PD-1-targeted il-15 / il-15rα fc fusion proteins with improved properties

Abstract

The present invention relates to fusion proteins comprising a variant IL-15 protein, fusion proteins comprising a variant anti-PD-1 antigen binding domain, and fusion proteins comprising a variant IL-15 protein and a variant anti-PD-1 antigen binding domain. The present invention also relates to nucleic acid molecules, expression vectors, host cells, and methods for making such fusion proteins, and the use of such fusion proteins in the treatment of cancer.

Description

[0001] Cross-reference to related applications

[0002] This application claims priority to U.S. Provisional Application No. 62 / 914,265, filed October 11, 2019, U.S. Provisional Application No. 62 / 914,317, filed October 11, 2019, and U.S. Provisional Application No. 63 / 011,208, filed April 16, 2020, the contents of which are incorporated herein by reference in their entirety.

[0003] sequence list

[0004] This application contains a sequence list, which has been submitted electronically in ASCII format and is incorporated herein by reference in its entirety. The ASCII copy was created on October 9, 2020, and is named 000218-0005-WO1_-_SEQUENCE_LISTING.TXT, with a size of 764,863 bytes. Background Technology

[0005] In cancer immunotherapy, two very promising approaches include cytokine-based therapy and the blocking of immune checkpoint proteins such as PD-1.

[0006] Cytokines such as IL-2 and IL-15 contribute to the proliferation and differentiation of B cells, T cells, and NK cells. Both cytokines exert their cellular signaling function by binding to a trimeric complex composed of two common receptors (a common γ chain (γc; CD132) and an IL-2 receptor β chain (IL-2Rβ; CD122)) and a specific α chain receptor for each cytokine (IL-2 receptor α (IL-2Rα; CD25) or IL-15 receptor α (IL-15Rα; CD215)). Both cytokines are considered potentially valuable therapeutic agents in oncology, and IL-2 has been approved for the treatment of patients with metastatic renal cell carcinoma and malignant melanoma. Currently, although several clinical trials are underway, recombinant IL-15 has not been approved for use. However, as potential drugs, both cytokines have very rapid clearance rates, with half-lives measured in minutes. High-dose IL-2 immunotherapy to overcome the rapid clearance problem can cause systemic toxicity. This systemic toxicity of IL-15 immunotherapy has also been documented in recent clinical trials (Guo et al., J Immunol, 2015, 195(5):2353-64).

[0007] Following T cell activation, immune checkpoint proteins such as PD-1 are upregulated, thereby depleting activated T cells through binding to immune checkpoint ligands such as PD-L1, thus preventing autoimmunity. However, immune checkpoint proteins are also upregulated in tumor-infiltrating lymphocytes (TILs), and immune checkpoint ligands are overexpressed on tumor cells, thereby promoting immune escape by tumor cells. Studies have shown that drugs can help prevent this. (nivolumab) and (Pembrolizumab) Blocking immune checkpoint interactions to eliminate TIL suppression can be an effective treatment for cancer. Although checkpoint blockade therapies such as nivolumab and pembrolizumab are promising, many patients still do not achieve a sufficient response to checkpoint blockade alone.

[0008] Therefore, therapeutic strategies incorporating cytokines remain necessary in cancer treatment, offering antitumor effects without the need for high-dose administration and avoiding systemic toxicity. Furthermore, it is crucial to identify other therapeutic modalities that combine with checkpoint blockade to improve patient response rates. This can be particularly complex, as other therapeutic modalities should not compete with checkpoint blockade agents. This invention addresses these needs and concerns by providing PD-1-targeting IL-15 fusion proteins with extended half-lives and higher selectivity in targeting TILs to improve safety profiles, and these do not compete with checkpoint blockade antibodies that may bind to them. Summary of the Invention

[0009] A first aspect of the present invention provides a targeting IL-15 / Rα heterodimer Fc fusion protein comprising: a) a first monomer comprising, from the N-terminus to the C-terminus: i) an IL-15 / Rαsushi domain; ii) a first domain linker; iii) an IL-15 domain; and iv) a first variant Fc domain; b) a second monomer comprising a heavy chain comprising VH-CH1-hinge-CH2-CH3, wherein the CH2-CH3 is the second variant Fc domain; c) a third monomer comprising a light chain comprising VL-CL; wherein the VH and VL domains form an antigen-binding domain (ABD) that binds to human PD-1, wherein, according to the Kabat number, the VH is a variant variable heavy domain comprising an F32L / W100F amino acid substitution, compared to SEQ ID NO:5; and wherein, compared to SEQ ID NO:5, the VH is a variant variable heavy domain comprising an F32L / W100F amino acid substitution; and wherein, compared to SEQ ID NO:5, the VH is a variant variable heavy domain. NO:168, according to the Kabat number, this VL is a variant variable light structure domain containing N27dH / K30Y / S93T.

[0010] In some embodiments, the VH comprises the amino acid sequence of SEQ ID NO:318, and the VL comprises the amino acid sequence of SEQ ID NO:176.

[0011] In some embodiments, the IL-15 domain is a variant IL-15 domain comprising amino acid substitutions selected from the group consisting of: D30N / E64Q / N65D, N1D / N4D / D8N, N1D / N4D / N65D, N1D / D30N, N1D / D61N, N1D / D61N / E64Q / Q108E, N1D / E64Q, N1D / N65D, N1D / Q108E, N4D / D3 0N, N4D / D61N, N4D / D61N / N65D, N4D / D61N / E64Q / Q108E, N4D / E64Q, N4D / N65D, D8N / D61N, D8N / E64Q, D30N / E64Q, D30N / N65D, D30N / Q180E, D61N / E64Q / N65D, E64Q / N65D, E64Q / Q108E, and N65D / Q108E. In some embodiments, the IL-15 domain is a variant IL-15 domain comprising amino acid substitutions selected from the group consisting of: N71Q, N79Q, N112Q, S114del, and S114A, or combinations thereof. In some embodiments, the IL-15 domain is a variant IL-15 domain comprising amino acid substitutions D30N / E64Q / N65D and N71Q / N79Q / N112Q.

[0012] In some embodiments, the first variant Fc structural domain includes all or part of the hinge structural domain. In some embodiments, the first monomer further includes a second structural domain joint between the IL-15 structural domain and the first variant Fc structural domain.

[0013] A second aspect of the present invention provides a targeting IL-15 / Rα heterodimer Fc fusion protein comprising: a) a first monomer comprising: i) an IL-15Rαsushi domain protein; ii) a first domain linker; iii) a variant IL-15 protein comprising, compared to SEQ ID NO:2, amino acid substitutions D30N / E64Q / N65D and N71Q / N79Q / N112Q; and iv) a first variant Fc domain; and b) a second monomer comprising a heavy chain comprising VH-CH1-hinge-CH2-CH3, wherein the CH2-CH3 is the second variant Fc domain; and c) a third monomer comprising a light chain comprising VL-CL; wherein the VH and VL domains form an antigen-binding domain (ABD) that binds to human PD-1.

[0014] In some embodiments, VH comprises the amino acid sequence of SEQ ID NO:5, and VL comprises the amino acid sequence of SEQ ID NO:168. In some embodiments, VH comprises the amino acid sequence of SEQ ID NO:318, and VL comprises the amino acid sequence of SEQ ID NO:176. In some embodiments, ABD does not competitively bind to the human PD-1 with nivolumab and / or pembrolizumab.

[0015] In some embodiments, the first variant Fc structural domain includes all or part of the hinge structural domain. In some embodiments, the first monomer further includes a second structural domain joint between the IL-15 structural domain and the first variant Fc structural domain.

[0016] A third aspect of the present invention provides a targeting IL-15 / Rα heterodimer Fc fusion protein comprising: a) a first monomer comprising: i) an IL-15Rαsushi domain protein; ii) a first domain linker; iii) a variant IL-15 protein comprising, compared to SEQ ID NO:2, amino acid substitutions D30N / E64Q / N65D and N71Q / N79Q / N112Q; and iv) a first variant Fc domain; and b) a second monomer comprising a heavy chain comprising VH-CH1-hinge-CH2-CH3, wherein the CH2-CH3 is the second variant Fc domain; c) a third monomer comprising a light chain comprising VL-CL; wherein the VH and VL domains form an antigen-binding domain (ABD) that binds to human PD-1, wherein, compared to SEQ ID NO:2, NO:5, according to the Kabat number, this VH is a variant variable heavy domain containing an amino acid substitution of F32L / W100F; and wherein, compared to SEQ ID NO:168, according to the Kabat number, this VL is a variant variable light domain containing N27dH / K30Y / S93T.

[0017] In some embodiments, VH comprises the amino acid sequence of SEQ ID NO:318, and VL comprises the amino acid sequence of SEQ ID NO:176.

[0018] In some embodiments, the first variant Fc structural domain includes all or part of the hinge structural domain. In some embodiments, the first monomer further includes a second structural domain joint between the IL-15 structural domain and the first variant Fc structural domain.

[0019] A fourth aspect of the present invention provides a targeting IL-15 / Rα heterodimer Fc fusion protein comprising: a) a first monomer comprising, from the N-terminus to the C-terminus: i) an IL-15Rαsushi domain protein; ii) a first domain linker; iii) a variant IL-15 protein comprising, compared to SEQ ID NO:2, an amino acid substitution selected from the group consisting of: N71Q, N79Q, N112Q, S114del, and S114A; and iv) a first variant Fc domain; and b) a second monomer comprising a heavy chain comprising VH-CH1-hinge-CH2-CH3, wherein the CH2-CH3 is the second variant Fc domain; c) a third monomer comprising a light chain comprising VL-CL; wherein the VH and VL domains form an antigen-binding domain that binds to human PD-1; wherein, according to the Kabat number, compared to SEQ ID NO:2, the first variant IL-15 / Rα heterodimer Fc domain comprises: a first monomer comprising ... NO:5, the VH is a variant variable recombinant domain comprising the W100F amino acid substitution and at least one additional amino acid substitution selected from the group consisting of: F32L, S52aG, R97E, R97Y, R97W, L98R, S100aT, R97A, V99T, V99L, S100aA, L98Q, R97Q, V99F, V99L, S100 aN, V99I, P100bS, G96H, L98V, V99A, V99Q, G96V, R97K, L98S, L98F, R97T, L98K, L98S, V99I, R97L, G96A, R97A, V99S, R97S, V99Y, R97H, L98R; wherein, the VL structural domain is selected from the group consisting of the following items: i) SEQ ID NO:168; and ii) the variant light chain domain, which, according to the Kabat number, compared to SEQ ID NO:168, contains amino acid substitutions selected from the group consisting of: N27dH, N27dS, K30Y, S93T, and Y94W; wherein the variant IL-15 protein contains amino acid substitutions selected from the group consisting of: N71Q / N79Q, N71Q / N79Q / N112Q, N71Q / N79Q / S114del, and N71Q / N79Q / S114A.

[0020] In some embodiments, the variant IL-15 protein further comprises amino acid substitutions selected from the group consisting of: D30N / E64Q / N65D, N1D / N4D / D8N, N1D / N4D / N65D, N1D / D30N, N1D / D61N, N1D / D61N / E64Q / Q108E, N1D / E64Q, N1D / N65D, N1D / Q108E, N4D / D30N, N 4D / D61N, N4D / D61N / N65D, N4D / D61N / E64Q / Q108E, N4D / E64Q, N4D / N65D, D8N / D61N, D8N / E64Q, D30N / E64Q, D30N / N65D, D30N / Q180E, D61N / E64Q / N65D, E64Q / N65D, E64Q / Q108E and N65D / Q108E.

[0021] In some embodiments, the variant light domain comprises the amino acid sequence of SEQ ID NO:318, and the variant light domain comprises the amino acid sequence of SEQ ID NO:176.

[0022] In some embodiments, the variant IL-15 protein comprises amino acid substitutions N71Q / N79Q / N112Q. In some embodiments, the variant IL-15 protein comprises amino acid substitutions N71Q / N79Q / N112Q and D30N / N65D. In some embodiments, the variant IL-15 protein comprises amino acid substitutions N71Q / N79Q / N112Q and D30N / E64Q / N65D. In some embodiments, the variant heavy domain is H1.176 (SEQ ID NO:318), the variant light domain is L1.140 (SEQ ID NO:176), and the variant IL-15 protein comprises amino acid substitutions N71Q / N79Q / N112Q and D30N / N65D. In some embodiments, the variant heavy domain is H1.176 (SEQ ID NO:318), the variant light domain is L1.140 (SEQ ID NO:176), and the variant IL-15 protein contains amino acid substitutions N71Q / N79Q / N112Q and D30N / E64Q / N65D.

[0023] In some embodiments, the first domain connector comprises a GGGGA (SEQ ID NO: 8). In some embodiments, the first variant Fc domain comprises all or part of a hinge domain. In some embodiments, the first monomer further comprises a second domain connector between the IL-15 domain and the first variant Fc domain.

[0024] A fifth aspect of the present invention provides a targeting IL-15 / Rα heterodimer Fc fusion protein, the fusion protein comprising: a) a first monomer comprising, from the N-terminus to the C-terminus: i) an IL-15Rαsushi domain protein; ii) a first domain linker; iii) a variant IL-15 protein; and iv) a first variant Fc domain; and b) a second monomer comprising a heavy chain comprising VH-CH1-hinge-CH2-CH3, wherein the CH2-CH3 is the second variant Fc domain; c) a third monomer comprising a light chain comprising VL-CL; wherein the VH and VL domains form an antigen-binding domain that binds to human PD-1; wherein, according to the Kabat number, compared to SEQ ID NO:5, the VH is a variant variable recombinant domain comprising the W100F amino acid substitution and at least one additional amino acid substitution selected from the group consisting of: F32L, S52aG, R97E, R97Y, R97W, L98R, S100aT, R97A, V99T, V99L, S100aA, L98Q, R97Q, V99F, V99L, S100 aN, V99I, P100bS, G96H, L98V, V99A, V99Q, G96V, R97K, L98S, L98F, R97T, L98K, L98S, V99I, R97L, G96A, R97A, V99S, R97S, V99Y, R97H, L98R; wherein the VL domain is selected from the group consisting of: i) SEQ ID NO:168; and ii) a variant light chain domain, which, according to the Kabat number, contains amino acid substitutions selected from the group consisting of: N27dH, N27dS, K30Y, S93T, and Y94W, compared to SEQ ID NO:168.

[0025] In some embodiments, the variable restructure domain is selected from the group consisting of: H1.176, H1.177, H1.178, H1.179, H1.180, H1.181, H1.182, H1.183, H1.184, H1.185, H1.186, H1.187, H1.188, H1.189, H1.190, H1.191, H1.192, H1.193, H1.194, H1.195, H1.196, H1.197, H1.19 8. H1.199, H1.200, H1.201, H1.202, H1.203, H1.204, H1.205, H1.206, H1.207, H1.208, H1.209, H1.210, H1.211 , H1.212, H1.213, H1.214, H1.215, H1.216, H1.217, H1.218, H1.219, H1.220, H1.221, H1.222, H1.223 and H1.224. In some embodiments, the variant light domain is selected from the group consisting of: L1.1, L1.3, L1.45, L1.117, L1.129, L1.135, L1.136, and L1.140. In some embodiments, the variant heavy domain comprises the amino acid sequence of SEQ ID NO:318, and the variant light domain comprises the amino acid sequence of SEQ ID NO:176.

[0026] In some embodiments, the first variant Fc structural domain includes all or part of the hinge structural domain. In some embodiments, the first monomer further includes a second structural domain joint between the IL-15 structural domain and the first variant Fc structural domain.

[0027] A sixth aspect of the present invention provides a targeting IL-15 / Rα heterodimer Fc fusion protein comprising: a) a first monomer comprising, from the N-terminus to the C-terminus: i) an IL-15Rαsushi domain protein; ii) a first domain linker; iii) a variant IL-15 protein comprising, compared to SEQ ID NO:2, amino acid substitutions selected from the group consisting of: N71Q, N79Q, N112Q, S114del, and S114A; and iv) a first variant Fc domain; and b) a second monomer comprising a heavy chain comprising VH-CH1-hinge-CH2-CH3, wherein the CH2-CH3 is the second variant Fc domain; c) a third monomer comprising a light chain comprising VL-CL; wherein the VH and VL domains form an antigen-binding domain that binds to human PD-1 and does not competitively bind to human PD-1 with nivolumab and / or pembrolizumab.

[0028] In some embodiments, the variant IL-15 protein comprises amino acid substitutions selected from the group consisting of: N71Q / N79Q, N71Q / N79Q / N112Q, N71Q / N79Q / S114del, and N71Q / N79Q / S114A. In some embodiments, the variant IL-15 protein comprises amino acid substitutions selected from the group consisting of: N1D / N4D / D8N, N1D / N4D / N65D, N1D / D30N, N1D / D61N, N1D / D61N / E64Q / Q108E, N1D / E64Q, N1D / N65D, N1D / Q108E, N4D / D30N, N4D / D61N, N4D / D61N / N65D, N4D / D61N / E64Q / Q108E, N4D / E64Q, N4D / N65D, D8N / D61N, D8N / E64Q, D30N / E64Q, D30N / N65D, D30N / E64Q / N65D, D30N / Q180E, D61N / E64Q / N65D, E64Q / N65D, E64Q / Q108E, and N65D / Q108E. In some embodiments, the variant IL-15 protein comprises amino acid substitutions for N71Q / N79Q / N112Q. In some embodiments, the variant IL-15 protein comprises amino acid substitutions for N71Q / N79Q / N112Q and D30N / N65D. In some embodiments, the variant IL-15 protein comprises amino acid substitutions N71Q / N79Q / N112Q and D30N / E64Q / N65D.

[0029] In some embodiments, the first domain connector is GGGGA (SEQ ID NO:8). In some embodiments, the first variant Fc domain includes all or part of the hinge domain. In some embodiments, the first monomer further includes a second domain connector between the IL-15 domain and the first variant Fc domain.

[0030] In some embodiments of any of the foregoing aspects, according to EU designations, the first and second variants' Fc domains comprise amino acid substitutions selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411E / K360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W, and T366S / L368A / Y407V / Y349C:T366W / S354C. In some embodiments of any of the foregoing aspects, according to EU designations, the first variant Fc domain comprises L368D / K370S and the second variant Fc domain comprises S364K / E357Q.

[0031] In some embodiments of any of the foregoing aspects, according to EU designations, the first and second variant Fc domains each independently comprise amino acid substitutions selected from the group consisting of: G236R / L328R, E233P / L234V / L235A / G236del / S239K, E233P / L234V / L235A / G236del / S267K, E233P / L234V / L235A / G236del / S239K / A327G, E233P / L234V / L235A / G236del / S267K / A327G, and E233P / L234V / L235A / G236del. In some embodiments of any of the foregoing aspects, according to EU designations, the first and second variants each contain amino acid substitutions E233P / L234V / L235A / G236del / S267K in their Fc domains.

[0032] In some embodiments of any of the foregoing aspects, according to EU designations, the first Fc domain comprises amino acid substitutions Q295E / N384D / Q418E / N481D.

[0033] In some embodiments of any of the foregoing aspects, according to EU designations, the first variant Fc domain and the second variant Fc domain each contain an amino acid substitution M428L / N434S.

[0034] In some embodiments of any of the foregoing aspects, the first monomer comprises the amino acid sequence of SEQ ID NO:225. In some embodiments of any of the foregoing aspects, the second monomer comprises the amino acid sequence of SEQ ID NO:244. In some embodiments of any of the foregoing aspects, the third monomer comprises the amino acid sequence of SEQ ID NO:196. In some embodiments of any of the foregoing aspects, the first monomer comprises the amino acid sequence of SEQ ID NO:225, the second monomer comprises the amino acid sequence of SEQ ID NO:244, and the third monomer comprises the amino acid sequence of SEQ ID NO:196.

[0035] A seventh aspect of the invention provides a composition comprising an anti-PD-1 antigen-binding domain (ABD), the anti-PD-1 antigen-binding domain comprising: a) a variant variable-recombination domain, which, according to the Kabat number, is compared to SEQ ID NO. NO:5, comprising the W100F amino acid substitution and at least one additional amino acid substitution selected from the group consisting of: F32L, S52aG, R97E, R97Y, R97W, L98R, S100aT, R97A, V99T, V99L, S100aA, L98Q, R97Q, V99F, V99L, S100aN, V99I, P100bS, G96H, L98V, V99A, V99Q, G96V, R97K, L98S, L98F, R97T, L98K, L98S, V99I, R97L, G96A, R97A, V99S, R97S, V99Y, R97H, L98R; and b) a variable light domain selected from the group consisting of: i) SEQ ID NO:168; and ii) variant light chain domain, which, according to the Kabat number, compared to SEQ ID NO:168, contains amino acid substitutions selected from the group consisting of: N27dH, N27dS, K30Y, S93T and Y94W; wherein the ABD binds to human PD-1.

[0036] In some embodiments, according to the Kabat number, the variant heavy domain has an amino acid substitution F32L / W100F; and according to the Kabat number, the variant light domain has an amino acid substitution N27dH / K30Y / S93T.

[0037] In some embodiments, the variable restructure domain is selected from the group consisting of: H1.176, H1.177, H1.178, H1.179, H1.180, H1.181, H1.182, H1.183, H1.184, H1.185, H1.186, H1.187, H1.188, H1.189, H1.190, H1.191, H1.192, H1.193, H1.194, H1.195, H1.196, H1.197, H1.19 8. H1.199, H1.200, H1.201, H1.202, H1.203, H1.204, H1.205, H1.206, H1.207, H1.208, H1.209, H1.210, H1.211 , H1.212, H1.213, H1.214, H1.215, H1.216, H1.217, H1.218, H1.219, H1.220, H1.221, H1.222, H1.223 and H1.224. In some embodiments, the variant light domain is selected from the group consisting of: L1.1, L1.3, L1.45, L1.117, L1.129, L1.135, L1.136, and L1.140. In some embodiments, the variant heavy domain comprises the amino acid sequence of SEQ ID NO:318, and the variant light domain comprises the amino acid sequence of SEQ ID NO:176.

[0038] In some embodiments, the composition comprises a full-length anti-PD-1 antibody. In some embodiments, the composition comprises a fusion protein. In some embodiments, the fusion protein is XENP32435.

[0039] The eighth aspect of the invention provides a composition comprising, compared to SEQ ID NO:2, a variant IL-15 protein; the variant IL-15 protein comprising amino acid substitutions selected from the group consisting of: N71Q, N79Q, N112Q, S114del, and S114A.

[0040] In some embodiments, the variant IL-15 protein further comprises amino acid substitutions selected from the group consisting of: N1D, N4D, D8N, D30N, D61N, E64Q, N65D, and Q108E.

[0041] In some embodiments, the variant IL-15 protein comprises amino acid substitutions selected from the group consisting of: N71Q / N79Q, N71Q / N79Q / N112Q, N71Q / N79Q / S114del, and N71Q / N79Q / S114A. In some embodiments, the variant IL-15 protein comprises amino acid substitutions selected from the group consisting of: N1D / N4D / D8N, N1D / N4D / N65D, N1D / D30N, N1D / D61N, N1D / D61N / E64Q / Q108E, N1D / E64Q, N1D / N65D, N1D / Q108E, N4D / D30N, N4D / D61N, N4D / D61N / N65D, N The variants are 4D / D61N / E64Q / Q108E, N4D / E64Q, N4D / N65D, D8N / D61N, D8N / E64Q, D30N / E64Q, D30N / N65D, D30N / E64Q / N65D, D30N / Q180E, D61N / E64Q / N65D, E64Q / N65D, E64Q / Q108E, and N65D / Q108E. In some embodiments, the variant IL 15 protein comprises amino acid substitutions for N71Q / N79Q / N112Q. In some embodiments, the variant IL 15 protein comprises amino acid substitutions for N71Q / N79Q / N112Q and D30N / N65D. In some embodiments, the variant IL-15 protein comprises amino acid substitutions N71Q / N79Q / N112Q and D30N / E64Q / N65D.

[0042] A ninth aspect of the present invention provides a heterodimeric protein comprising: a) a first fusion protein comprising: i) a variant IL-15 protein comprising, compared to SEQ ID NO:2, amino acid substitutions D30N / E64Q / N65D and N71Q / N79Q / N112Q; ii) a domain linker; and iii) a first variant Fc domain; and b) a second fusion protein comprising: i) an IL-15Rαsushi domain; ii) a domain linker; and iii) a second variant Fc domain.

[0043] In some embodiments, the variant IL-15 protein comprises the amino acid sequence of SEQ ID NO:319.

[0044] In some embodiments, according to EU designations, the first and second variant Fc domains comprise amino acid substitutions selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411E / K360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W, and T366S / L368A / Y407V / Y349C:T366W / S354C. In some embodiments, according to the EU designation, the first variant Fc structure domain includes L368D / K370S and the second variant Fc structure domain includes S364K / E357Q.

[0045] In some embodiments, the first and second variant Fc domains contain 428L / 434S.

[0046] In some embodiments, the first fusion protein comprises the amino acid sequence of SEQ ID NO:208, and the second fusion protein comprises the amino acid sequence of SEQ ID NO:95. In some embodiments, the first fusion protein comprises the amino acid sequence of SEQ ID NO:211, and the second fusion protein comprises the amino acid sequence of SEQ ID NO:206.

[0047] Other aspects of the invention provide nucleic acid molecules, nucleic acid compositions, expression vectors, expression vector compositions, host cells, and methods for expressing: (a) any of the above-described fusion proteins; (b) any of the above-described anti-PD-1ABD proteins; (c) any of the above-described compositions comprising a variant IL-15 protein; or (d) any of the above-described heterodimeric proteins.

[0048] Other aspects of the invention provide a pharmaceutical composition comprising a pharmaceutical carrier and (a) any of the above-described fusion proteins; (b) any of the above-described anti-PD-1ABD proteins; (c) any of the above-described compositions comprising a variant IL-15 protein or (d) any of the above-described heterodimeric proteins.

[0049] Other aspects of the invention provide a method of treating a subject with cancer, comprising administering to a subject in need a therapeutically effective amount of (a) any of the above-described fusion proteins; (b) any of the above-described anti-PD-1ABD; (c) any of the above-described compositions comprising a variant IL-15 protein; (d) any of the above-described heterodimeric proteins; or (e) any of the above-described pharmaceutical compositions.

[0050] In some embodiments, the method further includes administering a therapeutically effective amount of a checkpoint blocking antibody. In some embodiments, the checkpoint blocking antibody is an anti-PD-1 antibody or an anti-PD-L1 antibody. In some embodiments, the checkpoint blocking antibody is nivolumab or pembrolizumab.

[0051] Other aspects of the invention provide for use in the preparation of a medicament for treating cancer in a subject with such need: (a) any of the above-described fusion proteins; (b) any of the above-described anti-PD-1ABD proteins; (c) any of the above-described compositions comprising a variant IL-15 protein; (d) any of the above-described heterodimeric proteins; or (e) any of the above-described pharmaceutical compositions.

[0052] In some embodiments, the drug is formulated in combination with a therapeutically effective amount of a checkpoint blocking antibody for administration. In some embodiments, the checkpoint blocking antibody is an anti-PD-1 antibody or an anti-PD-L1 antibody. In some embodiments, the checkpoint blocking antibody is nivolumab or pembrolizumab.

[0053] Other aspects of the invention provide for treating cancer in subjects in need of treatment with: (a) any of the above-described fusion proteins; (b) any of the above-described anti-PD-1ABD proteins; (c) any of the above-described compositions comprising a variant IL-15 protein; (d) any of the above-described heterodimeric proteins; or (e) any of the above-described pharmaceutical compositions.

[0054] In some embodiments, a fusion protein anti-PD-1ABD, a composition comprising a variant IL-15 protein, a heterodimeric protein, or a pharmaceutical composition is administered in combination with a therapeutically effective amount of a checkpoint blocking antibody. In some embodiments, the checkpoint blocking antibody is an anti-PD-1 antibody or an anti-PD-L1 antibody. In some embodiments, the checkpoint blocking antibody is nivolumab or pembrolizumab. Attached Figure Description

[0055] The patent or application document contains at least one color drawing. Upon request and payment of the necessary fees, the patent office will provide a published copy of the patent or application document with one or more color drawings.

[0056] Figure 1A and Figure 1B The sequences of IL-15 and its receptor are shown.

[0057] Figure 2 The sequences of PD-1 in humans and cynomolgus monkeys are shown, which facilitate the formation of antigen-binding domains to bind to both, thus enabling clinical development.

[0058] Figures 3A-3EThe available Fc heterodimerization variants (including skew and pI variants) are shown. Some variants do not have a corresponding "monomer 2" variant; these are pI variants and can be used alone on any monomer.

[0059] Figure 4 A list of isosteric variant antibody constant regions and their corresponding substitutions is shown. pI_(-) indicates variants with lower pI, while pI_(+) indicates variants with higher pI. These can optionally and independently be combined with other heterodimerizing variants of the invention (and other variant types outlined herein).

[0060] Figure 5 Useful ablation variants (sometimes referred to as "knockout" or "KO" variants) that can ablate FcγR binding are shown. Typically, ablation variants are present on both monomers, although in some cases they may be present on only one monomer.

[0061] Figures 6A-6E This section shows a particularly useful example of the “non-cytokine” component of the IL-15 / Rα-Fc fusion protein of the present invention.

[0062] Figures 7A-7F This section shows a particularly useful embodiment of the “non-cytokine” / “non-Fv” component of the IL-15 / Rαx anti-PD-1 bifunctional protein of the present invention.

[0063] Figure 8 Numerous exemplary variable-length domain connectors for IL-15 / Rα-Fc fusion proteins are shown. In some embodiments, these domain connectors can be used to attach the C-terminus of IL-15 and / or IL-15Rα(sushi) to the N-terminus of the Fc region. In some embodiments, these domain connectors can be used to fuse IL-15 to IL-15Rα(sushi). In some embodiments, these domain connectors can be used to attach a single-stranded Fv to the Fc chain. In some embodiments, the domain connectors are scFv connectors for connecting the VH and VL domains and may optionally be charged. In some embodiments, these connectors may be used in combination. For example, the GGGGS connector may be used in combination with a “semi-hinge” connector.

[0064] Figure 9Numerous charged scFv linkers are shown, which can be used to increase or decrease the pI of heterodimeric antibodies utilizing one or more scFvs as components. (+H)-positive linkers are particularly useful in this paper. A single prior art scFv linker with a single charge is referred to as “Whitlow,” as described by Whitlow et al. in Protein Engineering 6(8):989-995 (1993). It should be noted that this linker is used to reduce aggregation in scFvs and enhance proteolytic stability. In some embodiments, these linkers are used to attach the C-terminus of IL-15 and / or IL-15Rα(sushi) to the N-terminus of the Fc region; and / or to fuse IL-15 to IL-15 / Rα(sushi).

[0065] Figures 10A-10DSequences of several useful IL-15 / Rα-Fc backbones based on human IgG1 are shown, which do not contain cytokine sequences (e.g., IL-15 and / or IL-15Rα(sushi)). Notably, these backbones can also be used in certain embodiments of PD-1-targeting IL-15 / Rα-Fc fusion proteins. Backbone 1 is based on human IgG1 (356E / 358M allotype) and contains C220S, S364K / E357Q:L368D / K370S skewed variants (Q295E / N384D / Q418E / N421D pI variants containing L368D / K370S skewed variants) on both strands, as well as E233P / L234V / L235A / G236del / S267K ablation variants on both strands. Main chain 2 is based on human IgG1 (356E / 358M isotype) and contains C220S, S364K:L368D / K370S skewed variants (Q295E / N384D / Q418E / N421D pI variants containing L368D / K370S skewed variants) and E233P / L234V / L235A / G236del / S267K ablation variants on both chains. Main chain 3 is based on human IgG1 (356E / 358M isotype) and contains C220S, S364K:L368E / K370S skewed variants (Q295E / N384D / Q418E / N421D pI variants containing L368E / K370S skewed variants) and E233P / L234V / L235A / G236del / S267K ablation variants on both chains. Main chain 4 is based on human IgG1 (356E / 358M allotype) and contains C220S, D401K:K360E / Q362E / T411E skewed variants (Q295E / N384D / Q418E / N421D pI variants containing K360E / Q362E / T411E skewed variants) on both chains and E233P / L234V / L235A / G236del / S267K ablation variants on both chains. Main chain 5 is based on human IgG1 (356D / 358L allotypes) and contains C220S, S364K / E357Q:L368D / K370S skewed variants (Q295E / N384D / Q418E / N421D pI variants containing L368D / K370S skewed variants) and E233P / L234V / L235A / G236del / S267K ablation variants on both chains.Main strand 6 is based on human IgG1 (356E / 358M allotype) and contains the C220S, S364K / E357Q:L368D / K370S skewed variant (containing the Q295E / N384D / Q418E / N421D pI variant of the L368D / K370S skewed variant) on both strands, the E233P / L234V / L235A / G236del / S267K ablation variant, and the N297A variant on both strands. Main strand 7 is identical to 6, except that it is mutated to N297S. Alternative forms of main strands 6 and 7 can exclude the ablation variant E233P / L234V / L235A / G236del / S267K from both strands. Main chain 8 is based on human IgG4 and includes the S364K / E357Q:L368D / K370S skewed variant (Q295E / N384D / Q418E / N421D pI variant with L368D / K370S skewed variant on the chain) and the S228P variant (designated S241P according to the EU number) on both chains (which ablates the Fab arm interchange in a manner known in the art). Main chain 9 is based on human IgG2 and includes the S364K / E357Q:L368D / K370S skewed variant (Q295E / N384D / Q418E / N421D pI variant with L368D / K370S skewed variant on the chain). Main chain 10 is based on human IgG2 and includes the S364K / E357Q:L368D / K370S skewed variant (Q295E / N384D / Q418E / N421D pI variant with L368D / K370S skewed variant on the chain) and the S267K variant on both chains. Main chain 11 is the same as main chain 1, except that it includes the M428L / N434S Xtend mutation. Main chain 12 is based on human IgG1 (356E / 358M isotype) and includes the C220S on both identical chains and the E233P / L234V / L235A / G236del / S267K ablation variant on both identical chains. Main chain 13 is based on human IgG1 (356E / 358M isotype) and contains C220S, S364K / E357Q:L368D / K370S skewed variants (containing P217R / P228R / N276K pI variants of the S364K / E357Q skewed variant) and E233P / L234V / L235A / G236del / S267K ablation variants on both chains.

