after removal of liquid, washed with PBS (200. Mu.L/2 times); PBS (200. Mu.L/well) was added to each well, and after resuspension of the cells, the cells were read by flow cytometry (brand BECKMAN, model BC 14115). As shown in FIG. 8, the results of FACS-blocking activity analysis show that BsAb-463 blocks the mutual binding of PD-1-mFc and PD-L1 monocyang cells at an IC50 value of 2722ng/ml, and the blocking activity is reduced by 2.8 times compared with that of PD-L1 bivalent monoclonal antibody Atezolizumab

Example 5 analysis of binding of CD47/PD-L1 bispecific antibodies to hCD 47-overexpressing Raji cells

Binding of 5.1CD47/PD-L1 bispecific antibodies to Raji cell surface CD47

Raji cells (purchased from beijing synergetic cell bank) were counted, resuspended in 96-well plates (bisil model: TCP 002096), 3E5 cells were added to each well, and washed 2 times with PBS. A blank control group, an experimental group and a PD-L1 monoclonal antibody control group (the atilizumab produced by the commercial original grinding manufacturer) are arranged on a 96-well plate, a CD47/PDL1 bispecific antibody BsAb-463 (the concentration is 20 mug/mL) is added into the hole of the experimental group, and the holes are sequentially diluted 5 times to 7 concentration points by adopting a gradient dilution method, and blank control holes are arranged at the same time. The 96-well plate was incubated at 4℃for 30mins, centrifuged (1000 rpm,2 min), and the wells were washed with PBS (200. Mu.L/2 times) after removal of the liquid from the wells; PE-anti-human-IgG-Fc-second-Antibody ((Jackson, cat. No. 109-116-098), diluted 1:200, 50. Mu.L) was added to each well, incubated at 4℃for 30mins, centrifuged (1000 rpm,2 min), and washed with PBS after removal of liquid (200. Mu.L/2 times); 5. PBS (200. Mu.L/well) was added to each well, and after resuspension of the cells, the cells were read by flow cytometry (brand BECKMAN, model BC 14115). The results of FACS-binding Activity analysis are shown in the following Table and graph, and the CD47/PD-L1 bispecific antibody BsAb-463 has binding activity to CD47 single-positive Raji cells, and the EC50 value of BsAb-463 is compared with that of CD47 bivalent monoclonal antibody Hu06-31, and the binding activity of double anti-BsAb-463 to CD47 on the cells is obviously weaker than that of double-bivalent CD47 monoclonal antibody Hu06-31, and the difference between the two is about 32 times.

5.2CD47/PD-L1 bispecific antibody blocked CD47 on Raji cells from binding to SIRPalpha and Raji cell counts, resuspended, and plated into 96 well plates (BIOFIL model: TCP 002096), 3E5 cells were added per well and washed 2 times with PBS. Simultaneously 25. Mu.L of SIRP-alpha-mFc (ex ACRO, cat. SIA-H52A 8) at a concentration of 5. Mu.g/mL and 25. Mu.L of CD47/PD-L1 bispecific antibody BsAb-463,2.5 at a concentration of 50. Mu.g/mL were added to the wells, 9 gradients of antibody were diluted, and a blank control was set; incubating at 4 ℃ for 30min; centrifugation at 1000rpm for 2min, the liquid in the wells was discarded, PBS wash, 200. Mu.l/Kong Xi times. mu.L of a PE-anti-Mouse-IgG-Fc-second-Antibody (Jackson, cat. No. 109-116-098) (1:200 dilution) was added to each well and incubated at 4℃for 30min; centrifuging at 1000rpm for 2min, discarding the liquid in the wells, and centrifuging at 1000rpm for 2min with 200 μl/Kong Xi times of PBS wash; 200. Mu.L/well PBS was added, the cells were resuspended, and read by flow cytometry (brand BECKMAN, model BC 14115). As shown in FIG. 10, the results of the FACS-blocking activity assay show that BsAb-463 blocks SIRP-alpha-mFc binding to CD47 single positive cells at an IC50 value of 2056ng/ml, and blocking activity was reduced by 6.5-fold compared to CD47 bivalent monoclonal antibody Hu 06-31.

Example 6 evaluation of Cross-species Activity of CD47/PD-L1 bispecific antibody

Analysis of binding Activity of 6.1CD47/PD-L1 bispecific antibodies against rat, mouse and cynomolgus monkey CD47 protein

mu-CD 47-His (ACRO, cat# CD7-M522 b), cyno-CD47-His (ACRO, cat# CD7-H52H 1), rat-CD47-His (cat# 80305-R08H) were diluted to 1. Mu.g/mL with a coating solution (0.2 mol/LNa2CO38mL,0.2mol/LNaHCO3 mL, 75mL distilled water was added thereto, pH was adjusted to 9.6), and the mixture was then added to wells of an ELISA plate (costar, cat# 3590) at a coating concentration of 50. Mu.L/well, followed by incubation at 37℃for 2 hours. The wells were discarded and 200. Mu.l/well of wash solution was washed with the Siemens WellwashVersa laboratory microplate reader (cat. No. 5165050)Washing 5 times. 200. Mu.L of blocking solution was added to each well overnight at 4 ℃. The wells were discarded and the plate was washed with 200. Mu.l/Kong Xi wash solution 5 times. 50 mu L of antibodies CD47/PD-L1 bispecific antibody BsAb-463, BT-hu-anti-CD47-006-H are added to each well ₃ L ₃ The concentrations of 5F9 and hIgG1 are 20 mug/mL, 5 times of dilution is carried out, 7 concentration points are carried out, blank control holes are arranged at the same time, and incubation is carried out for 60min at 37 ℃; the wells were discarded and the plate was washed with 200. Mu.l Kong Xi washes. mu.L of a secondary antibody of Peroxidase AffiniPure Goat Anti-Human IgG (Jackson, cat. No. 109-035-170) (1:10000 dilution) was added to each well and incubated at 37℃for 60min; the liquid in the wells was discarded, and the wells were washed 5 times with 200. Mu.l/wash solution using a plate washer. The color development solution was added at 50. Mu.L/well and incubated at 37℃for 15min. 50. Mu.L/well 2mol/L H are added ₂ SO ₄ The reaction was terminated and the OD450 values were read on a microplate reader (ThermoScientific, model multiskanFC).

