Bifunctional molecules targeting PD-l1 and TGF-beta

Abstract

To provide bifunctional molecules targeting PD-L1 and TGF-beta.SOLUTION: An anti-PD-L1 antibody having superior activity in blocking the PD-1 and PD-L1 interaction is provided. Also provided is a multifunctional molecule comprising an anti-PD-L1 antibody or a fragment thereof fused to extracellular domain of human TGF-β receptor type 2. The disclosure, in several embodiments, provides a bifunctional molecule targeting both PD-L1 protein and TGF-β. The disclosed PD-L1 targeting unit comprised of an anti-PD-L1 antibody is fused to the extracellular domain of human transforming growth factor-β(TGF-β) receptor II functioning as a trap for TGF-β. The experimental data show that these novel bifunctional molecules are more effective than M7824, the current lead candidate in clinical development.SELECTED DRAWING: None

Description

[Technical field]

[0001] background Exciting advances in cancer immunotherapy in recent years have led to a paradigm shift in oncology. The most notable results have been T cell-based therapies, including immune checkpoint inhibitors (ICIs), genetically engineered T cells and bispecific antibodies (BsAbs). T cells represent a major class of immune surveillance and tumor eradication with excellent specificity and long-term memory. However, in the tumor microenvironment, T cells can become exhausted or tolerant to tumor cells. T cell exhaustion is commonly associated with the overexpression of inhibitory receptors, including programmed death receptor-1 (PD-1), cytotoxic T lymphocyte antigen-4 (CTLA-4), lymphocyte activation gene-3 (LAG-3), T cell immunoglobulin domain and mucin domain-3 (TIM-3), IL-10 receptor, and killer immunoglobulin receptor. [Background technology]

[0002] Therapies based on monoclonal antibodies (mAbs) to counter these checkpoint molecules can remove the brakes that inhibit tumor-infiltrating T cells, thereby achieving significant clinical benefits in different malignancies. For example, blocking PD-1 / PD-L1 interaction can enhance immune normalization and strengthen anticancer responses. However, the notable deficiencies of PD-1 / PD-L1 blockade are inconsistent across homogenous study populations with similar tumor characteristics. In addition, PD-1 / PD-L1 blockade treatments can also cause certain inflammatory side effects in some patients. The limitations of monotherapy with PD-1 / PD-L1 blockade and the lack of promising alternatives have necessitated the exploration of combination treatment methods that can activate antitumor immunity and enhance treatment efficacy.

[0003] M7824 (bintrafusp alfa) is a bifunctional protein composed of a monoclonal antibody against programmed death ligand 1 (PD-L1) fused to the extracellular domain of human transforming growth factor-β (TGF-β) receptor II, acting as a "trap" for all three TGF-β isoforms. The PD-L1 portion is based on avelumab, which is approved for the treatment of Merkel cell carcinoma and urothelial carcinoma. However, current clinical data indicate that the use of M7824 is associated with unwanted skin growths and the overall response rate was only about 35%-40% in phase II trials for patients with HPV-positive malignancies. Improved treatments are therefore needed. Summary of the Invention [Means for solving the problem]

[0004] overview The present disclosure provides, in some embodiments, bifunctional molecules that target both PD-L1 protein and TGF-β. The disclosed PD-L1 targeting unit, which is composed of an anti-PD-L1 antibody, is fused to the extracellular domain of human transforming growth factor-β (TGF-β) receptor II, which acts as a trap for TGF-β. Experimental data show that these new bifunctional molecules are more effective than M7824, the current lead candidate in clinical development.

[0005] Thus, according to one embodiment of the present disclosure, there is provided a multifunctional molecule comprising an anti-PD-L1 (programmed death ligand 1) antibody, or a fragment thereof, and an extracellular domain of human TGF-βRII (TGF-β receptor type 2), wherein the anti-PD-L1 antibody, or a fragment thereof, has specificity for human PD-L1 protein, and comprises a heavy chain variable region (VH) comprising VH CDR1, VH CDR2, and VH CDR3, and a light chain variable region (VL) comprising VL CDR1, VL CDR2, and VL CDR3, wherein VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2, and VL CDR3 respectively comprise the amino acid sequences of SEQ ID NOs: 7-12 or SEQ ID NOs: 13-18, or wherein VH CDR1 comprises SEQ ID NO: 19, VH CDR2 comprises SEQ ID NO: 20, 91, or 92, VH CDR3 comprises SEQ ID NO: 21, VL CDR1 comprises SEQ ID NO: 22, and VL a multifunctional molecule, in which the CDR2 comprises SEQ ID NO: 23, the VL CDR3 comprises SEQ ID NO: 24 or 93, and the human TGF-βRII extracellular domain comprises the amino acid sequence of SEQ ID NO: 72, fused to an anti-PD-L1 antibody or fragment thereof.

[0006] In one embodiment, provided is an anti-PD-L1 (programmed death ligand 1) antibody or fragment thereof that has specificity for human PD-L1 protein, and comprises a heavy chain variable region (VH) comprising VH CDR1, VH CDR2, and VH CDR3, and a light chain variable region (VL) comprising VL CDR1, VL CDR2, and VL CDR3, wherein VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2, and VL CDR3 respectively comprise the amino acid sequence of SEQ ID NO:7-12 or SEQ ID NO:13-18, or wherein VH CDR1 comprises SEQ ID NO:19, VH CDR2 comprises SEQ ID NO:20, 91 or 92, VH CDR3 comprises SEQ ID NO:21, VL CDR1 comprises SEQ ID NO:22, VL CDR2 comprises SEQ ID NO:23, and VL CDR3 comprises SEQ ID NO:24 or 93.

[0007] Also provided is a multifunctional molecule comprising an antibody or antigen-binding fragment thereof fused to the N-terminus of the amino acid sequence of SEQ ID NO: 72 via a peptide linker, wherein the peptide linker is (a) at least 30 amino acid residues in length, or (b) at least 25 amino acid residues in length and comprises an alpha-helical motif.

[0008] Uses and methods for treating cancer with any of the molecules of the present disclosure are also provided. [Brief description of the drawings]

[0009] [Figure 1] Figure 1 shows that 47C6A3, 67F3G7 and 89C10H8 can bind to human PD-L1 with high affinity.

[0010] [Diagram 2] Figure 2 shows that the 47C6A3, 67F3G7 and 89C10H8 antibodies can potently bind to PD-L1 expressed on mammalian cells.

[0011] [Diagram 3] Figure 3 shows that the 47C6A3, 67F3G7 and 89C10H8 antibodies can bind to cynomolgus monkey PD-L1 with high affinity, but cannot bind to rat or mouse PD-L1.

[0012] [Figure 4] Figure 4 shows that 47C6A3, 67F3G7 and 89C10H8 can efficiently inhibit the binding of human PD-L1 to human PD1.

[0013] [Diagram 5] Figure 5 shows the binding kinetics of 47C6A3, 67F3G7 and 89C10H8 to recombinant PD-L1.

[0014] [Figure 6-1]Figures 6A-C show that all humanized antibodies tested have comparable binding efficiency to human PD-L1 as the chimeric antibodies. [Figure 6-2] Same as above. [Figure 6-3] Same as above.

[0015] [Figure 7] Figure 7 shows that the tested humanized antibodies can bind highly efficiently to PD-L1 expressed on mammalian cells, similar to the chimeric antibody.

[0016] [Figure 8-1] Figures 8A-8C show that several humanized antibodies can efficiently inhibit the binding of human PD-L1 to human PD1. [Figure 8-2] Same as above. [Figure 8-3] Same as above. [Figure 8-4] Same as above.

[0017] [Figure 9-1] Figures 9A-C show that several humanized antibodies can efficiently block the binding of human PD-L1 to human CD80. [Figure 9-2] Same as above. [Figure 9-3] Same as above. [Figure 9-4] Same as above.

[0018] [Figure 10] Figure 10 shows the binding kinetics of LP008-06, LP008-06a, LP008-06a-DA and LP008-06a-ES to recombinant human PD-L1.

[0019] [Figure 11] Figure 11 shows the binding kinetics of LP008-02 to human PD-L1 and human TGF-β1.

[0020] [Figure 12]Figure 12 shows that LP008-02 and LP008-06a-ES can block PD1 and PD-L1 interactions with higher affinity than M7824.

[0021] [Figure 13] FIG. 13 shows that M7824, LP008-02 and LP008-06a-ES can effectively block TGF-β canonical signaling.

[0022] [Figure 14] Figure 14 shows that LP008-02 and LP008-06a-ES bind to human PD-L1 with high affinity.

[0023] [Figure 15] Figure 15 shows that LP008-02 and LP008-06a-ES can bind with high affinity to cynomolgus monkey PD-L1, but cannot bind to rat PD-L1 or mouse PD-L1.

[0024] [Figure 16] FIG. 16 shows that LP008-02 and LP008-06a-ES have binding efficiencies to human TGF-β comparable to M7824.

[0025] [Figure 17] FIG. 17 shows that LP008-02 and LP008-06a-ES have comparable binding efficiency to cynomolgus monkey TGF-β, mouse TGF-β and rat TGF-β, similar to M7824.

[0026] [Figure 18-1] 18A-B show the drug effects of LP008-02 and LP008-06a-ES in an animal model. [Figure 18-2] Same as above.

[0027] [Figure 19]FIG. 19 shows that all tested modified bifunctional molecules had binding efficiencies to human TGF-β comparable to LP008-02-1.

[0028] [Figure 20] FIG. 20 shows that all tested modified bifunctional molecules are able to effectively block TGF-β canonical signaling.

[0029] [Figure 21] FIG. 21 shows that all tested modified bifunctional molecules had binding efficiencies to human TGF-β comparable to LP008-02-1.

[0030] [Figure 22] FIG. 22 shows that all tested modified bifunctional molecules are able to effectively block TGF-β canonical signaling.

[0031] [Figure 23] Figure 23 shows that antibodies MPDL3280A, 47C6A3, Hu67F3G7-22 and Hu89C10H8-7 can block PD1 and PD-L1 interactions with high affinity. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0032] definition It should be noted that the term "a" or "an" entity refers to one or more of that entity. For example, "an antibody" is understood to refer to one or more antibodies. Thus, the terms "a" (or "an"), "one or more," and "at least one" can be used interchangeably herein.

[0033] As used herein, "antibody" or "antigen-binding polypeptide" refers to a polypeptide or polypeptide complex that specifically recognizes and binds to an antigen. An antibody can be a whole antibody and any antigen-binding fragment or single chain thereof. Thus, the term "antibody" includes any protein or peptide-containing molecule that contains at least a portion of an immunoglobulin molecule that has biological activity of binding to an antigen. Examples of such include, but are not limited to, at least a portion of a heavy or light chain complementarity determining region (CDR) or ligand-binding portion thereof, a heavy or light chain variable region, a heavy or light chain constant region, a framework (FR) region or any portion thereof, or a binding protein.

[0034] The term "antibody fragment" or "antigen-binding fragment" as used herein is a portion of an antibody, such as F(ab')2, F(ab)2, Fab', Fab, Fv, scFv, etc. Regardless of structure, an antibody fragment binds to the same antigen recognized by the intact antibody. The term "antibody fragment" includes aptamers, spiegelmers, and diabodies. The term "antibody fragment" also includes any synthetic or genetically engineered protein that acts like an antibody by binding to a specific antigen to form a complex.

[0035] The term antibody encompasses various broad classes of polypeptides that can be distinguished biochemically. Those skilled in the art will appreciate that heavy chains are classified as gamma, mu, alpha, delta, or epsilon (γ, μ, α, δ, ε), with several subclasses therein (e.g., γ1-γ4). It is the nature of this chain that determines the "class" of an antibody as IgG, IgM, IgA, IgG, or IgE, respectively. Immunoglobulin subclasses (isotypes), e.g., IgG1, IgG2, IgG3, IgG4, IgG5, etc., are well characterized and are known to confer functional specialization. Modified versions of each of these classes and isotypes are readily discernible to one of skill in the art in light of the present disclosure, and thus are within the scope of the present disclosure. While all immunoglobulin classes are clearly within the scope of the present disclosure, the following discussion will generally be directed to the IgG class of immunoglobulin molecules. For IgG, a standard immunoglobulin molecule contains two identical light polypeptide chains of molecular weight approximately 23,000 daltons and two identical heavy polypeptide chains of molecular weight 53,000-70,000. The four chains are typically joined by disulfide bonds in a "Y" configuration, where the light chains support the heavy chains, which begin at the mouth of the "Y" and continue through the variable region.

[0036] "Specifically binds" or "has specificity" generally means that an antibody binds to an epitope through its antigen-binding domain, and that the binding involves some complementarity between the antigen-binding domain and the epitope. According to this definition, an antibody is said to "specifically bind" to an epitope if it binds to that epitope through its antigen-binding domain more readily than it binds to a random, unrelated epitope. The term "specificity" is used herein to qualify the relative affinity with which a particular antibody binds to a particular epitope. For example, antibody "A" may be considered to have a higher specificity for a given epitope than antibody "B", or antibody "A" may be said to bind epitope "C" with a higher specificity than it has for related epitope "D".

