Bispecific anti-MUC16 x anti-CD28 antibodies and uses thereof

Abstract

To provide bispecific antigen-binding molecules comprising a first antigen-binding domain that specifically binds human CD28, and a second antigen-binding molecule that specifically binds human MUC16.SOLUTION: Bispecific antigen-binding molecules of the present invention are capable of inhibiting the growth of tumors expressing MUC16, such as ovarian tumors. The antibodies and bispecific antigen-binding molecules of the invention are useful for the treatment of diseases and disorders in which an up-regulated or induced targeted immune response is desired and / or therapeutically beneficial.SELECTED DRAWING: None

Description

[Technical Field]

[0001] Related Applications This application is related to and claims priority to U.S. Provisional Patent Application No. 62 / 782,142, filed December 19, 2018, and U.S. Provisional Patent Application No. 62 / 815,861, filed March 8, 2019, the entire contents of which are expressly incorporated herein by reference.

[0002] Sequence Listing This application contains a Sequence Listing that has been submitted electronically in ASCII format and is incorporated herein by reference in its entirety. The above ASCII copy, created on December 18, 2019, is titled 10493WO01_118003_49320_SeqLst.txt and is 38,372 bytes in size.

[0003] The present invention also relates to bispecific antigen-binding molecules that bind to target molecules such as CD28 and MUC16, and methods of their use. [Background technology]

[0004] CD28 is a type I transmembrane protein with a single extracellular IgV-like domain assembled as a homodimer and expressed on the surface of T cells. CD28 is a receptor for the CD80 (B7.1) and CD86 (B7.2) proteins and is activated by CD80 or CD86 expressed on antigen-presenting cells (APCs). Binding of CD28 to CD80 or CD86 provides a costimulatory signal important for T cell activation and survival. T cell stimulation via CD28 in addition to the T cell receptor (TCR) provides a potent signal for the production of various interleukins. CD28 also enhances cellular signals, such as pathways controlled by the NFκB transcription factor, after TCR activation. CD28 costimulation is important for effective T cell activation, including T cell differentiation, proliferation, cytokine release, and cell death.

[0005] Anti-CD28 antibodies have been proposed for therapeutic purposes, including T cell activation. One particular anti-CD28 antibody, TGN1412 (an anti-CD28 superagonist), was used in clinical trials in 2006. Six healthy volunteers were intravenously administered TGN1412 at a dose of 0.1 mg / kg. Within two hours, all six patients developed a significant inflammatory response (cytokine storm), and all patients developed multiple organ failure within 16 hours. When the subjects were treated with corticosteroids, cytokine levels returned to normal within two to three days. The starting dose of 0.1 mg / kg in the phase 1 trial was based on 500 times the no-observed-adverse-effect level (NOAEL) of 50 mg / kg in cynomolgus monkeys (Non-Patent Document 1). Unfortunately, the cytokine storm induced by TGN1412 was not predicted in cynomolgus monkey toxicology studies or ex vivo human PBMC studies.

[0006] Mucin 16 (MUC16), also known as cancer antigen 125, cellular carcinoma antigen 125, carbohydrate antigen 125, or CA-125, is a highly glycosylated integral membrane glycoprotein. MUC16 contains three major domains: an extracellular N-terminal domain, a large tandem repeat domain interspersed with sea urchin sperm, enterokinase, and agrin (SEA) domains, and a carboxyl-terminal domain containing a segment of the transmembrane region and a short cytoplasmic tail. Proteolytic cleavage allows the extracellular portion of MUC16 to be released into the bloodstream. MUC16 is a cancer-associated protein involved in the development of ovarian, breast, pancreatic, non-small cell lung, and intrahepatic cholangiocarcinoma (ICC). MUC16 is overexpressed in cancers including uterine bladder cancer, cervical adenocarcinoma, and gastrointestinal adenocarcinoma, as well as in diseases and conditions including inflammatory bowel disease, liver cirrhosis, heart failure, peritoneal infection, and abdominal surgery (Non-Patent Document 2). Expression of MUC16 on cancer cells has been shown to protect tumor cells from the immune system (Non-Patent Document 3).

[0007] Methods for treating ovarian cancer using antibodies against MUC16 have been studied. However, the monoclonal antibodies oregovomab and abugovomab have had limited success (Non-Patent Document 4, supra). Therefore, there is a need in the art for improved MUC16 antibodies for treating cancer.

[0008] Furthermore, bispecific antigen binding molecules that bind to both CD28 and a target antigen, such as MUC16, are useful in therapeutic settings where specific targeting and T cell-mediated killing of tumor cells expressing the target antigen is desired. [Prior art documents] [Non-patent literature]

[0009] [Non-Patent Document 1] Suntharalingam, et al., Cytokine Storm in a Phase 1 Trial of the Anti-CD28 Monoclonal Antibody TGN1412, NEJM 355:1018-1028(2006) [Non-patent document 2] Haridas, D. et al., 2014, FASEB J., 28:4183-4199 [Non-patent document 3] Felder, M. et al., 2014, Molecular Cancer, 13:129 [Non-patent document 4] Felder, Das, S. and Batra, SK2015, Cancer Res. 75:4660-4674 Summary of the Invention

[0010] In a first aspect, the present invention provides bispecific antigen-binding molecules that bind to CD28 and MUC16, also referred to herein as "anti-CD28 / anti-MUC16 bispecific molecules." The anti-MUC16 portion of the anti-CD28 / anti-MUC16 bispecific molecule is useful for targeting tumor cells that express MUC16 (e.g., ovarian tumor cells), and the anti-CD28 portion of the bispecific molecule is useful for activating T cells. The simultaneous binding of MUC16 on tumor cells and CD28 on T cells promotes direct killing (cytolysis) of the targeted tumor cells by activated T cells. Thus, the anti-CD28 / anti-MUC16 bispecific molecules of the present invention are useful, inter alia, for treating diseases and disorders associated with or caused by MUC16-expressing tumors (e.g., ovarian cancer).

[0011] Bispecific antigen-binding molecules according to this aspect of the invention comprise a first antigen-binding domain that specifically binds human CD28 and a second antigen-binding domain that specifically binds MUC16. The invention includes anti-CD28 / anti-MUC16 bispecific molecules (e.g., bispecific antibodies) in which each antigen-binding domain comprises a heavy chain variable region (HCVR) paired with a light chain variable region (LCVR). In certain exemplary embodiments of the invention, the anti-CD28 antigen-binding domain and the anti-MUC16 antigen-binding domain each comprise a distinct and separate HCVR paired with a common LCVR.

[0012] The present invention provides anti-CD28 / anti-MUC16 bispecific molecules, wherein a first antigen-binding domain that specifically binds to CD28 comprises any of the HCVR amino acid sequences set forth in Table 3. The first antigen-binding domain that specifically binds to CD28 may also comprise any of the LCVR amino acid sequences set forth in Table 3. According to certain embodiments and the first antigen-binding domain that specifically binds to CD28 comprises any of the HCVR / LCVR amino acid sequence pairs listed in Table 3. The present invention also provides anti-CD28 / anti-MUC16 bispecific molecules, wherein the first antigen-binding domain that specifically binds to CD28 comprises any of the heavy chain CDR1-CDR2-CDR3 amino acid sequences listed in Table 3, and / or any of the light chain CDR1-CDR2-CDR3 amino acid sequences listed in Table 3.

[0013] According to certain embodiments, the present invention provides an anti-CD28 / anti-MUC16 bispecific molecule, wherein the first antigen-binding domain that specifically binds to CD28 comprises a heavy chain variable region (HCVR) having an amino acid sequence selected from the group consisting of SEQ ID NOs: 18 and 42, or a substantially similar sequence thereof having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

[0014] The present invention also provides anti-CD28 / anti-MUC16 bispecific molecules, wherein the first antigen-binding domain that specifically binds to CD28 comprises a light chain variable region (LCVR) having an amino acid sequence selected from the group consisting of SEQ ID NOs: 10 and 34, or a substantially similar sequence thereof having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

[0015] The present invention also provides anti-CD28 / anti-MUC16 bispecific molecules, wherein a first antigen-binding domain that specifically binds to CD28 comprises a HCVR and LCVR (HCVR / LCVR) amino acid sequence pair selected from the group consisting of SEQ ID NOs: 18 / 10 and 42 / 34.

[0016] The present invention also provides an anti-CD28 / anti-MUC16 bispecific molecule, wherein the first antigen-binding domain that specifically binds to CD28 comprises a heavy chain CDR3 (HCDR3) domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 24 and 48, or a substantially similar sequence thereof with at least 90%, at least 95%, at least 98%, or at least 99% sequence identity, and a light chain CDR3 (LCDR3) domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 16 and 40, or a substantially similar sequence thereof with at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

[0017] In certain embodiments, the first antigen-binding domain that specifically binds to CD28 comprises an HCDR3 / LCDR3 amino acid sequence pair selected from the group consisting of SEQ ID NOs: 24 / 16 and 48 / 40.

[0018] The present invention also provides an anti-CD28 / anti-MUC16 bispecific antigen binding molecule, wherein the first antigen-binding domain that specifically binds to CD28 comprises a heavy chain CDR1 (HCDR1) having an amino acid sequence selected from the group consisting of SEQ ID NOs: 20 and 44, or a substantially similar sequence thereof having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity, a heavy chain CDR2 (HCDR2) domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 22 and 46, or a substantially similar sequence thereof having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity. a light chain CDR1 (LCDR1) domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 12 and 36, or a substantially similar sequence thereof with at least 90%, at least 95%, at least 98%, or at least 99% sequence identity, and a light chain CDR2 (LCDR2) domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 14 and 38, or a substantially similar sequence thereof with at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

[0019] Certain non-limiting exemplary anti-CD28 / anti-MUC16 bispecific antigen binding molecules of the present invention comprise a first antigen binding domain that specifically binds to CD28, comprising HCDR1-HCDR2-HCDR3-LCDR1-LCDR2-LCDR3 domains having amino acid sequences selected from the group consisting of SEQ ID NOs: 20-22-24-12-14-16 and 44-46-48-36-38-40, respectively.

[0020] The present invention also provides anti-CD28 / anti-MUC16 bispecific molecules, wherein the second antigen-binding domain that specifically binds MUC16 comprises a heavy chain variable region (HCVR) having an amino acid sequence selected from the group consisting of SEQ ID NOs: 2 and 26, or a substantially similar sequence thereof having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

[0021] The present invention also provides anti-CD28 / anti-MUC16 bispecific molecules, wherein the second antigen-binding domain that specifically binds MUC16 comprises a light chain variable region (LCVR) having an amino acid sequence selected from the group consisting of SEQ ID NOs: 10 and 34, or a substantially similar sequence thereof having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

[0022] The present invention also provides anti-CD28 / anti-MUC16 bispecific molecules, wherein the second antigen-binding domain that specifically binds to MUC16 comprises a HCVR and LCVR (HCVR / LCVR) amino acid sequence pair selected from the group consisting of SEQ ID NOs: 2 / 10 and 26 / 34.

[0023] The present invention also provides anti-CD28 / anti-MUC16 bispecific molecules, wherein the second antigen-binding domain that specifically binds MUC16 comprises a heavy chain CDR3 (HCDR3) domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 8 and 32, or a substantially similar sequence thereof with at least 90%, at least 95%, at least 98%, or at least 99% sequence identity, and a light chain CDR3 (LCDR3) domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 16 and 40, or a substantially similar sequence thereof with at least 90%, at least 95%, at least 98%, or at least 99% sequence identity.

[0024] In certain embodiments, the second antigen-binding domain that specifically binds to MUC16 comprises an HCDR3 / LCDR3 amino acid sequence pair selected from the group consisting of SEQ ID NOs: 8 / 16 and 32 / 40.

[0025] The present invention also provides an anti-CD28 / anti-MUC16 bispecific antigen binding molecule, wherein the second antigen binding domain that specifically binds to MUC16 is a heavy chain CDR1 (HCDR1) having an amino acid sequence selected from the group consisting of SEQ ID NOs: 4 and 28, or a substantially similar sequence thereof with at least 90%, at least 95%, at least 98%, or at least 99% sequence identity, a heavy chain CDR2 (HCDR2) domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 6 and 30, or a substantially similar sequence thereof with at least 90%, at least 95%, at least 98%, or at least 99% sequence identity. a light chain CDR1 (LCDR1) domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 12 and 36, or a substantially similar sequence thereof having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity; and a light chain CDR2 (LCDR2) domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 14 and 38, or a substantially similar sequence thereof having at least 90%, at least 95%, at least 98%, or at least 99% sequence identity. Includes:

[0026] Certain non-limiting exemplary anti-CD28 / anti-MUC16 bispecific antigen binding molecules of the invention comprise a second antigen binding domain that specifically binds MUC16, comprising HCDR1-HCDR2-HCDR3-LCDR1-LCDR2-LCDR3, each having an amino acid sequence selected from the group consisting of SEQ ID NOs: 4-6-8-12-14-16 and 28-30-32-36-38-40.

[0027] In a related embodiment, the invention comprises an anti-CD28 / anti-MUC16 bispecific antigen binding molecule, wherein the second antigen binding domain that specifically binds MUC16 comprises heavy and light chain CDR domains contained within heavy and light chain variable region (HCVR / LCVR) sequences selected from the group consisting of SEQ ID NOs: 2 / 10 and 26 / 34.

[0028] In another aspect, the invention provides nucleic acid molecules encoding any of the HCVR, LCVR, or CDR sequences of the anti-CD28 / anti-MUC16 bispecific antigen-binding molecules disclosed herein, including nucleic acid molecules comprising a polynucleotide sequence set forth in Table 2 and / or Table 4 herein, and nucleic acid molecules comprising two or more of the polynucleotide sequences set forth in Table 2 and / or Table 4, in any functional combination or arrangement thereof. Recombinant expression vectors harboring nucleic acids of the invention, and host cells into which such vectors have been introduced, are also encompassed by the invention, as are methods of producing the antibodies by culturing the host cells under conditions that allow for the production of the antibodies, and recovering the antibodies produced.

[0029] The present invention includes anti-CD28 / anti-MUC16 bispecific antigen binding molecules in which any of the aforementioned antigen binding domains that specifically bind CD28 is combined, linked, or otherwise associated with any of the aforementioned antigen binding domains that specifically bind MUC16 to form a bispecific antigen binding molecule that binds CD28 and MUC16.

[0030] The present invention includes anti-CD28 / anti-MUC16 bispecific antigen binding molecules with modified glycosylation patterns. In some applications, modifications to remove undesired glycosylation sites or antibodies lacking fucose moieties present on the oligosaccharide chains, for example, to enhance antibody-dependent cellular cytotoxicity (ADCC) function, may be useful (see Shield et al. (2002) JBC 277:26733). In other applications, modifications to galactosylation can be performed to modify complement-dependent cytotoxicity (CDC).

[0031] In another aspect, the invention provides a pharmaceutical composition comprising an anti-CD28 / anti-MUC16 bispecific antigen-binding molecule disclosed herein and a pharmaceutically acceptable carrier. In a related aspect, the invention features a composition that is a combination of an anti-CD28 / anti-MUC16 bispecific antigen-binding molecule and a second therapeutic agent. In one embodiment, the second therapeutic agent is any agent that is advantageously combined with the anti-CD28 / anti-MUC16 bispecific antigen-binding molecule. Exemplary agents that can be advantageously combined with the anti-CD28 / anti-MUC16 bispecific antigen-binding molecule are discussed in detail elsewhere herein.

[0032] In yet another aspect, the present invention provides a therapeutic method for targeting / killing MUC16-expressing tumor cells using the anti-CD28 / anti-MUC16 bispecific antigen binding molecule of the invention, the therapeutic method comprising administering to a subject in need thereof a therapeutically effective amount of a pharmaceutical composition comprising the anti-CD28 / anti-MUC16 bispecific antigen binding molecule of the invention.

[0033] The present invention also relates to the use of an anti-CD28 / anti-MUC16 bispecific antibody of the invention in the manufacture of a medicament for the treatment of a disease or disorder associated with or caused by MUC16 expression. This includes the use of specific antigen-binding molecules.

[0034] In yet another aspect, the invention provides therapeutic methods for targeting / killing MUC16-expressing tumor cells using the anti-CD28 / anti-MUC16 bispecific antigen binding molecules of the invention, where the anti-CD28 / anti-MUC16 bispecific antigen binding molecule is combined with another anti-tumor bispecific antigen binding molecule that binds CD3 (e.g., anti-CD28 / anti-MUC16 combined with an anti-CD3 / anti-MUC16 antibody).

[0035] In yet another aspect, the present invention provides therapeutic methods for targeting / killing MUC16-expressing tumor cells using the anti-CD28 / anti-MUC16 bispecific antigen binding molecules of the invention, in which the anti-CD28 / anti-MUC16 bispecific antigen binding molecules are combined with checkpoint inhibitors that target, for example, PD-1, PD-L1, or CTLA-4 (e.g., anti-CD28 / anti-MUC16 are combined with an anti-PD-1 antibody). In certain embodiments, it is contemplated that the anti-CD28 / anti-MUC16 antibodies of the invention may be combined with agents that target PD-1, such as pembrolizumab (Keytruda®), nivolumab (Opdivo®), or cemiplimab (Libtayo®). In certain embodiments, it is contemplated that the anti-CD28 / anti-MUC16 antibodies of the invention may be combined with an agent that targets PD-L1, such as atezolizumab (Tecentriq®), avelumab (Bavencio®), or durvalumab (Imfinzi®). In certain embodiments, it is contemplated that the anti-CD28 / anti-MUC16 antibodies of the invention may be combined with an agent that targets CTLA-4, such as ipilimumab (Yervoy®).

[0036] In yet another aspect, the invention provides therapeutic methods for targeting / killing MUC16-expressing tumor cells using the anti-CD28 / anti-MUC16 bispecific antigen binding molecules of the invention, in combination with other anti-tumor bispecific antigen binding molecules that bind CD3 (e.g., anti-CD28 / anti-MUC16 combined with an anti-CD3 / anti-MUC16 bispecific antibody) and checkpoint inhibitors that target PD-1, PDL-1, or CTLA-4 (e.g., anti-CD28 / anti-MUC16 combined with an anti-PD-1 antibody).

[0037] Other embodiments will become apparent from review of the following detailed description. [Brief explanation of the drawings]

