Bispecific PD-L1x4-1BB antibodies and methods of use thereof

Abstract

The present disclosure provides bispecific antigen binding molecules (PD-L1 * 4-1BB) that bind to PD-L1 and 4-1BB. In certain embodiments, the present disclosure provides a bispecific PD-L1 * 4-1BB antibody, or antigen binding fragment thereof, comprising a first antigen binding domain that specifically binds to 4-1BB and a second antigen binding domain that specifically binds to PD-L1. In certain embodiments, the bispecific antibodies disclosed herein bind 4-1BB on T cells with a first antigen binding domain and bind PD-L1 expressed on tumor cells or antigen presenting cells with a second antigen binding domain. In certain embodiments, the antibodies of the present disclosure are useful in the treatment of cancer.

Description

Technical Field

[0001] This disclosure relates to antibodies that bind to PD-L1 and 4-1BB and methods of use thereof, for example, for the treatment or prevention of cancer.

[0002] sequence list

[0003] This application contains a sequence list that has been electronically submitted in XML format and is incorporated herein by reference in its entirety. The XML copy created on June 24, 2024, is named SeqList11375 and has a size of 112,849 bytes. This sequence list is part of the specification and is incorporated herein by reference in its entirety. The following sequences are below the minimum length allowed under the sequence list format ST.26: gctgcatcc (SEQ ID NO: 21), AAS (SEQ ID NO: 22), ggggcaagt (SEQ ID NO: 69), and GAS (SEQ ID NO: 70). Background Technology

[0004] The ability of T cells to recognize and kill their cellular targets, such as virus-infected cells or tumor cells, depends on a series of coordinated interactions. The most important of these is the recognition and binding of the T-cell receptor (TCR) complex (which comprises associated CD3 γ, δ, ε, and ζ chains) to target cells, and this interaction is known as “signal 1” for T cell activation. The TCR recognizes viral or tumor peptides presented on the grooves of MHC proteins expressed on the surface of target cells. Because such binding is typically low-affinity, successful triggering of “signal 1” requires the aggregation of numerous TCR complexes along the interface between the T cell and its target cell; this interface is called the “immune synapse.” T cell activation can be further promoted by additional interactions. For example, T cells have a molecule called 4-1BB on their surface that can provide a co-stimulatory “signal 2” to enhance activation via the TCR complex. T cell activation is enhanced when T cells recognize their target cells via their TCR complex and subsequently engage “signal 2” by binding to their homologous ligands on the target cells via 4-1BB; like “signal 1”, 4-1BB-mediated “signal 2” is thought to occur through co-aggregation at the immune synapse.

[0005] 4-1BB (CD137 / ILA / TNFRSF9) has attracted attention as a promising therapeutic target in the context of cancer and other diseases due to its broad expression profile and ability to stimulate multiple signaling pathways involved in generating potent immune responses. As a prominent mediator of immune responses expressed on multiple cell types, 4-1BB signaling not only plays a protective role but can also drive pathological conditions, such as the adverse effects observed after administration of therapeutic 4-1BB antibodies. It has been shown that sustained stimulation of 4-1BB signaling and consequently sustained activation of T cells can lead to granuloma formation in tumor-draining lymph nodes due to excessive recruitment of macrophages (Kim et al., Cell Mol Immunol (2021) 18(8):1956–68). Furthermore, hepatotoxicity has been reported as a common problem associated with therapeutic 4-1BB antibody therapy. It has been shown that 4-1BB antibody therapy leads to CD8+ toxicity. + T cell infiltration into the liver causes inflammation and increased transaminase expression (Dubrot et al., CancerImmunol Immunother (2010) 59(8):1223–33). However, such infiltration within or around the liver tissue in a tumor setting is not associated with clinical benefit. Local or targeted use of anti-4-1BB mAbs can be used to enhance antitumor immunity with lower risks.

[0006] Programmed death ligand 1 (PD-L1), the ligand of programmed cell death 1 (PD-1 or PD1), is expressed on antigen-presenting cells (such as activated monocytes and dendritic cells) and on some cancer cells. Stimulation with PD-L1 is known to inhibit the activation of PD-1-expressing T lymphocytes (induction of cell proliferation and various cytokine production).

[0007] Monoclonal antibodies (mAbs) designed to enhance T-cell activation are in clinical development as anti-tumor therapeutics. However, most current treatments struggle to overcome the inhibitory nature of the tumor microenvironment, thus failing to generate effective tumor-specific T-cell activation and subsequent tumor cell killing. Several blocking mAbs targeting checkpoint inhibitors such as CTLA-4 (cytotoxic T-lymphocyte-associated protein) and programmed cell death 1 (PD-1) / programmed cell death ligand 1 (PD-L1) have been clinically approved for melanoma, renal cell carcinoma, non-small cell lung cancer, and advanced metastatic squamous cell carcinoma of the skin. Blocking PD-1 release breaks the inhibition of T-cell activation, but its potency as a single agent is often insufficient to achieve tumor clearance and a durable anti-tumor response. Therefore, additional cancer-specific therapies, such as improved immunotherapies, are needed. Summary of the Invention

[0008] This disclosure provides bispecific antigen-binding molecules comprising bispecific antibodies containing a first antigen-binding arm (or domain) that specifically binds to 4-1BB (also known as CD137 and tumor necrosis factor receptor superfamily 9, TNFRSF9) and a second antigen-binding arm (or domain) that binds to programmed death-ligand 1 (PD-L1) (“PD-L1x4-1BB” or “4-1BBxPD-L1” antibody). These bispecific antigen-binding molecules induce durable anti-tumor immunity and promote robust intratumoral T cell activation, while reducing systemic cytokine secretion. The bispecific antigen-binding molecules of this disclosure provide a pan-tumor co-stimulation approach (tumor-agnostic targeting) in which the inhibitory checkpoint PD-L1, present in multiple tumor types, is combined with co-stimulatory signals from 4-1BB on activated T cells. A PD-L1x4-1BB bispecific antibody blocks the interaction between PD-L1 and PD-1 and leverages PD-L1 expression in the tumor or tumor microenvironment to intratumorally anchor the 4-1BB agonist antigen-binding fragment, providing a more spatially confined co-stimulatory signal to T cells. The PD-L1x4-1BB bispecific antigen-binding molecule of this disclosure binds and conjugates to PD-L1-expressing tumor cells, APCs, and other tumor-infiltrating immune cells, thus providing broad tumor-unaware targeting compared to bispecific agents targeting both 4-1BB and tumor-associated antigens (TAAs). In some embodiments, when used in combination with a CD3-based bispecific agent targeting this TAA, the bispecific antigen-binding molecule enhances T cell-mediated killing of tumor cells expressing both PD-L1 and TAA. In summary, combining 4-1BB agonism with PD-1 inhibition provides a safe biological solution with significantly enhanced, specific, and synergistic antitumor activity across multiple tumor types.

[0009] In one aspect, this disclosure provides an isolated bispecific antigen-binding molecule comprising a first antigen-binding domain and a second antigen-binding domain, wherein the first antigen-binding domain has a capacitance of less than about 2 × 10⁻⁶. -7 M of K D The second antigen-binding domain binds to human 4-1BB, as measured by surface plasmon resonance at 25°C; the second antigen-binding domain has a binding capacity of less than approximately 2 × 10⁻⁶. -10 M of K D Specific binding to programmed death ligand 1 (PD-L1), as measured by surface plasmon resonance at 25 °C.

[0010] In some implementations, the bispecific antigen-binding molecule is at a concentration of less than about 4 × 10⁻⁶. -9 M's EC 50 It binds to the surface of human T cells, as measured by in vitro FACS binding assays.

[0011] In some implementations, the bispecific antigen-binding molecule is at a concentration of less than about 4 × 10⁻⁶. -9 M's EC 50 It binds to the surface of cells expressing PD-L1, as measured by in vitro FACS binding assays.

[0012] In some embodiments, the bispecific antigen-binding molecule has an IC50 of less than about 1.3 nM. 50 Blocking the binding of PD-L1 to PD-1, as measured by an ELISA-based blocking assay.

[0013] On the other hand, when used in combination with an anti-mucoprotein 16 (MUC16) X CD3 bispecific antibody and tested on target cells expressing PD-L1, the isolated bispecific antigen-binding molecule showed co-stimulatory effects.

[0014] In some embodiments, the bispecific antigen-binding molecule is combined with a bispecific MUC16xCD3 antibody at a concentration of less than about 10. -10 M's EC 50 Mediates in vitro T cell killing of OVCAR-3 cells expressing PD-L1.

[0015] In certain embodiments of this disclosure, a bispecific antigen-binding molecule (e.g., an antibody or its antigen-binding fragment) binding PD-L1 and 4-1BB comprises: (1) a PD-L1 binding arm comprising: (a) a heavy chain immunoglobulin or its variable region comprising an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% amino acid sequence identity with the amino acid sequences shown in SEQ ID NO: 2 and 50, comprising the amino acid sequences shown in SEQ ID NO: 2 and 50, or consisting of the amino acid sequences shown in SEQ ID NO: 2 and 50, or variants thereof; and / or (b) a light chain immunoglobulin or its variable region comprising an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% amino acid sequence identity with the amino acid sequences shown in SEQ ID NO: 18 and 66, comprising the amino acid sequences shown in SEQ ID NO: 18 and 66, or consisting of the amino acid sequences shown in SEQ ID NO: 18 and 6 ... The amino acid sequences shown in SEQ ID NO: 18 and 66, or variants thereof; or (2) comprising a 4-1BB binding arm comprising: (c) a heavy chain immunoglobulin or its variable region comprising an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% amino acid sequence identity with the amino acid sequences shown in SEQ ID NO: 10, 32, 42, and 58, comprising the amino acid sequences shown in SEQ ID NO: 10, 32, 42, and 58, or comprising the amino acid sequences shown in SEQ ID NO: 10, 32, 42, and 58, or variants thereof; and / or (d) a light chain immunoglobulin or its variable region comprising an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% amino acid sequence identity with the amino acid sequences shown in SEQ ID NO: 18 and 66, comprising the amino acid sequences shown in SEQ ID NO: 18 and 6 ... comprising the amino acid sequences shown in SEQ ID NO: 10, 32, 42, and 58, The amino acid sequences shown in 18 and 66 are composed of [the amino acid sequence].