[0037] As used herein, the term "treat" or "treatment" refers to both therapeutic treatment and prophylactic or preventative measures, the purpose being to prevent or slow down (alleviate) undesirable physiological changes or disorders, such as the progression of cancer. Beneficial or desired clinical results include, but are not limited to, alleviation of symptoms, whether detectable or undetectable, reduction in the extent of disease, stabilization (i.e., not worsening) of disease, delay or slowing of disease progression, improvement or alleviation of disease symptoms, and remission (whether partial or total). "Treatment" can also mean prolonging survival compared to expected survival if not receiving treatment. Those in need of treatment include those already with the condition or disorder, as well as those prone to having the condition or disorder, or those in whom the condition or disorder is to be prevented.

[0038] "Subject" or "individual" or "animal" or "patient" or "mammal" means any subject for which diagnosis, prognosis, or treatment is desired, particularly a mammalian subject. Mammalian subjects include humans, domestic, farm, and zoo animals, sport or pet animals, such as dogs, cats, guinea pigs, rabbits, rats, mice, horses, cattle, dairy cows, and the like.

[0039] As used herein, phrases such as "in a patient in need of treatment" or "subject in need of treatment" include subjects, e.g., mammalian subjects, who would benefit from the administration of an antibody or composition of the disclosure, e.g., for use in detection, diagnostic procedures and / or treatment. multifunctional molecules

[0040] As demonstrated in the accompanying experimental examples, the inventors have been able to identify several bifunctional fusion proteins comprising an anti-PD-L1 unit and a TGF-β targeting unit. As shown in Example 14, for example, both tested bifunctional proteins, LP008-02 and LP008-06a-ES, showed greater efficacy than M7824 in the MC38 mouse model. M7824 is a PD-L1 / TGF-β dual targeting fusion protein currently in Phase II clinical trials for patients with HPV-positive malignancies. The anti-PD-L1 unit of M7824 is based on avelumab, a leading PD-L1 antibody, approved for the treatment of Merkel cell carcinoma and urothelial carcinoma. Therefore, the superior performance of the newly disclosed bifunctional protein compared to M7824 is surprising.

[0041] Furthermore, the presently disclosed bifunctional proteins have better species specificity, as shown in Example 12. Unlike M7824, which also reacts with mouse and rat PD-L1, the new bifunctional proteins bind only to human and cynomolgus PD-L1 in addition to their superior PD-L1 binding activity.

[0042] Thus, in one embodiment, the present disclosure provides a multifunctional molecule having at least an anti-PD-L1 unit and a TGF-β targeting unit. The anti-PD-L1 unit is , may comprise an anti-PD-L1 antibody or fragment of the present disclosure. The TGF-β targeting unit is preferably the extracellular domain of human transforming growth factor-β (TGF-β) receptor II (TGF-βRII or TGFBR2).

[0043] TGF-βRII has two isoforms. Isoform A (NP_001020018.1; SEQ ID NO: 70) has a longer extracellular fragment than isoform B (NP_003233.4; SEQ ID NO: 71), but they share the same core ectodomain (SEQ ID NO: 72). Their sequences are shown in Table A below. Table A. Sequences related to TGF-βRII (underlined and bold: core ectodomain; underlined and italic: residues that differ between isoforms; underline only: mutations) [Table A-1] [Table A-2]

[0044] In some embodiments, the TGF-βRII extracellular domain comprises the core ectodomain (SEQ ID NO: 72) as well as several flanking residues. For example, variant 1 (SEQ ID NO: 61), tested in Examples 8-16, contains an additional 25 residues on the N-terminus and 9 residues on the C-terminus. Another variant, variant 2 (SEQ ID NO: 73), contains only the 9 C-terminal flanking residues. Other variants, such as variants 4-7 (SEQ ID NOs: 75-78), contain alternative linkers that replace part of the N-terminal sequence of SEQ ID NO: 61.

[0045] In some embodiments, the TGF-βRII extracellular domain does not include the first 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, or 25 amino acids of SEQ ID NO: 61. In some embodiments, the TGF-βRII extracellular domain does not include the last 1, 2, 3, 4, 5, 6, 7, 8, or 9 amino acids of SEQ ID NO:61.

[0046] Yet another variant, variant 3, is based on variant 1 but contains at least an amino acid substitution at the X position in the N-terminal portion (SEQ ID NO: 88). These X positions are potential glycosylation sites. Thus, the substitutions are with amino acids other than K, S and N. Examples of substitutions include, but are not limited to, R, A, G, Q, I, L, D or E.

[0047] In some embodiments, the ani-PD-L1 unit is composed of an anti-PD-L1 antibody or fragment thereof, as described further below. The antibody or fragment may be in any antibody format, including but not limited to a traditional intact IgG format, a Fab fragment, a single chain fragment, or a single domain antibody. If the antibody or fragment thereof has a light chain and a separate heavy chain, the TGF-βRII extracellular domain may be fused to either the light chain or the heavy chain. If the antibody or fragment thereof has light and heavy chains on a single protein chain (e.g., scFv), the TGF-βRII extracellular domain may be fused more proximally to either the light chain or the heavy chain.

[0048] In some embodiments, the TGF-βRII extracellular domain is fused to the N-terminus of a chain of an anti-PD-L1 unit. In some embodiments, the TGF-βRII extracellular domain is fused to the C-terminus of a chain of an anti-PD-L1 unit. In preferred embodiments, the TGF-βRII extracellular domain is fused to the C-terminus of a heavy chain of an anti-PD-L1 unit, optionally via a peptide linker (e.g., SEQ ID NO: 60, or one, two, or three GGGGS (SEQ ID NO: 86) repeats).

[0049] In some embodiments, the anti-PD-L1 unit comprises a VH (heavy chain variable region) and a VL (light chain variable region). The VH and VL regions include VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3, for example, those shown in Tables 1A-1C.

[0050] In one embodiment, VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences of SDYAWN (SEQ ID NO: 7), YIIYSGSTSYNPSLKS (SEQ ID NO: 8), STMIATNWFAY (SEQ ID NO: 9), KASQDVSLAVA (SEQ ID NO: 10), WASTRHT (SEQ ID NO: 11) and QQHYITPWT (SEQ ID NO: 12), respectively. Examples of such VH sequences are provided in SEQ ID NOs: 25 (mouse) and 26-28 (humanized). Examples of such VL sequences are provided in SEQ ID NOs: 29 (mouse) and 30 (humanized). Exemplary humanized antibodies include those having a VH of SEQ ID NO: 26 or 27 or 28 and a VL of SEQ ID NO: 30.

[0051] In one embodiment, VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences of DFWVS (SEQ ID NO: 13), EIYPNSGVSRYNEKFKG (SEQ ID NO: 14), YFGYTYWFGY (SEQ ID NO: 15), RASKSVSTYMH (SEQ ID NO: 16), SASHLES (SEQ ID NO: 17) and QQSNELPVT (SEQ ID NO: 18), respectively. Examples of such VH sequences are provided in SEQ ID NOs: 31 (mouse) and 32-37 (humanized). Examples of such VL sequences are provided in SEQ ID NOs: 38 (mouse) and 39-43 (humanized). Exemplary humanized antibodies include those having a VH of SEQ ID NO: 34 and a VL of SEQ ID NO: 39, 40 or 43, those having a VH of SEQ ID NO: 35 and a VL of SEQ ID NO: 39, or those having a VH of SEQ ID NO: 37 and a VL of SEQ ID NO: 39. In one embodiment, the humanized antibody comprises a VH of SEQ ID NO:34 and a VL of SEQ ID NO:43.

[0052] In one embodiment, VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPDSVKG (SEQ ID NO: 20), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYNSGNT (SEQ ID NO: 24), respectively. Alternatively, VH CDR2 may comprise SITNTGSTFYPDAVKG (SEQ ID NO: 91) or SITNTGSTFYPESVKG (SEQ ID NO: 92). Alternatively, VL CDR3 may be SQYQSGNT (SEQ ID NO: 93).

[0053] In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPDSVKG (SEQ ID NO: 20), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYNSGNT (SEQ ID NO: 24), respectively. In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPDAVKG (SEQ ID NO: 91), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYNSGNT (SEQ ID NO: 24), respectively. In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPESVKG (SEQ ID NO: 92), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYNSGNT (SEQ ID NO: 24), respectively.

[0054] In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPDSVKG (SEQ ID NO: 20), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYQSGNT (SEQ ID NO: 93), respectively. In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPDAVKG (SEQ ID NO: 91), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYQSGNT (SEQ ID NO: 93), respectively. In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPESVKG (SEQ ID NO: 92), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYQSGNT (SEQ ID NO: 93), respectively.

[0055] Examples of such VH sequences are provided in SEQ ID NOs: 44 (murine), 45-49 (humanized), and 57-58 (humanized). Examples of such VL sequences are provided in SEQ ID NOs: 50 (murine), 51-55 (humanized), and 56 (humanized).

[0056] Exemplary humanized antibodies include those having a VH of SEQ ID NO: 49 and a VL of SEQ ID NO: 52 or 54, or those having a VH of SEQ ID NO: 48 and a VL of SEQ ID NO: 53 or 54. In one embodiment, the humanized antibody comprises a VH of SEQ ID NO: 48 and a VL of SEQ ID NO: 53. In one embodiment, the humanized antibody comprises a VH of SEQ ID NO: 48 and a VL of SEQ ID NO: 56. In one embodiment, the humanized antibody comprises a VH of SEQ ID NO: 57 and a VL of SEQ ID NO: 56. In one embodiment, the humanized antibody comprises a VH of SEQ ID NO: 58 and a VL of SEQ ID NO: 56.

[0057] In some embodiments, the antibody or fragment thereof further comprises a heavy chain constant region (e.g., CH1, CH2 and / or CH3) and / or a light chain constant region (e.g., CL). An exemplary heavy chain constant region is provided in SEQ ID NO:59, and an exemplary light chain constant region is provided in SEQ ID NO:67 (residues 108-214). TGF-βRII x antibody fusion

[0058] Tests with different fusion protein designs (e.g., Table 15) demonstrated that only the core ectodomain of TGF-βRII (SEQ ID NO: 72) is required for activity. Furthermore, the ectodomain of TGF-βRII should not be fused directly to the antibody. There should be sufficient distance provided by the peptide linker.

[0059] With respect to the ectodomain, the peptide linker (which may be a completely artificial linker or may include a portion of the extracellular fragment N-terminal to the ectodomain, SEQ ID NO: 89) should have a minimum length. If the distance is too short, the stability or activity of the fusion protein decreases. The minimum length is, in some embodiments, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, or 40 amino acid residues (residents). In some embodiments, the linker is no more than 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 110, 120, 130, 140, 150, 170 or 200 amino acid residues.

[0060] The inclusion of a flexible linker, such as one or more G4S (SEQ ID NO: 86) units, may be useful in some embodiments for the stability and / or activity of a multifunctional molecule. In some embodiments, the flexible linker comprises at least 40%, 50%, 60%, 70% or 80% glycine. In some embodiments, the flexible linker comprises one or more serine. In some embodiments, the flexible linker comprises 1, 2, 3, 4, 5 or 6 G4S (SEQ ID NO: 86) repeats.

[0061] (e.g., Example 17), in some embodiments, it has been shown that the native N-terminal fragment (IPPHVQKSVNNDMIVTDNNGAVKFP; SEQ ID NO:89) can be replaced with a replacement peptide to increase stability without sacrifice or improve activity. In some embodiments, the replacement peptide differs from SEQ ID NO:89 but has at least 30%, 40%, 50%, 60%, 70%, 80% or 90% sequence identity to SEQ ID NO:89.

[0062] An exemplary replacement peptide is IPPHVQXXVNNDMIVTDNXGAVKFP (SEQ ID NO: 88), where X is any amino acid except K, S, or N. In some embodiments, substitutions can be made to remove the rigid dipeptide PP, remove potential cleavage sites QK, N, and / or K, include multiple glycine residues for increased flexibility, and / or reduce hydrophobic residues. One such example is TAGHTQTSTGGGAITTGTSGAGHGP (SEQ ID NO: 87) or a variant having at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95% sequence identity to SEQ ID NO: 87. In some embodiments, the variant includes at least four Gs, no PP dipeptide, and includes no more than three hydrophobic amino acid residues selected from the group consisting of I, L, M, F, V, W, Y, and P. In some embodiments, the variant comprises at least five G and no more than one hydrophobic amino acid residue selected from the group consisting of I, L, M, F, V, W, Y and P.