[0066] As will be understood by those skilled in the art and as outlined below, these sequences can be used with any IL-15 and IL-15Rα(sushi) pairs outlined herein, including but not limited to those described below. Figure 14IL-15 / Rα-heteroFc, ncIL-15 / Rα, and scIL-15 / Rα are schematically shown in A–14G. Additionally, any IL-15 and / or IL-15Rα (sushi) variants can be incorporated into these in any combination. Figures 10A-10C In the main chain shown.

[0067] Each of these backbones includes a sequence that is 90%, 95%, 98%, and 99% identical (as defined herein) to the listed sequences, and / or contains 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 additional amino acid substitutions (which, as those skilled in the art will understand, already contain many amino acid modifications compared to the parent IgG1 (or IgG2 or IgG4, depending on the backbone)). That is, in addition to the skewed, pI, and ablation variants contained in the backbones shown in the figure, the listed backbones may also contain other amino acid modifications (typically amino acid substitutions).

[0068] Figure 11 The sequence of the backbone for several useful PD-1 targeting IL-15 / Rα-Fc fusion forms based on human IgG1 is shown, without cytokine sequences (e.g., IL-15 and / or IL-15Rα(sushi)) or VH, and further excludes... Figure 12The homologous light chain backbone is shown. Backbone 1 is based on human IgG1 (356E / 358M allotype) and includes the S364K / E357Q:L368D / K370S skewed variant (containing the C220S and Q295E / N384D / Q418E / N421D pI variants of the L368D / K370S skewed variant) and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Main chain 2 is based on human IgG1 (356E / 358M allotype) and includes the S364K / E357Q:L368D / K370S skewed variant (N208D / Q295E / N384D / Q418E / N421D pI variant containing the L368D / K370S skewed variant), the C220S variant containing the S364K / E357Q variant, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Main chain 3 is based on human IgG1 (356E / 358M allotype) and includes the S364K / E357Q:L368D / K370S skewed variant (N208D / Q295E / N384D / Q418E / N421D pI variant containing the L368D / K370S skewed variant), the Q196K / I199T / P217R / P228R / N276K pI variant containing the S364K / E357Q variant, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains.

[0069] In some embodiments, these sequences may be 356D / 358L allotypes. In other embodiments, these sequences may include N297A or N297S substitutions. In some other embodiments, these sequences may include the M428L / N434S Xtend mutation. In other embodiments, these sequences may alternatively be based on human IgG4 and include the S228P (S241P in Kabat numbering according to EU numbering) variant on both strands, which, as is known in the art, eliminates Fab arm exchanges. In still other embodiments, these sequences may alternatively be based on human IgG2. Furthermore, these sequences may alternatively utilize... Figures 3A-3E , Figure 4 and Figure 5 Other skew variants, pI variants, and ablation variants are shown.

[0070] As will be understood by those skilled in the art and as outlined below, these sequences can be used with any IL-15 and IL-15Rα (sushi) pairs outlined herein, including but not limited to scIL-15 / Rα, ncIL-15 / Rα, and dsIL-15Rα schematically illustrated in Figure 53. Furthermore, as will be understood by those skilled in the art and as outlined below, any IL-15 and / or IL-15Rα (sushi) variants can be incorporated into these main chains. Additionally, as will be understood by those skilled in the art and as outlined below, these sequences can be used with any VH and VL pairs (including scFv or Fab) outlined herein.

[0071] Each of these backbones includes a sequence that is 90%, 95%, 98%, and 99% identical (as defined herein) to the listed sequences, and / or contains 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 additional amino acid substitutions (which, as those skilled in the art will understand, already contain many amino acid modifications compared to the parent IgG1 (or IgG2 or IgG4, depending on the backbone)). That is, in addition to the skewed, pI, and ablation variants contained in the backbones shown in the figure, the listed backbones may also contain other amino acid modifications (typically amino acid substitutions).

[0072] Figure 12 The “non-Fv” backbone of the homologous light chain (i.e., constant light chain) is shown, which can be used in the PD-1 targeting IL-15 / Rα-Fc fusion protein of the present invention.

[0073] Figure 13A ~13G illustrates several forms of the IL-15 / Rα-Fc fusion protein of the present invention. The IL-15Rα heterodimeric Fc fusion protein "IL-15 / Rα-heteroFc" ( Figure 13A This fusion protein contains IL-15 recombinantly fused to one side of the heterodimer Fc and IL-15Rα(sushi) recombinantly fused to the other side of the heterodimer Fc. IL-15 and IL-15Rα(sushi) may have a variable-length Gly-Ser linker between the C-terminus and N-terminus of the Fc region. Single-chain IL-15 / Rα-Fc fusion protein or “scIL-15 / Rα-Fc” Figure 13B This comprises IL-15Rα(sushi) fused to IL-15 via a variable-length linker (called a “single-chain” IL-15 / IL-15Rα(sushi) complex or “scIL-15 / Rα”) and then fused to the N-terminus of the Fc region of the heterodimer, wherein the other side of the molecule is “Fc-only” or “empty Fc”. Non-covalent IL-15 / Rα-Fc or “ncIL-15 / Rα-Fc” ( Figure 13C This contains IL-15Rα fused to the Fc region of a heterodimer (sushi), while IL-15 is transfected separately, thus forming a non-covalent IL-15 / Rα complex, where the other side of the molecule is an "Fc-only" or "empty Fc". This is a bivalent non-covalent IL-15 / Rα-Fc fusion protein or "bivalent ncIL-15 / Rα-Fc" (sushi). Figure 13D This contains IL-15Rα (sushi) fused to the N-terminus of a homodimeric Fc region, while IL-15 is transfected alone, thus forming a non-covalent IL-15 / Rα complex. This is a bivalent single-chain IL-15 / Rα-Fc fusion protein or "bivalent scIL-15 / Rα-Fc" ( Figure 13E This contains IL-15 fused to IL-15Rα(sushi) via a variable-length linker (called a "single-chain" IL-15 / IL-15Rα (sushi) complex or "scIL-15 / Rα") and then fused to the N-terminus of the heterodimer's Fc region. The Fc-noncovalent IL-15 / Rα fusion protein, or "Fc-ncIL-15 / Rα" (…), is also known as… Figure 13F This contains IL-15Rα fused to the C-terminus of the heterodimer's Fc region (sushi), while IL-15 is transfected separately, thus forming a non-covalent IL-15 / Rα complex, where the other side of the molecule is either "Fc-only" or "empty Fc". This is an Fc-single-chain IL-15 / Rα fusion protein or "Fc-scIL-15 / Rα". Figure 13G It contains IL-15 fused to IL-15Rα(sushi) via a variable-length linker (called a “single-chain” IL-15 / IL-15Rα(sushi) complex or “scIL-15 / Rα”) and then fused to the C-terminus of the heterodimer Fc region, wherein the other side of the molecule is “Fc only” or “empty Fc”.

[0074] Figure 14 The sequence of an exemplary IL-15 / Rα-Fc fusion protein in the form of “IL-15 / Rα-heteroFc” is shown. IL-15 and IL-15Rα (sushi) are underlined, and the linker is underlined twice (although, as those skilled in the art will understand, the linker can be replaced by other linkers, some of which are in…). Figure 8 (as shown in the image), and the slash ( / ) indicates the boundary between IL-15, IL-15Rα, the connector, and the Fc region.

[0075] Figure 15The sequence of an exemplary IL-15 / Rα-Fc fusion protein in the form of “scIL-15 / Rα-Fc” is shown. IL-15 and IL-15Rα (sushi) are underlined, and the linker is underlined twice (although, as those skilled in the art will understand, the linker can be replaced by other linkers, some of which are in…). Figure 8 (as shown in the image), and the slash ( / ) indicates the boundary between IL-15, IL-15Rα, the connector, and the Fc region.

[0076] Figure 16 The sequence of an exemplary IL-15 / Rα-Fc fusion protein in the form of “ncIL-15 / Rα-Fc” is shown. IL-15 and IL-15Rα (sushi) are underlined, and the linker is underlined twice (although, as those skilled in the art will understand, the linker can be replaced by other linkers, some of which are in…). Figure 8 (as shown in the image), and the slash ( / ) indicates the boundary between IL-15, IL-15Rα, the connector, and the Fc region.

[0077] Figures 17A-17C This illustrates the effect of IL-15 / Rα-Fc fusion proteins on A)NK (CD56) via scIL-15 / Rα-Fc (XENP21478) and ncIL-15 / Rα-Fc (XENP21479) (based on Ki67 expression measured by FACS). + / CD16 + ) cells, B) CD4 + T cells and C)CD8 + Induction of T cell proliferation.

[0078] Figure 18 The structure of IL-15 complexed with IL-15Rα, IL-2Rβ, and the common γ chain is shown. Substitution sites designed to reduce potency are also shown.

[0079] Figures 19A-19C The sequences of exemplary IL-15 variants designed to reduce potency are provided. Each of these variant IL-15 sequences contains 90%, 95%, 98%, and 99% of the sequence as defined herein (as enumerated), and / or contains 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 additional amino acid substitutions. As will be apparent to those skilled in the art, the IL-15 variants can be used in any IL-15 / Rα-Fc fusion protein described herein and in PD-1-targeted IL-15 / Rα-Fc fusion proteins.

[0080] Figure 20The sequence of an exemplary IL-15 / Rα-Fc fusion protein in the form of “IL-15 / Rα-heteroFc” (which contains an IL-15 variant designed to reduce potency) is shown. IL-15 and IL-15Rα (sushi) are underlined, the linker is underlined twice (although, as those skilled in the art will understand, the linker can be replaced by other linkers, some of which are shown in Figure 65), and a slash ( / ) indicates the boundary between IL-15, IL-15Rα, the linker, and the Fc region.

[0081] Figures 21A-21B The sequence of an exemplary IL-15 / Rα-Fc fusion protein in the form of “scIL-15 / Rα-Fc” (which contains an IL-15 variant designed to reduce potency) is shown. IL-15 and IL-15Rα (sushi) are underlined, the linker is underlined twice (although, as those skilled in the art will understand, the linker can be replaced by other linkers, some of which are shown in Figure 65), and a slash ( / ) indicates the boundary between IL-15, IL-15Rα, the linker, and the Fc region.

[0082] Figures 22A-22B Describes the expression of Ki67 by A)CD4 after incubation with the specified test sample. + CD45RA- and B)CD8 + CD45RA - Percentage of cells.

[0083] Figure 23 The amino acid sequence of XENP15074 (a bivalent anti-RSV mAb based on movizumab and human IgG1 Fc, containing E233P / L234V / L235A / G236del / S267K substitutions) is shown. CDRs are underlined, and slashes indicate the boundaries of variable regions.

[0084] Figure 24 The amino acid sequences of A) XENP16432 (a bivalent anti-PD-1 mAb based on nivolumab and human IgG1 Fc, containing E233P / L234V / L235A / G236del / S267K substitutions), B) XENP21641 (pembrolizumab), and C) XENP28437 (a bivalent anti-PD-1 mAb based on pembrolizumab and human IgG1 Fc, containing E233P / L234V / L235A / G236del / S267K substitutions) are shown. CDRs are underlined, and slashes indicate the boundaries of variable regions.

[0085] Figure 25The sequence of XENP21575 (a chimeric and humanized anti-PD-1 antibody based on the variable region of hybridoma clone 1C11 and human IgG1, containing E233P / L234V / L235A / G236del / S267K substitutions in the heavy chain) is shown. CDRs are underlined, and slashes indicate the boundaries of the variable domains. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also those with V... H and V L CDRs contained within a structural domain (using other numbering systems). As those skilled in the art will understand, V H and V L The domain can be formatted as Fab or scFv for use in the PD-1 targeting IL-15 / Rα-Fc fusion protein of the present invention.

[0086] Figure 26 The sequence shown represents an exemplary humanized variant of the bivalent human IgG1 form of anti-PD-1 mAb A and mAb B, containing E233P / L234V / L235A / G236del / S267K substitutions in the heavy chain. CDRs are underlined, and slashes indicate the boundaries of variable domains. As described herein, and which holds true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). As those skilled in the art will understand, the VH and VL domains may be formatted as Fab or scFv for use in the PD-1-targeting IL-15 / Rα-Fc fusion protein of the present invention.

[0087] Figure 27 The diagram shows epitope clusters of XENP16432 (a bivalent anti-PD-1 mAb based on nivolumab), XENP21461 (pembrolizumab), chimeric mAb A (chmAb A), chimeric mAb B (chmAb B), and 1C11-based mAbs represented by the normalized BLI response Octet. A normalized BLI response greater than 0.5 indicates that the antibody pair does not bind to the same epitope.

[0088] Figure 28A ~28H illustrates several forms of the PD-1-targeting IL-15 / Rα-Fc fusion protein of the present invention. The “scIL-15 / Rαx scFv” form ( Figure 28AThis comprises IL-15Rα (sushi) fused to IL-15 via a variable-length connector (referred to as "scIL-15 / Rα") and then fused to the N-terminus of the heterodimer Fc region, wherein scFv is fused to the other side of the heterodimer Fc. The form is "scFv x ncIL-15 / Rα". Figure 28B This involves a scFv fused to the N-terminus of the Fc region of a heterodimer, where IL-15Rα (sushi) is fused to the other side of the Fc region of the heterodimer, while IL-15 is transfected separately, thus forming a non-covalent IL-15 / Rα complex. The “scIL-15 / Rαx Fab” form ( Figure 28C This involves IL-15Rα (sushi) fused to IL-15 via a variable-length connector (called "scIL-15 / Rα") and then to the N-terminus of the heterodimer's Fc region, where the variable heavy chain (VH) fuses to the other side of the heterodimer's Fc, while the corresponding light chain is transfected separately to form a Fab with the VH. The "ncIL-15 / RαxFab" form ( Figure 28D This involves a VH fused to the N-terminus of the Fc region of a heterodimer, where IL-15Rα (sushi) is fused to the other side of the Fc region of the heterodimer, while the corresponding light chain is transfected separately to form a Fab with the VH, and simultaneously IL-15 is transfected separately, thereby forming a non-covalent IL-15 / Rα complex. The “mAb-scIL-15 / Rα” form ( Figure 28E This involves a VH fused to the N-terminus of both the first and second heterodimer Fc, wherein IL-15 is fused to IL-15Rα (sushi) and then further fused to the C-terminus of one of the heterodimer Fc regions, while the corresponding light chain is transfected separately to form a Fab with the VH. The “mAb-ncIL-15 / Rα” form ( Figure 28F The VH comprises a first heterodimer Fc fused to the N-terminus of both the first and second heterodimer Fc regions, wherein IL-15Rα (sushi) is fused to the C-terminus of one of the heterodimer Fc regions, while the corresponding light chain is transfected separately to form a Fab with the VH, and IL-15 is transfected separately, thereby forming a non-covalent IL-15 / Rα complex. The “central IL-15 / Rα” form ( Figure 28G This includes a VH that is recombinantly fused to the N-terminus of IL-15 and then fused to one side of the heterodimer Fc, and a VH that is recombinantly fused to the N-terminus of IL-15Rα (sushi) and then further fused to the other side of the heterodimer Fc, while the corresponding light chains are transfected separately to form Fab with the VH. The “central scIL-15 / Rα” form ( Figure 28HThe product contains a VH fused to the N-terminus of IL-15Rα (sushi) (which is fused to IL-15 and then further fused to one side of the heterodimer Fc) and a VH fused to the other side of the heterodimer Fc, while the corresponding light chain is transfected separately to form a Fab with the VH.

[0089] Figure 29 The sequence of an illustrative [C]PD-1 targeting IL-15 / Rα-Fc fusion protein in the form of “scIL-15 / Rαx Fab” is shown. The CDR is shown in bold. As stated herein, and true for every sequence in which a CDR is included herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). IL-15 and IL-15Rα (sushi) are shown in italics, and the adapter is double-underlined (although, as those skilled in the art will understand, the adapter can be replaced by other adapters, some of which are shown in Figure 65 and...). Figure 66 (as shown in the diagram), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the linker, the variable region, and the constant / Fc region. As will be apparent to those skilled in the art, each of the targeted IL-15 / Rα-Fc fusion proteins may or may not include Xtend Fc (M428L / N434S). Figures 30A-30B The affinity of XENP22553 for PD-1 is shown by Octet (and associated sensor maps).

[0090] Figures 30A-30C The sequence of an illustrative [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein in the form of “scIL-15 / Rαx Fab” is shown. The CDR is underlined. As is stated herein, and true for every sequence in which a CDR is included herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). IL-15 and IL-15Rα (sushi) are shown in italics, and the adapter is double-underlined (although, as those skilled in the art will understand, the adapter can be replaced by other adapters, some of which are shown in Figures 65 and 66). Figure 66 (as shown in the diagram), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the linker, the variable region, and the constant / Fc region. As will be apparent to those skilled in the art, each of the targeted IL-15 / Rα-Fc fusion proteins may or may not include Xtend Fc (M428L / N434S). Figure 32A~32D shows the amino acid sequences of (A) XENP21641 (pembrolizumab) and (B) XENP28437 (a bivalent anti-PD-1 mAb based on pembrolizumab and human IgG1 Fc, containing E233P / L234V / L235A / G236del / S267K substitutions). CDRs are underlined, and slashes indicate the boundaries of variable regions.

[0091] Figures 31A-31B The sequence of the control RSV targeting the IL-15 / Rα-Fc fusion protein is shown. CDRs are underlined. As described herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). IL-15 and IL-15Rα (sushi) are shown in italics, and the adapter is double-underlined (although, as those skilled in the art will understand, the adapter can be replaced by other adapters, some of which are shown in Figure 65 and...). Figure 66 (as shown in the diagram), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the linker, the variable region, and the constant / Fc region. As will be apparent to those skilled in the art, each of the IL-15 / Rα-Fc fusion proteins may also contain Xtend Fc (M428L / N434S).

[0092] Figures 32A-32B The following diagram shows the results of incubation with PD-1-targeted IL-15 / Rα-Fc fusion proteins (XENP28532, XENP28692, and XENP25850), control RSV-targeted IL-15 / Rα-Fc fusion protein (XENP26007), and anti-PD-1 mAb (XENP28519) followed by CD8. + T cells and B)CD4 + T cell proliferation.

[0093] Figures 33A-33B This demonstrates the effect of XENP25850 (an exemplary PD-1-targeting IL-15 / Rα-Fc fusion protein) on A)CD4. + CD45RA - CD25 + and B)CD8 + CD45RA-CD25 + Induction of STAT5 phosphorylation. Fresh cells are shown by dashed lines, and activated cells by solid lines. All fresh cells are CD25 negative.

[0094] Figure 34A and Figure 34BThe results show that after pre-incubation with nivolumab-based XENP16432, pembrolizumab, or anti-RSVmAb XENP15074, A)[C]PD-1 targeting IL-15 / Rα-Fc fusion protein XENP25937 and B)[NC]PD-1 targeting IL-15 / Rα-Fc fusion protein XENP28532, CD8+ was significantly reduced. + CD45RACD25 + PD-1 + Induction of STAT5 phosphorylation in T cells.

[0095] Figure 35 The change in body weight over time (expressed as a percentage of initial body weight) is shown in NSG mice transplanted with huPBMCs after administration of the specified test substance.

[0096] Figures 36A-36B A)CD8 was observed on day 14 after the first administration of the specified test substance in NSG mice transplanted with human PBMCs. + T cell count and B)CD4 + T cell count.

[0097] Figures 37A-37B A)CD8 was observed on day 10 after the first administration of the specified test substance in NSG mice transplanted with human PBMCs. + T cells and B)CD4 + CD25 expression on T cells.

[0098] Figure 38 This shows CD8 levels on day 10 after the first administration of the specified test substance in NSG mice transplanted with human PBMCs. + T cells and CD4 + The ratio of T cells.

[0099] Figures 39A-39B A)CD8 was observed on day 10 after the first administration of the specified test substance in NSG mice transplanted with human PBMCs. + T cell count and B)CD4 + T cell count.

[0100] Figures 40A-40B A)CD8 was observed on day 10 after the first administration of the specified test substance in NSG mice transplanted with human PBMCs. + T cells and B)CD4 + CD25 expression on T cells.

[0101] Figure 41 The changes in serum concentrations of the specified test sample over time are shown in cynomolgus monkeys after the first administration of the specified relative concentration.

[0102] Figure 42 XENP29484 and XENP29485 (illustrative [NC]PD-1-targeting IL-15 / Rα-Fc fusion proteins with IL-15 (D30N / N65D) variants) are shown. CDRs are underlined. As described herein, and true for every sequence in which a CDR is included herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1. Therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). IL-15 and IL-15Rα (sushi) are shown in italics, the adapter is double-underlined (although, as those skilled in the art will recognize, the adapter can be replaced by other adapters, some of which are shown in the figure), and the slash ( / ) indicates the boundary between IL-15, IL-15Rα, the adapter, the variable region, and the constant / Fc region.

[0103] Figures 43A-43B Additional illustrative anti-PD-1ABD variants, including variable heavy and light chains, are shown that do not compete with nivolumab or pembrolizumab. CDRs are underlined. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems).

[0104] Figures 44A-44B The image shows the results of PD-1 treatment with XENP28532 (targeting the IL-15 / Rα-Fc fusion protein) and control XENP24306 (non-targeting the IL-15 / Rα-Fc fusion protein) and XENP26007 (RSV targeting the IL-15 / Rα-Fc fusion protein). + A)CD8 + CD45RA - CD45RO + T cells and B)CD4 + CD45RA - CD45RO + The percentage of T cells.

[0105] Figures 45A-45BThe sequence shown is an exemplary humanized variant of anti-PD-1 mAb C in the form of bivalent human IgG1, containing E233P / L234V / L235A / G236del / S267K substitutions in the heavy chain. CDRs are underlined, and slashes indicate the boundaries of variable domains. As described herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). As those skilled in the art will recognize, the VH and VL domains may be formatted as Fab or scFv for use in the PD-1-targeting IL-15 / Rα-Fc fusion protein of the present invention. Figure 46 (A)CD45 levels were observed on day 14 after the first administration of the designated test substance in NSG mice transplanted with human PBMCs. + Cell count, (B)CD3 + T cell count, (C)CD8 + T cell count and (D)CD4 + T cell count.

[0106] Figure 46 The affinity of XENP28536, XENP28537, XENP28538, XENP28539 and XENP28519 for human and macaque PD-1 was shown by Octet.

[0107] Figure 47 Epitope clustering is shown for XENP16432 (a bivalent anti-PD-1 mAb based on nivolumab), XENP21461 (pembrolizumab), and chimeric mAb C (chmAb C). A normalized BLI response greater than 0.5 indicates that the antibody pairs do not bind to the same epitope. The data show that anti-PD-1 mAb C does not bind to the same epitope as nivolumab and pembrolizumab.

[0108] Figures 48A-48CThe sequence shown is an exemplary “scIL-15 / Rαx Fab” form of a [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein containing a PD-1 targeting arm based on mAb C and various IL-15 potency variants. The CDR is underlined. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table X; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). The adapter is double-underlined (although, as those skilled in the art will recognize, the adapter can be replaced by other adapters, some of which are shown in Figure 65 and...). Figure 66 (as shown in the diagram), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the linker, the variable region, and the constant / Fc region. As will be apparent to those skilled in the art, each of the PD-1 targeting IL-15 / Rα-Fc fusion proteins may also contain Xtend Fc (M428L / N434S).

[0109] Figures 49A-49C The sequence shown is an exemplary “scIL-15 / Rαx Fab” form of a [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein containing a PD-1 targeting arm based on mAb C and various IL-15 potency variants, and also containing Xtend Fc (M428L / N434S). CDRs are underlined. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). Adapters are double underlined (although, as those skilled in the art will recognize, adapters can be replaced by other adapters, some of which are shown in Figure 65 and...). Figure 66 (as shown in the figure), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the linker, the variable region, and the constant / Fc region. Figures 50A-50B show (A)CD45 on day 10 after the first administration of the specified test item in NSG mice transplanted with human PBMCs. + Cell count, (B)CD3 + T cell count, (C)CD8 + T cell count and D)CD4 + T cell count.

[0110] Figures 50A-50B show the effects of [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) on A)CD8. + T cells and B)CD4+ Induction of T cell proliferation, as shown in the percentage of proliferating cells (based on CFSE dilution). Data showed that, compared to the non-targeted IL-15 (D30N / E64Q / N65D) / Rα-Fc fusion protein (and the control RSV-targeted IL-15 / Rα-Fc fusion protein), [NC]PD-1 targeting the IL-15 / Rα-Fc fusion protein significantly improved CD4 cell proliferation. + It is more effective in promoting T cell proliferation.

[0111] Figure 51 The [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) is shown to have an effect on LAG-3 positive CD8. + T cell induction, as shown by the percentage of proliferating cells (based on CFSE dilution). Data showed that XENP28532 significantly improved CD8 cell induction compared to the non-targeted IL-15 (D30N / E64Q / N65D) / Rα-Fc fusion protein (and the control RSV-targeted IL-15 / Rα-Fc fusion protein). + LAG-3 + It is more effective in T cell proliferation. Furthermore, XENP28543 induces CD8... + LAG-3 + T cell proliferation ratio induces somatic CD8 + T cell proliferation was more efficient (EC50 276.8 vs. 71.94). In summary, this supports the view that [NC]PD-1 targeting of the IL-15 / Rα-Fc fusion protein may be selective for T cells expressing checkpoints (such as those present in the tumor environment).

[0112] Figures 52A-52B show A)CD4 after incubation with [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control). + CD45RA - Memory T cells and B)CD4 + CD45RA + Activation of naïve T cells, as shown by the percentage of cells expressing CD25.

[0113] Figures 53A-53B show A)CD4 cells expressing PD-1 after incubation with [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control). + CD45RA - T cells and B)CD4 + CD45RA + The percentage of T cells.

[0114] Figures 54A-54CIFNγ secretion was induced in PBMCs pre-stimulated with anti-CD3 (OKT3) bound to a plate and incubated with the specified test sample using concentrations A) 50 ng / ml, B) 100 ng / ml, and C) 500 ng / ml. Data showed that both XENP28532 and XENP28543 effectively stimulated IFNγ secretion. Notably, XENP28532 (a mAb A-based PD-1 targeting arm) appeared to be more active than XENP28543 (a mAb C-based PD-1 targeting arm) in inducing IFNγ secretion.

[0115] Figures 55A-55B show the effects of XENP28543 (a mAb C-based [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein containing a PD-1 targeting arm) on A)CD8 after pre-incubation with nivolumab-based XENP16432, pembrolizumab, or anti-RSV mAb XENP15074. + CD45RA - CD25 + PD-1 + T cells and B)CD4 + CD45RA - CD25 + PD-1 + Induction of STAT5 phosphorylation in T cells. Data showed that PD-1 blockers did not interfere with the activity of XENP28543.

[0116] Figure 56 The change in body weight over time (expressed as a percentage of initial body weight) is shown in NSG mice transplanted with huPBMCs after administration of the specified test substance.

[0117] Figures 57A-57C The body weights (as a percentage of initial body weight) of NSG mice transplanted with huPBMCs are shown on days A) 11, B) 14, and C) 18 after the first administration of the designated test substance. p-values ​​were determined using an unpaired t-test. Data showed that, up to day 11, the combination of [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein and a PD-1 blocker significantly enhanced GVHD compared to treatment with [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein alone.

[0118] Figures 58A-58F The human A)CD45 level in the blood of NSG mice transplanted with huPBMCs is shown on day 14 after the first administration of the specified test substance. + Cells, B)CD3 + T cells, C)CD4 + T cells, D)CD8 +The number of T cells, E) γδT cells and F) NK cells.

[0119] Figures 59A-59B The human A)CD8 level in the blood of NSG mice transplanted with huPBMCs was shown on day 14 after the first administration of the specified test substance. + T cells and B)CD4 + T cell activation (as shown in CD25 MFI).

[0120] Figure 60 The changes in tumor volume (as measured by calipers) over time in NSG mice with transplanted pp65-MCF7 and huPBMCs after receiving [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein alone or in combination with a PD-1 blocker are shown.

[0121] Figures 61A-61F NSG mice with transplanted pp65-MCF7 and huPBMCs, receiving [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein alone or in combination with a PD-1 blocker, were shown at age 26 ( Figure 61A ), 28 Figure 61B ), 30 Figure 61C ), 33 ( Figure 61D ), 35 Figure 61E ) and the 37th Figure 61F Tumor volume (as measured by calipers) on day 28 (after PBMC transplantation and initial administration of the test sample). p-values ​​were determined by unpaired t-tests. Data showed that, up to day 28, XENP28543 in combination with a PD-1 inhibitor significantly reduced tumor size compared to PD-1 inhibitor treatment alone.

[0122] Figures 62A-62B show the human A)CD8 levels in the blood of NSG mice transplanted with pp65-MCF7 and huPBMC on day 7 after the first administration of the designated test substance. + T cells and B)CD4 + T cell activation (as shown by CD25 MFI). Data showed that [NC]PD-1 targeting the IL-15 / Rα-Fc fusion protein, used alone or in combination with PD-1 blockers, significantly enhanced early activation of CD8. + T cells. Unpaired t-tests were used to perform statistical analysis on the log-transformed data.

[0123] Figures 63A-63E The human A)CD45 level in the blood of NSG mice transplanted with pp65-MCF7 and huPBMC was shown on day 14 after the first administration of the specified test substance. + Cells, B)CD3 +T cells, C)CD4 + T cells, D)CD8 + T cells and E)NK cells. Data showed that [NC]PD-1 targeting the IL-15 / Rα-Fc fusion protein, alone or in combination with a PD-1 blocker, significantly enhanced the expansion of a large lymphocyte population at day 14 compared to PD-1 blocker alone. Unpaired t-tests were used for statistical analysis of the logarithmically transformed data.

[0124] Figures 64A-64B The sequence shown is an exemplary affinity-engineered variant of anti-PD-1 mAb C in the form of bivalent human IgG1, containing E233P / L234V / L235A / G236del / S267K substitutions in the heavy chain. CDRs are underlined, and slashes indicate the boundaries of variable domains. As described herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table X; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). As those skilled in the art will understand, the VH and VL domains may be formatted as Fab or scFv for use in the PD-1-targeting IL-15 / Rα-Fc fusion protein of the present invention. Figure 67 A shows epitope clustering for XENP16432 (a nivolumab-based bivalent anti-PD-1 mAb), XENP21461 (pembrolizumab), and chimeric mAb C (chmAb C). A normalized BLI response greater than 0.5 indicates that the antibody pairs do not bind to the same epitope. The data show that anti-PD-1 mAb C does not bind to the same epitope as nivolumab and pembrolizumab.

[0125] Figures 65A-65I The apparent dissociation constant (KDapp), binding rate (ka), and dissociation rate (kd) (as measured by Octet) of the affinity-engineered mAb C[PD-1]_H1L1 variant (in the form of bivalent IgG1, containing the E233P / L234V / L235A / G236_ / S267K ablation variant) are shown, along with the fold improvement relative to mAb C[PD-1]_H1L1. Substitutions in the heavy chain or variable light domains listed are based on Xencor numbering (the corresponding Kabat positions are listed in the next column). Of the 304 variants with single-point mutations in the heavy chain or light chain variable regions, we identified only 11 variants (including mAb C[PD-1]_H1_L1.1 and mab_C[PD-1]_H1_L1.3) with more than 2-fold higher affinity than WT.

[0126] Figure 66 The apparent dissociation constant (KDapp), binding rate (ka), and dissociation rate (kd) (as measured by Octet) of mAb C[PD-1]_H1L1 variants with affinity-engineered binding-favorable single-substituted VH variants and single-substituted VL variants (in the context of PD-1 targeting IL15 / Rα-Fc) are shown. Substitutions in the heavy chain or variable light domains listed are based on Xencor numbering (the corresponding Kabat positions are listed in the next column). Although H1.132_L1 provides a higher affinity than H1.19_L1, H1.19_L1.1 has a higher affinity than H1.132_L1.1.

[0127] Figure 67 The apparent dissociation constant (KDapp), binding rate (ka), and dissociation rate (kd) (as measured by Octet) of the affinity-engineered mAb C[PD-1]_H1L1 variant with multiple substitutions in VH and / or VL (in the context of PD-1 targeting IL15 / Rα-Fc) are shown. The substitutions in the heavy chain or variable light domains listed are based on Xencor numbering (the corresponding Kabat positions are listed in the next column). The trisubstituted VL variant N31H / K36Y / S99T (L1.140; Kabat numbered N27dH / K30Y / S93T) shows a 36-fold improvement in KD compared to the wild type and binds well to the VH variant, achieving an approximately 100-fold improvement compared to the wild type.

[0128] Figures 68A-68J The sequence shown is an exemplary “scIL-15 / Rαx Fab” form of a [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein containing an affinity-optimized mAb C ABD and various IL-15 potency variants. The CDR is underlined. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). The adapter is double-underlined (although, as those skilled in the art will recognize, the adapter can be replaced by other adapters, some of which are shown in Figure 65 and...). Figure 66 (as shown in the diagram), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the linker, the variable region, and the constant / Fc region. As will be apparent to those skilled in the art, each of the PD-1 targeting IL-15 / Rα-Fc fusion proteins may also include or exclude Xtend Fc (M428L / N434S).

[0129] Figures 69A-69C The sequence shown is an exemplary “scIL-15 / Rαx Fab” form of a [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein containing an affinity-optimized mAb C ABD and various IL-15 potency variants, and also containing an Xtend Fc (M428L / N434S) PD-1 targeting arm. CDRs are underlined. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table X; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). Adapters are double underlined (although, as those skilled in the art will recognize, adapters can be replaced by other adapters, some of which are shown in Figure 65 and...). Figure 66 (as shown in the diagram), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the connector, the variable region, and the constant / Fc region.

[0130] Figures 70A–70B illustrate the induction of CD8+ T cell and CD4+ T cell proliferation by [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants possessing different PD-1 affinity and IL-15 potency, as shown in the percentage of proliferating cells (based on CFSE dilution). The data show that XENP30046 (with an enhanced affinity PD-1 targeting arm) induced CD8+ and CD4+ T cell proliferation more effectively than XENP28543 (by a 2-fold increase). Furthermore, the data show that the IL-15 (D30N / N65D) variant had no significant effect on the activity of the PD-1 targeting IL-15 / Rα-Fc fusion protein.

[0131] Figures 71A-71B show A)CD8 after incubation with [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of [NC]PD-1 having different PD-1 affinity and IL-15 potency. + CD45RA - T cells and B)CD8 + CD45RA + T cell activation is indicated by the percentage of cells expressing CD25.