[0016] In one embodiment of this disclosure, a bispecific antigen-binding protein (e.g., an antibody or its antigen-binding fragment) binding PD-L1 and 4-1BB comprises: (1) a PD-L1 binding arm and a 4-1BB binding arm, the PD-L1 binding arm comprising: (a) a heavy chain immunoglobulin or its variable region comprising HCDR1, HCDR2, and HCDR3 of the heavy chain variable region, the heavy chain variable region comprising the amino acid sequence shown in SEQ ID NO: 2 or 50, and having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, and 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 2 or 50, respectively, comprising or consisting of the amino acid sequence shown in SEQ ID NO: 2 or 50; and / or (b) a light chain immunoglobulin or its variable region comprising LCDR1, LCDR2, and LCDR3 of the light chain variable region, the light chain variable region comprising SEQ ID NO: The amino acid sequence shown in SEQ ID NO: 18 or 66, and having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 18 or 66, respectively, comprising or consisting of the amino acid sequence shown in SEQ ID NO: 18 or 66; or (2) a 4-1BB binding arm and a PD-L1 binding arm, wherein the 4-1BB binding arm comprises: (c) a heavy chain immunoglobulin or its variable region, comprising HCDR1, HCDR2, and HCDR3 of the heavy chain variable region, wherein the heavy chain variable region comprises the amino acid sequence shown in SEQ ID NO: 10, 32, 42, or 58, and having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 10, 32, 42, or 58, respectively, comprising SEQ ID NO: The amino acid sequence shown in SEQ ID NO: 10, 32, 42 or 58 or composed of the amino acid sequence shown in SEQ ID NO: 10, 32, 42 or 58;And / or (d) light chain immunoglobulin or its variable region, comprising light chain variable regions LCDR1, LCDR2, and LCDR3, wherein the light chain variable regions comprise the amino acid sequences shown in SEQ ID NO: 18 or 66, and respectively have at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, and 99% amino acid sequence identity with the amino acid sequences shown in SEQ ID NO: 18 or 66, comprising or consisting of the amino acid sequences shown in SEQ ID NO: 18 or 66.

[0017] In some embodiments, the first antigen-binding domain comprises: (a) three heavy chain complementarity-determining regions (HCDR1, HCDR2, and HCDR3) contained within a heavy chain variable region (HCVR), the heavy chain variable region (HCVR) containing amino acid sequences selected from SEQ ID NO: 58, 42, 32, and 10 or variants thereof; and (b) three light chain complementarity-determining regions (LCDR1, LCDR2, and LCDR3) contained within a light chain variable region (LCVR), the light chain variable region (LCVR) containing amino acid sequences selected from SEQ ID NO: 66 and 18 or variants thereof.

[0018] In some embodiments, the isolated bispecific antigen-binding molecule comprises: HCDR1, which comprises an amino acid sequence selected from SEQ ID NO: 60, 34 and 12; HCDR2, which comprises an amino acid sequence selected from SEQ ID NO: 62, 44, 36 and 14; and HCDR3, which comprises an amino acid sequence selected from SEQ ID NO: 64, 46, 38 and 16.

[0019] In some embodiments, the isolated bispecific antigen-binding molecule comprises: LCDR1, which comprises an amino acid sequence selected from SEQ ID NO: 68 and 20; LCDR2, which comprises an amino acid sequence selected from SEQ ID NO: 70 and 22; and LCDR3, which comprises an amino acid sequence selected from SEQ ID NO: 72 and 24.

[0020] In some embodiments, the first antigen-binding domain comprises HCVR and LCVR, wherein the HCVR comprises the amino acid sequence of SEQ ID NO: 58 or a variant thereof, and the LCVR comprises the amino acid sequence of SEQ ID NO: 66 or a variant thereof.

[0021] In some embodiments, the first antigen-binding domain comprises HCVR and LCVR, wherein the HCVR comprises the amino acid sequence of SEQ ID NO: 42 or a variant thereof, and the LCVR comprises the amino acid sequence of SEQ ID NO: 18 or a variant thereof.

[0022] In some embodiments, the first antigen-binding domain comprises HCVR and LCVR, wherein the HCVR comprises the amino acid sequence of SEQ ID NO: 32 or a variant thereof, and the LCVR comprises the amino acid sequence of SEQ ID NO: 18 or a variant thereof.

[0023] In some embodiments, the first antigen-binding domain comprises HCVR and LCVR, wherein the HCVR comprises the amino acid sequence of SEQ ID NO: 10 or a variant thereof, and the LCVR comprises the amino acid sequence of SEQ ID NO: 18 or a variant thereof.

[0024] In some embodiments, the second antigen-binding domain comprises: (a) three heavy chain complementarity-determining regions (HCDR1, HCDR2, and HCDR3) contained within a heavy chain variable region (HCVR), the heavy chain variable region (HCVR) containing an amino acid sequence selected from SEQ ID NO: 50 and 2 or a variant thereof; and (b) three light chain complementarity-determining regions (LCDR1, LCDR2, and LCDR3) contained within a light chain variable region (LCVR), the light chain variable region (LCVR) containing an amino acid sequence selected from SEQ ID NO: 66 and 18 or a variant thereof.

[0025] In some embodiments, the second antigen-binding domain comprises: (a) HCDR1, which comprises the amino acid sequence of SEQ ID NO: 52 or SEQ ID NO: 4; (b) HCDR2, which comprises the amino acid sequence of SEQ ID NO: 54 or SEQ ID NO: 6; and (c) HCDR3, which comprises the amino acid sequence of SEQ ID NO: 58 or SEQ ID NO: 8.

[0026] In some embodiments, the second antigen-binding domain comprises: LCDR1, which comprises an amino acid sequence selected from SEQ ID NO: 68 and 20; LCDR2, which comprises an amino acid sequence selected from SEQ ID NO: 70 and 22; and LCDR3, which comprises an amino acid sequence selected from SEQ ID NO: 72 and 24.

[0027] In some embodiments, the second antigen-binding domain comprises: (a) HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 52, 54, and 56, respectively; and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 68, 70, and 72, respectively; or (b) HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 4, 6, and 8, respectively; and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, and 24, respectively.

[0028] In some embodiments, the second antigen-binding domain comprises: (a) an HCVR containing the amino acid sequence of SEQ ID NO: 50 or a variant thereof, and an LCVR containing the amino acid sequence of SEQ ID NO: 66 or a variant thereof; or (b) an HCVR containing the amino acid sequence of SEQ ID NO: 2 or a variant thereof, and an LCVR containing the amino acid sequence of SEQ ID NO: 18 or a variant thereof.

[0029] In some aspects, this disclosure provides isolated bispecific antigen-binding molecules comprising: (a) a first antigen-binding domain specifically binding to human 4-1BB, wherein the first antigen-binding domain comprises: HCDR1, HCDR2, HCDR3 domains containing the amino acid sequences of SEQ ID NO: 60, 62, 64, respectively, and LCDR1, LCDR2, LCDR3 domains containing the amino acid sequences of SEQ ID NO: 68, 70, 72, respectively; and (b) a second antigen-binding domain specifically binding to human PD-L1, wherein the second antigen-binding domain comprises: HCDR1, HCDR2, HCDR3 domains containing the amino acid sequences of SEQ ID NO: 52, 54, 56, respectively, and LCDR1, LCDR2, LCDR3 domains containing the amino acid sequences of SEQ ID NO: 68, 70, 72, respectively.

[0030] In some aspects, this disclosure provides isolated bispecific antigen-binding molecules comprising: (a) a first antigen-binding domain specifically binding to human 4-1BB, wherein the first antigen-binding domain comprises: HCDR1, HCDR2, HCDR3 domains containing the amino acid sequences of SEQ ID NO: 12, 44, 46, respectively, and LCDR1, LCDR2, LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, 24, respectively; and (b) a second antigen-binding domain specifically binding to human PD-L1, wherein the second antigen-binding domain comprises: HCDR1, HCDR2, HCDR3 domains containing the amino acid sequences of SEQ ID NO: 4, 6, 8, respectively, and LCDR1, LCDR2, LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, 24, respectively.

[0031] In other aspects, this disclosure provides isolated bispecific antigen-binding molecules comprising: (a) a first antigen-binding domain specifically binding to human 4-1BB, wherein the first antigen-binding domain comprises: HCDR1, HCDR2, HCDR3 domains containing the amino acid sequences of SEQ ID NO: 34, 36, 38, respectively, and LCDR1, LCDR2, LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, 24, respectively; and (b) a second antigen-binding domain specifically binding to human PD-L1, wherein the second antigen-binding domain comprises: HCDR1, HCDR2, HCDR3 domains containing the amino acid sequences of SEQ ID NO: 4, 6, 8, respectively, and LCDR1, LCDR2, LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, 24, respectively.

[0032] In some aspects, this disclosure provides isolated bispecific antigen-binding molecules comprising: (a) a first antigen-binding domain specifically binding to human 4-1BB, wherein the first antigen-binding domain comprises: HCDR1, HCDR2, HCDR3 domains containing the amino acid sequences of SEQ ID NO: 12, 14, 16, respectively, and LCDR1, LCDR2, LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, 24, respectively; and (b) a second antigen-binding domain specifically binding to human PD-L1, wherein the second antigen-binding domain comprises: HCDR1, HCDR2, HCDR3 domains containing the amino acid sequences of SEQ ID NO: 4, 6, 8, respectively, and LCDR1, LCDR2, LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, 24, respectively.

[0033] In some embodiments, the isolated bispecific antigen-binding molecule comprises: (a) a first antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 58 and an LCVR containing the amino acid sequence of SEQ ID NO: 66; and (b) a second antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 50 and an LCVR containing the amino acid sequence of SEQ ID NO: 66.

[0034] In some embodiments, the isolated bispecific antigen-binding molecule comprises: (a) a first antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 42 and an LCVR containing the amino acid sequence of SEQ ID NO: 18; and (b) a second antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 2 and an LCVR containing the amino acid sequence of SEQ ID NO: 18.

[0035] In some embodiments, the isolated bispecific antigen-binding molecule comprises: (a) a first antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 32 and an LCVR containing the amino acid sequence of SEQ ID NO: 18; and (b) a second antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 2 and an LCVR containing the amino acid sequence of SEQ ID NO: 18.

[0036] In some embodiments, the isolated bispecific antigen-binding molecule comprises: (a) a first antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 10 and an LCVR containing the amino acid sequence of SEQ ID NO: 18; and (b) a second antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 2 and an LCVR containing the amino acid sequence of SEQ ID NO: 18.

[0037] In some aspects, this disclosure provides isolated bispecific antigen-binding molecules that compete with a reference antibody for binding to PD-L1 or to the same epitope on PD-L1, wherein the reference antibody comprises a first antigen-binding domain and a second antigen-binding domain, the first antigen-binding domain comprising an HCVR / LCVR pair containing an amino acid sequence selected from SEQ ID NO: 58 / 66, 42 / 18, 32 / 18, and 10 / 18, and the second antigen-binding domain comprising an HCVR / LCVR pair containing an amino acid sequence selected from SEQ ID NO: 50 / 66 and 2 / 18.