[0038] [Figure 1] Figure 1 shows a graph demonstrating tumor growth inhibition in engineered cell lines transduced with costimulatory ligand expression. Three tumor cell lines, B16F10.9, EL4, and MC38, were engineered to express costimulatory ligand, GFP, or an empty vector as a control. Engineered tumor cells were injected into C57BL / 6 mice. Data represent the mean ± SEM. Data represent at least one experiment with five mice per group. The graph shows tumor growth as a percentage of control, calculated as follows: [Table 1] [Figure 2A] Figures 2A-2I are schematics and graphs showing that an exemplary anti-MUC16xCD28 antibody of the invention enhances T cell activity in the presence of anti-MUC16xCD3 and TCR stimulation with a cancer cell line (PEO1) harboring endogenous MUC16. Figures 2B-2E are graphs showing data from human PBMCs. Figures 2F-2H are graphs showing data from cynomolgus monkey PBMCs. Human T cells (as in Figures 2B-2E) were cultured with cancer target cells with endogenous MUC16 expression (ovarian cancer line PEO-1) and the indicated bispecific antibodies for 96 hours. Figure 2A is a schematic diagram of the assay setup. [Figure 2B] Figures 2A-2I are schematics and graphs showing that an exemplary anti-MUC16xCD28 antibody of the invention enhances T cell activity in the presence of anti-MUC16xCD3 and TCR stimulation with a cancer cell line (PEO1) harboring endogenous MUC16. Figures 2B-2E are graphs showing data from human PBMCs. Figures 2F-2H are graphs showing data from cynomolgus monkey PBMCs. Human T cells (as in Figures 2B-2E) were cultured with cancer target cells (ovarian cancer line PEO-1) with endogenous MUC16 expression and the indicated bispecific antibodies for 96 hours. Figure 2B is a graph showing tumor cell killing. The Y-axis values ​​refer to the percentage of surviving PEO1 cells. [Figure 2C] Figures 2A-2I are schematics and graphs showing that an exemplary anti-MUC16xCD28 antibody of the invention enhances T cell activity in the presence of anti-MUC16xCD3 and TCR stimulation with a cancer cell line (PEO1) harboring endogenous MUC16. Figures 2B-2E are graphs showing data from human PBMCs. Figures 2F-2H are graphs showing data from cynomolgus monkey PBMCs. Human T cells (as in Figures 2B-2E) were cultured with cancer target cells with endogenous MUC16 expression (ovarian cancer line PEO-1) and the indicated bispecific antibodies for 96 hours. Figure 2C is a graph showing IFNγ release. [Figure 2D] Figures 2A-2I are schematics and graphs showing that an exemplary anti-MUC16xCD28 antibody of the invention enhances T cell activity in the presence of anti-MUC16xCD3 and TCR stimulation with a cancer cell line (PEO1) harboring endogenous MUC16. Figures 2B-2E are graphs showing data from human PBMCs. Figures 2F-2H are graphs showing data from cynomolgus monkey PBMCs. Human T cells (as in Figures 2B-2E) were cultured with cancer target cells (ovarian cancer line PEO-1) with endogenous MUC16 expression and the indicated bispecific antibodies for 96 hours. Figure 2D is a graph showing the number of CD4 T cells and the frequency of CD25+ cells, expressed as a percentage of CD25+ cells among CD4 T cells. [Figure 2E]Figures 2A-2I are schematics and graphs showing that an exemplary anti-MUC16xCD28 antibody of the invention enhances T cell activity in the presence of anti-MUC16xCD3 and TCR stimulation with a cancer cell line (PEO1) harboring endogenous MUC16. Figures 2B-2E are graphs showing data from human PBMCs. Figures 2F-2H are graphs showing data from cynomolgus monkey PBMCs. Human T cells (as in Figures 2B-2E) were cultured with cancer target cells (ovarian cancer line PEO-1) with endogenous MUC16 expression and the indicated bispecific antibodies for 96 hours. Figure 2E is a graph showing the number of CD8 T cells and the frequency of CD25+ cells, expressed as a percentage of CD25+ cells among CD8 T cells. [Figure 2F] Figures 2A-2I are schematics and graphs showing that an exemplary anti-MUC16xCD28 antibody of the invention enhances T cell activity in the presence of anti-MUC16xCD3 and TCR stimulation with a cancer cell line (PEO1) harboring endogenous MUC16. Figures 2B-2E are graphs showing data from human PBMCs. Figures 2F-2H are graphs showing data from cynomolgus monkey PBMCs. Human T cells (as in Figures 2B-2E) were cultured with cancer target cells (ovarian cancer line PEO-1) with endogenous MUC16 expression and the indicated bispecific antibodies for 96 hours. Figure 2F is a graph showing tumor cell killing. The Y-axis values ​​refer to the percentage of surviving PEO1 cells. [Figure 2G] Figures 2A-2I are schematics and graphs showing that an exemplary anti-MUC16xCD28 antibody of the invention enhances T cell activity in the presence of anti-MUC16xCD3 and TCR stimulation with a cancer cell line (PEO1) harboring endogenous MUC16. Figures 2B-2E are graphs showing data from human PBMCs. Figures 2F-2H are graphs showing data from cynomolgus monkey PBMCs. Human T cells (as in Figures 2B-2E) were cultured with cancer target cells (ovarian cancer line PEO-1) with endogenous MUC16 expression and the indicated bispecific antibodies for 96 hours. Figure 2G is a graph showing the number of CD4 T cells and the frequency of CD25+ cells, expressed as a percentage of CD25+ cells among CD4 T cells. [Figure 2H]Figures 2A-2I are schematics and graphs showing that an exemplary anti-MUC16xCD28 antibody of the invention enhances T cell activity in the presence of anti-MUC16xCD3 and TCR stimulation with a cancer cell line (PEO1) harboring endogenous MUC16. Figures 2B-2E are graphs showing data from human PBMCs. Figures 2F-2H are graphs showing data from cynomolgus monkey PBMCs. Human T cells (as in Figures 2B-2E) were cultured with cancer target cells (ovarian cancer line PEO-1) with endogenous MUC16 expression and the indicated bispecific antibodies for 96 hours. Figure 2H is a graph showing the number of CD4 and CD8 T cells, as well as the frequency of CD25+ cells, expressed as a percentage of CD4 and CD8 T cells. [Figure 2I] Figures 2A-2I are schematics and graphs showing that an exemplary anti-MUC16xCD28 antibody of the invention enhances T cell activity in the presence of anti-MUC16xCD3 and TCR stimulation with a cancer cell line (PEO1) harboring endogenous MUC16. Figures 2B-2E are graphs showing data from human PBMCs. Figures 2F-2H are graphs showing data from cynomolgus monkey PBMCs. Human T cells (as in Figures 2B-2E) were cultured with cancer target cells with endogenous MUC16 expression (ovarian cancer line PEO-1) and the indicated bispecific antibodies for 96 hours. Figure 2I provides a graph showing antibody binding to cellular targets as measured by flow cytometry. [Figure 3A]Figures 3A-3C are graphs showing that an exemplary anti-MUC16xCD28 bispecific antibody of the invention enhances anti-tumor immunity through anti-MUC16xCD3-induced T cell activation. Figure 3A is a graph showing tumor burden over time as measured by mean radiance (mean radiance [p / s / cm2 / sr]). Values ​​represent group median and range. p-values ​​were calculated using the Mann-Whitney test for each time point. * indicates p<0.05 or ** indicates p<0.01 for the comparison of MUC16xCD3 vs. EGFRvIIIxCD3. ## indicates p<0.01 for the comparison of MUC16xCD3 + MUC16xCD28 vs. EGFRvIIIxCD3. NSG mice engrafted with human PBMCs were implanted with OVCAR3-Luc via intraperitoneal injection. Mice were dosed IV on days 5 and 8 (arrows). Mice received either 2.5 μg MUC16×CD3 or 2.5 μg EGFRvIII×CD3. Some mice also received 100 μg MUC16×CD28. Tumor burden was assessed at days 4, 8, 12, 15, 20, and 25 after tumor implantation by monitoring bioluminescence over time. N=5 mice / group. [Figure 3B] Figures 3A-3C are graphs showing that exemplary anti-MUC16xCD28 bispecific antibodies of the invention enhance anti-tumor immunity through anti-MUC16xCD3-induced T cell activation. Figure 3B provides a graph showing serum cytokine levels from blood obtained 4 hours after the first dose from the same experiment as shown in Figure 3A. p values ​​were calculated by one-way ANOVA. For the comparison of MUC16xCD3 + MUC16xCD28 vs. EGFRvIIIxCD3, ## indicates p<0.01, or #### indicates p<0.0001. For the comparison of MUC16xCD3 + MUC16xCD28 vs. MUC16xCD3, @@@ indicates p<0.005. For the comparison between MUC16×CD3+MUC16×CD28 and EGFRvIII×CD3+MUC16×CD28, ^ indicates p<0.01, ^^ indicates p<0.005, and ^^^^ indicates P<0.0001. [Figure 3C]Figures 3A-3C are graphs showing that an exemplary anti-MUC16xCD28 bispecific antibody of the invention enhances anti-tumor immunity through anti-MUC16xCD3-induced T cell activation. Figure 3C provides a graph showing the correlation between tumor burden and serum CA-125 levels at day 26. N=5 mice per group from the same experiment as shown in Figure 3A. [Figure 4A] Graphs show tumor burden over time as measured by mean radiance (mean radiance [p / s / cm2 / sr]). Values ​​represent group median and range. p-values ​​were calculated using the Mann-Whitney test for each time point. ** indicates p<0.01 for MUC16xCD3 vs. EGFRvIIIxCD3. ## indicates p<0.01 for MUC16xCD3 + MUC16xCD28 vs. EGFRvIIIxCD3. @ indicates p<0.05 for MUC16xCD3 + MUC16xCD28 vs. MUC16xCD3. NSG mice engrafted with human PBMCs were transplanted with OVCAR3-Luc via intraperitoneal injection. Mice were treated IV with 0.5 mg / kg MUC16xCD3 or 0.5 mg / kg EGFRvIIIxCD3. Some mice also received 0.2 mg / kg MUC16xCD28 on days 5 and 8 (arrows). Tumor burden was assessed by BL on days 4, 8, 11, 14, 21, 28, and 34 by monitoring bioluminescence over time. N=5 or 6 mice / group. [Figure 4B]Graphs are provided showing serum cytokine levels from blood obtained 4 hours after the first dose from the same experiment as shown in FIG. 4A. p-values ​​were calculated by one-way ANOVA. * indicates p<0.05 for comparisons between MUC16×CD3 and EGFRvIII×CD3; ## indicates p<0.01, or ### indicates p<0.001, or #### indicates p<0.0001 for comparisons between MUC16×CD3+MUC16×CD28 and EGFRvIII×CD3; @ indicates p<0.05, or @@@@ indicates p<0.0001 for comparisons between MUC16×CD3+MUC16×CD28 and MUC16×CD3. For the comparison of MUC16xCD3+MUC16xCD28 with EGFRvIIIxCD3+MUC16xCD28, ^^ denotes p<0.001, or ^^^ denotes P<0.001, or ^^^ denotes P<0.0001. [Figure 5] Figure 1 shows a graph of survival over time. ID8-VEGF / hMUC16 cells were implanted into the peritoneal cavity of mice humanized for hCD3 / hCD28 / hMUC16. As indicated by the arrows, mice were treated intravenously with EGFRvIIIxCD3 or MUC16xCD3 at 1 mg / kg or on days 3, 6, and 10 after tumor implantation. Some mice also received 1 mg / kg MUC16xCD28. p values ​​were calculated using the Mantel-Cox test for each time point. For the comparison of MUC16xCD3 vs. EGFRvIIIxCD3, ** indicates p<0.01. For the comparison of MUC16xCD3 + MUC16xCD28 vs. EGFRvIIIxCD3, ## indicates p<0.01. For the comparison of MUC16xCD3 + MUC16xCD28 vs. MUC16xCD3, @ indicates p<0.05. [Figure 6A]Figure 1 is a graph showing tumor volume over time. MC38 / hMUC16 tumor cells were implanted subcutaneously into hCD3 / hMUC16 humanized mice. Mice were treated with 0.01 mg / kg anti-MUC16xCD3, 0.5 mg / kg of an exemplary anti-MUC16xmCD28 bispecific antibody of the invention, as indicated (arrows), twice weekly starting on day 0. Tumor volume was monitored by caliper measurement over time. Values ​​shown are mean ± SEM. Data are representative of three experiments. N=7 per group. p-values ​​were calculated using two-way ANOVA compared to isotype control (**, p<0.01, and ****, p<0.0001 for MUC16xCD3+MUC16xmCD28 vs. isotype control; #, p<0.05 for MUC16xCD3 vs. isotype control; $, p<0.05 for MUC16xmCD28 vs. isotype control). [Figure 6B] 6A provides a graph showing serum cytokine levels from blood taken at the indicated time points from the same experiment as shown in FIG. 6A. [Figure 6C] Figure 1 shows graphs showing cytokine levels. Mice were bled for serum cytokines 4 hours post-dose on day 7. Statistical significance was calculated using one-way ANOVA compared to isotype, **p<0.01 and ****p<0.0001. n=7 mice / group. Data are representative of 3 experiments. [Figure 6D] Figure 1 shows graphs showing cytokine levels. Mice were bled for serum cytokines 4 hours post-dose on day 7. Statistical significance was calculated using one-way ANOVA compared to isotype, **p<0.01 and ****p<0.0001. n=7 mice / group. Data are representative of 3 experiments. [Figure 7] 1 is a graph showing that immobilization of MUC16xCD28 on assay plates using dry or wet coating methods does not induce T cell activation in the absence of CD3 stimulation, in contrast to CD28 superagonists. [Figure 8A]Figures 8A and 8B show that MUC16xCD28, alone or in combination, does not induce systemic T cell activation. Cynomolgus monkeys received a single dose of bispecific antibody at either 1 or 10 mg / kg (indicated in parentheses). Additional groups received a total of four doses, as indicated, as repeat administration. Blood was collected at the indicated times (hours) post-dose. Figure 8A shows serum cytokines, Figure 8B shows relative T cell counts, and Figure 8C shows the frequency of Ki67+ and ICOS+ T cells (% of CD3). Data represent mean ± SEM. N = 3 animals per group. p values ​​were calculated by two-way ANOVA compared to isotype control. (**, p < 0.01, ***, p < 0.001, and ****, p < 0.0001). [Figure 8B] Figures 8A and 8B show that MUC16xCD28, alone or in combination, does not induce systemic T cell activation. Cynomolgus monkeys received a single dose of bispecific antibody at either 1 or 10 mg / kg (indicated in parentheses). Additional groups received a total of four doses, as indicated, as repeat administration. Blood was collected at the indicated times (hours) post-dose. Figure 8A shows serum cytokines, Figure 8B shows relative T cell counts, and Figure 8C shows the frequency of Ki67+ and ICOS+ T cells (% of CD3). Data represent mean ± SEM. N = 3 animals per group. p values ​​were calculated by two-way ANOVA compared to isotype control. (**, p < 0.01, ***, p < 0.001, and ****, p < 0.0001). [Figure 8C] Figures 8A and 8B show that MUC16xCD28, alone or in combination, does not induce systemic T cell activation. Cynomolgus monkeys received a single dose of bispecific antibody at either 1 or 10 mg / kg (indicated in parentheses). Additional groups received a total of four doses, as indicated, as repeat administration. Blood was collected at the indicated times (hours) post-dose. Figure 8A shows serum cytokines, Figure 8B shows relative T cell counts, and Figure 8C shows the frequency of Ki67+ and ICOS+ T cells (% of CD3). Data represent mean ± SEM. N = 3 animals per group. p values ​​were calculated by two-way ANOVA compared to isotype control. (**, p < 0.01, ***, p < 0.001, and ****, p < 0.0001). [Figure 9A]We show that MUCxCD28 and MUC16xCD3 bispecific antibodies can bind to MUC-expressing cells in the presence of soluble CA-125. OVCAR-3 cells were incubated with 8 nM of the indicated antibody labeled with Alexa647 in the presence of increasing concentrations of soluble CA-125 (Figure 9A) or MUC16 clumps (Figure 9B) in flow cytometry buffer (PBS + 1% FBS) for 30 min at 4°C. After incubation, cells were washed with flow cytometry buffer and analyzed by flow cytometry. [Figure 9B] We show that MUCxCD28 and MUC16xCD3 bispecific antibodies can bind to MUC-expressing cells in the presence of soluble CA-125. OVCAR-3 cells were incubated with 8 nM of the indicated antibody labeled with Alexa647 in the presence of increasing concentrations of soluble CA-125 (Figure 9A) or MUC16 clumps (Figure 9B) in flow cytometry buffer (PBS + 1% FBS) for 30 min at 4°C. After incubation, cells were washed with flow cytometry buffer and analyzed by flow cytometry. [Figure 10] FIG. 1 is a schematic diagram of a T cell / antigen-presenting cell-based reporter bioassay. [Figure 11A]We show that bs24963D (also referred to as REGN5668) enhances NF-κB signaling in engineered T cells in the presence of stimulatory antigen-presenting cells expressing MUC16. Briefly, J.RT3.T3.5 / NF-κB-Luc / 1G4AB / hCD8αβ / hCD28 reporter cells were incubated with a range of concentrations (39 pM to 10 nM) of bs24963D or a CD28 non-crosslinking control bispecific antibody (non-TAA × CD28), including a no-antibody control, in the presence of 3T3 / hβ2M / HLA-A2 / NYESO1p / hMUC16 (Figure 11A) and 3T3 / hβ2M / HLA-A2 / NYESO1p cells (Figure 11B), at a reporter cell to stimulatory 3T3 cell ratio of 3.33:1. NF-κB signaling was detected as luciferase activity and measured by quantification of the luminescent signal, which is reported as relative light units (RLU). Data from assays performed in duplicate wells are plotted as mean ± SD. [Figure 11B] We show that bs24963D (also referred to as REGN5668) enhances NF-κB signaling in engineered T cells in the presence of stimulatory antigen-presenting cells expressing MUC16. Briefly, J.RT3.T3.5 / NF-κB-Luc / 1G4AB / hCD8αβ / hCD28 reporter cells were incubated with a range of concentrations (39 pM to 10 nM) of bs24963D or a CD28 non-crosslinking control bispecific antibody (non-TAA × CD28), including a no-antibody control, in the presence of 3T3 / hβ2M / HLA-A2 / NYESO1p / hMUC16 (Figure 11A) and 3T3 / hβ2M / HLA-A2 / NYESO1p cells (Figure 11B), at a reporter cell to stimulatory 3T3 cell ratio of 3.33:1. NF-κB signaling was detected as luciferase activity and measured by quantification of the luminescent signal, which is reported as relative light units (RLU). Data from assays performed in duplicate wells are plotted as mean ± SD. [Figure 12]We show that bs24963D (also referred to as REGN5668) mediates concentration-dependent IL-2 release from human primary T cells in the presence of REGN4018 (see WO2017 / 053856A1, BSMUC16 / CD3-001, which is REGN4018) using OVCAR-3 and PEO1 target cells. Briefly, enriched human primary T cells were incubated with a range of concentrations (7.6 pM to 500 nM) of bs24963D or CD28 non-crosslinking control bispecific antibody (non-TAA x CD28), including a no-antibody control, in the presence of a fixed concentration (5 nM) of either REGN4018 or a CD3 non-crosslinking control bispecific antibody (non-TAA x CD3) with the human ovarian cancer cell lines OVCAR-3 or PEO1 at effector-to-target cell ratios of 10:1 or 4:1, respectively. Data are from assays performed in triplicate wells and are plotted as mean ± SD. IL-2 release was measured using a human IL-2 immunoassay according to the manufacturer's protocol. [Figure 13] We show that bs24963D (also referred to as REGN5668) mediates concentration-dependent enhancement of human primary T cell proliferation in the presence of REGN4018 using OVCAR-3 and PEO1 target cells. Briefly, enriched human primary T cells were incubated with a range of concentrations (7.6 pM to 500 nM) of bs24963D or CD28 non-crosslinking control bispecific antibody (non-TAA x CD28), including a no-antibody control, in the presence of either a fixed concentration (5 nM) of REGN4018 or a CD3 non-crosslinking control bispecific antibody (non-TAA x CD3) at an effector-to-target cell ratio of 10:1 or 4:1, respectively, with the human ovarian cancer cell lines OVCAR-3 and PEO1. Data are from assays performed in triplicate wells and are plotted as mean ± SD. T cell proliferation was measured via detection of tritium decay (from tritiated thymidine incorporated into dividing cells) and reported as CPM. [Figure 14]These results demonstrate that bs24963D (also referred to as REGN5668) mediates concentration-dependent IL-2 release, and the addition of cemiplimab modestly increases IL-2 release from human primary T cells associated with SW1990 and SW1990 / hPD-L1 target cells. Briefly, enriched human primary T cells were incubated with a range of concentrations (7.6 pM to 500 nM) of bs24963D or a CD28 non-crosslinking control bispecific antibody (non-TAA x CD28), including a no-antibody control, in the presence of a fixed concentration (20 nM) of cemiplimab or an IgG4P control at a 2:1 effector-to-target cell ratio with SW1990 and SW1990 / hPD-L1 human pancreatic cancer cell lines. Data are from assays performed in triplicate wells and are plotted as mean ± SD. IL-2 release was measured using a human IL-2 immunoassay according to the manufacturer's protocol. Statistical analysis was performed using two-way ANOVA. Differences were considered statistically significant when p<0.05. bs24963D + cemiplimab demonstrated a statistically significant increase (p<0.0001) in IL-2 release compared to the REGN5668 + IgG4P control in SW1990 / hPD-L1 cells. [Figure 15]These results demonstrate that bs24963D (also referred to as REGN5668) mediates a concentration-dependent enhancement of proliferation, and the addition of cemiplimab modestly increases proliferation of human primary T cells with SW1990 and SW1990 / hPD-L1. Briefly, enriched human primary T cells were incubated with a range of concentrations (7.6 pM to 500 nM) of bs24963D or a CD28 non-crosslinking control bispecific antibody (non-TAA x CD28), including a no-antibody control, in the presence of a fixed concentration (20 nM) of cemiplimab or IgG4P control at a 2:1 effector-to-target cell ratio with SW1990 and SW1990 / hPD-L1 human pancreatic cancer cell lines. Data are from assays performed in triplicate wells and are plotted as mean ± SD. T cell proliferation was measured via detection of tritium decay (from tritiated thymidine incorporated into dividing cells) and reported as CPM. Statistical analysis was performed using two-way ANOVA. Differences were considered statistically significant when p<0.05. bs24963D + cemiplimab demonstrated a statistically significant increase in proliferation (p<0.0001) compared to the bs24963D + IgG4P control in SW1990 / hPD-L1 cells. DETAILED DESCRIPTION OF THE INVENTION

[0039] Before describing the present invention, it is to be understood that this invention is not limited to the particular methods and experimental conditions described, as such methods and conditions may vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present invention will be limited only by the appended claims.

[0040] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as that commonly understood by those skilled in the art to which this invention belongs.As used herein, the term "about" when used in relation to a specific listed value means that this value can vary by 1% or less from the listed value.For example, when used in the present invention, the expression "about 100" includes 99 and 101, and all values ​​therebetween (for example, 99.1, 99.2, 99.3, 99.4, etc.).

[0041] Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are now described. All patents, applications, and non-patent publications mentioned herein are incorporated by reference in their entirety.

[0042] definition As used herein, the term "CD28" refers to an antigen expressed on T cells as a costimulatory receptor. Human CD28 comprises the amino acid sequence set forth in SEQ ID NO: 50 and / or has the amino acid sequence set forth in NCBI accession number NP_006130.1. A human CD28 ectodomain (N19-P152) with a mouse Fc is set forth in SEQ ID NO: 52. A human CD28 ectodomain (N19-P152) with a myc-myc-his tag is set forth in SEQ ID NO: 53. All references to proteins, polypeptides, and protein fragments herein are intended to refer to the human version of the respective protein, polypeptide, or protein fragment unless specifically identified as being derived from a non-human species. Thus, the term "CD28" refers to human CD28 unless specifically identified as being derived from a non-human species, e.g., "mouse CD28," "monkey CD28," etc. The mouse CD28 (accession number NP_031668.3) ectodomain with a myc-myc-his tag is set forth in SEQ ID NO: 54.

[0043] As used herein, "antibodies that bind to CD28" or "anti-CD28 antibodies" include antibodies and antigen-binding fragments thereof that specifically recognize monomeric CD28, as well as antibodies and antigen-binding fragments thereof that specifically recognize dimeric CD28. The antibodies and antigen-binding fragments of the present invention can bind to soluble CD28 and / or cell surface-expressed CD28. Soluble CD28 includes the native CD28 protein, as well as membrane-expressed CD28. Included are recombinant CD28 protein variants, such as, for example, monomeric and dimeric CD28 constructs, that lack the transmembrane domain or that are not otherwise associated with the cell membrane.

[0044] As used herein, the phrase "cell surface-expressed CD28" refers to one or more CD28 proteins(s) expressed on the surface of a cell in vitro or in vivo, wherein at least a portion of the CD28 protein is exposed to the extracellular side of the cell membrane and is accessible to the antigen-binding portion of an antibody. "Cell surface-expressed CD28" includes CD28 protein in the context of a functional T cell costimulatory receptor within the cell membrane. The phrase "cell surface-expressed CD28" includes CD28 protein expressed as part of a homodimer on the surface of a cell. "Cell surface-expressed CD28" may comprise or consist of CD28 protein expressed on the surface of a cell that normally expresses CD28 protein. Alternatively, "cell surface-expressed CD28" may comprise or consist of CD28 protein expressed on the surface of a cell that does not normally express human CD28 on its surface but has been artificially engineered to express CD28 on its surface.

[0045] As used herein, the phrase "anti-CD28 antibody" includes both monovalent antibodies having a single specificity, and bispecific antibodies comprising a first arm that binds to CD28 and a second arm that binds to a second (target) antigen, wherein the anti-CD28 arm comprises any of the HCVR / LCVR or CDR sequences set forth in Table 3 herein. Examples of anti-CD28 bispecific antibodies are described elsewhere herein. The term "antigen-binding molecule" includes antibodies and antigen-binding fragments of antibodies, including, for example, bispecific antibodies.

[0046] As used herein, the term "MUC16" refers to the human MUC16 protein (e.g., "mouse MUC16," "monkey MUC16," etc.), unless specified as being from a non-human species. Human MUC16 protein has the amino acid sequence set forth in SEQ ID NO:49 and / or has the amino acid sequence set forth in NCBI accession number NP_078966. The human MUC16 membrane proximal domain (P13810-P14451) with a myc-myc-his tag is set forth as SEQ ID NO:51.

[0047] As used herein, "antibodies that bind MUC16" or "anti-MUC16 antibodies" include antibodies and antigen-binding fragments thereof that can bind to soluble MUC16 and / or cell surface-expressed MUC16. Soluble MUC16 includes native MUC16 protein as well as recombinant MUC16 protein variants that lack the transmembrane domain or are not associated with the cell membrane, such as MUC16 constructs.

[0048] As used herein, the phrase "anti-MUC16 antibody" includes both monovalent antibodies having a single specificity, as well as bispecific antibodies comprising a first arm that binds to MUC16 and a second arm that binds to a second (target) antigen, wherein the anti-MUC16 arm comprises any of the HCVR / LCVR or CDR sequences set forth in Table 1 herein. Examples of anti-MUC16 bispecific antibodies are described elsewhere herein. The term "antigen-binding molecule" includes antibodies and antigen-binding fragments of antibodies, including, for example, bispecific antibodies.

[0049] The term "antigen-binding molecule" includes antibodies and antigen-binding fragments of antibodies, including, for example, bispecific antibodies.

[0050] As used herein, the term "antibody" refers to any antigen-binding molecule or molecular complex that contains at least one complementarity-determining region (CDR) that specifically binds to or interacts with a particular antigen (e.g., CD28). Immunoglobulin molecules include four polypeptide chains, two heavy (H) chains and two light (L) chains, interconnected by sulfide bonds, as well as multimers thereof (e.g., IgM). Each heavy chain comprises a heavy chain variable region (abbreviated herein as HCVR or VH) and a heavy chain constant region. The heavy chain constant region is divided into three domains: C, C- ... H 1. C H 2, and C H Each light chain comprises a light chain variable region (abbreviated herein as LCVR or VL) and a light chain constant region. The light chain constant region comprises one domain (C L 1) V H Area and V L The regions can be further subdivided into regions of hypervariability called complementarity determining regions (CDRs) interspersed with more conserved regions called framework regions (FRs). H and V LEach CDR consists of three CDRs and four FRs arranged from amino-terminus to carboxy-terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. In different embodiments of the invention, the FRs of anti-CD28 and / or anti-MUC16 antibodies (or antigen-binding portions thereof) may be identical to human germline sequences or may be naturally or artificially modified. An amino acid consensus sequence may be defined based on a side-by-side analysis of two or more CDRs.

[0051] The term "antibody" as used herein also includes antigen-binding fragments of complete antibody molecules. "Antigen-binding portion" of an antibody, "antigen-binding fragment" of an antibody, and similar terms, as used herein, include naturally occurring, enzymatically obtained, synthetic, or genetically engineered polypeptides or glycoproteins that specifically bind antigens to form complexes. Antibody-binding fragments of antibodies can be derived from complete antibody molecules using any suitable standard techniques, such as, for example, proteolytic or recombinant genetic engineering techniques involving the manipulation and expression of DNA encoding antibody variable domains and, optionally, constant domains. Such DNA is known and / or readily available, for example, from commercial sources, DNA libraries (including, for example, phage antibody libraries), or can be synthesized. DNA can be sequenced and manipulated chemically or by using molecular biology techniques to, for example, place one or more variable and / or constant domains in a suitable configuration, or to introduce codons, create cysteine ​​residues, modify, add, or delete amino acids, etc.

[0052] Non-limiting examples of antibody-binding fragments include (i) Fab fragments, (ii) F(ab')2 fragments, (iii) Fd fragments, (iv) Fv fragments, (v) single-chain Fv (scFv) molecules, (vi) dAb fragments, and (vii) minimal recognition units consisting of amino acid residues mimicking a hypervariable region of an antibody (e.g., an isolated complementarity-determining region (CDR) such as a CDR3 peptide), or a constrained FR3-CDR3-FR4 peptide. Domain-specific antibodies, single-domain antibodies, domain-deleted antibodies, chimeric antibodies, CDR-grafted antibodies, diabodies, triabodies, tetrabodies, minibodies, nanobodies (e.g., monovalent nanobodies, bivalent nanobodies, etc.), small modular immunopharmaceuticals (SMIPs), and other engineered molecules such as shark variable IgNAR domains are also encompassed by the term "antigen-binding fragment" as used herein.

[0053] An antigen-binding fragment of an antibody will typically contain at least one variable domain, which may be of any size or amino acid composition and generally contains at least one CDR adjacent to or in-frame with one or more framework sequences. H Domain is V L In the antigen-binding fragment linked to the domain, V H Domains and V L The domains may be positioned relative to each other in any suitable configuration. For example, the variable region may be a dimer, with the V H -V H , V H -V L , or V L -V L Alternatively, the antigen-binding fragment of an antibody may comprise a dimer of monomeric V H Domain or V L It may also include a domain.

[0054] In certain embodiments, an antigen-binding fragment of an antibody comprises at least one constant domain. Non-limiting exemplary configurations of variable and constant domains that may be found in the antigen-binding fragments of antibodies of the invention include: (i) a V H -C H 1, (ii) V H -C H 2, (iii) V H -C H 3, (iv) V H -C H 1-C H 2. (v) V H -C H 1-C H 2-C H 3. (vi) V H -C H 2-C H 3, (vii)V H -C L , (viii) V L -C H 1, (ix)V L -C H 2. (x)V L -C H 3. (xi) V L -C H 1-C H 2, (xii)V L -C H 1-C H 2-C H 3, (xiii)V L -C H 2-C H 3, and (xiv) V L -C LIn any configuration of the variable and constant domains, including any of the exemplary configurations listed above, the variable and constant domains can be either directly linked to each other or linked by a complete or partial hinge or linker region. The hinge region can consist of at least two (e.g., 5, 10, 15, 20, 40, 60, or more) amino acids that provide a flexible or semi-flexible connection between adjacent variable and / or constant domains in a single polypeptide molecule. Moreover, antigen-binding fragments can be linked to each other and / or to one or more monomeric V H Domain or V L The variable domain and constant domain configurations may comprise homodimers or heterodimers (or other multimers) of any of the variable domain and constant domain configurations listed above in non-covalent association with the domains (e.g., by disulfide bond(s)).

[0055] Like intact antibody molecules, antibody-binding fragments can be monospecific or multispecific (e.g., bispecific). Multispecific antigen-binding fragments of antibodies typically contain at least two different variable domains, each capable of specifically binding to a separate antigen or to a different epitope on the same antigen. Any multispecific antibody format, including the exemplary bispecific antibody formats disclosed herein, can be adapted for use in connection with the antigen-binding fragments of antibodies of the present invention using routine techniques available in the art.

[0056] The antibodies of the present invention may function through complement-dependent cytotoxicity (CDC) or antibody-dependent cell-mediated cytotoxicity (ADCC). "Complement-dependent cytotoxicity" (CDC) refers to the lysis of antigen-expressing cells by the antibodies of the present invention in the presence of complement. "Antibody-dependent cell-mediated cytotoxicity" (ADCC) refers to a cell-mediated reaction in which nonspecific cytotoxic cells expressing Fc receptors (FcRs) (e.g., natural killer (NK) cells, neutrophils, and macrophages) recognize bound antibodies on target cells, thereby resulting in lysis of the target cells. CDC and ADCC can be measured using assays well known and available in the art. (See, e.g., U.S. Patent Nos. 5,500,362 and 5,821,337, and Clynes et al. (1998) Proc. Natl. Acad. Sci. (USA) 95:652-656.) The constant region of an antibody is important in the ability of an antibody to fix complement and mediate cell-dependent cytotoxicity. Thus, the isotype of the antibody can be selected based on whether it is desirable for the antibody to mediate cytotoxicity.

[0057] In certain embodiments of the invention, the anti-CD28 and / or MUC16 antibodies (monospecific or bispecific) of the invention are human antibodies. The term "human antibody," as used herein, is intended to include antibodies having variable and constant regions derived from human germline immunoglobulin sequences. Human antibodies of the invention may include, for example, amino acid residues in the CDRs, particularly CDR3, that are not encoded by human germline immunoglobulin sequences (e.g., mutations introduced by random or site-specific mutagenesis in vitro or by somatic mutation in vivo). However, the term "human antibody," as used herein, is also intended to include antibodies in which CDR sequences derived from the germline of another mammalian species (e.g., a mouse) have been grafted onto human framework sequences. It is not intended to be a diagram.

[0058] The antibodies of the present invention may, in some embodiments, be recombinant human antibodies. The term "recombinant human antibody," as used herein, is intended to include all human antibodies prepared, expressed, created, or isolated by recombinant means, such as antibodies expressed using a recombinant expression vector transfected into a host cell (described below), antibodies isolated from a recombinant combinatorial human antibody library (described below), antibodies isolated from an animal (e.g., a mouse) that is transgenic for human immunoglobulin genes (e.g., Taylor et al. (1992) Nucl. Acids Res. 20:6287-6295), or antibodies prepared, expressed, created, or isolated by any other means, including splicing of human immunoglobulin gene sequences into other DNA sequences. Such recombinant human antibodies have variable and constant regions derived from human germline immunoglobulin sequences. However, in certain embodiments, such recombinant human antibodies are subjected to in vitro mutagenesis (or, when animals transgenic for human Ig sequences are used, in vivo somatic mutagenesis), thus modifying the V and V regions of the recombinant antibody. H and V L The amino acid sequence of the region is human germline V H and V L These are sequences that are derived from and related to sequences, but may not naturally occur within the human antibody germline repertoire in vivo.

[0059] Human antibodies can exist in two forms related to hinge heterogeneity. In the first form, the immunoglobulin molecule contains a stable four-chain construct of approximately 150-160 kDa in which dimers are held together by interchain heavy chain disulfide bonds. In the second form, the dimers are not linked by interchain disulfide bonds, forming approximately 75-80 kDa molecules consisting of covalently linked light and heavy chains (half antibodies). These forms have been extremely difficult to separate, even after affinity purification.

[0060] The frequency of occurrence of the second form in various intact IgG isotypes is due to, but not limited to, structural differences associated with the antibody hinge region isotype. A single amino acid substitution in the hinge region of a human IgG4 hinge can significantly reduce the occurrence of the second form to levels typically observed with a human IgG1 hinge (Angal et al. al. (1993) Molecular Immunology 30:105). The present invention encompasses antibodies with one or more mutations in the hinge, CH2, or CH3 regions, which may be desirable, for example, in production, to improve the yield of the desired antibody form.

[0061] The antibody of the present invention may be an isolated antibody. As used herein, an "isolated antibody" refers to an antibody that has been identified, separated, and / or recovered from at least one component of its natural environment. For example, an antibody that has been separated or removed from at least one component of an organism, or from a tissue or cell in which it naturally occurs or is naturally produced, is an "isolated antibody" for purposes of the present invention. An isolated antibody also includes an antibody in situ within a recombinant cell. An isolated antibody is an antibody that has undergone at least one purification or isolation step. According to certain embodiments, an isolated antibody may be substantially free of other cellular material and / or chemicals.

[0062] The present invention also includes one-arm antibodies that bind to CD28 and / or MUC16. As used herein, "one-arm antibody" refers to an antigen-binding molecule comprising a single antibody heavy chain and a single antibody light chain. The one-arm antibodies of the present invention may comprise any of the HCVR / LCVR or CDR amino acid sequences listed in Tables 1 and 3.

[0063] The anti-CD28 and / or MUC16 antibodies or antigen-binding domains thereof herein may have one or more mutations in the framework and / or CDR regions of the heavy and light chain variable domains compared to the corresponding germline sequences from which the antigen-binding protein or antigen-binding domain is derived. The present invention also encompasses antibodies, including antigen-binding domains thereof, derived from any of the amino acid sequences disclosed herein, in which one or more amino acids in one or more framework and / or CDR regions are mutated to the corresponding residue(s) in the germline sequence from which the antibody was derived, or to the corresponding residue(s) in another human germline sequence, or to a conservative amino acid substitution of the corresponding germline residue(s) (such sequence changes are collectively referred to herein as "germline mutations"). Starting from the heavy and light chain variable region sequences disclosed herein, one skilled in the art can readily produce numerous antibodies and antibody-binding fragments containing one or more individual germline mutations or combinations thereof. In certain embodiments, the V H and / or V LAll framework and / or CDR residues within a domain are mutated back to the residue found in the original germline sequence from which the antibody was derived. In other embodiments, only certain residues are mutated back to the original germline sequence; for example, the mutated residue is found only within the first eight amino acids of FR1 or the last eight amino acids of FR4, or the mutated residue is found only in CDR1, CDR2, or CDR3. In other embodiments, one or more of the framework and / or CDR residue(s) are mutated to the corresponding residue(s) in a different germline sequence (i.e., a germline sequence different from the germline sequence from which the antibody was originally derived). Furthermore, antibodies of the present invention, or antigen-binding domains thereof, may contain any combination of two or more germline mutations within the framework and / or CDR regions; for example, certain individual residues are mutated to the corresponding residue in a particular germline sequence, while certain other residues that differ from the original germline sequence are maintained or mutated to the corresponding residue in a different germline sequence. Once obtained, antigen-binding domains containing one or more germline mutations can be readily tested for one or more desired properties, such as improved binding specificity, increased binding affinity, improved or enhanced antagonist or agonist biological properties (as the case may be), reduced immunogenicity, etc. Antibodies or antigen-binding fragments thereof obtained by this general method are encompassed within the scope of the present invention.