The results are shown in FIGS. 11-13 (FIG. 11 shows ELISA-binding assay for cross-binding activity of different antibodies to Cyno-CD47, FIG. 12 shows ELISA-binding assay for cross-binding activity of different antibodies to Mus-CD47, and FIG. 13 shows ELISA-binding assay for cross-binding activity of different antibodies to Rat-CD 47), hu06-31 mab and BsAb-463 diabody have species cross-binding activity to cynomolgus CD47 protein, and have no species cross-binding activity to Rat CD47 protein, mouse CD47 protein.

Analysis of binding Activity of 6.2CD47/PD-L1 bispecific antibodies against rat, mouse and cynomolgus monkey PD-L1 proteins

mu-PDL 1-His (ACRO, cat# PD 1-M5220), cyno-CD47-His (ACRO, cat# PD1-H52H 4), rat-CD47-His (cat# 80450-R08H, yizhushen) were diluted to 1. Mu.g/mL with a coating solution (0.2 mol/LNa2CO38mL,0.2mol/LNaHCO31 mL, 75mL distilled water was added, pH was adjusted to 9.6), and the mixture was then added to wells of an ELISA plate (costar, cat# 3590) to coat 50. Mu.L/well, and incubated at 37℃for 2 hours. The wells were discarded and 200. Mu.l/Kong Xi washes were performed with a Siemens WellwashVersa laboratory microplate reader (cat. 5165050). 200. Mu.L of blocking solution was added to each well overnight at 4 ℃. The wells were discarded and the plate was washed with 200. Mu.l/Kong Xi wash solution 5 times. 50 mu L of antibodies CD47/PD-L1 bispecific antibody BsAb-463 and Atezolizumab, hIgG1 were added to each well The degree is 20 mug/mL, 5 times of dilution is carried out, 7 concentration points are arranged, blank control holes are arranged at the same time, and incubation is carried out for 60min at 37 ℃; the wells were discarded and the plate was washed with 200. Mu.l Kong Xi washes. mu.L of a secondary antibody of Peroxidase AffiniPure Goat Anti-Human IgG (Jackson, cat. No. 109-035-170) (1:10000 dilution) was added to each well and incubated at 37℃for 60min; the liquid in the wells was discarded, and the wells were washed 5 times with 200. Mu.l/wash solution using a plate washer. The color development solution was added at 50. Mu.L/well and incubated at 37℃for 15min. 50. Mu.L/well 2mol/L H are added ₂ SO ₄ The reaction was terminated and the OD450 values were read on a microplate reader (ThermoScientific, model multiskanFC).

The results are shown in FIGS. 14-16 (FIG. 14 shows ELISA-binding assay for cross-binding activity of different antibodies to Cyno-PD-L1, FIG. 15 shows ELISA-binding assay for cross-binding activity of different antibodies to Mus-PD-L1, and FIG. 16 shows ELISA-binding assay for cross-binding activity of different antibodies to Rat-PD-L1), atezolizumab and BsAb-463 diabodies have species cross-binding activity to Rat, mouse and cynomolgus monkey PD-L1 proteins.

EXAMPLE 7 binding of diabodies to biscationic tumor cells

Binding of 7.1CD47/PD-L1 bispecific antibodies to Raji-hPDL1-GFP biscationic tumor cells

Raji-hPDL1-GFP (commission) Zhejiang Meisen cell technologies Co.LtdOverexpression construct cell lines), were counted, resuspended, and plated into 96-well plates (bisil model: TCP 002096), 3E5 cells were added per well and washed 2 times with PBS. Setting blank control group, experimental group and PD-L1 monoclonal antibody control group on 96 well plateAbilib monoclonal antibodyMonoclonal antibody injection, taishengqi), CD47/PD-L1 bispecific antibody BsAb-463 (concentration 20 μg/mL) was added to wells of the experimental group and diluted 5-fold sequentially to 7 concentration points by gradient dilution, while antibody dilutions were set as blank wells. Centrifugation (1000 rpm,2 min), washing with PBS (200. Mu.L/time, 2 times) after removal of the liquid in the wells; PE-anti-human-IgG-Fc-second-Antibody (Jackson, cat. No. 109-116-098,1:200 dilution, 50. Mu.L) was added to each well, incubated at 4deg.C for 30mins, centrifuged (1000 rpm,2 min), and washed with PBS (200. Mu.L/2 times) after removal of liquid; 5. PBS (200. Mu.L/well) was added to each well, and after cell resuspensionRead with a flow cytometer (brand BECKMAN, model BC 14115). The results of FACS-binding Activity analysis are shown in FIG. 17 (average fluorescence intensity of binding of CD47/PDL1 bispecific antibody to Raji-hPDL1-GFP double positive cells), and the fact that CD47/PD-L1 bispecific antibody BsAb-463 can bind to CD47+PD-L1+ double positive Raji cells, and that both BsAb-463 and CD47+PD-L1+ double positive Raji cells have fluorescence signal Max values higher than that of both CD47 diabody and PD-L1 diabody indicates that BsAb-463 diabody can simultaneously target to two targets of CD47 and PD-L1 on cells. BsAb-463 diabody has binding activity to PD-L1 reduced by 3.8 times compared with PD-L1 mab Atezolizumab, CD47 monoclonal antibody Hu06-31 reduced by 2.6 times compared with CD47, bsAb-463 diabody has binding activity to CD47 or PD-L1 monocyang cells, and BsAb-463 has higher affinity to CD47+PD-L1+ biscyang cells.

EXAMPLE 8 analysis of binding Activity of the diabody to human erythrocytes

To further examine the binding activity of BsAb-463-H01 to human erythrocytes, we took about 0.2mL of healthy human erythrocytes, added 10mL of PBS, washed 2 times and then diluted with Buffer to adjust the cell density to 1E7/mL. Diluted BsAb-463-H01 (20. Mu.g/mL initial concentration, 4-fold gradient dilution, 8 consecutive concentrations), ab-CD47 mab (456) (20. Mu.g/mL initial concentration, 4-fold gradient dilution, 8 consecutive concentrations) and Human IgG1 (HG 1K) (20. Mu.g/mL initial concentration, 4-fold gradient dilution, 8 consecutive concentrations) were added to 100. Mu.L of red blood cell suspension (2 multiplex wells per concentration), single cell control wells were resuspended with 50. Mu.L Buffer, mixed, incubated at 4℃for 30min, centrifuged at 1000rpm for 2min to remove supernatant, and washed 1 Xwith Buffer. Adding 50 mu L of secondary antibody working solution into each hole to resuspend the cells, incubating for 30min at 4 ℃, centrifuging, washing for 1 time by using a buffer, resuspend the cells by using 200 mu L of buffer, transferring the cell suspension into a 1.5mL EP tube after the cell suspension is uniformly mixed, detecting PE fluorescence intensity by using a flow cytometry, and detecting on-machine. The results of the Mean Fluorescence Intensity (MFI) of binding of the different test samples to erythrocytes are shown in fig. 18.