[0038] In some aspects, this disclosure provides isolated bispecific antigen-binding molecules that compete with a reference antibody for binding to human 4-1BB or to the same epitope on human 4-1BB, wherein the reference antibody comprises a first antigen-binding domain and a second antigen-binding domain, the first antigen-binding domain comprising an HCVR / LCVR pair containing an amino acid sequence selected from SEQ ID NO: 58 / 66, 42 / 18, 32 / 18, and 10 / 18, and the second antigen-binding domain comprising an HCVR / LCVR pair containing an amino acid sequence selected from SEQ ID NO: 50 / 66 and 2 / 18.

[0039] In some implementations, the isolated bispecific antigen-binding molecule is a human bispecific antigen-binding molecule.

[0040] In some implementations, the isolated bispecific antigen-binding molecule is a bispecific antibody.

[0041] In some embodiments, the antibody comprises a human IgG heavy chain constant region linked to an HCVR of each of the first antigen-binding domain and the second antigen-binding domain.

[0042] In some embodiments, the heavy chain constant region is isotype IgG1. In other embodiments, the heavy chain constant region is isotype IgG4.

[0043] In some embodiments, the heavy chain constant region linked to the HCVR of the first antigen-binding domain or the heavy chain constant region linked to the HCVR of the second antigen-binding domain contains amino acid modifications that reduce protein A binding, but not simultaneously, wherein the reduction in protein A binding is relative to the same isotype heavy chain without the modification.

[0044] In some embodiments, the modification involves an H435R substitution (EU number) in the heavy chain comprising isotype IgG1 or IgG4. In some embodiments, the modification involves both an H435R substitution and a Y436F substitution (EU number) in the heavy chain comprising isotype IgG1 or IgG4.

[0045] In some implementations, the bispecific antibody comprises a chimeric hinge that reduces the binding of the Fcγ receptor relative to the same isotype wild-type hinge.

[0046] In some embodiments, the antibody comprises a first heavy chain of HCVR containing a first antigen-binding domain and a second heavy chain of HCVR containing a second antigen-binding domain, wherein the first heavy chain comprises an amino acid sequence selected from SEQ ID NO: 76, 48, 40 and 28; and the second heavy chain comprises an amino acid sequence selected from SEQ ID NO: 74 and 26.

[0047] In some embodiments, the antibody comprises a common light chain of LCVR containing a first antigen-binding domain and a second antigen-binding domain, wherein the common light chain comprises an amino acid sequence selected from SEQ ID NO: 78 and 30.

[0048] In some embodiments, the antibody comprises a first heavy chain of HCVR containing a first antigen-binding domain and a second heavy chain of HCVR containing a second antigen-binding domain, wherein the first heavy chain comprises the amino acid sequence of SEQ ID NO: 76 and the second heavy chain comprises the amino acid sequence of SEQ ID NO: 74.

[0049] In some embodiments, the antibody comprises a common light chain of LCVR containing a first antigen-binding domain and a second antigen-binding domain, wherein the common light chain comprises the amino acid sequence of SEQ ID NO: 78.

[0050] In some aspects, this disclosure provides a bispecific antibody comprising a first antigen-binding domain that specifically binds to human 4-1BB and a second antigen-binding domain that specifically binds to human PD-L1, wherein the bispecific antibody comprises a first heavy chain containing the amino acid sequence of SEQ ID NO: 76, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 78; and a second heavy chain containing the amino acid sequence of SEQ ID NO: 74, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 78.

[0051] In some aspects, this disclosure provides a bispecific antibody comprising a first antigen-binding domain that specifically binds to human 4-1BB and a second antigen-binding domain that specifically binds to human PD-L1, wherein the bispecific antibody comprises a first heavy chain containing the amino acid sequence of SEQ ID NO: 48, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 30; and a second heavy chain containing the amino acid sequence of SEQ ID NO: 26, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 30.

[0052] In other aspects, this disclosure provides a bispecific antibody comprising a first antigen-binding domain that specifically binds to human 4-1BB and a second antigen-binding domain that specifically binds to human PD-L1, wherein the bispecific antibody comprises a first heavy chain containing the amino acid sequence of SEQ ID NO: 40, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 30; and a second heavy chain containing the amino acid sequence of SEQ ID NO: 26, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 30.

[0053] In other aspects, this disclosure provides a bispecific antibody comprising a first antigen-binding domain that specifically binds to human 4-1BB and a second antigen-binding domain that specifically binds to human PD-L1, wherein the bispecific antibody comprises a first heavy chain containing the amino acid sequence of SEQ ID NO: 28, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 30; and a second heavy chain containing the amino acid sequence of SEQ ID NO: 26, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 30.

[0054] In some implementations, the bispecific antibody is a human antibody.

[0055] In some aspects, this disclosure provides pharmaceutical compositions comprising the bispecific antigen-binding molecule described herein and a pharmaceutically acceptable carrier or diluent.

[0056] In other aspects, this disclosure provides pharmaceutical compositions comprising the bispecific antibodies described herein and pharmaceutically acceptable carriers or diluents.

[0057] This disclosure also provides a method for preparing the bispecific antigen-binding molecule described herein, comprising: (a) introducing one or more nucleic acid molecules containing nucleic acid sequences encoding an immunoglobulin chain of the antigen-binding molecule into a host cell (e.g., CHO cells); (b) culturing the host cell under conditions favorable for expression of the nucleic acid molecule; and (c) optionally isolating the antigen-binding molecule or the immunoglobulin chain from the host cell and / or the culture medium in which the host cell is cultured.

[0058] In some implementations, the host cell is a Chinese hamster ovary (CHO) cell.

[0059] In some embodiments, the method further includes formulating the antigen-binding molecule into a pharmaceutical composition comprising an acceptable carrier.

[0060] Any antigen-binding molecule or immunoglobulin chain that is a product of such a method is also part of this disclosure.

[0061] This disclosure also provides nucleic acid molecules comprising the nucleotide sequences of the bispecific antigen-binding molecules described herein; or sets of nucleic acid molecules comprising nucleotide sequences encoding the following: an HCVR that specifically binds to a first antigen-binding domain of human 4-1BB, an HCVR that specifically binds to a second antigen-binding domain of human PD-L1, and an LCVR of the bispecific antigen-binding molecules isolated herein. Expression vectors containing nucleic acid molecules of this disclosure, or sets of expression vectors containing sets of nucleic acid molecules of this disclosure, as well as host cells (e.g., CHO) containing nucleic acid molecules, vectors, or antigen-binding proteins of this disclosure, are also part of this disclosure.

[0062] This disclosure provides a method for generating a bispecific antigen-binding molecule that binds to PD-L1 and 4-1BB, comprising: (a) culturing host cells as described herein under conditions favorable for generating the bispecific antigen-binding molecule; and (b) optionally, isolating the antigen-binding molecule or immunoglobulin chain from the host cells and / or the culture medium in which the host cells are cultured. In some embodiments, the host cells are CHO cells. In some embodiments, the method further comprises formulating the antigen-binding molecule into a pharmaceutical composition comprising an acceptable carrier.

[0063] Antigen-binding molecules or immunoglobulin chains that are products of the methods disclosed herein are also part of this disclosure.

[0064] This disclosure also provides nucleic acid molecules comprising a nucleotide sequence encoding a bispecific antibody as described herein; or sets of nucleic acid molecules comprising a nucleotide sequence encoding a heavy chain of a first antigen-binding domain specifically binding to human 4-1BB, a heavy chain of a second antigen-binding domain specifically binding to human PD-L1, and a light chain of a bispecific antibody as described herein. Expression vectors of nucleic acids containing this disclosure, as well as host cells (e.g., CHO cells) containing nucleic acid molecules, vectors, or antibodies of this disclosure, are also part of this disclosure.

[0065] This disclosure provides a method for generating bispecific antibodies that bind to PD-L1 and 4-1BB, comprising: (a) culturing host cells as described herein under conditions favorable for generating the bispecific antigen-binding molecule; and (b) optionally, isolating the antigen-binding molecule or immunoglobulin chain from the host cells and / or the culture medium in which the host cells are cultured. In some embodiments, the host cells are CHO cells. In some embodiments, the method further comprises formulating the bispecific antibody into a pharmaceutical composition comprising an acceptable carrier.

[0066] Antibodies that are products of the methods disclosed herein are also part of this disclosure.

[0067] This disclosure also provides methods for treating hyperproliferative diseases (e.g., cancer) in subjects (e.g., humans) in which such treatment is desired, comprising administering (e.g., subcutaneously, intravenously, or intramuscularly) an effective amount of a bispecific antigen-binding protein or composition or formulation. In one embodiment of this disclosure, the cancer is B-cell carcinoma, basal cell carcinoma, urothelial carcinoma of the bladder, brain cancer, breast cancer, cervical cancer, cervical squamous cell carcinoma, colorectal cancer, diffuse large B-cell lymphoma, endometrial adenocarcinoma, endometrial cancer, esophageal cancer, gastroesophageal adenocarcinoma, gastroesophageal cancer, glioblastoma multiforme, head and neck squamous cell carcinoma, hepatocellular carcinoma, lung adenocarcinoma, lung cancer, lung squamous cell carcinoma, melanoma, multiple myeloma, non-small cell lung cancer, ovarian cancer, pancreatic cancer, prostate cancer, rectal cancer, skin cancer, or T-cell carcinoma. In one embodiment, the cancer comprises tumor cells expressing PD-L1.

[0068] In some aspects, this disclosure provides methods for inhibiting tumor growth in a subject, including administering to the subject an isolated bispecific antigen-binding molecule, or a bispecific antibody, or a pharmaceutical composition as described herein.

[0069] In some implementation schemes, the tumor is B-cell carcinoma, basal cell carcinoma, bladder urothelial carcinoma, brain cancer, breast cancer, cervical cancer, cervical squamous cell carcinoma, colorectal cancer, diffuse large B-cell lymphoma, endometrial adenocarcinoma, endometrial cancer, esophageal cancer, gastroesophageal adenocarcinoma, gastroesophageal cancer, glioblastoma multiforme, head and neck squamous cell carcinoma, hepatocellular carcinoma, lung adenocarcinoma, lung cancer, lung squamous cell carcinoma, melanoma, multiple myeloma, non-small cell lung cancer, ovarian cancer, pancreatic cancer, prostate cancer, rectal cancer, skin cancer, or T-cell carcinoma.

[0070] In some implementations, the tumor expresses PD-L1.

[0071] In some implementations, the method also includes administering a second therapeutic agent or treatment regimen.