[0064] The present invention also includes anti-CD28 and / or MUC16 antibodies and antigen-binding molecules comprising variants of any of the HCVR, LCVR, and / or CDR amino acid sequences disclosed herein. Exemplary variants included in this aspect of the invention include variants of any of the HCVR, LCVR, and / or CDR amino acid sequences disclosed herein with one or more conservative substitutions. For example, the present invention includes anti-CD28 antibodies and antigen-binding molecules having HCVR, LCVR, and / or CDR amino acid sequences with, for example, 10 or fewer, 8 or fewer, 6 or fewer, or 4 or fewer conservative amino acid substitutions relative to any of the HCVR, LCVR, and / or CDR amino acid sequences set forth in Table 3 herein.

[0065] The term "epitope" refers to an antigenic determinant that interacts with a specific antigen-binding site in the variable region of an antibody molecule, known as the paratope. A single antigen may have more than one epitope. Thus, different antibodies may bind to different regions on an antigen and have different biological effects. Epitopes may be conformational or linear. Conformational epitopes are generated by spatially juxtaposed amino acids from different segments of a linear polypeptide chain. Linear epitopes are generated by adjacent amino acid residues within a polypeptide chain. In certain circumstances, epitopes may include carbohydrate, phosphoryl, or sulfonyl moieties on an antigen.

[0066] The term "substantial identity" or "substantially identical," when referring to a nucleic acid or fragment thereof, refers to a nucleic acid having a sequence similar to that of another nucleic acid (or its complementary strand) with appropriate nucleotide insertions or substitutions. When optimally aligned with the deletion, they exhibit at least about 95%, more preferably at least about 96%, 97%, 98%, or 99% nucleotide sequence identity of the nucleotide bases as measured by any well-known algorithm for sequence identity, such as FASTA, BLAST, or Gap, as discussed below. A nucleic acid molecule having substantial identity to a reference nucleic acid molecule can, in certain cases, encode a polypeptide having the same or substantially similar amino acid sequence as the polypeptide encoded by the reference nucleic acid molecule.

[0067] When applied to polypeptides, the term "substantial similarity" or "substantially similar" means that two peptide sequences, when optimally aligned using a program such as GAP or BESTFIT with a predetermined gap weight, share at least 95% sequence identity, and even more preferably at least 98% or 99% sequence identity. Preferably, residue positions that are not identical differ by conservative amino acid substitutions. A "conservative amino acid substitution" is one in which an amino acid residue is replaced with another amino acid residue having a side chain (R group) with similar chemical properties (e.g., charge or hydrophobicity). Generally, conservative amino acid substitutions do not substantially alter the functional properties of a protein. When two or more amino acid sequences differ from each other in conservative substitutions, the percent sequence identity or degree of similarity may be adjusted upward to correct for the conservative nature of the substitution. Means for making this adjustment are well known to those of skill in the art. See, e.g., Pearson (1994) Methods Mol. Biol. 24:307-331. Examples of groups of amino acids having side chains with similar chemical properties include: (1) aliphatic side chains: glycine, alanine, valine, leucine, and isoleucine, (2) aliphatic-hydroxyl side chains: serine and threonine, (3) amide-containing side chains: asparagine and glutamine, (4) aromatic side chains: phenylalanine, tyrosine, and tryptophan, (5) basic side chains: lysine, arginine, and histidine, (6) acidic side chains: aspartic acid and glutamic acid, and (7) sulfur-containing side chains: cysteine ​​and methionine. Preferred conservative amino acid substitution groups are valine-leucine-isoleucine, phenylalanine-tyrosine, lysine-arginine, alanine-valine, glutamic acid-aspartic acid, and asparagine-glutamine. Alternatively, a conservative substitution is any change that has a positive value in the PAM250 log-likelihood matrix disclosed in Gonnet et al. (1992) Science 256:1443-1445. A "moderately conservative" substitution is any change that has a non-negative value in the PAM250 log-likelihood matrix.

[0068] Sequence similarity for polypeptides, also known as sequence identity, is typically measured using sequence analysis software. Protein analysis software matches similar sequences using measures of similarity assigned to various substitutions, deletions, and other modifications, including conservative amino acid substitutions. For example, GCG software includes programs such as Gap and Bestfit, which can be used with default parameters to determine sequence homology or sequence identity between closely related polypeptides, such as homologous polypeptides from different species of organisms, or between a wild-type protein and its mutant protein. See, for example, GCG version 6.1. Polypeptide sequences can also be compared using FASTA, a program in GCG version 6.1, with default or recommended parameters. FASTA (e.g., FASTA2 and FASTA3) provides alignment and percent sequence identity of the best overlapping regions between the query and search sequences (Pearson (2000) (see above)). Another preferred algorithm for comparing the sequences of the present invention to a database containing a large number of sequences from different organisms is the computer program BLAST, particularly BLASTP or TBLASTN, using default parameters. See, e.g., Altschul et al. (1990) J. Mol. Biol. 215:403-410 and Altschul et al. (1997) Nucleic Acids Res. 25:3389-402.

[0069] Bispecific antigen binding molecules The antibodies of the present invention may be monospecific, bispecific, or multispecific. Multispecific antibodies may be specific for different epitopes of a single target polypeptide or may contain antigen-binding domains specific for two or more target polypeptides. See, for example, Tutt et al., 1991, J. Immunol. 147:60-69; Kufer et al., 2004, Trends Biotechnol. 22:238-244. The anti-CD28 and / or MUC16 antibodies of the present invention can be linked to or coexpressed with another functional molecule, such as another peptide or protein. For example, an antibody or fragment thereof can be operatively linked (e.g., by chemical bonding, genetic fusion, noncovalent bonding, or other means) to one or more other molecular entities, such as another antibody or antibody fragment, to produce a bispecific or multispecific antibody with a second binding specificity.

[0070] The use of the phrases "anti-CD28 antibody" and / or "anti-MUC16 antibody" herein is intended to include both monospecific anti-CD28 antibodies and / or anti-MUC16 antibodies and bispecific antibodies comprising a CD28-binding arm or a MUC16-binding arm and a second arm that binds to a target antigen. Thus, the present invention includes bispecific antibodies in which one immunoglobulin arm binds to human CD28 or MUC16, and the other immunoglobulin arm is specific for a target antigen. The target antigen to which the other arm of a CD28 or MUC16 bispecific antibody binds can be any antigen expressed on or near a cell, tissue, organ, microorganism, or virus against which a targeted immune response is desired. The CD28-binding arm can comprise any of the HCVR / LCVR or CDR amino acid sequences listed in Table 3 herein. The MUC16-binding arm can comprise any of the HCVR / LCVR or CDR amino acid sequences listed in Table 1 herein. In certain embodiments, the CD28 binding arm binds to human CD28 and induces human T cell proliferation.

[0071] In the context of a bispecific antibody of the invention, in which one arm of the antibody binds to CD28 and the other arm binds to a target antigen, the target antigen can be a tumor-associated antigen such as MUC16.

[0072] According to certain exemplary embodiments, the present invention comprises bispecific antigen-binding molecules that specifically bind to CD28 and MUC16. Such molecules may be referred to herein as, for example, "anti-CD28 / anti-MUC16" or "anti-CD28xMUC16" or "CD28xMUC16" or "anti-MUC16 / anti-CD28" or "anti-MUC16xCD28" or "MUC16xCD28" bispecific molecules, or other similar terms.

[0073] According to certain exemplary embodiments shown in the figures, a bispecific antigen-binding molecule (e.g., a bispecific antibody) may have an effector arm and a targeting arm. The effector arm may be a first antigen-binding domain (e.g., an anti-CD28 antibody) that binds to an antigen on an effector cell (e.g., a T cell). The targeting arm may be a second antigen-binding domain (e.g., an anti-MUC16 antibody) that binds to an antigen on a target cell (e.g., a tumor cell). According to certain exemplary embodiments, the effector arm binds to CD28, and the targeting arm binds to MUC16. The bispecific anti-CD28 / MUC16 can provide a costimulatory signal to effector cells (e.g., T cells). The effector arm is ineffective at stimulating T cells without clustering. Once clustered, the effector arm alone has little effect at stimulating T cells. In combination with the targeting arm, the effector arm stimulates T cells. Tumor-targeting arm However, tumor specificity may be incomplete. The antigen targeted by the targeting arm (e.g., MUC16) may be expressed on a portion of tumor cells. The specificity of the tumor-targeting arm may be increased by overlapping with an anti-CD3 bispecific antigen-binding molecule (e.g., an anti-CD3 / MUC16 bispecific antibody).

[0074] As used herein, the term "antigen-binding molecule" refers to a protein, polypeptide, or molecular complex comprising or consisting of at least one complementarity-determining region (CDR) alone or in combination with one or more additional CDRs and / or framework regions (FRs), which specifically binds to a particular antigen. In certain embodiments, the antigen-binding molecule is an antibody or an antibody fragment, as those terms are defined elsewhere herein.

[0075] As used herein, the term "bispecific antigen-binding molecule" refers to a protein, polypeptide, or molecular complex comprising at least a first antigen-binding domain and a second antigen-binding domain. Each antigen-binding domain in a bispecific antigen-binding molecule comprises at least one CDR that specifically binds to a particular antigen, either alone or in combination with one or more additional CDRs and / or FRs. In the context of the present invention, the first antigen-binding domain specifically binds to a first antigen (e.g., CD28), and the second antigen-binding domain specifically binds to a second, different antigen (e.g., MUC16).

[0076] In certain exemplary embodiments of the present invention, the bispecific antigen-binding molecule is a bispecific antibody. Each antigen-binding domain of the bispecific antibody comprises a heavy chain variable domain (HCVR) and a light chain variable domain (LCVR). In the context of a bispecific antigen-binding molecule (e.g., a bispecific antibody) comprising a first and a second antigen-binding domain, the CDRs of the first antigen-binding domain may be designated with the prefix "D1", and the CDRs of the second antigen-binding domain may be designated with the prefix "D2". Thus, the CDRs of the first antigen-binding domain may be referred to herein as D1-HCDR1, D1-HCDR2, and D1-HCDR3, and the CDRs of the second antigen-binding domain may be referred to herein as D2-HCDR1, D2-HCDR2, and D2-HCDR3.

[0077] The first and second antigen-binding domains may be directly or indirectly linked to each other to form the bispecific antigen-binding molecule of the present invention. Alternatively, the first and second antigen-binding domains may each be linked to a separate multimerization domain. The association of one multimerization domain with another multimerization domain promotes the association between the two antigen-binding domains, thereby forming the bispecific antigen-binding molecule. As used herein, a "multimerization domain" is any polymer, protein, polypeptide, peptide, or amino acid capable of associating with a second multimerization domain of the same or similar structure or configuration. For example, a multimerization domain may be linked to a second multimerization domain of the same or similar structure or configuration. H A non-limiting example of a multimerizing component is a polypeptide comprising the Fc portion of an immunoglobulin (C H 2-C H 3 domains), for example the Fc domain of IgG selected from the isotypes IgG1, IgG2, IgG3, and IgG4, as well as any allotype within each isotype group.

[0078] The bispecific antigen-binding molecules of the present invention typically comprise two multimerization domains, e.g., two Fc domains that are each part of a separate antibody heavy chain. The first and second multimerization domains may be of the same IgG isotype, e.g., IgG1 / IgG1, IgG2 / IgG2, or IgG4 / IgG4. Alternatively, the first and second multimerization domains may be of different IgG isotypes, e.g., IgG1 / IgG2, IgG1 / IgG4, or IgG2 / IgG4.

[0079] In certain embodiments, the multimerization domain is an Fc fragment or an amino acid sequence of 1 to about 200 amino acids in length containing at least one cysteine ​​residue. In other embodiments, the multimerization domain is a cysteine ​​residue or a short cysteine-containing peptide. Other multimerization domains include peptides or polypeptides comprising or consisting of a leucine zipper, a helix loop motif, or a coiled-coil motif.

[0080] Any bispecific antibody format or technology can be used to generate the bispecific antigen-binding molecules of the present invention. For example, an antibody or fragment thereof having a first antigen-binding specificity can be operatively linked (e.g., by chemical conjugation, genetic fusion, non-covalent bonding, etc.) to one or more other molecular entities, such as another antibody or antibody fragment having a second antigen-binding specificity, to produce a bispecific antigen-binding molecule. Specific exemplary bispecific formats that can be used in the context of the present invention include, for example, scFv-based or diabody bispecific formats, IgG-scFv fusions, dual variable domain (OVO)-Ig, Quadroma, knobs-into-holes, common light chains (e.g., common light chains with knobs-into-holes), CrossMab, CrossFab, (SEEO)body, leucine zipper, Ouobody, IgG1 / IgG2, dual acting Fab (OAF)-IgG, and Mab. 2 These include, but are not limited to, bispecific formats (for a review of the aforementioned formats, see, e.g., Klein et al. 2012, mAbs 4:6, 1-11, and references cited therein).

[0081] In the context of the bispecific antigen-binding molecules of the present invention, the multimerization domain, e.g., the Fc domain, may contain one or more amino acid changes (e.g., insertions, deletions, or substitutions) compared to a wild-type, naturally occurring Fc domain. For example, the present invention includes bispecific antigen-binding molecules containing one or more modifications in the Fc domain that result in a modified Fc domain with modified binding interactions (e.g., enhanced or reduced) between Fc and FcRn. In one embodiment, the bispecific antigen-binding molecule comprises a C H 2 or C HThe FcRn comprises modifications in three regions that enhance the affinity of the Fc domain for FcRn in an acidic environment (e.g., in the endosome at a pH range of about 5.5 to about 6.0). Non-limiting examples of such Fc modifications include, for example, modifications at positions 250 (e.g., E or Q), 250 and 428 (e.g., L or F), 252 (e.g., LN / FIW or T), 254 (e.g., S or T), and 256 (e.g., S / R / Q / EID or T), or modifications at positions 428 and / or 433 (e.g., UR / S / P / Q or K) and / or 434 (e.g., H / F or V), or modifications at positions 250 and / or 428, or modifications at positions 307 or 308 (e.g., 308F, V308F) and 434. In one embodiment, the modifications include 428L (e.g., M428L) and 434S (e.g., N434S) modifications; 428L, 2591 (e.g., V2591), and 308F (e.g., V308F) modifications; 433K (e.g., H433K) and 434 (e.g., 434Y) modifications; 252, 254, and 256 (e.g., 252Y, 254T, and 256E) modifications, 250Q and 428L modifications (e.g., T250Q and M428L), and 307 and / or 308 modifications (e.g., 308F or 308P).

[0082] The present invention is the first C H 3 domain and second Ig C H Also included are bispecific antigen-binding molecules comprising three domains, a first and a second Ig C H The three domains differ from each other by at least one amino acid, and the at least one amino acid difference reduces binding of the bispecific antibody to Protein A compared to a bispecific antibody lacking that amino acid difference. H The 3 domain binds to protein A and the second Ig C H The 3 domains are H95R modified (according to IMGT exon numbering, H according to EU numbering) 435R) that reduces or eliminates Protein A binding. H3 may further comprise a Y96F modification (by IMGT, Y436F by EU). Additional modifications that may be found within the second CH3 include D16E, L18M, N44S, K52N, V57M, and V82I for IgG1 antibodies (by IMGT; D356E, L358M, N384S, K392N, V397M, and V422I by EU), N44S, K52N, and V82I for IgG2 antibodies (N384S, K392N, and V422I by IMGT; EU), and Q15R, N44S, K52N, V57M, R69K, E79Q, and V82I for IgG4 antibodies (by IMGT; Q355R, N384S, K392N, V397M, R409K, E419Q, and V422I by EU).

[0083] In certain embodiments, the Fc domain may be a chimera that combines Fc sequences from two or more immunoglobulin isotypes. For example, the chimeric Fc domain may be a chimeric Fc domain that combines Fc sequences from human IgG1, human IgG2, or human IgG4 C. H C derived from 2 regions H Part or all of the 2 sequence, and C derived from human IgG1, human IgG2, or human IgG4 H The chimeric Fc domain may comprise some or all of the three sequences. The chimeric Fc domain may also comprise a chimeric hinge region. For example, the chimeric hinge may comprise an "upper hinge" sequence derived from a human IgG1 hinge region, a human IgG2 hinge region, or a human IgG4 hinge region, combined with a "lower hinge" sequence derived from a human IgG1 hinge region, a human IgG2 hinge region, or a human IgG4 hinge region. A specific example of a chimeric Fc domain that may be comprised in any of the antigen-binding molecules described herein is a chimeric Fc domain that, from the N-terminus to the C-terminus, comprises [IgG4 C H 1]-[IgG4 upper hinge]-[IgG2 lower hinge]-[IgG4 CH2]-[IgG4 C H Another example of a chimeric Fc domain that may be included in any of the antigen-binding molecules described herein includes, from the N-terminus to the C-terminus, [IgG1 C H1]-[IgG1 upper hinge]-[IgG2 lower hinge]-[IgG4 C H 2]-[IgG1 C H These and other examples of chimeric Fc domains that can be included in any of the antigen-binding molecules of the invention are described in WO2014 / 022540A1, and chimeric Fc domains having these general structural arrangements, and variants thereof, can alter Fc receptor binding, which in turn affects Fc effector function.

[0084] Sequence variants The antibodies and bispecific antigen-binding molecules of the present invention may contain one or more amino acid substitutions, insertions, and / or deletions in the framework and / or CDR regions of the heavy and light chain variable domains compared to the corresponding germline sequences from which the individual antigen-binding domains were derived. Such mutations can be readily identified by comparing the amino acid sequences disclosed herein with germline sequences available, for example, from public antibody sequence databases. The antigen-binding molecules of the present invention may comprise antigen-binding fragments derived from any of the exemplary amino acid sequences disclosed herein, in which one or more amino acids in one or more framework and / or CDR regions are mutated to the corresponding residue(s) in the germline sequence from which the antibody was derived, or to the corresponding residue(s) in another human germline sequence, or to conservative amino acid substitutions of the corresponding germline residue(s) (such sequence changes are collectively referred to herein as "germline mutations"). Starting from the heavy and light chain variable region sequences disclosed herein, one skilled in the art can easily produce numerous antibodies and antibody-binding fragments containing one or more individual germline mutations or combinations thereof. In certain embodiments, V H and / or V LAll framework and / or CDR residues within the domain are mutated back to residues found in the original germline sequence from which the antigen-binding domain was originally derived. In other embodiments, only certain residues are mutated back to the original germline sequence, e.g., only mutated residues are within the first eight amino acids of FR1 or the last eight amino acids of FR4. Only amino acid residues found or mutated within the framework and / or CDR residue(s) are found in CDR1, CDR2, or CDR3. In other embodiments, one or more of the framework and / or CDR residue(s) are mutated to the corresponding residue(s) in a different germline sequence (i.e., a germline sequence different from the germline sequence from which the antigen-binding domain was originally derived). Furthermore, an antigen-binding domain may contain any combination of two or more germline mutations in the framework and / or CDR regions, e.g., certain individual residues are mutated to the corresponding residue in a particular germline sequence, while certain other residues that differ from the original germline sequence are either maintained or mutated to the corresponding residue in a different germline sequence. Once obtained, antigen-binding domains containing one or more germline mutations can be readily tested for one or more desired properties, such as improved binding specificity, increased binding affinity, improved or enhanced antagonist or agonist biological properties (as the case may be), reduced immunogenicity, etc. Bispecific antigen-binding molecules comprising one or more antigen-binding domains obtained in this general manner are encompassed within the scope of the present invention.

[0085] The present invention also includes antigen-binding molecules in which one or both antigen-binding domains comprise variants of any of the HCVR, LCVR, and / or CDR amino acid sequences disclosed herein with one or more conservative substitutions. For example, the present invention includes antigen-binding molecules comprising antigen-binding domains having HCVR, LCVR, and / or CDR amino acid sequences with, for example, 10 or fewer, 8 or fewer, 6 or fewer, or 4 or fewer conservative amino acid substitutions relative to any of the HCVR, LCVR, and / or CDR amino acid sequences disclosed herein. A "conservative amino acid substitution" is one in which an amino acid residue is replaced with another amino acid residue having a side chain (R group) with similar chemical properties (e.g., charge or hydrophobicity). Generally, conservative amino acid substitutions do not substantially alter the functional properties of a protein. Examples of groups of amino acids having side chains with similar chemical properties include: (1) aliphatic side chains: glycine, alanine, valine, leucine, and isoleucine, (2) aliphatic-hydroxyl side chains: serine and threonine, (3) amide-containing side chains: asparagine and glutamine, (4) aromatic side chains: phenylalanine, tyrosine, and tryptophan, (5) basic side chains: lysine, arginine, and histidine, (6) acidic side chains: aspartic acid and glutamic acid, and (7) sulfur-containing side chains: cysteine ​​and methionine. Preferred conservative amino acid substitution groups are valine-leucine-isoleucine, phenylalanine-tyrosine, lysine-arginine, alanine-valine, glutamic acid-aspartic acid, and asparagine-glutamine. Alternatively, a conservative substitution is any change that has a positive value in the PAM250 log-likelihood matrix disclosed in Gonnet et al. (1992) Science 256:1443-1445, which is incorporated herein by reference. A "moderately conservative" substitution is any change that has a non-negative value in the PAM250 log-likelihood matrix.

[0086] The present invention also includes antigen-binding molecules comprising antigen-binding domains having HCVR, LCVR, and / or CDR amino acid sequences substantially identical to any of the HCVR, LCVR, and / or CDR amino acid sequences disclosed herein. When referring to amino acid sequences, the term "substantial identity" or "substantially identical" means that two amino acid sequences, when optimally aligned using predefined gap weights, such as by the programs GAP or BESTFIT, share at least 95% sequence identity, and even more preferably at least 98% or 99% sequence identity. Preferably, residue positions that are not identical differ by conservative amino acid substitutions. When two or more amino acid sequences differ from each other by conservative substitutions, the percent sequence identity or degree of similarity may be adjusted upward to correct for the conservative nature of the substitutions. Means for making this adjustment are well known to those skilled in the art. See, for example, Pearson (1994) Methods Mol. Biol. 24:307-331. Please refer to.

[0087] Sequence similarity for polypeptides, also known as sequence identity, is typically measured using sequence analysis software. Protein analysis software matches similar sequences using measures of similarity assigned to various substitutions, deletions, and other modifications, including conservative amino acid substitutions. For example, GCG software includes programs such as Gap and Bestfit, which can be used with default parameters to determine sequence homology or sequence identity between closely related polypeptides, such as homologous polypeptides from different species of organisms, or between a wild-type protein and its mutant protein. See, for example, GCG version 6.1. Polypeptide sequences can also be compared using FASTA, a program in GCG version 6.1, with default or recommended parameters. FASTA (e.g., FASTA2 and FASTA3) provides alignment and percent sequence identity of the best overlapping regions between the query and search sequences (Pearson (2000) (see above)). Another preferred algorithm for comparing the sequences of the present invention to a database containing a large number of sequences from different organisms is the computer program BLAST, particularly BLASTP or TBLASTN, with default parameters. See, e.g., Altschul et al. (1990) J. Mol. Biol. 215:403-410 and Altschul et al. (1997) Nucleic Acids Res. 25:3389-402.

[0088] pH dependent binding The present invention includes anti-CD28 / anti-MUC16 bispecific antigen-binding molecules with pH-dependent binding properties. For example, anti-CD28 antibodies of the present invention may exhibit reduced binding to CD28 at acidic pH compared to neutral pH. Alternatively, anti-MUC16 antibodies of the present invention may exhibit enhanced binding to MUC16 at acidic pH compared to neutral pH. The term "acidic pH" includes pH values ​​below about 6.2, e.g., about 6.0, 5.95, 5.9, 5.85, 5.8, 5.75, 5.7, 5.65, 5.6, 5.55, 5.5, 5.45, 5.4, 5.35, 5.3, 5.25, 5.2, 5.15, 5.1, 5.05, 5.0, or lower. As used herein, the term "neutral pH" refers to a pH of about 7.0 to about 7.4. The expression "neutral pH" includes pH values ​​of about 7.0, 7.05, 7.1, 7.15, 7.2, 7.25, 7.3, 7.35, and 7.4.

[0089] In some cases, "decreased binding at acidic pH compared to neutral pH" refers to the K of an antibody that binds to its antigen at neutral pH. D K value of an antibody that binds to its antigen at acidic pH D For example, an antibody or antigen-binding fragment thereof may have an acidic / neutral K of about 3.0 or greater. D For purposes of the present invention, an antibody or antigen-binding fragment thereof may be considered to exhibit "reduced binding to CD28 at acidic pH compared to neutral pH" if it exhibits a ratio of 0.01 to 0.01. In certain exemplary embodiments, the acidic / neutral K D The ratio may be about 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, 8.0, 8.5, 9.0, 9.5, 10.0, 10.5, 11.0, 11.5, 12.0, 12.5, 13.0, 13.5, 14.0, 14.5, 15.0, 20.0, 25.0, 30.0, 40.0, 50.0, 60.0, 70.0, 100.0 or more.

[0090] Antibodies with pH-dependent binding properties can be obtained, for example, by screening a population of antibodies for reduced (or enhanced) binding to a specific antigen at acidic pH compared to neutral pH. Furthermore, modification of the antigen-binding domain at the amino acid level can produce antibodies with pH-dependent characteristics. For example, by substituting one or more amino acids in the antigen-binding domain (e.g., within the CDR) with histidine residues, an antibody can be obtained that has reduced antigen binding at acidic pH compared to neutral pH.

[0091] Antibodies containing Fc variants According to certain embodiments of the present invention, there are provided anti-CD28 / anti-MUC16 bispecific antigen binding molecules comprising an Fc domain comprising one or more mutations that enhance or decrease antibody binding to the FcRn receptor, e.g., at acidic pH compared to neutral pH. For example, the present invention provides an anti-CD28 / anti-MUC16 bispecific antigen binding molecule comprising an Fc domain comprising one or more mutations that enhance or decrease antibody binding to the FcRn receptor, e.g., at acidic pH compared to neutral pH. H 2 or C HThe present invention includes antibodies and antigen-binding molecules containing mutations in the FcRn region that increase the affinity of the Fc domain for FcRn in acidic environments (e.g., within endosomes at a pH range of about 5.5 to about 6.0). Such mutations can increase the serum half-life of the antibody when administered to an animal. Non-limiting examples of such Fc modifications include, for example, modifications at position 250 (e.g., E or Q), at positions 250 and 428 (e.g., L or F), at positions 252 (e.g., L / Y / F / W or T), 254 (e.g., S or T), and 256 (e.g., S / R / Q / E / D or T), or at positions 428 and / or 433 (e.g., H / L / R / S / P / Q or K) and / or 434 (e.g., H / F or Y), or modifications at positions 250 and / or 428, or at positions 307 or 308 (e.g., 308F, V308F), and 434. In one embodiment, the modifications include 428L (e.g., M428L) and 434S (e.g., N434S) modifications, 428L, 259I (e.g., V259I), and 308F (e.g., V308F) modifications, 433K (e.g., H433K) and 434 (e.g., 434Y) modifications, 252, 254, and 256 (e.g., 252Y, 254T, and 256E) modifications, 250Q and 428L modifications (e.g., T250Q and M428L), 307 and / or 308 modifications (e.g., 308F or 308P).

[0092] For example, the present invention includes anti-CD28 / anti-MUC16 bispecific antigen-binding molecules comprising an Fc domain comprising one or more pairs or groups of mutations selected from the group consisting of 250Q and 248L (e.g., T250Q and M248L); 252Y, 254T and 256E (e.g., M252Y, S254T and T256E); 428L and 434S (e.g., M428L and N434S); and 433K and 434F (e.g., H433K and N434F). All possible combinations of the foregoing Fc domain mutations, and other mutations in antibody variable domains disclosed herein, are contemplated within the scope of the present invention.

[0093] Biological properties of antibodies and antigen-binding molecules The present invention includes antibodies and antigen-binding fragments thereof that bind to human CD28 and / or MUC16 with high affinity. The present invention also includes antibodies and antigen-binding fragments thereof that bind to human CD28 and / or MUC16 with moderate or low affinity, depending on the therapeutic situation and the specific targeting properties desired. For example, in the context of a bispecific antigen-binding molecule in which one arm binds to CD28 and another arm binds to a target antigen (e.g., MUC16), it may be desirable for the target antigen-binding arm to bind to the target antigen with high affinity, while the anti-CD28 arm binds to CD28 with only moderate or low affinity. In this way, preferential targeting of the antigen-binding molecule to cells expressing the target antigen can be achieved while avoiding general / non-targeted CD28 binding and the resulting adverse side effects associated therewith.

[0094] According to certain embodiments, the present invention provides a method for the treatment of rhodopsin-related ... D In certain embodiments, the antibodies or antigen-binding fragments of the invention have an antibody binding affinity of less than about 150 nM, as measured by surface plasmon resonance, for example, using an assay format as defined in Example 4 herein, or a substantially similar assay. K of less than about 130 nM, less than about 120 nM, less than about 100 nM, less than about 50 nM, less than about 80 nM, less than about 60 nM, less than about 40 nM, or less than about 30 nM D It binds to CD28.