[0132] Figures 72A–72B show the activation of A) CD4+CD45RA-T cells and B) CD4+CD45RA+ T cells after incubation with [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of [NC]PD-1 with different PD-1 affinity and IL-15 potency, as shown by the percentage of cells expressing CD25.

[0133] Figures 73A-73B show the expression of PD-1 A)CD4 after incubation with [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of [NC]PD-1 having different PD-1 affinity and IL-15 potency. + CD45RA - T cells and B)CD4 + CD45RA + The percentage of T cells. Data shows that [NC]PD-1 targeting the IL-15 / Rα-Fc fusion protein induces CD4+. + PD-1 is downregulated in cells, and this downregulation is associated with the affinity of PD-1.

[0134] Figure 74 The changes in body weight (expressed as a percentage of initial body weight) over time in NSG mice with transplanted huPBMCs treated with [NC]PD-1-targeted IL-15 / Rα-Fc fusion proteins (and controls) with variants of PD-1 affinity and IL-15 potency are shown. For mice that died, 70% was used.

[0135] Figures 75A-75D The body weights of transplanted huPBMC NSG mice treated with [NC]PD-1-targeted IL-15 / Rα-Fc fusion protein (and control) with variants of PD-1 affinity and IL-15 potency are shown on days A) 11, B) 14, C) 18, and D) 21. For dead mice, 70% is used. p-values ​​were determined using an unpaired t-test. The data showed that XENP30046 alone (with an affinity-enhanced PD-1 targeting arm) resulted in a significant reduction in body weight measured on days 11 and 18 compared to PBS treatment, while XENP28543 alone did not produce a significant reduction in body weight (compared to PBS treatment).

[0136] Figures 76A–76C show the serum concentrations of A) IFNγ, B) IL-10, and C) IL-2Rα in NSG mice transplanted with huPBMCs on days 7, 11, and 14 after administration of [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of PD-1 affinity and IL-15 potency.

[0137] Figures 77A-77B show the blood levels of A)CD4 in NSG mice transplanted with huPBMCs on day 7 after administration of [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of different PD-1 affinity and IL-15 potency. + T cells and B)CD8 +T cell activation (as shown in CD25MFI) (statistical analysis of log-transformed data was performed using an unpaired t-test).

[0138] Figures 78A-78E The image shows the blood A)CD45 levels on day 11 in NSG mice transplanted with huPBMCs after administration of [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of PD-1 affinity and IL-15 potency. + Cells, B)CD3 + T cells, C)CD8 + T cells, D)CD4 + T cell and E)NK cell counts (statistical analysis of log-transformed data was performed using an unpaired t-test).

[0139] Figures 79A-79E The image shows the blood A)CD45 levels on day 14 after NSG mice transplanted with huPBMCs were administered [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of PD-1 affinity and IL-15 potency. + Cells, B)CD3 + T cells, C)CD8 + T cells, D)CD4 + T cell and ENK cell counts.

[0140] Figures 80A-80B show the effects of [NC]PD-1 targeting IL-15 / Rα-Fc fusion proteins (and controls) with different PD-1 affinity and IL-15 potency variants on A) CD8 + T cells and B)CD4 + Induction of T cell proliferation, as shown by the percentage of proliferating cells (measured based on CFSE dilution). Data showed that XENP30272 (a K+ inhibitor of PD-1) induced T cell proliferation. D (3.1 nM) was more effective than XENP30046 (for PD-1 K) in inducing the proliferation and activation of various T cell populations. D (5.4 nM) is more effective. It is worth noting that while XENP30429 (a PD-1-IL-15 / Rα-Fc fusion protein with the IL-15 (D30N / E64Q / N65D) variant) is more effective than XENP30046 (a PD-1-targeting IL-15 / Rα-Fc fusion protein with the IL-15 (N4D / N65D) variant) for CD8... + and CD4 +T cell activity was only 1.8 to 2.5 times lower, but XENP30432 (an alternative RSV targeting IL-15 / Rα-Fc with an IL-15 (D30N / E64Q / N65D) variant) was significantly less active against CD8 compared to XENP30046. + T cell activity was 12-fold lower and it was also affected by CD4. + T cell activity was 530-fold lower (based on proliferative activity). This indicates that PD-1 targeting the IL-15 / Rα-Fc fusion protein with the IL-15 (D30N / E64Q / N65D) variant remains active in the tumor environment but is essentially inactive outside the tumor environment.

[0141] Figures 81A-81C A)CD4 was shown after incubation with [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of [NC]PD-1 having different PD-1 affinity and IL-15 potency. + CD45RA - T cells and B)CD4 + CD45RA + T cell activation, as shown in CD25 MFI.

[0142] Figures 82A-82C A)CD8 was shown after incubation with [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of [NC]PD-1 having different PD-1 affinity and IL-15 potency. + CD45RA - T cells and B)CD8 + CD45RA + T cell activation, as shown in CD25 MFI.

[0143] Figures 83A-83C A)CD4 was shown after incubation with [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of [NC]PD-1 having different PD-1 affinity and IL-15 potency. + T cells, B)CD4 + CD45RA - T cells and C)CD4 + CD45RA + PD-1 expression on T cells, as shown in PD-1MFI.

[0144] Figures 84A-84C A)CD8 was shown after incubation with [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein (and control) with variants of [NC]PD-1 having different PD-1 affinity and IL-15 potency. + T cells, B)CD8 + CD45RA- T cells and C)CD8 + CD45RA + PD-1 expression on T cells, as shown in PD-1MFI.

[0145] Figures 85A-85D The image shows A) CD45 cells and B) CD3 cells in NSG mice on day 10 after the first administration of the specified test substance. + T cells, C)CD4 + T cells and D)CD8 + T cell expansion (as shown in cell counts). Data shows that the higher the potency of IL-15, the greater the expansion of various lymphocyte populations.

[0146] Figures 86A-86B show the A)CD4 levels in NSG mice on day 10 after the first administration of the designated test substance. + T cells and B)CD8 + T cell activation (as shown by CD25 staining). Data showed that the higher the potency of IL-15, the greater the expansion of various lymphocyte populations.

[0147] Figures 87A-87C The [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein and the control RSV targeting IL-15 / Rα-Fc fusion protein (with Xtend Fc) are shown to affect A)CD4. + T cell proliferation, B)CD8 + Induction of T cell and NK cell proliferation, as shown in the percentage of proliferating cells (based on CFSE dilution). Notably, the data showed that the EC50 of XENP30046 was comparable to that of XENP30290 (an Xtend analogue of XENP30046).

[0148] Figures 88A-88B The [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein and the control RSV targeting IL-15 / Rα-Fc fusion protein (with Xtend Fc) are shown to affect A)CD4. + T cells and B)CD8 + T cell activation, as shown in CD25 MFI.

[0149] Figures 89A-89B The [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein and the control RSV targeting IL-15 / Rα-Fc fusion protein (with Xtend Fc) are shown to affect A)CD4. + T cells and B)CD8 +PD-1 regulation in T cells, as shown in the PD-1 MFI. Consistent with Example 9B, the test sample downregulated PD-1 on T cells.

[0150] Figure 90 The changes in tumor volume (as measured by calipers) over time in NSG mice with transplanted pp65-MCF7 and huPBMCs after receiving [NC]PD-1 targeting IL-15 / Rα-Fc fusion protein alone or in combination with a PD-1 blocker are shown.

[0151] Figures 91A-91H Tumor volumes (as measured by calipers) in NSG mice with transplanted pp65-MCF7 and huPBMCs, administered [NC]PD-1-targeted IL-15 / Rα-Fc fusion protein alone or in combination with a PD-1 blocker, are shown at days 11, 14, 17, 19, 21, 24, 26, and 28 (after PBMC transplantation and first administration of the test sample). Baseline-corrected data were statistically analyzed using the Mann-Whitney test. * indicates that treatment significantly (p<0.05) enhanced amplification compared to the PBS control. # indicates that treatment significantly (p<0.05) enhanced amplification compared to XENP31123. This indicates that treatment significantly (p<0.05) enhanced tumor amplification compared to PD-1 blockade alone (XENP16432). Data showed that, up to day 28, all combinations of XENP30290 (0.1, 0.3, or 1 mg / kg) or XENP30516 (0.3, 1, or 3 mg / kg) with PD-1 significantly reduced tumor size compared to PD-1 blockade alone.

[0152] Figures 92A-92F The figures show the number of A) CD45+ cells, B) CD3+ T cells, C) NK cells, D) CD4+ T cells, and E) CD8+ T cells, and the CD8 to CD4 T cell ratio in the blood of NSG mice transplanted with pp65-MCF7 and huPBMC on day 14 after the first administration of the specified test substance. Unpaired t-tests were used for statistical analysis of logarithmically transformed CD45+ cell expansion. This indicates that treatment significantly enhanced amplification compared to using the PD-1 blocker (XENP16432) alone.

[0153] Figures 93A-93F This shows the levels of A)CD45 in the blood of NSG mice transplanted with pp65-MCF7 and huPBMC on day 21 after the first administration of the specified test substance. + Cells, B)CD3 + T cells, C) NK cells, D) CD4 +T cells and E)CD8 + The number of T cells and the F)CD8 to CD4 T cell ratio were also observed. This indicates that treatment significantly enhanced T cell proliferation compared to the use of the PD-1 blockade agent (XENP16432) alone.

[0154] Figure 94 CD34 + PD-1 expression levels (expressed as MFI) on various lymphocytes in Hu-NSG mice (before receiving any test drug treatment). Data showed that the PD-1 expression profile in mice was similar to that in humans, with higher expression in effector memory cell populations.

[0155] Figure 95 The study shows the fold increase in various lymphocyte populations on day 7 after CD34+Hu-NSG mice were treated with 0.3 mg / kg XENP30046 (a [NC]PD-1 targeting IL-15 / Rα-Fc variant containing mAb C_H1_L1.1 and IL-15 (N4D / N65D) variants). XENP30046 showed a 100-fold increase in the expansion of effector memory cell populations.

[0156] Figure 96 CD34 is shown + The fold increase in various lymphocyte populations on day 7 after Hu-NSG mice were treated with 0.3 mg / kg XENP30429 ([NC]PD-1 targeting IL-15 / Rα-Fc containing mAb C_H1_L1.1 and IL-15 (D30N / E64Q / N65D) variants).

[0157] Figure 97 CD34 is shown + The fold increase in various lymphocyte populations on day 7 after Hu-NSG mice were treated with 0.3 mg / kg XENP26007 (a control RSV targeting IL-15 / Rα-Fc containing the IL-15 (N4D / N65D) variant). XENP26007 showed very low expansion of various lymphocyte populations, indicating minimal peripheral lymphocyte expansion of the PD-1-targeting IL-15 / Rα-Fc fusion protein.

[0158] Figure 98 CD34 is shown +Hu-NSG mice treated with 0.3 mg / kg XENP30432 (a control RSV targeting IL-15 / Rα-Fc variant containing the IL-15 (D30N / E64Q / N65D) variant) showed a fold increase in various lymphocyte populations on day 7. XENP30432 exhibited very low expansion of various lymphocyte populations, indicating minimal peripheral lymphocyte expansion of the PD-1-targeting IL-15 / Rα-Fc fusion protein.

[0159] Figures 99A-99F CD34 is shown + Hu-NSG mice treated with 0.3 mg / kg XENP30046, XENP30429, XENP26007, or XENP30432 showed the following results: A) CD45 cells, B) CD3 cells. + Cells, C)CD4 + Cells, D)CD8 + fold changes in cells, E)γδ cells and F)NK cells.

[0160] Figures 100A-100D CD34 is shown + Hu-NSG mice treated with 0.3 mg / kg XENP30046, XENP30429, XENP26007, or XENP30432 showed changes in the fold increases of A) CD4 naïve cells, B) CD4 central memory cells, C) CD4 terminal effector cells, and D) CD4 effector memory cells. Data showed that XENP30046 and XENP30429 had an effect on PD-1... + Populations such as CD4 effector memory cells exhibit selectivity. Notably, XENP30429 of CD4 naïve cells showed minimal expansion, indicating that reducing the efficacy of the IL-15 arm improves selectivity.

[0161] Figures 101A-101D CD34 is shown + Hu-NSG mice treated with 0.3 mg / kg XENP30046, XENP30429, XENP26007, or XENP30432 showed changes in the fold increases of A) CD8 naïve cells, B) CD8 central memory cells, C) CD8 terminal effector cells, and D) CD8 effector memory cells. Data showed that XENP30046 and XENP30429 had an effect on PD-1... + Populations such as CD8 effector memory cells exhibit selectivity. Notably, XENP30429 amplification in CD8 naïve cells was minimal, indicating that reducing the efficacy of the IL-15 arm improves selectivity.

[0162] Figures 102A-102BThe study showed the correlation between lymphocyte expansion and baseline PD-1 expression on day 7 after CD34+Hu-NSG mice were treated with either A) 0.3 mg / kg XENP30046 or B) 0.3 mg / kg XENP30429.

[0163] Figures 103A-103B show the A)CD8 of cynomolgus monkeys that received 0.3X XENP22853, 1X XENP25937, or 0.3X XENP24306. + The expansion of T cells and beta-NK cells was observed. Data showed that PD-1 targeting the IL-15 / Rα-Fc fusion protein reduced NK cell expansion while maintaining CD8+. + T cell expansion.

[0164] Figures 104A-104B show the A)CD8 of cynomolgus monkeys that received 0.3X XENP22853, 1X XENP25937, or 0.3X XENP24306. + Naïve T cells and B)CD8 + Expansion of effector memory T cells. Data shows that PD-1 selectively expands CD8 by targeting the IL-15 / Rα-Fc fusion protein. + Effector memory T cells.

[0165] Figures 105A-105E The A)CD8 of cynomolgus macaques treated with either XENP30290 (mAb C_H1_L1.1 x IL-15 [N4D / N65D]) or XENP30362 (αRSV x IL-15 [N4D / N65D]) is shown. + PD1 + A) T cells, B) CD8 stem cell memory cells, C) CD8 naive cells, D) γδ T cells, and E) CD56 + NK cell expansion. In summary, data show that XENP30290 (with high PD-1 affinity and high IL-15 potency) can maintain peripheral pharmacodynamics for 2-3 weeks, achieving moderate PD-1... - Cell proliferation. Specifically, γδT cells showed the highest degree of proliferation; CD4 + and CD8 + Initial T cells represent the lowest degree of cell proliferation; and CD8... + Stem cell memory cells represent the most extensive related cell population.

[0166] Figures 106A-106EThe A)CD8 of cynomolgus macaques treated with either XENP30516 (mAb C_H1_L1.1 x IL-15 [D30N / E64Q / N65D]) or XENP30518 (αRSV x IL-15 [D30N / E64Q / N65D]) is shown. + PD1 + A) T cells, B) CD8 stem cell memory cells, C) CD8 naive cells, D) γδ T cells, and E) CD56 + NK cell expansion. In summary, the data show that XENP30516 (with high PD-1 affinity and low IL-15 potency) can maintain peripheral pharmacodynamics and achieve insignificant PD-1... - Cell proliferation. Specifically, γδT cells showed the highest degree of proliferation; CD4 + and CD8 + Initial T cells represent the lowest degree of cell proliferation; and CD8... + Stem cell memory cells represent the most extensive related cell population.

[0167] Figure 107 The changes in serum concentrations over time in cynomolgus macaques administered XENP30290, XENP30291, XENP29439, or XENP30516 are shown. Data indicate that XENP30290, with the highest PD-1 affinity and higher IL-15 potency, was cleared more rapidly than XENP30291 and XENP29439, which have lower PD-1 affinity. However, XENP30516, with the highest PD-1 affinity but lower IL-15 potency, resulted in slower clearance than XENP30290.

[0168] Figure 108 The changes in serum concentrations over time in cynomolgus macaques administered XENP30290, XENP30292, or XENP30516 are shown. Data indicate that XENP30516, with its lower IL-15 potency, resulted in a slower clearance rate than XENP30290.

[0169] Figure 109 The changes in serum concentrations over time in cynomolgus monkeys that received XENP30291 or XENP30293 are shown.

[0170] Figure 110 The changes in serum concentrations over time in cynomolgus monkeys that received XENP29439 or XENP30302 are shown.

[0171] Figure 111The changes in serum concentrations over time in cynomolgus macaques administered XENP30362 or XENP30518 are shown. Data indicate that higher IL-15 potency is associated with faster clearance.

[0172] Figures 112A-112C The changes in PD-1 expression over time in various lymphocyte populations in cynomolgus monkeys treated with A) XENP30290, B) XENP30516, or C) XENP30362 are shown. Data indicate that PD-1 targeting of the IL-15 / Rα-Fc fusion protein increased PD-1 expression, while targeting the IL-15 / Rα-Fc fusion protein with the control RSV did not have this effect.

[0173] Figure 113 The correlation between the peak amplification fold of all T cell memory subsets and the peak PD-1 expression induced by XENP30290 or XENP30516 is shown.

[0174] Figure 114 The sequences of exemplary IL-15 variants modified to eliminate glycosylation are shown. Each of these variant IL-15 sequences contains 90%, 95%, 98%, and 99% identical sequences to the enumerated sequences (as defined herein), and / or contains 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 additional amino acid substitutions. As will be apparent to those skilled in the art, the IL-15 variants can be used in any IL-15 / Rα-Fc fusion protein and PD-1-targeted IL-15 / Rα-Fc fusion protein described herein. Furthermore, the various IL-15 modifications described herein can be used alone or in combination with any other IL-15 modifications described herein.

[0175] Figures 115A-115C The sequence of an exemplary IL-15 / Rα-Fc fusion protein modified to remove glycosylation in the form of “IL-15 / Rα-heteroFc” is shown. IL-15 and IL-15Rα (sushi) are underlined, the linker is underlined twice (although, as those skilled in the art will understand, the linker can be replaced by other linkers, some of which are shown in Figure 7), and a slash ( / ) indicates the boundary between IL-15, IL-15Rα, the linker, and the Fc region. As will be apparent to those skilled in the art, each of the IL-15 / Rα-Fc fusion proteins may also include or exclude Xtend Fc (M428L / N434S).

[0176] Figures 116A-116GThe sequence of an exemplary PD-1-targeting IL-15 / Rα-Fc fusion protein modified to remove glycosylation, “scIL-15 / Rαx Fab,” is shown. CDRs are underlined. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). IL-15 and IL-15Rα (sushi) are indicated in italics, and the adapter is double-underlined (although, as those skilled in the art will recognize, the adapter can be replaced by other adapters, some of which are in…). Figure 9 and Figures 10A to 116G (as shown in the diagram), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the linker, the variable region, and the constant / Fc region. As will be apparent to those skilled in the art, each of the targeted IL-15 / Rα-Fc fusion proteins may also include or exclude Xtend Fc (M428L / N434S).

[0177] Figures 117A-117F The sequence of the control RSV targeting IL-15 / Rα-Fc fusion protein, modified to remove glycosylation of “scIL-15 / Rαx Fab,” is shown. CDRs are underlined. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). IL-15 and IL-15Rα (sushi) are shown in italics, and the adapter is double-underlined (although, as those skilled in the art will understand, the adapter can be replaced by other adapters, some of which are in…). Figure 9 (as shown in Figure 10), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the linker, the variable region, and the constant / Fc region. As will be apparent to those skilled in the art, each of the targeted IL-15 / Rα-Fc fusion proteins may also include or exclude Xtend Fc (M428L / N434S).

[0178] Figures 118A to 118B The following are shown: i) Chromatograms showing the purified fractions 2 of XENP30516, XENP31981, XENP31982, XENP31984, XENP31985, XENP31986 and XENP31987; and ii) Analytical anion exchange chromatography (AIEX) characterization results of the main peaks obtained by anion exchange separation (e.g., Figure 154A (As shown). Figure 118A XENP30516, XENP31981, XENP31982, XENP31984 and XENP31985 are shown. Figure 118B XENP31986 and XENP31987 are shown. The constructs are XENP30516 (mAb CH1_L1.1 x IL-15 [D30N / E64Q / N65D] w / (G4S) connector (SEQ ID NO:7)), XENP31981 (mAb CH1_L1.1 x IL-15 [D30N / E64Q / N65D / N71Q / N79Q] ​​w / (G4S) connector (SEQ ID NO:7)), XENP31982 (mAb CH1_L1.1 x IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q] w / (G4S) connector (SEQ ID NO:7)), and XENP31984 (mAb CH1_L1.1 x IL-15[D30N / E64Q / N65D]w / (G4A) connector (SEQ ID NO:8)), XENP31985(mAb CH1_L1.1 x IL-15[D30N / E64Q / N65D / N71Q / N79Q / S114A]w / (G4A) connector (SEQ ID NO:8)), XENP31986(mAb CH1_L1.1 xIL-15[D30N / E64Q / N65D / N71Q / N79Q / N112Q]w / (G4A) connector (SEQ ID NO:8)) and XENP31987(mAb CH1_L1.1 x IL-15[D30N / E64Q / N65D / N71Q / N79Q / S114_]w / (G4A) connector (SEQ ID NO:8)).

[0179] Figure 119 This study demonstrates the induction of CD8 effector memory T cell proliferation by a glycosylated PD-1-targeting IL-15 / Rα-Fc fusion protein. Various assays containing glycosylated IL-15 variants retained their activity, and their potency was unexpectedly greater than that of unmodified IL-15 glycosylated XENP30516 (and the corresponding XENP31984 with a Gly-Ala linker). The “(G4S) linker” is SEQ ID NO:7 and the “(G4A) linker” is SEQ ID NO:8.

[0180] Figure 120This study demonstrates the induction of CD4 and CD8 effector memory T cell proliferation by glycogen-modified PD-1 targeting IL-15 / Rα-Fc fusion protein and glycogen-modified RSV targeting IL-15 / Rα-Fc fusion protein. The study shows a fold increase in potency of the PD-1 targeting IL-15 / Rα-Fc fusion protein relative to the corresponding RSV targeting IL-15 / Rα-Fc fusion protein, indicating good selectivity (higher selectivity for TILs than for peripheral lymphocytes).

[0181] Figures 121A-121D The study illustrates the proliferation of CD8 effector memory T cells induced by A) XENP30290 vs. XENP30362; B) XENP31979 vs. XENP32163; C) XENP30516 vs. XENP30518; and D) XENP31986 vs. XENP32169. Data show that removal of N-linked glycosylation improves potency and selectivity. The “(G4S) adapter” is SEQ ID NO:7 and the “(G4A) adapter” is SEQ ID NO:8.

[0182] Figure 122 Tumor volume (as determined by caliper measurement) over time is shown in NSG mice transplanted with pp65-MCF7 and huPBMCs, administered alone or in combination with a glycogenotyped PD-1-targeting IL-15 / Rα-Fc fusion protein. The glycogenotype XENP319816 exhibits enhanced antitumor activity (even at lower doses) and binds efficiently to the PD-1 blocker. The “(G4S) adapter” is SEQ ID NO:7 and the “(G4A) adapter” is SEQ ID NO:8.

[0183] Figure 123 The changes in in vivo tumor volume (as measured by calipers) up to day 17 in NSG mice transplanted with pp65-MCF7 and huPBMCs after receiving the glycogen-modified PD-1-targeting IL-15 / Rα-Fc fusion protein alone or in combination with a PD-1 blocker are shown.

[0184] Figure 124 The CD8+ cell count in the blood of NSG mice transplanted with pp65-MCF7 and huPBMC was shown on day 14 after administration of the glycogen-modified PD-1-targeting IL-15 / Rα-Fc fusion protein alone or in combination with a PD-1 blocker.

[0185] Figures 125A-125EThe expansion of A) CD8+PD1+ T cells, B) CD8+ stem cell memory cells, C) CD8 naïve cells (low PD1), D) γδ cells, and E) NK cells in cynomolgus macaques receiving a low dose of 1X XENP30290 (PD1 x IL15 [N4D / N65D]) is shown. The data indicate that XENP30290 induced PD1+ cell expansion at a low dose of 1X.

[0186] Figures 126A-126E The expansion of A) CD8+PD1+ T cells, B) CD8+ stem cell memory cells, C) CD8 naïve cells (low PD1), D) γδ cells, and E) NK cells in cynomolgus monkeys receiving 10X high-dose XENP30290 (PD1 x IL15 [N4D / N65D]) and the corresponding RSV-targeting alternative XENP30362 is shown. Data show that XENP30290 induced good PD1+ cell expansion at the 10X high dose, but exhibited moderate activity against PD1 cells, as indicated by the activity of the RSV-targeting control XENP30362. The “(G4S) adapter” is SEQ ID NO:7.

[0187] Figures 127A-127E The A)CD8 of cynomolgus macaques receiving 10X high-dose XENP30516 (PD1 x IL15 [D30N / E64Q / N65D]) is shown. + PD1 + T cells, B)CD8 + Stem cell memory cells, C)CD8 naïve cells (PD1) 低 The expansion of D) γδ cells and E) NK cells. Data showed that XENP30516 induced good PD1 at a high dose of 10X. + Cell expansion.

[0188] Figures 128A-128E The expansion of A) CD8+PD1+ T cells, B) CD8+ stem cell memory cells, C) CD8 naïve (PD1 x IL15 [D30N / E64Q / N65D]) cells, D) γδ cells, and E) NK cells in cynomolgus macaques treated with 30X very high doses of XENP30516 (PD1 x IL15 [D30N / E64Q / N65D]) and the corresponding RSV-targeting alternative XENP30518 is shown. Data show that XENP30516 induces greater PD1+ expansion at 30X very high doses while maintaining beneficial selectivity. The “(G4S) adapter” is SEQ ID NO:7.

[0189] Figures 129A-129EThe A)CD8 of cynomolgus macaques receiving 1X low-dose XENP31986 (PD1 x IL15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q]) and the corresponding RSV-targeting alternative XENP32169 are shown. + PD1 + T cells, B)CD8 + Stem cell memory cells, C)CD8 initial (PD1) 低 The expansion of PD1 cells, D)γδ cells, and E)NK cells was observed. Data showed that XENP31986 induced good PD1 at low doses of 1X. + It amplifies while exhibiting excellent selectivity. The “(G4A) adapter” is SEQ ID NO:8.

[0190] Figures 130A-130E This study demonstrates the A)CD8 activity of cynomolgus macaques administered 3X moderate doses of XENP31986 (PD1xIL15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q]) and the corresponding RSV-targeting alternative XENP32169. + PD1 + T cells, B)CD8 + Stem cell memory cells, C)CD8 initial (PD1) 低 The expansion of PD1 cells, D)γδ cells, and E)NK cells was observed. Data showed that XENP31986 induced enhanced PD1 at moderate doses of 3X. + Cell expansion was achieved while maintaining excellent selectivity. The "(G4A) adapter" is SEQ ID NO:8.

[0191] Figures 131A-131E This study demonstrates the A)CD8 activity of cynomolgus monkeys administered 10X high doses of XENP31986 (PD1 x IL15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q]) and the corresponding RSV-targeting alternative XENP32169. + PD1 + T cells, B)CD8 + Stem cell memory cells, C)CD8 initial (PD1) 低 The expansion of PD1 cells, D)γδ cells, and E)NK cells was observed. Data showed that XENP31986 induced very large PD1 cell proliferation at a high dose of 10X. + Cell expansion was achieved while maintaining excellent selectivity. The "(G4A) adapter" is SEQ ID NO:8.

[0192] Figure 132This summary reflects the effects of various concentrations of PD-1 targeted or RSV targeted IL-15 / Rα-Fc fusion proteins containing IL-15 [N4D / N65D], IL-15 [D30N / E64Q / N65D], or IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q] on CD8+ levels in cynomolgus monkeys. + PD1 + T cell expansion.

[0193] Figures 133A-133E This image shows A)CD8 in cynomolgus macaques receiving various concentrations of PD-1-targeting or RSV-targeting IL-15 / Rα-Fc fusion proteins containing IL-15 [N4D / N65D], IL-15 [D30N / E64Q / N65D], or IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q]. + PD1 + T cells, B)CD8 + Stem cell memory cells, C)CD8 initial (PD1) 低 The correlation between the selectivity ratios of D)γδ cells and E)NK cells and peak amplification folds was investigated. In summary, the data showed that increasing the dose improved selectivity. Furthermore, the PD-1 targeting IL-15 / Rα-Fc fusion protein containing the IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q] variants showed higher selectivity than the PD-1 targeting IL-15 / Rα-Fc fusion protein containing the IL-15 [D30N / E64Q / N65D] variants, and subsequently higher selectivity than the PD-1 targeting IL-15 / Rα-Fc fusion protein containing the IL-15 [N4D / N65D] variants.

[0194] Figure 134 The changes in serum concentrations of the specified test substance in cynomolgus monkeys over time are shown. Surprisingly, at the same dose, compared to glycosylated PD-1 targeting IL-15 / Rα-Fc XENP31896 containing the IL-15[D30N / E64Q / N65D / N71Q / N79Q / N112Q] variants, XENP31896 exhibited enhanced pharmacokinetic characteristics, despite XENP31896 having enhanced potency / pharmacodynamics compared to XENP30516. The “(G4S) linker” is SEQ ID NO:7 and the “(G4A) linker” is SEQ ID NO:8.

[0195] Figure 135The changes in serum concentrations of the specified test substance in cynomolgus monkeys over time are shown. At the same dose, deglycosylated RSV targeting IL-15 / Rα-Fc XENP32169 containing the IL-15[D30N / E64Q / N65D / N71Q / N79Q / N112Q] variants exhibited enhanced pharmacokinetic characteristics compared to glycosylated RSV targeting IL-15 / Rα-Fc XENP32169 containing the IL-15[D30N / E64Q / N65D] variants. The “(G4S) linker” is SEQ ID NO:7 and the “(G4A) linker” is SEQ ID NO:8.

[0196] Figure 136 The Octet sensing plot shows the binding of the Trp-modified mAb C[PD-1]_H1L1 variant compared to WT mAb C[PD-1]_H1L1. Numbered according to Kabat. The data show that repair of unstable tryptophan leads to complete loss or significantly reduced binding to PD-1.

[0197] Figures 137A-137G The sequence shown is an exemplary Trp-modified and affinity-repaired variant of mAb C[PD-1]_H1L1 in the form of bivalent human IgG1, containing E233P / L234V / L235A / G236del / S267K substitutions in the heavy chain. CDRs are underlined, and slashes indicate the boundaries of variable domains. As described herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). As those skilled in the art will understand, the VH and VL domains may be formatted as Fab or scFv for use in the PD-1-targeting IL-15 / Rα-Fc fusion protein of the present invention.

[0198] Figures 138A-138GThe sequence shown is that of an exemplary PD-1 targeting IL-15 / Rα-Fc fusion protein containing an exemplary Trp-modified and affinity-repaired variant of mAb C[PD-1]_H1L1. The CDR is underlined. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). IL-15 and IL-15Rα (sushi) are shown in italics, and the adapter is double-underlined (although, as those skilled in the art will understand, the adapter can be replaced by other adapters, some of which are in…). Figure 9 (as shown in Figure 10), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the linker, the variable region, and the constant / Fc region. As will be apparent to those skilled in the art, each of the targeted IL-15 / Rα-Fc fusion proteins may also include or exclude Xtend Fc (M428L / N434S).

[0199] Figure 139 The apparent dissociation constant (KDapp), binding rate (ka), and dissociation rate (kd) of the Trp-modified mAb C[PD-1]_H1L1 variants containing restored PD-1 binding are shown, as measured by Octet. Substitutions in the heavy chain or variable light domains listed are based on Xencor numbering (the corresponding Kabat positions are listed in the next column). Several variants were identified whose restored PD-1 affinity binding is close to mAb C_H1_L1.1 (including mAb C_H1.176_L1.140, mAb C_H1.177_L1.140, and mAb C_H1.180_L1.140).

[0200] Figures 140A-140B The Octet sensing plots show the binding of A) XENP31986 (with mAb C[PD1]_H1_L1.1) and B) XENP32435 (with Trp-modified / affinity-repaired mAb C[PD1]_H1.176_L1.140) to PD-1. The data indicate that the repaired molecules have equivalent affinity for PD-1.

[0201] Figure 141This study demonstrates the induction of CD8 effector memory T cell proliferation by a Trp-modified PD-1 targeting IL-15 / Rα-Fc fusion protein. XENP32435, XENP32436, and XENP32439, modified to remove the Trp-binding domain, exhibited the same or higher CD8 proliferation rates compared to XENP31986. + The potency of effector memory T cells (despite having a similar affinity for PD-1).

[0202] Figure 142 The sequences of exemplary IL-15 / Rα-Fc fusion proteins XENP20818 (WT IL-15), XENP22821 (IL-15[N65D]), and XENP24045 (IL-15[D30N / E64Q / N65D]) in the form of “IL-15 / Rα-heteroFc” are shown. IL-15 and IL-15Rα (sushi) are underlined, the linker is underlined twice (although, as those skilled in the art will understand, the linker can be replaced by other linkers, some of which are shown in Figure 7), and the slash ( / ) indicates the boundary between IL-15, IL-15Rα, the linker, and the Fc region.

[0203] Figure 143 This study illustrates the proliferation of rapamycin-amplified Tregs following treatment with untargeted IL-15 / Rα-Fc fusion proteins XENP20818 and XENP24045, as well as various PD-1 targeted or control RSV-targeted IL-15 / Rα-Fc fusion proteins. Data show that IL-15 / Rα-Fc fusion proteins (targeted and untargeted) induce the proliferation of rapamycin-amplified Tregs (as measured by Tag-it UV dilution). Notably, PD-1 targeting of IL-15 / Rα-Fc fusion proteins is significantly less potent in inducing Treg proliferation compared to untargeted IL-15 / Rα-Fc fusion proteins.

[0204] Figures 144A-144B The diagram shows the incubation of 1 x 10 Tregs amplified with 5 μg / ml of the specified test sample and an increased amount of rapamycin. 5 Expansion of A) CD8 effector memory T cells and B) CD4 effector memory T cells after CFSE-labeled PBMCs (as measured by CFSE dilution). Data showed that PD-1 targeting of the IL-15 / Rα-Fc fusion protein altered (reduced) the inhibitory potency of Treg-induced CD8 and CD4 effector memory T cell proliferation. Notably, the alteration induced by RSV targeting of the IL-15 / Rα-Fc fusion protein was less than the potency reduction induced by PD-1 targeting of the IL-15 / Rα-Fc fusion protein.