[0072] In some implementations, the second therapeutic agent or treatment regimen includes chemotherapy drugs, DNA alkylating agents, immunomodulators, proteasome inhibitors, histone deacetylase inhibitors, radiotherapy, surgery, stem cell transplantation, bispecific antibodies that interact with tumor-associated antigens (TAAs) and T-cell or immune cell antigens, antibody-drug conjugates, oncolytic viruses, bispecific antibodies conjugated to antitumor agents, VEGF inhibitors, checkpoint inhibitors, GITR agonists, CD27 agonists, 4-1BB activators, PD-1 inhibitors, CTLA-4 inhibitors, EGFR inhibitors, Ang2 inhibitors, MUC16 inhibitors, cancer vaccines, cytokines, modified IL2, modified IL12, IL4 inhibitors, IL6 inhibitors, corticosteroids, or combinations thereof.

[0073] In some implementations, the T cell or immune cell antigen is CD3.

[0074] In some implementations, TAA is selected from AFP, ALK, BAGE protein, BCMA, BIRC5 (survival protein), BIRC7, β-linkin, brc-abl, BRCA1, BORIS, CA9, carbonic anhydrase IX, caspase-8, CALR, CCR5, CD19, CD20 (MS4A1), CD22, CD30, CD40, CDK4, CEA, CTLA4, cyclin-B1, CYP1B1, EGFR, EGFRvIII, ErbB2 / Her2, ErbB3, ErbB4, ETV6-AML, EpCAM, EphA2, Fra-1, FOLR1, GAGE ​​protein (e.g., GAGE-1, -2), GD2, GD3, GloboH, phosphatidylinositol proteoglycan-3, GM3, gp100, Her2, HLA / B-raf, HLA / k-ras, HLA / MAGE-A3, hT ERT, LMP2, MAGE proteins (e.g., MAGE-1, -2, -3, -4, -6, and -12), MART-1, mesothelin, ML-IAP, Muc1, Muc2, Muc3, Muc4, Muc5, Muc16 (CA-125), MUM1, NA17, NY-BR1, NY-BR62, NY-BR85, NY-ESO1, OX40, p15, p53, PAP, PAX3, PAX5 PCTA-1, PLAC1, PRLR, PRAME, PSMA (FOLH1), RAGE protein, Ras, RGS5, Rho, SART-1, SART-3, STEAP1, STEAP2, TAG-72, TGF-β, TMPRSS2, Thompson-nouvelle antigen (Tn), TRP-1, TRP-2, tyrosinase, and uroplakin-3.

[0075] In other respects, this disclosure provides the use of bispecific antigen-binding molecules, or bispecific antibodies, or pharmaceutical compositions as described herein in the treatment of tumors.

[0076] In some implementation schemes, the tumor is B-cell carcinoma, basal cell carcinoma, bladder urothelial carcinoma, brain cancer, breast cancer, cervical cancer, cervical squamous cell carcinoma, colon cancer, colorectal cancer, diffuse large B-cell lymphoma, endometrial adenocarcinoma, endometrial cancer, esophageal cancer, gastroesophageal adenocarcinoma, gastroesophageal cancer, glioblastoma multiforme, head and neck squamous cell carcinoma, hepatocellular carcinoma, melanoma, multiple myeloma, leukemia, lung adenocarcinoma, lung cancer, lung squamous cell carcinoma, lung adenocarcinoma, cervical squamous cell carcinoma, endometrial adenocarcinoma, bladder urothelial carcinoma, lung cancer, non-small cell lung cancer, colorectal cancer, ovarian cancer, pancreatic cancer, prostate cancer, rectal cancer, endometrial cancer, skin cancer, or T-cell carcinoma.

[0077] In some implementations, the tumor expresses PD-L1.

[0078] In some embodiments, the antigen-binding molecule or pharmaceutical composition is used in combination with a second therapeutic agent or treatment regimen, the second therapeutic agent or treatment regimen including chemotherapeutic agents, DNA alkylating agents, immunomodulators, proteasome inhibitors, histone deacetylase inhibitors, radiotherapy, surgery, stem cell transplantation, bispecific antibodies that interact with tumor-associated antigens (TAAs) and T-cell or immune cell antigens, antibody-drug conjugates, oncolytic viruses, bispecific antibodies conjugated to antitumor agents, VEGF inhibitors, checkpoint inhibitors, GITR agonists, CD27 agonists, 4-1BB activators, PD-1 inhibitors, CTLA-4 inhibitors, EGFR inhibitors, Ang2 inhibitors, MUC16 inhibitors, cancer vaccines, cytokines, modified IL2, modified IL12, IL4 inhibitors, IL6 inhibitors, corticosteroids, or combinations thereof. Attached Figure Description

[0079] Figure 1 This relates to Example 7, and is a graph showing the average tumor volume in mice administered the specified antibody.

[0080] Figures 2A to 2B This relates to Example 8. Figure 2A This is a graph showing the average tumor volume in mice that were administered a specified dose of a specified antibody. Figure 2B This is a graph showing the survival probability of mice administered a specified dose of a specified antibody. mpk = mg / kg body weight.

[0081] Figures 3A to 3B This relates to Example 9. Figure 3A This is a graph showing the average tumor volume in mice that were administered a specified dose of a specified antibody. Figure 3B This is a graph showing the survival probability of mice administered a specified dose of a specified antibody. mpk = mg / kg body weight.

[0082] Figure 4This relates to Example 10 and is a graph showing the average radiation in mice administered the specified antibody.

[0083] Figures 5A to 5E This relates to Example 11. Figure 5A This is a graph showing the average tumor volume in mice treated with the specified antibody. Figures 5B to 5E This is a graph showing the tumor volume in individual mice that have been given specific antibodies. Figure 5B The tumor volume in individual mice treated with the same type of control is shown. Figure 5C Tumor volume is shown in individual mice that were administered REGN6191. Figure 5D Tumor volume is shown in individual mice treated with anti-PD1. Figure 5E Tumor volume is shown in individual mice treated with REGN6191 + anti-PD1.

[0084] Figures 6A to 6D This relates to Example 12. Figure 6A This is a graph showing the concentration of IL-2 in the serum of mice administered the specified antibody on day 0. Figure 6B This is a graph showing the concentration of IFNg in the serum of mice administered the specified antibody on day 7. Figure 6C This is a graph showing the concentration of ALT in the serum of mice administered the specified antibody on day 7. Figure 6D This is a graph showing the concentration of IL-2 in the serum of mice administered the specified antibody on day 7.

[0085] Figures 7A to 7C This relates to Example 13, and is a graph showing tumor volume in individual mice administered a specified dose of a specified antibody. Figure 7A Tumor volumes in individual mice treated with 0.01 mg / kg EGFRxCD3 are shown. Figure 7B Tumor volumes in individual mice treated with 0.01 mg / kg EGFRxCD3 + 0.1 mg / kg PD-L1xBetV1 are shown. Figure 7C Tumor volumes are shown in individual mice treated with 0.01 mg / kg EGFRxCD3 + 0.1 mg / kg PD-L1x4-1BB. “mpk” is mg / kg. Detailed Implementation

[0086] Before describing this disclosure, it should be understood that this disclosure is not limited to the specific methods and experimental conditions described, as such methods and conditions can vary. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting, as the scope of this disclosure will be limited only by the appended claims.

[0087] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. When used with respect to a specific numerical value, the term “about” as used herein means that the value may vary by no more than 1% from the stated value. For example, the expression “about 100” as used herein includes 99 and 101 and all values ​​in between (e.g., 99.1, 99.2, 99.3, 99.4, etc.).

[0088] Although any methods and materials similar to or equivalent to those described and materials herein may be used in the practice or testing of this disclosure, preferred methods and materials are now described.

[0089] The PD-L1 binding arm of the bispecific antibody acts as a bridge to anchor the 4-1BB activating antibody to the tumor, providing a pan-tumor co-stimulatory approach (tumor unaware) to convert inhibitory checkpoints (PD-L1) that are widely present in a variety of tumor indications and to convert them into co-stimulatory signals to activate 4-1BB on T cells.

[0090] Just as CAR-T therapy has employed the artificial activation of chimeric receptors for both "signal 1" and "signal 2" to enhance its antitumor activity, combining PD-1 inhibition with 4-1BB agonist (which provides "signal 2") to enhance antitumor activity has potential benefits. This approach offers several practical advantages over CAR-T therapy: it eliminates the need for cumbersome cell therapy preparation tailored to each patient individually, and it avoids the need for patients to be "lymphocyte depleted" beforehand by toxic chemotherapy, which is often associated with adverse effects that prevent patients from receiving cell therapy. This bispecific approach offers the potential for improved efficacy and safety through the specificity of its action. Overall, the data presented here suggest that combining 4-1BB agonist with the blockade of the inhibitory checkpoint ligand PD-L1 can provide a well-tolerated biologic solution with significantly enhanced and synergistic antitumor activity.

[0091] definition

[0092] As used herein, “PD-L1” and “PD-L1 fragment” refer to the human PD-L1 protein (also known as CD274, B7-H, B7H1, PDCD1L1, and PDCD1LG1) or fragments thereof, unless specified as originating from a non-human species (e.g., “mouse PD-L1”, “mouse PD-L1 fragment”, “monkey PD-L1”, “monkey PD-L1 fragment”, etc.). In one embodiment of this disclosure, human PD-L1 comprises the amino acid sequence shown in NCBI accession no. AAH69381.1. In one embodiment, a human PD-L1 fragment (hPD-L1.mmH) with a C-terminal myc-myc-hexahistidine tag is shown (SEQ ID NO: 79).

[0093] As used herein, “4-1BB” refers to the human 4-1BB protein (also known as CD137 or TNFRS9) expressed on T cells as a co-stimulatory receptor, unless otherwise specified as originating from a non-human species. In one embodiment of this disclosure, human 4-1BB comprises the amino acid sequence shown in NCBI accession No. AAA53133.1. In one embodiment, a human 4-1BB fragment (hCD137 (4-1BB) mmH, SEQ ID NO: 80) is expressed with a C-terminal myc-myc-hexahistidine tag.

[0094] "Isolated" antigen-binding proteins (e.g., antibodies or antigen-binding fragments thereof), polypeptides, polynucleotides, and carriers are at least partially free of other biomolecules from the cells or cell cultures in which they are produced. Such biomolecules include nucleic acids, proteins, other antibodies or antigen-binding fragments, lipids, carbohydrates, or other substances such as cell debris and growth media. Isolated antigen-binding proteins may also be at least partially free of expression system components, such as biomolecules from host cells or their growth media. Generally, the term "isolated" is not intended to mean: the complete absence of such biomolecules; or the absence of water, buffers, or salts; or a component of a pharmaceutical formulation containing an antigen-binding protein (e.g., an antibody or antigen-binding fragment).