[0095] The present invention also includes antibodies and antigen-binding fragments thereof that bind to CD28 with a dissociation half-life (t1 / 2) of greater than about 2.1 minutes as measured by surface plasmon resonance at 37°C, e.g., using an assay format as defined in Example 4 herein, or a substantially similar assay. In certain embodiments, antibodies or antigen-binding fragments of the invention bind to CD28 with a t of greater than about 5 minutes, greater than about 10 minutes, greater than about 20 minutes, greater than about 30 minutes, greater than about 40 minutes, greater than about 50 minutes, greater than about 60 minutes, greater than about 70 minutes, greater than about 80 minutes, greater than about 90 minutes, greater than about 100 minutes, greater than about 200 minutes, greater than about 300 minutes, greater than about 400 minutes, greater than about 500 minutes, greater than about 600 minutes, greater than about 700 minutes, greater than about 800 minutes, greater than about 900 minutes, greater than about 1000 minutes, or greater than about 1200 minutes as measured by surface plasmon resonance at 25°C or 37°C using, for example, an assay format as defined in Example 4 herein, or a substantially similar assay.

[0096] The present invention includes bispecific antigen-binding molecules (e.g., bispecific antibodies) capable of simultaneously binding to human CD28 and human MUC16. According to certain embodiments, the bispecific antigen-binding molecules of the present invention specifically interact with cells expressing CD28 and / or MUC16. The extent to which a bispecific antigen-binding molecule binds to cells expressing CD28 and / or MUC16 can be assessed by fluorescence-activated cell sorting (FACS), as shown in Example 5 herein. For example, the present invention includes bispecific antigen-binding molecules that specifically bind to human or cynomolgus monkey cells (e.g., T cells) that express CD28 but not MUC16, and to human ovarian cancer cell lines (e.g., OVCAR-3 or PEO1) that express MUC16 but not CD28. The present invention provides a method for determining the binding of bispecific antigen-binding molecules to approximately 9.2 x 10 cells, as determined using a FACS assay as described in Example 4 or a substantially similar assay. -6 ~Approx. 2.8×10 -10 , or lower EC 50 and a bispecific antigen-binding molecule that binds to any of the aforementioned cells and cell lines at a value of 0.05.

[0097] The present invention also provides anti-CD28 / anti-MUC16 bispecific antigen binding molecules that induce or increase T cell-mediated killing of tumor cells. For example, the present invention provides anti-CD28 / anti-MUC16 bispecific antigen binding molecules that induce or increase T cell-mediated killing of tumor cells, with an EC of less than about 392 pM as measured in an in vitro T cell-mediated tumor cell killing assay using an assay format as defined in Example 7 herein (e.g., assessing the extent of PEO1 tumor cell killing by human or cynomolgus monkey PBMCs in the presence of anti-CD28xMUC16 antibodies) or a substantially similar assay. 50 In certain embodiments, the antibodies or antigen-binding fragments of the invention have an EC50 of less than about 200 pM, less than about 150 pM, less than about 100 pM, less than about 75 pM, less than about 50 pM, less than about 25 pM, less than about 10 pM, less than about 5.0 pM, less than about 4.0 pM, less than about 3.0 pM, less than about 2.5 pM, less than about 2.0 pM, less than about 1.5 pM, or less than about 1.45 pM, as measured by an in vitro T-cell-mediated tumor cell killing assay using an assay format as defined in Example 7 herein or a substantially similar assay. 50 At these values, it induces T cell-mediated tumor cell killing (e.g., PBMC-mediated killing of PEO1 cells).

[0098] The present invention also provides a compound having an EC value between 1.0 pM and 10 μM. 50 In certain embodiments, the anti-CD28 / anti-PSMA bispecific antigen binding molecule binds to CD28-expressing human and / or cynomolgus monkey T cells with an EC value between 9.2 μM and 120 nM. 50 For example, the present invention provides antibodies that bind to CD28-expressing human and / or cynomolgus monkey T cells at concentrations of about 1 pM, about 10 pM, about 100 pM, about 500 pM, about 1 nM, about 2 nM, about 5 nM, about 10 nM, about 20 nM, about 30 nM, about 40 nM, and about 50 nM. EC50 M, about 50 nM, about 60 nM, about 70 nM, about 80 nM, about 90 nM, about 100 nM, about 200 nM, about 300 nM, about 500 nM, about 800 nM, about 1000 nM, about 2 μM, about 4 μM, about 6 μM, about 8 μM, about 10 μM, or higher50 The present invention relates to an anti-CD28 / anti-MUC16 bispecific antigen-binding molecule that binds to CD28-expressing human T cells at a specific antibody level.

[0099] The present invention also includes anti-CD28 / anti-MUC16 bispecific antigen binding molecules that exhibit one or more properties selected from the group consisting of: (a) activate T cells and induce IL-2 release and upregulation of CD25+ and PD-1 in human PBMCs (see, e.g., Examples 6 and 7 herein); (b) increase human or cynomolgus monkey T cell-mediated cytotoxicity against MUC16-expressing cell lines (see, e.g., Example 7 herein); (c) induce naive primate T cell-mediated cytotoxicity against MUC16-expressing cell lines (see, e.g., Example 7 herein); (e) deplete tumor cells in mice (see, e.g., Example 8 herein); (f) enhance tumor clearance in mice (see, e.g., Example 8 herein); and (g) do not induce systemic T cell activation in cynomolgus monkeys.

[0100] The present invention includes anti-CD28 / anti-MUC16 bispecific antigen-binding molecules that can deplete tumor cells in a subject (see, e.g., Example 9). For example, according to certain embodiments, anti-CD28 / anti-MUC16 bispecific antigen-binding molecules are provided, wherein dual administration of the bispecific antigen-binding molecules to a subject (e.g., at a dose of about 5.0 mg / kg, about 2.5 mg / kg, about 1.0 mg / kg, about 0.5 mg / kg, about 0.2 mg / kg, about 0.1 mg / kg, about 0.05 mg / kg, about 0.02 mg / kg, about 0.01 mg / kg, or less) causes a reduction in the number of tumor cells in the subject. According to certain embodiments, anti-CD28 / anti-MUC16 bispecific antigen binding molecules are provided, wherein dual administration of the bispecific antigen binding molecules to a subject (e.g., at a dose of about 2500 mg, about 1000 mg, about 500 mg, about 200 mg, about 100 mg, about 50 mg, about 25 mg / kg, or less) causes a reduction in the number of tumor cells in the subject.

[0101] Epitope mapping and related techniques The epitope on CD28 or MUC16 bound by the antigen-binding molecules of the present invention may consist of a single contiguous sequence of three or more (e.g., 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or more) amino acids of the CD28 or MUC16 protein. Alternatively, the epitope may consist of multiple non-contiguous amino acids (or amino acid sequences) of CD28 or MUC16. The antibodies of the present invention may interact with amino acids contained within a CD28 monomer or with amino acids on two or more different CD28 chains of a CD28 dimer. The term "epitope," as used herein, refers to an antigenic determinant that interacts with a specific antigen-binding site in the variable region of an antibody molecule known as a paratope. A single antigen may have two or more epitopes. Therefore, different antibodies may bind to different regions of an antigen and have different biological effects. Epitopes can be either conformational or linear. Conformational epitopes are generated by spatially juxtaposed amino acids from different segments of a linear polypeptide chain. Linear epitopes are generated by adjacent amino acid residues within a polypeptide chain. In certain circumstances, epitopes may include saccharide, phosphoryl, or sulfonyl moieties on an antigen.

[0102] Various techniques known to those skilled in the art can be used to determine whether an antigen-binding domain of an antibody "interacts with one or more amino acids" within a polypeptide or protein. Exemplary techniques that can be used to determine the epitope or binding domain of a particular antibody or antigen-binding domain include, for example, Antibodies, Harlow and Lane(Cold Spring Harbor Press,Cold Spri These include routine cross-blocking assays such as those described in (Berkeley, NY), point mutagenesis (e.g., alanine scanning mutagenesis, arginine scanning mutagenesis, etc.), peptide blot analysis (Reineke, 2004, Methods Mol Biol 248:443-463), protease protection, and peptide cleavage analysis. Additionally, methods such as epitope excision, epitope extraction, and chemical modification of antigens can be employed (Tomer, 2000, Protein Science 9:487-496). Another method that can be used to identify amino acids within a polypeptide with which an antibody interacts is hydrogen / deuterium exchange detected by mass spectrometry. Generally speaking, the hydrogen / deuterium exchange method involves deuterium-labeling the protein of interest and then binding the antibody to the deuterium-labeled protein. The protein / antibody complex is then transferred to water, allowing hydrogen-deuterium exchange to occur at all residues except those protected by the antibody (which remain deuterium-labeled). After dissociation of the antibody, the target protein is subjected to protease cleavage and mass spectrometry, thereby revealing deuterium-labeled residues corresponding to the specific amino acids with which the antibody interacts. See, for example, Ehring (1999) Analytical Biochemistry 267(2):252-259; Engen and Smith (2001) Anal. Chem. 73:256A-265A. X-ray crystallography can also be used to identify the amino acids in a polypeptide with which the antibody interacts.

[0103] The present invention further includes anti-CD28 and anti-MUC16 antibodies (e.g., antibodies comprising any of the amino acid sequences as set forth in Tables 1 and 3 herein) that bind to the same epitope as any of the specific exemplary antibodies described herein. Similarly, the present invention also includes anti-CD28 and / or anti-MUC16 antibodies (e.g., antibodies comprising any of the amino acid sequences set forth in Table 1 herein) that compete with any of the specific exemplary antibodies described herein for binding to CD28 and / or MUC16.

[0104] The present invention also includes bispecific antigen-binding molecules comprising a first antigen-binding domain that specifically binds human CD28 and a second antigen-binding domain that specifically binds human MUC16, wherein the first antigen-binding domain binds to the same epitope on CD28 as any of the specific exemplary CD28-specific antigen-binding domains described herein, and / or the second antigen-binding domain binds to the same epitope on MUC16 as any of the specific exemplary MUC16-specific antigen-binding domains described herein.

[0105] Similarly, the present invention also includes bispecific antigen-binding molecules comprising a first antigen-binding domain that specifically binds human CD28 and a second antigen-binding domain that specifically binds human MUC16, wherein the first antigen-binding domain competes for binding to CD28 with any of the specific exemplary CD28-specific antigen-binding domains described herein, and / or the second antigen-binding domain competes for binding to MUC16 with any of the specific exemplary MUC16-specific antigen-binding domains described herein.

[0106] Whether a particular antigen-binding molecule (e.g., an antibody) or its antigen-binding domain binds to the same epitope as a reference antigen-binding molecule of the present invention or competes with a reference antigen-binding molecule of the present invention for binding can be easily determined using routine methods known in the art. For example, to determine whether a test antibody binds to the same epitope on CD28 (or MUC16) as a reference bispecific antigen-binding molecule of the present invention, the reference bispecific molecule is first bound to the CD28 protein (or MUC16 protein). The ability of the test antibody to bind to the CD28 (or MUC16) molecule is then evaluated. If the test antibody can bind to CD28 (or MUC16) after saturation binding with the reference bispecific antigen-binding molecule, it can be concluded that the test antibody binds to an epitope on CD28 (or MUC16) that is different from that of the reference bispecific antigen. On the other hand, if the test antibody does not bind to the reference bispecific antigen-binding molecule, the test antibody can be easily determined. If the test antibody cannot bind to the CD28 (or MUC16) molecule after saturation binding, it may bind to the same epitope of CD28 (or MUC16) as the epitope bound by the reference bispecific antigen-binding molecule of the present invention. Further routine experiments (e.g., peptide mutation and binding analysis) can then be performed to determine whether the observed loss of binding of the test antibody is actually due to binding to the same epitope as the reference bispecific antigen-binding molecule, or whether steric blocking (or another phenomenon) is responsible for the observed loss of binding. This type of experiment can be performed using ELISA, RIA, Biacore, flow cytometry, or any other quantitative or qualitative antibody binding assay available in the art. According to certain embodiments of the invention, two antigen-binding proteins bind to the same (or overlapping) epitope if, for example, a 1-fold, 5-fold, 10-fold, 20-fold, or 100-fold excess of one antigen-binding protein inhibits binding of the other by at least 50%, but preferably 75%, 90%, or even 99%, as measured in a competitive binding assay (see, e.g., Junghans et al., Cancer Res. 1990:50:1495-1502). Alternatively, two antigen-binding proteins are considered to bind to the same epitope if essentially all amino acid mutations in the antigen that reduce or eliminate binding of one antigen-binding protein also reduce or eliminate binding of the other. Two antigen-binding proteins are considered to have "overlapping epitopes" if only a subset of amino acid mutations that reduce or eliminate binding of one antigen-binding protein also reduce or eliminate binding of the other.

[0107] To determine whether an antibody or its antigen-binding domain competes with a reference antigen-binding molecule for binding, the above-mentioned binding method is carried out in two ways. In the first way, the reference antigen-binding molecule is allowed to bind to CD28 protein (or MUC16 protein) under saturating conditions, and then the binding of the test antibody to the CD28 (or MUC16) molecule is evaluated. In the second way, the test antibody is allowed to bind to CD28 (or MUC16) molecules under saturating conditions, and then the binding of the reference antigen-binding molecule to the CD28 (or MUC16) molecule is evaluated. In both ways, if only the first (saturating) antigen-binding molecule can bind to the CD28 (or MUC16) molecule, it is concluded that the test antibody and the reference antigen-binding molecule compete for binding to CD28 (or MUC16). As will be recognized by those skilled in the art, an antibody that competes for binding with a reference antigen-binding molecule does not necessarily bind to the same epitope as the reference antibody, but may sterically block the binding of the reference antibody by binding to an overlapping or adjacent epitope.

[0108] Preparation of antigen-binding domains and bispecific molecules Antigen-binding domains specific for a particular antigen can be prepared by any antibody production technique known in the art. Once obtained, two different antigen-binding domains specific for two different antigens (e.g., CD28 and MUC16) can be appropriately positioned relative to one another to produce a bispecific antigen-binding molecule of the present invention using conventional methods. (A discussion of exemplary bispecific antibody formats that can be used to construct bispecific antigen-binding molecules of the present invention is provided elsewhere herein.) In certain embodiments, one or more individual components (e.g., heavy and light chains) of a multispecific antigen-binding molecule of the present invention are derived from a chimeric antibody, a humanized antibody, or a fully human antibody. Methods for producing such antibodies are well known in the art. For example, one or more heavy and / or light chains of a bispecific antigen-binding molecule of the present invention can be prepared using VELOCIMMUNE™ technology. Using VELOCIMMUNE™ technology (or any other human antibody production technique), a high-affinity chimeric antibody against a specific antigen (e.g., CD28 or MUC16) is first isolated with a human variable region and a mouse constant region. Antibodies are characterized and selected for desirable characteristics including affinity, selectivity, epitope, etc. The murine constant regions are replaced with the desired human constant regions to generate fully human heavy and / or light chains that can be incorporated into the bispecific antigen-binding molecules of the invention.

[0109] Genetically engineered animals can also be used to produce human bispecific antigen-binding molecules. For example, genetically engineered mice can be used that are unable to rearrange and express endogenous mouse immunoglobulin light chain variable sequences, and the mice express only one or two human light chain variable domains encoded by human immunoglobulin sequences operably linked to the mouse kappa constant gene at the endogenous mouse kappa locus. Such genetically engineered mice can be used to produce fully human bispecific antigen-binding molecules that contain two different heavy chains associated with the same light chain that contains variable domains derived from one of two different human light chain variable region gene segments. (For a detailed discussion of such engineered mice and their use to produce bispecific antigen-binding molecules, see, for example, US2011 / 0195454).

[0110] biological equivalent The present invention encompasses antigen-binding molecules having amino acid sequences that differ from those of the described antibodies but retain the ability to bind to CD28 and / or MUC16. Such variant antibodies contain one or more amino acid additions, deletions, or substitutions compared to the parent sequence, but exhibit essentially equivalent biological activity to that of the described antigen-binding molecules. Similarly, DNA sequences encoding the antibody-binding molecules of the present invention encompass sequences that contain one or more nucleotide additions, deletions, or substitutions compared to the disclosed sequences, but are essentially biologically equivalent to the described antigen-binding molecules of the present invention. Examples of such variant amino acid and DNA sequences are discussed above.

[0111] The present invention includes antigen-binding molecules that are biologically equivalent to any of the exemplary antigen-binding molecules described herein. Two antigen-binding proteins or antibodies are considered to be bioequivalent if, for example, they are pharmaceutical equivalents or pharmaceutical alternatives that do not show significant differences in absorption rate and extent when administered at the same molar dose, either in a single dose or multiple doses, under similar experimental conditions. Some antibodies are considered to be equivalent or pharmaceutical alternatives if their absorption extent is equivalent but their absorption rate is not, and such differences in absorption rate are intentional and reflected in the labeling, and therefore can be considered bioequivalent, for example, they are not essential for achieving effective body drug concentrations in long-term use and are not considered medically significant for the particular drug product studied.

[0112] In one embodiment, two antigen binding proteins are bioequivalent if there are no clinically significant differences in their safety, purity, or efficacy.

[0113] In one embodiment, two antigen binding proteins are bioequivalent if a patient can be switched one or more times compared to therapy continued without switching between the reference product and the biological product without an expected increase in the risk of adverse effects, including clinically significant changes in immunogenicity or reduced efficacy.

[0114] In one embodiment, two antigen binding proteins are biologically equivalent if they both act by a common mechanism or mode of action for a condition or condition of use, to the extent that such mechanism is known.

[0115] Bioequivalence can be demonstrated by in vivo and in vitro methods. Bioequivalence measurements include, for example, (a) in vivo tests in humans or other mammals in which the concentration of the antibody or its metabolites is measured as a function of time in blood, plasma, serum, or other biological fluids; (b) in vitro tests that correlate with and reasonably predict human bioavailability data; (c) in vivo tests in humans or other mammals in which the relevant acute pharmacological effect of the antibody (or its target) is measured as a function of time; and (d) in vivo tests in humans or other mammals in which the relevant acute pharmacological effect of the antibody (or its target) is measured as a function of time. d) Includes well-controlled clinical trials that establish the safety, efficacy, or bioavailability or bioequivalence of the antibody.

[0116] Biologically equivalent variants of the exemplary bispecific antigen-binding molecules shown herein can be constructed, for example, by making various substitutions of residues or sequences or by deleting terminal or internal residues or sequences that are not required for biological activity. For example, cysteine ​​residues that are not essential for biological activity can be deleted or substituted with other amino acids to prevent the formation of unnecessary or incorrect intramolecular disulfide bridges during renaturation. In other contexts, biologically equivalent antibodies can include variants of the exemplary bispecific antigen-binding molecules described herein that contain amino acid changes that modify the glycosylation characteristics of the antibody, for example, mutations that eliminate or remove glycosylation.

[0117] Species selectivity and species cross-reactivity According to certain embodiments, the present invention provides antigen-binding molecules that bind to human CD28 but not to CD28 from other species. The present invention also provides antigen-binding molecules that bind to human MUC16 but not to MUC16 from other species. The present invention also includes antigen-binding molecules that bind to human CD28 and CD28 from one or more non-human species, and / or antigen-binding molecules that bind to human MUC16 and MUC16 from one or more non-human species.

[0118] According to certain exemplary embodiments of the present invention, there is provided an antigen-binding molecule that binds to human CD28 and / or human MUC16, and optionally binds to one or more of mouse, rat, guinea pig, hamster, gerbil, pig, cat, dog, rabbit, goat, sheep, cow, horse, camel, cynomolgus monkey, marmoset, rhesus monkey, or chimpanzee CD28 and / or MUC16. For example, in certain exemplary embodiments of the present invention, there is provided a bispecific antigen-binding molecule comprising a first antigen-binding domain that binds to human CD28 and cynomolgus monkey CD28, and a second antigen-binding domain that specifically binds to human MUC16.

[0119] immune complex The present invention encompasses antigen-binding molecules conjugated to a therapeutic moiety ("immunoconjugate"), such as a cytotoxin, a chemotherapeutic agent, an immunosuppressant, or a radioisotope. Cytotoxic agents include any agent that is harmful to cells. Examples of cytotoxic agents and chemotherapeutic agents suitable for forming immunoconjugates are known in the art (see, e.g., WO05 / 103081).

[0120] Therapeutic Formulations and Administration The present invention provides pharmaceutical compositions comprising the antigen-binding molecules of the present invention. The pharmaceutical compositions of the present invention are formulated with suitable carriers, excipients, and other agents that provide improved transfer, delivery, tolerance, etc. Many suitable formulations can be found in formularies known to all pharmacists: Remington's Pharmaceutical Sciences, Mack Publishing Company, Easton, PA. These formulations include, for example, powders, pastes, ointments, jellies, waxes, oils, lipids, lipid (cationic or anionic) containing vesicles (such as LIPOFECTIN™, Life Technologies, Carlsbad, CA), DNA complexes, anhydrous absorption pastes, oil-in-water and water-in-oil emulsions, emulsion carbowax (polyethylene glycol of various molecular weights), semi-solid gels, and semi-solid mixtures containing carbowax. Powell et al., "Compendium" of excipients for parenteral formulations”PDA(1998)J Pharm Sci Technol 52:238-3 See also 11.

[0121] The dose of an antigen-binding molecule administered to a patient may vary depending on the patient's age and size, target disease, pathological condition, route of administration, etc. Preferred doses are typically calculated according to body weight or body surface area. When the bispecific antigen-binding molecule of the present invention is used for the treatment of adult patients, it may be advantageous to administer the bispecific antigen-binding molecule of the present invention intravenously in a single dose of typically about 0.01 to about 20 mg / kg body weight, more preferably about 0.02 to about 7, about 0.03 to about 5, or about 0.05 to about 3 mg / kg body weight. The frequency and duration of treatment can be adjusted depending on the severity of the condition. Effective doses and schedules for administering bispecific antigen-binding molecules can be determined empirically; for example, the patient's progress can be monitored by periodic evaluation and the dose adjusted accordingly. Furthermore, interspecies scaling of dosages can be performed using methods well known in the art (e.g., Mordenti et al., 1991, Pharmaceut. Res. 8:1351).

[0122] Various delivery systems, e.g., encapsulation in liposomes, microparticles, microcapsules, recombinant cells capable of expressing mutant viruses, receptor-mediated endocytosis, are known and can be used to administer the pharmaceutical compositions of the present invention (see, e.g., Wu et al. (See, e.g., [Delta], [Delta], [Delta], [Delta], [Delta], [Delta], and [Delta]. al. (1987) J. Biol. Chem. 262:4429-4432. Methods of introduction include, but are not limited to, intradermal, transdermal, intramuscular, intraperitoneal, intravenous, subcutaneous, intranasal, epidural, and oral routes. The compositions can be administered by any convenient route, for example, by infusion or bolus injection, by absorption through epithelial or mucocutaneous linings (e.g., oral mucosa, rectal and intestinal mucosa, etc.), and can be administered together with other biologically active agents. Administration can be systemic or local.

[0123] The pharmaceutical composition of the present invention can be delivered subcutaneously or intravenously using a standard needle and syringe. In addition, for subcutaneous delivery, a pen delivery device facilitates application when delivering the pharmaceutical composition of the present invention. Such a pen delivery device can be reusable or disposable. Reusable pen delivery devices generally utilize a replaceable cartridge containing the pharmaceutical composition. Once the pharmaceutical composition in the cartridge has been administered and the cartridge is empty, the empty cartridge can be easily discarded and replaced with a new cartridge containing the pharmaceutical composition. The pen delivery device can then be reused. In disposable pen delivery devices, there is no replaceable cartridge. Rather, the disposable pen delivery device is pre-filled with the pharmaceutical composition held in a reservoir within the device. Once the reservoir is emptied of the pharmaceutical composition, the entire device is discarded.

[0124] A number of reusable pen and autoinjector delivery devices have utility for subcutaneous delivery of the pharmaceutical compositions of the present invention. Examples include AUTOPEN™ (Owen Mumford, Inc., Woodstock, UK), DISETRONIC™ pen (Disetronic Medical Systems, Bergdorf, Switzerland), HUMALOG MIX 75 / 25™ pen, HUMALOG™ pen, HUMALIN 70 / 30™ pen (Eli Lilly and Co., Indianapolis, IN), NOVOPEN™ I, II, and III (Novo Nordisk, Copenhagen, Denmark), NOVOPEN JUNIOR™ (Novo Nordisk, Copenhagen, Denmark), BD™ pen (Becton Dickinson, Franklin Lakes, NJ), OPTIPEN™, OPTIPEN PRO™, OPTIPEN STARLET™, to name just a few. and OPTICLIK™ (Sanofi-Aventis, Frankfurt, Germany). Examples of disposable pen delivery devices that have application in the subcutaneous delivery of the pharmaceutical compositions of the present invention include, but are not limited to, the SOLOSTAR™ pen (Sanofi-Aventis), FLEXPEN™ (Novo Nordisk), and KWIKPEN™ (Eli Lilly), the SURECLICK™ auto-injector (Amgen, Thousand Oaks, CA), PENLET™ (Haselmeier, Stuttgart, Germany), EPIPEN (Dey, LP), and the HUMIRA™ pen (Abbott Labs, Abbott Park IL), to name just a few.

[0125] In certain circumstances, pharmaceutical compositions can be delivered in a sustained-release system. In one embodiment, a pump can be used (see Langer, supra; Sefton, 1987, CRC Crit. Ref. Biomed. Eng. 14:201). In another embodiment, a polymeric material can be used; see Medical Applications of Controlled Release, Langer and Wise (eds.), 1974, CRC Pres., Boca Raton, Florida. In yet another embodiment, a sustained-release system can be placed in the vicinity of the target of the composition, thereby requiring only a fraction of the systemic dose (see, e.g., Goodson, 1984, in Medical Applications of Controlled Release, supra, vol. 2, pp. 115-138). Other sustained-release systems are discussed in the review by Langer, 1990, Science 249:1527-1533.

[0126] Injectable preparations may include dosage forms for intravenous injection, subcutaneous injection, intradermal injection, intramuscular injection, drip infusion, etc. These injectable preparations may be prepared by publicly known methods. For example, injectable preparations may be prepared by dissolving, suspending, or emulsifying the above-mentioned antibody or its salt in a sterile aqueous or oily medium conventionally used for injections. Aqueous media for injection include, for example, physiological saline, glucose-containing isotonic solutions, and other adjuvants, which may be used in combination with appropriate solubilizers such as alcohols (e.g., ethanol), polyalcohols (e.g., propylene glycol, polyethylene glycol), and nonionic surfactants [e.g., polysorbate 80, HCO-50 (polyoxyethylene (50 mol) adduct of hydrogenated castor oil)]. Oily media include, for example, sesame oil and soybean oil, which may be used in combination with solubilizers such as benzyl benzoate and benzyl alcohol. Injections prepared in this manner are preferably filled into appropriate ampoules.

[0127] Advantageously, the pharmaceutical compositions for oral or parenteral use described above are prepared into a suitable unit dose dosage form to accommodate the dose of the active ingredient. Such unit dose dosage forms include, for example, tablets, pills, capsules, injections (ampoules), suppositories, etc. The amount of antibody contained therein is generally about 5 to about 500 mg per unit dose dosage form, and particularly in the form of injection, the amount of antibody contained therein is preferably about 5 to about 100 mg, and for other dosage forms, about 10 to about 250 mg.

[0128] Therapeutic Uses of Antigen-Binding Molecules The present invention includes methods comprising administering to a subject in need thereof a therapeutic composition comprising an anti-CD28 antibody or a bispecific antigen-binding molecule that specifically binds to CD28 and a target antigen (e.g., MUC16). The therapeutic composition may comprise any of the antibodies or bispecific antigen-binding molecules disclosed herein and a pharmaceutically acceptable carrier or diluent. As used herein, the phrase "subject in need thereof" refers to a subject in need thereof who is suffering from one or more symptoms of cancer. Or a symptomatic human or non-human animal (e.g., a subject developing a tumor or suffering from any of the cancers described herein below), or who would otherwise benefit from inhibition or reduction of MUC16 activity or depletion of MUC16+ cells.

[0129] The antibodies and bispecific antigen-binding molecules of the present invention (and therapeutic compositions comprising them) are useful, inter alia, for the treatment of any disease or disorder in which stimulating, activating, and / or targeting an immune response is beneficial. In particular, the anti-CD28 / anti-MUC16 bispecific antigen-binding molecules of the present invention can be used for the treatment, prevention, and / or amelioration of any disease or disorder associated with or mediated by MUC16 expression or activity, or the proliferation of MUC16+ cells. The mechanism of action by which the therapeutic methods of the present invention are achieved involves the killing of MUC16-expressing cells in the presence of effector cells, e.g., T cells. MUC16-expressing cells that can be inhibited or killed using the bispecific antigen-binding molecules of the present invention include, for example, tumorigenic ovarian cells.

[0130] The antigen-binding molecules of the present invention can be used to treat primary and / or metastatic tumors occurring in, for example, subtypes of colon, lung, breast, renal, and bladder cancer. According to certain exemplary embodiments, the bispecific antigen-binding molecules of the present invention are used to treat ovarian cancer.

[0131] The present invention also includes methods for treating residual cancer in a subject. As used herein, the term "residual cancer" refers to the presence or persistence of one or more cancerous cells in a subject after treatment with an anti-cancer therapy.

