Albumin-binding domain fusion protein

Abstract

To solve the problem in which: while biologics have been useful in the treatment of many diseases, including cancers, the short circulatory half-life of such molecules poses a major obstacle.SOLUTION: Compositions that include an albumin binding domain and a fusion partner (e.g., a cytokine or a binding moiety) are provided. Such therapeutics have increased serum half-life and find use in applications where one or more such therapeutics are needed, for example, in oncology applications.SELECTED DRAWING: None

Description

Background Art

[0001] Biological agents are useful for the treatment of many diseases, including cancer, but the short circulating half-life of such molecules is a major obstacle.

[0002] Biological agents are useful for the treatment of cancer in various ways. Cytokine-based therapies can act directly on cancer cells by preventing the growth and proliferation of cancer cells. Cytokines can also stimulate the immune system by promoting the growth of killer T cells and other cells that attack cancer cells. Additionally, cytokines can prompt cancer cells to secrete chemicals that attract immune system cells. See, for example, Dranoff, Nature Reviews Cancer 4:11-22 (2004), and Zhang et al., Proc Natl Acad Sci USA 106(18):7513-7518 (2009). Antibodies are desirable as therapeutic agents because of their ability to recognize targets with both specificity and high affinity. Monoclonal antibody-based therapies, including those targeting tumor surface antigens and inhibitory signals that limit T cell activation, have been standard components of cancer treatment regimens for over 20 years. See, for example, Weiner, Nat Rev Cancer 15(6):361-370 (2015).

[0003] A short circulation half-life is a major obstacle for many biopharmaceuticals. See, for example, Perdreau et al., European Cytokine Network 21:297-307 (2010). Such short-acting therapeutic agents require a frequent dosing profile that can reduce their applicability in the clinic, especially in the case of chronic diseases. A long serum half-life is desirable to reduce the need for repeated injections of the molecule to achieve a therapeutically appropriate serum concentration. Methods for extending the half-life of therapeutic proteins include PEGylation, fusion to human serum albumin (HSA), fusion to the constant fragment (Fc) of human immunoglobulin IgG, and fusion to unstructured polypeptides such as XTEN. See, for example, Stohl, BioDrugs 29(4):215-239 (2015). Half-life extension technologies enable new and improved biological therapies that reduce the cost and burden of frequent dosing. Accordingly, there is a continuing need for novel reagents and methods useful for extending the half-life of protein and peptide-based therapeutic agents.

Summary of the Invention

[0004] Compositions comprising an albumin binding domain (ABD) are provided herein. As described herein, biopharmaceuticals comprising the subject albumin binding domain (i.e., albumin binding domain fusion proteins) advantageously exhibit an extended half-life and better in vivo pharmacokinetics compared to biopharmaceuticals without an ABD.

[0005] In one aspect, compositions comprising an albumin binding domain (ABD) are provided herein. The ABD comprises a) a variable heavy chain comprising any one of vhCDR1, vhCDR2, and vhCDR3 of the variable heavy chains shown in FIG. 2, and b) a variable light chain comprising any one of vlCDR1, vlCDR2, and vlCDR3 of the variable light chains shown in FIG. 2.

[0006] In some embodiments, according to any one of the vhCDR1 sequence, vhCDR2 sequence, and vhCDR3 sequence shown in FIG. 2, the vhCDR1 contains the vhCDR1 sequence, the vhCDR2 contains the vhCDR2 sequence, and the vhCDR3 contains the vhCDR3 sequence. In one embodiment, the variable heavy chain contains any one of the variable heavy chain sequences shown in FIG. 2. In some embodiments, according to any one of the vlCDR1 sequence, vlCDR2 sequence, and vhCDR3 sequence shown in FIG. 2, the vlCDR1 contains the vlCDR1 sequence, the vlCDR2 contains the vlCDR2 sequence, and the vlCDR3 contains the vlCDR3 sequence. In one embodiment, the variable light chain contains any one of the variable light chain sequences shown in FIG. 2. In an exemplary embodiment, the albumin binding domain comprises the variable heavy chain and variable light chain of A10m3 (FIG. 2D). In another aspect, a composition comprising variant IL-15 is provided herein. Variant IL-15 contains one or more amino acid substitutions selected from the group consisting of K86A, K86R, N112A, N112S, N112Q, K86A / N112A, K86R / N112A, K86A / N112S, K86R / N112S, K86A / N112Q, K86R / N112Q, K86A / N112A / N79A, K86R / N112A / N79A, K86A / N112A / N79D, K86R / N112A / N79D, K86A / N112A / N79Q, K86R / N112A / N79Q, K86A / N112A / N71D, K86R / N112A / N71D, K86A / N112A / N71Q, K86R / N112A / N71Q, K86A / N112A / N71D / N79A, K86A / N112A / N71D / N79D, K86A / N112A / N71Q / N79A, K86A / N112A / N71Q / N79D, K86R / N112A / N71D / N79A, K86R / N112A / N71D / N79D, K86R / N112A / N71D / N79Q, K86R / N112A / N71Q / N79A, K86R / N112A / N71Q / N79D, and K86R / N112A / N71Q / N79Q compared to the parental IL-15.

[0007] ​

[0008] In some embodiments, the IL-15 variant includes any one of the amino acid sequences of the IL-15 variants shown in FIG. 3. In certain embodiments, the IL-15 variant further includes IL-15 receptor α (IL-15Rα) bound to said IL-15.

[0009] In one aspect, an albumin binding domain (ABD) fusion protein comprising an ABD bound to a fusion partner is provided herein. The ABD includes a variable heavy chain comprising any one of vhCDR1, vhCDR2, and vhCDR3 of the variable heavy chains shown in FIG. 2, and a variable light chain comprising any one of vlCDR1, vlCDR2, and vlCDR3 of the variable light chains shown in FIG. 2.

[0010] In some embodiments, according to any one of the vhCDR1 sequence, vhCDR2 sequence, and vhCDR3 sequence shown in FIG. 2, vhCDR1 includes the vhCDR1 sequence, vhCDR2 includes the vhCDR2 sequence, and vhCDR3 includes the vhCDR3 sequence. In some embodiments, the variable heavy chain includes any one of the variable heavy chain sequences shown in FIG. 2.

[0011] In certain embodiments, according to any one of the vlCDR1 sequence, vlCDR2 sequence, and vhCDR3 sequence shown in FIG. 2, vlCDR1 includes the vlCDR1 sequence, vlCDR2 includes the vlCDR2 sequence, and vlCDR3 includes the vlCDR3 sequence. In certain embodiments, the variable light chain includes any one of the variable light chain sequences shown in FIG. 2.

[0012] In an exemplary embodiment, the variable heavy chain and variable light chain include the variable heavy chain and variable light chain of A10m3 (FIG. 2D), respectively.

[0013] In some embodiments, the fusion partner is a cytokine. In certain embodiments, the cytokine is selected from IL-2, IL-7, IL-12, IL-15, IL-18, IL-21, GM-CSF, and IFN-α.

[0014] In certain embodiments, the fusion partner is a binding moiety. In some embodiments, the binding moiety is a scFv comprising a scFv variable heavy chain and a scFv variable light chain. In some embodiments, the scFv is selected from anti-TGFβ scFv, anti-PD-L1 scFv, and anti-TNF scFv. In some embodiments, the scFv is an anti-interleukin scFv. In an exemplary embodiment, the scFv is an anti-IL-1, IL-6, IL-8, IL-17(A-F), or IL-23 scFv.

[0015] In some embodiments, the ABD is linked to the fusion partner by a linker. In an exemplary embodiment, the linker is (GGGGS) x wherein x is an integer from 1 to 10.

[0016] In another aspect, an IL15-albumin binding domain (ABD) fusion protein according to the formula (IL-15)-L-(ABD) is provided herein. The ABD comprises a variable heavy chain comprising any one of vhCDR1, vhCDR2, and vhCDR3 of the variable heavy chains shown in FIG. 2, and a variable light chain comprising any one of vlCDR1, vlCDR2, and vlCDR3 of the variable light chains shown in FIG. 2, and L is a linker.

[0017] In some embodiments, according to any of the vhCDR1 sequences, vhCDR2 sequences, and vhCDR3 sequences shown in FIG. 2, vhCDR1 comprises the vhCDR1 sequence, vhCDR2 comprises the vhCDR2 sequence, and vhCDR3 comprises the vhCDR3 sequence. In some embodiments, the variable heavy chain comprises any one of the variable heavy chain sequences shown in FIG. 2.

[0018] In certain embodiments, in accordance with any of the vlCDR1 sequence, vlCDR2 sequence, and vhCDR3 sequence shown in FIG. 2, the vlCDR1 contains the vlCDR1 sequence, the vlCDR2 contains the vlCDR2 sequence, and the vlCDR3 contains the vlCDR3 sequence. In certain embodiments, the variable light chain contains any one of the variable light chain sequences shown in FIG. 2.

[0019] In an exemplary embodiment, the variable heavy chain and the variable light chain contain the variable heavy chain and variable light chain of A10m3 (FIG. 2D), respectively.

[0020] 28. The IL-15-ABD fusion protein according to any one of claims 22 to 27, wherein the IL-15 is a variant IL-15 comprising one or more amino acid substitutions selected from the group consisting of K86A, K86R, N112A, N112S, N112Q, K86A / N112A, K86R / N112A, K86A / N112S, K86R / N112S, K86A / N112Q, K86R / N112Q, K86A / N112A / N79A, K86R / N112A / N79A, K86A / N112A / N79D, K86R / N112A / N79D, K86A / N112A / N79Q, K86R / N112A / N79Q, K86A / N112A / N71D, K86R / N112A / N71D, K86A / N112A / N71Q, K86R / N112A / N71Q, K86A / N112A / N71D / N79A, K86A / N112A / N71D / N79D, K86A / N112A / N71Q / N79A, K86A / N112A / N71Q / N79D, K86R / N112A / N71D / N79A, K86R / N112A / N71D / N79D, K86R / N112A / N71D / N79Q, K86R / N112A / N71Q / N79A, K86R / N112A / N71Q / N79D, and K86R / N112A / N71Q / N79Q compared to the parental IL-15. In some embodiments, the variant IL-15 contains any one of the amino acid sequences of the variant IL-15 shown in FIG. 3. In an exemplary embodiment, the variant IL-15 contains the amino acid sequence of IL15 K86R / N112A.

[0021] In certain embodiments, IL-15 is wild-type IL-15. In some embodiments, IL-15 comprises wild-type IL-15 bound to IL-15 receptor α (IL-15Rα).

[0022] In certain embodiments, the linker is selected from any of the linkers shown in FIG. 48. In some embodiments, the linker is (GGGGS) x wherein x is an integer from 1 to 10.

[0023] In one embodiment, the IL15-ABD fusion protein has an amino acid sequence according to any one of the amino acid sequences shown in FIG. 4.

[0024] In another aspect, provided herein is an IL12-albumin binding domain (ABD) fusion protein of formula (IL-12)-L-(ABD). ABD includes a variable heavy chain comprising any one of vhCDR1, vhCDR2, and vhCDR3 of the variable heavy chains shown in FIG. 2, and a variable light chain comprising any one of vlCDR1, vlCDR2, and vlCDR3 of the variable light chains shown in FIG. 2, and L is a linker.

[0025] In some embodiments, according to any of the vhCDR1 sequences, vhCDR2 sequences, and vhCDR3 sequences shown in FIG. 2, vhCDR1 contains the vhCDR1 sequence, vhCDR2 contains the vhCDR2 sequence, and vhCDR3 contains the vhCDR3 sequence. In some embodiments, the variable heavy chain contains any one of the variable heavy chain sequences shown in FIG. 2.

[0026] In certain embodiments, in accordance with any of the vlCDR1 sequences, vlCDR2 sequences, and vhCDR3 sequences shown in FIG. 2, the vlCDR1 contains the vlCDR1 sequence, the vlCDR2 contains the vlCDR2 sequence, and the vlCDR3 contains the vlCDR3 sequence. In certain embodiments, the variable light chain contains any one of the variable light chain sequences shown in FIG. 2.

[0027] In an exemplary embodiment, the variable heavy chain and variable light chain contain the variable heavy chain and variable light chain of A10m3 (FIG. 2D), respectively.

[0028] In one embodiment, IL-12 is a single-chain IL-12 comprising a p35 subunit, a p40 subunit, and an IL-12 linker, and the IL-12 linker covalently attaches the p35 subunit to the p40 subunit. In certain embodiments, the linker is selected from any of the linkers shown in FIG. 48. In an exemplary embodiment, the linker is (GGGGS) x where x is an integer from 1 to 10.

[0029] In an exemplary embodiment, the IL12-ABD fusion protein contains an amino acid sequence according to any one of the amino acid sequences in FIG. 20.

[0030] In another aspect, provided herein is an n albumin binding domain (ABD) fusion protein having a formula selected from a) (IL-12)-L1-(ABD)-L2-(IL-15), and b) (IL-15)-L1-(ABD)-L2-(IL-12) from the N-terminus to the C-terminus. The ABD contains a variable heavy chain containing any one of the vhCDR1, vhCDR2, and vhCDR3 of the variable heavy chains shown in FIG. 2 and a variable light chain containing any one of the vlCDR1, vlCDR2, and vlCDR3 of the variable light chains shown in FIG. 2, and L1 and L2 are the first and second linkers, respectively.

[0031] In some embodiments, according to any one of the vhCDR1 sequence, vhCDR2 sequence, and vhCDR3 sequence shown in FIG. 2, the vhCDR1 contains the vhCDR1 sequence, the vhCDR2 contains the vhCDR2 sequence, and the vhCDR3 contains the vhCDR3 sequence. In some embodiments, the variable heavy chain contains any one of the variable heavy chain sequences shown in FIG. 2.

[0032] In certain embodiments, according to any one of the vlCDR1 sequence, vlCDR2 sequence, and vhCDR3 sequence shown in FIG. 2, the vlCDR1 contains the vlCDR1 sequence, the vlCDR2 contains the vlCDR2 sequence, and the vlCDR3 contains the vlCDR3 sequence. In certain embodiments, the variable light chain contains any one of the variable light chain sequences shown in FIG. 2.

[0033] In an exemplary embodiment, the variable heavy chain and the variable light chain contain the variable heavy chain and variable light chain of A10m3 (FIG. 2D), respectively.

[0034] In some embodiments, IL-15 contains a wild-type IL-15 polypeptide. In certain embodiments, the wild-type IL-15 is bound to the IL-15 receptor α (IL-15Rα).

[0035] In one embodiment, the IL-15 is a variant IL-15 comprising one or more amino acid substitutions selected from K86A, K86R, N112A, N112S, N112Q, K86A / N112A, K86R / N112A, K86A / N112S, K86R / N112S, K86A / N112Q, K86R / N112Q, K86A / N112A / N79A, K86R / N112A / N79A, K86A / N112A / N79D, K86R / N112A / N79D, K86A / N112A / N79Q, K86R / N112A / N79Q, K86A / N112A / N71D, K86R / N112A / N71D, K86A / N112A / N71Q, K86R / N112A / N71Q, K86A / N112A / N71D / N79A, K86A / N112A / N71D / N79D, K86A / N112A / N71Q / N79A, K86A / N112A / N71Q / N79D, K86R / N112A / N71D / N79A, K86R / N112A / N71D / N79D, K86R / N112A / N71D / N79Q, K86R / N112A / N71Q / N79A, K86R / N112A / N71Q / N79D, and K86R / N112A / N71Q / N79Q as compared to the parental IL-15. In one embodiment, the IL-15 comprises an amino acid sequence according to any of the amino acid sequences shown in FIG. 3.

[0036] In certain embodiments, the IL-12 is a single-chain IL-12 comprising a p35 subunit, a p40 subunit, and an IL-12 linker, and the IL-12 linker binds the p35 subunit to the p40 subunit.

[0037] In some embodiments, the first linker and the second linker are each independently selected from any of the linkers shown in FIG. 48. In an exemplary embodiment, the linker is (GGGGS) x where x is an integer from 1 to 10.

[0038] In another aspect, provided herein is an ABD fusion protein comprising an albumin binding domain (ABD), a cytokine, and a linker (L) according to formula (Cytokine)-L-(ABD). The ABD includes a variable heavy chain comprising any one of vhCDR1, vhCDR2, and vhCDR3 of the variable heavy chains shown in FIG. 2, and a variable light chain comprising any one of vlCDR1, vlCDR2, and vlCDR3 of the variable light chains shown in FIG. 2, and L is a linker. The cytokine is selected from IL-2, IL-7, IL-12, IL-15, IL-18, IL-21, GM-CSF, and IFN-α is selected.