[0063] In some embodiments, the peptide linker between the antibody or fragment thereof and the ectodomain of TGF-βRII (SEQ ID NO: 72) comprises a flexible linker. In some embodiments, the peptide linker comprises a replacement peptide of SEQ ID NO: 89. In some embodiments, the peptide linker comprises both a flexible linker and a replacement peptide. In some embodiments, the flexible linker is at the N-terminus of the replacement peptide. In some embodiments, the flexible linker is at the C-terminus of the replacement peptide.

[0064] In some embodiments, the multifunctional molecule does not include at least the entire sequence of EEYNTSNPD (SEQ ID NO: 90). The multifunctional molecule may have the entire sequence of SEQ ID NO: 90 removed from the extracellular domain of TGF-βRII. In some embodiments, the multifunctional molecule does not include more than 1, 2, 3, 4, 5, 6, 7, or 8 amino acid residues of EEYNTSNPD (SEQ ID NO: 90).

[0065] The antibody or antigen-binding fragment of the multifunctional molecule can target any antigen. Non-limiting examples are PD-1, PD-L1, CTLA-4, LAG-3, CD28, CD122, 4-1BB, TIM3, OX-40, OX40L, CD40, CD40L, LIGHT, ICOS, ICOSL, GITR, GITRL, TIGIT, CD27, VISTA, B7H3, B7H4, BTLA, CD4, CD2, CD8, CD47 and CD73. They can also be any antibody or fragment as disclosed herein.

[0066] The ectodomain of TGF-βRII can be fused to any part of the antibody or fragment.In some embodiments, the ectodomain is fused to the C-terminus of the heavy or light chain of the antibody or fragment.In some embodiments, the ectodomain is fused to the C-terminus of the Fc fragment of the antibody or fragment. Anti-PD-L1 antibodies and fragments

[0067] Anti-PD-L1 antibodies and fragments are also provided and may be used as anti-PD-L1 entities in multifunctional molecules, bispecific or multispecific antibodies, or alone in monospecific antibodies.

[0068] Exemplary murine anti-PD-L1 antibodies, as well as their humanized and improved antibodies, have been prepared and tested in the accompanying experimental examples. All murine antibodies (47C6A3, 67F3G7, 89C10H8) and their corresponding humanized versions showed excellent binding affinity, cross-reactivity and efficacy in inhibiting PD-1 / PD-L1 binding.

[0069] Importantly, when compared with MPDL3280A (atezolizumab), humanized 67F3G7 and 89C10H8 showed higher activity than MPDL3280A in blocking the interaction between PD-1 and PD-L1 (see, for example, Example 18). Also, interestingly, all of the tested antibodies of the present disclosure showed lower hydrophobicity and lower viscosity than MPDL3280A. Higher hydrophobicity is known to reduce protein solubility. Similarly, high viscosity is also an obstacle to the development of high-concentration protein formulations. Therefore, such data demonstrate that the antibody of the present invention is more suitable for the preparation of high-concentration antibody formulations.

[0070] Furthermore, an antigen-binding fragment of the presently disclosed antibody was included as a unit in a bifunctional fusion protein that also contained a TGF-β targeting unit. The resulting bifunctional fusion protein showed higher efficacy than M7824 in the MC38 mouse model. M7824 is a PD-L1 / TGF-β dual targeting fusion protein currently in Phase II clinical trials for patients with HPV-positive malignancies. The anti-PD-L1 unit of M7824 is based on avelumab, a leading PD-L1 antibody approved for the treatment of Merkel cell carcinoma and urothelial carcinoma. Thus, these data demonstrate the unique advantages of the presently disclosed antibodies in preparing bifunctional or multifunctional molecules.

[0071] In some embodiments, the anti-PD-L1 antibody or fragment comprises a VH (heavy chain variable region) and a VL (light chain variable region). The VH and VL regions include VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3, including those shown in Tables 1A-1C.

[0072] In one embodiment, VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences of SDYAWN (SEQ ID NO: 7), YIIYSGSTSYNPSLKS (SEQ ID NO: 8), STMIATNWFAY (SEQ ID NO: 9), KASQDVSLAVA (SEQ ID NO: 10), WASTRHT (SEQ ID NO: 11) and QQHYITPWT (SEQ ID NO: 12), respectively. Examples of such VH sequences are provided in SEQ ID NOs: 25 (mouse) and 26-28 (humanized). Examples of such VL sequences are provided in SEQ ID NOs: 29 (mouse) and 30 (humanized). Exemplary humanized antibodies include those having a VH of SEQ ID NO: 26 or 27 or 28 and a VL of SEQ ID NO: 30.

[0073] In one embodiment, VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences of DFWVS (SEQ ID NO: 13), EIYPNSGVSRYNEKFKG (SEQ ID NO: 14), YFGYTYWFGY (SEQ ID NO: 15), RASKSVSTYMH (SEQ ID NO: 16), SASHLES (SEQ ID NO: 17) and QQSNELPVT (SEQ ID NO: 18), respectively. Examples of such VH sequences are provided in SEQ ID NOs: 31 (mouse) and 32-37 (humanized). Examples of such VL sequences are provided in SEQ ID NOs: 38 (mouse) and 39-43 (humanized). Exemplary humanized antibodies include those having a VH of SEQ ID NO: 34 and a VL of SEQ ID NO: 39, 40 or 43, those having a VH of SEQ ID NO: 35 and a VL of SEQ ID NO: 39, or those having a VH of SEQ ID NO: 37 and a VL of SEQ ID NO: 39. In one embodiment, the humanized antibody comprises a VH of SEQ ID NO:34 and a VL of SEQ ID NO:43.

[0074] In one embodiment, VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPDSVKG (SEQ ID NO: 20), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYNSGNT (SEQ ID NO: 24), respectively. Alternatively, VH CDR2 can comprise SITNTGSTFYPDAVKG (SEQ ID NO: 91) or SITNTGSTFYPESVKG (SEQ ID NO: 92). Alternatively, VL CDR3 can be SQYQSGNT (SEQ ID NO: 93).

[0075] In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPDSVKG (SEQ ID NO: 20), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYNSGNT (SEQ ID NO: 24), respectively. In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPDAVKG (SEQ ID NO: 91), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYNSGNT (SEQ ID NO: 24), respectively. In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPESVKG (SEQ ID NO: 92), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYNSGNT (SEQ ID NO: 24), respectively.

[0076] In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPDSVKG (SEQ ID NO: 20), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYQSGNT (SEQ ID NO: 93), respectively. In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPDAVKG (SEQ ID NO: 91), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYQSGNT (SEQ ID NO: 93), respectively. In one embodiment, the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the sequences NYWMT (SEQ ID NO: 19), SITNTGSTFYPESVKG (SEQ ID NO: 92), DTTIAPFDY (SEQ ID NO: 21), KASQNLNEYLN (SEQ ID NO: 22), KTNTLQA (SEQ ID NO: 23) and SQYQSGNT (SEQ ID NO: 93), respectively.

[0077] Examples of such VH sequences are provided in SEQ ID NOs: 44 (murine), 45-49 (humanized), and 57-58 (humanized). Examples of such VL sequences are provided in SEQ ID NOs: 50 (murine), 51-55 (humanized), and 56 (humanized).

[0078] Exemplary humanized antibodies include those having a VH of SEQ ID NO: 49 and a VL of SEQ ID NO: 52 or 54, or those having a VH of SEQ ID NO: 48 and a VL of SEQ ID NO: 53 or 54. In one embodiment, the humanized antibody comprises a VH of SEQ ID NO: 48 and a VL of SEQ ID NO: 53. In one embodiment, the humanized antibody comprises a VH of SEQ ID NO: 48 and a VL of SEQ ID NO: 56. In one embodiment, the humanized antibody comprises a VH of SEQ ID NO: 57 and a VL of SEQ ID NO: 56. In one embodiment, the humanized antibody comprises a VH of SEQ ID NO: 58 and a VL of SEQ ID NO: 56.

[0079] In some embodiments, the antibody or fragment thereof further comprises a heavy chain constant region (e.g., CH1, CH2 and / or CH3) and / or a light chain constant region (e.g., CL). An exemplary heavy chain constant region is provided in SEQ ID NO:59, and an exemplary light chain constant region is provided in SEQ ID NO:67 (residues 108-214).

[0080] It is envisioned that small changes (e.g., addition, deletion, or substitution of one amino acid) can be designed between these CDR sequences that can retain or even improve the activity of the antibody. Such modified CDR sequences are called CDR variants. It is also understood by those skilled in the art that the antibodies disclosed herein can be modified to have different amino acid sequences from the naturally occurring binding polypeptides from which they are derived. For example, the polypeptide or amino acid sequence derived from a designated protein can be similar, e.g., have a certain percent identity with the starting sequence, e.g., 60%, 70%, 75%, 80%, 85%, 90%, 95%, 98% or 99% identical to the starting sequence. In some embodiments, the modified antibody or fragment retains the designated CDR sequence.

[0081] In certain embodiments, the antibody comprises an amino acid sequence or one or more moieties that are not normally associated with antibodies. Exemplary variations are described in more detail below. For example, the antibody of the present disclosure may comprise a flexible linker sequence or may be modified to add a functional moiety (e.g., PEG, a drug, a toxin, or a label). Polynucleotides encoding proteins and methods for preparing proteins

[0082] The present disclosure also provides an isolated polynucleotide or nucleic acid molecule encoding the multifunctional protein, antibody, variant or derivative thereof of the present disclosure. The polynucleotide of the present disclosure may encode the entire heavy and light chain variable regions of the antigen-binding polypeptide, variant or derivative thereof on the same polynucleotide molecule or on separate polynucleotide molecules. Furthermore, the polynucleotide of the present disclosure may encode a portion of the heavy and light chain variable regions of the antigen-binding polypeptide, variant or derivative thereof on the same polynucleotide molecule or on separate polynucleotide molecules.

[0083] Methods for producing antibodies are well known in the art and described herein. In certain embodiments, both the variable and constant regions of the antigen-binding polypeptide of the present disclosure are fully human. Fully human antibodies can be produced using techniques described in the art and described herein. For example, fully human antibodies against a particular antigen can be prepared by administering the antigen to a transgenic animal whose endogenous locus has been disabled, but which has been modified to produce such antibodies in response to antigen challenge. Exemplary techniques that can be used to produce such antibodies are described in U.S. Patent Nos. 6,150,584, 6,458,592, and 6,420,140, ​​which are incorporated by reference in their entirety. Cancer Treatment

[0084] As described herein, the antibodies, variants or derivatives of the present disclosure may be used in certain treatment and diagnostic methods.

[0085] The present disclosure is further directed to multifunctional molecule or antibody-based therapies that involve administering the multifunctional molecules or antibodies of the present disclosure to patients (e.g., animals, mammals, and humans) to treat one or more disorders or conditions described herein. Therapeutic compounds of the present disclosure include, but are not limited to, the antibodies of the present disclosure (including variants and derivatives thereof as described herein) and nucleic acids or polynucleotides encoding the antibodies of the present disclosure (including variants and derivatives thereof as described herein).

[0086] The antibodies of the present disclosure can also be used to treat or inhibit cancer. PD-L1 can be overexpressed in tumor cells. Tumor-derived PD-L1 can bind to PD-1 on immune cells, thereby limiting anti-tumor T cell immunity. Results with small molecule inhibitors or monoclonal antibodies targeting PD-L1 in mouse tumor models indicate that targeted PD-L1 therapy is an important alternative and practical approach to effectively control tumor growth. As demonstrated in experimental examples, anti-PD-L1 antibodies activate adaptive immune response mechanisms, which can lead to improved survival rates for cancer patients.

[0087] Thus, in some embodiments, a method of treating cancer in a patient in need thereof is provided. The method, in one embodiment, involves administering to the patient an effective amount of a multifunctional molecule or antibody of the present disclosure. In some embodiments, at least one cancer cell (e.g., a stromal cell) of the patient expresses, overexpresses, or is induced to express PD-L1. Induction of PD-L1 expression can be achieved, for example, by administration of a tumor vaccine or radiation therapy.

[0088] Tumors that express PD-L1 protein include bladder cancer, non-small cell lung cancer, kidney cancer, breast cancer, urethral cancer, colorectal cancer, head and neck cancer, squamous cell carcinoma, Merkel cell carcinoma, gastrointestinal cancer, gastric cancer, esophageal cancer, ovarian cancer, renal cancer, and small cell lung cancer. Thus, the antibodies of the present disclosure can be used to treat any one or more of such cancers.

[0089] Cell therapy, such as chimeric antigen receptor (CAR) T cell therapy, is also provided in the present disclosure. Appropriate cells can be used that are contacted with the anti-PD-L1 antibody of the present disclosure (or alternatively engineered to express the anti-PD-L1 antibody of the present disclosure). Once such contacted or engineered, the cells can then be introduced into a cancer patient in need of treatment. The cancer patient can have any of the types of cancer disclosed herein. The cells (e.g., T cells) can be, for example, but are not limited to, tumor-infiltrating T lymphocytes, CD4+ T cells, CD8+ T cells, or a combination thereof.