[0205] Figures 145A-145B The diagram shows the incubation of 1 x 10 Tregs amplified with 5 μg / ml of the specified test sample and an increased amount of rapamycin. 5 The ratios of A) Tregs to CD8 effector memory T cells and B) Tregs to CD4 effector memory T cells were measured after CFSE-labeled PBMCs. Data showed that, compared to no assay, PD-1 targeting the IL-15 / Rα-Fc fusion protein increased the Treg / T ratio. EM The ratio, while PD-1 targeting the IL-15 / Rα-Fc fusion protein enhanced T EM Cell proliferation. This indicates that although Tregs are amplified, the amplified Tregs exhibit decreased inhibitory capacity.

[0206] Figures 146A-146B The proliferation of CD3-stimulated PBMCs after pre-culturing rapamycin-amplified Tregs for 6 days in complete Treg medium (RPMI containing 10% FBS, 0.5 μg / ml anti-CD28, 100 U / ml IL-2, and 100 ng / ml rapamycin); complete Treg medium without rapamycin; or complete Treg medium containing 100 ng / ml IL-15 (in RPMI containing 10% FBS, 0.5 μg / ml anti-CD28, and without IL-2 and rapamycin) is shown. A) CD8 + T cells and B)CD4 + The percentage of T cells (as measured by CFSE dilution). Data showed that the inhibitory capacity of Tregs pretreated with IL-15 was weakened.

[0207] Figures 147A-147B The image shows CD4 counts in PBMCs after treatment with either A) or B) 5 μg / ml IL-15 / Rα-Fc fusion protein XENP22821 for 14 days. + The expression of CD25 and FOXP3 on T cells was shown to be reduced by XENP22821 treatment. + Expression on T cell populations.

[0208] Figures 148A-148B The image shows CD4 counts in PBMCs after treatment with either A) or B) 5 μg / ml IL-15 / Rα-Fc fusion protein XENP22821 for 14 days. + Expression of CD45RA and FOXP3 on T cells. Data showed that XENP22821 treatment reduced CD45RA and FOXP3 expression. + CD45RA - Group from FoxP3 高 Change to FoxP3 低This indicates that treatment with the IL-15 / Rα-Fc fusion protein actually changed the population from eTreg (population decreased from 5.24% to 2.72%) to activated effector CD4 T cells (population increased from 18.2% to 28.6%).

[0209] Figures 149A-149B The image shows CD4 counts in PBMCs after treatment with either A) or B) 5 μg / ml IL-15 / Rα-Fc fusion protein XENP22821 for 14 days. + The expression of CD25 and CCR4 on T cells was shown to be reduced by XENP22821 treatment. + Expression on T cell populations.

[0210] Figures 150A-150B The proliferation of A) CD8 T cells and B) CD4 T cells after incubation with 100 ng / ml of 100 ng / ml PBMCs bound to anti-CD3 (OKT3) and labeled with CFSE (as measured by CFSE dilution) using 10 μg / ml of the IL-15 / Rα-Fc fusion protein XENP24045 and a specified concentration of TGFβ1 is shown. The data show that TGFβ inhibits T cell proliferation in a dose-dependent manner; however, notably, the IL-15 / Rα-Fc fusion protein prevented the inhibition of T cell proliferation by TGFβ at all tested doses.

[0211] Figures 151A-151B The apparent dissociation constant (Ki) of the Trp-modified mAb C[PD-1]_H1L1 variant containing the repaired PD-1 binding is shown. Dapp ), binding rate (k a ) and dissociation rate (k d (as measured by Octet). The substitutions in the listed variable restructure domains are based on Xencor numbering (the corresponding Kabat positions are listed in the next column).

[0212] Figures 152A-152N Additional illustrative anti-PD-1ABD variants, including variable heavy and light chains, are shown that do not compete with nivolumab or pembrolizumab. CDRs are underlined. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1; therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems).

[0213] Figure 153A and Figure 153B The sequence of an exemplary PD-1-targeted IL-15 construct is shown. Figure 153AXENP31326 is displayed. Figure 153B XENP31329 is shown. As stated herein, and true for every sequence containing a CDR herein, the exact identification of the CDR location may vary slightly depending on the numbering used as shown in Table 1. Therefore, this document includes not only underlined CDRs but also CDRs contained within the VH and VL domains (using other numbering systems). IL-15 and IL-15Rα (sushi) are shown in italics, with the connector underlined (although, as those skilled in the art will understand, the connector can be replaced by other connectors, some of which are in...). Figure 9 (as shown in Figure 10), and the slashes ( / ) indicate the boundaries between IL-15, IL-15Rα, the linker, the variable region, and the constant / Fc region. As will be apparent to those skilled in the art, each of the targeted IL-15 / Rα-Fc fusion proteins may also include or exclude Xtend Fc (M428L / N434S).

[0214] Figures 154A-154C The data show the induction of CD3 effector memory T cell proliferation by XENP31326 vs. XENP31329, XENP30516 and XENP32927 vs. XENP30518, and XENP31986 and XENP32435 vs. XENP32169. The data show that targeting the fusion protein alone (as shown in the comparison of XENP31326 vs. XENP31329) provided limited selectivity (the EC50 of the RSV-targeting fusion protein differed by 2.4-fold from that of the PD1-targeting fusion protein). PD1 targeting combined with decreased IL-15 potency (as shown in the comparisons of XENP30516 and XENP32927 vs. XENP30518 and XENP31986 and XENP32435 vs. XENP32169) provided significantly enhanced selectivity. Notably and surprisingly, the deglycosylated variant further improved selectivity (the deglycosylated variant had a selectivity of 246-fold, while the non-deglycosylated variant had a selectivity of only 105-fold).

[0215] Figure 155A and Figure 155B This shows the Treg amplification in the presence of XmAb24306 and various concentrations of rapamycin ( Figure 155A CD8+ and ( Figure 155B CD4+ response T cell proliferation.

[0216] Figure 156The proliferation of rapamycin-amplified Tregs after treatment with the IL-15 / Rα-Fc fusion proteins XENP20818 and XENP24045 is shown. Data indicate that the IL-15 / Rα-Fc fusion proteins induce the proliferation of rapamycin-amplified Tregs (as measured by Tag-it UV dilution). Notably, XENP24045 exhibits decreased potency in inducing Treg proliferation compared to XENP20818.

[0217] Figure 157A and Figure 157B The diagram shows the incubation of 1 x 10 Tregs amplified with 5 μg / ml of the specified test sample and an increased amount of rapamycin. 5 Expansion of A) CD8 effector memory T cells and B) CD4 effector memory T cells following CFSE-labeled PBMCs (as measured by CFSE dilution). Data showed that the IL-15 / Rα-Fc fusion protein altered (reduced) the inhibitory efficacy of Treg-induced CD8 and CD4 effector memory T cell proliferation.

[0218] Figure 158A and Figure 158B The diagram shows the incubation of 1 x 10 Tregs amplified with 5 μg / ml of the specified test sample and an increased amount of rapamycin. 5 The ratios of A) Tregs to CD8 effector memory T cells and B) Tregs to CD4 effector memory T cells were measured after CFSE-labeled PBMCs. Data showed that, compared to no test sample, the IL-15 / Rα-Fc fusion protein increased the Treg / T ratio. EM The ratio, while the IL-15 / Rα-Fc fusion protein enhanced T EM Cell proliferation. This indicates that although Tregs are amplified, the amplified Tregs exhibit decreased inhibitory capacity.

[0219] Figure 159A and Figure 159B The proliferation of CD3-stimulated PBMCs after pre-culturing rapamycin-amplified Tregs for 6 days in complete Treg medium (RPMI containing 10% FBS, 0.5 μg / ml anti-CD28, 100 U / ml IL-2, and 100 ng / ml rapamycin); complete Treg medium without rapamycin; or complete Treg medium containing 100 ng / ml IL-15 (in RPMI containing 10% FBS, 0.5 μg / ml anti-CD28, and without IL-2 and rapamycin) is shown. A) CD8 + T cells and B)CD4 + The percentage of T cells (as measured by CFSE dilution). Data showed that the inhibitory capacity of Tregs pretreated with IL-15 was weakened.

[0220] Figure 160A and Figure 160B The image shows CD4 counts in PBMCs after treatment with either A) or B) 5 μg / ml IL-15 / Rα-Fc fusion protein XENP22821 for 14 days. + The expression of CD25 and FOXP3 on T cells was shown to be reduced by XENP22821 treatment. + Expression on T cell populations.

[0221] Figure 161A and Figure 161B The image shows CD4 counts in PBMCs after treatment with either A) or B) 5 μg / ml IL-15 / Rα-Fc fusion protein XENP22821 for 14 days. + Expression of CD45RA and FOXP3 on T cells. Data showed that XENP22821 treatment reduced CD45RA and FOXP3 expression. + CD45RA - Group from FoxP3 高 Change to FoxP3 低 This indicates that treatment with the IL-15 / Rα-Fc fusion protein actually changed the population from eTreg (population decreased from 5.24% to 2.72%) to activated effector CD4 T cells (population increased from 18.2% to 28.6%).

[0222] Figure 162A and Figure 162B The image shows CD4 counts in PBMCs after treatment with either A) or B) 5 μg / ml IL-15 / Rα-Fc fusion protein XENP22821 for 14 days. + The expression of CD25 and CCR4 on T cells was shown to be reduced by XENP22821 treatment. + Expression on T cell populations.

[0223] Figure 163A and Figure 163B The proliferation of A) CD8 T cells and B) CD4 T cells after incubation with 100 ng / ml of 100 ng / ml PBMCs bound to anti-CD3 (OKT3) and labeled with CFSE (as measured by CFSE dilution) using 10 μg / ml of the IL-15 / Rα-Fc fusion protein XENP24045 and a specified concentration of TGFβ1 is shown. The data show that TGFβ inhibits T cell proliferation in a dose-dependent manner; however, notably, the IL-15 / Rα-Fc fusion protein prevented the inhibition of T cell proliferation by TGFβ at all tested doses.

[0224] Figure 164This study illustrates the changes in tumor volume (as measured by calipers) over time in pp65-MCF7-transplanted huCD34+ NSG mice after administration of a single dose of PD-1 blockade, the reduced-potency non-targeting IL-15 / Rα-Fc fusion protein XENP24045 (as a monotherapy or in combination with a PD-1 blockade), or the reduced-potency [NC]PD-1-targeting glycogen-modified IL-15 / Rα-Fc. All groups showed significantly enhanced tumor activity on day 15 compared to controls treated with PBS (p≤0.05), and exhibited a clear advantage in antitumor activity when used in combination with a PD-1 blockade.

[0225] Figure 165 This study demonstrates the expansion of CD4+ T cells and CD8+ T cells on days 7 and 13 in pp65-MCF7 transplanted huCD34+ NSG mice after administration of a single dose of PD-1 inhibitor, the reduced-potency non-targeted IL-15 / Rα-Fc fusion protein XENP24045 (administered as monotherapy or in combination with a PD-1 inhibitor), or [NC]PD-1-targeted reduced-potency glycogen-modified IL-15 / Rα-Fc. Compared to the non-targeted IL-15-Fc fusion protein in combination with a PD-1 inhibitor, the combination of PD-1-targeted IL-15 and a PD-1 inhibitor significantly enhanced the expansion of both CD8+ T cells and CD4+ T cells. Compared to PD-1-targeted IL-15 monotherapy, the combination of PD-1-targeted IL-15 and a PD-1 inhibitor significantly enhanced lymphocyte expansion.

[0226] Figure 166 The CD8 T cell to Treg ratio is shown on day 13 in pp65-MCF7 transplanted huCD34+ NSG mice after administration of a single dose of PD-1 inhibitor, the reduced-potency non-targeted IL-15 / Rα-Fc fusion protein XENP24045 (as monotherapy or in combination with a PD-1 inhibitor), or the reduced-potency [NC]PD-1 targeted glycogen-modified IL-15 / Rα-Fc. PD-1 targeting of IL-15 improved the CD8:Treg ratio (and was further improved when used in combination with a PD-1 inhibitor). Unpaired t-tests were used for statistical analysis of the logarithmically transformed data.

[0227] Figures 167A-167EThe following data are presented on day 9 after pp65-MCF7 transplanted huCD34+ NSG mice received a single dose of PD-1 inhibitor or [NC]PD-1 targeting potency-degraded glycogen-modified IL-15 / Rα-Fc (administered as monotherapy or in combination with PD-1 inhibitor): A) Percentage of CD3+ T cells in the total CD45 lymphocyte population, B) Percentage of effector memory CD8 T cells in the total CD8 cell population, C) Percentage of naïve CD8 cells in the total CD8 cell population, D) Percentage of Treg cells in the total CD3 cell population, and E) Ratio of effector memory CD8 T cells to Treg cells. Data show that XENP32986, used alone or in combination with PD-1 inhibitors, can increase the total CD3+ T cells in tumor tissue and change the T cell phenotype to effector memory CD8+; XENP32986, used alone or in combination with PD-1 inhibitors, can increase the CD8:Treg ratio in tumor tissue; and low-dose (0.01 mg / kg) XENP32986, when used in combination with PD-1 inhibitors, can significantly improve efficacy.

[0228] Figures 168A-168L The changes over time in tumor volume (as measured by calipers) of NSG-DKO mice transplanted with huPBMC and pp65-MCF7 after administration of single-dose PD-1 blockade, PD-1-targeted IL-15 [D30N / E64Q / N65D]XENP32927 (as a single agent at concentrations of 0.1, 0.3, or 1 mg / kg or in combination with 0.3 mg / kg of PD-1 blockade), PD-1-targeted IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q]XENP32435 (as a single agent at concentrations of 0.03, 0.1, or 0.3 mg / kg or in combination with 0.1 mg / kg of PD-1 blockade), and RSV-targeted IL-15 control in combination with PD-1 blockade are shown after baseline correction.

[0229] Figure 169The changes in tumor volume (as measured by calipers) over time in NSG-DKO mice transplanted with huPBMC and pp65-MCF7 after administration of a single dose of PD-1 inhibitor, PD-1-targeted IL-15 [D30N / E64Q / N65D] XENP32927 (used as a single agent at concentrations of 0.1, 0.3, or 1 mg / kg or in combination with a PD-1 inhibitor at 0.3 mg / kg), and RSV-targeted IL-15 [D30N / E64Q / N65D] control in combination with PD-1 inhibitor are shown after baseline correction. Compared with PD-1 blockade alone, XENP32927, when used in combination with PD-1 antagonists at a concentration of 0.3 mg / kg, resulted in significant tumor regression (p<0.05 at days 17, 20, 22, and 27), and when used in combination with PD-1 antagonists at a concentration of 1 mg / kg, it also resulted in significant tumor regression (p<0.05 at days 15 and 17) (statistical analysis of baseline-corrected data was performed using the Mann-Whitney test).

[0230] Figure 170 The changes in tumor volume (as measured by calipers) over time after administration of huPBMC and pp65-MCF7 transplanted NSG-DKO mice to single-dose PD-1 blockade, PD-1-targeted IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q] XENP32435 (as a single agent at concentrations of 0.03, 0.1, or 0.3 mg / kg or in combination with 0.1 mg / kg of PD-1 blockade), and RSV-targeted IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q] control in combination with PD-1 blockade are shown. Compared with PD-1 blockade alone, XENP32435, when used in combination with a PD-1 antagonist at a concentration of 0.1 mg / kg, resulted in significant tumor regression (p<0.05 at days 13 and 15), and also at a concentration of 0.3 mg / kg (p<0.05 at days 13, 15, and 17). Notably, XENP32435 demonstrated significant tumor regression in this model (p<0.05 at day 17 compared with PD-1 blockade monotherapy). Statistical analysis of baseline-corrected data was performed using the Mann-Whitney test.

[0231] Figure 171Tumor volume changes on day 17 in NSG-DKO mice transplanted with huPBMC and pp65-MCF7 after administration of single-dose PD-1 inhibitor, PD-1-targeted IL-15 [D30N / E64Q / N65D]XENP32927 (as monotherapy or in combination with 0.3 mg / kg PD-1 inhibitor), PD-1-targeted IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q]XENP32435 (as monotherapy or in combination with 0.1 mg / kg PD-1 inhibitor), and RSV-targeted IL-15 control in combination with PD-1 inhibitor. Baseline-corrected data were statistically analyzed using the Mann-Whitney test. Data shows that RSV targeting IL-15 (including IL-15 variants) does not enhance activity when used in combination with PD-1 inhibitors compared to using PD-1 blockers alone.

[0232] Figures 172A-172D The results show that NSG-DKO mice transplanted with huPBMC and pp65-MCF7 received a single dose of PD-1 inhibitor, PD-1-targeting IL-15 [D30N / E64Q / N65D] XENP32927 (used as a single agent at concentrations of 0.1, 0.3, or 1 mg / kg, or in combination with a 0.3 mg / kg PD-1 inhibitor), and an RSV-targeting IL-15 [D30N / E64Q / N65D] control in combination with a PD-1 inhibitor, on day 7, A) CD4+ T cell activation and B) CD8+ T cell activation (as shown by CD25 expression); and on day 14, C) CD4+ T cell count and D) CD8+ T cell count. The data show that XENP32927 is active in the concentration range of 0.1–1 mg / kg.

[0233] Figures 173A-173DThe results show that on day 7 after NSG-DKO mice transplanted with huPBMC and pp65-MCF7 received a single dose of PD-1 inhibitor, PD-1-targeting IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q] XENP32435 (used as a single agent at concentrations of 0.03, 0.1, or 0.3 mg / kg or in combination with 0.1 mg / kg of PD-1 inhibitor), and RSV-targeting IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q] control in combination with PD-1 inhibitor, A) activation of CD4+ T cells and B) activation of CD8+ T cells (as shown by CD25 expression); and on day 14, C) CD4+ T cell count and D) CD8+ T cell count. Data show that XENP32435 is active over a wide concentration range of 0.03–0.3 mg / kg.

[0234] Figure 174 The changes in serum IFNγ concentration over time in NSG-DKO mice transplanted with huPBMC and pp65-MCF7 after administration of a single dose of PD-1 inhibitor, PD-1-targeted IL-15 [D30N / E64Q / N65D] XENP32927 (used as a single agent at concentrations of 0.1, 0.3, or 1 mg / kg or in combination with a PD-1 inhibitor at 0.3 mg / kg), and RSV-targeted IL-15 [D30N / E64Q / N65D] control in combination with PD-1 inhibitor are shown.

[0235] Figure 175 The changes in serum IFNγ concentration over time in NSG-DKO mice transplanted with huPBMC and pp65-MCF7 after administration of a single dose of PD-1 inhibitor, PD-1-targeted IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q] XENP32435 (used as a single agent at concentrations of 0.03, 0.1, or 0.3 mg / kg or in combination with a PD-1 inhibitor at 0.1 mg / kg), and RSV-targeted IL-15 [D30N / E64Q / N65D / N71Q / N79Q / N112Q] control in combination with PD-1 inhibitor are shown.

[0236] Figures 176A-176E The expansion of A) CD8+ T cells, B) CD8+ stem cell memory T cells, C) CD8+ naïve T cells, D) γδ T cells and E) NK cells in cynomolgus monkeys that received an intravenous 1X dose of XENP32435 (PD1xIL15[D30N / E64Q / N65D / N71Q / N79Q / N112Q]) is shown.

[0237] Figures 177A-177E Showing compared to Figure 175 A~175E, expansion of A) CD8+ T cells, B) CD8+ stem cell memory T cells, C) CD8+ naïve T cells, D) γδ T cells and E) NK cells in cynomolgus monkeys that received intravenous administration of 12X doses of XENP32435 (PD1xIL15[D30N / E64Q / N65D / N71Q / N79Q / N112Q]).

[0238] Figures 178A-178E The expansion of A) CD8+ T cells, B) CD8+ stem cell memory T cells, C) CD8+ naïve T cells, D) γδ T cells and E) NK cells in cynomolgus monkeys that received an intravenous 1X dose of XENP32927 (PD1xIL15[D30N / E64Q / N65D]) is shown.

[0239] Figures 179A-179E Showing with Figures 177A-177E In contrast, the expansion of A) CD8+ T cells, B) CD8+ stem cell memory T cells, C) CD8+ naïve T cells, D) γδ T cells and E) NK cells in cynomolgus monkeys that received intravenous administration of 20X doses of XENP32927 (PD1xIL15[D30N / E64Q / N65D]) was observed.

[0240] Figure 180 The changes in serum concentrations of the test sample over time in cynomolgus monkeys treated with intravenous administration of 1X and 12X doses of deglycosylated XENP32435 are shown.

[0241] Figure 181 The changes in serum concentrations of the test sample over time in cynomolgus monkeys treated with intravenous administration of 1X and 20X doses of glycosylated XENP32927 are shown.

[0242] Figures 182A-182B The data shows the A) CD8+ T cell count and B) Ki-67 expression percentage in mouse spleen cells after incubation with the specified test sample. The data indicate that the mouse alternative PD-1 targeting IL-15 molecule exhibits high selectivity for PD1+ mouse T cells.

[0243] Figure 183A and 183BThe results show the following: A) CD8+ T cell count and B) Ki-67 expression percentage in mouse spleen cells after incubation with a specified test sample (containing XENP36217, an alternative mouse PD-1-targeting IL-15 molecule that binds to a different muPD-1 epitope than XENP33869) (i.e., non-competitive). The data also indicate that XENP36217 is highly selective for PD1+ mouse T cells.

[0244] Figures 184A-184K This study illustrates the antitumor and pharmacodynamic responses to administration of XENP33869, a PD-L1 blockade alternative to muPD1xIL15. A) Resected tumor weight recorded on days 3 and 6 post-grouping; B) Counts of CD8, CD4, Treg, NK, and NKT cells in the tumor, measured by flow cytometry; C) CD8:Treg ratio in the tumor; D) Frequency of CD8, CD4, Treg, NK, and NKT cells in the tumor, measured by flow cytometry; E) Frequency of CD8, CD4, Treg, NK, and NKT cells in the spleen, measured by flow cytometry; F) Counts of CD8, CD4, Treg, NK, and NKT cells in the spleen, measured by flow cytometry; G) CD8:Treg ratio in the spleen. H) Treg ratio; I) Frequency of GranzymeB+CD8, CD4, Treg, NK and NKT cells in tumors as measured by flow cytometry; J) Frequency of GranzymeB+CD8, CD4, Treg, NK and NKT cells in tumors as measured by flow cytometry; K) MFI of GranzymeB+CD8, CD4, Treg, NK and NKT cells in spleen as measured by flow cytometry.

[0245] Figures 185A-185HThe antitumor and pharmacodynamic responses of the PD-L1 blocking muPD1xIL15 alternative XEN33869 are shown compared with those of the non-blocking muPD1xIL15 alternatives XENP36213, XENP36216 and XENP36217. A) Tumor weight resected on day 6 after grouping; B) Counts of CD8, CD4, Treg, NK, and NKT cells in the tumor, measured by flow cytometry; C) Frequency of CD8, CD4, Treg, NK, and NKT cells in the tumor, measured by flow cytometry; D) CD8:Treg ratio in the tumor; E) Counts of CD8, CD4, Treg, NK, and NKT cells in the spleen, measured by flow cytometry; F) CD8:Treg ratio in the spleen; G) Frequency of CD8, CD4, Treg, NK, and NKT cells in the spleen, measured by flow cytometry; H) Frequency of Granzyme B+ CD8, CD4, Treg, NK, and NKT cells in the tumor and spleen, measured by flow cytometry.

[0246] Figure 186A and 186B This study shows the effect of intravenous administration of PD-L1 blockade muPD1xIL15 XENP33869 on the growth of an syngeneic tumor model in female C57Bl / 6 MC38 mice, compared to control and mouse reactive aPD-L1. A) Raw data and fitted results. B) Superposition of fitted growth curves for each group.

[0247] Figure 187A and 187B This study demonstrates the effect of PD-L1 blockade muPD1xIL15XENP33869 treatment on the growth of an MC38 syngeneic tumor model in female C57Bl / 6 mice after intravenous administration at various dose levels, compared to control and mouse-responsive aPD-L1. The combined activity of XENP33869 and anti-PDL1 was also evaluated. A) Raw data and fitted results. B) Superposition of fitted growth curves for each group.

[0248] Figure 188A and 188B This study demonstrates the effect of intravenous administration of the PD-L1 blockade muPD1xIL15 substitute XENP33869 on the growth of an MC38 syngeneic tumor model in female C57Bl / 6 mice compared to control and mouse-responsive aPD-L1 mice treated with the non-blockade muPD1xIL15 XENP36217. The combined activity of XENP33869 and XENP36217 with anti-PDL1 agents was also evaluated. A) Raw data and fitted results. B) Superposition of fitted growth curves for each group.

[0249] Figures 189A-189HThis image shows positron emission tomography (PET) imaging and quantification results of 18F-labeled anti-mouse CD8 tracers in female C57Bl / 6 mice carrying an MC38 syngeneic tumor model, treated with intravenous muPD1xIL15 substitute XENP33869 compared to those treated with the vector or mouse-responsive aPD-L1. A) Representative images of each treatment group, with tumor locations indicated by arrows. B) CD8 tracer uptake of tumors measured at 12 days post-treatment compared to the unbound control tracer. C) Time course of CD8 tracer uptake in blood, D) tumors, E) spleen, F) axillary lymph nodes, G) inguinal lymph nodes, and H) liver. Detailed Implementation

[0250] I. Naming Conventions

[0251] The heterodimeric fusion proteins of the present invention are listed in several different forms. Each polypeptide has a unique “XENP” number, although it will be understood in the art that longer sequences may contain shorter sequences. For example, a heavy chain containing a monomer for anti-PD-1 (see example...) Figure 28CThe first XENP number will be assigned, while the VH and VL domains may have different XENP numbers. Some molecules have three polypeptides, and therefore use the XENP number and its components as names. Thus, the molecule XENP29484 in “scIL-15 / RαX Fab” contains three sequences, commonly referred to as “XENP29484 chain 1,” “XENP29484 chain 2,” and “XENP29484 chain 3” or equivalents, although those skilled in the art will be able to easily identify these sequences by sequence alignment. These XENP numbers are used in sequence listings and identifiers, and are also used in the accompanying drawings. Furthermore, a molecule containing three components generates multiple sequence identifiers. For example, the Fab monomer list has three CDR sequences: full-length sequence, variable heavy chain sequence, and variable heavy chain sequence; the light chain has three CDR sequences: full-length sequence, variable light chain sequence, and variable light chain sequence; and the scFv-Fc domain has a full-length sequence, scFv sequence, variable light chain sequence, three light chain CDRs, scFv linker, variable heavy chain sequence, and three heavy chain CDRs. Note that all molecules containing the scFv domain in this paper use a single charged scFv linker (+H), but other linkers may also be used. Furthermore, the nomenclature for specific variable domains uses the format "Hx.xx_Ly.yy", where the number is a unique identifier for the specific variable chain sequence. Therefore, the variable domain on the Fab side of XENP30486 (binding to PD-1) is "H1.132", indicating that it uses the variable heavy chain domain H1.132, while in XENP30486, it binds to the light chain domain L1.135. Therefore, the name "mAbC[PD-1]_H1.132_L1.135" indicates the binding of the variable heavy chain domain H1.132 with the light chain domain L1.134. In the case of these sequences combining to form scFv, this name indicates that the scFv is positioned in the VH-linker-VL direction from the N-terminus to the C-terminus. A molecule with the same heavy chain and variable light chain domain sequences, but in reverse order, would be called "mAbC[PD-1]_L1.135_H1.132". Similarly, it is evident from the sequence listing and figures that different constructs can "mix and match" heavy and light chains.

[0252] II. Definition

[0253] To provide a more complete understanding of this application, several definitions are listed below. These definitions are intended to include syntactic equivalents.

[0254] As used herein, “ablation” refers to a reduction or elimination of activity. Thus, for example, “ablation of FcγR binding” means that the amino acid variant of the Fc region has less than 50% initial binding compared to the Fc region without the specific variant, wherein a loss of activity greater than 70-80-90-95-98% is preferred, and generally, the activity should be below the binding level detectable in a Biacore assay. Ablation of FcγR binding is particularly useful, as illustrated in Figure 6. However, unless otherwise stated, the Fc receptor of the present invention retains its binding to the FcRn receptor.

[0255] As used herein, "ADCC" or "antibody-dependent cell-mediated cytotoxicity" refers to a cell-mediated reaction in which non-specific cytotoxic cells expressing FcγR recognize antibodies bound to target cells, subsequently causing lysis of the target cells. ADCC is associated with binding to FcγRIIIa; increased binding to FcγRIIIa leads to increased ADCC activity. As described herein, many embodiments of the present invention completely eliminate ADCC activity.

[0256] The term “ADCP” or “antibody-dependent cell-mediated phagocytosis” used in this article refers to a cell-mediated reaction in which non-specific cytotoxic cells expressing FcγR recognize antibodies bound to target cells, subsequently inducing phagocytosis of the target cells.

[0257] As used herein, “antigen-binding domain” or “ABD” refers to a set of six complementary determining regions (CDRs) that, when present as part of a polypeptide sequence, specifically bind to the target antigens described herein. Therefore, the “PD-1 antigen-binding domain” binds to the human PD-1 antigen as outlined herein. As is known in the art, these CDRs are typically in the form of a first group of variable heavy chain CDRs (vhCDR or V…). H CDR) and the second group of variable light chain CDRs (vlCDR or V) L The CDRs exist in the form of variable heavy and light domains, each containing three CDRs of the heavy chain: vhCDR1, vhCDR2, and vhCDR3, and three CDRs of the light chain: vlCDR1, vlCDR2, and vlCDR3. The CDRs reside in the variable heavy and light domains, respectively, and together form the Fv region. Therefore, in some cases, the six CDRs of the antigen-binding domain are contributed by both the variable heavy and light chains. In the "Fab" form, the group consisting of six CDRs is contributed by two different polypeptide sequences: the variable heavy domain (vhCDR1, vhCDR2, and vhCDR3) and the light domain (vlCDR1, vlCDR2, and vlCDR3). H ;Contains vhCDR1, vhCDR2, and vhCDR3) and variable light structural domains (vl or V LThe VH and VL domains are typically covalently linked to a single polypeptide sequence (starting from the N-terminus) using the linkers outlined herein. This polypeptide sequence can be either vh-linker-vl or vl-linker-vh (inclusive of vlCDR1, vlCDR2, and vlCDR3).

[0258] The hypervariable region typically contains approximately amino acid residues 24-34 (LCDR1; "L" indicates light chain), 50-56 (LCDR2), and 89-97 (LCDR3) in the light chain variable region, and approximately amino acid residues 31-35B (HCDR1; "H" indicates heavy chain), 50-65 (HCDR2), and 95-102 (HCDR3) in the heavy chain variable region; Kabat et al., SEQUENCES OF PROTEINS OF IMMUNOLOGICAL INTEREST, 5th edition, Public Health Service, National Institutes. ofHealth, Bethesda, Md. (1991) and / or those residues forming the hypervariable ring in the light chain variable region (e.g., residues 26-32 (LCDR1), 50-52 (LCDR2), and 91-96 (LCDR3)) and those residues forming the hypervariable ring in the heavy chain variable region (e.g., residues 26-32 (HCDR1), 53-55 (HCDR2), and 96-101 (HCDR3)); Chothia and Lesk (1987) J. Mol. Biol. 196: 901-917. Specific CDRs of the present invention are described below.

[0259] As those skilled in the art will understand, the exact numbering and location of CDRs may vary in different numbering systems. However, it should be understood that the disclosure of variable heavy chain and / or variable light chain sequences includes the disclosure of the associated (inherent) CDRs. Thus, the disclosure of each heavy chain variable region is a disclosure of vhCDRs (e.g., vhCDR1, vhCDR2, and vhCDR3), and the disclosure of each light chain variable region is a disclosure of vlCDRs (e.g., vlCDR1, vlCDR2, and vlCDR3).

[0260] A useful comparison of CDR numbers is shown below, see Lafranc et al., Dev. Comp. Immunol. 27(1):55-77(2003):

[0261] Table 1

[0262]

[0263] Throughout this specification, when referring to residues in the variable domains (approximately residues 1-107 of the light chain variable region and residues 1-113 of the heavy chain variable region), the Kabat numbering system is generally used, and when referring to the Fc region, the EU numbering system is generally used (e.g., Kabat et al., ibid. (1991)).

[0264] This invention provides a large number of different CDR sets. In this case, a “complete CDR set” comprises three variable light chain CDRs and three variable heavy chain CDRs, such as vlCDR1, vlCDR2, vlCDR3, vhCDR1, vhCDR2, and vhCDR3. They may be part of a larger variable light domain or a variable heavy domain, respectively. Furthermore, as outlined more fully herein, when using heavy and light chains (e.g., when using Fab), the variable heavy and light domains may be on separate polypeptide chains, or, in the case of scFv sequences, on a single polypeptide chain.

[0265] CDRs facilitate antigen binding, or more specifically, the formation of epitope binding sites for antibodies. An "epitaph" is a determinant that interacts with a specific antigen-binding site (called a complementary site) in the variable region of an antibody molecule. Epitopes are molecular groups, such as amino acid or sugar side chains, and typically possess specific structural and charge characteristics. A single antigen may have multiple epitopes.

[0266] Epitopes can contain amino acid residues that directly participate in binding (also known as the immunodominant component of the epitope) and other amino acid residues that do not directly participate in binding, such as amino acid residues that are effectively blocked by specific antigen-binding peptides; in other words, the amino acid residues are within the coverage of specific antigen-binding peptides.

[0267] Epitopes can be conformational epitopes or linear epitopes. Conformational epitopes are formed by amino acids arranged side-by-side in a linear polypeptide chain. Linear epitopes are formed by adjacent amino acid residues in a polypeptide chain. The difference between conformational and non-conformational epitopes is that, in the presence of denaturing solvents, conformational epitopes lose their binding to the former but retain their binding to the latter.

[0268] An epitope typically contains at least three, and more commonly at least five, or eight to ten, amino acids with a unique spatial configuration. Antibodies that recognize the same epitope can be verified by a simple immunoassay that shows the ability of one antibody to block the binding of another antibody to the target antigen, such as "clustering." As outlined below, this invention includes not only the antigen-binding domains and antibodies listed herein, but also those epitopes that compete for binding with the listed antigen-binding domains (or, in the case of NC[PD-1]Fv, the anti-PD-1CDR of this invention does not compete for binding with the listed antibodies to the same epitope).