[0039] In some embodiments, according to any of the vhCDR1 sequences, vhCDR2 sequences, and vhCDR3 sequences shown in FIG. 2, vhCDR1 includes the vhCDR1 sequence, vhCDR2 includes the vhCDR2 sequence, and vhCDR3 includes the vhCDR3 sequence. In some embodiments, the variable heavy chain includes any one of the variable heavy chain sequences shown in FIG. 2.

[0040] In certain embodiments, according to any of the vlCDR1 sequences, vlCDR2 sequences, and vhCDR3 sequences shown in FIG. 2, vlCDR1 includes the vlCDR1 sequence, vlCDR2 includes the vlCDR2 sequence, and vlCDR3 includes the vlCDR3 sequence. In certain embodiments, the variable light chain includes any one of the variable light chain sequences shown in FIG. 2.

[0041] In an exemplary embodiment, the variable heavy chain and the variable light chain include the variable heavy chain and variable light chain of A10m3 (FIG. 2D), respectively.

[0042] In some embodiments, the linker is selected from any of the linkers shown in FIG. 48. In an exemplary embodiment, the linker is (GGGGS) x where x is an integer from 1 to 10.

[0043] In another aspect, an ABD fusion protein of the formula (FP1)-L1-(ABD)-L2-(FP2) is provided herein, where ABD is an albumin binding domain comprising a variable heavy chain and a variable light chain, FP1 and FP2 are a first fusion protein and a second fusion protein, respectively, and L1 and L2 are a first and a second linker, respectively. ABD includes a variable heavy chain comprising any one of vhCDR1, vhCDR2, and vhCDR3 of the variable heavy chains shown in FIG. 2, and a variable light chain comprising any one of vlCDR1, vlCDR2, and vlCDR3 of the variable light chains shown in FIG. 2.

[0044] In some embodiments, according to any of the vhCDR1 sequences, vhCDR2 sequences, and vhCDR3 sequences shown in FIG. 2, vhCDR1 includes the vhCDR1 sequence, vhCDR2 includes the vhCDR2 sequence, and vhCDR3 includes the vhCDR3 sequence. In some embodiments, the variable heavy chain includes any one of the variable heavy chain sequences shown in FIG. 2.

[0045] In certain embodiments, according to any of the vlCDR1 sequences, vlCDR2 sequences, and vhCDR3 sequences shown in FIG. 2, vlCDR1 includes the vlCDR1 sequence, vlCDR2 includes the vlCDR2 sequence, and vlCDR3 includes the vlCDR3 sequence. In certain embodiments, the variable light chain includes any one of the variable light chain sequences shown in FIG. 2.

[0046] In an exemplary embodiment, the variable heavy chain and the variable light chain include the variable heavy chain and the variable light chain of A10m3 (FIG. 2D), respectively.

[0047] In some embodiments, FP1 and FP2 are a first cytokine and a second cytokine, respectively. In an exemplary embodiment, the first cytokine and the second cytokine are IL-2 and IL-12, IL-7 and IL-15, IL-15 and IL-12, IL-18 and GM-CSF, IL-21 and IL-15, GM-CSF and IL-12, GM-CSF and IL-21, and IFN-α and selected from IL-15.

[0048] In some embodiments, the first and second fusion partners are anti-PD-L1 scFv and IL-12, anti-PD-L1 scFv and IL-15, anti-PD-L1 scFv and anti-TGFβ scFv, a first anti-PD-L1 scFv and a second PD-L1 scFv, anti-TGFβ scFv and IL-12, anti-TGFβ scFv and IL-15, anti-TGFβ scFv and PD-L1 scFv, and a first anti-TGFβ scFv and a second anti-TGFβ scFv.

[0049] The ABD fusion protein according to claim 68, wherein FP1 and FP2 are a first binding moiety and a second binding moiety, respectively. In certain embodiments, the first binding moiety and the second binding moiety are each an scFv. In an exemplary embodiment, the first binding moiety and the second binding moiety are selected from TNF scFv and IL-1 scFv, TNF scFv and IL-6 scFv, TNF scFv and IL-8 scFv, TNF scFv and IL-17 (isoforms A-F) scFv, TNF scFv and IL-23 scFv, and a first TNF scFv and a second TNF scFv.

[0050] In some embodiments, the first linker and the second linker are each independently selected from any of the linkers shown in FIG. 48. In one embodiment, the first linker and the second linker are each independently (GGGGS) xwhere x is an integer from 1 to 10.

[0051] In another aspect, an albumin-binding domain (ABD) fusion protein comprising a TGFβ-binding domain and an albumin-binding domain is provided herein. The albumin-binding domain includes an ABD variable heavy chain and an ABD variable light chain having the amino acid sequence of either the variable heavy chain and variable light chain of Figure 2.

[0052] In an exemplary embodiment, the ABD variable heavy chain and the ABD variable light chain include the amino acid sequences of the variable heavy chain and variable light chain of A10m3.

[0053] In one embodiment, the TGFβ-binding domain is a scFv (Figure 40B) comprising the variable heavy chain and variable light chain of 4D9. In some embodiments, the ABD further includes an IL-12, IL-15, PD-L1-binding domain, or a second TGFβ-binding domain.

[0054] In another aspect, an albumin-binding domain (ABD) fusion protein comprising a PD-L1-binding domain and an albumin-binding domain is provided herein. The albumin-binding domain includes an ABD variable heavy chain and an ABD variable light chain having the amino acid sequence of either the variable heavy chain and variable light chain of Figure 2.

[0055] In some embodiments, the ABD variable heavy chain and the ABD variable light chain include the amino acid sequences of the variable heavy chain and variable light chain of A10m3 (Figure 2D). In certain embodiments, the PD-L1-binding domain is a scFv (Figure 50) comprising the variable heavy chain and variable light chain of 10D12.

[0056] In some embodiments, the ABD further includes an IL-12, IL-15, TGFβ-binding domain, or a second PD-L1-binding domain.

[0057] In another aspect, provided herein are nucleic acids encoding any of the albumin binding domains, variant IL-15, or ABD fusion proteins described herein, host cells comprising any such nucleic acid, and methods of making such ABDs, variant IL-15, or ABD fusion proteins. In yet another aspect, provided herein is a method of suppressing or reducing a tumor in a subject in need thereof, the method comprising administering an ABD fusion protein to the subject.

[0058] In yet another aspect, provided herein is a method of suppressing or reducing a tumor in a subject in need thereof, the method comprising administering an ABD fusion protein to the subject. BRIEF DESCRIPTION OF THE DRAWINGS

[0059]

FIG. 1A-C

FIG. 2A-2Z

FIG. 3

FIG. 4

FIG. 5

FIG. 6

FIG. 7

FIG. 8

FIG. 9

FIG. 10

FIG. 11

FIG. 12

FIG. 13

FIG. 14

FIG. 15

FIG. 16

FIG. 17-18

FIG. 19A-B

FIG. 20

FIG. 21-22

FIG. 23

FIG. 24

FIG. 25

FIG. 26

FIG. 27

FIG. 28

FIG. 29

FIG. 30

FIG. 31

FIG. 32

FIG. 33

FIG. 34

FIG. 35

FIG. 36

FIG. 37-38

FIG. 39

FIG. 40A-B

FIG. 41

FIG. 42-43

FIG. 44

FIG. 45A

FIG. 45B

FIG. 46

FIG. 47

FIG. 48

FIG. 49A-G

FIG. 50

FIG. 51A-B

Modes for Carrying Out the Invention

[0060] A.Summary Biologics, including cytokines and antibody-based therapeutics, are useful in the treatment of cancer.

[0061] Current and potential cytokine-based therapeutics include those utilizing IL-2, IL-7, IL-12, IL-15, IL-18, IL-21 GM-CSF, and IFN-α.

[0062] IL-12 can mediate immune effector functions in a manner compatible with enhancing the inflammatory-induced endogenous antitumor immune response. (See, for example, Boggio et al., J Exp Med 188:589-96 (1998), Cavallo et al., Cancer Res 59:414-21 (1999), Yu et al., Int Immunol 8:855-65 (1996), Nastala et al., J Immunol 153:1697:706 (1994), Brunda et al., J Exp Med 178:1223-30 (1993).) IL-12 is known to induce an inflammatory Th1 CD4+ T cell response and enhance the cytotoxicity of CD8+ T cells. Studies have also shown that T cell secretion of IFNγ mediated by IL-12 can reverse T cell anergy and confer effector T cell resistance to immunosuppressive regulatory T cells. IL-12 is an ideal candidate for tumor immunotherapy due to its ability not only to activate the adaptive and innate immune systems but also to further modulate the tumor microenvironment hostile to the immune system.

[0063] IL-15 can stimulate T cell proliferation within tumors (e.g., Miecnik (see, e.g., et al., Sci Transl Med 6(228):228ra37(2014)), which extends the viability of effector memory CD8+ T cells and is important for NK cell development. Thus, IL-15 is thought to enhance the efficacy of checkpoint inhibitors and other immunotherapies that utilize T cells to attack cancer cells. However, the half-life of IL-15 monomer in vivo is short, less than 40 minutes. Modification of IL-15 monomer may improve its pharmacokinetics in cancer treatment in vivo. These modifications generally focus on improving the trans-presentation of IL-15 using IL-15Rα, the α subunit of the IL-15 receptor. Such modifications include: 1) pre-association of IL-15 with its soluble receptor a-subunit-Fc fusion to form an IL-15:IL-15Rα-Fc complex (see, e.g., Rubinstein et al., Proc Natl Acad Sci USA. 103:9166-71(2006)), 2) expression of hyperagonist IL-15-sIL-15Rα-sushi protein (see, e.g., Bessard et al., Molecular cancer therapeutics 8:2736-45(2009)), and 3) pre-association of human IL-15 mutant IL-15N72D with an IL-15Rα-Fc sushi-Fc fusion complex (see, e.g., Zhu et al., Journal of Immunology 183:3598-6007(2009)).

[0064] Monoclonal antibody-based therapies that target tumor surface antigens that limit T cell activation and inhibitory signals have been standard components of cancer treatment regimens for over 20 years. For example, see Weiner, Nat Rev Cancer 15(6):361-370(2015 )

[0065] Short circulating half-lives are a major obstacle for many biologic agents, including cytokines and antibody-based therapeutics. See, for example, Herrington-Symes et al., Advances in Bioscience and Biotechnology 4:689-698 (2013), and Perdreau et al., European Cytokine Network 21:297-307 (2010). Such short-acting therapeutic agents require frequent dosing profiles that can reduce their applicability to the clinic, especially in the case of chronic diseases. A long serum half-life is desirable to reduce the need for repeated injections of molecules to achieve therapeutically appropriate serum concentrations. Methods for extending the half-life of therapeutic proteins include PEGylation, fusion to human serum albumin (HSA), fusion to the constant fragment (Fc) of human immunoglobulin IgG, and fusion to unstructured fusion proteins such as XTEN. See, for example, Stohl, BioDrugs 29(4):215-239 (2015). Half-life extension technologies enable improved or novel biologic therapies that reduce the cost and burden of frequent dosing. Accordingly, there is a continuing need for novel reagents and methods that can extend the half-life of protein and peptide-based therapeutics.

[0066] Novel albumin-binding domain (ABD) fusion proteins useful for extending the half-life of biologic agents (e.g., interleukins and antibodies) are provided herein. Serum albumin has a long half-life in the range of 2 to 4 weeks by recycling via the neonatal Fc receptor (FcRn). Albumin is taken up by endothelial cells via macropinocytosis and binds to FcRn in a pH-dependent manner within the acidic environment of early endosomes. Albumin-FcRn binding diverts albumin molecules from degradation within the lysosomal compartment, redirects the albumin molecules towards the plasma membrane, where they are returned to the plasma by neutral pH.

[0067] The albumin-binding domain (ABD) described herein does not compete with FcRn for albumin binding and binds to albumin at a pH range that also allows the ABD to undergo FcRn-driven endosomal albumin recycling when bound to albumin. Thus, a biologic agent containing the subject albumin-binding domain (ABD) can avoid lysosomal degradation using the albumin-FcRn pathway and, as a result, exhibits a longer serum half-life than its counterpart lacking the ABD.

[0068] Furthermore, such ABD-containing therapeutics advantageously localize to tumors known to contain high levels of serum albumin. Thus, such ABD-containing therapeutics are particularly useful for the treatment of cancer.

[0069] B. Albumin-Binding Domain In one aspect, a composition comprising an albumin-binding domain is provided herein. As used herein, "serum albumin" refers to a member of a family of globular proteins produced by the liver that functions primarily as a carrier protein for steroids, fatty acids, and thyroid hormones in the blood. Serum albumin also plays a major role in stabilizing extracellular fluid volume by contributing to the colloid osmotic pressure of plasma and includes, but is not limited to, human serum albumin (HSA, Genbank accession numbers: NM_000477 and NP_000468), and mouse serum albumin (MSA, Genbank accession numbers: NM_009654 and NP_0033784). The structure of albumin is characterized by several long α-helices and contains 11 different binding domains for hydrophobic compounds. In humans, serum albumin is encoded by the ALB gene.

[0070] Fusion proteins containing the subject albumin binding domain (ABD) can bind to serum albumin (SA), thereby allowing the fusion protein to be taken up by cells via macropinocytosis. In certain embodiments, the ABDs described herein bind in a pH range of about pH 5.5 to about pH 7.2. In early endosomes, such SA-binding ABD fusion proteins bind to FcRn via SA at acidic pH (e.g., pH 5.5), which then diverts the SA-binding ABD fusion protein from the lysosomal compartment of the cell and back to the plasma membrane. At the plasma membrane, SA dissociates from FcRn due to neutral pH (e.g., pH 7.1 - 7.5), and SA and the ABD fusion protein are released and return to the bloodstream. Therapeutics containing the subject ABD can bind to albumin in the pH range of about pH 5.5 to about pH 7.2, such that such therapeutics advantageously also undergo FcRn-driven endosomal albumin recycling and thus avoid lysosomal degradation. Accordingly, therapeutics containing such ABDs advantageously exhibit a longer serum half-life than their counterparts lacking the ABD. Such therapeutics are particularly useful for the treatment of cancers known to contain high levels of serum albumin.

[0071] In some embodiments, the albumin binding domain binds to albumin at a site that does not interfere with the binding of SA to the neonatal Fc receptor (FcRn). As used herein, "FcRn" or "neonatal Fc receptor" refers to a protein that associates with the Fc region of an IgG antibody and is at least partially encoded by the FcRn gene. FcRn can be from any organism including, but not limited to, human, mouse, rat, rabbit, and monkey. As is known in the art, functional FcRn proteins often contain two fusion proteins often referred to as the heavy chain and the light chain. The light chain is β-2-microglobulin and the heavy chain is encoded by the FcRn gene. Unless otherwise stated herein, FcRn or FcRn protein refers to a complex of the FcRn heavy chain and β-2-microglobulin.

[0072] In some embodiments, the albumin-binding domain described or exemplified herein specifically binds to serum albumin (e.g., HSA) at an epitope on the serum albumin molecule that preferably does not participate in the interaction between the serum albumin molecule and FcRn. Thus, the binding of the SA-binding moiety to the serum albumin molecule preferably does not substantially interfere with, inhibit, prevent, or otherwise reduce the binding of the serum albumin molecule (e.g., HSA) to FcRn. Preferably, the albumin-binding domain does not compete with FcRn for binding to the serum albumin molecule. Preferably, the albumin-binding domain does not sterically inhibit the binding of serum albumin to FcRn. Preferably, the SA-binding moiety does not change the conformation of the serum albumin molecule such that albumin cannot interact with FcRn.

[0073] In some embodiments, the albumin-binding domain binds to SA (e.g., HSA) at a pH of 5.0±0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8, 2.9, or 3.0. In some embodiments, the albumin-binding domain binds to SA in a pH range of about pH 5.5 to about pH 7.2. In some embodiments, the SA-binding moiety binds to SA at a pH of 5.5.

[0074] In certain embodiments, the albumin-binding domain is a human serum albumin (HSA)-binding domain. The HSA-binding domain includes, but is not limited to, albumin-binding domains that can bind to an HSA molecule such as the entire HSA molecule or a fragment of HSA. In some aspects, the HSA-binding domain also binds to mouse serum albumin. In some aspects, the HSA-binding domain also binds to cynomolgus albumin. In certain embodiments, the HSA-binding domain does not bind to bovine serum albumin (BSA). No.

[0075] The albumin-binding domains provided herein can include a variable heavy chain alone or a variable heavy chain in combination with a variable light chain. In some embodiments, the albumin-binding domain includes a variable heavy chain. In certain embodiments, the variable heavy chain includes vhCDR1, vhCDR2, and vhCDR3 (variable heavy chain complementarity determining regions 1-3). In certain embodiments, the antigen-binding domain also includes a variable light chain. In certain embodiments, the variable light chain includes vlCDR1, vlCDR2, and vlCDR3 (variable light chain complementarity determining regions 1-3).