[0090] In some embodiments, the cells are isolated from the cancer patient themselves. In some embodiments, the cells are provided by a donor or from a cell bank. When the cells are isolated from the cancer patient, unwanted immune responses can be minimized.

[0091] Additional diseases or conditions associated with increased cell survival that may be treated, prevented, diagnosed and / or prognosed using the disclosed antibodies or variants, or derivatives thereof include leukemias (including acute leukemias (e.g., acute lymphocytic leukemia, acute myelocytic leukemia (including myeloblastic leukemia, promyelocytic leukemia, myelomonocytic leukemia, monocytic leukemia, erythroleukemia) and chronic leukemias (e.g., chronic myelocytic (granulocytic) leukemia and chronic lymphocytic leukemia)), polycythemia vera, lymphomas (e.g., Hodgkin's disease and non-Hodgkin's disease), multiple myeloma, Waldenstrom's macroglobulinemia, heavy chain disease, and fibrosarcoma, myxosarcoma, liposarcoma, chondrosarcoma, osteogenic sarcoma, chordoma, angiosarcoma, endothelial sarcoma, lymphangiosarcoma, lymphatic ... These include, but are not limited to, progression and / or metastasis of malignant tumors and related disorders, such as solid tumors, including, but not limited to, sarcomas and carcinomas such as sarcoma, synovium, mesothelioma, Ewing's tumor, leiomyosarcoma, rhabdomyosarcoma, colon cancer, pancreatic cancer, breast cancer, thyroid cancer, endometrial cancer, melanoma, prostate cancer, ovarian cancer, prostate cancer, squamous cell, basal cell carcinoma, adenocarcinoma, sweat gland carcinoma, sebaceous gland carcinoma, papillary carcinoma, papillary adenocarcinoma, cystadenocarcinoma, medullary carcinoma, bronchial carcinoma, renal cell carcinoma, hepatoma, cholangiocarcinoma, choriocarcinoma, seminoma, embryonal carcinoma, Wilms' tumor, cervical cancer, testicular tumor, lung cancer, small cell lung carcinoma, bladder carcinoma, epithelial carcinoma, glioma, astrocytoma, medulloblastoma, craniopharyngioma, ependymoma, pinealoma, hemangioblastoma, acoustic neuroma, oligodendroglioma, meningioma, melanoma, neuroblastoma, retinoblastoma, etc. composition

[0092] The present disclosure also provides pharmaceutical compositions. Such compositions comprise an effective amount of antibody and acceptable carrier. In some embodiments, the composition further comprises a second anticancer agent (e.g., immune checkpoint inhibitor).

[0093] In certain embodiments, the term "pharmaceutically acceptable" means approved by a federal or state regulatory agency or listed in the United States Pharmacopeia or other generally recognized pharmacopoeias for use in animals, and more particularly in humans. Moreover, a "pharmaceutically acceptable carrier" is generally a non-toxic solid, semi-solid or liquid filler, diluent, encapsulating material or formulation auxiliary of any type.

[0094] The term "carrier" refers to a diluent, adjuvant, excipient, or vehicle with which the therapeutic is administered. Such pharmaceutical carriers can be sterile liquids, such as water and oils, such as those of petroleum, animal, vegetable, or synthetic origin, such as peanut oil, soybean oil, mineral oil, sesame oil, and the like. Water is a preferred carrier when the pharmaceutical composition is administered intravenously. Saline solutions and aqueous dextrose and glycerol solutions can also be used 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. The composition can also contain minor amounts of wetting or emulsifying agents, or pH buffering agents, such as acetates, citrates, or phosphates, if desired. Antibacterial agents such as benzyl alcohol, methylparaben, antioxidants such as ascorbic acid, sodium bisulfite, chelating agents such as ethylenediaminetetraacetic acid, and agents for adjusting tonicity such as sodium chloride or dextrose are also contemplated. These compositions can take the form of solutions, suspensions, emulsions, tablets, pills, capsules, powders, sustained release formulations, and the like. The compositions can be formulated as suppositories with conventional binders and carriers such as triglycerides. Oral formulations can include standard carriers such as pharmaceutical grades of mannitol, lactose, starch, magnesium stearate, sodium saccharin, cellulose, magnesium carbonate, and the like. Examples of suitable pharmaceutical carriers are described in Remington's Pharmaceutical Sciences by EW Martin, which is incorporated herein by reference. Such compositions contain a therapeutically effective amount of the antigen-binding polypeptide, preferably in purified form, together with an appropriate amount of carrier to provide the form for proper administration to the patient. The formulation should suit the mode of administration. Parental preparations ) may be enclosed in ampoules, disposable syringes or multiple dose vials made of glass or plastic.

[0095] In one embodiment, the composition is formulated according to routine procedures as a pharmaceutical composition adapted for intravenous administration to humans. Typically, compositions for intravenous administration are solutions in sterile isotonic aqueous buffer. If necessary, the composition may also include a solubilizing agent and a local anesthetic, such as lignocaine, to ease pain at the injection site. Generally, the ingredients are supplied in unit dosage form, either separately or mixed together, for example as a dry lyophilized powder or water-free concentrate in a hermetically sealed container, such as an ampoule or sachette indicating the quantity of active agent. When the composition is administered by infusion, it can be dispensed with an infusion bottle containing sterile pharmaceutical grade water or saline. When the composition is administered by injection, an ampoule of sterile water for injection or saline can be provided so that the ingredients can be mixed prior to administration. EXAMPLES

[0096] Example 1: Generation of mouse monoclonal antibodies against human PD-L1 This example describes the generation of anti-human PD-L1 mouse monoclonal antibodies using hybridoma technology.

[0097] Antigen: human PDL1-Fc protein and human PD-L1 highly expressed on CHOK1 cell line (PDL1-CHOK1 cell line).

[0098] Immunization: To generate mouse monoclonal antibodies targeting human PD-L1, Balb / c mice and Wistar rats were first immunized with PD-L1-Fc protein. The immunized mice and rats were then boosted with PD-L1-Fc protein and CHO-K1 / PD-L1 stable cells, respectively. To select mice or rats producing antibodies that bind to PD-L1 protein, the sera of the immunized mice or rats were subjected to antibody titer assessment by ELISA. Briefly, microtiter plates were coated with 0.5μg / ml human PD-L1 protein in ELISA coating buffer, 100μl / well, overnight at 4℃, and then blocked with 150μl / well 1% BSA. Dilutions of serum from immunized mice were added to each well and incubated at 37℃ for 1-2 hours. Plates were washed with PBS / Tween® and then incubated with anti-mouse IgG antibody conjugated with horseradish peroxidase (HRP) or anti-rat IgG antibody conjugated with HRP for 1 hour at 37°C. After washing, plates were developed with TMB substrate and analyzed by spectrophotometer at OD 450nm. After three immunizations, immune responses were also tested by serum ELISA against rhPD-L1 protein and FACS against CHO-K1 / PDL-1 stable cell line using CHO-K1 parental cell line as negative control. Mice that showed sufficient titers of anti-PDL1 IgG were boosted with 25μg human PDL1-Fc protein after three immunizations. The resulting mice were used for fusions. Hybridoma supernatants were tested for anti-PD-L1 IgG by ELISA.

[0099] Cell fusion: Fusion was performed by electrofusion. The fused cells were plated into 50 96-well plates for each fusion.

[0100] Screening: Supernatants were screened by ELISA against recombinant human (rh)PD-L1-Fc protein and counter-screening antigen. Positive supernatants were then subjected to confirmatory screening from the primary screening by receptor-blocking FACS against CHO-K1 / PD-L1 stable cell line and rhPD-1-Fc protein.

[0101] Subcloning and screening: Positive primary clones from each fusion were subcloned by limiting dilution to ensure that the subclones were derived from a single parent cell. The subclones were screened using the same approach as the primary clones, and culture supernatants of positive clones were subjected to further confirmatory screening by affinity ranking.

[0102] Hybridoma clones 47C6A3, 67F3G7 and 89C10H8 were selected for further analysis. The amino acid sequences of the variable regions of 47C6A3, 67F3G7 and 89C10H8 are listed in Table 1 below. Table 1. Variable region sequences of 47C6A3, 67F3G7 and 89C10H8 [Table 1] Table 1A CDR sequence of 47C6A3 [Table 1A] Table 1B CDR sequence of 67F3G7 [Table 1B] Table 1C CDR sequence of 89C10H8 [Table 1C] Example 2: Binding activity to PD-L1 antigen ELISA Testing

[0103] To evaluate the binding activity of the hybridoma clones 47C6A3, 67F3G7 and 89C10H8, the chimeric mAbs derived from these clones were subjected to ELISA tests.

[0104] Briefly, microtiter plates were coated with 0.5μg / ml human PD-L1-Fc protein in PBS, 100μl / well overnight at 4℃, then blocked with 150μl / well 1% BSA. Three-fold dilutions of 47C6A3, 67F3G7 and 89C10H8 antibodies starting at 10μg / ml were added to each well and incubated for 1 hour at 37℃. Plates were washed with PBS / Tween® and then blocked with mouse anti-human IgG conjugated with horseradish peroxidase (HRP). The Fab antibodies were incubated for 30 minutes at 37°C. After washing, the plates were developed with TMB substrate and analyzed by spectrophotometer at OD 450 nm. As shown in Figure 1, 47C6A3, 67F3G7 and 89C10H8 bound to human PD-L1 with high affinity (EC 50 = 10.24ng / ml, EC for 67F3G7 50 = 10.76ng / ml, EC for 89C10H8 50 =8.112ng / ml).

[0105] Cell-based binding: FACS was used to assess the binding activity of 47C6A3, 67F3G7 and 89C10H8 chimeric mAbs to human PD-L1-overexpressing CHOK1 cells.

[0106] Briefly, PDL1-CHOK1 cells were first incubated with 3-fold serial dilutions of 47C6A3, 67F3G7 and 89C10H8 chimeric mAbs starting at 100 nM for 40 min at 4°C. After washing with PBS, Alexa Fluor® 647 AffiniPure goat anti-human IgG (H+L) was added to each well and incubated for 30 min at 4°C. Samples were washed twice with FACS buffer. The mean fluorescence intensity (MFI) of Alexa Fluor® 647 was assessed on a FACSCanto. As shown in Figure 2, 47C6A3, 67F3G7 and 89C10H8 bound with high affinity to PDL1-CHOK1 cells (EC for 47C6A3). 50 = 0.1476 nM, EC for 67F3G7 50 = 0.1035nM, EC for 89C10H8 50 =0.1696nM). Interspecies activity

[0107] ELISA studies were performed to evaluate the binding of the chimeric antibodies to human, mouse, rat and cynomolgus monkey PD-L1, respectively.

[0108] Briefly, microtiter plates were coated with 100 μl / well of human, mouse, rat and cynomolgus PD-L1 proteins at 0.5 μg / ml in PBS overnight at 4° C., then blocked with 150 μl / well of 1% BSA. Three-fold dilutions of chimeric antibodies starting at 10 μg / ml were added to each well and incubated at 37° C. for 1 hour. Plates were washed with PBS / Tween®, then incubated with mouse anti-human IgG Fab antibody conjugated with horseradish peroxidase (HRP) for 30 minutes at 37° C. After washing, plates were developed with TMB substrate and OD Analyzed by spectrophotometry at 450 nm, the 47C6A3, 67F3G7 and 89C10H8 antibodies bound to human and cynomolgus monkey PD-L1, but not to rat or mouse PD-L1 (Figure 3 and Table 2). Table 2 Interspecies activity of 47C6A3, 67F3G7 and 89C10H8 [Table 2] Example 3: Blockade of PD-L1 binding to PD-1 by antibodies

[0109] To evaluate the blocking effects of 47C6A3, 67F3G7 and 89C10H8 chimeric mAbs on the binding of recombinant human PD-L1 to its receptor PD-1, an ELISA-based receptor blocking assay was used.