[0132] According to certain embodiments, the present invention provides methods for treating a disease or disorder associated with MUC16 expression (e.g., a MUC16-expressing cancer, such as ovarian cancer), comprising administering to a subject one or more of the bispecific antigen binding molecules described elsewhere herein after the subject has been shown to be non-responsive to other types of anti-cancer therapy. For example, the present invention includes methods for treating ovarian cancer, comprising administering to a patient an anti-CD28 / anti-MUC16 bispecific antigen binding molecule 1 day, 2 days, 3 days, 4 days, 5 days, 6 days, 1 week, 2 weeks, 3 weeks, or 4 weeks, 2 months, 4 months, 6 months, 8 months, 1 year, or more after the subject has received standard treatment for patients with cancer, e.g., ovarian cancer. In other embodiments, a bispecific antigen-binding molecule of the invention comprising an IgG4 Fc domain (anti-CD28 / anti-MUC16 bispecific antigen-binding molecule) is administered to a subject first at one or more time points (e.g., to provide robust initial depletion of ovarian cancer cells), followed by administration of an equivalent bispecific antigen-binding molecule comprising a different IgG domain, such as an IgG1 Fc domain, at a subsequent time point. It is contemplated that the anti-CD28 / anti-MUC16 antibody of the invention can be used in conjunction with other bispecific antigen-binding molecules, such as an anti-PSMA / anti-MUC16 bispecific antibody. It is also contemplated that the bispecific antibody of the invention can be used in conjunction with checkpoint inhibitors, such as those targeting PD-1 and CTLA-4, as well as other targets. It may be advantageous to combine two bispecific antibodies targeting the same tumor antigen (e.g., MUC16), where one bispecific antibody targets CD3 on T cells and the other bispecific antibody targets a costimulatory molecule such as CD28. This combination can be used alone to enhance tumor cell killing, or in combination with a checkpoint inhibitor.

[0133] Exemplary MUC16-expressing cancers include, but are not limited to, ovarian cancer, breast cancer, endometrial cancer, pancreatic cancer, non-small cell lung cancer, intrahepatic cholangiocarcinoma mass-forming type, adenocarcinoma of the uterine cervix, and adenocarcinoma of the gastrointestinal tract.

[0134] Combination Therapies and Formulations The present invention includes compositions and therapeutic formulations comprising any of the exemplary antibodies and bispecific antigen-binding molecules described herein in combination with one or more additional therapeutically active ingredients, and methods of treatment comprising administering such combinations to a subject in need thereof.

[0135] Exemplary additional therapeutic agents that can be combined with or administered in combination with the antigen-binding molecules of the invention include, for example, chemotherapy, radiation therapy, checkpoint inhibitors that target PD-1 (e.g., anti-PD-1 antibodies such as pembrolizumab, nivolumab, or cemiplimab; see US 9,987,500), costimulatory agonist bivalent antibodies that target molecules such as CTLA-4, LAG3, TIM3, GITR, OX40, 4-1BB, CD3x bispecific antibodies (see, e.g., WO2017 / 053856A1, WO2014 / 047231A1, WO2018 / 067331A1, and WO2018 / 058001A1), other antibodies that target MUC16xCD3 (see, e.g., WO2017 / 053856A1), and other costimulatory CD28x bispecific antibodies.

[0136] Other agents that may be beneficially administered in combination with the antibodies of the invention include, for example, tamoxifen, aromatase inhibitors, and cytokine inhibitors, including small molecule cytokine inhibitors and antibodies that bind to cytokines such as IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, IL-8, IL-9, IL-11, IL-12, IL-13, IL-17, IL-18, or their respective receptors. Pharmaceutical compositions of the invention (e.g., pharmaceutical compositions comprising the anti-CD28 / anti-MUC16 bispecific antigen binding molecules disclosed herein) also include other anti-CD28 / anti-MUC16 drugs, such as "ICE": ifosfamide (e.g., Ifex®), carboplatin (e.g., Paraplatin®), etoposide (e.g., Etopophos®, Toposar®, VePesid®, VP-16), "DHAP": dexamethasone (e.g., Decadron®), cytarabine (e.g., Cytosar- U®, cytosine arabinoside, ara-C), cisplatin (e.g., Platinol®-AQ), and "ESHAP": etoposide (e.g., Etopofos®, Toposar®, VePesid®, VP-16), methylprednisolone (e.g., Medrol®), high-dose cytarabine, cisplatin (e.g., Platinol®-AQ).

[0137] The present invention also encompasses therapeutic combinations comprising any of the antigen-binding molecules described herein and one or more inhibitors of VEGF, Ang2, DLL4, EGFR, ErbB2, ErbB3, ErbB4, EGFRvIII, cMet, IGF1R, B-raf, PDGFR-o, PDGFR-I3, FOLH1, PRLR, STEAP1, STEAP2, TMPRSS2, MSLN, CA9, uroplakin, or any of the aforementioned cytokines, wherein the inhibitor is an aptamer, antisense molecule, ribozyme, siRNA, peptibody, nanobody, or antibody fragment (e.g., Fab fragment; F(ab')2 fragment; Fd fragment; Fv fragment; scFv; dAb fragment; or other engineered molecules such as diabodies, triabodies, tetrabodies, minibodies, and minimal recognition units). The antigen-binding molecules of the present invention can also be administered and / or co-formulated with antivirals, antibiotics, analgesics, corticosteroids, and / or NSAIDs. The antigen-binding molecules of the present invention may also be administered as part of a treatment regimen that also includes radiation therapy and / or conventional chemotherapy, or treatment with biologic agents, including checkpoint inhibitors or other bispecific antibodies.

[0138] The present invention includes compositions and therapeutic formulations comprising any of the antigen-binding molecules described herein in combination with one or more chemotherapeutic agents. Examples of chemotherapeutic agents include alkylating agents such as thiotepa and cyclophosphamide (Cytoxan™); alkylsulfonates such as busulfan, improsulfan, and piposulfan; aziridines such as benzodopa, carboquone, mesuredopa, and uredopa; ethylenimines and methylameramines, including altretamine, triethylenemelamine, triethylenephosphoramide, triethylenethiophosphatamide, and trimethylolmelamine; chlorambucil, ... Nitrogen mustards such as lunaphadine, chlorophosphamide, estramustine, ifosfamide, mechlorethamine, mechlorethamine oxide hydrochloride, melphalan, nobembine, phenesterine, prednimustine, trofosfamide, and uracil mustard; nitrosoureas such as carmustine, chlorozotocin, fotemustine, lomustine, nimustine, and ranimustine; aclacinomycin, actinomycin, autramycin, azaserine, bleomycin, cactinomycin, and camphor. Richeamicin, carabicin, carminomycin, carzinophilin, chromomycin, dactinomycin, daunorubicin, detorubicin, 6-diazo-5-oxo-L-norleucine, doxorubicin, epirubicin, esorubicin, idarubicin, marcellomycin, mitomycin, mycophenolic acid, nogalamycin, olivomycin, peplomycin, potofilomycin, puromycin, chelamycin, rodorubicin, streptonigrin, streptozocin, tubercidin, ubeni Antibiotics such as mexamic acid, zinostatin, and zorubicin; antimetabolites such as methotrexate and 5-fluorouracil (5-FU); folic acid analogues such as denopterin, methotrexate, pteropterin, and trimetrexate; purine analogues such as fludarabine, 6-mercaptopurine, thiamiprine, and thioguanine; pyrimidine analogues such as ancitabine, azacitidine, 6-azauridine, carmofur, cytarabine, dideoxyuridine, doxifluridine, enocitabine, and floxuridine;Androgens such as calcitriol, dromostanolone propionate, epitiostanol, mepitiostane, and testolactone; antiadrenal agents such as aminoglutethimide, mitotane, and trilostane; folic acid supplements such as furoic acid; aceglatone; aldophosphamide glycosides; aminolevulinic acid; amsacrine; bestravcil; bisantrene; edatraxate; defofamine; demecolcine; diaziquone; eflornithine; elliptinium acetate; etoglucide; gallium nitrate; hydroxyurea; lentinan; lonidamine; mitoguazone; mitoxant lon; mopidamol; nitracrine; pentostatin; phenamet; pirarubicin; podophyllic acid; 2-ethylhydrazide; procarbazine; PSK™; razoxane; sizofiran; spirogermanium; tenuazonic acid; triaziquone; 2,2',2''-trichlorotriethylamine; urethane; vindesine; dacarbazine; mannomustine; mitobronitol; mitolactol; pipobroman; gacytosine; arabinoside ("Ara-C"); cyclophosphamide; thiotepa; taxanes, such as paclitaxel (Taxol™, Bristol-Myers Squibb Oncology, Princeton, NJ) and docetaxel (Taxotere™; Aventis Antony, France); chlorambucil; gemcitabine; 6-thioguanine; mercaptopurine; methotrexate; platinum analogs such as cisplatin and carboplatin; vinblastine; platinum; etoposide (VP-16); ifosfamide; mitomycin C; mitoxantrone; vincristine; vinorelbine; navelbine; novantrone; teniposide; daunomycin; aminopterin; xeloda; ibandronate; CPT-11; topoisomerase inhibitor RFS2000; difluoromethylornithine (DMFO); retinoic acid; esperamicin; capecitabine;and pharmaceutically acceptable salts, acids, or derivatives of any of the above. This definition includes, for example, antiestrogens that modulate or inhibit hormone action on tumors, such as tamoxifen, raloxifene, aromatase-inhibiting 4(5)-imidazoles, 4-hydroxytamoxifen, trioxifene, ketoxifene, LY117018, onapristone, and toremifene (Fareston); Also included are antihormonal agents that act to inhibit growth of the ovarian canal; antiandrogens such as flutamide, nilutamide, bicalutamide, leuprolide, and goserelin; and pharmaceutically acceptable salts, acids, or derivatives of any of the above.

[0139] The additional therapeutically active ingredient(s) may be administered immediately before, simultaneously with, or immediately after administration of the antigen-binding molecule of the present invention. (For purposes of this disclosure, such administration regimens will be considered to be administration of the antigen-binding molecule "in combination" with the additional therapeutically active ingredient.

[0140] The present invention includes pharmaceutical compositions in which the antigen-binding molecules of the present invention are co-formulated with one or more of the additional therapeutically active ingredient(s) described elsewhere herein.

[0141] Dosing regimen According to certain embodiments of the present invention, multiple doses of an antigen-binding molecule (e.g., an anti-CD28 antibody or bispecific antigen-binding molecule that specifically binds to MUC16 and CD28) can be administered to a subject over a predetermined period of time. The method according to this aspect of the present invention comprises sequentially administering multiple doses of the antigen-binding molecule of the present invention to the subject. As used herein, "sequentially administering" means that each dose of the antigen-binding molecule is administered to the subject at different time points, for example, on different days separated by a predetermined interval (e.g., several hours, several days, weeks, or months). The present invention includes methods comprising sequentially administering to a patient a single initial dose of the antigen-binding molecule, followed by one or more secondary doses of the antigen-binding molecule, and then optionally one or more tertiary doses of the antigen-binding molecule.

[0142] The terms "primary dose," "secondary dose," and "tertiary dose" refer to the time sequence of administration of an antigen-binding molecule of the present invention. Thus, a "primary dose" is a dose administered at the beginning of a treatment regimen (also referred to as a "baseline dose"), a "secondary dose" is a dose administered after the primary dose, and a "tertiary dose" is a dose administered after the secondary dose. The primary, secondary, and tertiary doses may all contain the same amount of antigen-binding molecule, but generally may differ from each other in terms of administration frequency. However, in certain embodiments, the amount of antigen-binding molecule contained in the primary, secondary, and / or tertiary dose differs from each other (e.g., adjusted up or down as appropriate) during the course of treatment. In certain embodiments, two or more doses (e.g., two, three, four, or five) are administered as a "loading dose" at the beginning of a treatment regimen, followed by subsequent doses (e.g., "maintenance doses") administered at a lower frequency.

[0143] In certain exemplary embodiments of the invention, each secondary and / or tertiary dose is 1 to 26 (e.g., 1, 1 1 / 2, 2, 2 1 / 2, 3, 3 1 / 2, 4, 4 1 / 2, 5, 5 1 / 2, 6, 6 1 / 2, 7, 7 1 / 2, 8, 8 1 / 2, 9, 9 1 / 2, 10, 10 1 / 2, 11, 11 1 / 2, 12, 12 1 / 2, 13, The phrase "immediately preceding dose" as used herein refers to a dose of an antigen-binding molecule in a multiple administration series that is administered to a patient prior to administration of the immediately following dose with no intervening doses.

[0144] The method according to this aspect of the invention may include administering any number of secondary and / or tertiary doses of an antigen-binding molecule (e.g., an anti-CD28 antibody or bispecific antigen-binding molecule that specifically binds MUC16 and CD28) to the patient. For example, in certain embodiments, only a single secondary dose is administered to the patient. In other embodiments, two or more (e.g., two, three, four, five, six, seven, eight, or more) secondary doses are administered to the patient. Similarly, in certain embodiments, only a single tertiary dose is administered to the patient. In other embodiments, two or more (e.g., two, three, four, five, six, seven, eight, or more) tertiary doses are administered to the patient.

[0145] In embodiments involving multiple secondary doses, each secondary dose may be administered at the same frequency as the other secondary doses. For example, each secondary dose may be administered to the patient 1-2 weeks after the immediately preceding dose. Similarly, in embodiments involving multiple tertiary doses, each tertiary dose may be administered at the same frequency as the other tertiary doses. For example, each tertiary dose may be administered to the patient 2-4 weeks after the immediately preceding dose. Alternatively, the frequency with which the secondary and / or tertiary doses are administered to the patient may vary over the course of the treatment regimen. The administration frequency may also be adjusted by the physician during the course of treatment depending on the needs of the individual patient after clinical testing.

[0146] In one embodiment, the antigen-binding molecule (e.g., a bispecific antigen-binding molecule that specifically binds to MUC16 and CD28) is administered to a subject as a weight-based dose. A "weight-based dose" (e.g., a dose in mg / kg) is a dose of an antibody or antigen-binding fragment thereof or a bispecific antigen-binding molecule that will vary depending on the subject's weight.

[0147] In another embodiment, the antibody or antigen-binding fragment thereof, or bispecific antigen-binding molecule is administered to the subject as a fixed dose. A "fixed dose" (e.g., a dose in mg) means that one dose of the antibody or antigen-binding fragment thereof, or bispecific antigen-binding molecule is used for all subjects, regardless of any specific subject-related factors, such as body weight. In certain embodiments, the fixed dose of the antibody or antigen-binding fragment thereof, or bispecific antigen-binding molecule of the present invention is based on a predetermined weight or age.

[0148] Generally, suitable doses of antigen-binding molecules of the present invention can range from about 0.001 to about 200.0 milligrams per kilogram of recipient body weight, generally from about 1 to 50 mg per kilogram of body weight. For example, antibodies or antigen-binding fragments thereof, or bispecific antigen-binding molecules can be administered at about 0.1 mg / kg, about 0.2 mg / kg, about 0.5 mg / kg, about 1 mg / kg, about 1.5 mg / kg, about 2 mg / kg, about 3 mg / kg, about 5 mg / kg, about 10 mg / kg, about 15 mg / kg, about 20 mg / kg, about 25 mg / kg, about 30 mg / kg, about 40 mg / kg, or about 50 mg / kg per single dose. Values ​​and ranges intermediate to the recited values ​​are also intended to be part of the present invention.

[0149] In some embodiments, the antigen-binding molecules of the present invention are administered as a fixed dose of about 25 mg to about 2500 mg. In some embodiments, the antigen-binding molecules of the present invention are administered as a fixed dose of about 25 mg, about 30 mg, about 50 mg, about 75 mg, about 100 mg, about 125 mg, about 150 mg, about 175 mg, 200 mg, about 225 mg, about 250 mg, about 275 mg, about 300 mg, about 325 mg, about 350 mg, about 375 mg, about 400 mg, about 425 mg, about 450 mg, about 475 mg, about 500 mg, about 525 mg, or about 600 mg. and / or administered as a fixed dose of about 550 mg, about 575 mg, about 600 mg, about 625 mg, about 650 mg, about 675 mg, about 700 mg, about 725 mg, about 750 mg, about 775 mg, about 800 mg, about 825 mg, about 850 mg, about 875 mg, about 900 mg, about 925 mg, about 950 mg, about 975 mg, about 1000 mg, about 1500 mg, about 2000 mg, or about 2500 mg. Values ​​and ranges intermediate to the recited values ​​are also intended to be part of this invention.

[0150] Diagnostic Uses of Antibodies The bispecific antibodies of the invention can also be used to detect and / or measure CD28 or MUC16 or CD28- or MUC16-expressing cells in a sample, e.g., for diagnostic purposes. For example, an anti-CD28 antibody x MUC16 antibody, or fragments thereof, can be used to detect abnormal expression (e.g., excessive expression) of CD28 or MUC16. The present invention may also be used to diagnose conditions or diseases characterized by CD28 or MUC16 (expression, underexpression, lack of expression, etc.). An exemplary diagnostic assay for CD28 or MUC16 may involve, for example, contacting a sample obtained from a patient with an antibody of the invention, where the antibody is labeled with a detectable label or reporter molecule. Alternatively, an unlabeled antibody can be used in diagnostic applications in combination with a secondary antibody that is itself detectably labeled. The detectable label or reporter molecule may be 3 H, 14 C. 32 P, 35 S, or 125The CD28 or MUC16 signaling molecule may be a radioisotope such as I, a fluorescent or chemiluminescent moiety such as fluorescein isothiocyanate or rhodamine, or an enzyme such as alkaline phosphatase, β-galactosidase, horseradish peroxidase, or luciferase. Specific exemplary assays that can be used to detect or measure CD28 or MUC16 in a sample include enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), and fluorescence-activated cell sorting (FACS). Samples that can be used in the CD28 or MUC16 diagnostic assays of the present invention include any tissue or body fluid sample obtainable from a patient that contains a detectable amount of CD28 or MUC16 protein or fragments thereof under normal or pathological conditions. Generally, the level of CD28 or MUC16 in a particular sample obtained from a healthy patient (e.g., a patient not suffering from a disease or condition associated with abnormal CD28 or MUC16 levels or activity) is measured to initially establish a baseline or standard level of CD28 or MUC16. This baseline level of CD28 or MUC16 can then be compared to the level of CD28 or MUC16 measured in a sample obtained from an individual suspected of having a disease or condition associated with CD28 or MUC16. [Example]

[0151] The following examples are presented to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the methods and compositions of the present invention, and are not intended to limit the scope of what the inventors regard as the invention. Efforts have been made to ensure accuracy with respect to numbers used (e.g., amounts, temperatures, etc.), but some experimental error and deviation should be accounted for. Unless otherwise indicated, parts are parts by weight, molecular weight is average molecular weight, temperature is in degrees Celsius, and pressure is at or near atmospheric.

[0152] background T cell activation is initiated when the T cell receptor (TCR) / CD3 complex binds to a peptide-MHC complex ("signal 1"). Activation is then reinforced by the engagement of a second "costimulatory" receptor, such as the CD28 receptor on the T cell, which binds to its cognate ligand(s) on the target cell ("signal 2"). Recently described CD3-based "bispecifics" replace traditional signal 1 and act by attaching a tumor-specific antigen (TSA) to one arm of the bispecific, allowing the T cell to bind to the tumor cell, and bridging the TCR / CD3 with the other arm. Some of these so-called TSA x CD3 bispecific antibodies have shown promising antitumor effects in cancer patients, but their activity has yet to be optimized. As described elsewhere herein, the present invention introduces a novel class of bispecific antibodies that mimic signal 2 by bridging a second TSA to the costimulatory CD28 receptor on the T cell. These bispecific antibodies were designated TSA x CD28 bispecific antibodies. As described herein, one exemplary antibody of the invention is specific for an ovarian cancer antigen (e.g., MUC16). Unlike CD28 superagonists, which broadly activate T cells and resulted in severe toxicity in early clinical trials, these TSAxCD28 bispecific antibodies exhibit limited activity and no toxicity when used alone in genetically humanized immunocompetent mouse models or primates. However, when combined with a TSAxCD3 bispecific antibody, the exemplary antibodies of the invention activate T cells and their This strengthened the artificial synapse between CD28 and target cells, enhanced T cell activation, and significantly improved the antitumor activity of CD3 bispecific antibodies in various xenogeneic and syngeneic tumor models. Combining this novel class of CD28 costimulatory bispecific antibodies with the new class of TSA×CD3 bispecific antibodies may provide a well-tolerated "off-the-shelf" antibody therapy with enhanced antitumor efficacy.

[0153] The ability of T cells to recognize and kill cellular targets, such as virus-infected or tumor cells, depends on a coordinated set of interactions. The most important of these is the recognition and binding of target cells by the TCR complex (containing the associated CD3γ, δ, ε, and ζ chains). This interaction is referred to as "signal 1" of T cell activation. The TCR can recognize viral or tumor peptides present in the groove of the MHC protein expressed on the surface of target cells. This binding is usually low affinity; therefore, successful triggering of signal 1 requires the clustering of many TCR complexes along the interface between the T cell and its target cell, known as the immune synapse (J.B. Huppa, M.M. Davis, T-cell-antigen recognition and the immunological synapse. Nat Rev Immunol 3, 973–983 (2003)). T cell activation and proliferation are then further promoted by additional interactions with costimulatory receptors such as CD28 ("Signal 2") (J.H.E. Sensten, Y.A. Helou, G. Chopra, A. Weiss, J.A. Bluestone, CD28 Costimulation: From Mechanism to Therapy. Immunity 44, 973-988 (2016)). T cell activation is enhanced when T cells recognize target cells via the TCR complex and engage Signal 2 via CD28 binding to its cognate ligand(s) (CD80 / B7.1 and / or CD86 / B7.2) on professional antigen-presenting cells or target cells. Similar to Signal 1, CD28-mediated Signal 2 is thought to occur through co-clustering at the immune synapse.

[0154] Conventional monoclonal antibodies targeting tumor-specific antigens (TSAs) have been used as antitumor therapeutic agents for the past 20 years (G. Salles et al., Rituximab in B-Cell Hematologic Malignancies: A Review of 20 Years of Clinical Experience. Adv Ther 34, 2232-2273 (2017); M.V. Mateos et al., Daratumab plus Bortezomib, Melphalan, and Prednisone for Untreated Myeloma. N Engl J Med 378, 518-528 (2018); W. Eiermann, G. International Herceptin Study, Trastuzumab combined with chemotherapy for the treatment of HER2-positive metastatic breast cancer: pivotal trial data. Ann Oncol 12 Suppl 1, S57-62 (2001); J.M. Connors et al., Brentuximab Vedotin with Chemotherapy for Stage III or IV Hodgkin´s Lymphoma.N Engl J Med 378,331-344(2018), V.Dieras et al.,Trastuzumab emtansine versus capecitabine plus lapatinib in patients with previously treated HER2-positive advanced breast cancer(EMILIA): a descriptive analysis of final overall survival results from a randomized, open-label, Phase 3 trial. Lancet Oncol 18, 732-742 (2017)). However, this class of antibodies has limited ability to induce T cell-mediated cytotoxicity and instead acted by promoting antibody-dependent cellular cytotoxicity (ADCC) and / or complement-dependent cytotoxicity (CDC) or by delivering toxins to tumor cells. Recently, a new class of bispecific antibodies (TSA × CD3) has emerged that can efficiently induce T cell-mediated killing of tumor cells by binding T cells to tumor cells and activating the CD3 / TCR complex via a surrogate mechanism (usually via the e chain of CD3), thus mimicking signal 1. An early version of such a bispecific antibody (one arm binding to CD19 on leukemia cells and the other arm binding to CD3) recently received regulatory approval for B-cell acute lymphoblastic leukemia (R. Bargou et al., Tumor regression in cancer patients by very low doses of a T cell engaging antibody. Science 321, 974-977 (2008); H. Kantarjian et al., Blinatumomab versus Chemotherapy for Advanced Acute Lymphoblastic Leukemia. N Engl J Med 376, 836-847 (2017)).Recently, more advanced versions of bispecific antibodies have been shown to have good activity against non-Hodgkin's lymphomas, targeting CD20 on these lymphomas (EJ Smith et al., A novel, native-format bispecific antibody triggering T-cell killing of B cells is robustly active in mouse tumor models and cynomolgus monkeys. Sci Rep 5, 17943 (2015); L.L. Sun et al., Anti-CD20 / CD3 T cell-dependent bispecific antibody for the treatment of B cell malignancies. Sci Transl Med 7, 287ra270 (2015); M. Bacac. et al.,CD20-TCB with Obinutuzumab Pretreatment as Next-Generation Treatment of Hematologic Malignancies.Clin Cancer Res 24,4785-4797(2018), R. Bannerji et al., Emerging Clinical Activity of REGN1979, an Anti-CD20xAnti-CD3 Bispecific Antibody, in Patients with Relapsed / Refractory Follicular Lymphoma(FL), Diffuse Large B-Cell Lymphoma (DLBCL), and Other B-Cell Non-Hodgkin Lymphoma (B-NHL) Subtypes. American Society of Hematology, (2018), L. Budde et al., Mosunetuzumab, a Full-Length Bispecific CD20 / CD3 Antibody, Displays Clinical Activity in Relapsed / Refractory B-Cell Non-Hodgkin Lymphoma (NHL): Interim Safety and Efficacy Results from a Phase 1 Study. American Society of Hematology, (2018)). However, although TSA × CD3 bispecific antibodies have emerged as an important new class of immunotherapy in hematological malignancies, a cross-sectional comparison (E A Zhukovsky, R J Morse, M V Maus, Bispecific antibodies and CARs: generalized immunotherapeutics harnessing T cell redirection. Curr Opin Immunol 40, 24-35 (2016)) suggests that in some cases, they may not achieve the level of efficacy seen with personalized chimeric antigen receptor T cell (CAR-T) therapy. This suggests that there is a gender.

[0155] One of the reasons for the powerful efficacy of CAR-T therapy is that the chimeric antigen receptor (CAR) is engineered to provide both signal 1 (via a portion of the CD3z cellular domain) and signal 2 (e.g., via a portion of the CD28 cellular domain) upon binding to its target on a tumor cell. Two CAR-T cell therapies have recently received FDA approval for B-cell malignancies, both of which work by binding to and targeting the antigen CD19 (S.S. Neelapu et al., Axicabtagene Ciloleucel CAR T-Cell Therapy in Refractory Large B-Cell Lymphoma. N Engl J Med 377, 2531-2544 (2017); S.J. Chuster et al., Chimeric Antigen Receptor T Cells in Refractory B-Cell Lymphomas. N Engl J Med 377, 2545-2554 (2017)). CAR-T cell approaches have been shown to reduce the risk of serious adverse effects such as cytokine release syndrome (CRS) and neurotoxicity (S.S.Neelapu et al. al., Chimeric antigen receptor T-cell therapy-assessment and management of toxicities. Nat Rev Clin Oncol 15,47-62(2018), J. Gust et al., Endothelial Activation and Blood-Brain Barrier Disruption in Neurotoxicity after Adoptive Immunotherapy with CD19 CAR-T Cells. Cancer Discov 7, 1404-1419 (2017), A. Shimabukuro-Vornhagen et al., Cytokine release syndrome. J Immunother Cancer 6, 56 (2018)), as well as the requirement for highly personalized manufacturing processes and pre-adjustment of chemotherapy regimens (SS Neelapu et al., Axicabtagene Ciloleucel CAR T-Cell Therapy in Refractory Large Cells B Cell Lymphoma. N Engl J Med 377, 2531-2544 (2017), SJ Schuster et al., Chimeric Antigen Receptor T Cells in Refractory B-Cell Lymphomas. N Engl J Med 377, 2545-2554 (2017), P. Salmikangas, N. Kinsella, P. Chamberlain, Chimeric Antigen Receptor T-Cells (CART-Cells) for Cancer Immunotherapy - Moving Target for Industry? Pharm Res 35, 152 (2018), Many patients are not considered suitable candidates.

[0156] The advantages of TSAxCD3 bispecific antibodies as a relatively well-tolerated, "off-the-shelf" therapeutic solution for a broader patient population could be further enhanced if their antitumor activity could be further optimized, particularly without sacrificing tolerability, or perhaps even further enhanced specificity for tumor cells as opposed to normal cells. To this end, it was hypothesized that pairing TSAxCD3 bispecific antibodies with a novel class of bispecific antibodies that independently activate signal 2 might offer opportunities for potentially improved efficacy and enhanced specificity. Therefore, a second class of bispecific antibodies was designed. These bispecific antibodies could engage either a second epitope on the same tumor-specific antigen or a second, distinct tumor antigen in conjunction with the costimulatory receptor CD28 expressed on T cells (TSAxCD28 bispecific antibodies). Combining TSA1xCD3 and TSA2xCD28 allows for directed and enhanced surrogate activation of T cells by triggering both signal 1 and signal 2, with specificity directed against both epitopes or both. It was thought that this would allow for greater antitumor activity, with the opportunity for increased specificity, by targeting only tumor cells expressing one antigen.