[0095] The following references relate to the BLAST algorithm frequently used in sequence analysis: BLAST ALGORITHMS: Altschul et al., (2005) FEBS J. 272(20): 5101-5109; Altschul, SF et al., (1990) J.Mol. Biol. 215:403-410; Gish, W. et al., (1993) Nature Genet. 3:266-272; Madden, TL et al., (1996) Meth. Enzymol. 266:131-141; Altschul, SF et al., (1997) NucleicAcids Res. 25:3389-3402; Zhang, J. et al., (1997) Genome Res. 7:649-656; Wootton, JC et al., (1993) Comput. Chem. 17:149-163; Hancock, JM et al., (1994) Comput. Appl. Biosci. 10:67-70; ALIGNMENT SCORING SYSTEMS: Dayhoff, MO et al., "A model of evolutionary change in proteins." in Atlas of Protein Sequence and Structure, (1978) Vol. 5, Supplement 3. MO Dayhoff (ed.), pp. 345-352, Natl. Biomed. Res. Found., Washington, DC; Schwartz, RM et al., "Matrices for detecting distant relationships." in Atlas of Protein Sequence and Structure, (1978) Vol. 5, Supplement 3.'' MO Dayhoff (ed.), pp. 353-358, Natl. Biomed. Res. Found., Washington, DC; Altschul, SF, (1991) J. Mol. Biol. 219:555-565; States, DJ et al., (1991) Methods 3:66-70; Henikoff, S. et al., (1992) Proc. Natl. Acad. Sci.USA 89:10915-10919; Altschul, SF et al., (1993) J. Mol. Evol. 36:290-300; ALIGNMENTSTATISTICS: Karlin, S. et al., (1990) Proc. Natl. Acad. Sci. USA 87:2264-2268; Karlin, S. et al., (1993) Proc. Natl. Acad. Sci. USA 90:5873-5877; Dembo, A. et al., (1994) Ann. Prob. 22:2022-2039; and Altschul, SF "Evaluating the statistical significance of multiple distinct local alignments." in Theoretical and Computational Methods in Genome Research (S. Suhai, Editor), (1997) Pages 1-14, Plenum, NY. .

[0096] Antibodies are immunoglobulin molecules consisting of four polypeptide chains: two heavy chains (HC) and two light chains (LC) linked together by disulfide bonds. Each heavy chain (HC) contains a heavy chain variable region (abbreviated as HCVR or V in this article). H The heavy chain constant region (e.g., IgG, IgG1, or IgG4) contains three domains, C... H 1. C H 2 and C H 3. Each light chain (LC) contains a light chain variable region (abbreviated as LCVR or V in this document). L ) and light chain constant regions (e.g., λ or κ). The light chain constant region contains a structural domain (C L 1). V H and V L The region can be further subdivided into highly variable regions, called complementarity determining regions (CDRs), interspersed with more conservative regions, called framework regions (FRs). Each V... H and V LIt comprises three CDRs and four FRs, arranged in the following order from the amino terminus to the carboxyl terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. The heavy chain CDR may be referred to as HCDR, and the light chain CDR may be referred to as LCDR. In different embodiments of this disclosure, the FRs of the antibody (or its antigen-binding portion) may be identical to those of a human lineage, or may be natural or artificially modified.

[0097] The antigen-binding arm of a Y-shaped IgG antibody (e.g., a 4-1BB or PD-L1 binding arm) refers to the structural portion of the antibody that confers specificity for binding to the antigen. For example, the antigen-binding arm of an IgG antibody has a heavy chain (HC) that associates with a light chain (LC).

[0098] For example, a bispecific antibody includes an arm (or domain) that binds to a first antigen and another arm (or domain) that binds to a second antigen. For example, the PD-L1x4-1BB bispecific antibody includes an arm that binds to PD-L1 and another arm that binds to 4-1BB.

[0099] Bispecific antigen-binding molecules (e.g., bispecific antibodies) may have an effector arm that binds to a first antigen and a targeting arm that binds to a second antigen. The effector arm may be a first antigen-binding domain (e.g., anti-4-1BB) that binds to an antigen on an effector cell (e.g., T cells). The targeting arm may be a second antigen-binding domain that binds to an antigen on a target cell (e.g., tumor cells or immune cells). In the context of this disclosure, the effector arm binds to 4-1BB, while the targeting arm binds to the inhibitory checkpoint ligand PD-L1.

[0100] The terms "antigen-binding moiety," "antigen-binding fragment," etc., used herein refer to any naturally occurring, enzymatically obtainable, synthetic, or genetically modified polypeptide or glycoprotein that specifically binds to an antigen to form a complex. Multispecific antigen-binding fragments of antibodies bind to multiple antigens (e.g., if the fragment is bispecific, it binds to two different antigens). Antigen-binding fragments of antibodies can be derived from, for example, intact antibody molecules using any suitable standard technique, such as proteolytic digestion or recombinant genetic engineering techniques involving the manipulation and expression of DNA encoding variable and optionally constant domains of the antibody. Some non-limiting examples of antigen-binding fragments include: (i) Fab fragments; (ii) F(ab')2 fragments; (iii) Fd fragments; (iv) Fv fragments; (v) single-stranded Fv (scFv) molecules; and (vi) dAb fragments.

[0101] In one embodiment of this disclosure, the antigen-binding fragment of the antibody will include at least one variable domain. The variable domain may have any size or amino acid composition and will typically include at least one CDR, which is adjacent to or co-framed with one or more frame sequences. L V of domain association H In the antigen-binding fragment of the domain, V H and V L Domains can be positioned relative to each other in any suitable arrangement. For example, variable regions can be dimers and contain V. H -V H V H -V L or V L -V L Dimer. Alternatively, the antigen-binding fragment of the antibody may contain monomer V. H or V L Structural domain.

[0102] In some embodiments, the antigen-binding fragment of the antibody may contain at least one variable domain covalently linked to at least one constant domain. Some non-limiting exemplary configurations of the variable and constant domains that may be present within the antigen-binding fragment of the antibody of this disclosure include: In any configuration of variable and constant domains (including any of the exemplary configurations listed above), the variable and constant domains may be directly connected to each other or connected via all or part of a hinge or connector region. The hinge region may consist of at least two amino acids (e.g., 5, 10, 15, 20, 40, 60, or more), which allows for flexible or semi-flexible connections between adjacent variable and / or constant domains in a single polypeptide molecule. Furthermore, the antigen-binding fragment of the antibody of this disclosure may comprise non-covalently associated fragments with each other and / or with one or more monomers V H or V L Homodimers or heterodimers (or other polymers) of any of the variable and constant domain configurations listed above, whose domains (e.g., via disulfide bonds) are non-covalently associated.

[0103] The term "recombinant" antigen-binding protein, such as an antibody or an antigen-binding fragment thereof, refers to a molecule produced, expressed, isolated, or obtained by techniques or methods known in the art as recombinant DNA technologies, including, for example, DNA splicing and transgenic expression. The term includes antibodies expressed in non-human mammals (including transgenic non-human mammals, such as transgenic mice) or host cells (e.g., Chinese hamster ovary (CHO) cells) or cell expression systems, or antibodies isolated from recombinant human antibody libraries. This disclosure includes recombinant antigen-binding proteins as illustrated herein.

[0104] The term "specific binding" refers to those bindings that target antigens (such as PD-L1 or 4-1BB proteins) using K-type receptors. D The binding affinity is less than about 10. -6 M (e.g., 10) -7 M, 10 -8 M, 10 -9 M, 10 -10 M, 10 -11 M or 10 -12 Antigen-binding proteins (e.g., antibodies or their antigen-binding fragments) that specifically bind to PD-L1, such as antibodies or their antigen-binding fragments, are measured by real-time, label-free biolayer interference, for example at 25°C or 37°C, using the Octet® HTX biosensor, or by surface plasmon resonance, such as BIACORE™, or by solution affinity ELISA. "Anti-PD-L1" refers to an antigen-binding protein (or other molecule, such as an antigen-binding arm), such as an antibody or its antigen-binding fragment, that specifically binds to PD-L1; "Anti-4-1BB" refers to an antigen-binding protein (or other molecule, such as an antigen-binding arm), such as an antibody or its antigen-binding fragment, that specifically binds to 4-1BB. "PD-L1x4-1BB" refers to antigen-binding proteins (or other molecules), such as antibodies or their antigen-binding fragments, that specifically bind to both PD-L1 and 4-1BB.

[0105] This disclosure includes antigen-binding proteins, such as antibodies or antigen-binding fragments, that bind to the same PD-L1 and 4-1BB epitopes as the antigen-binding protein (e.g., REGN6191) of this disclosure.

[0106] The term "epitope" refers to a specific antigen-binding site that interacts with an antigen-binding protein, such as an antigenic determinant cluster interacting with a variable region called a complementary site in an antibody molecule (e.g., an antigenic determinant cluster on PD-L1 or 4-1BB). A single antigen may have more than one epitope. Therefore, different antibodies can bind to different regions of an antigen and may have different biological effects. The term "epitope" may also refer to a site on an antigen to which B cells and / or T cells respond, and / or to an antigenic region bound by an antibody. Epitopes can be defined as structural or functional. Functional epitopes are typically a subset of structural epitopes and contain those residues with an affinity that directly contributes to the interaction. Epitopes can be linear or conformational, i.e., composed of non-linear amino acids. In some embodiments, an epitope may comprise a determinant cluster as a chemically active surface group of a molecule such as an amino acid, a sugar side chain, a phosphoryl group, or a sulfonyl group, and in some embodiments may have specific three-dimensional structural features and / or specific charge features.

[0107] Methods for identifying epitopes of antigen-binding proteins (e.g., antibodies, fragments, or peptides) include alanine scanning mutation analysis, peptide blotting (Reineke (2004) Methods Mol. Biol. 248: 443-63), peptide cleavage analysis, crystallographic studies, and NMR analysis. Alternatively, methods such as epitope excision, epitope extraction, and antigen chemical modification can be used (Tomer (2000) Prot. Sci. 9: 487-496). Another method for identifying amino acids in peptides that interact with antigen-binding proteins (e.g., antibodies, fragments, or peptides) is hydrogen / deuterium exchange detected by mass spectrometry. See, for example, Ehring (1999) Analytical Biochemistry 267: 252-259; Engen and Smith (2001) Anal. Chem. 73: 256A-265A.