[0110] Briefly, microtiter plates were coated with 0.5μg / ml human PD-L1-Fc protein in PBS, 100μl / well, overnight at 4℃, then blocked with 150μl / well 1% BSA. 50μl of biotin-labeled human PD-1-Fc protein and 47C6A3, 67F3G7 and 89C10H8 antibodies, 3-fold diluted in 50μl starting from 10μg / ml, were added to each well and incubated at 37℃ for 1 hour. Plates were washed with PBS / Tween® and then incubated with streptavidin-HRP for 10 minutes at 37℃. After washing, plates were developed with TMB substrate and analyzed by spectrophotometer at OD 450nm. As shown in Figure 4, 47C6A3, 67F3G7 and 89C10H8 were detected at IC of 91.18ng / ml, 139.8ng / ml and 129.8ng / ml, respectively. 50 efficiently inhibited the binding of human PD-L1 to human PD1. Example 4: Binding affinity of mAbs

[0111] The binding of 47C6A3, 67F3G7 and 89C10H8 antibodies to recombinant PD-L1 protein (human PD-L1-his tag) was tested on Biacore using the capture method. 47C6A3, 67F3G7 and 89C10H8 mAbs were captured using a Protein A chip. Serial dilutions of human PD-L1-his tag protein were injected over the capture antibody at a flow rate of 30μl / min for 2 minutes. The antigen was allowed to dissociate for 480-1500 seconds. All experiments were performed on a Biacore T200. Data analysis was performed using the Biacore T200 Evaluation software. The results are shown in Figure 5 and Table 3 below. Table 3. Affinities measured by Biacore [Table 3] Example 5. Humanization of a mouse antibody

[0112] The 47C6A3, 67F3G7 and 89C10H8 variable region genes were used to generate humanized mAbs. In the first step of this process, the amino acid sequences of VH and VL or VK of 47C6A3, 67F3G7 and 89C10H8 were compared to the available database of human Ig gene sequences to find the overall best match human germline Ig gene sequence. For the light chain of 47C6A3, human Vk1-4 was the best match germline, and for the heavy chain, human VH1-2 was selected as the backbone. For the light chain of 67F3G7, the closest human match is the Vk1-39 / JK4 gene, and for the heavy chain, the closest human match is the VH1-2 / JH4-FW4 gene. For the light chain of 89C10H8 the closest human match is the Vk1-17 / JK2 gene, and for the heavy chain the closest human match is the VH3-21 / JH3 gene.

[0113] For the VL of 47C6A3, human Vk1-4 was the best-matched germline, and for the VH of 47C6A3, human VH1-2 was selected as the backbone.Then, the humanized variable domain sequence of 47C6A3 was designed, and CDRL1, L2 and L3 were grafted into the framework sequence of Vk1-4 gene, and CDRH1, H2 and H3 were grafted into the framework sequence of VH1-2 gene.Then, a 3D model was generated to determine whether there were any framework positions where replacing mouse amino acids with human amino acids could affect binding and / or CDR conformation.For the heavy chain, R, M and I in the framework were involved in back mutation.

[0114] Then, a humanized variable domain sequence of 67F3G7 was designed, and CDRL1, L2 and L3 were grafted into the framework sequence of Vk1-39 / JK4 gene, and CDRH1, H2 and H3 were grafted into the framework sequence of VH1-2 / JH4-FW4 gene. A 3D model was then generated to determine whether there were framework positions where replacing mouse amino acids with human amino acids could affect binding and / or CDR conformation. For the heavy chain, V, K, T and I in the framework were involved in back mutations. For the light chain, T, V, L and Q in the framework were involved in back mutations.

[0115] Then, a humanized variable domain sequence of 89C10H8 was designed, and CDRL1, L2 and L3 were grafted into the framework sequence of Vk1-17 / JK2 gene, and CDRH1, H2 and H3 were grafted into the framework sequence of VH3-21 / JH3 gene. A 3D model was then generated to determine whether there were framework positions where replacing mouse amino acids with human amino acids could affect binding and / or CDR conformation. For the heavy chain, A, T, I and S in the framework were involved in back mutations. For the light chain, Y, I, E and F in the framework were involved in back mutations.

[0116] The amino acid and nucleotide sequences of several humanized antibodies are listed in Table 4 below. Table 4. Humanized antibody sequences (underlined indicates CDRs, bold / italics indicates backmutations) [Table 4-1] [Table 4-2]

[0117] The genes were cloned into pcDNA 3.4 vector and transfected into 293F cells. Antibodies were produced according to the following table.

[0118] Humanized VH and VL genes were synthetically generated and then cloned into vectors containing the human gamma 1 and human kappa constant domains, respectively. Pairing of human VH and human VL generated 41 humanized antibodies (see Table 5). Table 5. Humanized antibodies with VH and VL regions [Table 5] Example 6: Antigen-binding properties of humanized antibodies Binding to recombinant human PD-L1

[0119] To evaluate the antigen-binding activity, the humanized antibodies were subjected to ELISA test. Briefly, microtiter plates were coated with 0.5μg / ml human PD-L1-Fc protein in PBS, 100μl / well, overnight at 4℃, then blocked with 200μl / well 1% BSA. Three-fold dilutions of humanized antibodies starting from 10μg / ml were added to each well and incubated at 37℃ for 1 hour. The plates were washed with PBS / Tween®, and then incubated with mouse anti-human IgG Fab antibody conjugated with horseradish peroxidase (HRP) for 1 hour at 37℃. After washing, the plates were developed with TMB substrate and analyzed by spectrophotometer at OD 450nm. As shown in Figure 6, all humanized antibodies showed similar binding efficiency to human PD-L1 as chimeric antibodies.

[0120] To investigate the binding kinetics of the humanized antibodies, affinity ranking was performed using Biacore in this example. As shown in Table 6, Hu67F3G7-2, Hu67F3G7-3, Hu67F3G7-5, Hu67F3G7-7, Hu67F3G7-22, Hu89C10H8-4, Hu89C10H8-7, Hu89C10H8-11, and Hu89C10H8-12 showed high affinity comparable to that of the chimeric antibody. Table 6. Affinity ranking of humanized antibodies [Table 6] Binding to human PD-L1 overexpressed on mammalian cells

[0121] To evaluate the antigen-binding properties, the humanized antibodies were analyzed for their binding to PD-L1 overexpressed on mammalian cells by FACS. Briefly, PDL1-CHOK1 cells were first incubated with 3-fold serious diluted humanized antibodies starting at 15μg / ml for 40 minutes at 4℃. After washing with PBS, Alexa Fluor® 647 AffiniPure goat anti-human IgG (H+L) antibody was added to each well and incubated for 30 minutes at 4℃. The MFI of Alexa Fluor® 647 was evaluated by FACSCanto. As shown in Figure 7, all humanized antibodies can bind to PD-L1 expressed on mammalian cells with high efficiency. Full kinetic affinity of humanized antibodies by Biacore

[0122] The binding of humanized antibodies to recombinant PD-L1 protein (human PD-L1-his tag) was tested by Biacore using the capture method. Hu47C6A3-1, Hu47C6A3-2, Hu47C6A3-3, Hu67F3G7-2, Hu67F3G7-3, Hu67F3G7-5, Hu67F3G7-7, Hu67F3G7-22, Hu89C10H8-4, Hu89C10H8-7, Hu89C10H8-11 and Hu89C10H8-12 mAbs were captured using a Protein A chip. Serial dilutions of human PD-L1-his tag protein were injected over the capture antibody at a flow rate of 30μl / min for 2 minutes. The antigen was allowed to dissociate for 1500 seconds. All experiments were performed on a Biacore T200. Data analysis was performed using Biacore T200 evaluation software and the results are shown in Table 7 below. Table 7. Affinity by Biacore [Table 7] Example 7: Blocking PDL1 Binding to PD1 by Humanized Antibodies Receptor blocking assay using recombinant human PD-L1

[0123] Human PD-L1 has two receptors, PD-1 and CD80, and a protein-based receptor blocking assay was used here to examine the blocking properties of humanized PD-L1 antibodies against these two proteins.

[0124] Briefly, microtiter plates were coated with 100μl / well of human PD-L1-Fc protein at 0.5μg / ml in PBS overnight at 4℃, then blocked with 150μl / well of 1% BSA at 37℃ for 2 hours. 50μl of biotin-labeled human PD-1-Fc or CD80-Fc protein and 50μl of 3-fold diluted PD-L1 antibody starting at 10μg / ml were added to each well and incubated at 37℃ for 1 hour. Plates were washed with PBS / Tween® and then incubated with streptavidin-HRP for 10 minutes at 37℃. After washing, plates were developed with TMB substrate and analyzed by spectrophotometer at OD 450nm. As shown in Figure 8, Hu47C6A3-1, Hu47C6A3-2, Hu47C6A3-3, Hu67F3G7-2, Hu67F3G7-3, Hu67F3G7-5, Hu67F3G7-7, Hu67F3G7-22, Hu89C10H8-4, Hu89C10H8-7, Hu89C10H8-11 and Hu89C10H8-12 efficiently inhibited the binding of human PD-L1 to human PD1. Furthermore, Hu47C6A3-1, Hu47C6A3-2, Hu47C6A3-3, Hu67F3G7-2, Hu67F3G7-3, Hu67F3G7-5, Hu67F3G7-7, Hu67F3G7-22, Hu89C10H8-4, Hu89C10H8-7, Hu89C10H8-11 and Hu89C10H8-12 efficiently inhibited human PD-L1 binding to human CD80 in a dose-dependent manner (Figure 9). Example 8. Bifunctional protein targeting in both the PD-L1 and TGF-β pathways

[0125] In this example, a bifunctional recombinant anti-PD-L1 antibody and TGF-βRII fusion protein was prepared and tested.

[0126] The light chain of the molecule is that of an anti-PDL1 mAb. The heavy chain is a fusion of the heavy chain of an anti-PDL1 mAb via a flexible (Gly4Ser)4Gly linker with the N-terminus of the soluble extracellular domain of TGF-βRII. At the fusion junction, the C-terminal lysine residue of the antibody heavy chain was mutated to alanine to reduce potential proteolytic cleavage.

[0127] In some cases, potential modification sites in the CDRs were mutated to similar amino acids. The sequences of the anti-PD-L1 moieties are shown in Table 8 below. Table 8. Variable region sequences of antibody moieties in bifunctional molecules [Table 8-1] [Table 8-2] Table 9. VH / VL of bifunctional molecules [Table 9]

[0128] In addition to the VH, the heavy chain of the bifunctional molecule further comprises a constant region (with the C-terminal K mutated to A), a (Gly4Ser)4Gly linker, and the N-terminus of the soluble extracellular domain of TGF-βRII. Their sequences are shown in Table 10. Table 10. Additional heavy and full heavy / light chain sequences [Table 10-1] [Table 10-2] [Table 10-3] Example 9: Binding Affinity of Bifunctional Molecules

[0129] The binding of LP008-06, LP008-06a, LP008-06a-DA and LP008-06a-ES bifunctional molecules to recombinant PD-L1 protein (human PD-L1-his tag) was tested on Biacore using the capture method.

[0130] The bifunctional molecules were captured using a Protein A chip. Serial dilutions of human PD-L1-his tagged protein were injected over the capture antibody at a flow rate of 30μl / min for 2 minutes. The antigen was allowed to dissociate for 1500 seconds. All experiments were performed on a Biacore T200. Data analysis was performed using the Biacore T200 evaluation software. The results are shown in Figure 10 and Table 11 below. Table 11. Affinity study by Biacore [Table 11]

[0131] The binding of LP008-02 to recombinant PD-L1 protein and human TGF-β1 was tested on Biacore using the capture method.

[0132] LP008-02 was captured using a Protein A chip. Serial dilutions of human PD-L1-his tagged protein and human TGF-β1 were injected over the capture antibody for 2 minutes at a flow rate of 30μl / min. PD-L1 was allowed to dissociate for 680 seconds and TGF-β1 was allowed to dissociate for 1000 seconds. All experiments were performed on a Biacore T200. Data analysis was performed using the Biacore T200 Evaluation software. The results are shown in Figure 11 and Table 12 below. Table 12. Affinity study by Biacore [Table 12] Example 10: Functional assays for PD-1 / PD-L1 blockade

[0133] The activity of the bifunctional molecules in blocking PD1 / PD-L1 interaction was measured in this example by a bioluminescent cell-based assay.

[0134] In this assay, PD1 effector cells are co-cultured with PD-L1 target cells, and the PD-1 / PD-L1 interaction inhibits TCR signaling and NFAT-RE-mediated luminescence. Addition of either anti-PD-1 or anti-PD-L1 antibodies that block the PD-1 / PD-L1 interaction will release an inhibitory signal, resulting in TCR activation and NFAT-RE-mediated luminescence.

[0135] As shown in Figure 12, LP008-02 and LP008-06a-ES blocked PD1 and PD-L1 interactions with significantly greater activity than M7824 (M7824 EC 50 =0.8504nM, LP008-02 EC 50 =0.3630nM, LP008-06a-ES EC 50 =0.4553nM). Example 11: Functional assay for TGF-β

[0136] This example evaluated the effects of LP008-02 and LP008-06a-ES on canonical TGF-β signaling using a luciferase assay.

[0137] Serial dilutions of M7824 (a bifunctional anti-PD-L1 / TGFβ trap fusion protein, see, e.g., Knudson et al., Oncoimmunology. 2018;7(5):e1426519), LP008-02 or LP008-06a-ES were incubated with SBE luciferase reporter-transfected 293 cells in the presence of recombinant human TGF-β for approximately 20 hours.