[0076] In some embodiments, the albumin-binding domain includes a variable heavy chain that includes any of vhCDR1, vhCDR2, and vhCDR3 of the variable heavy chains shown in FIG. 2. In some embodiments, the albumin-binding domain includes vhCDR1, vhCDR2, and vhCDR3 of the A10m3 variable heavy chain as shown in FIG. 2D. In certain embodiments, the albumin-binding domain includes vhCDR1, vhCDR2, and vhCDR3 of A10m3 as shown in FIG. 2D.

[0077] In certain embodiments, the albumin-binding domain also includes a variable light chain. In an exemplary embodiment, the albumin-binding domain includes a variable light chain that includes any of vlCDR1, vlCDR2, and vlCDR3 of the variable light chains shown in FIG. 2. In some embodiments, the albumin-binding domain includes vlCDR1, vlCDR2, and vlCDR3 of the A10m3 variable light chain as shown in FIG. 2D. In certain embodiments, the albumin-binding domain includes vlCDR1, vlCDR2, and vlCDR3 of A10m3 as shown in FIG. 2D.

[0078] In certain embodiments, the albumin-binding domain (e.g., HSA-binding domain) is an antibody or an antibody fragment. In some embodiments, the albumin-binding domain (e.g., HSA-binding domain) is a scFv.

[0079] In some embodiments where the ABD includes both a variable heavy chain and a variable light chain, the variable heavy chain and the variable light chain are joined to each other by a linker (e.g., an scFv linker). In certain embodiments, the linker is joined to the variable heavy chain at its C-terminus and to the variable light chain at its N-terminus. Suitable linkers are described herein and in Figure 48. In some embodiments, the linker is (Gly4Ser) x is a linker, and x is 1, 2, 3, 4, 5, 6, 7, or 8. In certain embodiments, the linker is a (Gly4Ser)5 linker.

[0080] In certain embodiments, the albumin binding domain also includes a variable heavy chain comprising vhCDR1, vhCDR2, and vhCDR3 of A10m3, and a variable light chain comprising vlCDR1, vlCDR2, and vlCDR3 of A10m3 (Figure 2D). In one embodiment, the albumin binding domain includes the variable heavy and variable light sequences of the A10m3 ABD shown in Figure 2D.

[0081] C. Interleukin-15 Variants In another aspect, compositions are provided herein that include variant IL-15 having improved in vivo stability and / or biological activity as compared to wild-type IL-15.

[0082] As used herein, "interleukin 15", "IL-15", and "IL15" all refer to the interleukin that binds to and signals through a complex consisting of the IL-15 specific receptor alpha chain, the IL-2 / IL-15 receptor beta chain (CD122), and the common gamma chain (γC, CD132) (Genbank accession numbers: NM_00000585 and NP_000576 (human), and NM_001254747 and NP_001241676 (mouse)). No. 18 refers to the interleukin that emits signals (Genbank accession numbers: NM_00000585 and NP_000576 (human), and NM_001254747 and NP_001241676 (mouse)).

[0083] IL-15 has been shown to stimulate T cell proliferation within tumors (see, e.g., Miecnik et al., Sci Transl Med 6(228):228ra37 (2014)). IL-15 can also extend the viability of effector memory CD8+ T cells and is important for NK cell development. IL-15 is thought to enhance the efficacy of checkpoint inhibitors and other immunotherapies that utilize T cells to attack cancer cells. Thus, without being bound by a particular theory of operation, the IL-15 described herein is considered useful for the treatment of cancer.

[0084] However, the half-life of IL-15 monomer in vivo is short, less than 40 minutes. Modification of IL-15 monomer can improve its pharmacokinetics in vivo for cancer treatment. These modifications generally focus on improving the trans-presentation of IL-15 using the α subunit of the IL-15 receptor, IL-15Rα. Such modifications include 1) pre-association of IL-15 with its soluble receptor a-subunit-Fc fusion to form an IL-15:IL-15Rα-Fc complex (see, e.g., Rubinstein et al., Proc Natl Acad Sci USA. 103:9166-71 (2006)), 2) expression of hyperagonist IL-15-sIL-15Rα-sushi protein (see, e.g., Bessard et al., Molecular cancer therapeutics 8:2736-45 (2009)), and 3) pre-association of human IL-15 mutant IL-15N72D with an IL-15Rα-Fc sushi-Fc fusion complex (see, e.g., Zhu et al., Journal of Immunology 183:3598-6007 (2009)).

[0085] In some embodiments, IL-15 is a variant of parental IL-15 with improved stability compared to wild-type IL-15. In certain embodiments, the variant IL-15 is a variant of wild-type human IL-15. In an exemplary embodiment, the variant IL-15 includes an amino acid substitution at position K86 of the parental IL-15 shown in Figure 3. As described herein, K86 is a putative site of ubiquitin-dependent degradation when made using certain cell types (e.g., HEK293 T cells) (see Example 2). Thus, without being bound by a particular theory of operation, it is believed that removal of the K86 ubiquitination site by amino acid substitution improves the stability of IL-15 (see Examples 2 and 3).

[0086] In certain embodiments, IL-15 is a variant IL-15 having an amino acid substitution at position N112. Amino acid position N112 is an important site for IL-15 bioactivity as it is important for proper IL-15 / IL-15 receptor γ interaction, particularly when IL-15 is bound to ABD. Thus, without being bound by a particular theory of operation, variants at position N112 are thought to be able to enhance one or more functions of IL-15 including, but not limited to, promoting T cell proliferation in the tumor environment, enhancing the viability of CD8+ T cells, and promoting the development of NK cells.

[0087] Certain amino acid substitutions that can improve the in vivo stability and / or bioactivity of IL-15 include K86A, K86R, N112A, N112S, N112Q, K86A / N112A, K86R / N112A, K86A / N112S, K86R / N112S, K86A / N112Q, K86R / N112Q, K86A / N112A / N79A, K86R / N112A / N79A, K86A / N112A / N79D, K86R / N112A / N79D, K86A / N112A / N79Q, K86R / N112A / N79Q, K86A / N112A / N71D, K86R / N112A / N71D, K86A / N112A / N71Q, K86R / N112A / N71Q, K86A / N112A / N71D / N79A, K86A / N112A / N71D / N79D, K86A / N112A / N71Q / N79A, K86A / N112A / N71Q / N79D, K86R / N112A / N71D / N79A, K86R / N112A / N71D / N79D, K86R / N112A / N71D / N79Q, K86R / N112A / N71Q / N79A, K86R / N112A / N71Q / N79D, and K86R / N112A / N71Q / N79Q, including but not limited to. An exemplary variant IL-15 containing one or more of such amino acid substitutions is shown in FIG. 3. In an exemplary embodiment, the variant IL-15 contains the amino acid substitutions K86A and N112A.

[0088] In one embodiment, the IL-15 (wild-type and variant IL-15) described herein is bound to IL-15Rα. Such IL-15 presented in trans with its receptor has been shown to have an extended half-life and higher potency compared to native IL-15 alone. See, for example, Wu, J Mol Genet Med 7, 85 (2013).

[0089] D. IL-12 In another aspect, a composition comprising IL-12 is provided herein. As used herein, "Interleukin 12", "IL-12", and "IL12" all refer to Interleukin, a heterodimeric cytokine encoded by the IL-12A and IL-12B genes (Genbank accession numbers: NM_000882 (IL-12A) and NM_002187 (IL-12B)). IL-12 is composed of a bundle of four α-helices and is involved in the differentiation of native T cells into TH1 cells. IL-12 binds to the IL-12 receptor, which is a heterodimeric receptor formed by IL-12R-β1 and IL-12R-β2. IL-12 is known as a T cell stimulatory factor that can stimulate the proliferation and function of T cells. Specifically, IL-12 can stimulate the production of interferon γ (IFN-γ) and tumor necrosis factor α (TNF-α) from T cells and natural killer (NK) cells, and reduce the IL-4-mediated suppression of IFN-γ. IL-12 can further mediate the enhancement of the cytotoxic activity of NK cells and CD8+ cytotoxic T lymphocytes. Furthermore, IL-12 can also have anti-angiogenic activity by increasing the production of interferon γ, which in turn increases the production of chemokine-induced protein-10 (IP-10 or CXCL10). IP-10 then mediates this anti-angiogenic effect. Without being bound by a particular theory of operation, it is believed that IL-12, through its ability to induce an immune response and anti-angiogenic activity, can be used to treat cancer.

[0090] In some embodiments, IL-12 is murine IL-12. In other embodiments, IL-12 is human IL-12.

[0091] In certain embodiments, IL-12 is a single-chain IL-12 polypeptide comprising an IL-12 p35 subunit linked to an IL-12 p40 subunit. Such an IL-12 single-chain polypeptide advantageously retains one or more of the biological activities of wild-type IL-12. In some embodiments, the single-chain IL-12 polypeptides described herein are, from the N-terminus to the C-terminus, of the formula (p40)-(L)-(p35), wherein "p40" is the IL-12 p40 subunit, "p35" is the IL-12 p35 subunit, and L is a linker. In other embodiments, the single-chain IL-12 is, from the N-terminus to the C-terminus, of the formula (p35)-(L)-(p40). Any suitable linker can be used within the single-chain IL-12 peptide as described herein and disclosed in FIG. 49C. Suitable linkers can include, for example, an amino acid sequence (GGGGS) x having, wherein x is an integer from 1 to 10. Other suitable linkers can include, for example, the amino acid sequence GGGGGGS. Exemplary single-chain IL-12 linkers that can be used with the subject single-chain IL-12 polypeptides are also described in Lieschke et al., Nature Biotechnology 15:35-40 (1997), which is hereby incorporated by reference in its entirety for its teachings regarding IL-12 polypeptide linkers in particular.

[0092] In an exemplary embodiment, the single-chain IL-12 polypeptide is a single-chain human IL-12 polypeptide (i.e., comprising human p35 and p40 IL-12 subunits). In certain embodiments, the single-chain IL-12 polypeptide is a single-chain mouse IL-12 polypeptide. Exemplary single-chain human and mouse IL-12 are shown in FIGS. 20 (shown as a fusion peptide with ABD) and 49C.

[0093] E.ABD fusion protein In one aspect, provided herein is an ABD composition comprising an albumin-binding domain coupled via a linker to one or more fusion partners (e.g., a first fusion partner, a second fusion partner, etc.). As discussed herein, the subject ABD fusion proteins can exhibit an extended half-life under FcRn-mediated endosomal recycling and, thus advantageously, compared to their counterparts that do not contain such ABDs.

[0094] Useful ABDs for such ABD fusion proteins include, but are not limited to, those described herein. The amino acid sequences of such ABDs, which include vhCDR1-3, vlCDR1-3, variable heavy chains, and variable light chain sequences, are disclosed, for example, in FIG. 2. In some embodiments, the ABD fusion protein includes a variable heavy chain comprising any of vhCDR1, vhCDR2, and vhCDR3 of the ABD variable heavy chain of FIG. 2, and a variable light chain comprising any of vlCDR1, v1CDR2, and vCDR3 of the ABD variable light chain of FIG. 2. In certain embodiments, the ABD includes a variable heavy chain having vhCDR1, vhCDR2, and vhCDR3 of the ABD disclosed in FIG. 2, and a variable light chain having vhCDR1, vhCDR2, and vhCDR3 of the ABD disclosed in FIG. 2. In some embodiments, the ABD fusion protein includes the variable heavy chain and variable light chain of the ABD disclosed in FIG. 2.

[0095] In an exemplary embodiment, the ABD fusion protein includes a variable heavy chain comprising vhCDR1, vhCDR2, and vhCDR3 of the A10m3 variable heavy chain, and a variable light chain comprising vlCDR1, vlCDR2, and vCDR3 of the A10m3 variable light chain (FIG. 2D). In certain embodiments, the ABD includes a variable heavy chain having vhCDR1, vhCDR2, and vhCDR3 of A10m3, and a variable light chain having vhCDR1, vhCDR2, and vhCDR3 of A10m3 (FIG. 2D). In some embodiments, the ABD fusion protein includes the variable heavy chain and variable light chain of A10m3 (FIG. 2D).

[0096] The ABD fusion proteins described herein include a fusion partner. In some embodiments, the fusion partner includes two fusion partners (a first fusion partner (FP1) and a second fusion partner (FP2)). In embodiments that include two fusion partners, the fusion partner can bind to the ABD in several orientations. In some embodiments, the ABD fusion protein is from the N-terminus to the C-terminus, according to the formula FP1-ABD-FP2, FP1-PF2-ABD or ABD-FP1-FP2, wherein FP1 is the first fusion partner and FP2 is the second fusion partner.

[0097] Any suitable fusion partner for which an extension of the half-life is desired can be included in the subject ABD fusion protein. The fusion partner can include, for example, cytokines (such as interferons and interleukins), growth factors, polypeptides, proteins, and hormones (such as growth hormone, parathyroid hormone).

[0098] In certain embodiments, the fusion partner is an antibody-derived binding portion that includes a variable heavy chain and a variable light chain. Such a binding portion can bind to any target of interest, including, for example, a tumor-specific target or a cytokine. In an exemplary embodiment, the fusion partner is a single-chain variable fragment (scFv). Antibody-derived fusion partners also include, but are not limited to, ds-scFv, single-domain antibodies (sdAb), diabodies, dsFv, ds-scFv, Fab, and full-length antibodies. Antibody-derived fusion partners also include multispecific (e.g., bispecific) antibodies and fragments.

[0099] The ABD fusion proteins of the subject matter described herein utilize linkers between components (ABD and fusion partner) and within components. For example, scFv fusion partners generally utilize standard peptide linkers based on glycine and serine to join the mutant heavy and light chains to form the scFv. Further, standard peptide linkers are utilized to join the ABD to the fusion partner (e.g., cytokine fusion partner). Further, linkers are used to join components of certain moieties, such as the p35 and p40 subunits of IL-12 and IL-15, to IL-15Rα. In some embodiments, the linker is (Gly4Ser) x is a linker, and x is 1, 2, 3, 4, 6, 7, or 8. In certain embodiments, the ABD is connected to the fusion partner by a (Gly4Ser)5 linker.

[0100] As shown herein, there are a number of suitable linkers that can be used, including conventional peptide linkages produced by recombinant techniques. Linker peptides can mainly contain the following amino acid residues: Gly, Ser, Ala, or Thr. Linker peptides should have a length sufficient to join the two molecules in such a way that they adopt the correct conformation relative to each other so as to retain the desired activity. In one embodiment, the linker is about 1 to 50 amino acids in length, preferably about 1 to 30 amino acids in length. In one embodiment, a linker of 1 to 20 amino acids in length may be used, and in some embodiments, about 5 to about 10 amino acids are useful. Useful linkers include, for example, glycine-serine polymers such as (GS)n, (GSGGS)n, (GGGGS)n, and (GGGS)n (where n is an integer of at least 1 (and generally 3 to 4)), glycine-alanine polymers, alanine-serine polymers, and other flexible linkers. Alternatively, various non-proteinaceous polymers including, but not limited to, polyethylene glycol (PEG), polypropylene glycol, polyoxyalkylene, or copolymers of polyethylene glycol and polypropylene glycol can be useful as linkers, i.e., can be useful as linkers.

[0101] Other linker sequences include any sequence of CL / CH1 domains of any length, but may not include all residues of the CL / CH1 domain, for example, the first 5-12 amino acid residues of the CL / CH1 domain. The linker can be derived from an immunoglobulin light chain, such as Cκ or Cλ. The linker can be derived from an immunoglobulin heavy chain of any isotype, including, for example, Cγ1, Cγ2, Cγ3, Cγ4, Cα1, Cα2, Cδ, Cε, and Cμ. The linker sequence can also be derived from other proteins such as Ig-like proteins (e.g., TCR, FcR, KIR), sequences derived from hinge regions, and other natural sequences derived from other proteins.

[0102] Often, the linker is not substantially sensitive to the extracellular environment. As used in the context of the linker herein, "not substantially sensitive to the extracellular environment" means that no more than about 20%, 15%, 10%, 5%, 3%, or about 1% of the linker in a sample of the antibody-drug conjugate compound is cleaved when the antibody-drug conjugate compound is present in the extracellular environment (e.g., in plasma).

[0103] Whether the linker is not substantially sensitive to the extracellular environment can be determined, for example, by incubating the antibody-drug conjugate compound with plasma for a predetermined time (e.g., 2, 4, 8, 16, or 24 hours) and then quantifying the amount of free drug present in the plasma.