[0108] This disclosure includes antigen-binding proteins that compete with the antigen-binding protein of this disclosure (e.g., REGN6188) for binding to 4-1BB and PD-L1. As used herein, the term "competition" means that an antigen-binding protein (e.g., an antibody or its antigen-binding fragment) binds to an antigen and inhibits or blocks the binding of another antigen-binding protein (e.g., an antibody or its antigen-binding fragment) to that antigen. Unless otherwise stated, the term also includes competition between two antigen-binding proteins (e.g., antibodies) in two directions, i.e., a first antibody binds to an antigen and blocks the binding of a second antibody, and vice versa. Therefore, in one embodiment of this disclosure, the competition occurs in one such direction. In some embodiments, the first antigen-binding protein (e.g., an antibody) and the second antigen-binding protein (e.g., an antibody) may bind to the same epitope. Alternatively, the first antigen-binding protein and the second antigen-binding protein (e.g., an antibody) may bind to different but, for example, overlapping or non-overlapping epitopes, wherein the binding of one inhibits or blocks the binding of the second antibody, for example, by steric hindrance. Competition between antigen-binding proteins (e.g., antibodies) can be measured by methods known in the art, such as by real-time, label-free biolayer interferometry. Furthermore, binding competition between antigen-binding proteins (e.g., monoclonal antibodies (mAbs)) can be determined using real-time, label-free biolayer interferometry on the Octet RED384 biosensor (Pall ForteBio Corp.).

[0109] Generally, the antibody or antigen-binding fragments of this disclosure modified in some way retain the ability to specifically bind to PD-L1 and 4-1BB, for example, retaining at least 10% of their PD-L1 and 4-1BB binding activity (compared to the parent antibody) when the activity is expressed in molar terms. Preferably, the antibody or antigen-binding fragments of this disclosure retain at least 20%, 50%, 70%, 80%, 90%, 95%, or 100% or more of the PD-L1 and 4-1BB binding affinity as a parent antibody. The antibody or antigen-binding fragments of this disclosure are also intended to include conserved or non-conserved amino acid substitutions (referred to as “conserved variants” or “functionally conserved variants” of the antibody) that substantially do not alter its biological activity.

[0110] Polypeptides (e.g., chains of immunoglobulins containing the amino acid sequences specifically shown herein, such as the chain of REGN6191) V H V L A “variant” (HC, LC, or CDR) refers to a sequence containing the same amino acid sequence as the reference amino acid sequence shown herein (e.g., SEQ ID NO: 2, 4, 6; 8; 10; 12; 14; 16; 18; 20; 22; 24; 26; 28; 30; 32; 34; 36; 38; 40; 42; 44; 46; 48; 50; 52; 54; 56; 58; 60; 62; 64; 66; 68; 70; 72; 74;) A polypeptide with at least approximately 70% to 99.9% (e.g., at least 70%, 72%, 74%, 75%, 76%, 79%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.5%, or 99.9%) amino acid sequence identity or similarity, compared via a BLAST algorithm, where the parameters of the algorithm are selected to give the maximum match between the sequences over the entire length of each reference sequence (e.g., expected threshold: 10; word length: 3; maximum match within the query range: 0; BLOSUM 62 matrix; gap cost: 11 for existence, 1 for extension; conditional combination score matrix adjustment).

[0111] Furthermore, variants of the polypeptide may include polypeptides whose immunoglobulin chains (e.g., the chain of REGN6191) V H V LHC or LC or CDR: may include the amino acid sequence of a reference polypeptide, the amino acid sequence of which is specifically listed herein, but with one or more mutations (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10), such as one or more missense mutations (e.g., conserved substitutions), nonsense mutations, deletions, or insertions. For example, this disclosure includes such a 4-1BBxPD-L1 antigen-binding protein: which includes the amino acid sequence shown in SEQ ID NO: 18, but with one or more such mutations in the PD-L1 binding arm immunoglobulin light chain (or V L Variants and / or immunoglobulin heavy chains (or V) containing the amino acid sequence shown in SEQ ID NO: 2, but with one or more such mutations. H Variants. In one embodiment of this disclosure, the 4-1BBxPD-L1 antigen-binding protein comprises: immunoglobulin light chain variants containing LCDR1, LCDR2, and LCDR3, wherein one or more (e.g., 1, 2, or 3) of such CDRs have one or more such mutations (e.g., conserved substitutions); and / or immunoglobulin heavy chain variants containing HCDR1, HCDR2, and HCDR3, wherein one or more (e.g., 1, 2, or 3) of such CDRs have one or more such mutations (e.g., conserved substitutions).

[0112] "Conservatively modified variants" or "conservative substitutions," for example, in the context of immunoglobulin chains described herein, refer to variants in which one or more amino acids in a polypeptide are substituted with other amino acids having similar characteristics (e.g., charge, side chain size, hydrophobicity / hydrophilicity, ossicular structure, and rigidity). Such changes can often be made without significantly impairing the biological activity of the antibody or fragment. Those skilled in the art will recognize that, in general, a single amino acid substitution in a non-essential region of a polypeptide does not substantially alter the biological activity (see, for example, Watson et al., (1987) Molecular Biology of the Gene, The Benjamin / CummingsPub. Co., p. 224 (4th edition)). Furthermore, substitutions of structurally or functionally similar amino acids are unlikely to significantly impair biological activity. This disclosure includes PD-L1x4-1BB antigen-binding proteins and / or binding arms comprising such conservedly modified variant immunoglobulin chains.

[0113] Examples of amino acid groups with side chains possessing 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. Alternatively, a conservative substitution is any variation with a positive value in the PAM250 log-likelihood matrix disclosed in Gonnet et al., (1992) Science 256: 1443-45.

[0114] Unless the context clearly specifies otherwise, nouns without quantifiers as used herein include one / something and more / somethings. Unless otherwise stated, the terms “comprising,” “including,” “containing,” or “having,” and their variations, as used herein, mean to cover the items listed thereafter and their equivalents, as well as other subjects. The phrases “in one embodiment,” “in multiple embodiments,” “in some embodiments,” etc., are used repeatedly herein. Such phrases do not necessarily refer to the same embodiment, but they may refer to the same embodiment unless the context indicates otherwise. The terms “and / or” or “ / ” as used herein mean any one of the items associated with the term, any combination of the items, or all of the items.

[0115] PD-L1x4-1BB antigen-binding molecule, antibody and its antigen-binding fragment

[0116] The antibodies or antigen-binding molecules of this disclosure can be bispecific or multispecific. Multispecific antibodies or antigen-binding molecules may be specific to different epitopes of a single target polypeptide, or may contain antigen-binding domains specific to more than one target polypeptide. See, for example, Tutt et al., 1991, J. Immunol. 147:60-69; Kufer et al., 2004, Trends Biotechnol. 22:238-244. The antibodies of this disclosure can be linked to or co-expressed with another functional molecule, such as another peptide or protein. For example, an antibody or fragment thereof can be functionally linked to one or more other molecular entities, such as another antibody or antibody fragment (e.g., through chemical coupling, gene fusion, non-covalent association, or other means), to produce bispecific or multispecific antibodies with a second binding specificity.

[0117] The term "anti-4-1BB antibody" as used herein is intended to include both monospecific anti-4-1BB antibodies and multispecific (e.g., bispecific) antibodies comprising a 4-1BB binding arm and a second arm that binds to a target antigen. Therefore, this disclosure includes bispecific antibodies in which one arm of an immunoglobulin binds human 4-1BB, and the other arm of the immunoglobulin is specific for a target antigen. The target antigen bound to the other arm of a 4-1BB bispecific antibody can be any antigen expressed on or near cells, tissues, organs, microorganisms, or viruses, against which an immune response is desired to be targeted. The 4-1BB binding arm may comprise any HCVR / LCVR or CDR amino acid sequence as shown in Tables 3 and 8 herein. In some embodiments, the 4-1BB binding arm binds human 4-1BB and induces human T cell proliferation.

[0118] According to certain exemplary embodiments, this disclosure includes a bispecific antigen-binding molecule that specifically binds to 4-1BB and PD-L1. Such a molecule may be referred to herein as, for example, "anti-4-1BB / anti-PD-L1", or "anti-4-1BBxPD-L1", or "4-1BBxPD-L1", or "PD-L1x4-1BB", or "anti-PD-L1 / anti-4-1BB", or "anti-PD-L1x4-1BB", or "PD-L1x4-1BB" bispecific molecule, or "anti-PD-L1 x anti-4-1BB" or "anti-4-1BBx anti-PD-L1", or other similar terms.

[0119] According to certain exemplary embodiments, a bispecific antigen-binding molecule (e.g., a bispecific antibody) may have an effector arm and a target arm. The effector arm may be a first antigen-binding domain (e.g., an anti-4-1BB antibody) that binds to an antigen on an effector cell (e.g., a T cell). The target arm may be a second antigen-binding domain (e.g., an anti-PD-L1 antibody) that binds to an antigen on a target cell (e.g., a tumor cell or an antigen-presenting cell). According to certain exemplary embodiments, the effector arm binds to 4-1BB, and the target arm binds to the inhibitory checkpoint ligand PD-L1. The bispecific anti-4-1BB / PD-L1 can provide a pan-tumor co-stimulatory approach (tumor unaware) to convert the inhibitory checkpoint (PD-L1), which is widely present in a variety of tumor indications, into a co-stimulatory signal to activate 4-1BB on T cells.

[0120] As used herein, the term "antigen-binding molecule" means a protein, polypeptide, or molecular complex that specifically binds to a particular antigen and comprises or consists of at least one complementary determinant region (CDR), alone or in combination with one or more additional CDRs and / or frame regions (FRs). In some embodiments, the antigen-binding molecule is an antibody or antibody fragment, as those terms are defined elsewhere herein.

[0121] As used herein, the term "bispecific antigen-binding molecule" refers to a protein, peptide, or molecular complex (e.g., an antibody or its antigen-binding fragment) comprising at least a first antigen-binding domain and a second antigen-binding domain. Each antigen-binding domain within a bispecific antigen-binding molecule contains at least one antigen-binding domain (CDR), which is alone or in combination with one or more additional CDRs and / or FRs, and binds specifically to a particular antigen. In the context of this disclosure, the first antigen-binding domain specifically binds to a first antigen (e.g., 4-1BB), and the second antigen-binding domain specifically binds to a second, different antigen (e.g., PD-L1).

[0122] In some exemplary embodiments of this disclosure, 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).

[0123] The first antigen-binding domain and the second antigen-binding domain may be directly or indirectly linked to each other to form the bispecific antigen-binding molecule of this disclosure. Alternatively, the first antigen-binding domain and the second antigen-binding domain may each be linked to a separate multimerizing domain. Association of one multimerizing domain with another multimerizing domain promotes association between the two antigen-binding domains, thereby forming a bispecific antigen-binding molecule. As used herein, "multimerizing domain" refers to any macromolecule, protein, polypeptide, peptide, or amino acid having the ability to associate with a second multimerizing domain of the same or similar structure or construction. For example, a multimerizing domain may be containing immunoglobulin C. H 3-domain polypeptides. A non-limiting example of a polymerized component is the Fc region of an immunoglobulin (containing C... H 2-C H 3. A domain, such as the Fc domain of IgG, wherein the IgG is selected from isotypes IgG1, IgG2, IgG3, and IgG4, and any allotypes within each isotype group. The Fc domain may contain wild-type or modified IgG isotypes.