[0138] As shown in FIG. 13, M7824, LP008-02 and LP008-06a-ES blocked TGF-β canonical signaling in a TGF-β SBE luciferase reporter assay system established in 293 cells (IC50=0.06687nM, IC50=0.07352nM, IC50=0.07167nM). Example 12: Binding activity to human PD-L1 ELISA using recombinant human PD-L1

[0139] To evaluate the binding activity of M7824, LP008-02 and LP008-06a-ES, the bifunctional molecules were subjected to ELISA tests.

[0140] Briefly, microtiter plates were coated with 0.5μg / ml human PD-L1-His protein in PBS, 100μl / well overnight at 4℃, then blocked with 150μl / well 1% BSA. Three-fold dilutions of M7824, LP008-02 and LP008-06a-ES starting at 1μg / ml were added to each well and incubated at 37℃ for 1 hour. Plates were washed with PBS / Tween® and then incubated with goat anti-human IgG antibody conjugated with horseradish peroxidase (HRP) for 30 minutes at 37℃. After washing, plates were developed with TMB substrate and analyzed by spectrophotometer at OD 450nm.

[0141] As shown in Figure 14, LP008-02 and LP008-06a-ES bound to human PD-L1 with significantly higher activity than M7824 (EC 50 = 11.82ng / ml and EC 50 = 14.36ng / ml vs EC 50 =23.68ng / ml). Interspecies activity

[0142] To assess the binding of the bispecific antibodies to mouse PD-L1, rat PD-L1, and cynomolgus PD-L1, the antibodies were tested by ELISA.

[0143] Briefly, microtiter plates were coated with mouse, rat and cynomolgus monkey PD-L1 proteins at 0.5μg / ml in PBS, 100μl / well overnight at 4℃, then blocked with 150μl / well 1% BSA. Three-fold dilutions of bispecific antibodies starting at 1μg / ml were added to each well and incubated at 37℃ for 1 hour. Plates were washed with PBS / Tween® and then incubated with goat anti-human IgG antibody conjugated with horseradish peroxidase (HRP) for 30 minutes at 37℃. After washing, plates were developed with TMB substrate and analyzed by spectrophotometer at OD 450nm.

[0144] LP008-02 and LP008-06a-ES were able to bind to cynomolgus monkey PD-L1 with higher affinity than M7824, while only M7824 could bind to rat and mouse PD-L1 (Figure 15 and Table 13). Table 13. Interspecies activity of M7824, CZ010-02 and CZ010-06a-ES [Table 13] Example 13: Binding activity to human TGF-β ELISA by using recombinant human TGF-β

[0145] To evaluate the binding activity of M7824, LP008-02 and LP008-06a-ES to human TGF-β, these bifunctional molecules were subjected to ELISA tests.

[0146] Briefly, microtiter plates were coated with 100 μl / well of human TGF-β protein at 1 μg / ml in PBS overnight at 4° C., then blocked with 150 μl / well of 1% BSA. Three-fold dilutions of M7824, LP008-02 and LP008-06a-ES bifunctional molecules starting at 10 μg / ml were added to each well and incubated for 1 hour at 37° C. The plates were washed with PBS / Tween®, then incubated with goat anti-human IgG antibody conjugated with horseradish peroxidase (HRP) for 30 minutes at 37° C. After washing, the plates were developed with TMB substrate and analyzed by spectrophotometer at OD 450 nm.

[0147] As shown in FIG. 16, M7824, LP008-02, and LP008-06a-ES all showed high activity (EC 50 = 43.43ng / ml, EC 50 = 28.58ng / ml, EC 50 =39.38ng / ml). Interspecies activity

[0148] To evaluate the binding of the bispecific antibodies to mouse, rat and cynomolgus monkey TGF-β, the bifunctional molecules were subjected to ELISA tests.

[0149] Briefly, microtiter plates were coated with mouse, rat and cynomolgus monkey TGF-β proteins at 1 μg / ml in PBS, 100 μl / well overnight at 4° C., then blocked with 150 μl / well 1% BSA. Three-fold dilutions of bispecific antibodies starting at 10 μg / ml were added to each well and incubated for 1 hour at 37° C. Plates were washed with PBS / Tween® and then incubated with goat anti-human IgG antibody conjugated with horseradish peroxidase (HRP) for 30 minutes at 37° C. After washing, plates were developed with TMB substrate and analyzed by spectrophotometer at OD 450 nm.

[0150] All bifunctional molecules tested bound with high activity to cynomolgus monkey, rat and mouse TGF-β (Figure 17 and Table 14). Table 14. Interspecies activity of M7824, LP008-02 and LP008-06a-ES [Table 14] Example 14: Efficacy in MC38 tumor mouse model

[0151] This example used a tumor mouse model to test the in vivo efficacy of the bifunctional molecule.

[0152] MC38 cells expressing human PD-L1, resuspended in PBS, were injected into the right skin of B-hPD-L1 humanized mice at 5 × 10 5 Cells were inoculated subcutaneously in a volume of 0.2 mL at a concentration of 100 μg / mL. The average tumor volume was approximately 55 mm. 3 When tumor volume reached 10 μg / g, 24 mice with appropriate individual tumor volumes were selected for groups, and the animals were randomly assigned to four experimental groups according to tumor volume, with six animals assigned to each group. After anti-mCD20 mAb injection, total human IgG, M7824, LP008-02 and LP008-06a-ES were administered by intraperitoneal injection three times a week. The dose was calculated based on the body weight of the experimental animals at 10 μg / g. The body weight and tumor size of the mice were examined twice a week.

[0153] The results are shown in Figure 18. The bifunctional molecules LP008-02 and LP008-06a-ES showed better efficacy than M7824 in these animal models in terms of tumor growth inhibition. Furthermore, animal deaths were observed in both IgG and M7824 groups, but not in LP008-02 and LP008-06a-ES groups, demonstrating the better safety profile of the new bifunctional molecules. Example 15. Modification of bifunctional molecules

[0154] In this example, certain modified bifunctional molecules (Table 15) were tested for their in vitro efficacy in functional assays for TGF-β. Some of them contained linker sequences of TAGHTQTSTGGGAITTGTSGAGHGP (SEQ ID NO: 87), HYP and / or G4S (SEQ ID NO: 86) repeats. These molecules are referred to as LP008-02-1 to LP008-02-7, respectively. Table 15. Linker and TGF-βRII modification sequence design [Table 15] ELISA using recombinant human TGF-β1

[0155] To assess the binding activity of the modified LP008-02 bifunctional molecules, these bifunctional molecules were tested in ELISA.

[0156] Briefly, microtiter plates were coated with 100 μl / well of human TGF-β1 protein (Acro, TG1-H4212) at 1 μg / ml in PBS overnight at 4° C., then blocked with 150 μl / well of 1% BSA. Three-fold dilutions of modified LP008-02 bifunctional molecules starting at 30 nM were added to each well and incubated for 1 hour at 37° C. The plates were washed with PBS / Tween®, then incubated with goat anti-human IgG (H+L) antibody conjugated with horseradish peroxidase (HRP) for 30 minutes at 37° C. After washing, the plates were developed with TMB substrate and analyzed by spectrophotometer at OD450 nm.

[0157] As shown in FIG. 19, all of the modified LP008-02 bifunctional molecules bound to human TGF-β1 with high activity comparable to that of LP008-02-1. TGF-β functional assay

[0158] Serial dilutions of the modified LP008-02 bifunctional molecule were incubated with 293 cells transfected with the SBE luciferase reporter in the presence of recombinant human TGF-β1 for approximately 22 hours.

[0159] As shown in FIG. 20, LP008-02-2, LP008-02-3 and LP008-02-4 effectively blocked TGF-β canonical signaling in a TGF-β SBE luciferase reporter assay system established in 293 cells (IC50=0.1435nM, IC50=0.1639nM, IC50=0.1882nM), compared with LP008-02-1. Example 16. Comparison of bifunctional molecules

[0160] Molecules 1-7 in Table 15 contained different sequences at the N-terminus and C-terminus of the ectodomain (SEQ ID NO: 72). They were tested for stability and activity to assess the effect of these sequences.

[0161] Molecule 1 (LP008-02-1) contained the entire excellular portion of the protein (SEQ ID NO: 61), including the N-terminal 25 amino acids of the extracellular domain (IPPHVQKSVNNDMIVTDNNGAVKFP, SEQ ID NO: 89, or amino acids 24-48 of isoform B, SEQ ID NO: 71) and the C-terminal fragment (EEYNTSNPD, SEQ ID NO: 90). In addition, this molecule contained several G4S (SEQ ID NO: 86) The repeats were added to the linker.

[0162] Molecule 2 (LP008-02-2), compared to molecule 1, replaced the N-terminal portion of the extracellular domain (amino acids 24-48 of isoform B, SEQ ID NO: 89) with an artificial linker TAGHTQTSTGGGAITTGTSGAGHGP (SEQ ID NO: 87). This linker was modeled based on SEQ ID NO: 89. Changes included (i) removal of the rigid dipeptide PP, (ii) removal of potential cleavage sites QK, N and K, (iii) inclusion of multiple glycine residues to increase flexibility, and (iv) partial removal of hydrophobic residues (e.g., retaining only one I). These changes are shown in Table 16 below. Molecule 2 also contained a single G4S unit at the N-terminus. Table 16 Artificial linkers [Table 16]

[0163] Molecule 3 (LP008-02-3) contained a longer G4S linker than molecule 2. On top of molecule 3, molecule 4 (LP008-02-4) had a deletion of the C-terminal fragment EEYNTSNPD (SEQ ID NO: 90). Molecule 5 (LP008-02-5) replaced the artificial linker SEQ ID NO: 87 with the short linker HYP. Molecules 6 (LP008-02-6) and 7 (LP008-02-7) contained G4S linkers of different lengths on the N-terminal side of the HYP linker. Example 17. Binding activity and stability of bifunctional molecules

[0164] In this example, SEC-HPLC and CE-SDS were used to evaluate the stability of several of the modified bifunctional molecules, including LP008-02-1, and four additional modified molecules, LP008-02-2, LP008-02-3, LP008-02-6 and LP008-02-7.

[0165] The five sequences were expressed in CHO-K1 cells by polyethyleneimine (PEI)-mediated transient transfection and the supernatants were harvested after 10 days. The bifunctional molecules were purified from the culture supernatants by Protein A and then by Superdex 200 pg with purity levels of >99% as detected by SEC-HPLC (Table 17). Table 17. SEC-HPLC and CE-SDS results of test items on day 0 [Table 17]

[0166] To assess the binding activity of the modified LP008-02 bifunctional molecules, these bifunctional molecules were tested by ELISA.

[0167] Briefly, microtiter plates were coated with 1 μg / ml human TGF-β1 protein (Acro, TG1-H4212) in PBS, 100 μl / well overnight at 4° C., then blocked with 150 μl / well 1% BSA. 4-fold serial dilutions of modified LP008-02 bifunctional molecules starting from 30 nM were added to each well and incubated for 1 hour. Plates were washed with PBS / Tween® and then incubated with goat anti-human IgG Fc antibody conjugated with horseradish peroxidase (HRP) for 30 minutes. After washing, plates were incubated with TMB substrate for color development and analyzed by spectrophotometer at OD 450 nm.

[0168] As shown in FIG. 21, all of the other modified LP008-02 bifunctional molecules bound to human TGF-β1 with high activity comparable to that of LP008-02-1.

[0169] To evaluate the effect of modified LP008-02 bifunctional molecules on canonical TGF-β signaling, modified bifunctional molecules were tested in luciferase assay. Serial dilutions of bifunctional molecules were incubated with 293 cells transfected with SBE luciferase reporter in the presence of recombinant human TGF-β1 for 24 hours. As shown in Figure 22, LP008-02-1, LP008-02-2, LP008-02-3, LP008-02-6 and LP008-02-7 effectively blocked TGF-β canonical signaling in the TGF-β SBE luciferase reporter assay system constructed in 293 cells (IC50=0.04231nM, IC50=0.0527nM, IC50=0.09616nm and IC50=0.1962nM).

[0170] The bifunctional molecules were dissolved in two buffers separately for antibody stability detection. The buffer information is listed as follows: Buffer A: 20 mM acetic acid-sodium acetate, 250 mM sorbitol, 0.02% polysorbate 80, pH 4.9; Buffer B: 20 ​​mM His / His-HCl, 250 mM trehalose, pH 5.4.