[0269] With regard to the antibodies and their components used in this invention, the carboxyl-terminal portion of each chain defines the constant region primarily responsible for effector function. Kabat et al. collected numerous primary sequences of the variable regions of the heavy and light chains. Based on the degree of sequence conservation, they categorized individual primary sequences into CDRs and frames and created a list of them (see SEQUENCES OFIMMUNOLOGICAL INTEREST, 5th edition, NIH publication, No. 91-3242, EA Kabat et al., the full text of which is incorporated herein by reference).

[0270] Within the IgG subclass of immunoglobulins, several immunoglobulin domains exist in the heavy chain. The term "immunoglobulin (Ig) domain" as used herein refers to immunoglobulin regions with different tertiary structures. This invention focuses on the heavy chain domain, which includes a heavy chain constant (CH) domain and a hinge domain. In the context of IgG antibodies, each IgG isotype has three CH regions. Therefore, in the context of IgG antibodies, the "CH" domains are as follows: "CH1" refers to positions 118-220 according to the EU index as described by Kabat; "CH2" refers to positions 237-340 according to the EU index as described by Kabat; and "CH3" refers to positions 341-447 according to the EU index as described by Kabat. As shown herein and described below, pI variants may reside in one or more CH regions and the hinge region, as described below.

[0271] Another Ig domain of the heavy chain is the hinge region. As used herein, “hinge,” “hinge region,” “antibody hinge region,” or “immunoglobulin hinge region” refers to a flexible polypeptide containing amino acids between the first and second constant domains of the antibody. Structurally, the IgG CH1 domain terminates at EU position 215, while the IgG CH2 domain begins at residue EU position 231. Therefore, for IgG, the antibody hinge is defined herein as encompassing positions 216 (E216 in IgG1) through 230 (P230 in IgG1), where the numbering is based on the EU index as described in Kabat. In some embodiments, such as in the context of an Fc region, a lower hinge is included, where the starting position of the “lower hinge” typically refers to position 226. As noted herein, pI variants can also be prepared within the hinge region.

[0272] As those skilled in the art will recognize, the exact numbering and location of the heavy chain constant region may differ in different numbering systems. A useful comparison of heavy chain constant numbering obtained from the EU and Kabat is shown below, see Edelman et al. (1969, Proc Natl Acad Sci USA 63:78-85) and Kabat et al. (1991, Sequences of Proteins of Immunological Interest, 5th ed., United States Public Health Service, National Institutes of Health, Bethesda), both of which are incorporated herein by reference in full.

[0273] Table 2

[0274] <![CDATA[ EU number ]]> <![CDATA[ Kabat number ]]> CH1 118-215 114-223 hinge 216-230 226-243 CH2 231-340 244-360 CH3 341-447 361-478

[0275] Light chains typically contain two structural domains: the variable light structural domain (containing the light chain CDR and forming the Fv region together with the variable heavy structural domain) and the light chain constant region (usually called CL or Cκ).

[0276] Another additional target substitution region outlined in this article is the Fc region.

[0277] Therefore, this invention provides different protein domains. As described herein and known in the art, the heterodimeric proteins of this invention contain different domains within the heavy and light chains, and these domains may overlap. These domains include, but are not limited to: Fc domain, CH1 domain, CH2 domain, CH3 domain, hinge domain, heavy chain constant domain (CH1-hinge-Fc domain or CH1-hinge-CH2-CH3), variable heavy domain, variable light domain, light chain constant domain, Fab domain, and scFv domain.

[0278] As used herein, “Fc”, “Fc region”, or “Fc domain” refers to a polypeptide containing the constant region of an antibody, excluding the first constant region, the immunoglobulin domain (e.g., CH1), and in some cases, a portion of the hinge. For IgG, the Fc domain contains the immunoglobulin domains CH2 and CH3 (Cγ2 and Cγ3) and a hinge region between CH1 (Cγ1) and CH2 (Cγ2). Although the boundaries of the Fc region may vary, the Fc region of the human IgG heavy chain is generally defined as containing residues C226 or P230 at its carboxyl terminus, where the numbering is based on the EU index as described in Kabat. Thus, in the context of IgG antibodies, the “CH” domain is defined as follows: “CH1” refers to positions 118-215 according to the EU index as described in Kabat; “hinge” refers to positions 216-230 according to the EU index as described in Kabat; “CH2” refers to positions 231-340 according to the EU index as described in Kabat; and “CH3” refers to positions 341-447 according to the EU index as described in Kabat. Therefore, the “Fc domain” includes the -CH2-CH3 domain and optionally includes the hinge domain (hinge-CH2-CH3).

[0279] Therefore, the “Fc domain” includes the -CH2-CH3 domain and optionally includes a hinge domain, which in many cases serves as a domain connector. In the embodiments herein, when scFv is attached to the Fc domain, the C-terminus of the scFv construct is attached to all or part of the hinge of the Fc domain; for example, it is typically attached to the sequence EPKS (SEQ ID NO:9), which is the hinge opening. Similarly, when an IL-15 component (whether it is an IL-15 complex, an IL-15 domain, or an IL-15Rα domain) is attached to the Fc domain, it is typically attached to all or part of the Fc domain (as a domain connector); for example, it is typically attached to the sequence EPKS (SEQ ID NO:9), which is the hinge opening.

[0280] This invention relates to Fc domains, typically based on IgG classes, which have several subclasses, including but not limited to IgG1, IgG2, IgG3, and IgG4. Generally, IgG1, IgG2, and IgG4 are more commonly used than IgG3. It should be noted that IgG1 has different allotypes, exhibiting polymorphism at 356 (D or E) and 358 (L or M). The sequences shown herein use the 356E / 358M allotype, but other allotypes are also included herein. That is, any sequence containing an IgG1 Fc domain included herein may have 356D / 358L, replacing the 356E / 358M allotype.

[0281] Furthermore, many sequences in this paper have at least one cysteine ​​residue at position 220 substituted with a serine residue; for most sequences shown in this paper, this substitution is typically on the “scFv monomer” or “IL-15 complex” side, but it can also be on the “Fab monomer” side or both sides simultaneously to reduce disulfide bond formation. The sequences in this paper specifically include one or both of these substituted cysteine ​​residues (C220S).

[0282] The term "heavy chain" or "heavy chain structural domain" as used in this paper refers to the VH-CH1-hinge-CH2-CH3 structural domain from the N-terminus to the C-terminus (where CH2-CH3 includes the Fc structural domain). A heavy chain comprises a variable heavy structural domain and a constant structural domain, including the CH1-optional hinge-Fc structural domain (containing CH2-CH3). A light chain comprises a variable light structural domain and a light chain constant structural domain (VL-CL).

[0283] As used herein, “modification” refers to an alteration in the polypeptide sequence involving the substitution, insertion, and / or deletion of amino acids or their chemical linkage to a protein. For example, a modification could be a change in the structure of a carbohydrate or PEG linked to a protein. As used herein, “amino acid modification” refers to the substitution, insertion, and / or deletion of amino acids in a polypeptide sequence. For clarity, unless otherwise stated, amino acid modification always refers to amino acids encoded by DNA, such as the 20 amino acids that have a codon in DNA and RNA.

[0284] As used herein, “amino acid substitution” or “replacement” refers to replacing an amino acid at a specific position in the parental polypeptide sequence with a different amino acid. Specifically, in some embodiments, substitution involves an amino acid at a specific position that is not naturally present or not naturally present in an organism or any organism. For example, substitution of E272Y or 272Y refers to a variant polypeptide, in this case, the Fc variant, in which glutamic acid at position 272 is replaced by tyrosine. For clarity, a protein engineered to alter the nucleic acid coding sequence but not the starting amino acid (e.g., exchanging CGG (encoding arginine) for CGA (still encoding arginine) to increase expression levels in a host organism) is not “amino acid substitution”; that is, although a new gene encoding the same protein is created, it is not an amino acid substitution if the protein has the same amino acid at a specific position at its starting position.

[0285] As used in this article, "amino acid insertion" or "insertion" refers to the addition of an amino acid sequence at a specific position in the parental polypeptide sequence. For example, -233E or 233E indicates the insertion of glutamic acid after position 233 and before position 234. Furthermore, -233ADE or A233ADE indicates the insertion of AlaAspGlu after position 233 and before position 234.

[0286] As used herein, “amino acid deletion” or “deletion” refers to the removal of an amino acid residue or sequence at a specific position in the parent polypeptide sequence. For example, E233-, E233#, E233(), E233_, or E233del indicate the deletion of glutamic acid at position 233. Furthermore, EDA233- or EDA233# indicates the deletion of the sequence GluAspAla starting at position 233.

[0287] As used herein, "variant protein," "protein variant," or "variant" refers to a protein that differs from a parent protein due to at least one amino acid modification. A protein variant can refer to the protein itself, a composition comprising the protein, or the amino sequence encoding the protein. Preferably, the protein variant has at least one amino acid modification compared to the parent protein, for example, about 1 to about 70 amino acid modifications compared to the parent protein, more preferably about 1 to about 5 amino acid modifications. As described below, in some embodiments, the parent polypeptide (e.g., the Fc parent polypeptide) is a human wild-type sequence, such as from the Fc region of IgG1, IgG2, IgG3, or IgG4. The protein variant sequence described herein preferably has at least about 80% identity with the parent protein sequence, and most preferably at least about 90% identity, more preferably at least about 95-98-99% identity. A variant protein can refer to the variant protein itself, a composition comprising the protein variant, or the DNA sequence encoding the protein variant.

[0288] Therefore, as used herein, "Fc variant" or "variant Fc" refers to a protein containing amino acid modifications in its Fc domain. The Fc variants of this invention are defined according to the amino acid modifications that constitute them. Thus, for example, N434S or 434S is an Fc variant having a substituted serine residue at position 434 relative to the parental Fc polypeptide, where numbering is based on the EU index. Similarly, M428L / N434S defines an Fc variant with substituted M428L and N434S relative to the parental Fc polypeptide. The characteristics of the WT amino acids may be uncertain; in such cases, the above variants are referred to as 428L / 434S. Note that the order of substituents provided herein is arbitrary; that is, for example, M428L / N434S is the same Fc variant as N434S / M428L, and so on. For all antibody-related positions described in this invention, unless otherwise stated, amino acid position numbering is based on the EU index. The EU index, or the EU index or EU numbering scheme as described by Kabat, refers to the numbering of EU antibodies (Edelman et al., 1969, ProcNatl Acad Sci USA 63:78-85, the full text of which is incorporated herein by reference). Modifications can be additions, deletions, or substitutions. Substitutions may include naturally occurring amino acids and, in some cases, synthetic amino acids. Examples include U.S. Patent Nos. 6,586,207; WO 98 / 48032; WO 03 / 073238; US2004 / 0214988A1; WO 05 / 35727A2; WO05 / 74524A2; JWChin et al. (2002), Journal of the American Chemical Society 124:9026-9027; JWChin and PGSchultz (2002), ChemBioChem 11:1135-1137; JWChin et al. (2002), PICAS United States of America 99:11020-11024; and L.Wang and PGSchultz (2002), Chem. 1-10, all of which are incorporated herein by reference in their entirety.

[0289] As used herein, “protein” refers to at least two covalently linked amino acids, including proteins, polypeptides, oligopeptides, and peptides. As used herein, “residue” refers to a position in a protein and its associated amino acid characteristics. For example, asparagine 297 (also known as Asn297 or N297) is a residue at position 297 in human antibody IgG1.

[0290] As used in this article, "Fab" or "Fab region" refers to a polypeptide containing the VH, CH1, VL, and CL immunoglobulin domains. Fab can refer to the region alone, or to a full-length antibody, antibody fragment, or Fab fusion protein.

[0291] As used herein, “Fv” or “Fv fragment” or “Fv region” refers to a polypeptide containing the VL and VH domains of a single antibody and forming an ABD. As those skilled in the art will understand, they typically consist of two chains, or can be combined (often used with the linkers described herein) to form scFv.

[0292] As used in this paper, "single-chain Fv" or "scFv" refers to a variable heavy structure domain that is covalently connected to a variable light structure domain using an scFv connector as described in this paper to form an scFv or scFv structure domain. The orientation of the scFv structure domain can be from the N end to the C end (vh-connector-vl or vl-connector-vh).

[0293] As used herein, "IgG subclass modification" or "isotype modification" refers to amino acid modification, which converts an amino acid of one IgG isotype into the corresponding amino acid of a different matching IgG isotype. For example, since IgG1 contains tyrosine at EU position 296 and IgG2 contains phenylalanine at the same position, the F296Y substitution in IgG2 is considered an IgG subclass modification.

[0294] Therefore, as used herein, "isotype" refers to any subclass of immunoglobulin defined by the chemical and antigenic properties of its constant region. It should be understood that therapeutic antibodies may also contain hybrids of isotypes and / or subclasses.

[0295] As used herein, "variable region" refers to a region of an immunoglobulin containing one or more Ig domains encoded by any Vκ, Vλ, and / or VH genes that respectively constitute the κ, λ, and heavy chain immunoglobulin loci. Regarding the variable heavy and light domains of this invention, the amino-terminal portion of each heavy and light chain antibody includes a variable region of about 100 to 110 or more amino acids primarily used for antigen recognition, which is commonly referred to in the art and herein as the "Fv domain" or "Fv region". Within the variable region, each V domain of the heavy and light chains aggregates three loops to form an antigen-binding site. Each loop is called a complementarity-determining region (hereinafter referred to as "CDR"), where the amino acid sequence variation is most pronounced. "Variable" refers to the fact that certain segments of the variable region differ significantly in the sequence between antibodies. The variation within the variable region is not uniformly distributed. Conversely, the V region consists of relatively unchanging segments of framework regions (FRs) consisting of 15 to 30 amino acids, which are separated by much more variable shorter regions called "hypervariates," each containing 9 to 15 or more amino acids.

[0296] Each VH and VL consists of three CDRs and four framework regions (FRs), arranged in the following order from the amino terminus to the carboxyl terminus: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4. Therefore, the variable heavy domain comprises VHFR1-VHCDR1-VHFR2-VHCDR2-VHFR3-VHCDR3-VHFR4, and the variable light domain comprises VLFR1-VLCDR1-VLFR2-VLCDR2-VLFR3-VLCDR3-VLFR4.

[0297] As used in this article, “non-naturally occurring modification” refers to an amino acid modification that is not of the same type. For example, since no IgG contains serine at position 434, the 434S substituent in IgG1, IgG2, IgG3, or IgG4 (or their hybrids) is considered a non-natural modification.

[0298] As used in this article, “amino acid” and “amino acid properties” refer to one of the 20 naturally occurring amino acids encoded by DNA and RNA.

[0299] As used in this article, "effective function" refers to a biochemical event caused by the interaction between the antibody's Fc region and the Fc receptor or ligand. Effector functions include, but are not limited to, ADCC, ADCP, and CDC.

[0300] As used in this article, “Fcγ receptor,” “FcγR,” or “FcgammaR” refers to any member of the protein family that binds to the Fc region of IgG antibodies and is encoded by the FcγR gene. In humans, this protein family includes, but is not limited to, FcγRI (CD64), which includes isoforms FcγRIa, FcγRIb, and FcγRIc; FcγRII (CD32), which includes isoforms FcγRIIa (including alloforms H131 and R131), FcγRIIb (including FcγRIIb-1 and FcγRIIb-2), and FcγRIIc; and FcγRIII (CD16), which includes isoforms FcγRIIIa (including alloforms V158 and F158) and FcγRIIIb (including alloforms FcγRIIb-NA1 and FcγRIIb-NA2) (Jefferis et al., 2002, Immunol Lett 82:57-65, the full text of which is incorporated herein by reference) and any undiscovered human FcγR or FcγR isoforms or alloforms.

[0301] As used herein, “FcRn” or “novice Fc receptor” refers to a protein that binds to the Fc region of an IgG antibody and is at least partially encoded by the FcRn gene. As is known in the art, a functional FcRn protein comprises two polypeptides, commonly referred to as a heavy chain and a 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. Various FcRn variants can be used to increase binding to the FcRn receptor and, in some cases, to prolong the serum half-life. Generally, unless otherwise stated, the Fc receptor of the present invention retains its binding to the FcRn receptor (and, as described below, may include amino acid variants to increase binding to the FcRn receptor).

[0302] As used herein, “parental polypeptide” refers to the initial polypeptide that is subsequently modified to generate a variant. Parental polypeptides can be naturally occurring polypeptides, variants of natural polypeptides, or engineered forms. Parental polypeptides can refer to the polypeptide itself, the composition constituting the parental polypeptide, or the amino acid sequence encoding the parental polypeptide.

[0303] The term "heavy chain constant region" as used in this article refers to the CH1-hinge-CH2-CH3 portion of the antibody.

[0304] As used herein, “Fc fusion protein” or “immunoadhesin” refers to an Fc region that is typically linked (optionally via a linker portion as described herein) to different proteins such as IL-15 and / or IL-15R as described herein. In some instances, two Fc fusion proteins may form a homodimeric Fc fusion protein or a heterodimeric Fc fusion protein, the latter being preferred. In some cases, one monomer of the heterodimeric Fc fusion protein contains only an Fc domain (e.g., an empty Fc domain), and the other monomer is an Fc fusion protein containing a variant Fc domain and a protein domain, such as the IL-15 complex. As outlined herein, one monomer of the heterodimeric protein is an Fc fusion protein containing an IL-15 complex, and the other monomer is a conventional heavy chain (with an associated light chain).

[0305] As used in this article, "position" refers to a location within a protein sequence. Positions can be numbered sequentially or according to a pre-defined format (such as the EU index used for antibody numbering).

[0306] In the context of the heterodimeric antibody monomers of the present invention disclosed herein, "chain-like" refers to two DNA strands that are similar to "matched," with the heterodimerizing variant binding to each monomer, thereby maintaining the ability to "match" to form a heterodimer. For example, if certain pI variants are engineered to monomer A (e.g., enhancing pI), then "charge-pair" spatial variants that can be used do not interfere with the pI variants, for example, charge-pairing variants of pI are placed on the same "chain" or "monomer" to retain both functions. Similarly, for paired "skewed" variants, as outlined more fully below, those skilled in the art will consider the pI when determining which chain or monomer to bind to in a paired manner, such that using pI-skewed variants also maximizes separation.

[0307] As used in this article, “targeting cells” refers to cells that express target antigens, in this case, PD-1 and / or IL-15 receptors.

[0308] As used in this article, "wild-type or WT" refers to an amino acid or nucleotide sequence found in nature, including allelic variations. WT proteins have an unmodified amino acid or nucleotide sequence.

[0309] The bispecific heterodimeric proteins of this invention are typically isolated proteins or recombinant proteins. When used to describe the various polypeptides disclosed herein, “isolated” means a polypeptide that has been identified, isolated, and / or recovered from cells or cell cultures expressing it. Typically, isolated polypeptides are prepared by at least one purification step. “Isolated protein” means a protein that is substantially free of other proteins with different binding specificities. “Recombinant” means a protein generated in a foreign host cell using recombinant nucleic acid technology.

[0310] The "percentage (%) amino acid residue identity" relative to a protein sequence refers to the percentage of amino acid residues in a candidate sequence that are identical to those in a specific (parental) sequence. This percentage of sequence identity is achieved by aligning sequences and introducing differences (if necessary), without considering any conserved substitutions as part of the sequence identity. Alignments performed to determine the percentage of amino acid sequence identity can be performed in various ways within the scope of the art, for example, using publicly available computer software such as BLAST, BLAST-2, ALIGN, or Megalign (DNASTAR) software. Those skilled in the art to which this invention pertains can determine suitable parameters for measuring alignments, including any algorithms required to achieve maximum alignment across the full length of the sequences being compared. One particular program is the ALIGN-2 program, which is outlined in U.S. Patent Publication No. 20160244525, paragraphs

[0279] to

[0280] , which is incorporated herein by reference.

[0311] The degree of identity between the amino acid sequence of the present invention (“Inventive Sequence”) and the parental amino acid sequence is calculated by dividing the number of exact matches in the two sequence alignments by the length of the “Inventive Sequence” or the length of the parental sequence (whichever is shorter). The result is expressed as a percentage of identity.

[0312] In some embodiments, two or more amino acid sequences are at least 50%, 60%, 70%, 80%, or 90% identical. In some embodiments, two or more amino acid sequences are at least 95%, 97%, 98%, 99%, or even 100% identical.

[0313] "Specific binding," or "specific to," or "specific to" a particular antigen or epitope (in this case, human PD-1), refers to binding that is distinctly different from nonspecific interactions. Specific binding can be measured, for example, by determining the binding of a molecule against a control molecule, which is typically a similarly structured molecule that does not have binding activity. For instance, specific binding can be determined by competing with a control molecule that resembles the target.

[0314] Specific binding to a particular antigen or epitope can manifest as, for example, a KD of at least about 10 for ABD to the antigen or epitope. -4 M, at least about 10 -5 M, at least about 10 -6 M, at least about 10 -7 M, at least about 10 -8 M, at least about 10 -9 M or at least about 10 -10 M, at least about 10 -11 M, at least about 10 -12M or greater, where KD refers to the dissociation rate of a specific ABD-antigen interaction. Typically, the KD of an ABD that specifically binds to an antigen is 20, 50, 100, 500, 1000, 5,000, 10,000 or more times higher than the KD of a control molecule to an antigen molecule or epitope.

[0315] Furthermore, specific binding to a particular antigen or epitope can manifest as an increase in KA or Ka between the antibody and the antigen or epitope by 20, 50, 100, 500, 1000, 5000, or 10000 times compared to the control epitope, where KA or Ka refers to the binding rate of the specific antibody-antigen interaction. Binding affinity is typically measured using surface plasmon resonance (SPR)-based assays (e.g., Biacore) or biolayer interferometry (BLI)-based assays (e.g., Octet).

[0316] III. Introduction

[0317] Some aspects of the present invention provide a targeting heterodimeric fusion protein that can bind to the checkpoint inhibitor PD-1 antigen and can complex with common γ chain (γc; CD132) and / or IL-2 receptor β chain (IL-2Rβ; CD122) (due to the presence of the IL-15 complex). Typically, the heterodimeric fusion protein of the present invention has three functional components: an IL-15 / IL-15Rα (sushi) component, generally referred to herein as the “IL-15 complex” or “IL-15 / Rα complex”; an anti-PD-1 component, which serves as the “targeting” portion, bridging the fusion protein to PD-1-expressing cells; and an Fc component, wherein each component may take different forms, and the various components may be combined with other components of any conformation.

[0318] However, the anti-PD-1 component of the heterodimeric fusion protein of the present invention does not compete with known anti-PD-1 antibodies such as nivolumab or pembrolizumab for binding. That is, by incorporating an anti-PD-1 (αPD-1) antigen-binding domain (ABD) that does not compete with approved αPD-1 antibodies (“NC[PD-1]”), the fusion protein of the present invention allows for highly efficient binding to anti-PD-1 antibody therapies. Specifically, by incorporating an anti-PD-1 ABD (“NC-αPD-1ABD”) that does not compete with approved therapies for binding, the non-competitive ABD can be used to target the fusion protein to tumors, but still allows for additional anti-PD-1 antibody therapy, as both can non-competitively bind to PD-1. Furthermore, in some embodiments, the NC-αPD-1ABD can also completely block the PD-1:PD-L1 interaction (in embodiments based on mAb A variant), partially block the PD-1:PD-L1 interaction (mAbC variant), or completely not block any interaction (mAbB variant).

[0319] As will be understood by those skilled in the art and as outlined herein, many different forms of non-competitive constructs of different targeting heterodimer fusion proteins, “NC-αPD-1X IL-15 / Rα”, are shown in Figure 28.

[0320] Furthermore, some aspects of the present invention rely on a comparison of this embodiment with a “non-targeting IL-15 / Rα-Fc fusion protein” (which does not contain an antigen-binding domain for human PD-1), as shown in Figure 13.

[0321] Furthermore, the non-targeted or targeted heterodimer fusion protein of the present invention can be combined with other antibodies against checkpoint receptors (including anti-PD-1, anti-TIM-3, anti-LAG-3, anti-TIGIT, etc.).

[0322] Therefore, the present invention provides a number of different functional components that can be assembled in a variety of different ways to generate the heterodimeric fusion protein of the present invention. As described above, the fusion protein includes an IL-15 complex comprising an IL-15 domain and an IL-15 receptor component.

[0323] Some approaches offer IL-15 / Rα-Fc fusion proteins that can induce the proliferation of regulatory T cells (Tregs) with reduced or minimal immunosuppressive activity. In one aspect, this heterodimeric fusion protein promotes the proliferation of effector memory T cells (Tregs). EM (Amplification.) In one embodiment, the heterodimeric fusion protein increased the interaction between Treg and T... EM (Treg / T EMThe ratio of ) to ). In some instances, treatment with any of the IL-15 / Rα-Fc fusion proteins outlined herein can convert Tregs from a suppressor Treg cell type to non-suppressor activated effector CD4 T cells. In one embodiment, the effector Treg FOXP3 高 CD45RA – CD4 + Differentiate into effector CD4 T cells FOXP3 低 CD45RA – CD4 + In some embodiments, activated effector CD4 T cells exhibit reduced CCR4 expression.

[0324] Furthermore, this article provides an IL-15 / Rα-Fc fusion protein that reverses the inhibition of T cell proliferation by TGFβ. In the tumor environment, TGFβ is expressed by malignant cells and immune cells, including Tregs. TGFβ also inhibits T cell proliferation, thereby suppressing the anti-tumor immune response. Treatment with the IL-15 / Rα-Fc fusion protein of this invention prevents TGFβ from inhibiting T cell proliferation. In one embodiment, administration of the IL-15 / Rα-Fc fusion protein in the tumor environment counteracts the activity of TGFβ on T cells, thereby promoting T cell proliferation and the anti-tumor immune response.

[0325] Some aspects of this invention relate to heterodimeric Fc fusion proteins comprising IL-15 and IL-15 receptor α (IL-15Rα) protein domains in different orientations. The Fc domains may be derived from IgG Fc domains, such as IgG1, IgG2, IgG3, or IgG4 Fc domains, with the IgG1 Fc domain being particularly suitable for use in this invention.

[0326] Therefore, some aspects of the present invention provide different antibody domains. As described herein and known in the art, the heterodimeric proteins of the present invention comprise different domains, which may also overlap. These domains include, but are not limited to, the Fc domain, CH1 domain, CH2 domain, CH3 domain, hinge domain, and heavy chain constant domain (CH1-hinge-Fc domain or CH1-hinge-CH2-CH3).

[0327] In some constructs and sequences of Fc domain proteins outlined herein, the C-terminus of an IL-15 or IL-15Rα protein fragment is attached to the N-terminus of a domain linker, the C-terminus of which is attached to the N-terminus of a constant Fc domain (N-IL-15 or IL-15Rα protein fragment-linker-Fc domain-C), but may be switched (N-Fc domain-linker-IL-15 or IL-15Rα protein fragment-C). In other constructs and sequences outlined herein, the C-terminus of a first protein fragment is attached to the N-terminus of a second protein fragment, optionally via a domain linker, and the C-terminus of the second protein fragment is attached to the N-terminus of a constant Fc domain, optionally via a domain linker. In other constructs and sequences outlined herein, a constant Fc domain not attached to either the first or second protein fragment is provided. Heterodimeric Fc fusion proteins may comprise two or more of the exemplary monomeric Fc domain proteins described herein. In yet another construct, the N-terminus of the first protein fragment is linked to the C-terminus of the second protein fragment, optionally via a domain linker, and the N-terminus of the second protein fragment is linked to the C-terminus of the constant Fc domain, optionally via a domain linker.

[0328] A.IL-15 complex

[0329] As shown in the figure, the IL-15 complex can take various forms. As mentioned above, the stability of the IL-15 protein itself is lower than that of the IL-15Rα protein after complexation. As is known in the art, the IL-15Rα protein contains a "sushi domain," which is the shortest region in the receptor that retains IL-15 binding activity. Therefore, although heterodimeric fusion proteins containing the entire IL-15Rα protein can be prepared, the preferred embodiments described herein include complexes using only the sushi domain, the sequence of which is shown in the figure.

[0330] Therefore, the IL-15 complex typically contains the IL-15 protein and the sushi domain of IL-15Rα (unless otherwise specified, "IL-15Rα", "IL-15Rα(sushi)", and "sushi" are used interchangeably throughout the text). When complexed together, it is named "IL-15 / Rα" with a forward slash (" / ") to indicate the presence of both the IL-15 and IL-15Rα domains.

[0331] 1. IL-15 domain

[0332] As those skilled in the art will understand, the IL-15 domain can be a wild-type human sequence or can be engineered to include variants, particularly the efficacy variants described below.

[0333] In some embodiments, the human IL-15 protein has the amino acid sequence described in NCBI Reference Sequence No. NP_000576.1 or SEQ ID NO:1, which is a precursor sequence. In some cases, the coding sequence for human IL-15 is described in NCBI Reference Sequence No. NM_000585. The exemplary IL-15 protein of the Fc fusion protein heterodimer protein outlined herein may have the amino acid sequence of SEQ ID NO:2 (mature IL-15), which corresponds to amino acids 49-162 of SEQ ID NO:1. In some embodiments, the IL-15 protein has at least 90%, for example, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or higher sequence identity with SEQ ID NO:2.

[0334] In some embodiments, the IL-15 domain has been engineered to include amino acid substitutions.

[0335] Therefore, in some embodiments, the IL-15 protein is a variant of the amino acid sequence of SEQ ID NO:2, and one or more amino acid substitutions are selected from the group consisting of: C42S, L45C, Q48C, V49C, L52C, E53C, E87C, and E89C. The IL-15 protein of the heterodimeric fusion protein may have 1, 2, 3, 4, 5, 6, 7, 8, or 9 amino acid substitutions.

[0336] a. IL-15 efficacy variant

[0337] Furthermore, in some embodiments, the human IL-15 protein is engineered to reduce potency, as broadly described in PCT / US2019 / 028107, the entire disclosure of which is incorporated herein by reference. That is, as described herein, the reduced potency of IL-15 in the heterodimeric fusion proteins of the present invention (optionally having and not having the Xtend-Fc substitutions described herein, such as M428L / N434S) can enhance the pharmacodynamic and pharmacokinetic characteristics in subjects receiving such protein. Similarly, as shown in Example 7 of PCT / US2019 / 028107, potency-reduced IL-15 / Rα-Fc variants such as XENP22821 can amplify lymphocyte counts for a longer time compared to the wild-type IL-15 / Rα-Fc fusion protein described herein, such as XENP20818. It is worth noting that XENP23343 is an Xtend analogue of XENP22821, which further prolongs the duration of lymphocyte expansion compared to XENP22821. Furthermore, the decrease in IL-15 potency improves the therapeutic index (i.e., lower toxicity at higher doses).

[0338] “Non-targeting molecules” such as XmAb24306, as shown in Example 8 of PCT / US2019 / 028107, and IL-15 / Rα-Fc fusion proteins (such as those incorporating NC-αPD-1X IL-15 / Rα-Fc fusion proteins in this paper) can overcome Treg inhibition of induced effector T cell proliferation.

[0339] Similarly, as shown in Example 4 below, within a certain dose range, the NC-αPD-1X IL-15 / Rα-Fc fusion protein can promote leukocyte proliferation and worsen xenogeneic GVHD. Notably, combination therapy with the NC-αPD-1X IL-15 / Rα-Fc fusion protein and anti-PD-1 antibodies exhibits a synergistic effect (e.g., a synergistic effect), especially at low doses.

[0340] Therefore, the present invention provides a number of suitable IL-15 amino acid variants that impart reduced potency and enhanced pharmacokinetic characteristics, including but not limited to variant IL-15 proteins comprising one or more amino acid substitutions selected from the group consisting of: N1D; N4D; D8N; D30N; D61N; E64Q; N65D; Q108E; N1D / N4D / D8N; N1D / N4D / N65D; N1D / D30N; N1D / D61N; N1D / D61N / E64Q / Q108E; N1D / E64Q; N1D / N6 5D; N1D / Q108E; N4D; N4D / D30N; N4D / D61N; N4D / D61N / N65D; N4D / D61N / E64Q / Q108E; N4D / E64Q; N4D / N65D; D8N / D61N; D8N / E64Q; D30N / E64Q; D30N / N65D; D30N / E64Q / N65D; D30N / Q180E; D61N / E64Q / N65D; E64Q; E64Q / N65D; E64Q / Q108E; and N65D / Q108E. In some embodiments, the variant IL-15 protein comprises the amino acid sequence SEQ ID NO:319.

[0341] In some embodiments, the one or more amino acid substitutions may be isochronous substitutions of IL-15:IL-2β and IL-15:common γ chain interfaces.

[0342] In some embodiments, the human IL-15 protein, such as the Fc fusion protein, has the same amino acid sequence as SEQ ID NO:2. In some cases, the human IL-15 protein, such as the human mature IL-15 protein, does not contain amino acid substitutions.

[0343] In some embodiments, the mature human IL-15 variant protein has one or more amino acid mutations (e.g., substitution, insertion, and / or deletion). In some instances, the mutation introduces a cysteine ​​residue that can form a disulfide bond with the human IL-15 receptor α (IL-15Rα) protein.

[0344] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and an amino acid variant D30N. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and a D30N substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least one D30N substitution.

[0345] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and the amino acid substitution N4D / N65D. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and the N4D / N65D substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least the N4D / N65D substitution.

[0346] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and an amino acid variant N1D. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and an N1D substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least one N1D substitution.

[0347] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and an amino acid variant N4D. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and an N4D substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least one N4D substitution.

[0348] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and an amino acid variant E64Q. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and an E64Q substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least one E64Q substitution.

[0349] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and an amino acid variant N65D. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and an N65D substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least one N65D substitution.

[0350] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and an amino acid substitution N1D / D30N. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and an N1D / D30N substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least an N1D / D30N substitution.

[0351] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and the amino acid substitution N4D / D30N. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and the N4D / D30N substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least the N4D / D30N substitution.

[0352] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and the amino acid substitution D30N / E64Q. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and the D30N / E64Q substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least the D30N / E64Q substitution.

[0353] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and the amino acid substitution D30N / N65D. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and the D30N / N65D substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least the D30N / N65D substitution.

[0354] In some embodiments, the present invention provides a protein comprising a human IL-15 variant and the amino acid substitution D30N / E64Q / N65D. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and the D30N / E64Q / N65D substitution. In some embodiments, the protein comprises the amino acid sequence of SEQ ID NO:2 and at least the D30N / E64Q / N65D substitution.