[0157] Described herein is the generation and testing of a TSA×CD28 costimulatory bispecific antibody (MUC16×CD28) that targets ovarian cancer and binds to MUC16, a large intramembrane glycoprotein expressed at high levels in keratinocyte carcinomas (H. Suh, K. Pillai, DL Morris, Mucins in pancreatic cancer: biological role, implications in carcinogenesis and applications in diagnosis and therapy. Am J Cancer Res 7, 1372-1383 (2017)), which is cleaved to release the ovarian tumor biomarker CA-125 (I. Mylonas et al., Immunohistochemical expression of the tumor marker CA-125 in normal, hyperplastic and malignant endometrial tissue. Anticancer Res 23, 1075-1080 (2003)). Toxicology studies in genetically humanized immunocompetent mice and cynomolgus monkeys demonstrate that these bispecific antibodies exhibit limited activity and no toxicity as single agents. However, these novel costimulatory bispecific antibodies can be effectively combined with a new class of TSA x CD3 bispecific antibodies to enhance antitumor responses in both xenogeneic and syngeneic tumor models. Collectively, these data suggest that combining this novel class of CD28-based bispecific antibodies (TSA x CD28) with CD3-based bispecific antibodies (TSA x CD3) could provide a well-tolerated, "off-the-shelf" biologic solution with significantly enhanced synergistic antitumor activity.

[0158] Example 1. Construction of anti-MUC16xCD28 antibodies Generation of anti-CD28 antibodies Anti-CD28 antibodies were obtained by immunizing VELOCIMMUNE® mice (i.e., engineered mice containing DNA encoding human immunoglobulin heavy chain and kappa light chain variable regions) with human CD28 fused to the Fc portion of mouse IgG2a or with cells expressing CD28 or DNA encoding CD28. Antibody immune responses were monitored by CD28-specific immunoassays. When the desired immune response was achieved, splenocytes were harvested and fused with mouse myeloma cells to maintain their viability and form hybridoma cell lines. The hybridoma cell lines were screened and selected to identify cell lines producing CD28-specific antibodies. Using this technique, several anti-CD28 chimeric antibodies (i.e., antibodies with human variable domains and mouse constant domains) were obtained. Furthermore, as described in US2007 / 0280945A1, several fully human anti-CD28 antibodies were isolated directly from antigen-positive B cells without fusion to myeloma cells.

[0159] Certain biological properties of exemplary anti-CD28 antibodies generated according to the methods of this example are described in detail in the Examples set forth below.

[0160] Generation of anti-MUC16 antibodies Anti-MUC16 antibodies were obtained by immunizing genetically modified mice with the human MUC16 antigen or by immunizing engineered mice containing DNA encoding the human immunoglobulin heavy chain and kappa light chain variable regions with the human MUC16 antigen.

[0161] Genetically modified mice were immunized with hMUC16.nub (a truncated form of mucin 16 (SEQ ID NO: 49) encompassing the last five SEA domains) or with an hMUC16-expressing cell line such as OVCAR-3 cells. After immunization, splenocytes were harvested from each mouse and analyzed for: (1) mucin 16.nub (a truncated form encompassing the last five SEA domains of mucin 16 (SEQ ID NO: 49)) or hMUC16-expressing cell lines such as OVCAR-3 cells. (1) B cells were fused with mouse myeloma cells to maintain their viability and form hybridoma cells that were screened for MUC16 specificity, or (2) B cells were selected using human MUC16 fragments as selection reagents that bind to and identify reactive antibodies (antigen-positive B cells) (as described in US2007 / 0280945A1).

[0162] Chimeric antibodies against MUC162 were first isolated, each with a human variable region and a mouse constant region. The antibodies were characterized and selected for desirable characteristics, including affinity, selectivity, etc. If necessary, the mouse constant region was replaced with a desired human constant region, such as a wild-type or modified IgG1 or IgG4 constant region, to produce a fully human anti-MUC16 antibody. While the constant region selected can vary depending on the specific application, the high-affinity antigen-binding and target specificity characteristics reside in the variable region.

[0163] The specific biological properties of exemplary anti-MUC16 antibodies produced according to the methods of this example are described in detail in the Examples below.

[0164] Generation of a bispecific antibody that binds CD28 and MUC16 Bispecific antibodies comprising an anti-MUC16-specific binding domain and an anti-CD28-specific binding domain were constructed using standard methods, where the anti-MUC16 antigen-binding domain and the anti-CD28 antigen-binding domain each comprise a distinct HCVR paired with a common LCVR. In some cases, the bispecific antibodies were constructed utilizing a heavy chain from an anti-CD28 antibody, a heavy chain from an anti-MUC16 antibody, and a common light chain (see Table 5).

[0165] The bispecific antibody generated according to this example comprises two separate antigen-binding domains (i.e., binding arms). The first antigen-binding domain comprises a heavy chain variable region derived from an anti-CD28 antibody ("CD28-VH"), and the second antigen-binding domain comprises a heavy chain variable region derived from an anti-MUC16 antibody ("MUC16-VH"). Both anti-MUC16 and anti-CD28 share a common light chain. The CD28-VH / MUC16-VH pairing creates an antigen-binding domain that specifically recognizes CD28 on T cells and MUC16 on tumor cells.

[0166] Example 2. Amino acid and nucleic acid sequences of heavy and light chain variable regions Table 1 shows the amino acid sequence identifiers for the heavy and light chain variable regions and CDRs of selected anti-MUC16 antibodies of the invention. The corresponding nucleic acid sequence identifiers are listed in Table 2. [Table 2] [Table 3]

[0167] Table 3 shows the amino acid sequence identifiers for the heavy and light chain variable regions (HCVRs and LCVRs) and CDRs of selected anti-CD28 antibodies of the invention. The details are shown in Table 4. [Table 4] [Table 5]

[0168] A summary of the components of the various anti-MUC16xCD3 bispecific antibodies that were constructed is shown in Table 5. Tables 6 and 7 list the HCVRs, LCVRs, CDRs, and heavy and light chain sequence identifiers of the bispecific antibodies. [Table 6]

[0169] Table 6 shows the amino acid sequence identifiers for the bispecific anti-MUC16 x anti-CD28 antibodies exemplified herein. The corresponding nucleic acid sequence identifiers are set forth in Table 7. [Table 7] [Table 8]

[0170] Example 3. CD28 is a potent costimulatory receptor To determine which costimulatory receptors are effective in providing the costimulatory signal critical for T cell activation, a blind screen of costimulatory pathways performed by forced expression of costimulatory ligands on a panel of syngeneic tumors (Table 8 and Figure 1) again established CD28 as one of the most potent costimulatory receptors, along with 4-1BB. Table 8 summarizes the number of tumor-free mice in the blind screen. The assay was performed with three different tumor cell lines engineered to express seven different costimulatory ligands. The number in each square represents the number of tumor-free mice out of a total of five mice. [Table 9]

[0171] Example 4. Surface Plasmon Resonance-Derived Binding Affinity and Rate Constants of Anti-MUC16 x CD28 Bispecific Antibodies To determine the binding kinetics of exemplary anti-MUC16xCD28 bispecific monoclonal antibodies of the invention, the surface plasmon resonance-derived binding affinities and rate constants of anti-MUC16xCD28 bispecific antibodies to MUC16 and / or CD28 were determined.

[0172] The equilibrium dissociation constants (K) for the binding of hMUC16.mmh (SEQ ID NO: 51), hCD28.mmh (SEQ ID NO: 53), and mCD28.mmh (murine CD28.mmh, SEQ ID NO: 54) to purified exemplary anti-MUC16xCD28 bispecific monoclonal antibodies of the invention are shown. D The RI (RI value) was determined using a real-time surface plasmon resonance biosensor with a Biacore T-200 instrument. The CM5 Biacore sensor surface was derivatized by amine coupling with a monoclonal mouse anti-human Fc antibody to capture purified exemplary anti-MUC16 x CD28 bispecific antibodies of the invention. Two exemplary bispecific antibodies, bs24963D and REGN4615, were tested. REGN4615 is an antibody / scFv that recognizes human MUC16 and mouse CD28 and is sometimes referred to as anti-MUC16xmsCD28 antibody. The MUC16 arm of REGN4615 utilizes the VH and VK-ULC1-39 sequences as shown in Table 1 above for mAb8794P2. mCD28 (PV-1), with a light chain of SEQ ID NO: 11 (see also Figure 15A of US2004 / 0116675) and a heavy chain of SEQ ID NO: 13 (see Figure 15), is described in US2004 / 0116675 and was reformatted as an scFv for the experiments described herein.

[0173] The SPR binding studies were carried out in a pH 7.4 buffer (HBS-ET running buffer) consisting of 0.01 M HEPES (pH 7.4), 0.15 M NaCl, and 0.05% v / v surfactant P20. Different concentrations of hMUC16 with a C-terminal myc-myc-6xHis tag (hMUC16.mmh), hCD28 with a C-terminal myc-myc-6xHis tag (hCD28.mmh), and mCD28 with a C-terminal myc-myc-6xHis tag (mCD28.mmh) were prepared in HBS-ET running buffer as 3-fold serial dilutions ranging from 3.33 nM to 90 nM (for hMuc16) or 22.2 nM to 600 nM (for hCD28 or mCD28) for affinity determination for the anti-MUC16xCD28 and anti-MUC16xmCD28 bispecific antibodies.

[0174] The MASS-2 high-capacity amine sensor surface was first derivatized by amine coupling with a monoclonal mouse anti-human Fc antibody to capture approximately 500–900 RU of anti-MUC16×CD28 or anti-MUC16×mCD28 bispecific monoclonal antibodies. As defined by the manufacturer, 1 RU (response unit) is 1 mm 2 Each antibody represents 1 pg of protein. For affinity determination of anti-MUC16×CD28 and anti-MUC16×mCD28 bispecific antibodies, different concentrations of hMUC16 with a C-terminal myc-myc-6×His tag (hMUC16.mmh), hCD28 with a C-terminal myc-myc-6×His tag (hCD28.mmh), and mCD28 with a C-terminal myc-myc-6×His tag (mCD28.mmh) were prepared in HBS-ET running buffer as 3-fold serial dilutions ranging from 3.33 nM to 90 nM (hMuc16) or 22.2 nM to 600 nM (hCD28 or mCD28) and injected over anti-human Fc-captured anti-MUC16×CD28 or anti-MUC16×mCD28 bispecific monoclonal antibody surfaces at a flow rate of 50 μL / min for 5 min. Dissociation of bound hMUC16, hCD28, and mCD28 reagents was monitored for 10 min in HBS-ET running buffer. a ) and dissociation (kd The binding-dissociation equilibrium constant (K) was determined by fitting the real-time binding sensorgrams to a 1:1 binding model with mass transport limitation using Scrubber evaluation software version 2.0c. D ) and dissociation half-life (t1 / 2) were calculated from the kinetic rate constants as follows:

number

[0175] The binding kinetic parameters of exemplary bispecific antibodies binding to purified hMUC16, hCD28, mCD28 recombinant proteins at 37° C. are shown in Tables 9-12 below. [Table 10] [Table 11] [Table 12] [Table 13]

[0176] Example 5. Binding of anti-MUC16xCD28 bispecific antibodies to T cells and target cells To determine the binding of exemplary bispecific antibodies of the invention to human and cynomolgus monkey T cells and target cells, flow cytometry analysis was used to determine the binding of MUC16xCD28 bispecific antibodies to OVACR-3, PEO1, negative control Raji cells, human and cynomolgus monkey T cells, followed by detection with phycoerythrin (PE)- or Alexa-647-labeled anti-human IgG antibodies. Briefly, 1 x 10 5Cells / wells were incubated for 30 minutes at 4°C with serial dilutions of an exemplary MUC16 x CD28 bispecific antibody or an IgG4 isotype control that binds to human antigens without cross-reactivity with human or cynomolgus CD28. The dilutions ranged from 133 nM to 32.6 pM for human and cynomolgus T cells, and from 133 nM to 8.14 pM for Muc16-expressing cells and negative control Raji cells. After incubation, cells were washed twice with cold PBS containing 1% filtered FBS, and PE- or Alexa647-conjugated anti-human secondary antibodies were added to MUC16-expressing cells or human / cynomolgus T cells, respectively, and incubated for an additional 30 minutes. Live / dead dye was added to human and cynomolgus T cell incubations. Wells containing no antibody or only secondary antibody were used as controls.

[0177] After incubation with MUC16-expressing cells, cells were washed, resuspended in 200 μL of cold PBS containing 1% filtered FBS, and filtered on a BD FACS Canto II. Analysis was performed by flow cytometry.

[0178] After incubation with human or cynomolgus T cells, cells were washed and stained with a cocktail of anti-CD2, anti-CD16, anti-CD4, and anti-CD8 antibodies in Brilliant Stain Buffer for an additional 20 min incubation at 4° C. After washing, cells were resuspended in 200 μL of cold PBS containing 1% filtered FBS, gated into Live / CD2+ / CD4+ / CD16 or Live / CD2+ / CD8+ / CD16 gates, and analyzed by flow cytometry on a BD FACS LSR-Fortessa-X20.

[0179] Binding of exemplary MUC16xCD28 bispecific antibodies to the surface of human T cells was tested by flow cytometry. bs24963D bound 2.61x10 CD4+ T cells. -7 M, which bound to CD8+ T cells with an EC50 value of 2.53 × 10-7 M's EC 50 bs32897D bound to CD4+ T cells at a concentration of 9.16 × 10 -6 M's EC 50 It also bound weakly to CD8+ T cells with a value of 7.58 × 10 -6 M's EC 50 The results are summarized in Table 13. [Table 14]

[0180] Binding of exemplary MUC16xCD28 bispecific antibodies to the surface of cynomolgus monkey T cells was tested by flow cytometry. The exemplary bs24963D bound 2.03x10 CD4+ T cells. -7 M's EC 50 This resulted in a binding of 1.22 × 10 -7 M's EC 50 The exemplary bs32897D bound to OVCAR-3 and PEO1 cells at 2.87× values, respectively. -10 M and 5.96 x 10 -10 M's EC 50 The exemplary bs24963D did not exhibit any binding to the MUC16 negative control RAJI cells. The results are summarized in Table 14. [Table 15]

[0181] The binding of exemplary MUC16xCD28 bispecific antibodies to the surface of MUC16-expressing cell lines was tested by flow cytometry. bs24963D bound 6.09 x 10 OVCAR-3 and PEO1 cells, respectively. -10 M and 4.67 × 10 -10 M's EC 50 bs32897D did not exhibit any binding to the MUC16 negative control RAJI cells. bs24963D bound to OVCAR-3 and PEO1 cells at 2.87 x 10 -10 M and 5.96 x 10-10 M's EC 50 bs24963D did not exhibit any binding to MUC16-negative control RAJI cells. The isotype control antibody did not exhibit any binding to human or cynomolgus T cells, nor did it bind to MUC16-expressing cell lines. The results are summarized in Table 15. [Table 16]

[0182] Example 6. Primary bioassay of MUC16xCD28 bispecific antibodies T cell activation is achieved by stimulating the T cell receptor (TCR), which recognizes specific peptides presented by major histocompatibility complex class I or II (MHC1 or MHCII) proteins on antigen-presenting cells (APCs) (Goldrath et al., Selecting and maintaining a diverse T-cell repertoire, Nature 402, 255-262 (1999)). Activated TCRs, in turn, initiate a cascade of signaling events that can be monitored by reporter genes driven by various transcription factors, such as activator protein 1 (AP-1), nuclear factor of activated T cells (NFAT), or nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB). T cell responses are then further refined through the engagement of coreceptors expressed either constitutively or inducibly on T cells, such as CD28, CTLA-4 (cytotoxic T lymphocyte-associated protein 4), PD-1 (programmed cell death protein 1), LAG-3 (lymphocyte activation gene 3), or other molecules (Sharpe et al., The B7-CD28 Superfamily, Nat. Rev. Immunol., 2(2):116-26 (2002)). CD28, a costimulatory molecule, is activated by its endogenous ligands CD80 or CD86 expressed on APCs. Following TCR activation, CD28 enhances cellular signals, such as pathways controlled by the NFκB transcription factor. CD28 co-signaling is important for effective T cell activation, including T cell differentiation, proliferation, cytokine release, and cell death (Smeets et al., NFκB activation induced by T cell receptor / CD28 costimulation is mediated by protein kinase C-Θ, PNAS, 97(7):3394-3399 (2012)).

[0183] To identify antibodies that enhance T cell activity in the presence of both a primary stimulus and MUC16 target expression, exemplary anti-MUC16xCD28 bispecific antibodies of the invention were characterized in cell-based assays using human primary T cells. These assays evaluate the behavior of anti-MUC16 / CD28 bispecific antibodies in the presence and absence of a primary stimulus, and in the presence and absence of target expression.

[0184] Primary human CD4 + IL-2 Functional Assay Using T Cells: First generation CD4 + A T cell / APC functional assay was developed to assess the effect of CD28 activation on IL-2 production upon engagement with anti-MUC16xCD28 bispecific antibodies.

[0185] a) Isolation of human primary CD4+ T cells: Human peripheral blood mononuclear cells (PBMCs) were isolated from healthy donor leukocyte packs. PBMC isolation was achieved by density gradient centrifugation using 50 mL SepMate™ tubes according to the manufacturer's recommended protocol. Briefly, 15 mL of Ficoll-Paque PLUS was layered onto a 50 mL SepMate tube, followed by the addition of 30 mL of leukocytes diluted 1:2 with D-PBS. Subsequent steps were performed using the SepMate tube. The manufacturer's protocol was followed. + T cells were isolated from PBMCs using a human CD4 microbead kit from Miltenyi Biotec according to the manufacturer's protocol. + T cells were added at 5 x 10 per vial. 6 The cells were frozen in FBS containing 10% DMSO at a concentration of 100%.

[0186] b) Primary CD4 cells treated with CD28 antibody + IL-2 release from T cells: This assay uses human primary CD4 +T cells are activated via crosslinking of CD3 molecules complexed with the T cell receptor (TCR) using the αMuc16 × αCD3 bispecific antibody (REGN4018) incubated with human target cells OVCAR3 or PEO-1, which express Muc16 on their cell surface. Binding of the Muc16 arm of REGN4018 to Muc16-expressing target cells promotes clustering of CD3 molecules, providing the initial signal required for T cell stimulation in the absence or addition of an allogeneic response. However, in this assay, costimulation provided by crosslinking of CD28 molecules is required to complete T cell activation and increase the level of IL-2 release. Here, the bispecific anti-CD28 antibody inhibits CD4 + It interacts with CD28 on T cells and Muc16 on OVCAR3 or PEO-1 cells, promoting the clustering and activation of the costimulatory molecule CD28. The combined TCR-CD28 engagement enhances IL-2 production, which is released into the cell culture medium. IL-2 is detected and quantified in the cell supernatant using a homogenous, no-wash PerkinElmer AlphaLisa kit.

[0187] Previously isolated and frozen human CD4 from donor 104 + T cells were thawed on the day of the assay in stimulation medium (X-VIVO 15 cell culture medium supplemented with 10% FBS, HEPES, NaPyr, NEAA, and 0.01 mM BME) containing 50 U / ml benzonase nuclease. Cells were centrifuged at 1200 rpm for 10 min, resuspended in stimulation medium, and diluted to 1 × 10 5 Target cells were plated in 96-well round-bottom plates at a concentration of 1 × 10 cells / well. OVCAR3 and PEO-1 cells were treated with mitomycin C in primary stimulation medium using 25 μg / mL mitomycin C for OVCAR3 and 10 μg / mL for PEO-1 cells. After 1 hour of incubation at 37°C and 5% CO2, target cells were washed three times with wash buffer (PBS + 2% FBS) and plated at 1 × 10 cells / well. 4 OVCAR3 or 2.5 x 10 4 At a final concentration of PEO-1 cells, CD4+ The CD28 bispecific antibody or control antibody, diluted 1:4 from 3 pM to 200 nM, was then added to the wells in the presence of a constant 5 nM of REGN4018 (αMuc16 x αCD3) or a negative control antibody (hIgG4 isotype control = H4sH). The final dilution of the 10-point series contained no CD28 antibody, which resulted in background signal. After incubation of the plates at 37°C and 5% CO2 for 72 hours, they were centrifuged to pellet the cells, and 20 μL of the medium supernatant was collected. From this, 5 μL was tested in the human IL-2 Alpha LISA assay according to the manufacturer's protocol. Measurements were acquired using the Envision multilabel plate reader, and raw RLU (relative light units) values ​​were plotted. All serial dilutions were tested in duplicate.

[0188] Antibody EC 50 Values ​​were determined by fitting the data to a four-parameter logistic equation on a 10-point dose-response curve using GraphPad Prism™ software. Maximum fold induction was calculated using the following equation: [Table 17]

[0189] CD4+ T cell activation (measured by IL-2 release) was measured using a primary stimulus (anti-MUC16×C D3) and enhanced by hMUC16×hCD28 in the presence of MUC16 expressed on target cells.

[0190] c) Primary human CD4 + IL-2 Functional Assay Results Using T Cells: Anti-Muc16 × anti-CD28 bispecific antibody was used to detect CD4 T cells isolated in the absence or presence of a TCR-stimulating bispecific antibody (REGN4018 = anti-Muc16 × anti-CD3). +The ability of isolated human CD4 T cells to provide costimulation through CD28 on T cells was assessed by incubating them with target cells (OVAR3 and PEO-1 cells) that endogenously express Muc16 on their cell surface. + This was assessed in a functional IL-2 release assay using T cells.

[0191] CD4 co-cultured with OVCAR3 or PEO-1 cells in addition to a constant 5 nM of either hIgG4H4sH isotype control or REGN4018 anti-Muc16 × anti-CD3 + The fold induction values ​​for T cells are summarized in Tables 16 and 17.

[0192] Isolated CD4 + When T cells are incubated with OVCAR3 or PEO-1 target cells in the absence of directed TCR stimulation via REGN4018 using a constant amount of H4sH isotype control, the amount of IL-2 detected is similar between the CD28 parental antibody, the bispecific anti-Muc16 x anti-CD28 antibody (bs32897D and bs24963D), and the negative H4sH isotype control antibody (Table 16).

[0193] In contrast, elevated IL-2 levels are detected in samples treated with anti-Muc16 x anti-CD3 (REGN4018). Under these conditions, human CD4 + When T cells were co-cultured with OVCAR3 or PEO-1 cells, both CD28 bispecific antibodies increased IL-2 levels more than their respective parent CD28 antibodies. As expected, minimal IL-2 release was not observed with the isotype control (Table 17).

[0194] Using OVCAR3 as target cells, both CD28 bispecific antibodies (bs32897D: 5.63× and EC 50 = 606 pM) and bs24963D: 5.32 × and EC 50= 401 pM), a similar dose-dependent IL-2 release is measured. On the other hand, in PEO-1 cells, differences in fold induction of IL-2 levels were observed between both bispecific molecules. Here, bs24963D (10.94× and EC 50 =996 pM) produces higher IL-2 levels than bs32897D (5.22× and EC50 could not be determined because the dose-response curve did not reach saturation).

[0195] In the absence of TCR stimulation, either through allogeneic responses or promoted by anti-MUC16xCD3, no measurable IL-2 release is observed with CD28 antibody in wells containing a fixed amount of isotype control in the presence of OVCAR3 or PEO-1 cells (Table 16). Table 16 shows the release of IL-2 from CD4 cells co-cultured with OVCAR3 or PEO-1 and a fixed 5 nM isotype control. + EC of IL-2 release from T cells 50 Values ​​and fold induction are summarized. [Table 18]

[0196] In contrast, measurable IL-2 levels (RLU) were detected in samples treated with anti-MUC16xCD3. Under these conditions, human CD4 + When T cells were co-cultured with OVCAR3 or PEO-1 cells, both CD28 bispecific antibodies increased IL-2 levels more than their parent CD28 antibodies. As expected, no IL-2 release was observed with the isotype control (Table 17). Table 17 shows the results of CD4 T cells co-cultured with OVCAR3 or PEO-1 and a constant 5 nM anti-MUC16xCD3. + EC of IL-2 release from T cells 50 Values ​​and fold induction are summarized. [Table 19]

[0197] Example 7. It is concluded that the anti-MUC16xCD28 bispecific antibody enhances T cell activation and cytotoxicity against ovarian tumor cells in the presence of TCR stimulation with anti-MUC16xCD3. To investigate whether exemplary anti-MUC16xCD28 bispecific antibodies of the invention could enhance anti-MUC16xCD3-mediated T cell activation and cytotoxicity against ovarian tumor cells, FACS was used to examine tumor cell and phenotypic T cell survival after in vitro coculture with increasing doses of MUC16xCD3 alone or in combination with MUC16xCD28 (Figure 2A). Human peripheral blood mononuclear (PBMC) cells, including T cells, were cultured with PEO-1 ovarian cancer cells, which express endogenous levels of MUC16 (Coscia, F. et al., Nat. Commun. (2016), August 26;7:12645).

[0198] Two FACS-based cytotoxicity studies were performed. In the first study, FACS-based cytotoxicity was performed on MUC16+ cells in the presence of human peripheral blood mononuclear cells (PBMCs) and anti-MUC16xCD3, with or without anti-MUC16xCD28 stimulation (FACS-based cytotoxicity against MUC16+ cells + human PBMCs + / - MUC16xCD28 stimulation (MUC16xCD28 and MUC16xCD3 matrix setting)). The second study was identical to the first study, except that cynomolgus monkey PBMCs were used instead of human PBMCs (FACS-based cytotoxicity against MUC16+ cells + cynomolgus monkey PBMCs + / - MUC16xCD28 stimulation (MUC16xCD28 and MUC16xCD3 matrix setting)).

[0199] Experimental procedure To monitor the killing of MUC16+ cells in the presence of a combination of an anti-MUC16xCD3 antibody of the invention and an exemplary anti-MUC16xCD28 antibody, cell lines that endogenously express MUC16 (PEO1, MUC16 +) were labeled with 1 μM Violet Cell Tracker and plated overnight at 37°C. Separately, human PBMCs (New York Blood Center) or cynomolgus monkey PBMCs (Covance, Cranford NJ) were plated at 1 × 10 6 Cells were seeded at 1000 cells / mL in supplemented RPMI medium and incubated overnight at 37°C to enrich for lymphocytes by depleting adherent macrophages, dendritic cells, and some monocytes. The following day, target cells were co-incubated for 96 hours at 37°C with naive human PBMCs (effector / target cell ratio 4:1) depleted of adherent cells and serial dilutions of either anti-MUC16xCD3 or a non-targeting CD3-based bispecific antibody (bs17664D) alone or in combination with a fixed concentration (2.5 μg / mL) of an exemplary anti-MUC16xCD28 bispecific antibody. After incubation, cells were removed from the cell culture plate using trypsin-EDTA dissociation buffer and analyzed by flow cytometry (FACS).

[0200] For FACS analysis, cells were stained with a far-red cell tracker (Invitrogen) for viability, and antibodies directly conjugated to CD2, CD4, CD8, and CD25 (BD). Samples were run with calibration beads for cell counting. To assess the specificity of killing, target cells were gated as the violet cell tracker-positive population. The percent of live target cells was calculated as follows: Percentage of live cells = (R1 / R2) * 100 (where R1 = percentage of live target cells in the presence of antibody, and R2 = percentage of live target cells in the absence of test antibody). T cell activation was assessed by CD2 + / CD4 + or CD2 + / CD8 + ) Activation of T cells (CD25 +Upregulation of the PD-1 marker was assessed by incubating cells with directly conjugated antibodies against CD2, CD4, CD8, CD25, and PD-1 and reporting the percentage of PD-1+ T cells among total T cells (CD2+). T cell counts were calculated as the percentage of live CD4+ T cells per calibration bead. + or CD8 + The number of cells was measured by counting. The levels of cytokines accumulated in the culture medium were analyzed using the BD Cytometric Bead Array (CBA) Human Th1 / Th2 / Th17 Cytokine Kit according to the manufacturer's protocol.

[0201] Results, Summary, and Conclusions: The anti-MUC16xCD3 bispecific antibody was tested for its ability to induce naive human T cells to kill PEO1 target cells expressing human MUC16, either as a single agent or in the presence of costimulatory anti-MUC16xCD28 antibody. The anti-MUC16xCD3 bispecific antibody was induced to activate human T cells and deplete PEO1 cells. Furthermore, MUC16xCD3 alone induced moderate T cell killing of PEO-1 cancer cells, thereby reducing their viability to approximately 60% in a dose-dependent manner (Figure 2B and Table 18). Target cell killing was observed in the presence of the anti-MUC16xCD3 bispecific antibody, and PEO1 cells were killed at picomolar (pM) levels of EC 50 Target cell killing was observed in a dose-dependent manner at 100 μg / mL (Table 2B). In the absence of anti-MUC16×CD3, no target cell killing was observed (Figure 2B). The observed target cell lysis was again observed at picomolar (pM) levels of EC 50 This was associated with upregulation of CD25+ and PD-1+ T cells on CD2+ T cells in situ (Table 18).

[0202] Anti-MUC16xCD3 induced the release of human cytokines. The cytotoxic activity observed with the anti-MUC16xCD3 bispecific antibody as a single agent was enhanced in the presence of the exemplary anti-MUC16xCD28 costimulatory molecules bs24963D and bs32897D of the invention.