[0104] In other, non-exclusive embodiments, the linker promotes endocytosis. In certain embodiments, the linker promotes endocytosis when conjugated to a therapeutic agent (i.e., in the context of the linker-therapeutic agent moiety of an antibody-drug conjugate compound as described herein). In still other embodiments, the linker promotes endocytosis when conjugated to both an auristatin compound and an ABD fusion protein of the invention.

[0105] A variety of exemplary linkers that can be used in the present compositions and methods are described in International Patent Application No. 2004 / 010957, U.S. Patent Publication No. 2006 / 0074008, U.S. Patent Publication No. 2005 / 0238649, and U.S. Patent Publication No. 2006 / 0024317, each of which is hereby incorporated by reference in its entirety for all purposes.

[0106] Exemplary linkers that can be utilized with the subject ABD as binding components of domain linkers, scFv linkers, and certain fusion partners are further shown in FIG. 48.

[0107] In some embodiments, a linker is a "domain linker" used to covalently join any two domains as outlined herein (e.g., a cytokine fusion partner (e.g., an interleukin) and an ABD). Any suitable linker can be used, and many embodiments utilize glycine - serine polymers such as (GS)n, (GSGGS)n, (GGGGS)n, and (GGGS)n where n is at least 1 (and generally 3 - 4 - 5), and any peptide sequence that allows for recombinant joining of two domains with sufficient length and flexibility such that each domain can retain its biological function.

[0108] In some embodiments, the fusion partner is an scFv comprising a variable heavy chain and a variable light chain. In such embodiments, the ABD variable heavy chain is linked to the variable light chain using an scFv linker.

[0109] Exemplary ABD fusion proteins are further discussed below.

[0110] 1. Cytokine - ABD fusion proteins In some embodiments, the ABD fusion protein includes a cytokine fusion partner, namely, a cytokine-albumin binding domain (cytokine-ABD) fusion protein (Figure 1). In some embodiments, the cytokine-ABD fusion protein includes IL-2, IL-7, IL-12, IL-15, IL-18, IL-21, IL-27, GM-CSF or IFN-α. The subject cytokine-ABD fusion protein is useful for immunomodulatory effects in a subject in which such immunomodulatory effects are needed (e.g., for the treatment of cancer or autoimmune diseases). Further, the subject cytokine-ABD has a longer half-life and improved pharmacokinetic properties compared to cytokine therapeutic agents alone. shows a longer half-life and improved pharmacokinetic properties compared to cytokine therapeutic agents alone.

[0111] Any ABD can be used together with the subject cytokine-ABD fusion protein described herein. In some embodiments, the ABD includes an ABD variable heavy chain. In certain embodiments, the ABD includes an ABD variable light chain. In an exemplary embodiment, the ABD is a scFv comprising a variable heavy chain bound to a variable light chain by a linker (e.g., any of the linkers disclosed herein and in Figure 48).

[0112] In certain embodiments, the variable heavy chain includes any of the ABD variable heavy chains described herein, including these ABD variable heavy chains shown in Figure 2, of vhCDR1-3. In certain embodiments, the ABD variable heavy chain includes vhCDR1-3 of the A10m3 variable heavy chain (Figure 2D). In some embodiments, the ABD variable heavy chain includes A10m3 vhCDR1-3 as shown in Figure 2D. In an exemplary embodiment, the ABD variable heavy chain has the amino acid sequence of the A10m3 variable heavy chain.

[0113] In certain embodiments, the ABD includes an ABD variable light chain. In some embodiments, the ABD variable light chain includes any of the vlCDR1-3 of the ABD variable light chains described herein, including those shown in Figure 2. In certain embodiments, the ABD variable light chain includes the vlCDR1-3 of the A10m3 variable light chain (Figure 2D). In some embodiments, the ABD variable light chain includes A10m3 vlCDR1-3 as shown in Figure 2D. In an exemplary embodiment, the ABD variable light chain has the sequence of the A10m3 variable light chain.

[0114] The amino acid sequences of exemplary cytokines that can be used in the subject cytokine-ABD fusions, as well as exemplary cytokine-ABD fusion proteins where the ABD is A10m3, are shown in Figures 49A-G.

[0115] In some embodiments, the IL-ABD is from the N-terminus to the C-terminus by the formula cytokine-L-ABD or ABD-L-cytokine, where L is a linker (e.g., a peptide linker) that binds the cytokine to the ABD. In certain embodiments, the cytokine-ABD includes a variable heavy chain, and the cytokine is bound to the N-terminus of the ABD variable heavy chain. In some embodiments, the cytokine is bound to the C-terminus of the ABD variable heavy chain. In an exemplary embodiment, the interleukin-ABD includes an ABD that also includes a variable light chain (e.g., an ABD scFv). In some embodiments, the cytokine is bound to the C-terminus of the variable light chain. In certain embodiments, the N-terminus of the cytokine is bound to the ABD. In other embodiments, the C-terminus of the cytokine is bound to the ABD.

[0116] In certain embodiments, the cytokine is a cytokine selected from IL-2, IL-7, IL-12, IL-15, IL-18, IL-21, IL-27, GM-CSF, and IFN-α. Cytokine molecules include, for example, full-length cytokines, and cytokine fragments, such as portions important for the function of a particular cytokine (e.g., the portion of an interleukin that binds to its receptor).

[0117] In some embodiments, the cytokine-ABD comprises an IL-2 molecule or a fragment thereof. As used herein, "interleukin 2", "IL-2", and "IL2" refer to members of the cytokine having four α-helix bundles that signal through the IL-2 receptor (Genbank accession numbers: NM_000586 and NP_000577 (human), and NM_008366 and NP_032392 (mouse)). IL-2 plays an important role in the function, tolerance, and immunity of the immune system, mainly through its direct effect on T cells. In the thymus, IL-2 prevents autoimmune diseases by promoting the differentiation of certain immature T cells into regulatory T cells, which kill other T cells that are primed to attack normal healthy cells in the body. IL-2 has been used in the treatment of cancer (malignant melanoma, renal cell carcinoma) with high-dose intermittent administration and is widely used with continuous administration. An exemplary IL-2-ABD is shown in FIG. 49A. plays an important role in immunity. In the thymus, IL-2 prevents autoimmune diseases by promoting the differentiation of certain immature T cells into regulatory T cells, which kill other T cells that are primed to attack normal healthy cells in the body. IL-2 has been used in the treatment of cancer (malignant melanoma, renal cell carcinoma) with high-dose intermittent administration and is widely used with continuous administration. An exemplary IL-2-ABD is shown in FIG. 49A.

[0118] In some embodiments, the cytokine-ABD fusion protein comprises an IL-7 molecule or a fragment thereof. As used herein, "interleukin 7", "IL-7", and "IL7" (Genbank accession numbers: NM_000880 and NP_000871 (human), and NM_008371 and NP_032397 (mouse)) all refer to members of the cytokine family of hematopoietic growth factors secreted by bone marrow and thymic stromal cells and that bind to the IL-7 receptor. Interleukin-7 (IL-7) is a non-hematopoietic cell-derived cytokine that plays a central role in the adaptive immune system. It promotes lymphocyte development in the thymus, which maintains the homeostatic survival of naive and memory T cells in the periphery. Furthermore, it is important for lymph node (LN) organogenesis and the maintenance of activated T cells that populate secondary lymphoid organs (SLO). IL-7 is an ideal solution for immune reconstitution in immunosuppressed cancer patients by promoting the proliferation of peripheral T cells. In animal models, IL-7 has been shown to extend the survival of tumor-bearing hosts. An exemplary IL-7-ABD is shown in Figure 49B.

[0119] In certain embodiments, the cytokine-ABD comprises an IL-12 molecule or a fragment thereof. In some embodiments, IL-12 is a single-chain IL-12 as shown in Figure 20 (shown as a fusion peptide with ABD) and Figure 49C as described herein.

[0120] In certain embodiments, the cytokine-ABD comprises an IL-15 molecule or a fragment thereof. In some embodiments, the IL-15-ABD comprises the polymorphic IL-15 as described herein and shown in Figures 3 and 4. In some embodiments, IL-15 is wild-type IL-15 as shown in Figure 3 (e.g., "parent IL-15"). In certain embodiments, wild-type IL-15 is bound to IL-15 receptor α (IL-15Rα).

[0121] In certain embodiments, the cytokine-ABD comprises an IL-21 molecule or a fragment thereof. As used herein, "interleukin 21", "IL-21", and "IL21" (Genbank accession numbers: NM_001207006 and NP_001193935 (human), and NM_0001291041 and NP_001277970 (mouse)) all refer to members of the cytokine that bind to the IL-21 receptor and have a potent regulatory effect on cells of the immune system including natural killer (NK) cells and cytotoxic cells, and can destroy virus-infected cells or cancer cells. Thus, without being bound by a particular theory of operation, the cytokine-ABD of the subject matter having IL-21 is considered useful for the treatment of various cancers. An exemplary IL-21-ABD is shown in FIG. 49E.

[0122] Other useful cytokines that may be included in the cytokine-ABD of the subject matter include, but are not limited to, IL-27, IFN-α, and GM-CSF. In certain embodiments, the interleukin-ABD polypeptide also includes an interferon-α, interferon-β, or GM-CSF molecule. Such molecules are useful for delivery to tumors and reduction of toxicity.

[0123] In some embodiments, the cytokine-ABD includes, hereinafter, IL-2 and IL-1 2, IL-7 and IL-15, IL-15 and IL-12, IL-18 and GMC SF, IL-21 and IL-15, GMC-SF and IL-12, GMC-SF and IL-21, and two cytokines bound to the ABD selected from IFN-α and IL-15. FIG. 1 shows some exemplary orientations in which two cytokines can bind to the ABD. Any ABD can be used for such fusion proteins, including those that include either the ABD variable heavy domain or the variable light domain shown in FIG. 2. In some embodiments, the ABD is A10m3.

[0124] Any linker can be used to couple each cytokine to the ABD, which includes any linker shown in Figure 48. An exemplary backbone linker-A10m3-linker sequence for use in such fusion proteins is shown in Figure 50. Exemplary linkers that can be used include (GGGGS) x linkers, where X is from 1 to 10. In certain embodiments, (GGGGS)5 is used to couple the cytokine to the ABD.

[0125] In some embodiments, the cytokine-ABD is of the formula selected from: (from N-terminus to C-terminus) cytokine1-L1-ABD-L2-cytokine2, cytokine1-L1-cytokine2-L2-ABD, and ABD-L1-cytokine1-L2-cytokine2, where L1 and L2 are linkers that connect the cytokine and the ABD components, and cytokine1 and cytokine2 are selected from the following cytokine pairs: IL-2 and IL-12, IL-7 and IL-15, IL-15 and IL-12, IL-18 and GM-CSF, IL-21 and IL-15, GM-CSF and IL-12, GM-CSF and IL-21, and IFN-α and IL-15. In some embodiments, the ABD is A10m3 (Figure 2D).

[0126] In such cytokine-ABDs, either cytokine can be “cytokine1” and the other cytokine can be “cytokine2”. For example, in one embodiment of the IL-2 and IL-12 cytokine pair, IL-2 is “cytokine1” and IL-12 is “cytokine2”. In another embodiment of the IL-2 and IL-12 cytokine pair, IL-12 is “cytokine1” and IL-2 is “cytokine2”.

[0127] In some embodiments, the cytokine-ABD contains two identical interleukins. In other embodiments, the interleukin-ABD polypeptide contains two different interleukins (e.g., IL-12 and IL-15) as described above. ABD fusion proteins containing specific combinations of cytokines are disclosed in Figure 51. Exemplary cytokine-ABD fusion proteins containing different combinations of cytokines are discussed in more detail below.

[0128] a. IL-12 and IL-15 In certain embodiments, the cytokine-ABD fusion protein contains IL-12 and IL-15. Exemplary IL-12 and IL-15 sequences that may be included in such embodiments are shown in Figures 3 and 49C. Cytokine-ABD containing both IL-12 and IL-15 fusion partners is thought to exhibit the anti-tumor effects of both interleukins. The combination of IL-12 and IL-15 has been shown to induce enhanced anti-tumor activity compared to the cytokines alone. Such enhanced anti-tumor activity correlated with the reciprocal upregulation of the receptors of each cytokine via the synergistic induction of IFN-γ. The combination of IL-12 and IL-15 was further shown to promote anti-tumor activity in peritoneal macrophages via the synthesis of nitric oxide. Without being bound by a particular theory of operation, a polypeptide having both IL-12 and IL-15 fusion partners is thought to be able to rapidly activate the natural response (IL-12), strongly stimulate the proliferation of T cells, and maintain memory CD8+ T cells (IL-15). Animal studies have shown that continuous delivery of IL-12- and IL-15-expressing cells also cured mice in an established tumor treatment setting. Depletion of CD8+ cells eliminated this treatment protection, suggesting the expansion of tumor CTL clones. See, for example, Croce et al., Clin Cancer Res 11(2 Pt 1):735-742(2005). See Cancer Res 11(2 Pt 1):735-742(2005).

[0129] In some embodiments, the cytokine-ABD is, from the N-terminus to the C-terminus, according to the following formula.

[0130] a) (IL-12)-L1-(ABD)-L2-(IL-15), or

[0131] b) (IL-15)-L1-(ABD)-L2-(IL-12).

[0132] For example, any suitable ABD can be used that includes an ABD variable heavy chain having any of the vhCDR1-3 of the ABD variable heavy chains listed in Figure 2. In an exemplary embodiment, the variable heavy chain includes vhCDR1-3 of the A10m3 variable heavy chain (Figure 2D). In certain embodiments, the ABD includes an ABD variable light chain having vlCDR1-3 of the A10m3 variable light chain. In an exemplary embodiment, the ABD is an A10m3 scFv.

[0133] Any suitable IL-12 and IL-15 can be used. In some embodiments, IL-15 is a variant IL-15 as described herein (see, for example, Figure 3). IL-15 can also be wild-type IL-15 or wild-type IL-15 bound to IL-15Rα. In one embodiment, IL-15 is a variant IL-15 selected from those shown in Figure 3. In an exemplary embodiment, Il-15 is a variant IL-15 having the amino acid substitutions K86R and N112A.

[0134] IL-12 that can be used includes these IL-12s having p35 and p40 domains. In some embodiments, IL-12 is a single-chain IL-12 as described herein (see, for example, Figure 20 shown as part of an IL-12-ABD fusion polypeptide, and Figure 49C).

[0135] In such embodiments, L1 and L2 are first and second linkers, respectively, and L1 and L2 can be any linkers suitable for connecting the IL-15 and IL-12 domains to the ABD domain (e.g., the linkers listed in Figure 48). Exemplary linkers that can be used include (GGGGS) x A linker is included, wherein X is 1 to 10. In certain embodiments, L1 and L2 are each (GGGGS)5.

[0136] The sequences of exemplary ABD polypeptides, including IL-12 and IL-15, are shown in FIGS.

[0137] b.IL-2 and IL-12 In certain embodiments, the cytokine-ABD includes IL-2 and IL-12.

[0138] IL-2 and IL-12 mutually upregulate each other's receptors and signal differently. IL-2 and IL-12 induce different but complementary biological effects using signal transduction pathways. Both IL-2 and IL-12 can stimulate mitogen- or CD3-activated T cells to proliferate and produce IFN-γ. Furthermore, studies have shown that delivery of both IL-2 and IL-12 genes to mice bearing B16 melanoma induced a significant reduction in tumor burden and enhanced overall survival (see, e.g., Dietrich et al., Arch Surg 387(34):177-182 (2002)). Therefore, cytokine-ABD fusion proteins containing both IL-2 and IL-12 are believed to be useful for tumor reduction and cancer treatment.

[0139] c.IL-2 and IL-15 In certain embodiments, the cytokine-ABD fusion proteins include IL-2 and IL-15. Both IL-2 and IL-15 can stimulate the proliferation of NK cells and activated T cells and support the proliferation of effector T cells. Cytokine-ABD fusion proteins containing both IL-2 and IL-15 are thought to be useful for tumor shrinkage and cancer treatment.

[0140] d. IL-7 and IL-12 In certain embodiments, the cytokine-ABD fusion proteins include IL-7 and IL-12 fusion partners.

[0141] Interleukin-7 (IL-7) is a cytokine derived from non-hematopoietic cells that plays a central role in the adaptive immune system. It promotes lymphocyte development in the thymus, which maintains the homeostatic survival of naive and memory T cells in the periphery. Furthermore, it is important for lymph node (LN) organogenesis and the maintenance of activated T cells replenished in secondary lymphoid organs (SLO). The immunity of cancer patients is suppressed, characterized by a decrease in T cell numbers, a decrease in the infiltration of effector immune cells, an increase in the level of exhausted effector cells, and an increase in the level of immunosuppressive cytokines such as transforming growth factor-β (TGF-β). IL-7 is an ideal solution for immune reconstitution in immunosuppressed cancer patients by promoting the proliferation of peripheral T cells. In animal models, IL-7 has been shown to extend the survival of tumor-bearing hosts. See Gao et al., Int. J. Mol. Sci. 16:10267-10280 (2015).