[0355] b. IL-15 glycosylation variants

[0356] Furthermore, the IL-15 domain may have amino acid variants used alone or in combination with potency variants to reduce glycosylation and / or glycosylation heterogeneity. Suitable glycosylation variants include, but are not limited to, N71Q, N79Q, N112Q, S114del, and S114A compared to SEQ ID NO:2, which may be used alone or in combination with each other. Particularly useful sets of glycosylation variants in many embodiments of the invention are N71Q / N79Q, N71Q / N79Q / N112Q, N71Q / N79Q / S114del, and N71Q / N79Q / S114A. In some embodiments, the variant IL-15 protein comprises the amino acid sequence SEQ ID NO:319.

[0357] c. Combination of potency and glycosylation variants

[0358] In many embodiments, a combination of potency variants and glycosylation variants is used in the IL-15 domain to form part of the IL-15 complex with the sushi domain (typically a truncated wild-type sequence of SEQ ID NO:20).

[0359] Therefore, in some embodiments, the IL-15 domain includes the potency variants D30N / N65D and the glycosylation variants N71Q / N79Q / N112Q.

[0360] In some embodiments, the IL-15 domain includes the potency variant D30N / N65D and the glycosylation variant N71Q / N79Q.

[0361] In some embodiments, the IL-15 domain includes the potency variants D30N / E64Q / N65D and the glycosylation variants N71Q / N79Q / N112Q.

[0362] In some embodiments, the IL-15 domain includes the potency variants D30N / E64Q / N65D and the glycosylation variants N71Q / N79Q.

[0363] In some embodiments, the variant IL-15 protein comprises the amino acid sequence of SEQ ID NO:319.

[0364] 2. IL-15 / Rα domain

[0365] In addition to the IL-15 domain, the heterodimeric fusion protein of the present invention optionally includes amino acid variants as outlined above, and includes the "sushi" domain.

[0366] In some embodiments, the human IL-15 receptor α (IL-15Rα) protein has the amino acid sequence as described in NCBI Reference Sequence Number NP_002180.1 or SEQ ID NO:3. In some cases, the coding sequence of human IL-15Rα is as described in NCBI Reference Sequence Number NM_002189.3. The exemplary IL-15Rα protein of the Fc fusion protein heterodimer protein summarized herein may comprise or consist of the sushi domain of SEQ ID NO:3 (e.g., amino acids 31-95 of SEQ ID NO:3) or, in other words, the amino acid sequence of SEQ ID NO:20. That is, certain embodiments utilize a truncated version of the extracellular domain of the receptor.

[0367] In some embodiments, the IL-15Rα protein has the amino acid sequence of SEQ ID NO:20 and one or more amino acid insertions selected from the group consisting of: D96, P97, A98, D96 / P97, D96 / C97, D96 / P97 / A98, D96 / P97 / C98, and D96 / C97 / A98, wherein the amino acid position is determined relative to the full-length human IL-15Rα protein or SEQ ID NO:3. For example, one or more amino acids such as D (e.g., Asp), P (e.g., Pro), A (e.g., Ala), DP (e.g., Asp-Pro), DC (e.g., Asp-Cys), DPA (e.g., Asp-Pro-Ala), DPC (e.g., Asp-Pro-Cys), or DCA (e.g., Asp-Cys-Ala) may be added to the C-terminus of the IL-15Rα protein of SEQ ID NO:20. In some embodiments, the IL-15Rα protein has the amino acid sequence of SEQ ID NO:20 and one or more amino acid substitutions selected from the group consisting of K34C, A37C, G38C, S40C, and L42C, wherein the amino acid position is determined relative to SEQ ID NO:20. The IL-15Rα (sushi) protein of SEQ ID NO:20 may have 1, 2, 3, 4, 5, 6, 7, 8, or more amino acid mutations (e.g., substitutions, insertions, and / or deletions). When the amino acid modification is performed on the sushi domain, the variant sushi domain must retain biological activity, such as binding to IL-15.

[0368] In some embodiments, the IL-15Rα protein has the amino acid sequence of SEQ ID NO:4. In some embodiments, the IL-15Rα protein has the amino acid sequence of SEQ ID NO:4 and amino acid insertions selected from the group consisting of: D96, P97, A98, D96 / P97, D96 / C97, D96 / P97 / A98, D96 / P97 / C98, and D96 / C97 / A98, wherein the amino acid positions are determined relative to the full-length human IL-15Rα protein or SEQ ID NO:3. In some embodiments, the IL-15Rα protein has the amino acid sequence of SEQ ID NO:4 and one or more amino acid substitutions selected from the group consisting of: K34C, A37C, G38C, S40C, and L42C, wherein the amino acid positions are determined relative to SEQ ID NO:4. When amino acid modifications are performed on the IL-15Rα protein, the variant IL-15Rα protein must retain biological activity, such as binding to IL-15.

[0369] 3. Different forms of IL-15 complex

[0370] Therefore, the IL-15 complex typically contains the sushi domains of the IL-15 protein and IL-15Rα. When combined, it is named "IL-15 / Rα" with a forward slash (" / ") to indicate the presence of both the IL-15 and IL-15Rα domains.

[0371] As shown in Figure 28, the IL-15 / Rα complex can exist in two different forms; it can be either a non-covalent or a covalent complex. Figure 28B As shown in 28D, 28F, and 28G, the IL-15 protein and IL-15Rα (sushi) are not covalently linked, but rather self-assembled through regular ligand-ligand interactions. As described more fully in this paper, the IL-15 domain or the sushi domain can be covalently linked to the Fc domain (typically using an optional domain linker).

[0372] Or, such as Figure 28A As shown in 28C, 28E, and 28H, the IL-15 / Rα complex is a covalently linked compound using domain connectors. In each case, the orientation of IL-15 and IL-15Rα from the N-terminus to the C-terminus can be switched: that is, in Figure 28A In this invention, the IL-15 domain is located at the N-terminus, wherein the IL-15Rα domain is connected to the Fc domain using a connector. Similarly, in Figure 28C The invention also includes a case where the IL-15 domain is located at the N-terminus, wherein the IL-15Rα domain is connected to the Fc domain using a connector. Figure 28EThe invention also includes a case where the IL-15 domain is at the C-end of the Fc domain (connected using a domain connector), and the IL-15Rα domain is connected to the IL-15 domain using a connector. Finally, Figure 28H This also includes the case where the IL-15 domain is at the N-terminus of the IL-15Rα domain.

[0373] B. Non-competitive anti-PD-1 antigen-binding domain

[0374] The anti-PD-1 component of the present invention (e.g., an anti-PD-1 antigen-binding domain (ABD)) is typically a set of six CDRs comprising a variable heavy domain and a variable light domain, forming an Fv domain that can bind to human PD-1 (whose sequence is shown below). Figure 2 (as shown), but does not compete with commercially available anti-PD-1 antibodies such as pembrolizumab and nivolumab. This allows the heterodimeric fusion protein of the present invention to have excellent tumor targeting, thereby allowing the IL-15 / Rα complex to exert local effects, and also allowing for combination therapy with effective anti-PD-1 antibodies without competing for the same PD-1 epitopes.

[0375] The NC-αPD-1 antigen-binding domain of the present invention can have two universal forms, such as Figure 28A And the scFv domain shown in 28B, or the Fab domain present on two different peptides (such as...) Figure 28C And 28D (the monovalent binding of PD-1) and Figure 28E (Those shown as 28F, 28G, and 28H (divalent binding of PD-1), as described more fully below.

[0376] Compared to the initial mAbC H1_L1 sequence, many useful amino acid variants were identified.

[0377] 1. Oxidized variants

[0378] This invention relates to the fact that the particularly useful non-competitive anti-PD-1 Fv "[NC]mAb C" containing tryptophan (Xencor designation W112; Kabat designation W100) in VH-CDR3 is readily oxidized (data not shown) and subsequently loses PD-1 binding. That is, amino acid substitutions that eliminate the potential for oxidation at this position can reduce the affinity of ABD for human PD-1, as shown in Figure 151. Therefore, as described above, other variants conferring higher binding affinity were bound to the W100F variant to investigate whether this binding affinity could be restored, as described below.

[0379] 2. Affinity variants

[0380] like Figures 65A-65IAs shown, there are many variables in the variable restructured domain compared to SEQ ID. NO:5 can be used to increase the affinity of W100F oxidation variants. Suitable variants include, but are not limited to: F34L, S52aG, R97E, R97Y, R97W, L98R, S100aT, R97A, V99T, V99L, S100aA, L98Q, R97Q, V99F, V99L, S100aN, V99I, P100bS, G96H, L98V, V99A, V99Q, G96V, R97K, L98S, L98F, R97T, L98K, L98S, V99I, R97L, G96A, R97A, V99S, R97S, V99Y, R97H, and L98R (Kabat designations), which can be used individually or in combination.

[0381] In addition, compared to SEQ ID NO:168, there are many suitable variants of the variable light structural domain that can be used alone or in combination, including but not limited to: N27dH, N27dS, K30Y, S93T and Y94W (Kabat number).

[0382] 3. Combinations of oxidation and affinity variants

[0383] Therefore, the present invention provides NC-mAbC ABD with variants, which contain the oxidized variant W100F in the VH domain and combine with one or more affinity variants to restore suitable affinity binding.

[0384] In some embodiments, the VH amino acid substitution (compared to the parental VH domain SEQ ID NO:5) is selected from the group consisting of: F32L / W100F; F32L / S52aG / W100F; F32L / S52aG / R97F / W100F; F32L / S52aG / R97Y / W100F; F32L / S52aG / R97E / W100F; F32L / S52aG / R97Q / W100F; F32L / S52aG / R97L / W100F; F32L / S52aG / R97V / W100F; F32L / S52aG / R97D / W100F; F32L / S 52aG / R97H / W100F); F32L / S52aG / R97A / W100F; F32L / S52aG / R97W / W100F; F32L / S52aG / R97T / W100F; 97A / W100F; V99T / W100F; V99L / W100F / S100aA; L98Q / V99L / W100F; R97Q / W100F; L98Q / W100F; V99F / W100F; V99L / W100F; W100 F / S100aN; V99I / W100F / P100bS; G96H / L98V / W100F; V99A / W100F; V99Q / W100F / S100aT; G96V / R97A / V99A / W100F / P100bS; R9 7Q / L98Q / W100F / S100aA; R97K / W100F / S100aA; W100F; W100F / S100aT; L98S / SW100F; L98F / W100F; R97W / L98H / W100F; W100F / S100aA; R97T / L98K / W100F; L98S / V99I / W100F; R97L / V99I / W100F; G96A / W100F; R97S / L98V / V99L / W100F; V99S / W100F; L98Q / W100F / S100aT; R97S / V99Y / W100F; V99Y / W100F; L98R / W100F; W100F / P100bS; R97H / L98Q / W100F; H1.224L98R / V99L / W100F.

[0385] In some embodiments, the VL of NCPD-1 mAbC ABD has an amino acid substitution N27dH (compared to the parental VL domain SEQ ID NO:168), and the VH amino acid substitution (compared to the parental VH domain SEQ ID NO:168) NO:5) Select the group consisting of the following items: F32L / W100F; F32L / S52aG / W100F; F32L / S52aG / R97F / W100F; F32L / S52aG / R97Y / W100F; F32L / S52aG / R97E / W100F; F32L / S52aG / R97Q / W100F; F32L / S52aG / R97L / W100F; F32L / S52aG / R97V / W100F; F32L / S52aG / R97D / W100F; F3 2L / S52aG / R97H / W100F); F32L / S52aG / R97A / W100F; F32L / S52aG / R97W / W100F; F32L / S52aG / R97T / W100F; L98R / W100F / S1 00aT; R97A / W100F; V99T / W100F; V99L / W100F / S100aA; L98Q / V99L / W100F; R97Q / W100F; L98Q / W100F; V99F / W100F; V99L / W1 00F; W100F / S100aN; V99I / W100F / P100bS; G96H / L98V / W100F; V99A / W100F; V99Q / W100F / S100aT; G96V / R97A / V99A / W100F / P100bS; R97Q / L98Q / W100F / S100aA; R97K / W100F / S100aA; W100F; W100F / S100aT; L98S / SW100F; L98F / W100F; R97W / L98H / W100F; W100F / S100aA; R97T / L98K / W100F; L98S / V99I / W100F; R97L / V99I / W100F; G96A / W100F; R97S / L98V / V99L / W100F; V 99S / W100F; L98Q / W100F / S100aT; R97S / V99Y / W100F; V99Y / W100F; L98R / W100F; W100F / P100bS; R97H / L98Q / W100F; H1.224 L98R / V99L / W100F.

[0386] In some embodiments, the VL of NCPD-1 mAbC ABD has an amino acid substitution N27dH / K30Y / S93T (compared to the parental VL domain SEQ ID NO:168), and the VH amino acid substitution (compared to the parental VH domain SEQ ID NO:168). NO:5) Select the group consisting of the following items: F32L / W100F; F32L / S52aG / W100F; F32L / S52aG / R97F / W100F; F32L / S52aG / R97Y / W100F; F32L / S52aG / R97E / W100F; F32L / S52aG / R97Q / W100F; F32L / S52aG / R97L / W100F; F32L / S52aG / R97V / W100F; F32L / S52aG / R97D / W100F; F3 2L / S52aG / R97H / W100F); F32L / S52aG / R97A / W100F; F32L / S52aG / R97W / W100F; F32L / S52aG / R97T / W100F; L98R / W100F / S1 00aT; R97A / W100F; V99T / W100F; V99L / W100F / S100aA; L98Q / V99L / W100F; R97Q / W100F; L98Q / W100F; V99F / W100F; V99L / W1 00F; W100F / S100aN; V99I / W100F / P100bS; G96H / L98V / W100F; V99A / W100F; V99Q / W100F / S100aT; G96V / R97A / V99A / W100F / P100bS; R97Q / L98Q / W100F / S100aA; R97K / W100F / S100aA; W100F; W100F / S100aT; L98S / SW100F; L98F / W100F; R97W / L98H / W100F; W100F / S100aA; R97T / L98K / W100F; L98S / V99I / W100F; R97L / V99I / W100F; G96A / W100F; R97S / L98V / V99L / W100F; V 99S / W100F; L98Q / W100F / S100aT; R97S / V99Y / W100F; V99Y / W100F; L98R / W100F; W100F / P100bS; R97H / L98Q / W100F; H1.224 L98R / V99L / W100F.

[0387] In some embodiments, the NC-PD-1 mAbC ABD has a VH selected from those shown in FIG. 43, including but not limited to: H1.176, H1.177, H1.178, H1.179, H1.180, H1.181, H1.182, H1.183, H1.184, H1.185, H1.186, H1.187, H1.188, H1.189, H1.190, H1.191, H1.192, H1.193, H1.194, H1.195, H1.196, H1.197, H1 .198, H1.199, H1.200, H1.201, H1.202, H1.203, H1.204, H1.205, H1.206, H1.207, H1.208, H1.209, H1.210, H1.2 11. H1.212, H1.213, H1.214, H1.215, H1.216, H1.217, H1.218, H1.219, H1.220, H1.221, H1.222, H1.223, H1.224.

[0388] In some embodiments, the NC-PD-1 mAbC ABD has a VL selected from those shown in FIG43, including but not limited to: L1.1, L1.3, L1.45, L1.117, L1.129, L1.135, L1.136 and L1.140.

[0389] In some embodiments, the NC-PD-1 mAbC ABD has the H1.176_L1.140 sequence.

[0390] In some embodiments, the NC-PD-1 mAbC ABD has an H1.176_L1.1 sequence.

[0391] It should be noted that any VH domain can be combined with any VL domain, including the specific combinations described herein that are of use.

[0392] As shown in this document, in addition to using the "Fab" form of anti-PD-1ABD, anti-PD-1ABD can also be used in the form of scFv, where the vh and vl structural domains are connected by scFv connectors, which can optionally be charged scFv connectors. As those skilled in the art will understand, scFv can be assembled from the N-end to the C-end as N-vh-scFv connector-vl-C or N-vl-scFv connector-vh-C, where the C-end of the scFv structural domain is typically linked to the hinge-CH2-CH3 Fc structural domain. Suitable Fv (including CDR groups and heavy chain / variable light structural domains) can be used in the scFv or Fab form shown in Figure 43. As will be further understood by those skilled in the art, all or part of the hinge (which may also be a wild-type hinge derived from IgG1, IgG2 or IgG4 or variants thereof, such as the IgG4 S241P or S228P hinge variants having a substituted proline at position 228 relative to the parental IgG4 hinge polypeptide (wherein, S228P is obtained according to the EU index and S241P is a Kabat number)) may be used as a domain linker between the scFv and CH2-CH3 domains, or other domain linkers as shown in the figure may be used.

[0393] C.Fc structural domain

[0394] In addition to the IL-15 complex and the NC-αPD-1 targeting Fv domain, this invention further provides a heterodimeric Fc domain as a component. As shown in Figure 28, these heterodimeric Fc domains are used to integrate the IL-15 / Rα and anti-PD-1 targeting domains into a single construct, which typically comprises two polypeptide chains (e.g., Figure 28A As shown, one monomer contains an IL-15 / Rα complex, and the other monomer contains anti-PD-1scFv, three polypeptide chains (e.g., Figure 28C As shown, one monomer contains an IL-15 / Rα complex, the second monomer contains a heavy chain and the third monomer is a light chain, etc.

[0395] The Fc domain components of this invention, as described herein, typically comprise the skewed variants and / or optional pI variants and / or ablation variants outlined herein. For example, see the disclosure of WO2017 / 218707 entitled "IV Heterodimeric Antibody," including portions IV.A, IV.B, IV.C, IV.D, IV.E, IV.F, IV.G, IV.H, and IV.I, the entire contents of which are expressly incorporated herein by reference. The Fc domains outlined herein, comprising "skewed variants," "pI variants," "ablation variants," and FcRn variants, are particularly suitable for use in the heterodimeric proteins of this invention. Particularly useful Fc domains are... Figure 8 Those shown.

[0396] The Fc domain may be derived from the IgG Fc domain, such as the IgG1, IgG2, IgG3, or IgG4 Fc domain, with the IgG1 Fc domain being particularly suitable for use in this invention. The Fc domains of IL-15 / IL-15Rα Fc fusion protein monomers and checkpoint antibody fragments that can be used in the heterodimeric Fc protein of this invention are described below.

[0397] The carboxyl terminus of each chain defines the constant region primarily responsible for effector function. Kabat et al. collected numerous primary sequences of the variable regions of the heavy and light chains. Based on the degree of sequence conservation, they divided individual primary sequences into CDRs and frames and created a list of them (see SEQUENCES OF IMMUNOLOGICAL INTEREST, 5th edition, NIH publication, No. 91-3242, EA Kabat et al., the full text of which is incorporated herein by reference). Throughout this specification, when referring to residues in the variable domains (approximately residues 1-107 of the light chain variable region and residues 1-113 of the heavy chain variable region), the Kabat numbering system is generally used, and when referring to the Fc region, the EU numbering system is generally used (e.g., Kabat et al., ibid. (1991)).

[0398] Within the IgG subclass of immunoglobulins, several immunoglobulin domains exist in the heavy chain. The term "immunoglobulin (Ig) domain" as used herein refers to immunoglobulin regions with different tertiary structures. This invention focuses on the heavy chain domain, which includes a heavy chain constant (CH) domain and a hinge domain. In the context of IgG antibodies, each IgG isotype has three CH regions. Therefore, in the context of IgG antibodies, the "CH" domains are as follows: "CH1" refers to positions 118-220 according to the EU index as described by Kabat; "CH2" refers to positions 237-340 according to the EU index as described by Kabat; and "CH3" refers to positions 341-447 according to the EU index as described by Kabat. As shown herein and described below, pI variants may reside in one or more CH regions and the hinge region, as described below.

[0399] Another Ig domain of the heavy chain is the hinge region. As used herein, “hinge,” “hinge region,” “antibody hinge region,” or “immunoglobulin hinge region” refers to a flexible polypeptide containing amino acids between the first and second constant domains of the antibody. Structurally, the IgG CH1 domain terminates at EU position 220, while the IgG CH2 domain begins at residue EU position 237. Therefore, for IgG, the antibody hinge is defined herein as including positions 221 (D221 in IgG1) through 236 (G236 in IgG1), where the numbering is based on the EU index as described in Kabat. In some embodiments, such as in the context of an Fc region, a lower hinge is included, where “lower hinge” generally refers to position 226 or 230. As noted herein, pI variants can also be prepared within the hinge region.

[0400] Therefore, the present invention provides different antibody domains, such as different Fc domains. As described herein and known in the art, the heterodimeric proteins of the present invention comprise different domains, which may also overlap. These domains include, but are not limited to, Fc domains, CH1 domains, CH2 domains, CH3 domains, hinge domains, and heavy chain constant domains (CH1-hinge-Fc domains or CH1-hinge-CH2-CH3).

[0401] Therefore, the “Fc domain” includes the -CH2-CH3 domain and optionally includes a hinge domain, and may be derived from human IgG1, IgG2, IgG3, or IgG4, wherein the Fc domain is derived from IgG1. In some embodiments described herein, when a protein fragment (e.g., IL-15 or IL-15Rα) is attached to the Fc domain, the C-terminus of the IL-15 or IL-15Rα construct is attached to all or part of the hinge of the Fc domain; for example, it is typically attached to the sequence EPKS (SEQ ID NO: 9), which is the hinge opening. In other embodiments, when a protein fragment (e.g., IL-15 or IL-15Rα) is attached to the Fc domain, the C-terminus of the IL-15 or IL-15Rα construct is attached to the CH1 domain of the Fc domain.

[0402] In some constructs and sequences of Fc domain proteins outlined herein, the C-terminus of an IL-15 or IL-15Rα protein fragment is attached to the N-terminus of a domain linker, the C-terminus of which is attached to the N-terminus of a constant Fc domain (N-IL-15 or IL-15Rα protein fragment-linker-Fc domain-C), but may be switched (N-Fc domain-linker-IL-15 or IL-15Rα protein fragment-C). In other constructs and sequences outlined herein, the C-terminus of a first protein fragment is attached to the N-terminus of a second protein fragment, optionally via a domain linker, and the C-terminus of the second protein fragment is attached to the N-terminus of a constant Fc domain, optionally via a domain linker. In other constructs and sequences outlined herein, a constant Fc domain not attached to either the first or second protein fragment is provided. Heterodimeric Fc fusion proteins may comprise two or more of the exemplary monomeric Fc domain proteins described herein.

[0403] In some embodiments, the connector is a "domain connector," as outlined more fully below, used to connect any two domains described herein, some of which are... Figure 8 As shown. While any suitable linker can be used, many embodiments utilize glycine-serine polymers, including, for example, (GS)n (SEQ ID NO: 10), (GSGGS)n (SEQ ID NO: 11), (GGGGS)n (SEQ ID NO: 12), and (GGGS)n (SEQ ID NO: 13), where n is an integer of at least 1 (typically 1 to 2 to 3 to 4 to 5) and any peptide sequence that allows the two domains to recombine with sufficient length and flexibility so that each domain retains its biological function. In some cases, and as outlined herein, the focus on “chain-like” linkages is directed at domain linkers.

[0404] Therefore, in some embodiments of the present invention, a heterodimeric Fc fusion protein is provided, which relies on the self-assembly of two different heavy chain variant Fc sequences to form a heterodimeric Fc domain fusion polypeptide.

[0405] This invention relates to novel constructs of heterodimeric Fc fusion proteins that allow binding to one or more binding partners, ligands, or receptors. The heterodimeric Fc fusion proteins are based on the self-assembly of two Fc domains (e.g., two “monomers” assembling into a “dimer”) of the heavy chain of an antibody. The heterodimeric Fc fusion proteins are prepared by altering the amino acid sequence of each monomer, as detailed below. Therefore, this invention generally relates to the formation of heterodimeric Fc fusion proteins that can co-bind with binding partners, ligands, or receptors in various ways, depending on amino acid variants in different constant regions on each chain, to facilitate heterodimer formation and / or facilitate the purification of the heterodimer from homodimers.

[0406] Numerous mechanisms are available for generating the heterodimers of this invention. Furthermore, as those skilled in the art will understand, these mechanisms can be used in combination to ensure high heterodimerization. Therefore, amino acid variants used to produce heterodimers are referred to as “heterodimerizing variants.” As described below, heterodimerizing variants may include stereovariates (e.g., the “mortar” or “skew” variants and the “charge pair” variants described below) and “pI variants” that can purify homodimers from heterodimers. As generally described in WO2014 / 145806 (the entire text of which is incorporated herein by reference) and specifically discussed below under “Heterodimerization Variants”, available heterodimerization mechanisms include “kidney” (“KIH”; sometimes referred to herein as the “skew” variant (see discussion in WO2014 / 145806)), “electrostatic orientation” or “charge pair” as described in WO2014 / 145806, pI variant as described in WO2014 / 145806, and general additional Fc as outlined in WO2014 / 145806 and below.

[0407] In this invention, several basic mechanisms can be employed to simplify the purification of heterodimeric antibodies; one mechanism relies on the use of pI variants, such that each monomer has a different pI, thereby achieving isoelectric purification of AA, AB, and BB dimer proteins. Alternatively, certain forms also allow for size-based separation. As further outlined below, heterodimers can also be "skewed" to form rather than homodimers. Therefore, combinations of stereodimerizing variants with pI or charge-pair variants can be particularly useful in this invention.

[0408] Generally, embodiments particularly useful in this invention rely on a group of variants including skewed variants, used in combination with pI variants (which increase the pI difference between two monomers) to promote heterodimerization rather than homodimerization.

[0409] Furthermore, as more fully outlined below, depending on the form of the heterodimeric Fc fusion protein, the pI variant may be contained within a constant domain and / or an Fc domain of the monomer, or a domain linker may be used. That is, the present invention provides a pI variant on one or both monomers along with a charged domain linker. Additionally, other amino acid engineering for achieving alternative functions can also confer pI variations, such as Fc, FcRn, and KO variants.

[0410] In this invention, when using pI as a separation mechanism to purify heterodimeric proteins, amino acid variants can be introduced into one or two monomeric polypeptides; that is, the pI of one monomer (referred to herein as "monomer A") can be designed to be different from that of monomer B, or variations can be made between monomers A and B, increasing the pI of monomer A and decreasing the pI of monomer B. As described herein, changes in the pI of one or both monomers can be achieved by removing or adding charged residues (e.g., replacing neutral amino acids with positively or negatively charged amino acid residues, such as replacing glycine with glutamic acid), changing charged residues from positive or negative to opposite charges (e.g., replacing aspartic acid with lysine), or changing charged residues to neutral residues (e.g., charge loss; replacing lysine with serine). Many of these variants are shown in the figure.

[0411] Therefore, this embodiment of the invention provides for generating sufficient pI variation in at least one monomer to allow the heterodimer to be separated from the homodimer. As will be understood by those skilled in the art and further described below, this can be accomplished by using a constant region of the "wild-type" heavy chain and variant regions modified to increase or decrease its pI (wt A-+B or wtA--B) or by increasing one region and decreasing another (A+-B- or A-B+).

[0412] Therefore, in general, some embodiments of the present invention are composed of amino acid variants in the constant region, which are intended to alter the isoelectric point (pI) of at least one (if not two) monomers of the dimer monomer by adding an amino acid substituent (“pI variant” or “pI substitute”) to one or two monomers. As shown herein, the heterodimer can be separated from the two homodimers if the pIs of the two monomers differ by only 0.1 pH units (0.2, 0.3, 0.4, and 0.5 or greater may be used in the present invention).

[0413] As those skilled in the art will understand, to achieve good separation, the number of pI variants contained in each or both monomers will depend in part on the initial pI of the components. As is known in the art, different Fcs have different initial pIs than those used in this invention. Generally, as outlined herein, pIs are modified such that the total pIs of each monomer differ by at least about 0.1 log, and preferably by 0.2 to 0.5, as outlined herein.

[0414] As those skilled in the art will understand, to achieve good separation, the number of pI variants contained in each or both monomers will depend in part on the starting pI of the components. That is, to determine which monomers to modify or to determine the “direction” of modification (e.g., increase or decrease), the sequence of the Fc domain is calculated, and in some cases, the protein domains linked to the Fc domain are calculated, and a decision is made accordingly. As is known in the art, different Fc domains and / or protein domains have starting pIs different from the starting pIs used in this invention. Generally, as outlined herein, pIs are modified such that the total pI of each monomer differs by at least about 0.1 log, and preferably by 0.2 to 0.5, as outlined herein.

[0415] Furthermore, as understood by those skilled in the art and as outlined herein, in some embodiments, heterodimers can be separated from homodimers based on size. As shown in the figures, for example, several forms allow for the separation of heterodimers and homodimers based on size.

[0416] In cases where pI variants are used to achieve heterodimerization, the use of constant regions of the Fc domain provides a more modular approach to designing and purifying heterodimeric Fc fusion proteins. Therefore, in some embodiments, heterodimerizing variants (including skewed and purified heterodimerizing variants) must be engineered. Furthermore, in some embodiments, by introducing pI variants from different IgG isotypes, the likelihood of the pI variants generating immunogenicity is significantly reduced without inducing significant immunogenicity. Therefore, another issue that needs to be addressed is elucidating low pI constant regions with high human sequence content, such as minimizing or avoiding the presence of non-human residues at any particular position.

[0417] One potential side benefit of this pI engineering is the extension of serum half-life and increased FcRn binding. Specifically, as described in USSN 13 / 194,904 (in its entirety incorporated herein by reference), pIs that reduce antibody constant domains (including those present in both the antibody and the Fc fusion protein) can result in longer in vivo serum retention times. These pI variants with extended serum half-life also facilitate the modification of pIs for purification.

[0418] Furthermore, it should be noted that the pI variant of the heterodimerized variant brings additional benefits to the analysis and quality control process of Fc fusion proteins, as the ability to eliminate, minimize, and distinguish homodimers is important when they are present. Similarly, the ability to reliably detect the reproducibility of preparations of heterodimerized Fc fusion proteins is also important.

[0419] 1. Heterodimer variant

[0420] This invention provides heterodimeric proteins, including various forms of heterodimeric Fc fusion proteins, which utilize heterodimeric variants to achieve heterodimer formation and / or purification from homodimers. The heterodimeric fusion proteins are based on the self-assembly of two Fc domains (e.g., two “monomers” assembling into a “dimer”).

[0421] There are many suitable heterodimerization skew variant sets. These variant sets appear "pairs." That is, one pair of variant sets binds to the first monomer, while another pair of variant sets binds to the second monomer. It should be noted that these sets do not necessarily exhibit "mortar and pestle" variants, where there is a one-to-one correspondence between residues on one monomer and residues on the other; that is, these pairs form an interface between the two monomers that favors heterodimer formation and inhibits homodimer formation, resulting in a proportion of spontaneously formed heterodimers exceeding 90% under biological conditions, rather than the expected 50% (25% homodimer A / A: 50% heterodimer A / B: 25% homodimer B / B).

[0422] A list of suitable skew variants is shown in Figure 3. Particularly useful variant pairs in many embodiments include, but are not limited to: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411E / K360E / Q362E:D401K; L368D / K370S:S364K / E357L; K370S:S364K / E357Q and T366S / L368A / Y407V:T366W (optionally including the bridging disulfide T366S / L368A / Y407V / Y349C:T366W / S354C). In terms of nomenclature, the “S364K / E357Q:L368D / K370S” pair refers to one monomer having a dual variant group S364K / E357Q and another having a dual variant group L368D / K370S; as mentioned above, the “chain type” of these pairs depends on the starting pI.

[0423] 2. Stereo variants

[0424] In some embodiments, stereovariants can be added to promote the formation of heterodimers. That is, by altering the amino acids in each heavy chain, different heavy chains are more likely to combine to form a heterodimer structure rather than a homodimer with the same Fc amino acid sequence. Suitable stereovariants are shown in USSN 15 / 141,350 (the entire text of which is incorporated herein by reference). Figure 29 And as shown in this article Figure 8 middle.

[0425] One mechanism, commonly referred to in the art as a "mortar and pestle," refers to an amino acid engineering technique that optionally employs steric effects that favor heterodimer formation and discourage homodimer formation; this technique is sometimes referred to as a "mortar and pestle" as described in the following references: USSN 61 / 596,846; Ridgway et al., Protein Engineering 9(7):617(1996); Atwell et al., J. Mol. Biol. 1997 270:26; U.S. Patent No. 8,216,805, the entire contents of which are incorporated herein by reference. The figure identifies numerous "monomer A–monomer B" pairs that depend on the "mortar and pestle." Furthermore, as Merchant et al. (Nature Biotech. 16:677, 1998) describe, these "mortar and pestle" mutations can bind to disulfide bonds to favor heterodimer formation.

[0426] Another mechanism that can be used to generate heterodimers is sometimes referred to as “electrostatic reversal,” as described by Gunasekaran et al. (J. Biol. Chem. 285(25):19637, 2010), the full text of which is incorporated herein by reference. It is sometimes referred to herein as “charge pair.” In this embodiment, electrostatics are used to bias the formation toward heterodimerization. As those skilled in the art will understand, they may also affect pI, thereby affecting purification, and thus may be considered pI variants in some cases. However, since these variants are generated to promote heterodimerization and are not used as purification tools, they are classified as “stereovariants.” These include, but are not limited to, D221E / P228E / L368E paired with D221R / P228R / K409R (e.g., these are “monomer-corresponding groups”) and C220E / P228E / 368E paired with C220R / E224R / P228R / K409R.

[0427] Other monomer A and monomer B variants that can be combined with other variants may be used optionally and independently in any amount, such as the pI variants outlined herein or other stereovariates shown in Figure 37 of US 2012 / 0149876, all of which are explicitly incorporated herein by reference.

[0428] In some embodiments, the stereovariates outlined herein may optionally and independently bind to one or two monomers with any pI variant (or other variants, such as Fc variants, FcRn variants, etc.) and may optionally include or exclude the proteins described herein.

[0429] 3. pI (isoelectric point) variants of heterodimers

[0430] Generally, as those skilled in the art will recognize, pI variants fall into two main categories: those that increase the type of protein pI (basic changes) and those that decrease the type of protein pI (acidic changes). As described herein, all combinations of these variants are possible: one monomer may be wild-type, or a variant that does not exhibit a pI significantly different from the wild-type, and another may have a higher basicity or acidity. Alternatively, each monomer may be modified to increase its basicity or acidity.