FIG. 18 (mean fluorescence intensity of binding (MFI) of different antibodies to erythrocytes) shows that BsAb-463 and Hu06-31 both bind to erythrocytes in a dose-response curve with an EC50 of 2153ng/mL for double anti-BsAb-463 and an EC50 of 74.18ng/mL for Hu06-31, the binding activity to erythrocytes of double anti-BsAb-463 being weaker than that of Ab-CD47 monoclonal antibody Hu06-31 by a factor of about 29.

EXAMPLE 9 detection of human hemagglutination by diabodies

To observe the dissolution and aggregation of BsAb-463 antibodies of different concentrations on human erythrocytes in vitro, healthy human erythrocytes were taken, and approximately 10 volumes of physiological saline (0.9% sodium chloride injection) were added, centrifuged at 1500rpm for 10min, and washed 3 times. The obtained erythrocytes were prepared as a 2% suspension (V/V) by volume with physiological saline. 2% erythrocyte suspensions were added to 5mL clear centrifuge tubes, 400. Mu.L each, and 100. Mu.L of BsAb-463 (400. Mu.g/mL initial concentration, 2-fold gradient dilution, 6 consecutive concentrations) at different concentrations were added, each group was replicated 2. After being placed in a constant temperature incubator at 37 ℃, the solution was observed for 3 hours, and as shown in FIG. 19, the BsAb-463 double antibody was found to have no hemolysis and aggregation.

Example 10 detection of Selective binding Activity of double antibodies against tumor cells

Human erythrocytes (from Huizhiand sources, cat. No. A08218.11, lot No. 22L 001) were isolated from purchased human whole blood, counted, and cell density adjusted to 1X10 ⁷ /mL. Raji-hPDL1-GFP tumor cells (commission)Zhejiang Meisen powder Cell technology Co.LtdConstruction of overexpressing cell lines) and cell density was adjusted to 5x10 ⁵ /mL. Firstly, 50 mu L/Kong Gong cell suspension is paved on a round bottom 96-well plate (BIOFIL model: TCP 002096), then 50 mu L/well Raji-hPDL1-GFP tumor is paved on the same round bottom 96-well plate, and cells are mixed, so that the number of red blood cells is 20 times of that of tumor cells. Centrifuge at 1000rpm for 5min and remove supernatant. 4 antibodies to be tested, namely CD47/PD-L1 bispecific antibodies BsAb-463, BT-hu-anti-CD47-006-H3L1, 5F9 and hIgG1, are diluted to 20 mug/mL by buffer as initial concentration, then diluted 4 times to 8 concentration points, and 2 duplicate wells are made for each concentration. 50 μl of each concentration was added to each cell pellet, the cells were resuspended, mixed well, and incubated at 4deg.C for 30min. Centrifuge at 1000rpm for 2min, buffer wash once, discard supernatant. The secondary antibody R-PE-AffiniPureF (ab) 2-Fragment-go-Anti-Human-IgG (vendor Jackson, cat. No. 109-116-098) was diluted 1:200 with buffer, and EBioscienceTM Fixable viability D was added at the same timeYe eFluor 450 (ThermoFisher, cat. No. 48-4317-82) was diluted 1:10 with buffer, the two diluted fluorescent antibodies were added to 50. Mu.L staining solution/well, each cell pellet was resuspended, the cells were mixed, incubated at 4℃for 30min, centrifuged at 1000rpm for 2min, buffer washed 2 times, and approximately 150. Mu.L buffer resuspended. PE/FITC/PB450 fluorescence intensity was examined by flow cytometry, and 2E5 cells were analyzed by injection per well. And (3) detecting and analyzing by using a flow meter, and reading the respective fluorescence intensities of the tumor cells and the red blood cells in the mixed cells. FIG. 20 shows the Mean Fluorescence Intensity (MFI) of binding of different antibodies to tumor cells. FIG. 21 shows the Mean Fluorescence Intensity (MFI) of binding of different antibodies to erythrocytes

As shown in fig. 20 and 21, hu06-31 mab was substantially indistinguishable in binding activity to the mixed incubated Raji-hPDL1-GFP and human erythrocytes, whereas BsAb-463 mab was more potent in binding to Raji-hPDL1-GFP tumor cells expressing CD47, PD-L1 dual targets, whereas the binding ability to human erythrocytes expressing CD47 had been greatly reduced compared to mab, indicating that BsAb-463 mab had binding activity to preferentially tumor cells.

Example 11 ADCP Effect detection of diabodies

Mouse macrophage cell line Ana-1 (Shang En organism, cat# SNL-020) (ATCC, cat# HTX 1568) was inoculated as effector cells into 96-well cell culture plates, 1X10 per well ⁵ Individual cells were incubated at 37℃and 5% CO2 for 18 hours. HCC827 cells (Prunocel, CL-0094) (ATCC, cat# XY-XB-1356) as target cells were washed once with 1640 medium (Gibco cat# 72400047) and then the cell density was adjusted to 1X10 ⁶ Per mL, cells were stained in an incubator with 5. Mu.M CFSE (Invitrogen, cat. No. 65-0850-84) for 30min. After staining, the cells were resuspended in 1640 medium to adjust the cell density to 3×10 ⁶ /mL. The BsAb-463 diabodies, hu-mab and hIgG1 were each diluted in a 4-fold gradient with 1640 medium at 8 concentrations, 100. Mu.L of HCC827 cells were mixed with 100. Mu.L of the diluted samples and preincubated at 37℃and 5% CO2 for 50 min. Subsequently, ana-1 cells were depleted of supernatant. 200. Mu.L of the above-described pre-incubated HCC827 cells and sample mixture was added, and the cell plate was left to incubate at 37℃and 5% CO2 for 4-5 hours, and the incubation was completed The cells in each well were then repeatedly blown 4-5 times, washed 2 times with PBS, and then incubated with APC-F4/80 fluorescent antibody (Biolegend, cat# 123116) at 4℃for 30 minutes, and FACS buffer (PBS+2% BSA) washed 2 times for analysis by flow cytometry. Phagocytosis rate was analyzed. The phagocytosis rate is calculated by the following steps: FITC-F4/80 double-positive macrophage count/total macrophage count X100, results are shown in FIG. 22 (results of analysis of ADCC effect by double antibody).