[0203] The addition of an exemplary anti-MUC16xCD28 antibody of the present invention was found to increase the potency and depth of cytotoxicity induced by MUC16xCD3, resulting in a further reduction in PEO-1 cancer cell survival to less than 20% (a more than three-fold increase in T cell killing) (Figure 2B). Furthermore, an exemplary anti-MUC16xCD28 antibody of the present invention increased the level of IFNγ release induced by MUC16xCD3 by more than 10-fold (Figure 2C). Combination of MUC16xCD28 with MUC16xCD3 resulted in proliferation of CD4 and CD8 T cells and increased expression levels of the activation marker CD25 (Figures 2D-E). Notably, MUC16xCD28 in combination with a non-targeting CD3 bispecific antibody did not induce T cell cytotoxicity or activation (Figure 2B).

[0204] In summary, costimulation increased T cell activation, PD-1 upregulation, and cytokine release when compared to that observed with anti-MUC16xCD3 as a single agent. Tables 18 and 19 and Figures 2A-2E summarize the results of experiments using human PBMCs. [Table 20]

[0205] The anti-MUC16xCD3 bispecific antibody was also tested for its ability to induce naive cynomolgus monkey T cells to kill target cells expressing human MUC16, either as a single agent or in the presence of a costimulatory anti-MUC16xCD28 bispecific antibody. Similar results were obtained when the same assay was performed using cynomolgus monkey PBMCs (Figure 2F-H). Figure 2I shows that an exemplary anti-MUC16xCD28 bispecific antibody of the invention binds to cellular targets as measured by flow cytometry. These results support the ability of the anti-MUC16xCD28 bispecific antibody of the invention to induce naive cynomolgus monkey T cells to kill target cells expressing human MUC16. We demonstrated that the anti-MUC16xCD28 bispecific antibody can potently enhance MUC16xCD3-mediated T cell activation not only through proliferation and cytokine release but also through cytotoxicity. At selected antibody titrations, the anti-MUC16xCD3 bispecific antibody activated human T cells but did not induce the T cells to deplete PEO1 cells (Table 19). Costimulation with exemplary anti-MUC16xCD28 antibodies of the invention increased T cell activation, enhanced cytotoxic activity, and upregulated the PD-1 marker on T cells (Table 19). [Table 21]

[0206] Example 8. In vivo studies of anti-MUC16xCD28 antibodies Combining anti-CD3xMUC16 and anti-CD28xMUC16 bispecific antibodies targeting tumor antigens enhanced tumor clearance in a mouse model. As shown in detail below, OVCAR-3 tumor growth was significantly inhibited in mice treated with an exemplary anti-CD3xMUC16 and exemplary anti-CD28xMUC16 of the invention compared to mice treated with anti-CD3xMUC16 alone or a control isotype.

[0207] To investigate whether MUC16×CD28 could enhance the antitumor effect of MUC16×CD3 in vivo, we used two different tumor models, a tumor xenograft ascites model and a tumor syngeneic mouse model, as detailed below.

[0208] Tumor xenogeneic ascites model In a tumor xenograft ascites model, OVCAR-3 ovarian cancer cells, a high-grade serous carcinoma of human origin that expresses endogenous high levels of MUC16, are implanted intraperitoneally into NSG mice pre-engrafted with human PBMCs. (Crawford A, Haber L, Kelly MP, Vazzana K, Canova L, Ram P, Pawashe A, Finney J, Jalal S, Chiu D, Colleton CA, Garnova E, Makonnen S, Hickey C, Krueger P, Delfino F, Potocky T, Kuhnert J, Godin S, Retter MW, Duramad P, MacDonald D, Olson WC, Fairhurst J, Huang T, Martin J, Lin JC, Smith E, Thurston G, Kirshner JR. A Mucin 16 bispecific T Cell-engaging antibodies for the treatment of ovarian cancer. Science Translational Medicine 19 Jun 2019:Vol. 11, Issue 497, eaau7534). OVCAR-3 cells were engineered with a luciferase reporter and tumor growth was tracked over time using in vivo bioluminescence (BLI).

[0209] Experimental procedure The experiment was carried out by (Crawford A, Haber L, Kelly MP, Vazzana K, Canova L, Ram P, Pawashe A, Finney J, Jalal S, Chiu D, Colleton CA, Garnova E, Makonnen S, Hickey C, Krueger P, Delfino F, Potoc Ky T, Kuhnert J, Godin S, Retter MW, Duramad P, MacDonald D, Olson WC, Fairhurst J, Huang T, Martin J, Lin JC, Smith E, Thurston G, Kirshner JR. A Mucin 16 bispecific T cell-engaging antibody for the treatment of ovarian cancer. Science Translational Medicine 19 Jun 2019: Vol. 11, Issue 497, eaau7534). Briefly, mice were IP injected with 150 mg / kg of the luciferase substrate D-luciferin (Perkin Elmer) suspended in PBS. Ten minutes later, mice were subjected to BLI imaging under isoflurane anesthesia using a Xenogen IVIS system (Perkin Elmer). Image acquisition was performed at a field of view of 100 mm, a subject height of 1.5 cm, and a medium binning level with a 0.5-minute exposure time. BLI signals were extracted using Living Image software (Xenogen; Alameda, CA). A region of interest was drawn around each tumor volume and the photon intensity was calculated in p / s / cm 2 BLI activity was recorded as / sr (photons per second per square centimeter per steradian). Mice that did not receive OVCAR-3 / Luc cells served as a baseline reading for BLI activity. These baseline tumor-free mice (N=3) were imaged each day, and the limit of detection (LOD) was calculated as the average BLI reading across all tumor-free mice imaged.

[0210] 5 × 10 NSG (NOD SCID gamma chain knockout) mice (Jackson Laboratory, MD) aged 8–10 weeks were injected with 5 × 10 6 Each mouse was injected with 1000 human PBMCs (ReachBio, Seattle, WA). 10–14 days later, blood was collected from the tail vein of the mouse to measure human T cell engraftment. Within 2 weeks of PBMC transfer, 2 × 10 human T cell lines from the OVCAR-3 / Luc cell line, which had been previously passaged in vivo, were injected. 6Ascites cells were administered intraperitoneally (IP) within 2 weeks (day 0). Mice were checked for T cell engraftment by flow cytometry and then assigned to groups using BLI to ensure similar tumor burden. Four days after tumor implantation, mice were dosed with 1.49 x 10 cells for two studies. 5 or 3.03 × 10 5 P / S / cm 2 Mice were divided into groups of 5 mice each, with a median BLI of 100 / sr. Mice were treated with the indicated bispecific antibody or control antibody on days 5 and 8. Mice were administered two doses of anti-MUC16xCD3 or a CD3-binding control by intravenous (IV) injection, with or without the exemplary anti-MUC16xCD28 (bs24963D) of the invention. Multiple imaging sessions were performed throughout the study to track tumor growth.

[0211] Serum cytokine levels from blood were also obtained at the indicated time points. Blood was collected by submandibular puncture into microtainer serum tubes (BD365967). Cytokine levels were analyzed using the V-plex Human ProInflammatory-10 Plex kit according to the manufacturer's instructions (Meso Scale Diagnostics, Rockville, MA).

[0212] All procedures were performed in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. The protocol was approved by the Regeneron Pharmaceuticals Institutional Animal Care and Use Committee. A total of two studies were completed, with five mice per group.

[0213] Results, Summary, and Conclusions Two models were used in the xenogeneic tumor studies. For the first xenogeneic model, NSG mice were injected intraperitoneally (IP) with OVCAR-3 / Luc cells that had been previously passaged in vivo (day 0) 13 days after human PBMC engraftment. Mice were treated IV on days 5 and 8. Mice were treated with 12.5 μg of anti-MUC16×CD3 or 12.5 μg of CD3-conjugated anti-MUC16×CD3. Control (hIgG4P-PVA Mice were given either an anti-MUC16xCD28 antibody or an anti-MUC16xCD3 antibody (isotype). Some mice also received 100 μg of an exemplary anti-MUC16xCD28 of the invention (bs24963D). Tumor burden was assessed by BLI at days 4, 8, 12, 15, 20, and 25 after tumor implantation. When exemplary bs24963D was administered without anti-MUC16xCD3, no tumor reduction was observed by BLI. In contrast, treatment with 12.5 μg of anti-MUC16xCD3 significantly reduced tumors evident by BLI, while exemplary anti-MUC16xCD28 of the invention significantly enhanced efficacy over anti-MUC16xCD3 alone (Tables 20-22).

[0214] Table 20 summarizes the bioluminescence levels at day 4 after tumor implantation in the first OVCAR-3 / Luc xenogeneic model. [Table 22]

[0215] Table 21 summarizes the bioluminescence levels at day 25 after tumor implantation in the first OVCAR-3 / Luc xenogeneic model. [Table 23]

[0216] Table 22 summarizes the fold change in BLI between days 4 and 25 after tumor implantation in the first OVCAR-3 / Luc xenogeneic model. [Table 24]

[0217] For the second xenogeneic model, NSG mice were injected with OVCAR-3 / Luc cells that had been previously passaged in vivo (day 0) 10 days after engraftment with human PBMCs. Mice were treated IV with 0.5 mg / kg anti-MUC16xCD3 or 0.5 mg / kg CD3-binding control on days 5 and 8. Tumor burden was assessed by BLI on days 4, 8, 11, 14, 21, 28, and 34. Some mice were also administered an exemplary anti-MUC16xCD28 (bs24963D) of the present invention at 0.2 mg / kg, 1 mg / kg, or 5 mg / kg. The exemplary bs24963D did not reduce tumor burden when administered without anti-MUC16xCD3. In contrast, treatment with 0.5 mg / kg anti-MUC16xCD3 significantly reduced BLI-evident tumors, while exemplary anti-MUC16xCD28 enhanced efficacy over anti-MUC16xCD3 alone (Tables 23-25 ​​and Figure 4A).

[0218] Table 23 summarizes bioluminescence levels at day 4 post tumor implantation in the second OVCAR-3 / Luc xenogeneic model. [Table 25]

[0219] Table 24 summarizes bioluminescence levels at day 25 post tumor implantation in the second OVCAR-3 / Luc xenogeneic model. [Table 26]

[0220] Table 25 summarizes the fold change in BLI between days 4 and 34 post tumor implantation in the second OVCAR-3 / Luc xenogeneic model. [Table 27]

[0221] Further results from a second xenogeneic model using different dosages are shown in Figure 3A. Mice treated with 2.5 μg of MUC16×CD3 on days 5 and 8 after tumor implantation significantly reduced tumor burden compared with mice treated with a CD3-binding control antibody (EGFRvIII×CD3), but did not completely eliminate OVCAR-3 / Luc tumor cells (Figure 3A). Combining 2.5 μg of MUC16×CD3 with 100 μg of MUC16×CD28 further inhibited tumor growth and resulted in more sustained tumor cell rejection over time (Figure 3A). Serum cytokine levels were also obtained in the same experiment. Figure 3B shows cytokine levels (pg / ml) in mice treated with different antibodies and / or antibody combinations. Figure 3C shows the correlation between tumor burden and serum CA-125 levels on day 26.

[0222] To test the ability of CD28 and CD3 bispecific antibodies to promote tumor killing in vivo, we used the well-established xenogeneic intraperitoneal ovarian OVCAR-3 tumor model. In this model, tumor cells are introduced into immunodeficient mice reconstituted with human PBMCs. Like other ovarian cancer cell lines, OVCAR-3 cells express MUC16. Prior to implantation, OVCAR-3 cells were engineered with a luciferase reporter to allow in vivo tracking of tumor growth over time using bioluminescence imaging (BLI). Implanted OVCAR-3 tumors grew unabated in mice treated with the EGFRvIII x CD3 bispecific antibody, a control CD3 bispecific antibody that did not bind to these cells, and in mice treated with the MUC16 x CD28 bispecific antibody alone (Figure 3A). The MUC16 x CD3 bispecific antibody alone was , showed significant antitumor activity but did not completely eliminate OVCAR-3 tumors (Figure 3A), whereas the addition of the MUC16xCD28 bispecific to the MUC16xCD3 bispecific antibody enhanced the in vivo antitumor effect compared to MUC16xCD3 alone (Figure 3A). Consistent with the enhanced antitumor activity, the combination of both bispecific antibodies also increased the secretion of circulating cytokines (Figure 3B).

[0223] The MUC16 bispecific antibody binds to the remaining "clumps" of MUC16 on the surface of ovarian cancer cells after proteolytic cleavage releases the prognostic ovarian cancer biomarker CA-125 (cell surface debris following cleavage and release of CA-125) (I. Mylonas et al. al., Immunohistochemical expression of the tumor marker CA-125 in normal, hyperplastic, and malignant endometrial tissue. Anticancer Res 23, 1075-1080 (2003)), but not soluble CA-125 (Figures 9A and 9B). To determine whether the MUC16×CD28 bispecific antibody impaired the ability to use CA-125 as a biomarker of ovarian tumor burden, we measured CA-125 levels in mice. CA-125 levels correlated with tumor burden regardless of treatment. The lowest CA-125 levels were seen in mice treated with the bispecific antibody combination, as previously shown for the MUC16×CD3 bispecific antibody (Figure 3C).

[0224] Syngeneic tumor models Experimental procedure Syngeneic studies were performed in mice genetically modified to express CD3 and portions of human MUC16 for the MC38 study using VelociGene® technology as previously described (Valenzuela et al., (2003) Nat. Biotechnol. June;21(6):652-9). (Crawford A, Haber L, Kelly MP, Vazzana K, Canova L, Ram P, Pawashe A, Finney J, Jalal S, Chiu D, Colleton CA, Garnova E, Makonnen S, Hickey C, Krueger P, Delfino F, Potocky T, Kuhnert J, Godin S, Retter MW, Duramad P, MacDonald D, Olson WC, Fairhurst J, Huang T, Martin J, Lin JC, Smith E, Thurston G, Kirshner JR. A Mucin 16 bispecific T cell-engaging antibody for the treatment of ovarian cancer.Science Translational Medicine 19 Jun 2019: Vol. 11, Issue 497, eaau7534). Mice expressing portions of human CD3, human CD28, and human MUC16 were used for the ID8-VEGF study. For CD3 humanization, a targeting vector was engineered to replace the extracellular portion (γδε) of the mouse CD3 gene with the corresponding human region of the gene. For CD28 humanization, a targeting vector was engineered to replace the extracellular portion of the mouse CD28 gene with the corresponding human region of the gene. For MUC16, mouse SEA repeats 13–17 were replaced with human SEA repeats 12–16. For each humanized mouse, correct gene targeting in F1H4 (C57BL / 6x129 hybrid) embryonic stem (ES) cell clones was identified by allelic loss assays as previously described (Poueymirou et al. (2007), Nat. Biotechnol. January;25(1):91–9). Targeted ES cells were injected into 8-cell stage Swiss Webster embryos to generate fully heterozygous F0 mice for homozygous breeding with C57BL / 6N mice (Taconic, Rensselaer, NY). Targeted ES cells were then transfected to express the human extracellular portion of CD3 (γδε), the human extracellular portion of CD28, and a portion of human MUC16. The resulting mice were bred to homozygosity (referred to as hCD3 / hMuc16 or hCD3 / hCD28 / hMUC16 humanized mice).

[0225] To investigate efficacy in immunocompetent models, knock-in mice were generated. The T cells of these mice express human CD3, and instead of mouse MUC16, a chimeric molecule containing a portion of human MUC16 to which an exemplary bispecific antibody of the present invention binds is expressed. Therefore, an anti-MUC16×CD3 molecule can be used in this study. To investigate whether the addition of a CD28-targeting bispecific molecule could enhance efficacy in these mice, a surrogate bispecific antibody was also generated. The surrogate antibody recognized human MUC16 but not mouse CD28 to test the effect of CD28 costimulation, and is sometimes referred to as anti-MUC16×mCD28. In a syngeneic tumor model, the MC38 cell line, engineered to express a portion of human MUC16, was used. Mice were implanted subcutaneously (SC) with MC38 / huMUC16 cells and treated with 0.01 mg / kg anti-MUC16×CD3 twice weekly on the day of implantation until day 21. Treatment with 0.01 mg / kg anti-MUC16xCD3 produced a significant antitumor effect, which was enhanced by the addition of MUC16xmCD28 (see Figures 6A, 6B, 6C, and 6D). Syngeneic Tumor Implantation and Measurement

[0226] 1 × 10 mice expressing human-mouse chimeras of human CD3 and MUC16 at the corresponding mouse loci 6 MC38 / huMUC16 cells were implanted subcutaneously (SC). Mice were administered anti-MUC16xCD3 or isotype control intraperitoneally (IP) twice weekly throughout the study until day 21, with or without a surrogate bispecific antibody recognizing human MUC16 and mouse CD28. Treatment began on the day of implantation. Tumor growth was measured twice weekly using calipers. Mice were sacrificed 50 days after tumor implantation.

[0227] Calculation of syngeneic tumor growth and inhibition To determine tumor volume with external calipers, the maximum major axis (length) and maximum transverse axis (width) were determined. Based on the caliper measurements, tumor volume was calculated using the formula: volume = (length × width) 2) / 2. Caliper measurements of the X and Y diameters were used to monitor tumor growth over time. 3 Mice were euthanized when the mean age exceeded 1. Statistical significance was determined using an unpaired, nonparametric Mann-Whitney t-test.

[0228] result The tumor sizes in the MC38 / huMUC16 model under different treatments are summarized in Table 26. [Table 28]

[0229] We tested whether an exemplary anti-MUC16xCD28 bispecific antibody of the invention enhanced the antitumor effect of MUC16xCD3 in a syngeneic mouse model of mice with a fully intact immune system. Mice were genetically engineered to express human CD3 and human MUC16 in place of the mouse genes using Velocigene technology (Crawford, 2004). A, Haber L, Kelly MP, Vazzana K, Canova L, Ram P, Pawashe A, Finney J, Jalal S, Chiu D,C Olleton CA, Garnova E, Makonnen S, Hickey C, Krueger P, Delfino F, Potocky T, Kuhnert J, Godin S, Retter MW, Duramad P, MacDonald D, Olson WC, Fairhurst J, Huang T, Martin J, Lin JC, Smith E, Thurston G, Kirshner JR. A Mucin 16 bispecific T cell-engaging antibody for the treatment of ovarian cancer. Science Translational Medicine 19 Jun 2019: Vol 11, Issue 497, eaau7534). The MC38 colon cancer cell line was engineered to express human MUC16 (pLVX.EF1a.MUC16, MC38 / hMUC16) and implanted subcutaneously. Mice were dosed twice weekly by intraperitoneal injection starting on the day of implantation (day 0) with isotype control (Iso Ctl), 0.01 mg / kg MUC16xCD3, 0.5 mg / kg MUC16xmCD28, or a combination thereof. Tumor growth was monitored over time (Figure 6A). Monotherapy with MUC16xCD3 or MUC16xCD28 significantly inhibited tumor growth. Tumor growth was even more significantly inhibited by combined treatment with MUC16xCD3 and MUC16xCD28 (Table 26). Serum cytokine levels were also obtained in the same experiment. Figure 6B shows cytokine levels in mice treated with different antibodies and / or antibody combinations.

[0230] Appropriate humanized mice, MC38 / hMUC16, were given implanted tumor cells and treated with control, individual CD3 or CD28 bispecific antibodies, or their combination (Figures 6A, 6C, and 6D). In the MUC16 tumor model, the combination of CD3 and CD28 bispecific antibodies produced the best antitumor response (Figure 6A), as indicated by cytokine production assays (Figures 6C and 6D).

[0231] To investigate whether the addition of MUC16xCD28 lead targeting could enhance efficacy in a syngeneic model, we used mice expressing human CD3 instead of mouse MUC16, human CD28 instead of mouse CD28, and a chimeric molecule containing the portion of human MUC16 to which an exemplary bispecific antibody of the present invention binds. The ID8-VEGF cell line was engineered to express human MUC16 (ID8-VEGF / hMUC16) and implanted intraperitoneally. Mice were dosed with 1 mg / kg of EGFRvIIIxCD3 or MUC16xCD3 alone or in combination with MUC16xCD28 on days 3, 6, and 10 after tumor implantation. Tumor growth was monitored using weight gain (Figure 5). Tumor growth was inhibited by MUC16xCD3, and the combination with MUC16xCD28 further delayed tumor growth.

[0232] Notably, unlike previous in vitro and in vivo analyses (see above) in which CD28 bispecific antibodies had very limited single-agent activity, the CD28 bispecific antibodies in this syngeneic MC38 / MUC16 model had more pronounced activity as single agents, suggesting that "signal 1" is already activated to some extent in these MC38 models. Consistent with this, MC38 tumor cells express high levels of reactivated endogenous retroviral proteins, such as p15E, and C57BL6 mice have previously been shown to generate endogenous T cells that recognize and respond to this neoepitope (JC Yang, D. Perry-Lalley, The envelope protein of an endogenous murine retrovirus is a tumor-associated T-cell antigen for multiple murine tumors. J Immunother 23, 177-183 (2000); HJ Zeh, 3rd, D. Perry-Lalley, M. E. Dudley, S. A. Rosenberg, JC Yang, High avidity CTLs for two self-antigens demonstrate superior in vitro and in vivo antitumor J Immunol 162, 989-994 (1999)). Indeed, in the MC38 model of the present invention, intratumoral T cells responding to this p15E neoantigen were readily detected (data not shown). Thus, the CD28 bispecific antibody in this MUC16 syngeneic tumor model can promote endogenous TCR / CD3-dependent T cell responses, which can be further enhanced by providing additional "signal 1" activation via the CD3 bispecific antibody.

[0233] It has long been recognized that T cell activation via the TCR complex ("signal 1") can be significantly enhanced by costimulatory signals ("signal 2"), such as those mediated when the CD28 receptor on the T cell engages its ligands (CD80 / B7.1 and CD86 / B7.2) on the target cell (JH E Sensten, YA Helou, G. Chopra, A. Weiss, J.A. Bluestone, CD28 Costimulation: From Mechanism to Therapy. Immunity 44, 973-988 (2016)). Consistent with our data, the potential of CD28 costimulation to enhance T cell antitumor activity was first demonstrated in studies in which B7 ligands were overexpressed on tumor cells (R.H. Schwartz, "Costimulation of T lymphocytes: the role of CD28, CTLA-4, and B7 / B1 in interleukin-2 production and immunotherapy," Cell 71, 1065-1068 (1992); L. Chen et al., "Costimulation of antitumor immunity by the B7 counterreceptor for the T lymphocyte molecules CD28 and CTLA-4," Cell 71, 1093-1102 (1992)) and demonstrated improved T cell rejection of such B7-expressing tumors. This potential motivated efforts to evaluate CD28-activating antibodies in human clinical trials. Unfortunately, a 2006 trial of such an antibody (TGN1412) resulted in life-threatening complications in all six human volunteers due to multiple organ failure caused by massive cytokine release syndrome (CRS) (G. Suntharalingam et al., Cytokine storm in a phase 1 trial of the anti-CD28 monoclonal antibody TGN1412. N Engl J Med 355, 1018-1028 (2006)). This catastrophe led to the halt of further trials of CD28 activating antibodies in humans.

[0234] CD28 bispecific antibodies, which do not directly activate CD28 unless clustered on the tumor cell surface, offer the possibility of facilitating costimulation exclusively at the tumor site without the systemic toxicity of conventional CD28-activating antibodies. Early versions of such CD28 bispecific antibodies were proposed and evaluated in the 1990s (C. Renner et al., Cure of xenografted human tumors by bispecific monoclonal antibodies and human T cells. Science 264, 833-835 (1994); G. Jung et al., Local immunotherapy of glioma patients with a combination of two bispecific antibody fragments and resting autologous lymphocytes: evidence for in situ T-cell activation and therapeutic efficacy. Int J Cancer 91, 225-230 (2001); M. Brandl, L. Grosse-Hovest, E. Holler, H.J. Kolb, G. Jung, Bispecific antibody fragments with CD20 x CD28 specificity allow effective autologous ous and allogeneic T-cell activation against malignant cells in peripheral blood and bone marrow cultures from patients with B-cell lineage leukemia and lymphoma. Exp Hematol 27, 1264-1270 (1999)). However, the early technologies available at the time required chemical crosslinking or hybrid / hybridoma fusion to create the proposed biopharmaceuticals, resulting in suboptimal reagents with extensive activity independent of clustering on tumor cells (possibly due to nonspecific aggregation of these bispecific antibodies, resembling conventional CD28 antibodies). Furthermore, these early approaches required in vitro T cell activation before any antitumor activity could be observed in vivo. Combined, the disastrous clinical results with the TGN1412 CD28 activating antibody and the limitations of these early CD28 bispecific antibody approaches led to the abandonment of further exploration of these approaches.

[0235] Described herein is a novel class of CD28 costimulatory bispecific antibodies that can significantly and safely enhance anti-tumor activity by providing a costimulatory "signal 2." While these CD28 bispecific antibodies have limited activity on their own (in the absence of "signal 1"), significantly enhanced anti-tumor activity in the context of "signal 1" can be provided by pairing these CD28 bispecific antibodies with a new class of CD3 bispecific antibodies (or when these CD28 bispecifics are used in situations where an endogenous population of tumor-specific T cells already exists). The creation, testing, and success of this new CD28 bispecific antibody approach is based on (1) the development of technologies (EJ Smith et al.) that were initially developed to produce CD3 bispecific antibodies, and for which these CD3 bispecific antibodies have been developed. al.,A novel,native-format bispecific antibody triggering T-cell killing of B cells is robustly active in mouse tumor models and cynomolgus monkeys.Sci Rep 5,17943(2015)) and clinical (A.Crawford et al.,REGN4018,a novel MUC16xCD3 bispecific T-cell engager for the treatment of ovarian cancer.Proceedings of the American Association for Cancer Research Annual Meeting (2) the utilization of a novel bispecific platform that has been recently validated on both fronts (Clinicaltrials.gov:NCT02290951, Clinicaltrials.gov:NCT03564340) and subsequently adapted to efficiently generate CD28 bispecific antibodies that exhibit minimal activity in the absence of specific "signal 1"; (3) the utilization of multiple xenogeneic and syngeneic genetically humanized CD28 bispecifics to evaluate these CD28 bispecifics alone and in combination with CD3 bispecifics (DM Valenzuela et al., High-throughput engineering of the mouse genome coupled with high-resolution expression analysis. Nat Biotechnol 21, 652-659 (2003); WT Poueymirou et al., F0 generation mice fully derived from gene-targeted embryonic stem cells allowing immediate phenotypic analyses. Nat Biotechnol 25, 91-99 (2007)) development of animal tumor models, (3) much deeper knowledge of cytokine release syndrome and its clinical development (A. Shimabukuro-Vornhagen et al.,Cytokine release syndrome.J Immunother Cancer 6,56(20. 18), DW Lee et al., Current concepts in the diagnosis and management of cytokine release syndrome. Blood 124, 188-195 (2014); C.L. Bonifant, H.J. Jackson, R.J. Brentjens, K.J. Curran, Toxicity and management in CAR T-cell therapy. Mol Ther Oncolytics 3, 16011 (2016)), along with validation in monkey models where the potential toxicity of any of these CD28 bispecific antibodies can be compared with that of conventional CD28 activating antibodies.

[0236] Described herein is the generation and testing of a TSA x CD28 costimulatory bispecific antibody (MUC16 x CD28) targeting the TSA of ovarian cancer. In the absence of "Signal 1," these CD28 bispecific antibodies were shown to have minimal activity in vitro or in vivo. However, these CD28 bispecific antibodies can be paired with CD3 bispecific antibodies to form an artificial "immune synapse" containing tumor antigens and TCR and CD28 complexes. Furthermore, when paired with an appropriate CD3 bispecific antibody in vitro, these CD28 bispecific antibodies can efficiently and specifically promote T cell activation and tumor cell killing in an antigen-dependent manner. Furthermore, these CD28 bispecific antibodies also efficiently enhance the antitumor activity of CD3 bispecific antibodies in vivo in a tumor antigen-specific manner in xenogeneic and syngeneic tumor models. In such models, CD28 bispecific antibodies have minimal single-agent activity unless tumor-specific T cells are already present, and in such settings, they appear to enhance this specific activity in a tumor antigen-dependent manner. Furthermore, combination therapy with TSAxCD28 and TSAxCD3 significantly enhances the expansion of intratumoral activated / memory T cell phenotypes in vivo. Finally, toxicology studies in genetically humanized immunocompetent mice and cynomolgus monkeys demonstrate that these bispecific antibodies exhibit limited activity and no toxicity as single agents in direct comparison with conventional CD28-activating antibodies.