[0142] IL-12 acts directly on CD8+ T cells to enhance IL-7-mediated proliferation. Cytokine-ABD fusion proteins containing the IL-7 and IL-12 binding domains are thought to advantageously promote the proliferation of CD8+ T cells and enhance the cytolytic activity against tumors.

[0143] e. IL-7 and IL-15 In certain embodiments, the cytokine-ABD fusion proteins include IL-7 and IL-15. As described above, interleukin 7 and 15 are considered to be potent pro-inflammatory cytokines with the ability to reduce tumor formation. Cytokine ABD fusion proteins containing both IL-7 and IL-15 are thought to be useful for tumor shrinkage and cancer treatment.

[0144] f. IL-12 and IL-21 In certain embodiments, the cytokine-ABD includes IL-12 and IL-21.

[0145] As described above, IL-12 can stimulate the proliferation of NK cells and activated T cells and support the proliferation of effector T cells. IL-21 is a regulator of NK and T cell functions that bridges the innate and adaptive immune systems . IL-21 promotes the maturation of NK cells from bone marrow progenitor cells, activates human peripheral NK cells, promotes the proliferation and maturation of NK, and enhances CD8+ T cell-mediated effector function.

[0146] Cytokine-ABD fusion proteins containing both IL-12 and IL-21 are thought to be useful for tumor shrinkage and cancer treatment.

[0147] g. IL-12 and IL-18 In certain embodiments, the cytokine-ABD includes IL-12 and IL-18.

[0148] IL-18 is known to induce IFN-γ production and promote the development of Th1 cells and NK activation. IL-12 is known to induce IFN-γ-dependent upregulation of the IL-18 receptor. Administration of SCK mouse breast cancer cells co-expressing IL-18 and IL-12 to mice reduced the tumor tissue mass and inhibited angiogenesis (see, for example, Coughlin et al., J Clin Invest 101(6):1441-1452(1998)). When IL-18 is used in combination with IL-12 in tumor-bearing mice, extended serum levels of IFN-γ are synergistically induced, while tumor-bearing mice treated with IL-18 or IL-12 alone induced minimal serum IFN-γ that rapidly decayed (see, for example, Subleski et al., Cancer Res 66(22):11005-11012(2006)). An exemplary IL-18-ABD is shown in FIG. 37.

[0149] Cytokine-ABD fusion proteins containing both IL-12 and IL-18 fusion partners are thought to be useful for tumor shrinkage and cancer treatment.

[0150] h.GM-CSF and IL-12 In certain embodiments, the cytokine-ABD includes GM-CSF and IL-18. GM-CSF regulates the differentiation and proliferation of hematopoietic progenitor cells. GM-CSF also enhances the ability of APCs to process and present antigens, which in turn leads to activation of cytotoxic T cells, increased IFN-γ production, and ultimately tumor regression. Both GM-CSF and IL-12 can induce significant antitumor responses in several different preclinical tumor models, including liver tumor models and lung tumor models. (See, for example, Kilinc et al., J Immunol 177(10):6962-6973(2006)).

[0151] Cytokine-ABD fusion proteins that include both GM-CSF and IL-12 fusion partners are thought to be useful for tumor shrinkage and cancer treatment.

[0152] i. IFNα and IL-12 The synergistic nature of these two cytokines extends beyond mere similar biological effects. For example, IL-12, which is well known to induce IFN-γ production, can result in the production of additional soluble factors that enhance IFN-α signaling. Interferons, including interferon α, are known to induce apoptosis in malignant cells. See, for example, Thyrell, L. et al., Oncogene 21, 1251-1262 (2002).

[0153] Cytokine-ABD fusion proteins that include both IFNα and IL-12 are thought to be useful for tumor shrinkage and cancer treatment.

[0154] Additional cytokine-cytokine combinations that can be included in the subject cytokine-ABD fusion proteins are shown in Figure 51A.

[0155] 2. Binding moiety-ABD fusion proteins In some embodiments, the ABD fusion protein includes a binding moiety (e.g., scFv) fusion partner, i.e., a BM-ABD fusion protein. In some embodiments, the BM-ABD includes one binding moiety. In certain embodiments, the BM-ABD includes two binding moieties (e.g., scFv). In other embodiments, the BM-ABD includes a cytokine and a binding moiety (Figure 1B). Figure 1 shows some exemplary orientations in which a binding moiety or a combination of binding moiety-cytokine / binding moiety-binding moiety can bind to the ABD.

[0156] Binding moieties useful for practicing with major BM-ABD binding moieties based on antibody variable heavy and variable light domains. In some embodiments, the binding moiety includes a variable heavy domain and a variable light domain. In some embodiments, the binding moiety is a single-chain variable fragment (scFv).

[0157] Any ABD can be used in such fusion proteins, including those that include any of the ABD variable heavy and variable light domains shown in Figure 2. In certain embodiments, the ABD includes the variable heavy and light domains of A10m3. In some embodiments, the ABD is an A10m3 scFv.

[0158] Any linker can be used to couple the cytokine and / or binding moiety to the ABD, including any linker shown in Figure 48. In an exemplary embodiment, the linker is (GGGGS)5. An exemplary backbone linker-A10m3-linker sequence for use in such fusion proteins is shown in Figure 48.

[0159] Two binding moieties (e.g., scFv), or an exemplary combination of a cytokine and a binding moiety (which can be such an ABD fusion protein), are shown in Figure 51B.

[0160] In some embodiments, the BM-ABD includes an anti-TGFβ binding domain (e.g., an anti-TGFβ scFv). An exemplary anti-TGFβ scFv sequence is shown in Figure 45. In some embodiments, the anti-TGFβ scFv includes the variable heavy and variable light domains of the 4D9 anti-TGFβ scFv. In some embodiments, the ABD fusion protein includes an anti-TGFβ scFv and a cytokine or additional binding moiety, and the cytokine or additional binding moiety is a second anti-TGFβ scFv, IL-15, IL-12, or an anti-PD-L1 binding domain (10D12). See Figure 51B.

[0161] In some embodiments, the BM-ABD includes an anti-PD-L1 binding domain (e.g., an anti-T PD-L1 scFv). In some aspects, the anti-PD-L1 scFv includes the variable heavy domain and the variable light domain of the 4D9 anti-PD-L1 10D12 scFv. In some embodiments, the ABD fusion protein includes an anti-PD-L1 scFv and a cytokine or additional binding moiety, which is a second anti-anti-PD-L1 scFv, IL-15, IL-12, or an anti-TGFβ binding domain (4D9). See Figure 51B.

[0162] a. TGF-β binding moiety In certain embodiments, the BM-ABD provided herein includes a TGF-β binding moiety. As used herein, "TGF-β", "TGFβ", "TGFb", and All “transforming growth factor-β”s are involved in controlling the proliferation, cell differentiation, and other functions of most cells and refer to members of a cytokine family that exists in at least three isoforms, TGFβ1 (Genbank accession numbers: NM_000660 and NP_000651 (human), and NM_011577 and NP_035707 (mouse)), TGFβ2 (Genbank accession numbers NM_001135599 and NP_001129071 (human), and NM_009367 and NP_33393 (mouse)), and TGFβ3 (Genbank accession number: NM_003239). Members of the TGFβ family have an N-terminal signal peptide of 20-30 amino acids required for secretion from cells, a pro-region, and a C-terminal region of 112-113 amino acids that becomes the mature TGFβ molecule according to its released form, and a proteolytic pro-region. In certain embodiments, the mature TGFβ protein dimerizes to produce a 25 kDa active molecule with many conserved structural motifs, contains nine cysteine residues, eight of which can form disulfide bonds within the TGFβ molecule to create a cystine knot structure. The ninth conserved cysteine forms a bond with the ninth cysteine of another TGFβ to produce a dimer.

[0163] Without being bound by a particular theory of operation, the BM-ABD of the subject that binds to TGFβ is considered to be useful for treating a subject having cancer (e.g., advanced cancer). In certain embodiments, the BM-ABD includes a TGFβ1 binding portion. In certain embodiments, the BM-ABD includes a TGFβ2 binding portion. In certain embodiments, the BM-ABD includes a TGFβ3 binding portion. In some embodiments, the multivalent binding polypeptide includes a binding portion capable of binding to TGFβ1, TGFβ2, and / or TGFβ3, or any combination thereof (e.g., binding of TGFβ1 and TGFβ2, binding of TGFβ2 and TGFβ3, binding of TGFβ1 and TGFβ3, or binding of TGFβ1, TGFβ2, and TGFβ3). In some embodiments, the TGFβ binding portion binds to TGFβ1, TGFβ2, and TGFβ3. In some embodiments, the TGFβ binding portion includes the variable heavy domain and variable light domain of the TGFβ binding portion of FIG. 40. In one embodiment, the TGFβ binding portion includes the variable heavy domain and variable light domain of the TGFβ binding portion 4D9 (FIG. 40B). In certain embodiments, the TGFβ binding portion is 4D9 scFv. 4D9 is CD4 + FoxP3 + It has been shown to prevent the proliferation of regulatory T cells, prevent the activation of Smad (e.g., phosphorylation of Smad2), and prevent the epithelial-to-mesenchymal transition of cells and / or the migration of cancer cells.

[0164] In some embodiments, the TGFβ binding portion-ABD fusion protein further binds to another binding portion or cytokine. In certain embodiments, the other binding portion is a PD-L1 binding portion or another TGFβ binding portion. In some embodiments, the cytokine is IL-15 or IL-12 (see FIG. 51B).

[0165] b. PD-L1 binding portion In certain embodiments, the BM-ABD provided herein includes a programmed cell death 1 ligand 1 (PD-L1) binding moiety. As used herein, "programmed cell death 1 ligand 1", "programmed death ligand 1", "PDL1", and "PD-L1" (Genbank accession numbers NM_001267706 and NP_001254635 (human), and NM_021893 and NP_068693 (mouse)) all refer to members of a 40 kDa type I transmembrane protein that binds to the Pd1 receptor found on activated T cells, B cells, and myeloid cells in order to regulate activation or inhibition. Upregulation of PD-L1 allows cancer to evade the immune system. See, for example, Hamanishi et al., Proc Natl Acad Sci USA 104(9):3360-5 (2007). Thus, BM-ABDs of the subject matter that include a PD-L1 binding moiety are thought to be useful in the treatment of cancer.

[0166] In one embodiment, the PD-L1 binding moiety includes the variable heavy domain and variable light domain of the PD-L1 binding moiety 10D12 (Figure 50). In certain embodiments, the PD-L1 binding moiety is a 10D12 scFv (Figure 50). 10D12 binds to hPD-L1 at low pH and cross-reacts with mPD-L1. 10D12 does not bind to hPD-L2 or mPD-L2. Additionally, 10D12 blocks the PD-1 / PD-L1 interaction, as well as the B71 / PD-L1 interaction.

[0167] In some embodiments, the PD-L1 binding moiety-ABD fusion protein further binds to another binding moiety or cytokine. In certain embodiments, the other binding moiety is a TGFβ binding moiety or another PD-L1 binding moiety. In some embodiments, the cytokine is IL-15 or IL-12 (see Figure 51B).

[0168] c. TNF and other binding moieties In one embodiment, the ABD fusion proteins provided herein include a tumor necrosis factor (TNF) binding portion. Without being bound by a particular theory of operation, such ABD fusion proteins are thought to be useful as anti-inflammatory and / or cancer therapeutics. In some embodiments, the TNF binding portion is a scFv. In certain embodiments, the TNF binding portion-ABD fusion protein is further bound to another fusion partner that is a binding portion or inhibitor peptide. In certain embodiments, the second binding portion is a second TNF binding portion, an IL-1 binding portion, an IL-6 binding portion, an IL-8 binding portion, an IL-17 (isoforms A-F) binding portion, or an IL-23 binding portion.

[0169] In another embodiment, the ABD fusion protein includes a binding portion selected from an IL-1, IL-6, IL-8, IL-17 (A-F), and IL-23 binding portion.

[0170] In some embodiments, such TNF and interleukin binding portion-ABDs are useful for the treatment of diseases such as rheumatoid arthritis, Crohn's disease, psoriatic arthritis, ankylosing spondylitis, ulcerative colitis, chronic plaque psoriasis, and TNF-mediated diseases.

[0171] F. Diagnostic Uses In another aspect, methods for imaging and / or detecting a tumor are provided herein. In some aspects, the methods include contacting tumor cells, tumor cell cultures, tumor vascular cells, tumor vascular cell cultures, tumor tissue, and other tissues and cells with a labeled ABD fusion protein of the subject invention.

[0172] The ABD fusion proteins are also useful for in vivo or in vitro imaging of tumors or autoimmune disease states associated with the antigen binding partners of the ABD fusion proteins herein. In some embodiments, the fusion proteins described herein are used for both diagnosis and treatment, or for diagnosis only. In some embodiments, the subject ABD fusion proteins are labeled.

[0173] Diagnosis can be performed in vivo by administration of diagnostic proteins that allow for whole-body imaging, as described below, or in vitro in a sample removed from the patient. A "sample" in this context includes any number of samples, including but not limited to, bodily fluids (including but not limited to blood, urine, serum, lymph, saliva, anal and vaginal secretions, sweat, and semen), and tissue samples, such as those obtained from biopsies of relevant tissues.

[0174] As used herein, "labeled" refers to an ABD fusion protein disclosed herein that is tagged with one or more elements, isotopes, or other elements to allow for detection in a screening or diagnostic procedure. means to attach a chemical compound. Generally, labels are divided into several classes: a) immunolabels, which may be epitopes incorporated as fusion partners recognized by antibodies; b) isotopic labels, which may be radioactive or heavy isotopes; c) small molecule labels, which may include molecules such as biotin, which allow for fluorescent and colorimetric dyes or other labeling methods; and d) labels such as particles (including bubbles for ultrasound labeling) or paramagnetic labels, which allow for body imaging. Labels may be incorporated into proteins at any position (e.g., via one or more of the linkers described herein) or may be incorporated in vitro or in vivo during protein expression, as known in the art. Specific labels include optical dyes, including, but not limited to, chromophores, fluorophores, and fluorophores, the latter of which are often specific. Fluorophores may be either "small molecule" fluorophores or proteinaceous fluorophores.

[0175] "Fluorescent label" means a molecule that can be detected through its intrinsic fluorescent properties. Suitable fluorescent labels include, but are not limited to, fluorescein, rhodamine, tetramethylrhodamine, eosin, erythrosin, coumarin, methylcoumarin, pyrene, malachite green, stilbene, lucifer yellow, cascade blue J, Texas red, IAEDANS, EDANS, BODIPY FL, LC red 640, Cy5, Cy5.5, LC red 705, Oregon green, Alexa-Fluor dyes (Alexa Fluor 350, Alexa Fluor 430, Alexa Fluor 488, Alexa Fluor 546, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor 633, Alexa Fluor 660, Alexa Fluor 680), cascade blue, cascade yellow, and R-phycoerythrin (PE) (Molecular Probes, Eugene, Oregon), FITC, rhodamine, and Texas red (Pierce, Rockford, Ill), Cy5, Cy5.5, Cy7 (Amersham Life Science, Pittsburgh, Pa). Suitable optical dyes containing fluorescent dye molecules are described in the Molecular Probes Handbook by Richard P. Haugland, which is hereby incorporated by reference in its entirety.

[0176] Suitable proteinaceous fluorescent labels include green fluorescent proteins including the Aequorea species of GFP of Renilla, Ptilosarcus, or Aequorea (Chalfie et al., Science 263:802-805 (1994)), EGFP (Clontech Laboratories, Inc., Genbank accession number U55762), blue fluorescent protein (BFP, Quantum Biotechnologies, Inc., 1801 de Maisonneuve Blvd. West, 8th Floor, Montreal, Quebec, Canada H3H 1J9; Stauber, 1998, Biotechniques 24:462-471, Heim et al., 1996, Curr. Biol. 6:178-182), enhanced yellow fluorescent protein (EYFP, Clontech Laboratories, Inc.), luciferase (Ichiki et al., J. Immunol. 150:5408-5417 (1993)), β-galactosidase (Nolan Proc. Natl. Acad. Sci. USA 85:2603-2607 (1998)) and jellyfish (WO92 / 15673, WO95 / 07463, WO98 / 14605, WO98 / 26277, WO99 / 49019, U.S. Patent No. 5,292,658, U.S. Patent No. 5,418,155, U.S. Patent No. 5,683,888, U.S. Patent No. 5,741,668, U.S. Patent No. 5,777,079, U.S. Patent No. 5,804,387, U.S. Patent No. 5,874,304, U.S. Patent No. 5,876,995, U.S. Patent No. 5,925,558) are included. In this section, all of the above-cited references are expressly incorporated herein by reference.