As shown in FIG. 22, both the Hu06-31 mab and the BsAb-463 mab can induce the macrophage line to phagocytize tumor cells, and the phagocytosis rate change curve of BsAb-463 mab under different antibody concentrations shows that the phagocytosis effect of the BsAb-463 mab on the tumor cells of CD47/PD-L1 double positive is better than that of the Hu06-31 mab, and even slightly higher than that of the CD47 mab of the positive reference antibody 5F9.

Example 12 mediates Antibody Dependent Cellular Cytotoxicity (ADCC)

To observe ADCC effect of BsAb-463 antibody on tumor cells in vitro, NK cells were incubated with Raji-PD-L1 cells and HCC827 cells, respectively, and ADCC effect of BsAb-463 diabody, ab-CD47 mab (Hu 06-31) and PD-L1 mab (Atezolizumab) at different concentrations was observed. NK cells were recovered, resuspended in complete medium (1640+10% FBS), cultured overnight with the addition of hIL2 at a final concentration of 10ng/mL, and Raji-PD-L1 cells in the logarithmic phase were collected, washed 1 time with medium, and cell density was adjusted to 1E 6/mL, and 2mL was added with 3.4uL of BATDA to perform target cell labeling. The labeled target cell density was adjusted to 1E 5/ml, cell suspensions were plated, 100. Mu.L of cells were added to each well in a V-bottom transparent 96-well plate, and 3 wells were multiplexed. BsAb-463 antibody (Raji-PD-L1 system: 0.008, 0.04, 0.2. Mu.g/ml final concentration; HCC827 system: 0.04, 0.2, 1.0. Mu.g/ml final concentration), CD47 mab (Raji-PD-L1 system: 0.008, 0.04, 0.2. Mu.g/ml final concentration; HCC827 system: 0.04, 0.2, 1.0. Mu.g/ml final concentration) and PD-L1 antibody (Raji-PD-L1 system: 0.008, 0.04, 0.2. Mu.g/ml final concentration; HCC827 system: 0.04, 0.2, 1.0. Mu.g/ml final concentration; HCC827 system: 1.0. Mu.g/ml final concentration) were diluted to the corresponding concentrations, and added to the corresponding well plate 50. Mu.L per well from low to high; mixing, standing at 37deg.C, 5% CO2, and incubating for 30min. NK cells were collected, the cell density was adjusted to 2.5E5 cells/mL with 1640 complete medium (1 mM sulfopirone (supplier: widli, cat. HBW-B112), 100. Mu.L was added to tumor cell wells, the target ratio was 2.5:1, incubation was continued for 4h after the end of incubation time, centrifugation was carried out at 1500rpm for 5min, 20uL supernatant was pipetted into PE bottom-permeabilized white plates, 200uL europium was added, TRF was detected after shaking for 15min in the dark, the maximum release well was added with 25. Mu.L of lysate (HBSS+0.5% TritonX-100), and the cell killing rate was calculated FIG. 23 shows the ADCC effect of BsAb-463 antibody on Raji-PD-L1 cells and HCC827 cells, in the Raji-PD-L1 system, there was no obvious ADCC effect with respect to either Ab-CD47 monoclonal antibody alone or Ab-CD47 monoclonal antibody administration, no obvious ADCC effect was observed after shaking for 15min in the dark place, but no obvious ADCC effect was observed with respect to Ab-463 antibody 4 alone, no obvious ADCC effect was observed with respect to Ab-463 antibody administration with obvious contrast with either Ab-CD47 monoclonal antibody administration, and no obvious contrast was observed with obvious contrast with respect to the concentration of Ab-463 antibody was increased with respect to the concentration of Ab-463-4 antibody in the system, as shown in FIG. 23.

EXAMPLE 13 in vivo efficacy experiment of diabody-RAJI-PDL 1 animal model for subcutaneous tumor transplantation

To observe the inhibition of tumor proliferation by the CD47/PD-L1 bispecific antibody BsAb-463 of the present invention in antibodies, a Raji/PD-L1 subcutaneous engrafting tumor animal model was established using CB17/SCID mice (Beijing Bei Fu Biotechnology Co., ltd.). Raji/PD-L1 cells in logarithmic growth phase (entrusting Zhejiang Meisen cell technology Co., ltd. To construct an over-expression cell strain) are collected, the cell density is adjusted to 5E7/ml, the cells are mixed with matrigel in equal volume of 1:1 before inoculation, each mouse is inoculated with 0.1ml of cell mixture under the skin at the left anterior axilla, and the inoculation cell quantity is 5E 6/mouse. Tumor volume reaches 80-150 mm after 13 days of inoculation ³ Grouping was performed, and 36 mice were selected to be randomly grouped into 6 groups: igG1 control group (5.0 mg/kg), ab-CD47 mab group (5.0 mg/kg), PD-L1 mab group (5.0 mg/kg), ab-CD47 mab combined with PD-L1 mab group (5.0 mg/kg each), bsAb-463 low-high dose group (5.0, 10 mg/kg), 6 groups each, starting dosing on the day of grouping, and continuous observation for 21 days.Weighing weights at D0, D4, D7, D11, D14, D18 and D21 respectively after grouping, measuring and recording tumor length and diameter, calculating tumor volume, drawing a tumor growth curve according to the tumor volume, and calculating relative tumor proliferation rate; tumor weight inhibition was calculated by tumor weighing after euthanasia of the animals.

FIG. 24 shows the inhibition of CB17/SCID mice subcutaneously transplanted with Raji/PD-L1 cytoma model by BsAb-463 antibody.

As shown in FIG. 24, the present findings provided double anti-BsAb-463 has more remarkable anti-tumor effect in the tumor model of human Burkitt's lymphoma Raji cell subcutaneous transplantation. The tumor inhibition rates of the two dose groups of BsAb-463 diabodies exhibited dose-dependence. The BsAb-463 tumor inhibiting effect of the 5mg/kg dose group is superior to that of the CD47 monoclonal antibody-Hu 06-31 administration group and the PD-L1 monoclonal antibody administration group with the same dose. Analysis of q-value of drug combination effect by the golden formula method: the addition effect is achieved between the single Ab-CD47 monoclonal antibody (5 mg/kg) and the PD-L17 monoclonal antibody (5 mg/kg) combined drug; and the equal dose BsAb-463 double antibody (5 mg/kg) is equivalent to the synergistic effect between the combined drug of Ab-CD47 (5 mg/kg) and PD-L1 monoclonal antibody (5 mg/kg).