[0124] The bispecific antigen-binding molecules disclosed herein typically comprise two multimerizing domains, such as two Fc domains, each independently forming part of a separate antibody heavy chain. The first and second multimerizing domains can be the same IgG isotype, such as IgG1 / IgG1, IgG2 / IgG2, or IgG4 / IgG4. Alternatively, the first and second multimerizing domains can be different IgG isotypes, such as IgG1 / IgG2, IgG1 / IgG4, or IgG2 / IgG4.

[0125] In some embodiments, the polymerizing domain is an amino acid sequence or Fc fragment of 1 to 200 amino acids containing at least one cysteine ​​residue. In other embodiments, the polymerizing domain is a cysteine ​​residue, or a short cysteine-containing peptide. Still other polymerizing domains include peptides or polypeptides comprising or composed of leucine zippers, helical-cyclic motifs, or coiled-coil motifs.

[0126] Any bispecific antibody form or technology can be used to prepare the bispecific antigen-binding molecules of this disclosure. For example, an antibody or antigen-binding fragment having a first antigen-binding specificity can be functionally linked with one or more other molecular entities, such as another antibody or antibody fragment having a second antigen-binding specificity (e.g., by chemical coupling, genetic fusion, non-covalent association, or other means) to produce a bispecific antigen-binding molecule. Specific exemplary bispecific forms that can be used in the context of this disclosure include, but are not limited to, scFv-based forms or bispecific antibody forms, IgG-scFv fusions, dual variable domain (OVO)-Ig, quadromas, knocks-in-holes, common light chains (e.g., common light chains with bulge-in-hole structures, etc.), CrossMab, CrossFab, (SEEO) bodies, leucine zippers, Ouobody, IgG1 / IgG2, dual-action Fab (OAF)-IgG, and Mab. 2 Bispecific form (for a review of the aforementioned forms, see, for example, Klein et al., 2012, mAbs 4:6, 1-11 and references cited therein).

[0127] In the context of the bispecific antigen-binding molecules of this disclosure, compared to the wild-type, naturally occurring Fc domain form, the polymerized domain, such as the Fc domain, may contain one or more amino acid variations (e.g., insertions, deletions, or substitutions). For example, this disclosure includes bispecific antigen-binding molecules that contain one or more modifications in the Fc domain such that the modified Fc domain has a modified binding interaction (e.g., enhanced or reduced) between Fc and FcRn. In one embodiment, the bispecific antigen-binding molecule is located at C... H 2 or C H Region 3 contains a modification that enhances the affinity of the Fc domain for FcRn in acidic environments (e.g., in endosomes with a pH range of about 5.5 to about 6.0). Some 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).

[0128] This disclosure also includes the first C H 3 structural domains and second Ig C H A bispecific antigen-binding molecule with three domains, wherein the first and second IgC H The three domains differ from each other by at least one amino acid, and wherein at least one amino acid difference reduces the binding of the bispecific antibody to protein A compared to a bispecific antibody lacking amino acid difference. In one embodiment, the first IgC H 3-domain binding to protein A, and second IgC H Domain 3 contains mutations that reduce or eliminate protein A binding, such as H95R modification (IMGT exon number; H435R (EU number)). Second C H3 may also include Y96F modification (IMGT; Y436F (EU)). Other modifications that may exist in the second CH3 include: in the case of IgG1 antibody, D16E, L18M, N44S, K52N, V57M, and V821 (IMGT; D356E, L358M, N384S, K392N, V397M, and V4221 (EU)); in the case of IgG2 antibody, N44S, K52N, and V821 (IMGT; N384S, K392N, and V4221 (EU)); and in the case of IgG4 antibody, Q15R, N44S, K52N, V57M, R69K, E79Q, and V821 (IMGT; Q355R, N384S, K392N, V397M, R409K, E419Q, and V4221 (EU)).

[0129] In some implementations, the Fc domain can be chimeric, combining Fc sequences derived from more than one immunoglobulin isotype. For example, a chimeric Fc domain may contain partial or complete Fc sequences derived from human IgG1, human IgG2, or human IgG4C. H C in Zone 2 H 2 sequences, and C sequences derived partially or entirely from human IgG1, human IgG2, or human IgG4. H 3. Sequence. The chimeric Fc domain may also include a chimeric hinge region. For example, the chimeric hinge may include an "upper hinge" sequence derived from the hinge region of human IgG1, human IgG2, or human IgG4, combined with a "lower hinge" sequence derived from the hinge region of human IgG1, human IgG2, or human IgG4. A specific example of a chimeric Fc domain that may be included in any antigen-binding molecule described herein includes, from the N-terminus to the C-terminus: [IgG4C H 1]-[IgG4 upper hinge]-[IgG2 lower hinge]-[IgG4 CH2]-[IgG4 C H 3). Another example of a chimeric Fc domain that may be included in any antigen-binding molecule shown herein includes, from the N-terminus to the C-terminus: [IgG1 C H 1]-[IgG1 upper hinge]-[IgG2 lower hinge]-[IgG4 C H 2]-[IgG1 C H 3]. These and other examples of chimeric Fc domains that may be included in any antigen-binding molecule contained in this disclosure are described in WO2014 / 022540 A1. Chimeric Fc domains having these general structural arrangements and their variants may have altered Fc receptor binding, which in turn affects Fc effector function.

[0130] The antibody and antigen-binding fragments of this disclosure comprise immunoglobulin chains containing the amino acid sequences (and variants thereof) specifically shown herein, along with cellular and in vitro post-translational modifications to the antibody or fragment. For example, this disclosure includes antibodies and antigen-binding fragments that specifically bind to PD-L1 and 4-1BB, comprising the heavy chain and / or light chain amino acid sequences shown herein; and antibodies and fragments with one or more asparagine, serine, and / or threonine residues glycosylated, one or more asparagine residues deamidated, one or more residues (e.g., Met, Trp, and / or His) oxidized, N-terminal glutamine being pyroglutamic acid (pyroE), and / or C-terminal lysine or other amino acid deletions.

[0131] The bispecific antigen-binding molecule of this disclosure comprises a first antigen-binding arm (“4-1BB binding arm” or “4-1BB binding domain”) that specifically binds to 4-1BB. In some embodiments, the 4-1BB binding arm comprises HCVR and LCVR containing the amino acid sequences disclosed herein. The bispecific antigen-binding molecule also comprises a second antigen-binding arm (“PD-L1 binding arm” or “PD-L1 binding domain”) that specifically binds to PD-L1. In some embodiments, the PD-L1 binding arm comprises HCVR and LCVR containing the amino acid sequences disclosed herein. In some embodiments, the PD-L1 binding arm comprises a heavy chain immunoglobulin and a corresponding light chain immunoglobulin, said heavy chain immunoglobulin comprising a V-type immunoglobulin containing a combination of heavy chain CDRs (HCDR1, HCDR2, and HCDR3). H The light chain immunoglobulin comprises a V containing a combination of light chain CDRs (LCDR1, LCDR2, and LCDR3). L It is described in this article or in WO2014 / 004427.

[0132] This disclosure includes multispecific (e.g., bispecific) antigen-binding proteins (e.g., antibodies or antigen-binding fragments) that include one or more 4-1BB binding arms and one or more PD-L1 binding arms.

[0133] The 4-1BB binding arm is the portion of a multispecific antigen-binding protein that confers 4-1BB binding. For example, the 4-1BB binding arm of a Y-shaped IgG antibody refers to the structural portion of the antibody that confers specificity for 4-1BB binding. For example, in one embodiment of this disclosure, the 4-1BB binding arm includes HCDR1, LCDR1, HCDR2, LCDR2, HCDR3, and LCDR3, which specifically bind to 4-1BB; HVCR (V H ) and LCVR (V L ); and / or HC and LC.

[0134] In one embodiment of this disclosure, the 4-1BB binding arm comprises a heavy chain immunoglobulin and a corresponding light chain immunoglobulin, said heavy chain immunoglobulin comprising a combination of heavy chain CDRs (HCDR1, HCDR2, and HCDR3) as described herein. H The corresponding light chain immunoglobulin comprises a V containing a combination of light chain CDRs (LCDR1, LCDR2, and LCDR3) as described herein. L In one embodiment of this disclosure, the 4-1BB binding arm includes the heavy chain variable region (V) described herein. H ) and the corresponding light chain variable region (V L ).

[0135] The bispecific PD-L1x4-1BB antigen-binding protein of this disclosure, which binds to 4-1BB on the surface of T cells and activates 4-1BB signaling, thereby enhancing T cell activation and / or proliferation, is referred to herein as “co-stimulatory.” T cell activation is initiated upon the binding of the T cell receptor (TCR) / CD3 complex to the peptide-MHC complex (“Signal 1”); activation is then enhanced by the binding of a second “co-stimulatory” receptor, such as the 4-1BB receptor on the T cell, to its homologous ligand on the target cell (“Signal 2”). For example, T cell activation by a 4-1BB bispecific antibody may be caused by signal amplification in response to the recognition of endogenous tumor antigens by the TCR / CD3 complex, or in response to the activation of “Signal 1” via a CD3 bispecific agent.

[0136] Polynucleotides and their preparation methods

[0137] Isolated polynucleotide molecules or groups of polynucleotide molecules containing polynucleotide sequences that encode an immunoglobulin chain of any PD-L1x4-1BB bispecific antigen-binding molecule described herein are part of this disclosure. This disclosure also includes vectors or vector groups containing said polynucleotide molecules and / or host cells (e.g., Chinese hamster ovary (CHO) cells) containing said polynucleotide molecules, vectors, or antigen-binding proteins described herein.

[0138] Polynucleotide molecules or sequences include DNA and RNA. This disclosure includes any polynucleotide molecules or sequences of this disclosure, for example, immunoglobulin V encoding the PD-L1 binding arm and / or the 4-1BB binding arm. H V LCDR-H, CDR-L, HC, or LC, optionally, are operatively linked to a promoter or other expression control sequence. For example, this disclosure provides any polynucleotide (e.g., DNA) comprising the nucleotide sequences shown in Table 2 and Table 4.

[0139] This disclosure includes polynucleotides comprising the nucleotide sequences shown in SEQ ID NO: 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 29, 31, 33, 35, 39, 41, 43, 45, 49, 51, 53, 55, 57, 59, 61, 63, 65, 67, 69, 71, 73, 75 and / or 77, optionally operatively linked to a promoter or other expression control sequence or other polynucleotide sequence.