[0431] Preferred combinations of pI variants are shown in Figure 30 of USSN 15 / 141,350, the entire contents of which are incorporated herein by reference. As outlined herein and as shown in the figures, these variations are shown relative to IgG1, but all isotypes and isotype hybrids can be modified in this manner. R133E and R133Q may also be used where the heavy chain constant domain originates from IgG2-4.

[0432] In one embodiment, if one of the Fc monomers contains a CH1 domain, a preferred combination of pI variants comprises a monomer containing the 208D / 295E / 384D / 418E / 421D variant (N208D / Q295E / N384D / Q418E / N421D relative to human IgG1). In some instances, a second monomer contains a positively charged domain linker comprising (GKPGS)4 (SEQ ID NO: 14). In some cases, a first monomer contains a CH1 domain, including the domain at position 208. Therefore, in constructs that do not contain a CH1 domain (e.g., for heterodimeric Fc fusion proteins that do not use a CH1 domain in one of the domains), a preferred set of negative pI variant Fc variants comprises the 295E / 384D / 418E / 421D variant (Q295E / N384D / Q418E / N421D relative to human IgG1).

[0433] In some embodiments, the mutation occurs in the hinge domain of the Fc domain, including positions 221, 222, 223, 224, 225, 233, 234, 235, and 236. It should be noted that positions 233-236 can be modified to increase effector function (and 327A) in the IgG2 backbone. Therefore, pI mutations, and particularly substitutions, can occur at one or more of positions 221-225, wherein 1, 2, 3, 4, or 5 mutations can be used in this invention. Similarly, all possible combinations are contemplated in other domains, either alone or in combination with other pI variants.

[0434] Specific substitutions that can be used to reduce pI in the hinge domain include, but are not limited to: deletion at position 221, non-natural valine or threonine at position 222, deletion at position 223, non-natural glutamic acid at position 224, deletion at position 225, deletion at position 235, or non-natural alanine at position 236. In some cases, pI substitution occurs only in the hinge domain, while in others, these substitutions are added in any combination to other pI variants in other domains.

[0435] In some embodiments, mutations may occur in the CH2 region, including positions 274, 296, 300, 309, 320, 322, 326, 327, 334, and 339. Similarly, these 10 positions can be used in all possible combinations; for example, a pI antibody may have 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 CH2 pI substitutions.

[0436] Specific substitutions that can be used to reduce the pI of the CH2 domain include, but are not limited to: non-natural glutamine or glutamic acid at position 274, non-natural phenylalanine at position 296, non-natural phenylalanine at position 300, non-natural valine at position 309, non-natural glutamic acid at position 320, non-natural glutamic acid at position 322, non-natural glutamic acid at position 326, non-natural glycine at position 327, natural glutamic acid at position 334, non-natural threonine at position 339, and all possible combinations of CH2 with other domains.

[0437] In this embodiment, the mutation may be independently and optionally selected from positions 355, 359, 362, 384, 389, 392, 397, 418, 419, 444, and 447. Specific substitutions that can be used to reduce the pI of the CH3 domain include, but are not limited to: non-natural glutamine or glutamic acid at position 355, non-natural serine at position 384, non-natural asparagine or glutamic acid at position 392, non-natural methionine at position 397, non-natural glutamic acid at position 419, non-natural glutamic acid at position 359, non-natural glutamic acid at position 362, non-natural glutamic acid at position 389, non-natural glutamic acid at position 418, non-natural glutamic acid at position 444, and deletion or non-natural aspartic acid at position 447. Exemplary examples of pI variants are provided in the figures (including...). Figure 5 )middle.

[0438] Furthermore, in some cases, the domain junctions between the IL-15 and IL-15Rα domains may be charged, and the scFv junctions (if present in a particular form) may also be charged.

[0439] 4. Isomorphic variants

[0440] Furthermore, many embodiments of the present invention rely on “introducing” pI amino acids from one IgG isoform to another at specific positions, thereby reducing or eliminating the possibility of introducing unwanted immunogenicity into the variant. Some of these are shown in Figure 21 of U.S. Patent Application No. 2014 / 0370013, which is incorporated herein by reference. That is, IgG1 is a universal isoform for therapeutic antibodies for a variety of reasons, including high-effects function. However, the pI of the constant region of the IgG1 heavy chain is higher than that of IgG2 (8.10 vs 7.31). By introducing IgG2 residues at specific positions in the IgG1 backbone, the resulting monomer has a lower (or higher) pI and exhibits a longer serum half-life. For example, IgG1 has glycine at position 137 (pI 5.97), and IgG2 has glutamate (pI 3.22); introducing glutamate will affect the pI of the resulting protein. As described below, many amino acid substitutions are typically required to significantly affect the pI of variant Fc fusion proteins. However, it should be noted that, as described below, even changes in IgG2 molecules can prolong serum half-life.

[0441] In other embodiments, non-isotype amino acid changes occur to reduce the overall charge state of the resulting protein (e.g., by changing a higher-charge pI amino acid to a lower-charge pI amino acid), or to allow structural adjustments to improve stability, etc., as detailed below.

[0442] Furthermore, by modifying the pI of the heavy and light chain constant domains, significant changes in each monomer of the heterodimer can be observed. As described herein, two monomers with a pI difference of at least 0.5 can be separated by ion exchange chromatography, isoelectric focusing, or other methods sensitive to the isoelectric point.

[0443] 5. Calculate pI

[0444] The pI of each monomer depends on the pI of the variable heavy chain constant domain and the pI of the total monomers (including the variable heavy chain constant domain and the fusion mating body). Therefore, in some embodiments, the variation of pI is calculated based on the variable heavy chain constant domain, using the graph in Figure 19 of U.S. Patent Application No. 2014 / 0370013. As described herein, the monomer to be modified typically depends on the inherent pI of each monomer.

[0445] 6. It can also provide better FcRn in vivo binding pI variants.

[0446] In cases where pI variants reduce the pI of the monomer, they may have the additional benefit of increasing serum retention in vivo.

[0447] Although still under investigation, the Fc region is believed to have a longer half-life in vivo because binding to FcRn at pH 6 in the endosome induces Fc chelation (Ghetie and Ward, 1997 Immunol Today 18(12):592-598, the full text of which is incorporated herein by reference). The endosome chamber then circulates Fc to the cell surface. Once the chamber opens to the extracellular space, a higher pH of approximately 7.4 induces the release of Fc back into the bloodstream. In mice, Dall'Acqua et al. found that Fc mutants with enhanced FcRn binding at pH 6 and pH 7.4 actually had lower serum concentrations and the same half-life compared to wild-type Fc (Dall'Acqua et al., 2002, J. Immunol. 169:5171-5180, the full text of which is incorporated herein by reference). At pH 7.4, the increased affinity of Fc for FcRn prevents the release of Fc back into the bloodstream. Therefore, Fc mutations that prolong the in vivo half-life of Fc would ideally increase FcRn binding at lower pH levels while still allowing Fc release at higher pH levels. The amino acid histidine alters its charge state within a pH range of 6.0 to 7.4. Therefore, the presence of His residues at key positions in the Fc / FcRn complex is not surprising.

[0448] 7. Additional Fc variants with additional functions

[0449] In addition to pI amino acid variants, many useful Fc amino acid modifications are performed for various reasons, including but not limited to altering the binding to one or more FcγR receptors and altering the binding to FcRn receptors.

[0450] Therefore, the proteins of the present invention may include amino acid modifications, including the heterodimerization variants outlined herein, which include pI variants and stereo variants. Each group of variants may be independently and optionally included in or excluded from any particular heterodimer protein.

[0451] 8. FcγR variant

[0452] Therefore, many useful Fc substitutions can be used to alter binding to one or more FcγR receptors. Substitutions that result in both increased and decreased binding can be used. For example, increased binding to FcγRIIIa is known to lead to elevated ADCC (antibody-dependent cell-mediated cytotoxicity; a cell-mediated reaction in which nonspecific cytotoxic cells expressing FcγR recognize bound antibodies on target cells and subsequently cause target cell lysis). Similarly, in some cases, decreased binding to FcγRIIb (inhibitory receptor) may also be beneficial. Amino acid substitutions that can be used in this invention include USSN 11 / 124,620 (particularly...) Figure 41 Those listed in 11 / 174,287, 11 / 396,495, and 11 / 538,406, all of which are expressly incorporated herein by reference in their entirety, and especially the variants disclosed herein. Specific variants available include, but are not limited to: 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D, 332E / 330L, 243A, 243L, 264A, 264V, and 299T.

[0453] Furthermore, amino acid substitutions that enhance the affinity of FcγRIIc can also be included in the Fc domain variants outlined herein. Substitutions described, for example, in USSN 11 / 124,620 and 14 / 578,305 can be used.

[0454] In addition, there are additional Fc substitutions that can be used to increase binding to the FcRn receptor and prolong serum half-life, specifically as disclosed in USSN 12 / 341,769 (the entire contents of which are incorporated herein by reference), including but not limited to: 434S, 434A, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L and 259I / 308F / 428L.

[0455] 9. Ablation variants

[0456] Similarly, another class of functional variants are “FcγR ablation variants” or “Fc knockout (FcKO or KO) variants.” In these embodiments, for certain therapeutic applications, it is desirable to reduce or eliminate the normal binding of the Fc domain to one or more Fcγ receptors (e.g., FcγR1, FcγRIIa, FcγRIIb, FcγRIIIa, etc.) to avoid additional mechanisms of action. This is particularly evident in, for example, bispecific immunomodulatory antibodies specifically used in many embodiments to eliminate FcγRIIIa binding, which are designed to eliminate or significantly reduce ADCC activity such that one of the Fc domains contains one or more Fcγ receptor ablation variants. These ablation variants are shown in Figure 31 of USSN 15 / 141,350, all of which are incorporated herein by reference in their entirety and may be independently and optionally included or excluded, wherein preferred aspects utilize ablation variants selected from the group consisting of: G236R / L328R, E233P / L234V / L235A / G236del / S239K, E233P / L234V / L235A / G236del / S267K, E233P / L234V / L235A / G236del / S239K / A327G, E233P / L234V / L235A / G236del / S267K / A327G, and E233P / L234V / L235A / G236del (according to the EU index). It should be noted that the ablation variants described in this article eliminate FcγR binding, but generally do not eliminate FcRn binding.

[0457] Exemplary embodiments of the pI variant are provided in the figures (including...) Figure 5 )middle.

[0458] 10. Combinations of heterodimers and Fc variants

[0459] As will be understood by those skilled in the art, all the listed heterodimer variants (including skewed and / or pI variants) may optionally and independently be combined in any manner, as long as they maintain their “chain” or “monomer-spaced” configuration. Furthermore, all these variants may be combined into any heterodimer form.

[0460] Regarding pI variants, while particularly useful examples are shown in the figure, other combinations can be formed by following the basic rule of altering the pI difference between the two monomers to facilitate purification.

[0461] In addition, as generally outlined in this article, any of the heterodimerization variants (skewed and pI variants) can also independently and optionally bind to the Fc ablation variant, Fc variant, and FcRn variant.

[0462] In addition, the Fc domain of the monomer may contain a set of amino acid substituents, including C220S / S267K / L368D / K370S or C220S / S267K / S364K / E357Q.

[0463] In addition, heterodimeric Fc fusion proteins may contain skewed variants (e.g., such as USSN15 / 141,350). Figure 1A The amino acid substitutions shown in ~1C (the entire contents of which are incorporated herein by reference) are particularly useful, among which the skewed variants are selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411E / K360E / Q362E:D401K; L368D / K370S:S364K / E357L; K370S:S364K / E357Q; T366S / L368A / Y407V:T366W and T366S / L368A / Y407V / Y349C:T366W / S354C, optionally including ablation variants, optionally including charged domain connectors, and heavy chains including pI variants.

[0464] In some embodiments, the Fc domain comprises amino acid substitutions selected from the group consisting of: 236R, 239D, 239E, 243L, M252Y, V259I, 267D, 267E, 298A, V308F, 328F, 328R, 330L, 332D, 332E, M428L, N434A, N434S, 236R / 328R, 239D / 332E, M428L, 236R / 328F, V259I / V308F The Fc domains are: 267E / 328F, M428L / N434S, Y436I / M428L, Y436V / M428L, Y436I / N434S, Y436V / N434S, 239D / 332E / 330L, M252Y / S254T / T256E, V259I / V308F / M428L, E233P / L234V / L235A / G236del / S267K, G236R / L328R, and PVA / S267K. In some cases, the Fc domain contains an amino acid substitution for 239D / 332E. In other cases, the Fc domain contains an amino acid substitution for G236R / L328R or PVA / S267K.

[0465] In one embodiment, a specific combination of skew and pI variants that can be used in this invention is T366S / L368A / Y407V:T366W (optionally including the bridging disulfide T366S / L368A / Y407V / Y349C:T366W / S354C), where one monomer comprises Q295E / N384D / Q418E / N481D, and the other is a positively charged structural domain connector. As those skilled in the art will recognize, the "mortar and pestle" variant does not alter the pI and can therefore be used in either monomer.

[0466] In one embodiment, a specific combination of the skew and pI variants that can be used in the present invention is L368D / K370S:S364K / E357Q, wherein one of the monomers comprises Q295E / N384D / Q418E / N421D.

[0467] Available Fc dimer variant sets (including skewed and pI variants) are provided for Figure 4 A through 4E. Other pI variants are provided in Figure 5 Available ablation variants are provided in Figure 6. Useful examples of the non-cytokine component of the IL-15 / Rαx anti-PD1 ABD heterodimer fusion protein of the present invention are provided in... Figures 7A-7E and Figure 8 A through 8F.

[0468] D. Structural domain connector

[0469] The three components of this invention (anti-PD-1 Fv, IL-15 / Rα complex, and the heterodimerized Fc domain of this invention) are optionally linked together using domain linkers. While direct covalent linkages can occur (e.g., linking the C-terminus of the IL-15 complex to the N-terminus of the CH2 domain of the Fc domain), linkers that provide flexibility and sometimes play a role are typically used.

[0470] In some embodiments, the IL-15 protein is attached to the N-terminus of the Fc domain, and the IL-15Rα protein is attached to the N-terminus of the IL-15 protein. In other embodiments, the IL-15Rα protein is attached to the N-terminus of the Fc domain and is non-covalently attached to the IL-15 protein. In still other embodiments, the IL-15Rα protein is attached to the C-terminus of the Fc domain and is non-covalently attached to the IL-15 protein.

[0471] In some embodiments, the IL-15 and IL-15Rα proteins are linked together via domain linkers (e.g., in the form of “scIL-15 / Rα”). Optionally, these proteins are not linked via linkers but utilize natural self-assembly or disulfide bonds, as outlined herein. In other embodiments, the IL-15 and IL-15Rα proteins are non-covalently linked. In some embodiments, the IL-15 protein is linked to the Fc domain via a linker. In some embodiments, the IL-15 protein is directly linked to the Fc domain (e.g., without a linker). In a particular embodiment, the IL-15 protein is linked to the Fc domain via a hinge region or a fragment thereof. In other embodiments, the IL-15Rα protein is linked to the Fc domain via a linker. In other embodiments, the IL-15Rα protein is directly linked to the Fc domain (e.g., without a linker). In a particular embodiment, the IL-15Rα protein is linked to the Fc domain via a hinge region or a fragment thereof. Optionally, the linker is not used to link the IL-15 or IL-15Rα protein to the Fc domain.

[0472] In some cases, the PD-1ABD is covalently connected to the N-terminus of the Fc structural domain via a structural domain connector. In some embodiments, the PD-1ABD is directly connected to the Fc structural domain (e.g., without a connector). In a particular embodiment, the PD-1ABD is connected to the Fc structural domain via a hinge region or a segment thereof.

[0473] In some embodiments, the linker is a “domain linker” used to connect any two domains outlined herein. The linker peptide may primarily comprise the following amino acid residues: Gly, Ser, Ala, or Thr. The linker peptide should be long enough to link two molecules to give them the correct conformation to each other, thereby maintaining the desired activity. In one embodiment, the linker is about 1 to 50 amino acids long, preferably about 1 to 30 amino acids. 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 may be used. Useful linkers include: glycine-serine polymers, including, for example, (GS)n (SEQ ID NO:15), (GSGGS)n (SEQ ID NO:16), (GGGGS)n (SEQ ID NO:17), and (GGGS)n (SEQ ID NO:18), where n is an integer of at least 1 (typically 3 to 4); glycine-alanine polymers; alanine-serine polymers; and other flexible linkers. Alternatively, various non-protein polymers (including but not limited to polyethylene glycol (PEG), polypropylene glycol, polyoxyethylene, or copolymers of polyethylene glycol and polypropylene glycol) can be used as connectors.

[0474] In some embodiments, the domain connector includes hinge regions of all or part of IgG1, IgG2, and IgG4, the former being used in many embodiments. Several hinged domain connectors are shown in... Figure 8 As understood by those skilled in the art, Figure 8 The structural domain connectors can also be combined.

[0475] In some embodiments, the domain connector is an scFv connector. Generally, the scFv connector is flexible and long enough to allow the VH and VL domains to connect correctly. In some cases, the scFv connector may be charged, approximately as follows: Figure 8 As outlined in the discussion below, its pI depends on the form and composition of the heterodimeric protein.

[0476] In some embodiments, the scFv connector is a charged scFv connector, with many of these connectors shown in Figure 7 of WO2017 / 218707. Therefore, the present invention further provides charged scFv connectors to facilitate pI separation between the first and second monomers (e.g., IL-15 / IL-15Rα monomers and PD-1ABD monomers). That is, by incorporating a positively or negatively charged scFv connector (or both, in the case of using an scFv scaffold on different monomers and / or using a charged connector and a connector that connects the IL-15 and sushi domains or the IL-15 / Rα component to the Fc domain), the monomer containing the charged connector changes the pI without further altering the Fc domain. These charged connectors can be substituted for any scFv containing a standard connector. Similarly, as those skilled in the art will understand, the charged scFv connector is used on the correct “chain” or monomer depending on the desired pI change. For example, as described herein, in order to prepare the heterodimeric fusion protein of the present invention, the original pI of the Fv region of each desired domain is calculated, and one of them is selected to prepare scFv, and a positively or negatively charged linker is selected according to the pI.

[0477] Charged structural domain connectors can also be used to increase the pI separation of the monomers of the present invention, and therefore those included in FIG10 can be used in any embodiment using connectors herein.

[0478] Other linker sequences may include any sequence of any length of CL / CH1 domain, but not all residues of the CL / CH1 domain; for example, the first 5-12 amino acid residues of the CL / CH1 domain. Linkers may be derived from immunoglobulin light chains, such as Cκ or Cλ. Linkers may be derived from any isotype of immunoglobulin heavy chain, including, for example, Cγ1, Cγ2, Cγ3, Cγ4, Cα1, Cα2, Cδ, Cε, and Cμ. Linker sequences may 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.

[0479] 1. Glycosylated variant linkers

[0480] As discussed above regarding anti-PD-1 Fv, glycosylation can lead to heterogeneity, which may be undesirable in some cases. Therefore, in some cases, for domain linkers connecting the IL-15 domain and the IL-15Rα domain to form the IL-15 complex, the (GGGGA)n linker (SEQ ID NO:19) is used, such as... Figure 8 As shown, n is between 1 and 5.

[0481] IV. Useful Forms of the Invention

[0482] like Figure 28A As shown in ~28H, the present invention provides many available forms of PD-1-targeting IL-15 / IL-15Rα(sushi)Fc fusion proteins. Generally, the heterodimeric fusion protein of the present invention has three functional components: an IL-15 / IL-15Rα(sushi) component, an anti-PD-1 component, and an Fc component, each of which can take different forms as outlined herein, and each component can be combined with other components in any conformation.

[0483] In any of the following forms, the first Fc domain and the second Fc domain may have a set of amino acid substitutions selected from the group consisting of: a) S267K / L368D / K370S:S267K / S364K / E357Q; b) S364K / E357Q:L368D / K370S; c) L368D / K370S:S364K; d) L368E / K370S:S364K; e) T411E / K360E / Q362E:D401K; f) L368D / K370S:S364K / E357L and g) K370S:S364K / E357Q (according to EU designation).

[0484] In some embodiments, the first Fc domain and / or the second Fc domain may have an additional set of amino acid substitutions comprising Q295E / N384D / Q418E / N421D (according to EU designation).

[0485] Optionally, the first Fc domain and / or the second Fc domain may have an additional set of amino acid substitutions consisting of G236R / L328R, E233P / L234V / L235A / G236del / S239K, E233P / L234V / L235A / G236del / S267K, E233P / L234V / L235A / G236del / S239K / A327G, E233P / L234V / L235A / G236del / S267K / A327G and E233P / L234V / L235A / G236del (according to EU designation).

[0486] Optionally, the first Fc domain and / or the second Fc domain have an M428L / N434S variant to extend the half-life. In some embodiments, the first Fc domain and / or the second Fc domain have a 428L / 434S variant to extend the half-life.

[0487] Therefore, particularly useful variants for any of the following forms include: positively charged monomers comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and negatively charged monomers comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D (when the monomer contains a CH1 domain) or the pI variant Q295E / N384D / Q418E / N421D (when the monomer does not contain a CH1 domain).

[0488] A.scIL-15 / RαX scFv

[0489] An implementation example Figure 28A As shown, it contains two monomers. It is commonly referred to as "scIL-15 / RαX scFv", where "sc" stands for "single-chain," referring to the connection between the IL-15 and sushi domains using a covalent domain connector. The form "scIL-15 / Rαx scFv" (see...) Figure 28A It contains IL-15Rα (sushi) fused to IL-15 via a variable-length connector (referred to as "scIL-15 / Rα") and then fused to the N-terminus of the heterodimer Fc region, wherein scFv is fused to the other side of the heterodimer Fc.

[0490] exist Figure 28A In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, the pI variant Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO: 20), and the anti-PD-1 Fv domain is selected from the group consisting of: mAbC H.176_L1.140, H1.19_L1.140, H1_L1.1, H1.19_L1, H1.48_L1, H1.125_L1, H1.30_L1, H1.132_L1, H1_L1.1 Combinations of VH and VL structural domains of H1_L1.3, H1_L1.45, H1_L1.117, H1_L1.129, H1.19_L1.1, H1.32_L1.1, H1.169_L1.1, H1.169_L1.1, H1.175_L1.1, H1.175_L1.1, H1_L1.140, H1_L1.135, H1_L1.136, H1.132_L1.135, H1.132_L1.140, H1.175_L1.135, and H1.175_L1.140.

[0491] exist Figure 28A In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, the pI variant Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H.176_L1.140, and the IL-15 domain contains variants selected from N4D / N65D, D30N / N65D, and D30N / E64Q / N65D.

[0492] exist Figure 28A In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, the pI variant Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), and the anti-PD-1 Fv domain is selected from the group consisting of the VH and VL domains of H1_L1.1, and the IL-15 domain contains variants selected from N4D / N65D, D30N / N65D, and D30N / E64Q / N65D.

[0493] B.scFv X ncIL-15 / Rα

[0494] The implementation is, for example Figure 28B The diagram shows three monomers. It is typically referred to as “ncIL-15 / RαX scFv” or “scFv X ncIL-15 / Rα”, where “nc” stands for “non-covalent,” referring to the non-covalent self-assembly of IL-15 with the sushi domain. The form “scFv x ncIL-15 / Rα” (see...) Figure 34B The scFv contains fused to the N-terminus of the heterodimer Fc region, wherein IL-15Rα(sushi) is fused to the other side of the heterodimer Fc, while IL-15 is transfected separately, thereby forming a non-covalent IL-15 / Rα complex.

[0495] exist Figure 28BIn the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, the pI variant Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H.176_L1.140, and the IL-15 domain contains variants selected from N4D / N65D, D30N / N65D, and D30N / E64Q / N65D.

[0496] exist Figure 28B In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, the pI variant Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), and the anti-PD-1 Fv domain is selected from the group consisting of the VH and VL domains of H1_L1.1, and the IL-15 domain contains variants selected from N4D / N65D, D30N / N65D, and D30N / E64Q / N65D.

[0497] C.scIL-15 / RαX Fab

[0498] The implementation is, for example Figure 28C As shown, it contains three monomers. It is commonly referred to as "scIL-15 / RαX Fab" or "FabX scIL-15 / Rα," the two terms being interchangeable, where "sc" stands for "single-chain." The scIL-15 / Rαx Fab form (see...) Figure 28CThe IL-15Rα(sushi) comprises IL-15 fused to IL-15 via a variable-length connector (referred to as "scIL-15 / Rα") and then fused to the N-terminus of the Fc region of the heterodimer (containing a hinge, which here serves as a second domain connector). That is, from the N-terminus to the C-terminus, the first monomer is the variant IL-15-first domain connector-IL-15Rαsushi domain-second domain connector-CH2-CH3. In some cases, the second domain connector is as follows: Figure 8 The full-length hinge domain is present. The second monomer is the heavy chain VH-CH1-hinge-CH2-CH3, while the corresponding light chain (the third monomer) is transfected separately to form Fab with VH.

[0499] In some Figure 28C In this embodiment, the anti-PD-1 Fv domain is selected from the group consisting of: mAbCH.176_L1.140, H1_L1, H1.19_L1.140, H1_L1.1, H1.19_L1, H1.48_L1, H1.125_L1, H1.30_L1, H1.132_L1, H1_L1.1 Combinations of VH and VL structural domains of H1_L1.3, H1_L1.45, H1_L1.117, H1_L1.129, H1.19_L1.1, H1.32_L1.1, H1.169_L1.1, H1.169_L1.1, H1.175_L1.1, H1.175_L1.1, H1_L1.140, H1_L1.135, H1_L1.136, H1.132_L1.135, H1.132_L1.140, H1.175_L1.135, and H1.175_L1.140.

[0500] exist Figure 28CIn the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H1.176_L1.140, and the IL-15 domain contains amino acid substitutions D30N / N65D for efficacy and N71Q / N79Q / N112Q for glycosylation. Furthermore, the domain linker between the IL-15 and sushi domains is GGGGA (SEQ ID NO:8).

[0501] exist Figure 28C In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and the Xtend428L / 434S variant; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, the Xtend428L / 434S variant, and the pI variant N208D / Q295E / N384D / Q418E / N421D. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H1.176_L1.140, and the IL-15 domain contains amino acid substitutions D30N / N65D for efficacy and N71Q / N79Q / N112Q for glycosylation. Furthermore, the domain linker between the IL-15 and sushi domains is GGGGA (SEQ ID NO:8).

[0502] exist Figure 28CIn the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H1.176_L1.140, and the IL-15 domain contains amino acid substitutions D30N / E64Q / N65D for efficacy and N71Q / N79Q / N112Q for glycosylation. Furthermore, the domain linker between the IL-15 and sushi domains is GGGGA (SEQ ID NO:8).

[0503] exist Figure 28C In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and the Xtend428L / 434S variant; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, the Xtend428L / 434S variant, and the pI variant N208D / Q295E / N384D / Q418E / N421D. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H1.176_L1.140, and the IL-15 domain contains amino acid substitutions D30N / E64Q / N65D for efficacy and N71Q / N79Q / N112Q for glycosylation. Furthermore, the domain linker between the IL-15 and sushi domains is GGGGA (SEQ ID NO:8).

[0504] In a particularly useful embodiment, the fusion protein is XENP32435, the sequence of which is as follows: Figure 138A As shown.

[0505] exist Figure 28CIn the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), and the anti-PD-1 Fv domain is selected from the group consisting of the VH and VL domains of H1_L1.1, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0506] The amino acid sequence of an exemplary non-competitive PD-1 targeting IL-15 / Rα-Fc fusion protein in the form of scIL-15 / Rαx Fab ( Figure 28C (See Figures 30, 48, 49 and 68.)

[0507] In some embodiments, the first monomer comprises a sushi domain-domain connector-variant IL-15-domain connector-CH2-CH3 from the N-terminus to the C-terminus, and the second monomer comprises a heavy chain VH-CH1-hinge-CH2-CH3. The third monomer is a light chain VL-CL. Preferred combinations of variations of this embodiment are described in PCT / US2017 / 056826. Figure 7C .

[0508] In the scIL-15 / RαX Fab form, a preferred embodiment utilizes a skewed variant of S364K / E357Q:L368D / K370S.

[0509] In the scIL-15 / RαX Fab form, a preferred embodiment utilizes the skew variants S364K / E357Q (on the scFv-Fc monomer) and L368D / K370S (on the IL-15 complex monomer), the pI variants Q295E / N384D / Q418E / N421D (on one side of the IL-15 complex), the ablation variants E233P / L234V / L235A / G236_ / S267K on both monomers, and optionally the 428L / 434S variants on both sides.

[0510] D.ncIL-15 / RαX Fab

[0511] The implementation is, for example Figure 28DThe diagram shows a complex containing three monomers. It is commonly referred to as “ncIL-15 / RαX Fab” or “FabX ncIL-15 / Rα”, which are interchangeable. “nc” stands for “non-covalent,” referring to the non-covalent self-assembly of IL-15 with the sushi domain. The ncIL-15 / Rαx Fab form (see Figure 34D) contains a VH fused to the N-terminus of the heterodimer Fc region, with IL-15Rα (sushi) fused to the other side of the heterodimer Fc. The corresponding light chain is transfected separately to form a Fab containing the VH, and simultaneously, IL-15 is transfected separately, thus forming a non-covalent IL-15 / Rα complex.

[0512] In some embodiments, the first monomer comprises a sushi domain-domain connector-CH2-CH3 from the N-terminus to the C-terminus, and the second monomer comprises a heavy chain VH-CH1-hinge-CH2-CH3. The third monomer is an IL-15 domain. In the ncIL-15 / RαX Fab form, a preferred embodiment utilizes a skewed variant of S364K / E357Q:L368D / K370S.

[0513] exist Figure 28D In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H.176_L1.140, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0514] exist Figure 28DIn the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), and the anti-PD-1 Fv domain is selected from the group consisting of the VH and VL domains of H1_L1.1, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0515] E.mAb-scIL-15 / Rα

[0516] The implementation is, for example Figure 28E As shown, it contains three monomers (although the fusion protein is a tetramer). This is commonly referred to as “mAb-scIL-15 / Rα”, where “sc” stands for “single-chain”. The mAb-scIL-15 / Rα form (see Figure 34E) contains a VH fused to the N-terminus of the first heterodimer Fc and the second heterodimer Fc, wherein IL-15 is fused to IL-15Rα (sushi) and then further fused to the C-terminus of one of the heterodimer Fc regions, while the corresponding light chain is transfected separately to form a Fab containing the VH.

[0517] In some embodiments, the first monomer comprises a heavy chain VH-CH1-hinge-CH2-CH3. The second monomer comprises a heavy chain containing the scIL-15 complex, VH-CH1-hinge-CH2-CH3-domain connector-sushi domain-domain connector-IL-15. The third (and fourth) monomers are light chains VL-CL. This is commonly referred to as “mAb-scIL-15 / Rα”, where “sc” stands for “single chain”. In the mAb-scIL-15 / Rα form, a preferred embodiment utilizes a skewed variant of S364K / E357Q:L368D / K370S.

[0518] exist Figure 28EIn the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H.176_L1.140, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0519] exist Figure 28E In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), and the anti-PD-1 Fv domain is selected from the group consisting of the VH and VL domains of H1_L1.1, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0520] F.mAb-ncIL-15 / Rα

[0521] The implementation is, for example Figure 28F As shown, it contains four monomers (although the heterodimeric fusion protein is a pentamer). This is commonly referred to as “mAb-ncIL-15 / Rα”, where “sc” stands for “non-covalent”. The “mAb-ncIL-15 / Rα” form (Fig. 34F) contains a VH fused to the N-terminus of the first heterodimer Fc and the second heterodimer Fc, wherein IL-15Rα (sushi) is fused to the C-terminus of one of the heterodimer Fc regions, while the corresponding light chain is transfected separately to form a Fab containing the VH, and IL-15 is transfected separately, thereby forming a non-covalent IL-15 / Rα complex.

[0522] In some embodiments, the first monomer comprises a heavy chain VH-CH1-hinge-CH2-CH3. The second monomer comprises a heavy chain containing an IL-15Rα (sushi) domain, a VH-CH1-hinge-CH2-CH3 domain connector, and a sushi domain. The third monomer is an IL-15 domain. The fourth (and fifth) monomers are light chains VL-CL. In the mAb-ncIL-15 / Rα form, a preferred embodiment utilizes a skewed variant of S364K / E357Q:L368D / K370S.

[0523] exist Figure 28F In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H.176_L1.140, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0524] exist Figure 28F In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), and the anti-PD-1 Fv domain is selected from the group consisting of the VH and VL domains of H1_L1.1, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0525] G. Central IL-15 / Rα

[0526] The implementation is, for example Figure 28G As shown, it contains four monomers that form a tetramer. This is commonly referred to as the "central IL-15 / Rα". The central IL-15 / Rα form (see...) Figure 28G The product contains a VH that is recombined and fused to the N-terminus of IL-15 and then fused to one side of the heterodimer Fc, and a VH that is recombined and fused to the N-terminus of IL-15Rα(sushi) and then further fused to the other side of the heterodimer Fc, while the corresponding light chains are transfected separately to form a Fab containing VH.

[0527] exist Figure 28G In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H.176_L1.140, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0528] exist Figure 28G In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), and the anti-PD-1 Fv domain is selected from the group consisting of the VH and VL domains of H1_L1.1, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0529] H. Central scIL-15 / Rα

[0530] The implementation is, for example Figure 28H As shown, it contains four monomers forming a tetramer. This is often referred to as the “central scIL-15 / Rα”, where “sc” stands for “single chain”. The central scIL-15 / Rα form (see Figure 34H) contains a VH fused to the N-terminus of IL-15Rα (sushi) (which is fused to IL-15 and then further fused to one side of the heterodimer Fc) and a VH fused to the other side of the heterodimer Fc, while the corresponding light chain is transfected separately to form a Fab containing the VH.

[0531] exist Figure 28H In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), the anti-PD-1 Fv domain is a combination of the VH and VL domains of H.176_L1.140, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0532] exist Figure 28H In the form shown, some aspects include an Fc domain comprising: a positively charged monomer comprising the skew variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K; and a negatively charged monomer comprising the skew variant L368D / K370S, the ablation variant E233P / L234V / L235A / G236del / S267K, and the pI variant N208D / Q295E / N384D / Q418E / N421D, and the optional 428L / 434S FcRn variant. In these embodiments, the sushi domain is wild-type (SEQ ID NO:20), and the anti-PD-1 Fv domain is selected from the group consisting of the VH and VL domains of H1_L1.1, and the IL-15 domain contains variants selected from D30N / N65D and D30N / E64Q / N65D.