Claims

1. A bispecific antibody or antigen binding fragment comprising a binding arm a of CD47 binding specificity and a binding arm B of PD-L1 binding specificity, characterized in that the binding arm a comprises a light chain a (LA), a heavy chain a (HA) and a linker 1 (X1) connecting the light chain a (LA), heavy chain a (HA) and the binding arm B comprises a light chain B (LB), a heavy chain B (HB) and a linker 2 (X2) connecting the light chain B (LB), heavy chain B (HB) derived from an anti-PD-L1 antibody or antigen binding fragment, wherein the heavy chain a (HA) and the heavy chain B (HB) interact to form a FIH structure, and the binding arm a and the binding arm B form a heterodimer via the KIH structure.

2. The bispecific antibody or antigen-binding fragment of claim 1, wherein in the binding arm a, the heavy chain a (HA) comprises a heavy chain variable region a (VHA) and the light chain a (LA) comprises a light chain variable region a (VLA); wherein, the liquid crystal display device comprises a liquid crystal display device,

(1) The VHA comprises or is selected from:

(A1) An amino acid sequence as set forth in any one of SEQ ID NOs 22 to 28;

(A3) An amino acid sequence having at least 80% sequence identity to the amino acid sequence set forth in (A1); and/or (2) the VLA comprises or is selected from

(A4) An amino acid sequence as set forth in any one of SEQ ID NOs 57 to 63;

3. The bispecific antibody or antigen-binding fragment of any one of claims 1-2, wherein the PD-L1-specific binding arm B comprises a heavy chain variable region B (VHB) and a light chain variable region B (VLB); wherein the VHB comprises or is selected from:

(B1) An amino acid sequence as shown in SEQ ID NO. 85;

The VLB comprises or is selected from:

(B4) An amino acid sequence as shown in SEQ ID NO. 86;

4. The bispecific antibody or antigen-binding fragment of claim 1, wherein the heavy chain a (HA) comprises a heavy chain variable region a (HA), and the light chain a (LA) comprises a light chain variable region a (VLA), wherein the VHA comprises an amino acid sequence as set forth in SEQ ID No. 22; the VLA comprises an amino acid sequence as set forth in any one of SEQ ID NOs 57;

5. The bispecific antibody or antigen-binding fragment of any one of claim 1-4, wherein the heavy chain A comprises a heavy chain variable region A (VHA) and the light chain A comprises a light chain variable region A (VLA), wherein,

the heavy chain variable region a (VHA) comprises one or more amino acid sequences of (a) - (d):

(a) A VHA CDR1 comprising an amino acid sequence as set forth in any one of SEQ ID NOs 1 to 7;

(b) A VHA CDR2 comprising an amino acid sequence as set forth in any one of SEQ ID NOs 8 to 14;

(c) A VHA CDR3 comprising an amino acid sequence as set forth in any one of SEQ ID NOs 15 to 21;

(d) An amino acid sequence having at least 90% sequence identity to an amino acid sequence set forth in any one of (a) - (c); and/or

(f) VLA CDR2 comprising the amino acid sequence ATS;

(h) An amino acid sequence having at least 90% sequence identity to an amino acid sequence set forth in any one of (e) - (g).

6. The bispecific antibody or antigen-binding fragment of claims 1-5, wherein the binding arm a comprises the VHA CDR1, VHA CDR2, VHA CDR3, VLA CDR1, VLA CDR2, and VLA CDR3.

7. The bispecific antibody or antigen-binding fragment of any one of claims 1-6, wherein the binding arm a comprises a heavy chain variable region a (VHA) and a light chain variable region a (VLA), wherein

The VHA comprises a HACDR1, a HACDR2 and a HACDR3, wherein the VHA CDR1 comprises an amino acid sequence as shown in SEQ ID NO. 1; the VHA CDR2 comprises an amino acid sequence as set forth in any one of SEQ ID NOs 8-14; the VHA CDR3 comprises an amino acid sequence as shown in SEQ ID NO. 19; or (b)

The VHA CDR1 comprises an amino acid sequence shown as SEQ ID NO. 2; the VHA CDR2 comprises an amino acid sequence as shown in SEQ ID NO. 9; the VHA CDR3 comprises an amino acid sequence as shown in SEQ ID NO. 16; and

the VLA comprises VLACDR1, VLACDR2 and VLACDR3, wherein the VLA CDR1 comprises an amino acid sequence as shown in SEQ ID NO. 37; the VLA CDR2 comprises the amino acid sequence ATS; the VLA CDR3 comprises the amino acid sequence as shown in SEQ ID NO:51, or

The VLA CDR1 comprises the amino acid sequence as shown in SEQ ID NO. 36; the VLA CDR2 comprises the amino acid sequence ATS; the VLACDR3 comprises an amino acid sequence as set forth in SEQ ID NO. 50.

8. The bispecific antibody or antigen-binding fragment of claim 1-7, wherein the binding arm B comprises a heavy chain variable region B (VHB) and a light chain variable region B (VLB), wherein the VHB comprises a VHB CDR1, a VHB CDR2 and a VHB CDR3, and the VLB comprises a VLB CDR1, a VLB CDR2 and a VLB CDR3, wherein,

Wherein the VHB comprises the amino acid sequence of one or more VHB CDRs of (i) - (n):

(i) VHB CDR1 comprising an amino acid sequence as shown in SEQ ID NO. 80 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 43;

(j) VHB CDR2 comprising an amino acid sequence as shown in SEQ ID NO. 81 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 44;

(k) VHB CDR3 comprising the amino acid sequence shown as SEQ ID NO. 83 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown as SEQ ID NO. 45;

(l) An amino acid sequence having at least 85% sequence identity to an amino acid sequence set forth in any one of (i) - (k); and

the VLB comprises the amino acid sequence of one or more VLB CDRs of (m) - (p):

(m) a VLB CDR1 comprising the amino acid sequence as shown in SEQ ID NO. 83 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence shown in SEQ ID NO. 46;

(n) VLB CDR2 comprising the amino acid sequence SAS;

(o) a VLB CDR3 comprising the amino acid sequence as set forth in SEQ ID NO. 47 or an amino acid sequence having one or more conservative amino acid substitutions compared to the amino acid sequence set forth in SEQ ID NO. 84;

9. The bispecific antibody or antigen-binding fragment of claim 7, wherein the binding arm B comprises the VHBCDR1, VHBCDR2, VHB CDR3, VLB CDR1, VLB CDR2, and VLB CDR3.