[0177] G. Albumin Binding Domain and Production of Fusion Proteins As will be understood by those skilled in the art, standard protocols are used to generate the subject ABD. General methods for antibody molecular biology, expression, purification, and screening are described in Antibody Engineering, edited by Kontermann & Dubel, Springer, Heidelberg, 2001; and Hayhurst & Georgiou, Curr Opin Chem Biol 5:683-689 (2001), Maynard & Georgiou, Annu Rev Biomed Eng 2:339-76 (2000).

[0178] In one embodiment disclosed herein, a nucleic acid encoding an ABD fusion protein is generated, and then the nucleic acid can be cloned into a host cell, expressed, and assayed as needed. Thus, nucleic acids, particularly DNA, encoding each protein sequence can be generated. These procedures are carried out using well-known methods. For example, similar to antibody production, various methods that may be useful for the production of ABD fusion proteins are disclosed herein and are described in Molecular Cloning - A Laboratory Manual, 3rd edition (Maniatis, Cold Spring Harbor Laboratory Press, New York, 2001) and Current Protocols in Molecular Biology (John Wiley & Sons), both of which are hereby incorporated by reference in their entirety. There are various techniques that can be used to efficiently generate the DNA encoding ABD disclosed herein. Such methods include, but are not limited to, gene assembly methods, PCR-based methods and methods using PCR modifications, ligase chain reaction-based methods, pool oligo methods such as those used in synthetic shuffling, error-prone amplification methods, methods using oligos with random mutants, classical site-directed mutagenesis methods, cassette mutagenesis, and other amplification and gene synthesis methods. As is known in the art, there are various commercially available kits and methods for gene assembly, mutagenesis, vector subcloning, etc., and such commercial products are useful for generating nucleic acids encoding ABD fusion proteins.

[0179] The ABDs disclosed herein can be produced by culturing host cells transformed with a nucleic acid, such as an expression vector, containing a nucleic acid encoding an ABD fusion protein under appropriate conditions that induce or cause the expression of the protein. The conditions suitable for expression vary depending on the choice of the expression vector and the host cell and can be readily determined by one of ordinary skill in the art through routine experimentation. A variety of suitable host cells can be used, including but not limited to mammalian cells, bacteria, insect cells, yeast, and plant cells. For example, various cell lines that may be useful for the production of the ABD fusion proteins disclosed herein are described in the ATCC® cell line catalog available from the American Type Culture Collection.

[0180] In one embodiment, the ABD is expressed in a mammalian cell expression system that includes a system in which an expression construct is introduced into mammalian cells using a virus such as a retrovirus or an adenovirus. Any mammalian cell, e.g., human, mouse, rat, hamster, primate cells, can be used. Suitable cells include, but are not limited to, Jurkat T cells, NIH3T3, CHO, BHK, COS, HEK293, PER C.6, HeLa, Sp2 / 0, NS0 cells, and variants thereof, as well as known research cells. In an alternative embodiment, the library protein is expressed in bacterial cells. Bacterial expression systems are well known in the art and include Escherichia coli (E. coli), Bacillus subtilis, Streptococcus cremoris, and Streptococcus lividans. In another embodiment, the ABD fusion protein is produced in insect cells (e.g., Sf21 / Sf9, Trichoplusia ni Bti-Tn5b1-4) or yeast cells (e.g., Saccharomyces cerevisiae, Pichia, etc.). In an alternative embodiment, the ABD polypeptide is expressed in vitro using a cell-free translation system. Prokaryotic cells In vitro translation systems derived from both prokaryotes (such as Escherichia coli) and eukaryotic cells (wheat germ, rabbit reticulocytes) are available and can be selected based on the expression level and functional characteristics of the protein of interest. For example, as will be understood by those skilled in the art, some display technologies, such as ribosome display, require in vitro translation. In addition, ABD fusion proteins can be produced by chemical synthesis methods. Also, transgenic expression systems in both animals (such as milk, sheep milk, or goat milk, chicken eggs, whole insect larvae, etc.) and plants (such as corn, tobacco, duckweed, etc.).

[0181] The nucleic acids encoding the ABD fusion proteins disclosed herein can be incorporated into expression vectors for protein expression. Various expression vectors can be used for protein expression. The expression vector can include a self-replicating episomal vector or a vector integrated into the host genome. The expression vector is constructed to be compatible with the host cell type. Thus, expression vectors useful for the production of the antibodies disclosed herein include, but are not limited to, those that enable protein expression in mammalian cells, bacteria, insect cells, yeast, and in vitro systems. As is known in the art, various expression vectors that may be useful for expressing the antibodies disclosed herein are commercially or otherwise available.

[0182] The disclosed ABD fusion proteins can be encoded by multiple nucleic acid molecules. For example, the variable heavy chain and variable light chain can be independently introduced into host cells. Although present in separate nucleic acids, their expression produces a single polypeptide.

[0183] Expression vectors typically contain a protein operably linked to a control or regulatory sequence, a selectable marker, any fusion partner, and / or additional elements. As used herein, "operably linked" means that the nucleic acid is in a functional relationship with another nucleic acid sequence. Generally, these expression vectors contain transcription and translation regulatory nucleic acids operably linked to a nucleic acid encoding a multivalent ABD fusion protein and are typically suitable for the host cell used for protein expression. Generally, transcription and translation regulatory sequences can include a promoter sequence, a ribosome binding site, transcription start and stop sequences, translation start and stop sequences, and enhancer or activator sequences. As is known in the art, expression vectors typically contain a selectable gene or marker that allows for the selection of transformed host cells containing the expression vector. Selectable genes are well known in the art and vary depending on the host cell used.

[0184] In one embodiment, the ABD is purified or isolated after expression. The ABD and ABD fusion proteins can be isolated or purified by various methods known to those skilled in the art. As described herein, purification can be particularly useful for separating heterodimeric heavy chain species from homodimeric heavy chain species. Standard purification methods include chromatographic techniques such as ion exchange, hydrophobic interaction, affinity, sizing or gel filtration, and reverse phase, which are performed at atmospheric or high pressure using systems such as FPLC and HPLC. Purification methods also include electrophoresis, isoelectric focusing, immunology, precipitation, dialysis, and chromatofocusing techniques. Ultrafiltration and diafiltration techniques are also useful in combination with protein concentration. If a fusion is used and a His tag is used, Ni+2 affinity chromatography is used, and if a flag tag is used, an immobilized anti-flag antibody is used. For general guidance on suitable purification techniques, see, for example, Protein Purification, Principles and Practice, 3rd Edition, Scopes, Springer-Verlag, NY, 1994, which is incorporated herein by reference in its entirety. The degree of purification required varies depending on the antibody screening or application. In some cases, purification is not required.

[0185] H. Therapeutic Uses of Albumin-Binding Domains and ABD Fusion Proteins The subject ABDs and ABD fusion proteins are useful in a variety of therapeutic applications, as described herein.

[0186] In one aspect, provided herein is a method of inhibiting tumor growth in a subject in need thereof by administering the albumin-binding domain (ABD) fusion proteins of the subject described herein. Useful ABD fusion proteins include, but are not limited to, those disclosed in FIGS. 4, 20, 34, 36, 40, 45, 50-51.

[0187] In some embodiments, the ABD fusion protein includes IL-12 or IL-15 (e.g., an IL-12-ABD or IL-15-ABD fusion protein). As described herein, the IL-15 ABD fusion protein can inhibit tumor growth in a dose-dependent manner. Such IL-15-mediated tumor growth inhibition is accompanied by an increase in tumor-infiltrating lymphocytes including cytotoxic T lymphocytes (CTLs) and activated natural killer (NK) cells. In certain embodiments, the ABD fusion protein includes an IL-12 molecule. In some aspects, the ABD fusion protein includes IL-15. In yet other specific embodiments, the cytokine-ABD includes IL-12 and IL-15. In some embodiments, the IL-15-ABD includes the IL-15 variant as described herein (see, e.g., FIG. 3).

[0188] Also provided herein is a method of treating a subject having cancer by administering a subject albumin-binding domain (ABD) fusion protein. In some embodiments, the ABD fusion protein includes IL-12 or IL-15 (e.g., an IL-12-ABD or IL-15-ABD fusion protein). IL-12 and IL-15 are cytokines for immunomodulation of the tumor microenvironment by their ability to enhance the survival, proliferation, and expansion of CD8+ T cells. Other useful ABD fusion proteins include, but are not limited to, those disclosed in FIGS. 4, 20, 34, 36, 40, 45, 50-51.

[0189] Examples of cancers treated herein include, but are not limited to, carcinomas, blastomas, sarcomas, certain leukemias or lymphoid malignancies. More specific examples of such cancers include squamous cell carcinoma (e.g., cutaneous squamous cell carcinoma), small cell lung cancer, lung cancers including non-small cell lung cancer, lung adenocarcinoma and lung squamous cell carcinoma, peritoneal cancer, hepatocellular carcinoma, gastric cancer including gastrointestinal cancer (gastric cancer) or gastric cancer, pancreatic cancer, glioblastoma, cervical cancer, ovarian cancer, liver cancer, bladder cancer, hepatoma, breast cancer, colorectal cancer, rectal cancer, colorectal cancer, endometrial or uterine cancer, salivary gland cancer, kidney cancer or renal carcinoma, prostate cancer, vulvar cancer, thyroid cancer, liver cancer, anal cancer, penile cancer, skin cancer / melanoma, and head and neck cancers and metastases associated with any of the primary tumors are included.

[0190] In another aspect provided herein is a method of enhancing the proliferation and / or viability of CD8 + T cells. In certain embodiments, the method comprises contacting the cells with an albumin-binding domain fusion protein comprising IL-12 and / or IL-15 (eg, an IL-12-ABD or IL-15-ABD fusion protein). In certain embodiments, the ABD fusion protein comprises an IL-12 molecule. In some embodiments, the ABD fusion protein comprises an IL-15 molecule. In still other embodiments, the fusion protein ABD fusion protein comprises IL-12 and IL-15.

[0191] I. Pharmaceutical Formulations, Administration and Dosage In another aspect, a therapeutic composition comprising an albumin-binding domain (ABD) polypeptide and a carrier of any subject is provided herein. The subject used to carry out the aforementioned method The therapeutic composition can be formulated into a pharmaceutical composition comprising a carrier suitable for the desired delivery method. Suitable carriers include any substance that, when combined with the therapeutic composition, retains the anti-tumor function of the therapeutic composition and is generally non-reactive with the patient's immune system. Examples include any of several standard pharmaceutical carriers, such as sterile phosphate buffered saline, bacteriostatic water, etc., but are not limited thereto (generally Remington´s Pharmaceutical Sciences 16 th Edition, A.Osal., Ed., 1980).

[0192] 1. Compositions for In Vivo Administration The formulations of the albumin-binding domain (ABD) fusion proteins used in the present invention are prepared for storage in the form of lyophilized formulations or aqueous solutions by mixing an ABD fusion protein having a desired purity with any pharmaceutically acceptable carrier, excipient, or stabilizer (Remington´s Pharmaceutical Sciences 16th edition, Osol, A. Ed.

[1980] ). Acceptable carriers, excipients, or stabilizers are non-toxic to the recipient at the dosages and concentrations used and include buffers such as phosphoric acid, citric acid, and other organic acids; antioxidants including ascorbic acid and methionine; preservatives (octadecyldimethylbenzylammonium chloride; hexamethonium chloride; benzalkonium chloride, benzetonium chloride; phenol, butyl, or benzyl alcohol; alkyl parabens such as methyl or propyl paraben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight (less than about 10 residues) fusion proteins; proteins such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers such as polyvinylpyrrolidone; amino acids such as glycine, glutamine, asparagine, histidine, arginine, lysine; monosaccharides, disaccharides, and other carbohydrates including glucose, mannose, or dextrin; chelating agents such as EDTA or DPTA; sugars such as sucrose, mannitol, trehalose or sorbitol; salt-forming counterions such as sodium; metal complexes (e.g., Zn-protein complexes); and / or nonionic surfactants such as TWEEN (trademark), PLURONICS (trademark) or polyethylene glycol (PEG) of various molecular weights.

[0193] The formulations of the present specification may also contain two or more active compounds necessary for the specific indication to be treated, preferably those having complementary activities that do not adversely affect each other. For example, it may be desirable to confer other specificities to the ABD fusion protein. Alternatively, or in addition, the composition may contain a cytotoxic agent, a cytokine, a growth inhibitor, and / or a small molecule antagonist. Such molecules are preferably present in combination in an amount effective for the intended purpose.

[0194] The active ingredient may be entrapped, for example, in microcapsules prepared by coacervation techniques or interfacial polymerization, such as in hydroxy methylcellulose or gelatin microcapsules, and poly-(methyl methacrylate) microcapsules, respectively, in colloidal drug delivery systems (such as liposomes, albumin microspheres, microemulsions, nanoparticles, and nanocapsules), or in macroemulsions. Such techniques are disclosed in Remington´s Pharmaceutical Sciences 16th Edition, Osol, A. Ed (1980).

[0195] The formulations used for in vivo administration must be sterile or substantially sterile. This is readily accomplished by filtration through sterile filtration membranes.

[0196] Sustained release formulations can be prepared. Suitable examples of sustained release preparations include semipermeable matrices of solid hydrophobic polymers containing antibodies, and the matrix is a molded article, such as a It is in the form of pills or microcapsules. Examples of sustained-release matrices include polyesters, hydrogels (e.g., poly(2-hydroxyethyl methacrylate) or poly(vinyl alcohol)), polylactides (U.S. Patent No. 3,773,919), copolymers of L-glutamic acid and γ-ethyl-L-glutamic acid, non-degradable ethylene-vinyl acetate copolymers, degradable lactic acid-glycolic acid copolymers, such as LUPRON DEPOT™ (injectable microspheres composed of lactic acid-glycolic acid copolymer and leuprolide acetate), and poly-D-(-)-3-hydroxybutyric acid. Polymers such as ethylene-vinyl acetate and lactic acid-glycolic acid enable the release of molecules for more than 100 days, while certain hydrogels release proteins for a shorter period.

[0197] When the encapsulated albumin-binding domain fusion protein remains in the body for a long time, it may denature or aggregate as a result of being exposed to moisture at 37°C, losing its biological activity and potentially changing its immunogenicity. Depending on the mechanism involved, rational strategies for stabilization can be devised. For example, if the aggregation mechanism is found to be intermolecular S-S bond formation via thiol-disulfide exchange, stabilization can be achieved by denaturing sulfhydryl residues, lyophilization from acidic solutions, controlling the water content, using appropriate additives, and developing specific polymer matrix compositions.

[0198] 2. Administrative modality The subject albumin-binding domain fusion protein and therapeutic agent are administered to a subject according to known methods, for example, by intravenous administration as a bolus or continuous infusion over a period of time, intramuscular, intraperitoneal, intrathecal, subcutaneous, intra-articular, intrasynovial, subarachnoid, oral, topical, or inhalation routes. Intravenous or subcutaneous administration of the antibody is preferred.

[0199] 3. Treatment In the methods provided herein, treatment is used to provide a positive treatment response with respect to a disease or medical condition. A "positive treatment response" is intended to result in an improvement in the disease or medical condition and / or an improvement in the symptoms associated with the disease or medical condition. For example, a positive treatment response can refer to one or more of the following improvements in a disease: (1) a decrease in the number of neoplastic cells, (2) an increase in neoplastic cell death, (3) inhibition of neoplastic cell survival, (5) inhibition of tumor growth (i.e., some degree of deceleration, preferably cessation), (6) an increase in patient survival rate, and (7) some relief from one or more symptoms associated with the disease or medical condition.

[0200] A positive treatment response in any given disease or medical condition can be determined by standardized response criteria specific to that disease or medical condition. Tumor response can be evaluated for changes in tumor morphology (i.e., total body tumor tissue mass, tumor size, etc.) using screening techniques such as magnetic resonance imaging (MRI) scans, radiographs, computed tomography (CT) scans, bone scans, endoscopies, and tumor biopsy sampling including bone marrow aspiration (BMA) and counting of circulating tumor cells.

[0201] In addition to these positive treatment responses, a subject undergoing treatment can experience the beneficial effects of an improvement in the symptoms associated with the disease.

[0202] Thus, for example, in the case of B cell tumors, the subject can experience a reduction in so-called B symptoms, i.e., night sweats, fever, weight loss, and / or pruritus. In the case of pre-malignant conditions, treatment with a multi-therapeutic agent can block and / or extend the time to onset of a related malignancy, e.g., the onset of multiple myeloma in a subject suffering from monoclonal gammopathy of undetermined significance (MGUS).