[0533] V. Useful Embodiments of the Invention

[0534] This document provides a PD-1-targeting IL-15 / Rα-Fc fusion protein comprising one or more engineered amino acid substitutions of the IL-15 protein, and an anti-PD-1 ABD that does not compete with selected approved anti-PD-1 binding antibodies. In some embodiments, the IL-15 variant of the Fc fusion protein has an N4D / N65D substitution. In some embodiments, the IL-15 variant of the Fc fusion protein has a D30N substitution. In some embodiments, the IL-15 variant of the Fc fusion protein has a D30N / E64Q / N65D substitution. In some embodiments, the IL-15 variant of the Fc fusion protein has a D30N / N65D substitution. This IL-15 / Rα-Fc fusion protein comprising the fusion protein can induce or promote immune cells (including NK cells, CD8 cells, etc.). + T cells and CD4 + Proliferation of T cells. Notably, fusion proteins containing IL-15 / Rα-Fc without an exogenous adaptor on the IL-15Fc side (e.g., the hinge domain is the only domain adaptor) exhibited weaker proliferative activity.

[0535] This document describes a PD-1-targeting IL-15 / Rα-Fc fusion protein with lower potency, higher pharmacokinetic characteristics, and / or prolonged serum half-life. The PD-1-targeting IL-15 / Rα-Fc fusion protein described herein is engineered to reduce its potency compared to the parental construct. In some embodiments, one or more amino acid substitutions are introduced into the Fc domain of the IL-15 / Rα complex and / or the heterodimeric Fc fusion protein. In some embodiments, the PD-1-targeting IL-15 / Rα-Fc fusion protein with lower potency compared to a control construct (e.g., the parental construct) has a significantly prolonged serum half-life. In some embodiments, the serum half-life is increased by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, or more times.

[0536] This article presents an enhancement of GVHD in animal models (e.g., human PBMC-transplanted NSG mice) of PD-1-targeted IL-15 / Rα-Fc fusion protein by combining it with an anti-PD-1 antibody, compared to a modified control scIL-15 / Rα-Fc fusion protein with reduced potency. Administration of the exemplary non-competitive PD-1-targeted IL-15 / Rα-Fc fusion protein produced a greater effect than the combination of IL-15 and PD-1 blockers.

[0537] The PD-1 targeting IL-15 / Rα-Fc fusion proteins described herein (including non-competitive PD-1 targeting IL-15 / Rα-Fc fusion proteins) can induce STAT5 phosphorylation in immune cells, including but not limited to activated lymphocytes, activated T cells (e.g., activated CD4+ T cells and activated CD8+ cells), and activated tumor-infiltrating lymphocytes (e.g., activated TILs).

[0538] VI. Non-targeted heterodimer Fc fusion protein

[0539] One aspect of the present invention provides a heterodimeric protein comprising: a) a first fusion protein comprising i) a variant IL-15 protein comprising amino acid substitutions for D30N / E64Q / N65D and N71Q / N79Q / N112Q compared to SEQ ID NO:2; ii) a domain linker; and iii) a first variant Fc domain; and b) a second fusion protein comprising i) an IL-15Rαsushi domain; ii) a domain linker; and iii) a second variant Fc domain.

[0540] In some embodiments, the variant IL-15 protein comprises the amino acid sequence of SEQ ID NO:319.

[0541] In some embodiments, according to EU designations, the first and second variant Fc domains comprise amino acid substitutions selected from the group consisting of: S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411E / K360E / Q362E:D401K; L368D / K370S:S364K / E357L, K370S:S364K / E357Q, T366S / L368A / Y407V:T366W, and T366S / L368A / Y407V / Y349C:T366W / S354C. In some embodiments, according to the EU designation, the first variant Fc structure domain includes L368D / K370S and the second variant Fc structure domain includes S364K / E357Q.

[0542] In some embodiments, the first and second variant Fc domains each independently contain 428L / 434S. In some embodiments, both the first and second variant Fc domains contain 428L / 434S.

[0543] In some embodiments, the IL-15 / Rα heterodimer Fc fusion protein is selected from... Figure 115A-115CThe IL-15 / Rα heterodimeric Fc fusion protein is selected from the group consisting of XENP31967, XENP31968, XENP31969, XENP31970, XENP31971, XENP31972, and XENP31973. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is XENP31969. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is XENP31972. In some embodiments, the first fusion protein comprises the amino acid sequence of SEQ ID NO:208, and the second fusion protein comprises the amino acid sequence of SEQ ID NO:95. In some embodiments, the first fusion protein comprises the amino acid sequence of SEQ ID NO:211, and the second fusion protein comprises the amino acid sequence of SEQ ID NO:206.

[0544] In one aspect, this article provides a method for inducing the proliferation of regulatory T cells (Tregs) with reduced or minimal immunosuppressive activity in patients with such need, the method comprising administering to the patient a therapeutically effective amount of an IL-15 / Rα heterodimeric Fc fusion protein comprising: (a) a first monomer comprising, from the N-terminus to the C-terminus: i) a variant IL-15 protein; ii) a first domain adapter; and iii) a first variant Fc domain comprising CH2-CH3; and (b) a second monomer comprising, from the N-terminus to the C-terminus: i) an IL-15Rαsushi domain protein; ii) a second domain adapter; and iii) a second variant Fc domain comprising CH2-CH3.

[0545] In some embodiments, the variant IL-15 protein of the heterodimeric Fc fusion protein comprises the amino acid sequence of SEQ ID NO:2 and amino acid substitutions selected from the group consisting of: N1D; N4D; D8N; D30N; D61N; E64Q; N65D; Q108E; N1D / N4D / D8N; N1D / N4D / N65D; N1D / D30N; N1D / D61N; N1D / D61N / E64Q / Q108E; N1D / E64Q; N1D / N65D; N1D / Q108E; N4D; N4D / D30N; N4D / D61N; N4D / D61N / N65D; N4D / D61N / E64Q / Q108E; N4D / E64Q; N4D / N65D; D8N / D61N; D8N / E64Q; D30N / E64Q; D30N / N65D; D30N / E64Q / N65D; D30N / Q180E; D61N / E64Q / N65D; E64Q; E64Q / N65D; E64Q / Q108E; and N65D / Q108E. In one embodiment, the variant IL-15 protein comprises the amino acid sequence of SEQ ID NO:2 and amino acid substitutions selected from the group consisting of: N65D; D30N / E64Q / N65D; N4D / N65D; D30N / E64Q; and D30N / N65D.

[0546] In some embodiments, the variant IL-15 protein comprises amino acid substitutions D30N / E64Q / N65D and N71Q / N79Q / N112Q, compared to SEQ ID NO:2. In some embodiments, the variant IL-15 protein comprises the amino acid sequence of SEQ ID NO:319.

[0547] In some embodiments, the IL-15Rαsushi domain protein comprises the amino acid sequence of SEQ ID NO:20. In some embodiments, the IL-15Rα protein comprises the amino acid sequence of SEQ ID NO:4.

[0548] In some embodiments, according to EU designations, the first variant Fc domain and the second variant Fc domain have a set of amino acid substitutions selected from the group consisting of: S267K / L368D / K370S:S267K / S364K / E357Q; S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411E / K360E / Q362E:D401K; L368D / K370S:S364K / E357L and K370S:S364K / E357Q. In some embodiments, according to EU designations, the first variant Fc domain and the second variant Fc domain comprise amino acid substitutions of S267K / L368D / K370S:S267K / S364K / E357Q. In some embodiments, according to EU designations, the first variant Fc domain and the second variant Fc domain each comprise amino acid substitutions of M428L / N434S.

[0549] In some embodiments, the IL-15 / Rα heterodimer Fc fusion protein is selected from... Figure 115A-115C The IL-15 / Rα heterodimeric Fc fusion protein is selected from the group consisting of XENP31967, XENP31968, XENP31969, XENP31970, XENP31971, XENP31972, and XENP31973. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is XENP31969. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is XENP31972.

[0550] In some embodiments, the non-targeted heterodimeric Fc fusion protein comprises a first monomer comprising a first fusion protein comprising a variant IL-15 protein (compared to SEQ ID NO:2, comprising amino acid substitutions D30N / E64Q / N65D and N71Q / N79Q / N112Q) linked to a first variant Fc domain using a domain linker. A second monomer comprises a second fusion protein comprising an IL-15Rαsushi domain linked to a second variant Fc domain using a domain linker. In some embodiments, the heterodimeric protein is selected from XENP 31969 and XENP 31972.

[0551] In some embodiments, the heterodimeric fusion protein promotes effector memory T cells (T cells). EM Amplification. In one embodiment, the heterodimeric fusion protein increases Treg and T EM The ratio (Treg / TEM In some cases, treatment with any of the IL-15 / Rα Fc fusion proteins outlined herein converts Tregs from a suppressor Treg cell type to non-suppressor activated effector CD4 T cells. In one embodiment, FOXP3... 高 CD45RA – CD4 + The effect of Treg differentiation into FOXP3 低 CD45RA - CD4 + Activated effector CD4 T cells. In some embodiments, activated effector CD4 T cells have reduced CCR4 expression.

[0552] In some embodiments, the subjects had cancer.

[0553] Useful non-targeting IL-15 / IL-15Rα heterodimer Fc fusion proteins are described in detail in, for example, the following patents: U.S. Provisional Application No. 62 / 408,655, filed October 15, 2016; U.S. Provisional Application No. 62 / 416,087, filed November 1, 2016; U.S. Provisional Application No. 62 / 443,465, filed January 6, 2017; U.S. Provisional Application No. 62 / 477,926, filed March 28, 2017; U.S. Patent Publication No. US2018 / 0118805, filed October 16, 2017; and WO Publication No. WO20180719, filed October 16, 2017. U.S. Provisional Application No. 62 / 659,563, filed April 18, 2018; U.S. Provisional Application No. 62 / 684,143, filed June 12, 2018; U.S. Provisional Application No. 62 / 724,396, filed August 29, 2018; U.S. Provisional Application No. 62 / 756,800, filed November 7, 2018; U.S. Application No. 16 / 388,174, filed April 18, 2019; and PCT Application No. PCT / US19 / 28107, filed April 18, 2019, are incorporated herein by reference in their entirety, particularly the accompanying drawings, illustrations, sequence lists, and claims.

[0554] One particularly useful non-targeting IL-15 / IL-15Rα heterodimer Fc fusion protein is XmAb24306 (also known as "XENP24306"), which is used to reverse the inhibition of TGFβ-mediated T cell proliferation in patients with this need. Its sequence is as follows: Figure 20 As shown.

[0555] Furthermore, targeting the IL-15 / RαX PD-1 heterodimer Fc fusion protein is also used in methods for inducing the proliferation of regulatory T cells (Tregs) with reduced or minimal immunosuppressive activity in patients with this need. Any targeting structure shown in the figure can be found for use in this application, with XENP32435 having a particular application.

[0556] In one aspect, this article provides a method for reversing TGFβ-mediated T cell proliferation inhibition in patients in need, comprising administering to a patient a therapeutically effective amount of an IL-15 / Rα heterodimer Fc fusion protein comprising: (a) a first monomer comprising, from the N-terminus to the C-terminus: i) a variant IL-15 protein; ii) a first domain adapter; and iii) a first variant Fc domain comprising CH2-CH3; and (b) a second monomer comprising, from the N-terminus to the C-terminus: i) an IL-15Rαsushi domain protein; ii) a second domain adapter; and iii) a second variant Fc domain comprising CH2-CH3.

[0557] In some embodiments, the variant IL-15 protein of the heterodimeric Fc fusion protein comprises the amino acid sequence of SEQ ID NO:2 and one or more amino acid substitutions selected from the group consisting of: N1D; N4D; D8N; D30N; D61N; E64Q; N65D; Q108E; N1D / N4D / D8N; N1D / N4D / N65D; N1D / D30N; N1D / D61N; N1D / D61N / E64Q / Q108E; N1D / E64Q; N1D / N65D; N1D / Q108E; N4D; N4D / D30N; N 4D / D61N; N4D / D61N / N65D; N4D / D61N / E64Q / Q108E; N4D / E64Q; N4D / N65D; D8N / D61N; D8N / E64Q; D30N / E64Q; D30N / N65D; D30N / E64Q / N65D; D30N / Q180E; D61N / E64Q / N65D; E64Q; E64Q / N65D; E64Q / Q108E; and N65D / Q108E. In one embodiment, the variant IL-15 protein comprises the amino acid sequence of SEQ ID NO:2 and one or more amino acid substitutions selected from the group consisting of: N65D; D30N / E64Q / N65D; N4D / N65D; D30N / E64Q; and D30N / N65D. In some embodiments, the variant IL-15 protein comprises the amino acid substitutions D30N / E64Q / N65D and N71Q / N79Q / N112Q compared to SEQ ID NO:2. In some embodiments, the variant IL-15 protein comprises the amino acid sequence SEQ ID NO:319.

[0558] In some embodiments, the IL-15Rαsushi domain protein comprises the amino acid sequence of SEQ ID NO:20. In some embodiments, the IL-15Rα protein comprises the amino acid sequence of SEQ ID NO:4.

[0559] In some embodiments, the first variant Fc domain and the second variant Fc domain have a set of amino acid substitutions selected from the group consisting of: S267K / L368D / K370S:S267K / S364K / E357Q; S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411E / K360E / Q362E:D401K; L368D / K370S:S364K / E357L and K370S:S364K / E357Q (according to EU designations). In some embodiments, the first variant Fc domain and the second variant Fc domain comprise amino acid substitutions of S267K / L368D / K370S:S267K / S364K / E357Q (according to EU designation). In some embodiments, the first variant Fc domain and the second variant Fc domain each comprise amino acid substitutions of M428L / N434S (according to EU designation).

[0560] In some embodiments, the IL-15 / Rα heterodimer Fc fusion protein is selected from... Figures 115A-115C The diagram shows the group of IL-15 / Rα heterodimeric Fc fusion proteins. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is selected from the group consisting of: XENP31967, XENP31968, XENP31969, XENP31970, XENP31971, XENP31972, and XENP31973. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is XENP31969. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is XENP31972.

[0561] In some embodiments, the non-targeted heterodimeric Fc fusion protein comprises a first monomer comprising a first fusion protein comprising an isoform of IL-15 protein (comprising amino acid substitutions D30N / E64Q / N65D and N71Q / N79Q / N112Q (compared to SEQ ID NO:2)) connected to a first variant Fc domain via a domain linker. A second monomer comprises a second fusion protein comprising an IL-15Rαsushi domain connected to a second variant Fc domain via a domain linker. In some embodiments, the heterodimeric protein is selected from XENP 31969 and XENP 31972.

[0562] In some embodiments, the patient exhibits enhanced T cell proliferation after administration. In one embodiment, the T cell proliferation is CD4 T cell proliferation. In one embodiment, the T cell proliferation is CD8 T cell proliferation. In some embodiments, the T cell proliferation is both CD4 and CD8 T cell proliferation.

[0563] In some embodiments, the patient has cancer.

[0564] Useful non-targeting IL-15 / IL-15Rα heterodimer Fc fusion proteins are described in detail in, for example, the following patents: U.S. Provisional Application No. 62 / 408,655, filed October 15, 2016; U.S. Provisional Application No. 62 / 416,087, filed November 1, 2016; U.S. Provisional Application No. 62 / 443,465, filed January 6, 2017; U.S. Provisional Application No. 62 / 477,926, filed March 28, 2017; U.S. Patent Publication No. US2018 / 0118805, filed October 16, 2017; and WO Publication No. WO20180719, filed October 16, 2017. U.S. Provisional Application No. 62 / 659,563, filed April 18, 2018; U.S. Provisional Application No. 62 / 684,143, filed June 12, 2018; U.S. Provisional Application No. 62 / 724,396, filed August 29, 2018; U.S. Provisional Application No. 62 / 756,800, filed November 7, 2018; U.S. Application No. 16 / 388,174, filed April 18, 2019; and PCT Application No. PCT / US19 / 28107, filed April 18, 2019, are incorporated herein by reference in their entirety, particularly the accompanying drawings, illustrations, sequence lists, and claims.

[0565] One particularly useful non-targeting IL-15 / IL-15Rα heterodimer Fc fusion protein is XmAb24306 (also known as "XENP24306"), which is used to reduce the expression level of FOXP3 in T cells in patients with this need. Its sequence is as follows: Figure 20 As shown.

[0566] Furthermore, targeting the IL-15 / RαX PD-1 heterodimer Fc fusion protein is also used in methods to reduce FOXP3 expression levels in T cells of patients requiring this. Any targeting structures shown in the figure can be found for use in this application, with XENP32435 having a particular application.

[0567] On the other hand, this article provides a method for reducing FOXP3 expression levels in T cells of patients in need, comprising administering to a patient a therapeutically effective amount of an IL-15 / Rα heterodimer Fc fusion protein comprising: (a) a first monomer comprising, from the N-terminus to the C-terminus: i) a variant IL-15 protein; ii) a first domain adapter; and iii) a first variant Fc domain comprising CH2-CH3; and (b) a second monomer comprising, from the N-terminus to the C-terminus: i) an IL-15Rαsushi domain protein; ii) a second domain adapter; and iii) a second variant Fc domain comprising CH2-CH3.

[0568] In some embodiments, the variant IL-15 protein of the heterodimeric Fc fusion protein comprises the amino acid sequence of SEQ ID NO:2 and one or more amino acid substitutions selected from the group consisting of: N1D; N4D; D8N; D30N; D61N; E64Q; N65D; Q108E; N1D / N4D / D8N; N1D / N4D / N65D; N1D / D30N; N1D / D61N; N1D / D61N / E64Q / Q108E; N1D / E64Q; N1D / N65D; N1D / Q108E; N4D; N4D / D30N; N 4D / D61N; N4D / D61N / N65D; N4D / D61N / E64Q / Q108E; N4D / E64Q; N4D / N65D; D8N / D61N; D8N / E64Q; D30N / E64Q; D30N / N65D; D30N / E64Q / N65D; D30N / Q180E; D61N / E64Q / N65D; E64Q; E64Q / N65D; E64Q / Q108E; and N65D / Q108E. In one embodiment, the variant IL-15 protein comprises the amino acid sequence of SEQ ID NO:2 and one or more amino acid substitutions selected from the group consisting of: N65D; D30N / E64Q / N65D; N4D / N65D; D30N / E64Q; and D30N / N65D. In some embodiments, the variant IL-15 protein comprises the amino acid substitutions D30N / E64Q / N65D and N71Q / N79Q / N112Q compared to SEQ ID NO:2. In some embodiments, the variant IL-15 protein comprises the amino acid sequence SEQ ID NO:319.

[0569] In some embodiments, the IL-15Rαsushi domain protein comprises the amino acid sequence of SEQ ID NO:20. In some embodiments, the IL-15Rα protein comprises the amino acid sequence of SEQ ID NO:4.

[0570] In some embodiments, the first variant Fc domain and the second variant Fc domain have a set of amino acid substitutions selected from the group consisting of: S267K / L368D / K370S:S267K / S364K / E357Q; S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411E / K360E / Q362E:D401K; L368D / K370S:S364K / E357L and K370S:S364K / E357Q (according to EU designations). In some embodiments, the first variant Fc domain and the second variant Fc domain comprise amino acid substitutions of S267K / L368D / K370S:S267K / S364K / E357Q (according to EU designation). In some embodiments, the first variant Fc domain and the second variant Fc domain each comprise amino acid substitutions of M428L / N434S (according to EU designation).

[0571] In some embodiments, the IL-15 / Rα heterodimer Fc fusion protein is selected from... Figures 115A-115C The diagram shows the group of IL-15 / Rα heterodimeric Fc fusion proteins. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is selected from the group consisting of: XENP31967, XENP31968, XENP31969, XENP31970, XENP31971, XENP31972, and XENP31973. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is XENP31969. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is XENP31972.

[0572] In some embodiments, the patient has an expanded population of non-inhibitory regulatory T cells (Tregs) after administration.

[0573] In some embodiments, the patient has an expanded and activated effector CD4 T cell population after administration.

[0574] In some embodiments, patients have an increased ratio of Tregs to effector memory T cells (Treg / T) after administration. EM ).

[0575] In any embodiment of the invention, the patient has cancer.

[0576] Useful non-targeting IL-15 / IL-15Rα heterodimer Fc fusion proteins are described in detail in, for example, the following patents: U.S. Provisional Application No. 62 / 408,655, filed October 15, 2016; U.S. Provisional Application No. 62 / 416,087, filed November 1, 2016; U.S. Provisional Application No. 62 / 443,465, filed January 6, 2017; U.S. Provisional Application No. 62 / 477,926, filed March 28, 2017; U.S. Patent Publication No. US2018 / 0118805, filed October 16, 2017; and WO Publication No. WO20180719, filed October 16, 2017. U.S. Provisional Application No. 62 / 659,563, filed April 18, 2018; U.S. Provisional Application No. 62 / 684,143, filed June 12, 2018; U.S. Provisional Application No. 62 / 724,396, filed August 29, 2018; U.S. Provisional Application No. 62 / 756,800, filed November 7, 2018; U.S. Application No. 16 / 388,174, filed April 18, 2019; and PCT Application No. PCT / US19 / 28107, filed April 18, 2019, are incorporated herein by reference in their entirety, particularly the accompanying drawings, illustrations, sequence lists, and claims.

[0577] One particularly useful non-targeting IL-15 / IL-15Rα heterodimer Fc fusion protein is XmAb24306 (also known as "XENP24306"), which is used to reduce the expression level of FOXP3 in T cells in patients with this need. Its sequence is as follows: Figure 20 As shown.

[0578] Furthermore, targeting the IL-15 / RαX PD-1 heterodimer Fc fusion protein is also used in methods to reduce FOXP3 expression levels in T cells of patients requiring this. Any targeting structures shown in the figure can be found for use in this application, with XENP32435 having a particular application.

[0579] On another aspect, this article provides a method for expanding activated effector CD4 T cells in a patient with cancer, the method comprising administering to the patient a therapeutically effective amount of an IL-15 / Rα heterodimeric Fc fusion protein comprising: (a) a first monomer comprising, from the N-terminus to the C-terminus: i) a variant IL-15 protein; ii) a first domain adapter; and iii) a first variant Fc domain comprising CH2-CH3; and (b) a second monomer comprising, from the N-terminus to the C-terminus: i) an IL-15Rαsushi domain protein; ii) a second domain adapter; and iii) a second variant Fc domain comprising CH2-CH3.

[0580] In some embodiments, the variant IL-15 protein of the heterodimeric Fc fusion protein comprises the amino acid sequence of SEQ ID NO:2 and amino acid substitutions selected from the group consisting of: N1D; N4D; D8N; D30N; D61N; E64Q; N65D; Q108E; N1D / N4D / D8N; N1D / N4D / N65D; N1D / D30N; N1D / D61N; N1D / D61N / E64Q / Q108E; N1D / E64Q; N1D / N65D; N1D / Q108E; N4D; N4D / D30N; N4D / D61N; N4D / D61N / N65D; N4D / D61N / E64Q / Q108E; N4D / E64Q; N4D / N65D; D8N / D61N; D8N / E64Q; D30N / E64Q; D30N / N65D; D30N / E64Q / N65D; D30N / Q180E; D61N / E64Q / N65D; E64Q; E64Q / N65D; E64Q / Q108E; and N65D / Q108E. In one embodiment, the variant IL-15 protein comprises the amino acid sequence of SEQ ID NO:2 and amino acid substitutions selected from the group consisting of: N65D; D30N / E64Q / N65D; N4D / N65D; D30N / E64Q; and D30N / N65D. In some embodiments, compared to SEQ ID NO:2, the variant IL-15 protein comprises the amino acid substitutions D30N / E64Q / N65D and N71Q / N79Q / N112Q. In some embodiments, the variant IL-15 protein comprises the amino acid sequence of SEQ ID NO:319.

[0581] In some embodiments, the IL-15Rαsushi domain protein comprises the amino acid sequence of SEQ ID NO:20. In some embodiments, the IL-15Rα protein comprises the amino acid sequence of SEQ ID NO:4.

[0582] In some embodiments, according to EU designations, the first variant Fc domain and the second variant Fc domain have a set of amino acid substitutions selected from the group consisting of: S267K / L368D / K370S:S267K / S364K / E357Q; S364K / E357Q:L368D / K370S; L368D / K370S:S364K; L368E / K370S:S364K; T411E / K360E / Q362E:D401K; L368D / K370S:S364K / E357L and K370S:S364K / E357Q. In some embodiments, according to EU designations, the first variant Fc domain and the second variant Fc domain comprise amino acid substitutions of S267K / L368D / K370S:S267K / S364K / E357Q. In some embodiments, according to EU designations, the first variant Fc domain and the second variant Fc domain each comprise amino acid substitutions of M428L / N434S.

[0583] In some embodiments, the IL-15 / Rα heterodimer Fc fusion protein is selected from... Figure 115A-115C The IL-15 / Rα heterodimeric Fc fusion protein is selected from the group consisting of XENP31967, XENP31968, XENP31969, XENP31970, XENP31971, XENP31972, and XENP31973. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is XENP31969. In some embodiments, the IL-15 / Rα heterodimeric Fc fusion protein is XENP31972.

[0584] In some embodiments, the activated effector CD4 T cells are FOXP3 cells. 低 CD45RA - CD4 + T cells.

[0585] In some embodiments, activated effector CD4 T cells have reduced CCR4 expression or are CCR4-positive. lo / - .

[0586] In some embodiments, patients exhibit an increased Treg to effector memory T cell ratio (Treg / T) after administration. EM ).

[0587] Available non-targeted IL-15 / IL-15Rα heterodimer Fc fusion proteins are detailed in, for example: U.S. Provisional Application No. 62 / 408,655, filed October 15, 2016; U.S. Provisional Application No. 62 / 416,087, filed November 1, 2016; U.S. Provisional Application No. 62 / 443,465, filed January 6, 2017; U.S. Provisional Application No. 62 / 477,926, filed March 28, 2017; U.S. Patent Publication No. US2018 / 0118805, filed October 16, 2017; WO Publication No. WO2018071919, filed October 16, 2017; 2018 The following U.S. Provisional Application No. 62 / 659,563, filed April 18; U.S. Provisional Application No. 62 / 684,143, filed June 12, 2018; U.S. Provisional Application No. 62 / 724,396, filed August 29, 2018; U.S. Provisional Application No. 62 / 756,800, filed November 7, 2018; U.S. Patent Application No. 16 / 388,174, filed April 18, 2019; and PCT Patent Application No. PCT / US19 / 28107, filed April 18, 2019, are incorporated herein by reference in their entirety, and especially with reference to the drawings, illustrations, sequential listings, and claims therein.

[0588] A particularly useful non-targeted IL-15 / IL-15Rα heterodimer Fc fusion protein, XmAb24306 (also known as "XENP24306"), for expanding activated effector CD4 T cells in patients in need, has the following sequence: Figure 20 As shown.

[0589] Furthermore, targeting the IL-15 / RαX PD-1 heterodimer Fc fusion protein is also used in methods for expanding activated effector CD4 T cells in patients with this need. Any targeting structure shown in the figure can be found for use in this application, with XENP32435 having a specific application.

[0590] VII. Nucleic Acids of the Invention

[0591] The present invention further provides a nucleic acid composition encoding the heterodimeric Fc fusion protein of the present invention (or, in the case of monomeric Fc domain proteins, further comprising the nucleic acid encoding them).

[0592] As those skilled in the art will understand, the nucleic acid composition will depend on the form of the heterodimeric IL-15 / Rα Fc fusion protein. Therefore, for example, when the form requires a three-amino acid sequence, the three nucleic acid sequences can be incorporated into one or more expression vectors for expression. Similarly, some forms require only two nucleic acids; again, they can be incorporated into one or two expression vectors.

[0593] As is known in the art, nucleic acids encoding the components of the present invention can be incorporated into expression vectors as is known in the art, and depend on the host cell used to generate the heterodimeric Fc fusion protein of the present invention. Generally, nucleic acids are operatively linked to any number of regulatory components (promoters, origins of replication, selection markers, ribosome binding sites, inducers, etc.). The expression vector can be extrachromosomal or integrated.

[0594] The nucleic acids and / or expression vectors of the present invention are then converted into any number of different types of host cells known in the art, including mammalian, bacterial, yeast, insect and / or fungal cells, and mammalian cells (e.g., CHO cells) can be used in many embodiments.

[0595] Depending on the form, nucleic acids encoding various monomers are typically contained in a single expression vector, which is usually controlled by different or the same promoter. In embodiments specific to the purpose of this invention, each of the two or three nucleic acids is contained in a different expression vector.

[0596] The heterodimeric Fc fusion protein of the present invention is prepared by culturing host cells containing expression vectors known in the art. Once prepared, conventional fusion protein or antibody purification steps, including ion exchange chromatography, can be performed. As described herein, two monomers having a pI difference of at least 0.5 can be separated by ion exchange chromatography, isoelectric focusing, or other methods sensitive to the isoelectric point. That is, pI substitutions are incorporated to alter the isoelectric point (pI) of each monomer, such that each monomer has a different pI, and the heterodimer also has a different pI, thereby facilitating isoelectric purification of the heterodimer (e.g., anion exchange columns, cation exchange columns). These substitutions also help identify and monitor any contaminating homodimers after purification (e.g., IEF gels, cIEF, and analytical IEX columns).

[0597] VIII. Bispecific immune checkpoint antibody x IL-15 / IL-15Rα heterodimer immunomodulatory fusion

[0598] Biological and biochemical functions of proteins

[0599] Generally, the Fc fusion protein of the present invention is administered to cancer patients in various ways described herein, and its efficacy is evaluated. Therefore, while standard efficacy assays such as cancer burden, tumor size, assessment of the presence and extent of metastasis can be performed, immuno-oncology therapy can also be evaluated based on immune status assessment results. This can be done in various ways, including in vitro and in vivo assays. For example, changes in immune status (e.g., assessing the presence of ICOS+CD4+ T cells after IPI treatment) and "conventional" measurements such as tumor burden, size, invasiveness, LN involvement, metastasis, etc., can be assessed. Therefore, any or all of the following can be evaluated: the fusion protein's effect on CD4+... + T cell activation or proliferation, CD8 + T(CTL) cell activa...

Claims

1. A composition comprising a variant IL-15 protein, wherein, In the variant IL-15 protein, compared to SEQ ID NO:2, N at position 71 is replaced with Q, N at position 79 is replaced with Q, and S at position 114 is replaced with A.

2. A composition comprising a variant IL-15 protein, wherein, In the variant IL-15 protein, compared to SEQ ID NO:2, N at position 71 is replaced with Q, N at position 79 is replaced with Q, and S at position 114 is missing.

3. A composition comprising a variant IL-15 protein, wherein, In the variant IL-15 protein, compared to SEQ ID NO:2, N at position 71 is replaced with Q, N at position 79 is replaced with Q, and N at position 112 is replaced with Q.

4. A composition comprising a variant IL-15 protein, wherein, In the variant IL-15 protein, compared to SEQ ID NO:2, N at position 71 is replaced with Q, N at position 79 is replaced with Q, S at position 114 is replaced with A, D at position 30 is replaced with N, and N at position 65 is replaced with D.

5. A composition comprising a variant IL-15 protein, wherein, In the variant IL-15 protein, compared to SEQ ID NO:2, N at position 71 is replaced with Q, N at position 79 is replaced with Q, S at position 114 is deleted, D at position 30 is replaced with N, and N at position 65 is replaced with D.

6. A composition comprising a variant IL-15 protein, wherein, In the variant IL-15 protein, compared to SEQ ID NO:2, N at position 71 is replaced with Q, N at position 79 is replaced with Q, N at position 112 is replaced with Q, D at position 30 is replaced with N, and N at position 65 is replaced with D.

7. A composition comprising a variant IL-15 protein, wherein, In the variant IL-15 protein, compared to SEQ ID NO:2, N at position 71 is replaced with Q, N at position 79 is replaced with Q, S at position 114 is replaced with A, D at position 30 is replaced with N, E at position 64 is replaced with Q, and N at position 65 is replaced with D.

8. A composition comprising a variant IL-15 protein, wherein, In the variant IL-15 protein, compared to SEQ ID NO:2, N at position 71 is replaced with Q, N at position 79 is replaced with Q, S at position 114 is deleted, D at position 30 is replaced with N, E at position 64 is replaced with Q, and N at position 65 is replaced with D.

9. A composition comprising a variant IL-15 protein, wherein, In the variant IL-15 protein, compared to SEQ ID NO:2, N at position 71 is replaced with Q, N at position 79 is replaced with Q, N at position 112 is replaced with Q, D at position 30 is replaced with N, E at position 64 is replaced with Q, and N at position 65 is replaced with D.

10. The composition according to claim 9, wherein the amino acid sequence of the variant IL-15 protein is composed of SEQ ID NO:

319.

11. A nucleic acid composition comprising a nucleic acid molecule encoding a variant IL-15 protein according to any one of claims 1 to 10.

12. An expression vector comprising the nucleic acid molecule according to claim 11.

13. A host cell comprising the nucleic acid molecule of claim 11 or the expression vector of claim 12.

14. A method for preparing a composition comprising a variant IL-15 protein, the method comprising culturing a host cell according to claim 13 under conditions in which the variant IL-15 protein is produced and recovering the variant IL-15 protein.

15. A pharmaceutical composition comprising the composition comprising the variant IL-15 protein according to any one of claims 1 to 10 and a pharmaceutical carrier.

16. Use of the variant IL-15 protein according to any one of claims 1 to 10 for the treatment of cancer in a subject with such need, wherein said cancer is breast cancer or colon cancer.

17. The use according to claim 16, wherein the composition comprising the variant IL-15 protein is formulated for administration in combination with a therapeutically effective amount of a checkpoint blocking antibody, wherein the checkpoint blocking antibody is an anti-PD-1 antibody or an anti-PD-L1 antibody.

18. Use of the pharmaceutical composition according to claim 15 for preparing a medicament for treating cancer in a subject with such need, wherein the cancer is breast cancer or colon cancer.

19. The use according to claim 18, wherein the pharmaceutical composition is formulated for administration in combination with a therapeutically effective amount of a checkpoint blocking antibody, wherein the checkpoint blocking antibody is an anti-PD-1 antibody or an anti-PD-L1 antibody.

20. The use according to claim 17 or 19, wherein the checkpoint blocking antibody is nivolumab or pembrolizumab.

Images