10. The bispecific antibody or antigen-binding fragment of any one of claim 1-9, wherein the binding arm A specific for CD47 comprises a heavy chain variable region A (VHA) and a light chain variable region A (VLA),

the VHA CDR1 comprises an amino acid sequence shown as SEQ ID NO. 1; the VHA CDR2 comprises an amino acid sequence shown as SEQ ID NO. 8; the VHA CDR3 comprises an amino acid sequence as shown in SEQ ID NO. 15; the VLACDR1 comprises an amino acid sequence as set forth in SEQ ID NO. 36; the VLA CDR2 comprises the amino acid sequence ATS; the VLA CDR3 comprises the amino acid sequence as shown in SEQ ID NO. 50; and

the VHB CDR1 comprises an amino acid sequence as shown in SEQ ID NO. 145; the VHB CDR2 comprises an amino acid sequence as shown in SEQ ID NO. 146; the VHB CDR3 comprises the amino acid sequence shown as SEQ ID NO 147; the VLB CDR1 comprises the amino acid sequence depicted as SEQ ID NO 84; the VLB CDR2 comprises the amino acid sequence SAS; the VLB CDR3 comprises the amino acid sequence as shown in SEQ ID NO. 84.

11. The antibody or antigen-binding fragment of any one of claim 1 to 10,

the heavy chain A (HA) comprises an amino acid sequence shown in any one of SEQ ID NO. 29-35, or an amino acid sequence with at least 80% homology with the amino acid sequence shown in any one of SEQ ID NO. 29-35; the heavy chain B (HB) comprises an amino acid sequence shown as SEQ ID NO. 87 or an amino acid sequence having at least 80% homology with the amino acid sequence shown as SEQ ID NO. 87; the light chain A (LA) comprises an amino acid sequence as shown in any one of SEQ ID NO. 64-70, or an amino acid sequence having at least 80% homology with the amino acid sequence as shown in any one of SEQ ID NO. 64-70; and/or an amino acid sequence having at least 80% homology with the amino acid sequence shown in SEQ ID NO. 87; the light chain B (LB) comprises an amino acid sequence shown as SEQ ID NO. 72 or an amino acid sequence with at least 80% homology with the amino acid sequence shown as SEQ ID NO. 88.

12. The bispecific antibody or antigen-binding fragment thereof according to any one of claims 1 to 11, wherein

The anti-CD 47 antibody comprises a polypeptide comprising SEQ ID NO:29 and the light chain of the amino acid sequence shown in SEQ ID NO. 64

The PD-L1 antibody comprises a polypeptide comprising SEQ ID NO:87 and the heavy chain of the amino acid sequence of SEQ ID NO:88, and a light chain of the amino acid sequence of seq id no.

13. The bispecific antibody or antigen-binding fragment of any one of claims 1-12, wherein the binding arm a further comprises a heavy chain constant region a (CHA) comprising an amino acid sequence as shown in SEQ ID No. 75 or an amino acid sequence having NO less than 80%, or NO less than 85%, or NO less than 90%, or NO less than 95%, or NO less than 98%, or NO less than 99.5%, or NO less than 99.8%, or an amino acid sequence having a sequence identity to amino acid sequence as shown in SEQ ID No. 73, and a light chain constant region a (CLA) comprising an amino acid sequence as shown in SEQ ID No. 73 or an amino acid sequence having NO less than 80%, or NO less than 85%, or NO less than 90%, or NO less than 95%, or NO less than 98%, or NO less than 99.5%, or NO less than 99.8%, or an amino acid sequence having a sequence identity to amino acid sequence as shown in SEQ ID NO 75, and a light chain constant region (CLA) comprising an amino acid sequence as shown in SEQ ID No. 73 or an amino acid sequence having NO less than 80%, or NO less than 85%, or NO less than 95%, or NO less than 98%, or NO less than 99.5%, or NO less than 99.8%, or a constant region (CLA) comprising a constant region of amino acid sequence as shown in SEQ ID No. 90%, or NO sequence having a sequence identity to amino acid sequence as shown in SEQ ID No. 73, and a light chain constant region a (CLA) comprising an amino acid sequence as shown in NO 95% or NO less than 80%, or NO 95%, or not less than 90%, or not less than 95%, or not less than 98%, or not less than 99.5%, or not less than 99.8% sequence identity.

14. A bispecific antibody or antigen-binding fragment comprising a binding arm a specific for CD47 and a binding arm B specific for PD-L1, characterized in that the binding arm a comprises a heavy chain variable region a (VHA), a light chain variable region (VLA), and a linker 1 (X1) between the heavy chain variable region a (VHA) and the light chain variable region (VLA), the binding arm B comprising a heavy chain variable region B (VHB), a light chain variable region B (VLB), and a linker 2 (X2) between the heavy chain variable region B (VHB) and the light chain variable region B (VLB); wherein the heavy chain variable region a (VHA) and the light chain variable region (VLA) pair to form a CD47 antigen binding site and the heavy chain variable region B (VHB) and the light chain variable region B (VLB) pair to form a PD-L1 antigen binding site.

15. A bispecific antibody or antigen-binding fragment comprising binding arm a that specifically binds CD47 and binding arm B that specifically binds PD-L1; wherein, the liquid crystal display device comprises a liquid crystal display device,

the binding arm a comprises a scFab (a) comprising a light chain variable region a (VLA), a light chain constant region a (CLA), a heavy chain variable region a (VHA), and a first heavy chain constant region a (CH 1A) derived from an anti-CD 47 antibody, and a linker 1 (X1) connecting the light chain constant region a (CLA) and the heavy chain variable region a (VHA), wherein the scFab (a) comprises an amino acid sequence as shown in SEQ ID NO:77 or an amino acid sequence having NO less than 80% sequence identity to the amino acid sequence shown in SEQ ID NO: 77; and/or the number of the groups of groups,

The binding arm B comprises a scFab (B) comprising a light chain variable region B (VLB), a light chain constant region B (CLB), a heavy chain variable region B (VHB), and a first heavy chain constant region B (CH 1B) of an anti-PD-L antibody, and a linker 2 (X2) connecting the light chain constant region B (CLB) and the heavy chain variable region B (VHB), the scFab (B) comprising an amino acid sequence as set forth in SEQ ID NO:93 or an amino acid sequence having NO less than 80% sequence identity to the amino acid sequence set forth in SEQ ID NO: 93.