[0203] Improvement of the disease can be characterized as a complete response. "Complete response" means multiple myeloma In the case of, it means that there is no clinically detectable disease with normalization of any previous abnormal radiation examinations, bone marrow, and cerebrospinal fluid (CSF) or abnormal monoclonal proteins.

[0204] Such an efficacy may persist for at least 4 to 8 weeks, or sometimes 6 to 8 weeks, after the following treatment by the subject method. Alternatively, the improvement of the disease may be classified as a partial response. "Partial response" means at least about a 50% decrease in all measurable tumor tissue amounts (i.e., the number of malignant cells present in the subject, or the measured bulk of the tumor mass, or the amount of abnormal monoclonal protein), which may persist for 4 to 8 weeks, or 6 to 8 weeks.

[0205] Treatment includes a "therapeutically effective amount" of the agent used. "Therapeutically effective amount" refers to an amount effective at the required dosage and duration to achieve the desired therapeutic result.

[0206] The therapeutically effective amount may vary depending on factors such as the individual's disease state, age, gender, and weight, as well as the ability of the agent to induce the desired response in the individual. The therapeutically effective amount is also an amount where the therapeutically beneficial effects outweigh any toxic or detrimental effects of the antibody or antibody portion.

[0207] The "therapeutically effective amount" for tumor therapy can also be measured by the ability to stabilize the progression of the disease. The ability of a compound to inhibit cancer may be evaluated in an animal model system that predicts efficacy in human tumors.

[0208] Alternatively, this property of the composition may be evaluated by in vitro assays known to those skilled in the art by examining the ability of the compound to inhibit cell proliferation or induce apoptosis. A therapeutically effective amount of a therapeutic compound can reduce tumor size or otherwise alleviate the symptoms of the subject. One skilled in the art would be able to determine such amounts based on factors such as the size of the subject, the severity of the subject's symptoms, and the particular composition or route of administration selected.

Example

[0209] Example 1: Screening and Identification of Human Serum Albumin Binding Domains To identify the scFv albumin binding domains (ABDs) of phage, solid-phase panning and solution panning methods were performed. Subsequently, the albumin binding domains selected from the primary screening of human serum albumin binding domains were screened for cross-reactivity to mouse serum albumin using standard ELISA techniques. The primary albumin binding domain candidates obtained using the screening method were sequenced and subsequently assayed for target concentration-dependent binding, pH stability, FcRn binding interference, and kinetic binding. In particular, candidates ABDs were selected for their ability to bind human serum albumin (kD approximately 20 - 60 nM), mouse serum albumin (kD approximately 10 - 30 nM), and cynomolgus monkey serum albumin (kD 20 - 60 nM) at low pH (pH 5.5) and neutral pH (pH 7.7). The candidate ABDs were assayed to confirm that they did not compete with FcRn binding to serum albumin. As described herein, ABDs that bound at such pHs and did not compete with FcRn binding are capable of FcRn-mediated endosomal recycling. Thus, biopharmaceuticals containing such ABDs (e.g., cytokines and antibody-based biopharmaceuticals) are also capable of such FcRn-mediated recycling and thus exhibit a longer half-life compared to their counterparts that do not contain such ABDs.

[0210] Based on these criteria, five albumin binding clones, A9, A10, A6, 2B4, 2H10 were selected. Of these five clones, A10 was selected based on high expression levels and the highest activity profile. A10 was then mutated to eliminate regions that might potentially cause immunogenicity. From these A10 variants, Based on high affinity for serum albumin, A10m3 was selected as the lead.

[0211] The sequence of an exemplary human serum albumin binding domain is shown in Figure 2, including A10m3 (Figure 2D).

[0212] Example 2: Variants IL-15 and IL-15-ABD Expression of the IL15-A10m3 protein was very poor by HEK293T cells and could not be explained by transcription. The IL-15-ABD (IL-15-A10m3) construct was produced in HEK293 cells within three independently transfected cell cultures and evaluated by Western blotting using either an anti-His tag antibody (Figure 5A, left) or functional ELISA binding to mouse serum (Figure 5A, right). As shown in Figure 5A, expression of IL-15-A10m3 could not be evaluated by either of these methods. To evaluate whether the lack of IL-15-A10m3 expression in HEK293 cells was due to low transcription levels, mRNA was prepared from four independent IL-15-A10m3 transfected cells and RT-PCR was performed to quantify the mRNA levels of IL15-A10m3 mRNA (Figure 5B, lanes 2-5) compared to the housekeeping gene GAPDH (Figure 5B, lane 6). As shown in Figure 5B, IL15-A10m3 mRNA was detected in significant amounts from the transfected cells relative to its control GAPDH, indicating that the low expression level of IL-15-A10m3 produced in HEK293 cells is not due to transcription and is likely associated with the translation process or a post-translational process.

[0213] Identification of a putative ubiquitination site within IL15 adjacent to the IL15 receptor α binding site The study showed that IL15 protein was expressed intracellularly but was very unstable with a short half-life. Co-expression of IL15Rα and IL15 in the same cells significantly increased the amounts of IL15Rα and IL15 on the cell surface. Further studies confirmed that IL15Rα acts as a chaperone for IL15, binds IL15 intracellularly, and protects and stabilizes IL15 before secretion. These findings suggest that translation cannot account for the low productivity of IL15. Rather, post-translational modification (PTM) may be involved in the intracellular instability of IL15, and the intracellular instability of IL15 can be overcome by blocking a specific yet unknown post-translational modification by IL15Rα. Ubiquitination is a well-established mechanism that allows cells to mark intracellular proteins for degradation.

[0214] Assuming that IL15 is a very potent pro-inflammatory cytokine and its expression is tightly regulated by cells, it is possible for cells to actively control IL15 protein levels by utilizing ubiquitination. Potential ubiquitination sites on IL-15 that are presumably protected by binding to IL-15Rα were identified (Figure 6). In particular, amino acid K86 is a putative ubiquitination site adjacent to the IL-15 / IL-15Rα binding site (Figure 6A), suggesting that binding of IL15Rα to IL15 may block the accessibility of ubiquitin ligases (e.g., E3) to K86 on the IL15 protein. K86 was further confirmed as a ubiquitination site using UbPred, an online ubiquitination site database (www.ubpred.org) (Figure 6B).

[0215] Mutation of K86 on IL15 restores the expression of IL15-A10m3 protein by HEK293T cells. To evaluate whether ubiquitination at K86 affects the intracellular stability of IL-15, IL-15 variants containing amino acid substitutions at K86, including K86A and K86R, were generated. Sequences of some of these IL-15 variants are shown in Figure 3. Without being bound by a particular theory of operation, amino acid substitutions at these specific ubiquitination sites are thought to result in ubiquitination-resistant IL-15 that exhibits higher stability than wild-type IL-15. Such IL-15 variants were conjugated to ABD (A10m3) to further extend the half-life (Figure 4). Figure 7 provides a schematic of such an IL-15-ABD fusion protein, which includes an IL-15-ABD (A) having an IL-15Rα / IL15 "sushi domain" and an IL-15-ABD fusion protein (B) that includes an IL-15 variant having an amino acid substitution at the ubiquitination site K86. HEK293 cell-produced IL-15-ABD K86R, and K86A variants, as well as IL-15Rα / IL15-ABD, were evaluated for their ability to bind to mouse serum albumin (MSA) and IL-15Rα. As shown in Figure 8A, IL-15-ABD K86R and K86A variants (12 clones) produced in HEK293, as well as IL-15Rα / IL15-ABD (12 clones), showed higher expression compared to wild-type IL-15-ABD (12 clones) produced in HEK-293. Such constructs were also able to bind to MSA. Additionally, as shown in Figure 8B, substitutions of K86R (clone R6, green star) and K86A (clone A3, yellow star) did not interfere with the ability of the variants to bind to IL-15Rα. Interestingly, IL15Rα-fused IL15-A10m3 did not show binding to IL15Rα, suggesting that the internal IL15Rα sushi domain bound to IL15 intramolecularly and thus blocked binding to exogenous IL15Rα coated on the plate. This is consistent with our and others' findings that binding of IL15 to the IL15Rα sushi domain increases IL15 expression (Figure 7B).

[0216]

[0217] ​ One specific variant HEK293 cell-produced IL-15-ABD containing the amino acid substitution K86R (IL-15 K86R-A10m3) was subjected to scale-up production, and the IL-15 K86R-A10m3 construct was evaluated for in vitro serum albumin binding. As shown in Figure 9, IL-15-ABD with the K86 mutation (IL-15 K86R-A10m3) and IL-15Rα / IL15-ABD (IL-15Rα / IL15-A10m3) could be produced in scaled-up quantities in HEK 293 cells, as confirmed by SDS-PAGE (Figure 9C, left) and Western blotting using an anti-His tag antibody (right). Furthermore, as shown in Figure 10, K86R-A10m3 showed binding to mouse serum albumin (MSA). The biological activity of IL15 K86R-A10m3 produced from HEK293T cells is impaired, but it can be rescued by deglycosylation.

[0218] The CTLL2 proliferation assay was used to test the biological activities of various IL-15-ABDs. The IL-15-ABDs tested included wild-type IL-15-A10m3 and three different IL-15 K86R-A10m3s produced in HEK293T cells. Both commercially available IL-15 produced using E. coli and in-house produced IL-15-ABD were used as controls. As shown in Figure 11, IL-15 K86R-A10m3 produced from HEK293T showed significantly reduced ability to promote CTLL2 proliferation compared to controls produced in E. coli, including commercially available wild-type IL-15 (R&D) and in-house produced IL-15-A10m3. IL15 from HEK293T Considering the difference in the CTLL2 proliferation assay between IL15 K86R-A10m3 from HEK293T and IL15-A10m3 from E. coli, a hypothesis was proposed that IL15-A10m3 produced from E. coli does not undergo basic N-glycosylation like mammalian cells, and the glycosylation of IL15 K86R-A10m3 in HEK293 T cells may interfere with the interaction between IL15 K86R-A10m3 and its receptor.

[0219] To evaluate whether the reduced bioactivity of IL-15 K86R-A10m3 produced by HEK cells is due to glycosylation, IL-15 K86R-A10m3 was deglycosylated using PNGase and a CTLL2 proliferation assay was performed to evaluate the bioactivity of deglycosylated IL15 K86R-A10m3. After mixing and treating IL-15 K86R-A10m3 with PNGase under native conditions, the glycans were completely removed and visualized by SDS-PAGE followed by glycan staining (Figure 12A, left) and Coomassie blue staining (Figure 12A, right). 1) IL-15R-A10m3 + 5 ul PNGase mix, 2) protein + 10 ul PNGase mix, 3) no enzyme control. As shown in Figure 12B, deglycosylation of IL-15 K86R-A10m3 (blue) almost completely rescued its activity in the CTLL2 proliferation assay compared to untreated samples (purple, yellow). WT IL-15 from R&D Systems (red) and in-house E. coli-produced IL15-A10m3 (black) were used as positive controls. N112 of IL-15 K86R-A10m3 is important for its bioactivity to promote CTLL2 proliferation.

[0220] The amino acid position N112 of IL-15 K86R-A10m3 is important for properly establishing the interaction between IL-15 and the IL15 receptor γ, and thus is a site important for IL-15 bioactivity, particularly in the context of IL-15-ABD. The variant IL-15 K86R was further mutated to N112A, and it was determined whether this site mutation could restore the bioactivity of IL-15, similar to deglycosylated IL-15. In particular, IL15 K86R-A10m3 was mutated to further include N112Q, N112A, or N112S IL-15 amino acid substitutions, and the CTTLL2 proliferation assay was performed to test bioactivity. As shown in Figure 13A, introduction of the amino acid substitution N112A into IL-15 K86R-A10m3 (blue) restored bioactivity comparable to deglycosylated IL-15 K86R-A10m3 (green) in the CTLL2 proliferation assay, while the N112Q mutation (red) had no effect on bioactivity compared to the parental IL-15R-A10m3 without deglycosylation (yellow). WT IL-15 from R&D Systems (black) functioned as a positive control.

[0221] Mutations with different side chains at N112 were tested to further demonstrate the size effect on bioactivity. As shown in Figure 13B, N112Q (large, red), N112S (medium, green), and N112A (small, blue) showed increased bioactivity inversely proportional to the size of the side chain. Furthermore, the proposed hydrogen bond established by N112 of IL-15 and Y103 of the IL15 receptor γ does not seem to be important for this activity, as N112A cannot form such a bond. WT IL-15 from R&D Systems (black) and in-house E. coli-produced IL15-A10m3 (purple) were used as positive controls, and the parental IL15 K86R-A10m3 without deglycosylation (yellow) was used as a negative control.

[0222] Example 3: In Vivo Activity of Variant IL-15 and IL-15-ABD The ability of IL-15 and IL-15-ABD to inhibit tumor growth was evaluated using the B16-F10 mouse melanoma model. As summarized in Figure 14, mice were treated with IL-15, PBS placebo, or various doses of IL-15-ABD by IV injection at four different time points at 48-hour intervals. As shown in Figure 16, IL-15-ABD inhibits tumor growth in a dose-dependent manner.

[0223] To further evaluate the profile of the tumor-infiltrating lymphocyte population in IL-15-ABD-treated mice from these studies, FACS analysis was performed. As shown in Figure 16, tumors in IL-15-ABD-treated mice showed an increase in the NK cell population. This data, combined with the observation of increased tumor accumulation and retention in IL-15-ABD-treated mice as described above, suggests that the ABD enhances the pro-inflammatory effect of IL-15 within the tumor. Treatment of IL-15-ABD on lymphocyte populations in the spleen and tumor effects are summarized in Figures 17 and 18. As shown in Figures 17 and 18, FACS analysis of lymphocyte populations shows a 3- to 6-fold increase in the tumor-infiltrating CTL and NK cell populations in the tumors of IL-15 ABD-treated mice. No significant difference was observed in the spleen. Overall, the results of these studies demonstrate the tumor immunomodulatory ability of IL-15-ABD in vivo.

[0224] To evaluate the ability of the ABD fusion protein to extend the half-life of IL-15, 5 μg of IL-15-ABD or IL-15 alone was intravenously injected into C57B mice, and then the serum concentrations of IL-15-ABD and IL-15 were evaluated. As shown in Figure 20A, IL-15-ABD showed higher PK compared to IL-15 WT. IL-15 T 1 / 2β = 0.6 hours, similar to that reported in the public domain (about 0.5 hours). The research results indicate that ABD extends IL-15 T 1 / 2β to about 7.0 hours, which is an approximately 10-fold increase. IL-15-ABD was also assayed for stability in human serum using a cell line assay. As shown in Figure 19B, IL-15-ABD was more stable in human serum compared to a commercially available IL-15 control without ABD.

[0225] Example 4: IL-12-ABD The mouse IL-12 single-chain - ABD construct was prepared in HEK293T cells and purified by size exclusion chromatography. IL-12-A10m3 produced from HEK293T cells is fully active in both in vitro assays and cell line assays. As shown in Figure 21A, IL12-A10m3 can bind to mouse serum albumin, and the equilibrium dissociation constant (KD) is 2.1 nM. IL12-A10m3 produced from HEK293T was also able to stimulate human PBMC proliferation comparable to mouse IL12 produced in-house and commercially available mouse IL-12 (R&D) (Figure 21B). Furthermore, IL12-A10m3 produced from HEK293T stimulated the secretion of interferon γ from human PBMC comparable to the secretion of mouse IL-12 produced in-house and commercially available mouse IL-12 (R&D) (Figure 22). Treatment with IL-12-ABD reduces tumor volume in vivo.

[0226] The ability of IL-12 and IL-12-ABD to inhibit tumor growth was evaluated using a B16-F10 mouse melanoma model. As summarized in Figure 23, the tumor volume was 100 mm on day 7 after tumor inoculation (day 0).3 When this was reached, mice were treated with either IL-12-ABD or IL-12 at three similar doses by IV injection. Tumor growth was monitored every two days for 10 days after treatment. PBS placebo functioned as a control. As shown in Figures 24-26 and 28, both IL-12 and IL-12-ABD were able to reduce tumor growth in a dose-dependent manner. Furthermore, IL-12-ABD was able to more effectively reduce tumor volume compared to IL-12 alone at similar concentrations (e.g., day 10 in Figure 26, and the median number of days until 50% of the tumors reached 2000 mm 3 see Figure 28). In the longitudinal body weight measurements of the mice obtained in these studies, a slight change in body weight was shown in all IL-12-ABD treatment groups (Figure 27). The lack of a significant change in body weight observed suggests a lack of IL-12-ABD toxicity in the treatment groups over 12 days after treatment.