[0171] The prepared 3.0 mg / ml samples were incubated at 40° C. and then detected by SEC-HPLC and CE-SDS on days 0 and 14, respectively. As shown in Table 18, LP008-02-2, LP008-02-3, LP008-02-6 and LP008-02-7 formulated in both Buffer A and Buffer B have higher stability than LP008-02-1 in SEC-HPLC, non-reduced CE-SDS and reduced CE-SDS. Table 18. SEC-HPLC and CE-SDS results of test items on day 14 [Table 18]

[0172] Thus, this example shows that the modified bifunctional molecules LP008-02-2, LP008-02-3, LP008-02-6 and LP008-02-7 exhibited similar activity to LP008-02-1, but significantly higher stability than LP008-02-1. Replacing the N-terminal portion of TGF-βRII in LP008-02-1 (IPPHVQKSVNNDMIVTDNNGAVKFP, SEQ ID NO: 89) with an artificial linker (e.g., TAGHTQTSTGGGAITTGTSGAGHGP (SEQ ID NO: 87) or HYP) significantly improved stability. Example 18. High concentration formulation of anti-PD-L1 antibody

[0173] This example used HIC-HPLC and viscosity studies to assess the potential and risks of developing a highly concentrated anti-PD-L1 molecule formulation.

[0174] The four anti-PD-L1 molecules were expressed in CHO-K1 or 293F cells by transient transfection. The constant region of the heavy chain is human IgG1(N297A)-Fc. The purified MPDL3280A (atezolizumab), 47C6A3, Hu67F3G7-22 and Hu89C10H8-7 antibodies were tested by HIC-HPLC to obtain the ammonium sulfate concentrations corresponding to the hydrophobic elution times, which were used to predict the solubility ranges of these molecules. As shown in Table 19, the ammonium sulfate concentrations corresponding to the hydrophobic elution times of MPDL3280A, 47C6A3, Hu67F3G7-22 and Hu89C10H8-7 are 0.41M, 0.78M, 0.97M and 1.10M, respectively. All of the newly developed antibodies are less hydrophobic than the reference antibody MPDL3280A. Table 19. Antibody hydrophobicity from HIC-HPLC testing [Table 19]

[0175] The activity of anti-PD-L1 antibodies in blocking PD1 / PD-L1 interaction was then measured in a bioluminescent cell-based assay. In this assay, PD1 effector cells were co-cultured with PD-L1 target cells, and PD-1 / PD-L1 interaction inhibited TCR signaling and NFAT-RE-mediated luminescence. Addition of either anti-PD-1 or anti-PD-L1 antibodies that block PD-1 / PD-L1 interaction would release an inhibitory signal, resulting in TCR activation and NFAT-RE-mediated luminescence. As shown in Figure 23, MPDL3280A, 47C6A3, Hu67F3G7-22 and Hu89C10H8-7 blocked PD1 and PD-L1 interaction with fairly high activity (MPDL3280A EC 50 =0.1327nM, 47C6A3 EC 50 =0.1501nM, Hu67F3G7-22 EC 50 =0.1034nM, Hu89C10H8-7 EC 50 =0.2138nM).

[0176] MPDL3280A and Hu67F3G7-22 with human IgG1 Fc were expressed in CHO-K1 cells by transient transfection. The purified MPDL3280A-hIgG1 Fc and Hu67F3G7-22-hIgG1 Fc antibodies were tested by HIC-HPLC to obtain the ammonium sulfate concentration corresponding to the hydrophobic elution time, which was used to predict the solubility range of the two molecules. As shown in Table 20, the ammonium sulfate concentrations corresponding to the hydrophobic elution time of MPDL3280A-hIgG1 Fc and Hu67F3G7-22-hIgG1 Fc are 0.42 M and 0.99 M, respectively. Again, with the same Fc fragment, Hu67F3G7-22 showed lower hydrophobicity than MPDL3280A. Table 20. HIC-HPLC test results [Table 20]

[0177] To further confirm the solubility and viscosity properties of the antibodies, the two purified candidates were directly concentrated in phosphate buffer (containing 60 mM NaCl) by ultrafiltration. During the ultrafiltration process, the concentration, SEC-HPLC, and viscosity properties were measured at different stages. As shown in Table 21, the viscosity of MPDL3280A-hIgG1 Fc is much higher than that of Hu 67F3G7-22-hIgG1 Fc at similar concentrations. Also, for high concentration formulations, antibodies with lower viscosity are generally preferred over antibodies with higher viscosity. Therefore, as a therapeutic protein, the Hu67F3G7-22 antibody has higher potential than MPDL3280A. Table 21 Solubility test results [Table 21]

[0178] The present disclosure should not be limited in scope by the described specific embodiments, which are intended as single illustrations of individual aspects of the present disclosure, and any composition or method that is functionally equivalent is within the scope of the present disclosure.It is clear to those skilled in the art that various modifications and variations can be made to the methods and compositions of the present disclosure without departing from the spirit or scope of the present disclosure.Therefore, the present disclosure is intended to cover the modifications and variations of the present disclosure as long as they fall within the scope of the appended claims and their equivalents.

[0179] All publications and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference. The present invention provides, for example, the following items. (Item 1) A multifunctional molecule comprising an anti-PD-L1 (programmed death ligand 1) antibody or a fragment thereof and an extracellular domain of human TGF-βRII (TGF-β receptor type 2), the anti-PD-L1 antibody or fragment thereof has specificity for the human PD-L1 protein, and comprises a heavy chain variable region (VH) comprising VH CDR1, VH CDR2, and VH CDR3, and a light chain variable region (VL) comprising VL CDR1, VL CDR2, and VL CDR3; the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 each comprise an amino acid sequence of SEQ ID NO: 7 to 12 or SEQ ID NO: 13 to 18, or the VH CDR1 comprises SEQ ID NO: 19, the VH CDR2 comprises SEQ ID NO: 20, 91 or 92, the VH CDR3 comprises SEQ ID NO: 21, the VL CDR1 comprises SEQ ID NO: 22 and the VL CDR2 comprises SEQ ID NO: 23; the VL CDR3 comprises SEQ ID NO: 24 or 93, A multifunctional molecule, wherein the human TGF-βRII extracellular domain comprises the amino acid sequence of SEQ ID NO:72, and is fused to the anti-PD-L1 antibody or fragment thereof. (Item 2) The multifunctional molecule of item 1, wherein the anti-PD-L1 antibody or fragment thereof comprises a heavy chain that comprises the VH and a separate light chain that comprises the VL. (Item 3) 3. The multifunctional molecule of item 2, wherein the TGF-βRII extracellular domain is fused to the heavy chain of the anti-PD-L1 antibody or fragment thereof. (Item 4) 4. The multifunctional molecule of item 3, wherein the TGF-βRII extracellular domain is fused to the C-terminus of the heavy chain of the anti-PD-L1 antibody or fragment thereof. (Item 5) 5. The multifunctional molecule of item 3 or 4, wherein the TGF-βRII extracellular domain is fused to the heavy chain of the anti-PD-L1 antibody or fragment thereof via a peptide linker. (Item 6) 6. The multifunctional molecule according to any one of items 1 to 5, wherein the TGF-βRII extracellular domain comprises SEQ ID NO: 72 and at least a partial deletion of amino acid residues 24 to 48 of SEQ ID NO: 71. (Item 7) 6. The multifunctional molecule according to any one of items 1 to 5, wherein the TGF-βRII extracellular domain comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 61 and 73 to 78, wherein for SEQ ID NO: 74, X is any amino acid except K, S or N. (Item 8) 6. The multifunctional molecule of any one of items 1 to 5, comprising at least 30 amino acid residues between SEQ ID NO: 72 and the anti-PD-L1 antibody or fragment thereof. (Item 9) 6. The multifunctional molecule of any one of items 1 to 5, comprising an alpha-helical motif between SEQ ID NO: 72 and the anti-PD-L1 antibody or fragment thereof. (Item 10) 10. The multifunctional molecule according to any one of items 1 to 9, wherein the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 each comprise an amino acid sequence of SEQ ID NO: 13 to 18. (Item 11) 11. The multifunctional molecule according to item 10, wherein the VH comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 31 to 37, and the VL comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 38 to 43. (Item 12) 12. The multifunctional molecule of item 11, wherein the VH comprises the amino acid sequence of SEQ ID NO: 34 and the VL comprises the amino acid sequence of SEQ ID NO: 43. (Item 13) The multifunctional molecule according to any one of items 1 to 9, wherein the VH CDR1 comprises SEQ ID NO: 19, the VH CDR2 comprises SEQ ID NO: 20, 91 or 92, the VH CDR3 comprises SEQ ID NO: 21, the VL CDR1 comprises SEQ ID NO: 22, the VL CDR2 comprises SEQ ID NO: 23, and the VL CDR3 comprises SEQ ID NO: 24 or 93. (Item 14) Item 14. The multifunctional molecule according to item 13, wherein the VH comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 44 to 49 and 57 to 58, and the VL comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 50 to 56. (Item 15) 14. The multifunctional molecule of item 13, wherein the VH comprises the amino acid sequence of SEQ ID NO: 48, 57 or 58, and the VL comprises the amino acid sequence of SEQ ID NO: 53 or 56. (Item 16) 10. The multifunctional molecule according to any one of items 1 to 9, wherein the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 comprise the amino acid sequences of SEQ ID NOs: 19, 92, 21, 22, 23 and 93, respectively. (Item 17) 17. The multifunctional molecule of item 16, wherein the VH comprises the amino acid sequence of SEQ ID NO: 58 and the VL comprises the amino acid sequence of SEQ ID NO: 56. (Item 18) 18. The multifunctional molecule according to any one of items 1 to 17, comprising a light chain comprising the VL and a light chain constant region, and a heavy chain comprising the VH, a heavy chain constant region, a peptide linker, and the TGF-βRII extracellular domain. (Item 19) 19. The multifunctional molecule of item 18, wherein the heavy chain constant region comprises the amino acid sequence of SEQ ID NO:59. (Item 20) an anti-PD-L1 (programmed death ligand 1) antibody or a fragment thereof, which has specificity for the human PD-L1 protein, and comprises a heavy chain variable region (VH) comprising VH CDR1, VH CDR2, and VH CDR3, and a light chain variable region (VL) comprising VL CDR1, VL CDR2, and VL CDR3, wherein the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2, and VL CDR3 comprise the amino acid sequences of SEQ ID NOs: 7 to 12 or SEQ ID NOs: 13 to 18, respectively, or the VH CDR1 comprises SEQ ID NO: 19, the VH CDR2 comprises SEQ ID NO: 20, 91 or 92, the VH CDR3 comprises SEQ ID NO: 21, the VL CDR1 comprises SEQ ID NO: 22, the VL CDR2 comprises SEQ ID NO: 23, and the VL CDR3 comprises SEQ ID NO: 24 or 93. (Item 21) 21. The anti-PD-L1 antibody or fragment thereof of item 20, wherein the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 respectively comprise the amino acid sequences of SEQ ID NOs: 13 to 18. (Item 22) 22. The anti-PD-L1 antibody or fragment thereof of item 21, wherein the VH comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 31 to 37, and the VL comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 38 to 43. (Item 23) 22. The anti-PD-L1 antibody or fragment thereof of item 21, wherein the VH comprises the amino acid sequence of SEQ ID NO: 34 and the VL comprises the amino acid sequence of SEQ ID NO: 43. (Item 24) 21. The anti-PD-L1 antibody or fragment thereof of item 20, wherein the VH CDR1 comprises SEQ ID NO: 19, the VH CDR2 comprises SEQ ID NO: 20, 91 or 92, the VH CDR3 comprises SEQ ID NO: 21, the VL CDR1 comprises SEQ ID NO: 22, the VL CDR2 comprises SEQ ID NO: 23, and the VL CDR3 comprises SEQ ID NO: 24 or 93. (Item 25) 25. The anti-PD-L1 antibody or fragment thereof of item 24, wherein the VH comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 44 to 49 and 57 to 58, and the VL comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 50 to 56. (Item 26) Said VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CD 25. The anti-PD-L1 antibody or fragment thereof of item 24, wherein the R2 and VL CDR3 comprise the amino acid sequences of SEQ ID NOs: 19, 92, 21, 22, 23, and 93, respectively. (Item 27) 27. The anti-PD-L1 antibody or fragment thereof of item 26, wherein the VH comprises the amino acid sequence of SEQ ID NO: 58 and the VL comprises the amino acid sequence of SEQ ID NO: 56. (Item 28) 21. The anti-PD-L1 antibody or fragment thereof of item 20, wherein the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 respectively comprise the amino acid sequences of SEQ ID NOs: 7 to 12. (Item 29) 29. The anti-PD-L1 antibody or fragment thereof of item 28, wherein the VH comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 25 to 28, and the VL comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 29 to 30. (Item 30) 29. The anti-PD-L1 antibody or fragment thereof of item 28, wherein the VH comprises the amino acid sequence of SEQ ID NO: 26, 27, or 28, and the VL comprises the amino acid sequence of SEQ ID NO: 30. (Item 31) A multifunctional molecule comprising an antibody or antigen-binding fragment thereof fused to the N-terminus of the amino acid sequence of SEQ ID NO: 72 via a peptide linker. (Item 32) 32. The multifunctional molecule of item 31, wherein the peptide linker comprises a flexible linker and / or a replacement peptide of IPPHVQKSVNNDMIVTDNNGAVKFP (SEQ ID NO: 89), and the replacement peptide is different from SEQ ID NO: 89. (Item 33) 33. The multifunctional molecule of item 32, wherein the substituted peptide comprises the amino acid sequence of IPPHVQXXVNNDMIVTDNXGAVKFP (SEQ ID NO: 88), wherein X is any amino acid except K, S, or N. (Item 34) 34. The multifunctional molecule of item 33, wherein the substituted peptide has at least 50% sequence identity with SEQ ID NO:88. (Item 35) 33. The multifunctional molecule of item 32, wherein the substituted peptide comprises an amino acid sequence of TAGHTQTSTGGGAITTGTSGAGHGP (SEQ ID NO: 87), or a variant having at least 75% sequence identity with SEQ ID NO: 87, wherein the variant comprises at least four Gs, no PP dipeptide, and no more than three hydrophobic amino acid residues selected from the group consisting of I, L, M, F, V, W, Y, and P. (Item 36) 36. The multifunctional molecule of item 35, wherein the variant comprises at least five Gs and no more than one hydrophobic amino acid residue selected from the group consisting of I, L, M, F, V, W, Y and P. (Item 37) 37. The multifunctional molecule according to any one of items 32 to 36, wherein the flexible linker comprises S and at least 50% G. (Item 38) 38. The multifunctional molecule of item 37, wherein the flexible linker comprises one or more GGGGS (SEQ ID NO: 86) units. (Item 39) 39. The multifunctional molecule according to any one of items 31 to 38, wherein the C-terminus of the antibody or antigen-binding fragment thereof is at least 20 amino acid residues from the N-terminus of the amino acid sequence of SEQ ID NO: 72. (Item 40) 40. The multifunctional molecule according to any one of items 31 to 39, which does not contain at least the entire sequence of EEYNTSNPD (SEQ ID NO: 90). (Item 41) 41. The multifunctional molecule according to any one of items 31 to 40, wherein the antibody or fragment thereof is specific for an antigen selected from the group consisting of PD-1, PD-L1, CTLA-4, LAG-3, CD28, CD122, 4-1BB, TIM 3, OX-40, OX40L, CD40, CD40L, LIGHT, ICOS, ICOSL, GITR, GITRL, TIGIT, CD27, VISTA, B7H3, B7H4, BTLA, CD4, CD2, CD8, CD47 and CD73. (Item 42) 42. The multifunctional molecule according to any one of items 31 to 41, wherein the antibody or fragment thereof comprises an Fc fragment. (Item 43) 43. The multifunctional molecule of item 42, wherein the peptide linker is fused to the C-terminus of the Fc fragment. (Item 44) A cell comprising one or more polynucleotides encoding the multifunctional molecule of any one of items 1 to 19 and 31 to 43, or the anti-PD-L1 antibody or fragment thereof of any one of items 20 to 30. (Item 45) One or more polynucleotides encoding the multifunctional molecule of any one of items 1 to 19 and 31 to 43, or the anti-PD-L1 antibody or fragment thereof of any one of items 20 to 30. (Item 46) A composition comprising the multifunctional molecule of any one of items 1 to 19 and 31 to 43, or the anti-PD-L1 antibody or fragment thereof of any one of items 20 to 30, and a pharma- ceutically suitable carrier. (Item 47) Use of the multifunctional molecule of any one of items 1 to 19 and 31 to 43, or the anti-PD-L1 antibody or fragment thereof of any one of items 20 to 30, for the manufacture of a medicament for treating cancer. (Item 48) A method for treating cancer in a patient in need thereof, comprising administering to the patient an effective amount of the multifunctional molecule of any one of items 1 to 19 and 31 to 43, or the anti-PD-L1 antibody or fragment thereof of any one of items 20 to 30. (Item 49) The use according to item 47 or the method according to item 48, wherein the cancer is a solid tumor. (Item 50) 50. The use according to item 47 or 49 or the method according to item 48 or 49, wherein the cancer is selected from the group consisting of bladder cancer, liver cancer, colon cancer, rectal cancer, endometrial cancer, leukemia, lymphoma, pancreatic cancer, small cell lung cancer, non-small cell lung cancer, breast cancer, urethral cancer, head and neck cancer, gastrointestinal cancer, stomach cancer, esophageal cancer, ovarian cancer, renal cancer, melanoma, prostate cancer and thyroid cancer.