16. The bispecific antibody or antigen-binding fragment of claim 15, wherein the scFab (a) and Fc (a) are fused to a single chain peptide chain a that specifically binds to CD 47; the scFab (B) and Fc (B) are fused to a single-chain peptide chain B that specifically binds PD-L1; wherein the Fc (B) is subjected to amino acid modification to form a knob structure, the Fc (A) is subjected to amino acid modification to form a hole structure, the CD47 specific binding arm A and the PD-L1 specific binding arm B form a KIH structure between the knob and the hole, and the CD47 specific binding arm A and the PD-L1 specific binding arm B form a heterodimer through the KIH structure; wherein the Fc (A) comprises an amino acid sequence as shown in SEQ ID NO. 78 or an amino acid sequence having not less than 80% sequence identity to the amino acid sequence as shown in SEQ ID NO. 78, and the Fc (B) comprises an amino acid sequence as shown in SEQ ID NO. 94 or an amino acid sequence having not less than 80% sequence identity to the amino acid sequence as shown in SEQ ID NO. 94; optionally, the heavy chain constant region CHB of the PD-L1 binding arm comprises the knob structure, the amino acid sequence of the CHB is shown as SEQ ID NO. 75 or an amino acid sequence with not less than 80% sequence identity with the amino acid sequence shown as SEQ ID NO. 75, the heavy chain constant region CHA of the CD47 binding arm comprises the hole structure, and the amino acid sequence of the CHA is shown as SEQ ID NO. 90 or an amino acid sequence with not less than 80% sequence identity with the amino acid sequence shown as SEQ ID NO. 90.

17. A bispecific antibody or antigen-binding fragment comprising a binding arm a that specifically binds CD47 and a binding arm B that specifically binds PD-L1, wherein the binding arm a that specifically binds CD47 comprises a light chain constant region a (CLA) derived from an anti-CD 47 antibody or light chain variable region a (VLA) or from an anti-CD 47 antibody or variant thereof, a heavy chain variable region a (VHA) derived from an anti-CD 47 antibody or variant thereof, and a heavy chain constant region a (CHA) derived from an anti-CD 47 antibody or variant thereof; wherein the antigen binding arm A that specifically binds CD47 is N-terminal to C-terminal comprising the structure shown by the formula VLA-CLA-X1-VHA-CHA,

wherein the "-" is a peptide bond,

the binding arm A which specifically binds CD47 comprises the amino acid sequence shown as SEQ ID NO:79 or comprises an amino acid sequence having a sequence homology of more than 85%, preferably more than 90%, more preferably more than 95%, more preferably more than 98.5% or most preferably more than 99.5% with the amino acid sequence shown as SEQ ID NO: 79;

the CLA comprises the amino acid sequence as shown in SEQ ID NO. 73, or comprises an amino sequence having a sequence homology of more than 80%, preferably more than 90%, more preferably more than 95%, more preferably more than 98.5% or most preferably more than 99.5% with the amino acid sequence shown in SEQ ID NO. 73;

The CHA comprises an amino acid sequence as set forth in SEQ ID No. 75, or comprises an amino acid sequence having greater than 80%, preferably greater than 90%, more preferably greater than 95%, more preferably greater than 98.5%, or most preferably greater than 99.5% sequence homology with the amino acid sequence set forth in SEQ ID No. 75;

and/or

The binding arm B that specifically binds PD-L1 comprises a light chain variable region B (VLB), a light chain constant region B (CLB), a heavy chain variable region B (VHB), and a heavy chain constant region B (CHB) of an anti-PD-L1 antibody; the binding arm B of the specific binding PD-L1 is a structure from N end to C end and comprises VLB-CLB-X2-VHB-CHB,

wherein "-" is a peptide bond and X2 is a linker structure; the X2 comprises an amino acid sequence shown as SEQ ID NO. 77;

the antigen binding arm B that specifically binds PD-L1 comprises an amino acid sequence as set forth in SEQ ID NO. 95, or an amino acid sequence having greater than 85%, preferably greater than 90%, more preferably greater than 95%, more preferably greater than 98.5% or most preferably greater than 99.5% sequence homology with the amino acid sequence set forth in SEQ ID NO. 95; wherein the CLB comprises the amino acid sequence as set forth in SEQ ID No. 89 or comprises an amino sequence having greater than 80%, preferably greater than 90%, more preferably greater than 95%, more preferably greater than 98.5% or most preferably greater than 99.5% sequence homology with the amino acid sequence set forth in SEQ ID No. 89; the CHB comprises the amino acid sequence shown as SEQ ID No. 90 or comprises an amino sequence having greater than 80%, preferably greater than 90%, more preferably greater than 95%, more preferably greater than 98.5% or most preferably greater than 99.5% sequence homology with the amino acid sequence shown as SEQ ID No. 90.

18. The bispecific antibody or antigen-binding fragment of any one of claims 13-17, wherein the light chain constant region CLA of the anti-CD 47 antibody and the CLB of the anti-PD-L1 antibody optionally comprise light chain constant regions of kappa, lambda chains or variants thereof, and the CHA of the anti-CD 47 antibody and the CHB of the anti-PD-L1 antibody optionally comprise heavy chain constant regions of IgG1, igG2, igG3 or IgG4 or variants thereof; or the light chain constant region a (CLA) of said anti-CD 47 antibody and the light chain constant region (CLB) of said anti-PD-L1 antibody optionally a human antibody kappa chain or lambda chain constant region, or the CHA and the CHB optionally a human antibody IgG1 or IgG4 constant region, or the CHA and the CHB optionally a human antibody IgG1 constant region.

19. The bispecific antibody or antigen-binding fragment of any one of claims 1-18, wherein the CD47 antibody is a humanized anti-CD 47 antibody and the anti-PD-L1 antibody is a humanized anti-PD-L antibody.

20. The bispecific antibody or antigen-binding fragment of any one of claims 1-19, wherein said antigen-binding fragment is selected from the group consisting of Fab, fab ' -SH, fv, scFv or (Fab ') 2 fragments, or wherein said antigen-binding fragments are each selected from the group consisting of Fab and (Fab ') 2, scFab.