[0237] Characterization of human-specific clinical candidates in the field of immuno-oncology is often limited to testing in xenogeneic tumor models using engrafted human immune cells. While these xenogeneic models (such as the utilized OVCAR3 model) are very convenient, they have limitations. The mice used in such xenogeneic models do not express the human tumor target in their normal tissues, thereby preventing the evaluation of test agents in the context of the target's normal tissue expression. Indeed, if the target is normally expressed at high levels in normal tissues, this could limit antitumor efficacy by diverting the test agent away from the tumor and could result in toxicity to these normal tissues, neither of which could be evaluated in xenogeneic models. Additional limitations may involve the activity of engrafted human peripheral blood mononuclear cells (PBMCs) transferred into immunodeficient mice, which may differ from the activity of normal host T cells found in the immunocompetent system. To overcome these limitations and provide a better model for testing human-specific clinical candidates, we created double and triple genetically humanized mice. In these models, tumor antigens were genetically humanized (for MUC16) to allow normal expression in appropriate host tissues, and CD3 and / or CD28 components were genetically humanized to allow immunocompetent host cells to respond to human-specific clinical candidates. As in xenogeneic animal models, these genetically humanized immunocompetent syngeneic animal models demonstrated enhanced antitumor activity of the appropriate CD3 bispecific antibody over the MUC16 tumor-targeted CD28 bispecific antibody. Similar enhancement of antitumor efficacy by different TSA x CD28 bispecific antibodies (e.g., MUC16 and PSMA (data not shown)) across multiple preclinical models suggests this therapy is robust and not limited to specific tumor models, and may have broader utility as a novel combination target class for immunotherapy. Overall, the findings suggest that TSA x CD28 bispecific antibodies synergize with TSA x CD3 bispecifics. These findings highlight the potential of this compound to significantly enhance the efficacy of well-studied TSAxCD3 bispecific antibodies in a reasonably safe and well-tolerated manner, potentially providing a biologic solution that could justify testing in human clinical trials.

[0238] TSA × CD3 bispecific antibodies represent a promising new class of immunotherapy, but further optimization of their antitumor activity is certainly required in many cases. Similar to the CAR-T approach, which employs chimeric receptors that artificially activate both "signal 1" and "signal 2" to improve antitumor activity (E.A. Zhukovsky, R.J. Morse, M.V. Maus, Bispecific antibodies and CARs: generalized immunotherapeutics harnessing T cell redirection. Curr Opin Immunol 40, 24-35 (2016); S.L. Maude et al., Tisagenlecleucel in Children and Young Adults with B-Cell Lymphoblastic Leukemia. N Engl J Med 378, 439-448 (2018)), we demonstrate here the potential benefits of combining a CD3-specific antibody (providing "signal 1") with a CD28 bispecific antibody (providing "signal 2") to enhance antitumor activity. In addition to the practical advantages such an approach offers over CAR-T therapy, it does not require laborious cell therapy preparations that must be individually customized for each patient, and allows patients to accept this cell therapy without the need to preemptively "lymphocyte deplete" it through toxic chemotherapy, which often involves side effects (A. Shimabukuro-Vornhagen et al., Cytokine release syndrome. J Immunother Cancer 6, 56 (2018); C.H. June, R.S. Connor, O.U. Kawalekar, S. Ghassemi, M.C. Milone, CAR T cell immunotherapy for human cancer. Science 359, 1361-1365 (2018)) - The bispecific approach according to the present invention offers the potential for improved efficacy as well as improved safety and specificity of action. That is, by pairing a CD3 bispecific antibody specific for one antigen with a CD28 bispecific antibody specific for a second antigen, it is possible to utilize "combinatorial targeting," where improved efficacy occurs only in tumor cells expressing both antigens, thus focusing T cell killing only on tumor cells expressing both antigens while limiting "off-target toxicity" in normal tissues expressing only one of the antigens. Taken together, the data presented here demonstrate that combining a CD28-based bispecific antibody with a CD3-based bispecific antibody may provide a well-tolerated, "off-the-shelf" biologic solution with significantly enhanced and synergistic antitumor activity. The first tests of this potential in human trials are expected this year.

[0239] Example 9. TSAxCD28 alone or in combination therapy does not induce systemic T cell activation compared to CD28 superagonists in cynomolgus monkeys Exemplary MUC16xCD28 antibodies of the invention enhance MUC16xCD3 activation of T cells from cynomolgus monkeys (Figures 2F-2H). To determine the safety and tolerability of exemplary anti-MUC16xCD28 bispecific antibodies of the invention alone or in combination with anti-MUC16xCD3, single-dose toxicology studies were conducted in cynomolgus monkeys. Female or male cynomolgus monkeys were assigned to treatment groups as shown in Table 27.

[0240] Cynomolgus monkey studies were conducted in accordance with IACUC guidelines. Male cynomolgus monkeys (3 per group) received a single dose of each test article via intravenous infusion over approximately 30 minutes (combined treatments were administered as separate infusions totaling 1 hour). Toxicity assessment was based on clinical observations, qualitative food consumption, body weight, neurological examination, vital signs (temperature, heart rate, pulse oximetry, and respiratory rate), and clinical and anatomic pathology. Cytokine analysis, Blood and tissue samples were collected for immunophenotyping, histopathology, and toxicokinetic evaluation. CRP levels were analyzed using a Roche Modular P800 system. Cytokines were measured by Meso Scale Diagnostics (MSD, Rockville, MD). For peripheral blood flow cytometry, blood was collected in potassium EDTA tubes, lysed, stained with anti-CD3, anti-Ki67, and anti-ICOS (BD Biosciences), and analyzed on a FACS Canto II.

[0241] Animals received a single dose of each test article via intravenous infusion over approximately 30 minutes (combination treatments were administered as separate infusions totaling 1 hour). Toxicity assessment was based on clinical observations, qualitative food consumption, body weight, neurological examination, vital signs (temperature, heart rate, pulse oximetry, and respiratory rate), and clinical and anatomic pathology. Blood samples were collected for cytokine analysis, FACS immunophenotyping, and toxicokinetic assessment. No significant cytokine release, T cell margination, or upregulation of T cell activation markers was observed after single administration of exemplary anti-MUC16xCD28 of the invention at 1 or 10 mg / kg, MUC16xCD3 at 1 or 10 mg / kg, or combination treatments. Table 27 summarizes various readouts, including absolute T cell counts, T cell activation markers (ki67), CRP, and serum cytokine levels from blood obtained from individual animals at the indicated time points. These findings were further validated using dry-coated and wet-coated human T cell proliferation assays, which demonstrated that immobilizing MUC16xCD28 on assay plates using either the dry- or wet-coating method did not induce T cell activation in the absence of CD3 stimulation, in contrast to CD28 superagonist antibodies (Figure 7). Indeed, it was found that exemplary anti-MUC16xCD28 bispecific antibodies of the invention, as well as the parent bivalent CD28 antibody, failed to induce human T cell proliferation compared to CD28 superagonist antibodies. Overall, exploratory single-dose toxicology studies in monkeys and in vitro human T cell-based assays suggest that exemplary anti-MUC16xCD28 bispecific antibodies of the invention are safe and well-tolerated. [Table 29]

[0242] Blood samples were collected for cytokine and flow cytometry immunophenotyping analysis It is noteworthy that CD28-SA administered to monkeys induced significant cytokine release, lymphocyte margination, and T cell activation, whereas no cytokine release, T cell margination, or T cell activation was observed after administration of MUC16×CD28 (Figures 8A-8C and Table 27). Overall, these preliminary observations suggest that the TSA×CD28 bispecific antibody is well tolerated in primates and does not induce cytokine release and T cell activation as seen with CD28-SA (data not shown). It should be noted that previous studies using CD28-SA in monkeys failed to predict the massive cytokine release and T cell activation seen in humans (Tegenaro AG, www.circare.org / foia5 / tgn1412investigatorbrochure.pdf), which was attributed to lower CD28 expression in monkeys (D. Eastwood et al., Monoclonal antibody TGN1412 trial failure explained by species differences in CD28 expression on CD4+effector memory T-cells.Br J Pharmacol 161, 512-526 (2010)). Even though tolerability studies in cynomolgus monkeys are not predictive of CRS in humans, the strong signal demonstrated by CD28-SA in monkeys suggests that Tegenaro et al. missed this by simply not investigating an appropriate early time point at which these responses could be robustly observed.

[0243] Example 10: bs24963D (MUC16xCD28, also referred to as REGN5668) Binding of Ab) and REGN4018 (MUC16xCD3) to cell lines expressing human or cynomolgus monkey MUC16, primary cells from human or cynomolgus monkey PBMCs, and T cell lines Materials and Methods - Summary of Experimental Procedures Flow cytometry analysis was used to determine the binding of bs24963D to human ovarian cancer cell lines (OVCAR-3 and PEO1) that endogenously express human MUC16, to murine ID8 cells engineered to express human or cynomolgus MUC16, to human and cynomolgus T cells, and to engineered reporter T cell lines.

[0244] In short, 1 x 10 5 Cells / well were incubated with bs24963D, REGN4018, and control antibody (IgG4 P-PVA The cells were incubated for 30 minutes at 4°C with serial dilutions of antibodies, including a non-binding control mAb, a CD28 non-crosslinking control bispecific antibody, or a parental CD28 or CD3 control.

[0245] Antibody dilutions ranged from 12.2 pM to 200 nM for human and cynomolgus primary T cells and engineered reporter T cells, while MUC16 + For target cells, 8.1 pM to 133 nM was selected.

[0246] After incubation, cells were washed twice with cold PBS containing 1% filtered FBS, followed by phycoerythrin (PE)-labeled anti-human IgG (MUC16 + cells) or Alexa647-labeled anti-human IgG (CD28 + Detection was performed using ELISA kits (cells).

[0247] Near-infrared (IR)-reactive LIVE / DEAD dye was added to human and cynomolgus T cells. Wells containing no antibody or secondary antibody alone served as controls.

[0248] MUC16 +After incubation with cells or the J.RT3.T3.5 / NF-κB-Luc / 1G4AB / hCD8αβ / hCD28 cell line, cells were washed, resuspended in 200 μL of FACS buffer (cold PBS containing 1% filtered FBS and 1 mM EDTA), and analyzed by flow cytometry on a BD FACS Canto II.

[0249] After incubation with human or cynomolgus T cells, cells were washed and stained with a cocktail of anti-CD2, anti-CD16, anti-CD4, and anti-CD8 antibodies in FACS buffer for 20 minutes at 4°C. After washing, cells were resuspended in FACS buffer and stained for Live / CD2 + / CD4 + / CD16 - or Live / CD2 + / CD8 + / CD16 - and analyzed by flow cytometry on a BD LSRFortessa X-20.

[0250] EC 50 For determination, the measured MFI was analyzed using a four-parameter logistic equation on a nine-point response curve using GraphPad Prism. The fold increase in maximum MFI was determined by taking the ratio of the highest MFI detected to the MFI of wells containing only secondary antibody.

[0251] Using flow cytometry, we investigated the MUC16 activity of bs24963D and a commercially available anti-PD-L1 antibody. + Binding to human pancreatic cancer cells, SW1990 and SW1990 / hPD-L1 cells, was determined. Briefly, 2 × 10 5Cells were incubated with 5μL (66.7nM) of bs24963D, anti-PD-L1 (2.5μL), or non-binding control antibody conjugated to AlexaFluor647 (bs24963D) or APC (anti-PD-L1) and incubated on ice for 30 minutes. Cells were washed once with staining buffer, centrifuged, and washed with D-PBS. Cells were stained with 100μL of a 1:1000 dilution of LIVE / DEAD Fixable Violet viability dye and incubated for 15 minutes at room temperature. Cells were washed three times in staining buffer, resuspended in 100μL of a 1:1 ratio of staining buffer and cytofix solution, and analyzed by flow cytometry using a Cytoflex cytometer. Fold binding relative to viability was calculated by dividing the MFI of the antibody of interest by the MFI of viability alone.

[0252] material and method NF-κB luciferase reporter bioassay As shown in Figure 10, the ability of bs24963D to enhance TCR-mediated signaling was assessed in an engineered T cell / antigen-presenting cell-based reporter assay. TCRs recognize specific MHC / peptide complexes and activate T cells through a number of transcription factors, such as activator protein 1 (AP-1), nuclear factor of activated T cells (NFAT), or nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) (Goldrath, 1999; Nature 402:255-62) (Shapiro, 1998; J. Immunology; 161(12):6455-8). T cell responses are further refined through the engagement of costimulatory receptors, such as CD28, which are subsequently activated by their endogenous ligands, CD80 or CD86, subsequently enhancing cellular signals, such as pathways controlled by the NF-κB transcription factor, following TCR activation.

[0253] In this assay, engineered T cells are directly activated via the 1G4 TCR (IG4AB), which recognizes the NY-ESO-1 157-165 peptide (NYESO1p) complexed with the human MHC class I molecule HLA-A2 and hβ2M displayed on engineered antigen-presenting 3T3 cells (Robbins, 2008; J. of Immunology; 180(9):6116-31). TCR activation results in the production of luciferase, which is promoted by the NF-κB transcription factor in the engineered reporter T cells. CD8 enhances TCR signaling through phosphorylation of intracellular immunoreceptor tyrosine-based activation motifs (ITAMs) by recruiting lymphocyte-specific protein tyrosine kinase (Lck) to the TCR / CD3 complex, facilitating TCR / MHC interaction and promoting T cell activation (Cole, 2012; Immunology; 137(2):139-48) (Guirado, 2002; Biochem. Biophys. s.Res.Comm.291(3):574-81).

[0254] A non-crosslinking control, bs24963D (non-TAA x CD28 bispecific antibody), or two-fold serial dilutions of a non-binding control (39 pM to 10 nM) were administered to MUC16 - (3T3 / hβ2M / HLA-A2 / NYESO1p) or MUC16 + (3T3 / hβ2M / HLA-A2 / NYESO1p / hMUC16) 4 5 x 10 antigen-presenting cells per well in the presence of 4 Engineered reporter T cells (J.RT3.T3.5 / NF-κB-Luc / 1G4AB / hCD8αβ / hCD28) were added in duplicate. Antibody dilutions and bioassays were performed in complete medium (RPMI supplemented with 10% FBS and a cocktail of penicillin, streptomycin, and L-glutamine). Wells without any antibody were included as an additional control to assess activity and EC 50The fold increase in values ​​was used to calculate the fold increase. After incubating the plate at 37°C and 5% CO2 for 5 hours, ONE-Glo luciferase substrate (100 μL) was added to each well. Luciferase activity was recorded as luminescence signal using an ENVISION plate reader and expressed as relative light units (RLU). Detected RLUs were analyzed by a four-parameter logistic equation on a 10-point response curve using GraphPad Prism.

[0255] The maximum activation signal was determined as the average maximum RLU response detected within the antibody concentration range tested. The fold increase in activity was calculated as the ratio of the highest average RLU value recorded within the antibody concentration range tested to the average RLU value recorded in the absence of antibody.

[0256] T cell activation assays for T cell proliferation and IL-2 release The ability of bs24963D to mediate IL-2 release and T cell proliferation in the presence of a fixed concentration of REGN4018 (assessed in the human ovarian cancer cell lines OVCAR-3 and PEO1) or a fixed concentration of cemiplimab (assessed in the human pancreatic cancer cell lines [SW1990 and SW1990 / hPD-L1]) was determined using a T cell activation assay with enriched human primary T cells from three or two donors, respectively.

[0257] Isolation of human primary T cells Human PBMCs were isolated from leukocyte packs of four healthy donors. For donors 555014 and 555109, PBMCs were isolated from peripheral blood using density gradient centrifugation. Briefly, 15 mL of Ficoll Plaque Plus was added to a 50 mL conical tube, followed by overlaying with 30 mL of blood diluted 1:1 with PBS containing 2% FBS. After centrifugation at 400 × g for 30 minutes with the brake off, the mononuclear cell layer was transferred to a new tube, diluted 5-fold with PBS containing 2% FBS, and centrifuged at 300 × g for 8 minutes. For donors 555131 and 555129, PBMCs were isolated from peripheral blood of healthy donors using the EasySep Direct Human PBMC Isolation Kit from Stem Cell Technologies according to the manufacturer's protocol. Isolated PBMCs were frozen in FBS containing 10% DMSO. CD3 + To isolate T cells, a vial of frozen PBMCs was thawed in a 37°C water bath and diluted in stimulation medium (X-VIVO 15 cell culture medium supplemented with 10% FBS, HEPES, NaPyr, NEAA, and 0.01 mM β-mercaptoethanol [BME]) containing 50 U / mL Benzonase® Nuclease. Cells were centrifuged at 1200 rpm for 10 minutes, resuspended in EasySep buffer, and isolated using StemCell Technologies' EasySep T-Cell Isolation kit according to the manufacturer's protocol.

[0258] Human OVCAR-3, PEO1, SW1990, SW1990 / hPD-L1 cells, and T cell activation assay using human primary T cells CD3 resuspended in stimulation medium (X-VIVO 5 cell culture medium supplemented with 10% FBS, HEPES, NaPyr, NEAA, and 0.01 mM BME) + T cells, 1 x 10 5OVCAR-3, PEO1, SW1990, or SW1990 / hPD-L1 cells were plated into 96-well round-bottom plates at a concentration of 1 × 10 cells / well. OVCAR-3, PEO1, SW1990, or SW1990 / hPD-L1 cells were treated with mitomycin C at 25 μg / mL (OVCAR-3), 10 μg / mL (PEO1), or 30 μg / mL (SW1990 and SW1990 / hPD-L1) to inhibit proliferation. After 1 hour of incubation at 37°C and 5% CO2, the mitomycin C-treated cells were washed three times with D-PBS containing 2% FBS, followed by a final resuspension in stimulation medium. OVCAR-3, PEO1, SW1990, and SW1990 / hPD-L1 cells were cultured at a concentration of 1 × 10 cells / well for OVCAR-3, PEO1, and both SW1990 cells, respectively. 4 , 2.5×10 4 cells, or 5 x 10 4 At a final concentration of cells, CD3 + A fixed concentration of REGN4018 or a CD3 non-crosslinking control bispecific antibody (5 nM), or cemiplimab or non-binding IgG4 was added to wells containing T cells. P Control (20 nM) was added to wells containing OVCAR-3, PEO1, SW1990, or SW1990 / hPD-L1 cells. Subsequently, bs24963D, non-TAA x CD28 control, or non-binding control antibodies were titrated in a 1:4 dilution series from 7.6 pM to 500 nM and added to the wells. The end point of the 10-point concentration curve contained no antibody and was used to calculate the fold increase in activity. Plates were incubated at 37°C, 5% CO2 for 72 (OVCAR-3 and PEO1) or 96 (SW1990 and SW1990 / hPD-L1) hours, after which 50 μL of media supernatant was collected and used to quantify proliferation [methyl- 3 IL-2 release was measured before treatment with [H]-thymidine.

[0259] Five microliters (for assays using OVCAR-3 and PEO1 cells) or 20 μL (for assays using SW1990 and SW1990 / hPD-L1 cells) of supernatant were tested for IL-2 release using a human IL-2 AlphaLISA kit according to the manufacturer's protocol. IL-2 measurements were taken on a Perkin Elmer Envision multilabel plate reader and reported as relative fluorescence units (RFU).

[0260] For proliferation assays, 50 μL of [methyl- 3 [H]-thymidine was added to the wells and the plates were incubated for either 6 hours (for assays using OVCAR-3 and PEO1 cells) or 16 hours (for assays using SW1990 and SW1990 / hPD-L1 cells). 3 H]-thymidine will be incorporated in higher amounts into dividing cells. After incubation, cells were harvested onto filter plates and prepared for measurement in a Microplate Scintillation & Luminescence Counter TopCount NXT instrument.

[0261] All serial dilutions were tested in triplicate for IL-2 release and proliferation. 50 Values ​​were determined from a four-parameter logistic equation on a 10-point dose-response curve using GraphPad Prism™ software. Maximum levels of IL-2 release and proliferation are given as the mean maximum response detected within the dose range tested. The fold increase in maximum IL-2 release or T cell proliferation mediated by bs24963D was calculated relative to the maximum IL-2 release or proliferation not mediated by either antibody.

[0262] The ability of bs24963D to activate T cells was assessed in an assay in which stimulatory antigen-presenting cells provide a signal. This assay utilizes human CD8, human CD28, a published TCR (1G4) that recognizes the NY-ESO-1 peptide (NYESO1p) complexed with HLA-A2, and an NF-κB-luciferase reporter. J.RT3.T3.5 reporter T cells engineered to express HLA-A2, hβ2M, and NYESO1p were used. The stimulatory antigen-presenting cells providing signal 1 were 3T3 cells engineered to express HLA-A2, hβ2M, and NYESO1p, with or without human MUC16 (hMUC16). A CD28 non-crosslinking control bispecific antibody (non-TAA × CD28) and a non-binding control mAb (IgG4 P-PVA ) were tested in parallel with bs24963D. NF-κB signaling was measured using luminescence reagents to detect luciferase reporter activity. The results are summarized in Table 28.

[0263] Antigen: In this test system, bs24963D detects MUC16 + It mediated a concentration-dependent increase in NF-κB signaling in reporter T cells in the presence of antigen-presenting cells, and no activity was observed when cells lacking MUC16 expression were used (Figures 11A and 11B). No increase in NF-κB signaling was observed with a CD28 non-crosslinking control bispecific antibody. [Table 30]

[0264] Example 11. Evaluation of bs24963D (anti-MUC16 x anti-CD28)-mediated IL-2 release and proliferation of human primary T cells in the presence or absence of REGN4018 (anti-MUC16 x anti-CD3) or cemiplimab (PD-1 antagonist antibody) The ability of bs24963D to activate human primary T cells, as determined by IL-2 release and T cell proliferation, was assessed using two different MUC16 +The results were evaluated in the presence of human ovarian cancer cell lines (OVCAR-3 and PEO1). Because these cells do not provide sufficient signal 1 from an allogeneic response, a fixed concentration of REGN4018 (MUC16xCD3 bispecific antibody) was included to provide signal 1. Results for OVCAR-3 and PEO1 cells are summarized in Table 29 for IL-2 release and Table 30 for proliferation.

[0265] The ability of bs24963D to activate human primary T cells, as determined by IL-2 release and T cell proliferation, was assessed by the expression of MUC16 + This was evaluated in the presence of a human pancreatic cancer cell line (SW1990) and SW1990 engineered to overexpress human PD-L1 (SW1990 / hPD-L1). Both cell lines provide a sufficient allogeneic response to function as signal 1. In addition, the ability of a fixed concentration of cemiplimab (20 nM) to augment the effect of bs24963D was also evaluated. Results for SW1990 and SW1990 / hPD-L1 cells are summarized in Table 31 for IL-2 release and Table 32 for proliferation.

[0266] Ability of bs24963D (REGN5668) to enhance IL-2 release from human primary T cells and their proliferation in the presence or absence of REGN4018 by OVCAR-3 and PEO1 target cells When incubated with OVCAR-3 and PEO1 cancer cells, bs24963D inhibited IL-2 release from human T cells (Figure 12) and IL-2 release from human T cells only in the presence of REGN4018. REGN4018 mediated a concentration-dependent enhancement of IL-2 release in the IL-2-deficient cells and their proliferation (Figure 13). A CD3 and CD28 non-crosslinking control bispecific antibody did not enhance IL-2 release in the presence or absence of REGN4018.

[0267] In this assay, 5 nM REGN4018 alone did not increase IL-2 release compared to the non-binding control, but did show a modest enhancement of T cell proliferation. [Table 31] [Table 32]

[0268] The ability of bs24963D (REGN5668) to enhance IL-2 release from human primary T cells and their proliferation in the presence or absence of cemiplimab by SW1990 and SW1990 / hPD-L1 target cells SW1990 and SW1990 / hPD-L1 MUC16 + When incubated with human pancreatic cancer cells, bs24963D mediated a concentration-dependent enhancement of IL-2 release from human T cells (Figure 14) and their proliferation (Figure 15) in the presence and absence of cemiplimab. Overexpression of human PD-L1 in SW1990 cells suppressed IL-2 and T cell proliferation in the presence of bs24963D, which were somewhat increased by the addition of cemiplimab. At high concentrations, the CD28 non-crosslinking control bispecific antibody mediated some IL-2 release in the presence of SW1990 and SW1990 / hPD-L1 cells. In the absence of bs24963D, cemiplimab did not increase IL-2 release or T cell proliferation compared to the CD28 non-crosslinking control bispecific antibody. [Table 33] [Table 34]

[0269] The present invention is not to be limited in scope by the specific embodiments described herein. Indeed, various modifications of the invention in addition to those described herein may be made in accordance with the foregoing description and the accompanying drawings. Such modifications will be apparent to those skilled in the art from the accompanying figures and are intended to fall within the scope of the appended claims.

Claims

1. 1. An isolated bispecific antigen-binding molecule comprising: (a) a first antigen-binding domain that binds to human CD28 and comprises a heavy chain variable region (HCVR) comprising HCDR1, HCDR2, and HCDR3 domains comprising the amino acid sequences of SEQ ID NOs: 44, 46, and 48, respectively, and a light chain variable region (LCVR) comprising LCDR1, LCDR2, and LCDR3 domains comprising the amino acid sequences of SEQ ID NOs: 36, 38, and 40, respectively; (b) a second antigen-binding domain that specifically binds to the human mucin 16 membrane antigen (MUC16) and comprises an HCVR comprising HCDR1, HCDR2, and HCDR3 domains comprising the amino acid sequences of SEQ ID NOs: 28, 30, and 32, respectively, and an LCVR comprising LCDR1, LCDR2, and LCDR3 domains comprising the amino acid sequences of SEQ ID NOs: 36, 38, and 40, respectively; 2. The isolated bispecific antigen-binding molecule comprising:

2. (a) the first antigen-binding domain comprises an HCVR comprising the amino acid sequence of SEQ ID NO: 42 and an LCVR comprising the amino acid sequence of SEQ ID NO: 34; (b) the second antigen-binding domain comprises an HCVR comprising the amino acid sequence of SEQ ID NO: 26 and an LCVR comprising the amino acid sequence of SEQ ID NO: 34; The isolated bispecific antigen-binding molecule of claim 1.

3. 3. The isolated bispecific antigen-binding molecule of claim 1 or 2, which is a bispecific antibody.

4. 4. The isolated bispecific antigen-binding molecule of claim 3, wherein the bispecific antibody comprises a human IgG heavy chain constant region.

5. 5. The isolated bispecific antigen-binding molecule of claim 4, wherein the human IgG heavy chain constant region of the bispecific antibody is of isotype IgG1 or IgG4.

6. 6. The isolated bispecific antigen-binding molecule of any one of claims 3 to 5, comprising a first heavy chain comprising a first CH3 domain and a second heavy chain comprising a second CH3 domain, wherein the first CH3 domain or the second CH3 domain comprises a H435R modification (according to EU numbering) and a Y436F modification (according to EU numbering), but not both.

7. The isolated bispecific antigen-binding molecule of any one of claims 3 to 6, wherein the bispecific antibody comprises a hinge that has reduced Fcγ receptor binding compared to a wild-type hinge of the same isotype.

8. A pharmaceutical composition comprising the bispecific antigen-binding molecule of any one of claims 1 to 7 and a pharmaceutically acceptable carrier or diluent.

9. A nucleic acid molecule comprising a nucleotide sequence encoding the bispecific antigen-binding molecule of any one of claims 1 to 7, or three different nucleic acid molecules comprising nucleotide sequences encoding the HCVR of the first antigen-binding domain, the HCVR of the second antigen-binding domain, and the LCVR of the first and second antigen-binding domains, respectively, of the bispecific antigen-binding molecule of any one of claims 1 to 7.

10. 10. An expression vector comprising the nucleic acid molecule of claim 9, or three different expression vectors each comprising one of the three different nucleic acid molecules of claim 9.

11. A host cell comprising the bispecific antigen-binding molecule of any one of claims 1 to 7, the nucleic acid molecule of claim 9 or three different nucleic acid molecules, or the expression vector of claim 10 or three different expression vectors.

12. 12. A method for producing an anti-CD28 x anti-MUC16 bispecific antigen-binding molecule, comprising culturing the host cell of claim 11 under conditions that allow the production of said bispecific antigen-binding molecule, and recovering the bispecific antigen-binding molecule so produced.

13. 13. The method of claim 12, further comprising formulating the bispecific antigen-binding molecule as a pharmaceutical composition with a suitable carrier.

14. An anti-CD28 x anti-MUC16 bispecific antigen-binding molecule obtainable by the method of claim 12.

15. 9. The pharmaceutical composition of claim 8 for treating a MUC16-expressing tumor in a patient.

16. The pharmaceutical composition of claim 15, wherein the MUC16-expressing tumor is ovarian cancer.

17. 17. The pharmaceutical composition of claim 15 or 16, in combination with a second therapeutic agent, for treating a MUC16-expressing tumor.

18. 18. The pharmaceutical composition of claim 17, wherein the second therapeutic agent comprises an anti-tumor agent, radiation therapy, an antibody-drug conjugate, a bispecific antibody conjugated to an anti-tumor agent, a checkpoint inhibitor, or a combination thereof.

19. 18. The pharmaceutical composition of claim 17, wherein the second therapeutic agent is an anti-PD-1 antibody.

20. 20. The pharmaceutical composition of claim 19, wherein the anti-PD-1 antibody is cemiplimab.

21. 18. The pharmaceutical composition of claim 17, wherein the second therapeutic agent is an anti-MUC16 x anti-CD3 bispecific antibody.

Images