[0227] Further characterization of the pharmacodynamic effects of a single administration of IL-12-ABD (4.5 μg IL-12-ABD, the same molar dose as 3 μg IL-12 control) in B16-F10 tumor-bearing mice 5 days later demonstrated that IL-12-ABD showed a similar large suppression of tumor growth compared to the same molar dose of IL-12 control. IL-12-ABD-treated mice also showed a corresponding increase in immune activation without affecting the body weight of the mice compared to the control, as shown by the increase in spleen weight and IFN-γ (Figure 29).

[0228] Figure 30 further shows the research results comparing the tumor volumes of B16-F10 tumor-bearing mice on the 10th day after injection with either IL-12-ABD (1.3 μg), IL-12 (30 μg), or placebo. IL-12 (1 μg) and IL-12-ABD (1.3 μg) are molar equivalents and have the same biological activity in vitro, but IL-12-ABD is approximately 30-fold more potent than IL-12 in vivo (comparing the results on the 10th day in Figure 30, 1.3 μg of IL-12-ABD ≧ 30 μg of IL-12). Figure 31 further shows the hematopoietic effects of IL-12-ABD and IL-12 in mice from the study shown in Figure 30 on the 3rd and 7th days. As shown in Figure 31, mice treated with IL-12-ABD showed a temporary decrease in WBC, neutrophils, and lymphocytes on the 3rd day compared to IL-12-treated mice and placebo controls. However, such cell populations returned to normal by the 7th day. Furthermore, the IFN-γ levels in mice treated with IL-12-ABD were higher on the 3rd and 7th days compared to mice treated with IL-12 and controls.

[0229] Evaluation of the antitumor effects of IL-12-ABD or IL-12 in combination with anti-PD-1 antibody in vivo. The effect of single combination therapy using IL-12-ABD or IL-12, and anti-PD-1 antibody was evaluated in B16-F10 tumor-bearing mice on the 8th day (Figure 32).

[0230] B16-F10 tumor cell inoculation (2x10 4 cells / mouse), 10 days later, animals (7 - 10 weeks old) were assigned to 8 groups (8 animals per group). Animals were assigned based on tumor volume. At the time of assignment, the average tumor volume per group was 100 mm 3 . On day 0 (when the tumor reached 100 mm 3 ), each group received a single I.V. administration of either PBS (placebo), IL12-ABD (1.5 μg, 5 μg, 15 μg), or IL15-ABD-IL12 (1.7 μg, 6 μg, 17 μg).

[0231] The groups were examined for body weight, tumor volume, and pseudosurvival. Body weight was measured before tumor inoculation and at the time of tumor measurement. Tumor size was measured two-dimensionally every two days using calipers, and the volume was calculated using the formula V = 0.5 × a × b 2 in mm 3 where a and b are the major and minor diameters of the tumor, respectively. This study was performed as pseudosurvival. Each mouse was euthanized when the tumor reached 2000 mm 3に or when it was determined to be moribund.

[0232] As shown in Figure 33, IL-12-ABD was more effective than treatment with either anti-PD-1 or molar equivalent recombinant IL-12. Furthermore, IL-12-ABD was as effective as the combination of recombinant IL-12 and anti-PD-1 treatment. Interestingly, adding anti-PD-1 Ab to recombinant IL-12 improved the efficacy of either treatment alone, but anti-PD-1 treatment did not provide any additional benefit over IL-12-ABD.

[0233] To evaluate the ability of ABD fusion protein to extend the half-life of IL-12, 5 μg of IL-12-ABD or IL-12 alone was injected intravenously into C57B mice, and the serum concentrations of IL-12-ABD and IL-12 were evaluated. As shown in Figure 33, IL-12-ABD showed higher PK than IL-12 WT. IL-12 T 1 / 2β = 2.5 hours, similar to that reported in the public domain (about 3.5 hours). The study results showed that ABD extended IL-12 T 1 / 2β to about 9.5 hours, which is an approximately 4-fold increase.

[0234] Example 5: Bispecific IL-15-ABD-IL-12 IL-15-ABD-IL-12, hIL15(K86R / N112A)-A10m3-mIL-12sc and mIL-12sc-A10m3-hIL15(K86R / N1 12A) Constructs were made in HEK293T cells and purified by size exclusion chromatography. The sequences of these constructs are shown in Figure 34. The binding abilities of the hIL15(K86R / N112A)-A10m3-mIL-12sc and mIL-12sc-A10m3-hIL15(K86R / N112A) constructs to MSA, IL12 receptor β2, and IL-15 receptor α were evaluated by ELISA. As shown in Figure 35, both IL-15-ABD-IL-12 constructs were able to bind MSA in a dose-dependent manner in cell culture medium. Furthermore, both bispecific constructs were able to bind IL12 receptor β2 and IL15 receptor α in a dose-dependent manner in cell culture medium. As shown in Figure 35, IL-12 / IL-15-ABD having the IL-15-ABD-IL-12 orientation from the N-terminus to the C-terminus showed better antigen binding compared to IL-12-ABD-IL-15. Additional bispecificities containing IL-15 and IL-12 are disclosed in Figure 36.

[0235] Evaluation of IL-15-ABD-IL-12 for IL-12 and IL-15 activities The IL-12 and IL-15 activities of IL-15-ABD-IL-12 were further evaluated (Figures 37 and 38).

[0236] To evaluate IL-12 activity, lymphocytes from PBMCs were induced to undergo blastogenesis by treatment with PHA-P for 4 days and rhIL-2 on day 3. The lymphoblasts were then treated with either IL-15-ABD-IL-12 or an IL-12 control for 2 days, and IL-12 activity was evaluated based on lymphoblast proliferation and IFN-γ secretion (Figure 37A). IL-15 activity was evaluated using a CTLL-2 cell cytotoxic T lymphocyte proliferation assay (Figure 38A).

[0237] As shown in Fig. 37, IL-15-ABD-IL-12 exhibited IL-12 activity as evaluated by the proliferation of lymphoblasts (Fig. 37B) and the secretion of IFN-γ (Fig. 37C). Furthermore, IL-15-ABD-IL-12 exhibited IL-15 in the CTLL-2 proliferation assay (Fig. 38B). Thus, the subject IL-15-ABD-IL-12 exhibited the biological activities of both IL-12 and IL-15.

[0238] Antitumor effect of IL-15-ABD-IL-12 in B16-F10 mouse melanoma model Without being bound by a particular theory of operation, IL-15 / IL-12 ABD is thought to provide synergistic biological activity. In particular, IL-12 increases the IL-15α receptor, IFN-γ, NK / T cells, and TH1 immunity, while downregulating Treg cells. IL-15 increases the IL-12β1 receptor and NK cells, while reducing the loss of memory in CD8 cells.

[0239] The antitumor effect and pseudosurvivability of IL12-ABD on IL15-ABD-IL12 were evaluated using a B16-F10 mouse melanoma model (Fig. 39).

[0240] On day 10 after inoculation with B16-F10 tumor cells (2x10 4 cells / mouse), animals (7 - 10 weeks old) were assigned to 8 groups (8 animals per group). The animals were assigned based on tumor volume. At the time of assignment, the average tumor volume per group was 100 mm 3 On day 0 (when the tumor reached 100 mm 3 ), each group received a single I.V. administration of PBS (placebo), or a molar equivalent dose of IL12-ABD (1.5 μg, 5 μg, 15 μg), or IL15-ABD-IL12 (1.7 μg, 6 μg, 17 μg).

[0241] As shown in Figure 39, IL-15-ABD-IL-12 had excellent antitumor activity compared to IL-12-ABD in the B16-F10 mouse model at equimolar concentrations. In other similar in vivo studies, free IL-12 (5 μg) combined with I-15 (1 μg) has been shown to have less than 50% of the potency of IL-15-ABD-IL-12 (6 μg) (data not shown).

[0242] Example 6: Anti-TGFβ-ABD After biopanning and screening using surface plasmon resonance technology, anti-hTGFβ1 binding domains were identified: 1A10, 1F11, 2H6, 4B9, 4C10, 4D9, 4G3, 4G6, 4H4, 4H7, and 6H11. These clones showed cross-reactivity to hTGFβ2, 3, and mTGFβ1 and potentially inhibited the binding of hTGFβ1 to its receptor II. Subsequently, the clones were selected for purification and further characterization as scFvs.

[0243] Using standard ELISA technology, the scFvs were screened for cross-reactivity to mouse and human TGFβ-1. Binding ELISA showed that 2H6, 4G3, 4H7, 4B9, 4D9, and 6H11 had good cross-reactivity to both mTGFβ-1 and hTGFβ-1.

[0244] Binding and blocking ELISAs were performed to determine whether the anti-TGFβ-1 scFvs could bind to TGFβ-1 and block its interaction with TGFβR-II. 2H6, 4G3, 4H7, 4B9, and 4D9 all showed good blocking effects and inhibition of the mTGFβ and mTGFβR-II interaction.

[0245] Some of these anti-TGFβ-1 scFvs were tested for their ability to interfere with the biological activity of TGFβ-1. The sequences of anti-TGFβ-1 scFvs 4H7 and 4D9 are shown in Figures 40A and B.

[0246] CD4+ Foxp3 + Blocking TGFβ1-induced proliferation of regulatory T cells Regulatory T cells (Tregs) can affect the homeostasis of the immune system. Such Tregs are essential for maintaining self-tolerance because defects can lead to severe autoimmune diseases. In cancer, tumor cells can secrete cytokines that affect the homeostasis of the immune system. In particular, tumor cells can secrete TGFβ, which can affect the number of circulating Tregs. Exposure to TGFβ1 is CD4 from CD4 + Foxp3 - CD4 from T + Foxp3 + It has previously been demonstrated to result in the proliferation of Treg subsets. These induced Tregs can contribute to the induction of anergic responses of T cells by inhibiting the activation of tumor antigen-specific cytotoxic CD8 + T cells. As shown in Figure 41, recombinant TGFβ1 can stimulate the proliferation of CD4 + Foxp3 + Tregs from a mixed T cell population isolated from healthy human donors. However, the blockade of TGFβ using anti-TGFβ1 D11 antibody or TGF-β1 scFv 2H6, 4H7, and 4D9 all significantly inhibits the TGFβ-induced proliferation of CD4 + Foxp3 + Tregs in a dose-dependent manner. Therefore, such TGFβ1 scFv is useful for reducing TReg proliferation in cancer.

[0247] Blocking TGFβ1-induced epithelial-to-mesenchymal transition (EMT) Exposure to TGFβ is known to induce epithelial-to-mesenchymal transition. During this process, epithelial cells transform from an organized, polarized, and tightly connected epithelial sheet of cobblestone-shaped cells into unorganized and motile cells with a mesenchymal-like morphology. During EMT, the invasive capacity of the cells is activated, thereby enhancing the tumor-forming ability of the cells. E-cadherin is a commonly used marker of epithelial cells and is localized at the adherens junctions between epithelial cells. Loss of E-cadherin is a strong marker of EMT indicating the differentiation conversion process. (Figure 42A) Furthermore, vimentin is associated with highly motile cells. Therefore, induction of vimentin expression in cells also indicates an increase in motility and an increase in local invasion in vivo.

[0248] As shown in Figures 42 and 43, blockade of TGF-β by anti-TGFβ1 D11 antibody or TGF-β1 scFv reverses TGF-β1-induced epithelial-to-mesenchymal transition (Figure 42) and migration (Figure 43). Mouse 4T1 cells were cultured in growth medium supplemented with TGF-β1 (panel 2), TGF-β1 and D11 (panel 3), or TGF-β1 and anti-TGF-β1 scFv (panel 4), then fixed and stained with E-cadherin antibody (green) and vimentin antibody (purple). Nuclei were counterstained with DAPI (blue). Treatment with TGF-β1 induced loss of E-cadherin from cell junctions and increased vimentin expression. This effect was reversed by the addition of D11 or the subject anti-TGF-β1 scFv described herein (Figure 42, panels 3 and 4). Furthermore, the anti-TGF-β1 scFv described herein can block TGF-β1-mediated cancer cell migration (Figure 43).

[0249] Neutralization of TGFβ1-induced Smad activation The TGFβ superfamily consists of multifunctional cytokines that regulate various biological processes, including cell proliferation, differentiation, migration, cell survival, angiogenesis, wound healing, and immune surveillance. In humans, the major isoform is TGFβ1, which is expressed in various tissue types. TGFβ inhibits the proliferation of most normal epithelial cells. Furthermore, at the early stages of cancers of epithelial origin, TGFβ functions as a cell growth inhibitor. Thus, at the onset of cancer, TGFβ functions as a tumor suppressor. However, in the later stages of cancer progression, tumor cells become resistant to the growth-inhibitory effects of TGFβ, and TGFβ plays a role as a tumor promoter. Indeed, TGFβ1 has been shown to be overexpressed in various tumors. Activation of the TGFβ pathway occurs through binding of the TGFβ ligand to the type II TGFβ receptor (TβRII), which then induces binding and oligomerization between TβRII and TβRI. Once this oligomer is formed, Smad2 and Smad3 are recruited and phosphorylated by TβRI. The phosphorylated Smad2 or Smad3 then binds to cytoplasmic Smad4, and this complex translocates to the nucleus, where it interacts with the promoter region to activate transcription of target genes. Thus, activation of the TGFβ pathway can be measured by the rapid phosphorylation of Smad2 after addition of TGFβ to serum-starved cells. However, effective blockade of TGFβ inhibits phosphorylation of Smad2. Here, the absence of Smad2 phosphorylation can be used as a measure of effective blockade of TGFβ by the subject anti-TGFβ scFv construct.

[0250] Using human cells (Figure 44A) or mouse cells (Figure 44B) in a serum - deficient state, it was determined that the addition of human recombinant TGFβ1 (Figure 44A) or mouse TGF - β1, - β2, and - β3 (Figure 44B) induces phosphorylation of Smad2. Such phosphorylation is dose - dependently reduced when TGFβ is pre - incubated with the control anti - TGFβ1 D11 antibody or TGF - β1 scFv constructs 2H6, 4H7, and 4D9. These data suggest that the 2H6, 4H7, and 4D9 scFv constructs can sequester TGFβ1 and inhibit the interaction with TβRII / TβRI, thereby inhibiting the TGFβ activation cascade among advanced - stage cancers.

[0251] 4D9 anti - TGFβ - 1 - ABD The sequences of exemplary TGF - β1 scFv - ABD constructs (4D9M - A6m and 4H7 - A6m) are shown in Figure 45A. As shown in Figure 45B, the anti - TGFβ - 1 - ABD extended the anti - TGFβ - 1 scFv T1 / 2β from 106 minutes to 10.6 hours.

[0252] Anti - TGF - β1 scFv - ABD (bivalent of TGF - β1) produced in E. coli and HEK cells was evaluated for binding to mouse serum albumin (Figure 47). Three different orientations of the constructs were evaluated. Two anti - TGF - β1 scFvs bound to the N - terminus of ABD (「Bi N - terminus」), two anti - TGF - β1 scFvs bound to the C - terminus of ABD (「Bi C - terminus」), or one anti - TGF - β1 scFv bound to each of the N - terminus and C - terminus of ABD (「Bi Mid」). As shown in Figure 46, all constructs showed binding to mouse serum albumin. For constructs produced in E. coli, Bi Mid showed better binding to MSA than the N - terminal orientation. For constructs produced in HEK cells, the Bi N - terminal orientation showed better binding to MSA than the Bi Mid or Bi C - terminal orientations.

[0253] As shown in FIG. 47, TGFβ-1-mediated inhibition of T cell proliferation was reversed by such constructs (i.e., T cell proliferation increased) (FIG. 46B). Furthermore, 4D9M-ABD has been shown to block human TGFβ-1 and human TGFβ-3 that binds to the cognate receptor (data not shown).

[0254] All cited references are hereby expressly incorporated by reference in their entirety herein.

[0255] While specific embodiments of the invention have been described above for purposes of illustration, it will be understood by those skilled in the art that numerous variations in detail may be made without departing from the invention as set forth in the appended claims.

Claims

**Claim 1** An IL-12 / mutated IL-15 / albumin-binding domain fusion protein comprising the amino acid sequence of SEQ ID NO:

225. **Claim 2** A nucleic acid encoding the IL-12 / mutated IL-15 / albumin-binding domain fusion protein according to Claim 1. **Claim 3** An expression vector comprising the nucleic acid according to Claim 2. **Claim 4** A host cell comprising the nucleic acid according to Claim 2 or the expression vector according to Claim 3. **Claim 5** A method for producing an IL-12 / mutated IL-15 / albumin-binding domain fusion protein, comprising: culturing the host cell according to Claim 4 under conditions under which the IL-12 / mutated IL-15 / albumin-binding domain fusion protein is expressed; recovering the IL-12 / mutated IL-15 / albumin-binding domain fusion protein. A method comprising the above steps.

Images