21. A polynucleotide encoding the bispecific antibody or antigen-binding fragment of any one of claims 1-20, comprising: (1) a polynucleotide encoding a heavy or light chain that binds arm a; (2) a polynucleotide encoding a heavy or light chain that binds arm B; (3) A polynucleotide encoding a heavy chain variable region a (VHA) or a light chain variable region a (VLA) of binding arm a; or (4) a polynucleotide encoding a seed heavy chain variable region B (VHB) or a light chain variable region B (VLB) that binds to arm B; (5) a polynucleotide encoding a knob structure; (6) a polynucleotide encoding a hole structure;

or alternatively, the first and second heat exchangers may be,

the polynucleotide comprises: (1) A nucleotide sequence encoding a light chain or variant thereof as set forth in any one of SEQ ID NOs 64-70, (2) a nucleotide sequence encoding a heavy chain or variant thereof having at least 90% sequence homology thereto as set forth in any one of SEQ ID NOs 29-35; (3) A nucleotide sequence encoding a binding arm a as set forth in SEQ ID No. 79 or a variant having the same function with at least 90% sequence homology thereto: (4) A nucleotide sequence encoding a variant having the same function as binding arm B shown in SEQ ID No. 95 or at least 90% sequence homology thereto; (5) A nucleotide sequence encoding scFab (A) as shown in SEQ ID NO. 77 or a variant thereof having at least 90% sequence homology thereto; (6) A nucleotide sequence encoding scFab (B) as shown in SEQ ID NO. 93 or a variant having at least 90% sequence homology thereto; (7) A nucleotide sequence encoding a light chain variable region A (VHA) as set forth in any one of SEQ I NO 57-63 or a variant having at least 90% sequence homology thereto; (8) A nucleotide sequence encoding a heavy chain variable region B (VHB) as set forth in any one of SEQ I NOs 22 to 28 or a variant having the same function with at least 90% sequence homology thereto;

Or further alternatively

The polynucleotide comprises: (1) A nucleotide sequence encoding a light chain a (LA) of binding arm a as set forth in SEQ ID No. 29 or a variant having the same function with at least 90% sequence homology thereto, (2) a nucleotide sequence encoding a heavy chain a (HA) of binding arm a as set forth in SEQ ID No. 64 or a variant having the same function with at least 90% sequence homology thereto; (3) A nucleotide sequence encoding a light chain B (LB) of binding arm B as set forth in SEQ ID NO 88 or a variant having the same function with at least 90% sequence homology thereto, (4) a nucleotide sequence encoding a heavy chain B (HB) of binding arm B as set forth in SEQ ID NO 87 or a variant having the same function with at least 90% sequence homology thereto; (4) A nucleotide sequence as set forth in SEQ ID No. 106 encoding a binding arm a as set forth in SEQ ID No. 79 or a variant having the same function with at least 90% sequence homology thereto: (5) A nucleotide sequence as set forth in SEQ ID NO. 107 encoding a binding arm B as set forth in SEQ ID NO. 95 or a variant having at least 90% sequence homology thereto; (5) A nucleotide sequence encoding scFab (A) as shown in SEQ ID NO. 77 or a variant thereof having at least 90% sequence homology thereto; (6) A nucleotide sequence encoding scFab (B) as shown in SEQ ID NO. 93 or a variant having at least 90% sequence homology thereto; (7) A nucleotide sequence of a nucleotide sequence shown as SEQ ID NO. 96 or having homology of not less than 90% thereto encoding an amino acid sequence shown as SEQ ID NO. 57 or a light chain variable region A (VHA) or a variant having at least 90% sequence homology thereto; (8) A nucleotide sequence encoding a heavy chain variable region A (VHA) as shown in SEQ ID NO. 22 or a variant having the same function with at least 90% sequence homology thereto as shown in SEQ ID NO. 99 or a nucleotide sequence having not less than 90% homology thereto; (7) A nucleotide sequence of a nucleotide sequence shown as SEQ ID NO. 101 or having homology of not less than 90% thereto encoding an amino acid sequence shown as SEQ ID NO. 86 or a light chain variable region B (VHB) or a variant having at least 90% sequence homology thereto; (8) A nucleotide sequence encoding a heavy chain variable region B (VHB) as shown in SEQ ID NO. 85 or a variant having the same function with at least 90% sequence homology thereto as shown in SEQ ID NO. 104 or a nucleotide sequence having not less than 90% homology thereto.

22. An expression vector comprising a polynucleotide encoding the bispecific antibody or antigen-binding fragment of any one of claims 1-21.

23. A cell comprising a polynucleotide encoding the bispecific antibody or antigen binding fragment of any one of claims 1-20, or the expression vector of claim 22.

24. A method of producing the bispecific antibody or antigen-binding fragment of any one of claims 1-20, comprising:

culturing a host cell according to claim under conditions that cause the production of a bispecific antibody according to any one of claims 1-20, and purifying the produced bispecific antibody.

25. A composition comprising the bispecific antibody or antigen-binding fragment of any one of claims 1-20, the polynucleotide of claim 21, the expression vector of claim 22, or the cell of claim 23, and a pharmaceutically acceptable excipient.

26. Use of the antibody or antigen binding fragment of any one of claims 1-20, the polynucleotide of claim 21, the expression vector of claim 22, the cell of claim 23, or the composition of claim 25 in the manufacture of a medicament for treating a disease associated with simultaneous overexpression of CD47 and PD-L1, wherein the disease associated with simultaneous overexpression of CD47 and PD-L1 comprises or is selected from autoimmune diseases, acute and chronic inflammatory diseases, infectious diseases, cancer.

27. The use of claim 26, wherein the disease associated with simultaneous overexpression of CD47 and PD-L1 comprises a neoplastic disorder selected from at least one of a hematological disorder selected from the group consisting of acute myelogenous leukemia, chronic myelogenous leukemia, acute lymphocytic leukemia, chronic lymphocytic leukemia, adult T-cell leukemia, multiple myeloma, mixed leukemia, non-hodgkin's lymphoma; the solid tumor is at least one of lymphoma, breast cancer, head and neck cancer, gastric cancer, lung cancer, esophageal cancer, intestinal cancer, ovarian cancer, cervical cancer, liver cancer, renal cancer, pancreatic cancer, bladder cancer, colorectal cancer, glioma and melanoma.

28. A kit or article of manufacture comprising the bispecific antibody or antigen binding fragment of any one of claims 1-20, the polynucleotide of claim 21 or the expression vector of claim 22, or the cell of claim 23, or the composition of claim 25.