Claims

1. A multifunctional molecule comprising an anti-PD-L1 (programmed death ligand 1) antibody or an antigen-binding fragment thereof and an extracellular domain of human TGF-βRII (TGF-β receptor type 2), the anti-PD-L1 antibody or antigen-binding fragment thereof has specificity for the human PD-L1 protein and comprises a heavy chain variable region (VH) comprising a VH CDR1, a VH CDR2, and a VH CDR3, and a light chain variable region (VL) comprising a VL CDR1, a VL CDR2, and a VL CDR3; the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3 each comprise an amino acid sequence of SEQ ID NO: 13-18; The multifunctional molecule, wherein the human TGF-βRII extracellular domain comprises the amino acid sequence of SEQ ID NO:72 and is fused to the anti-PD-L1 antibody or antigen-binding fragment thereof.

2. The multifunctional molecule described in claim 1, wherein the anti-PD-L1 antibody or antigen-binding fragment thereof comprises a heavy chain comprising the VH and a separate light chain comprising the VL.

3. The multifunctional molecule of claim 2, wherein the TGF-βRII extracellular domain is fused to the heavy chain of the anti-PD-L1 antibody or its antigen-binding fragment.

4. The multifunctional molecule of claim 3, wherein the TGF-βRII extracellular domain is fused to the C-terminus of the heavy chain of the anti-PD-L1 antibody or antigen-binding fragment thereof.

5. The multifunctional molecule described in claim 3 or 4, wherein the TGF-βRII extracellular domain is fused to the heavy chain of the anti-PD-L1 antibody or its antigen-binding fragment via a peptide linker.

6. A multifunctional molecule described in any one of claims 1 to 5, wherein the TGF-βRII extracellular domain comprises SEQ ID NO:72 and at least a partial deletion of amino acid residues 24 to 48 of SEQ ID NO:

71.

7. The multifunctional molecule of claim 1, wherein the TGF-βRII extracellular domain comprises an amino acid sequence selected from the group consisting of SEQ ID NO:61 and 73-78, wherein for SEQ ID NO:74, X is any amino acid except K, S or N.

8. The multifunctional molecule of any one of claims 1 to 5, comprising at least 30 amino acid residues between SEQ ID NO: 72 and the anti-PD-L1 antibody or antigen-binding fragment thereof.

9. A multifunctional molecule described in any one of claims 1 to 5, comprising an alpha helix motif between sequence number 72 and the anti-PD-L1 antibody or its antigen-binding fragment.

10. The multifunctional molecule described in claim 1, wherein the VH comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 31 to 37, and the VL comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 38 to 43.

11. The multifunctional molecule described in claim 10, wherein the VH comprises the amino acid sequence of SEQ ID NO: 34 and the VL comprises the amino acid sequence of SEQ ID NO:

43.

12. A multifunctional molecule according to any one of claims 1 to 11, comprising a light chain comprising the VL and a light chain constant region, and a heavy chain comprising the VH, a heavy chain constant region, a peptide linker, and the TGF-βRII extracellular domain.

13. The multifunctional molecule described in claim 12, wherein the heavy chain constant region comprises the amino acid sequence of SEQ ID NO:

59.

14. An anti-PD-L1 antibody or antigen-binding fragment thereof having specificity for human PD-L1 (programmed death ligand 1) protein and comprising a heavy chain variable region (VH) comprising VH CDR1, VH CDR2, and VH CDR3, and a light chain variable region (VL) comprising VL CDR1, VL CDR2, and VL CDR3, wherein the VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2, and VL CDR3 each comprise the amino acid sequence of SEQ ID NO: 13 to 18.

15. The anti-PD-L1 antibody or antigen-binding fragment thereof according to claim 14, wherein the VH comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 31 to 37, and the VL comprises an amino acid sequence selected from the group consisting of SEQ ID NOs: 38 to 43.

16. The anti-PD-L1 antibody or antigen-binding fragment thereof described in claim 14, wherein the VH comprises the amino acid sequence of SEQ ID NO: 34 and the VL comprises the amino acid sequence of SEQ ID NO:

43.

17. A multifunctional molecule comprising an antibody or antigen-binding fragment thereof fused to the N-terminus of the amino acid sequence of SEQ ID NO: 72 via a peptide linker.

18. The multifunctional molecule described in claim 17, wherein the peptide linker comprises a flexible linker and / or a substituted peptide of IPPHVQKSVNNDMIVTDNNGAVKFP (sequence number 89), wherein the substituted peptide is different from sequence number 89.

19. The multifunctional molecule of claim 18, wherein the substituted peptide comprises an amino acid sequence of IPPHVQXXVNNDMIVTDNXGAVKFP (SEQ ID NO:88), where X is any amino acid except K, S, or N.

20. The multifunctional molecule described in claim 19, wherein the substituted peptide has at least 50% sequence identity with sequence number 88.

21. The multifunctional molecule of claim 18, wherein the substituted peptide comprises an amino acid sequence of TAGHTQTSTGGGAITTGTSGAGHGGP (SEQ ID NO: 87), or a variant having at least 75% sequence identity to SEQ ID NO: 87, said variant containing at least four Gs, no PP dipeptide, and containing three or less hydrophobic amino acid residues selected from the group consisting of I, L, M, F, V, W, Y and P.

22. The multifunctional molecule described in claim 21, wherein the variant contains at least five Gs and one or less hydrophobic amino acid residues selected from the group consisting of I, L, M, F, V, W, Y and P.

23. A multifunctional molecule described in any one of claims 18 to 22, wherein the flexible linker comprises S and at least 50% G.

24. The multifunctional molecule of claim 23, wherein the flexible linker comprises one or more GGGGS (sequence number 86) units.

25. A multifunctional molecule described in any one of claims 17 to 24, wherein the C-terminus of the antibody or antigen-binding fragment thereof is at least 20 amino acid residues from the N-terminus of the amino acid sequence of SEQ ID NO:

72.

26. A multifunctional molecule described in any one of claims 17 to 25, which does not contain at least the entire sequence of EEYNTSNPD (sequence number 90).

27. The multifunctional molecule of any one of claims 17 to 26, wherein the antibody or antigen-binding fragment thereof is specific for an antigen selected from the group consisting of PD-1, PD-L1, CTLA-4, LAG-3, CD28, CD122, 4-1BB, TIM 3, OX-40, OX40L, CD40, CD40L, LIGHT, ICOS, ICOSL, GITR, GITRL, TIGIT, CD27, VISTA, B7H3, B7H4, BTLA, CD4, CD2, CD8, CD47, and CD73.

28. A multifunctional molecule described in any one of claims 17 to 27, wherein the antibody or its antigen-binding fragment comprises an Fc fragment.

29. The multifunctional molecule described in claim 28, wherein the peptide linker is fused to the C-terminus of the Fc fragment.

30. A cell comprising one or more polynucleotides encoding a multifunctional molecule described in any one of claims 1 to 13 and 17 to 29, or an anti-PD-L1 antibody or antigen-binding fragment thereof described in any one of claims 14 to 16.

31. One or more polynucleotides encoding a multifunctional molecule described in any one of claims 1 to 13 and 17 to 29, or an anti-PD-L1 antibody or antigen-binding fragment thereof described in any one of claims 14 to 16.

32. A composition comprising a multifunctional molecule according to any one of claims 1 to 13 and 17 to 29, or an anti-PD-L1 antibody or antigen-binding fragment thereof according to any one of claims 14 to 16, and a pharma- ceutical suitable carrier.

33. Use of a multifunctional molecule described in any one of claims 1 to 13 and 17 to 29, or an anti-PD-L1 antibody or antigen-binding fragment thereof described in any one of claims 14 to 16, for the manufacture of a medicament for treating cancer.

34. A composition for treating cancer in a patient in need thereof, comprising an effective amount of a multifunctional molecule according to any one of claims 1 to 13 and 17 to 29, or an anti-PD-L1 antibody or antigen-binding fragment thereof according to any one of claims 14 to 16.

35. The use described in claim 33 or the composition described in claim 34, wherein the cancer is a solid tumor.

36. The use of claim 33 or 35 or the composition of claim 34 or 35, wherein the cancer is selected from the group consisting of bladder cancer, liver cancer, colon cancer, rectal cancer, endometrial cancer, leukemia, lymphoma, pancreatic cancer, small cell lung cancer, non-small cell lung cancer, breast cancer, urethral cancer, head and neck cancer, gastrointestinal cancer, stomach cancer, esophageal cancer, ovarian cancer, renal cancer, melanoma, prostate cancer and thyroid cancer.

Images