[0140] Generally, a “promoter” or “promoter sequence” is a DNA regulatory region capable of binding to RNA polymerases in a cell (e.g., proteins or substances that bind directly or via other promoters) and initiating transcription of a coding sequence. Promoters can be operatively linked to other expression control sequences, including enhancer and repressor sequences and / or polynucleotides of this disclosure. Promoters that can be used to control gene expression include, but are not limited to, cytomegalovirus (CMV) promoters (US Patent Nos. 5,385,839 and 5,168,062), the early promoter region of SV40 (Benoist et al., (1981) Nature 290:304-310), promoters contained in the 3' long terminal repeat of Rous sarcoma virus (Yamamoto et al., (1980) Cell 22:787-797), herpes simplex kinase promoters (Wagner et al., (1981) Proc. Natl. Acad. Sci. USA 78:1441-1445), regulatory sequences of metallothionein genes (Brinster et al., (1982) Nature 296:39-42); and prokaryotic expression vectors such as β-lactamase promoters (VIIIa-Komaroff et al., (1978) Proc. Natl. Acad. Sci. USA). 75:3727-3731) or tac promoter (DeBoer et al., (1983) Proc. Natl. Acad. Sci. USA 80:21-25); see also "Useful proteins from recombinant bacteria" in Scientific American (1980) 242:74-94; and promoter elements from yeast or other fungi, such as Gal4 promoter, ADC (alcohol dehydrogenase) promoter, PGK (phosphoglycerol kinase) promoter or alkaline phosphatase promoter.

[0141] The polynucleotide encoding the polypeptide is "operably linked" to a promoter or other expression control sequence: in a cell or other expression system, this sequence directs the RNA polymerase-mediated transcription of the coding sequence into RNA, preferably mRNA, which can then be spliced ​​(if it contains introns) and, optionally, translated into a protein encoded by the coding sequence.

[0142] This disclosure includes polynucleotides encoding immunoglobulin polypeptide chains that are variants of those nucleotide sequences specifically shown herein. A “variant” of a polynucleotide is a polynucleotide containing at least about 70% to 99.9% (e.g., 70%, 72%, 74%, 75%, 76%, 79%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.5%, 99.9%) of the identity with the reference nucleotide sequences shown herein; this is done by comparison using a BLAST algorithm, where the parameters of the algorithm are selected to give the maximum match between the sequences over the entire length of the respective reference sequences (e.g., expected threshold: 10; word length: 28; maximum match within the query range: 0; match / non-match score: 1, -2; gap penalty: linear). In one embodiment of this disclosure, the variant of the nucleotide sequence specifically illustrated herein comprises one or more (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12) point mutations, insertions (e.g., in-frame insertions), or deletions (e.g., in-frame deletions) of one or more nucleotides. In one embodiment of this disclosure, such mutations may be missense or nonsense mutations. In one embodiment of this disclosure, such variant polynucleotides encode immunoglobulin polypeptide chains that can be incorporated into the PD-L1 binding arm and / or the 4-1BB binding arm, i.e., such that the protein retains specific binding to PD-L1 and / or 4-1BB.

[0143] Eukaryotic and prokaryotic host cells, including mammalian cells, can be used as hosts for expressing PD-L1x4-1BB antigen-binding proteins (e.g., antibodies or their antigen-binding fragments) or their antigen-binding arms. Such host cells are well known in the art, and many are available from the American Type Culture Collection (ATCC). These host cells include, in particular, Chinese hamster ovary (CHO) cells, NSO, SP2 cells, HeLa cells, baby hamster kidney (BHK) cells, monkey kidney cells (COS) cells, human hepatocellular carcinoma cells (e.g., Hep G2), A549 cells, 3T3 cells, HEK-293 cells, and many other cell lines. Mammalian host cells include cells from humans, mice, rats, dogs, monkeys, pigs, goats, cattle, horses, and hamsters. Other cell lines that can be used include insect cell lines (such as fall armyworm (Spodoptera frugiperda) or white armyworm (Trichoplusia ni)), amphibian cells, bacterial cells, plant cells, and fungal cells.Fungal cells include cells of yeast and filamentous fungi, including, for example, *Pichia*, *Pichia pastoris*, *Pichia finlandica*, *Pichia trehalophila*, *Pichia koclamae*, *Pichia membranaefaciens*, *Pichia minuta* (Ogataea minuta, Pichia lindneri), *Pichia opuntiae*, *Pichia athermotolerans*, *Pichia salictaria*, *Pichia guercuum*, *Pichia pijperi*, *Pichia stiptis*, *Pichia methanolica*, *Pichia* sp., *Saccharomyces cerevisiae*, *Saccharomyces* sp., and *Hansenula polymorpha*. The fungi include *C. polymorpha*, *Kluyveromyces* sp., *Kluyveromyces lactis*, *Candida albicans*, *Aspergillus niduLans*, *Aspergillus niger*, *Aspergillus oryzae*, *Trichoderma reesei*, *Chrysosporium lucknowense*, *Fusarium* sp., *Fusarium gramineum*, *Fusarium venenatum*, *Physcomitrella patens*, and *Neurospora crassa*.This disclosure includes isolated host cells (e.g., CHO cells or any of the types of host cells described above) containing anti-PD-L1 x anti-4-1BB antigen-binding proteins of this disclosure, such as REGN6188, REGN6189, REGN6190, and REGN6191, as well as the anti-PD-L1 x anti-4-1BB antigen-binding proteins shown in Table 9, or polynucleotides encoding the heavy and / or light chains of immunoglobulins (Ig); and / or one or more polynucleotides encoding the PD-L1 binding arm and the 4-1BB binding arm of the multispecific antigen-binding protein of this disclosure.

[0144] This disclosure also includes cells expressing PD-L1 and / or 4-1BB or antigen fragments or fusions thereof (e.g., His6, Fc, and / or myc), wherein the PD-L1 and / or 4-1BB or antigen fragments or fusions thereof are bound to: PD-L1x4-1BB antigen-binding proteins of this disclosure (e.g., antibodies or antigen-binding fragments thereof), such as REGN6188, REGN6189, REGN6190, and REGN6191; and bispecific antibodies prepared by combining any PD-L1 HCVR arm of Table 1 (e.g., the HCVR arms of parental monoclonal antibodies 9364P3 and 9373P2) with any 4-1BB HCVR arm of Table 3 (e.g., the HCVR arms of parental 25894P2, 25898P2, 25907P2, and 25921P2), or any bispecific antibody shown in Table 9; for example, wherein the cells are in vivo or in vitro.

[0145] Additionally, this disclosure provides a complex comprising a PD-L1x4-1BB antigen-binding protein, such as an antibody or antigen-binding fragment thereof, conjugated to a PD-L1 and / or 4-1BB peptide or an antigen fragment thereof or a fusion thereof, and / or a secondary antibody or antigen-binding fragment thereof that specifically binds to a PD-L1x4-1BB antibody or fragment thereof (e.g., a detectable labeled secondary antibody). In one embodiment of this disclosure, the complex is in vitro (e.g., immobilized to a solid substrate) or in vivo. In one embodiment of this disclosure, PD-L1 is on the surface of tumor cells or antigen-presenting cells, and 4-1BB is on the surface of immune cells (e.g., T cells). In one embodiment of this disclosure, the T cells are activated.

[0146] Several methods for generating recombinant antibodies are known in the art. One example of a method for generating recombinant antibodies is disclosed in US4816567. Transformation can be performed by any known method for introducing polynucleotides into host cells. Methods for introducing heterologous polynucleotides into mammalian cells are well known in the art and include dextran-mediated transfection, calcium phosphate precipitation, polybrene-mediated transfection, protoplast fusion, electroporation, encapsulation of polynucleotides in liposomes, biolistic injection, and direct microinjection of DNA into the cell nucleus. Additionally, nucleic acid molecules can be introduced into mammalian cells via viral vectors. Methods for transforming cells are well known in the art. See, for example, US Patent Nos. 4399216; 4912040; 4740461 and 4959455.

[0147] This disclosure includes a method for preparing the anti-PD-L1 x anti-4-1BB (e.g., REGN6188, REGN6189, REGN6190, and REGN6191) antigen-binding protein of this disclosure, such as the antibody or antigen-binding fragment thereof, or the recombinant method thereof, the recombinant method comprising:

[0148] (i) Introducing one or more polynucleotides encoding immunoglobulin light and heavy chains into a host cell, the immunoglobulin light and heavy chains encoding antigen-binding arms of PD-L1x4-1BB antigen-binding protein, for example, wherein the polynucleotides are located in a vector; and / or integrating into the host cell chromosome and / or operatively linking to a promoter;

[0149] (ii) Culture host cells (e.g., CHO or Pichia pastoris or Pichia pastoris) under conditions favorable to the expression of the polynucleotide, and

[0150] (iii) Optionally, antigen-binding proteins (e.g., antibodies or antigen-binding fragments) or chains are isolated from host cells and / or the culture medium in which the host cells are cultured. This disclosure also includes PD-L1x4-1BB antigen-binding proteins, such as antibodies and their antigen-binding fragments, as products of the methods described herein (and optionally, the purification methods described herein).

[0151] In one embodiment of this disclosure, a method for preparing PD-L1x4-1BB (e.g., REGN6188, REGN6189, REGN6190, and REGN6191) antigen-binding proteins, such as antibodies or antigen-binding fragments thereof, includes methods for purifying the antigen-binding proteins, such as by column chromatography, precipitation, and / or filtration. As discussed, the products of such methods also constitute part of this disclosure.

[0152] Sequence variants

[0153] Compared to corresponding germline sequences derived from a single antigen-binding domain, antibodies and bispecific antigen-binding molecules of this disclosure may contain one or more amino acid substitutions, insertions, and / or deletions in the frame regions and / or CDR regions of the heavy and light chain variable domains. Such mutations can be readily identified by comparing the amino acid sequences disclosed herein with germline sequences available from, for example, public antibody sequence databases. Antigen-binding molecules of this disclosure may contain antigen-binding fragments derived from any exemplary amino acid sequence disclosed herein, wherein one or more amino acids in one or more frame regions and / or CDR regions are mutated to corresponding residues of the germline sequence from which the antibody is derived, or mutated to corresponding residues of another human germline sequence, or mutated to conserved amino acid substitutions of the corresponding germline residues (such sequence changes are collectively referred to herein as “germline mutations”). Those skilled in the art can readily generate numerous antibody and antigen-binding fragments containing one or more single germline mutations or combinations thereof, starting from the heavy and light chain variable region sequences disclosed herein. In some embodiments, V H and / or V L All frame and / or CDR residues within the domain mutate back to residues present in the original germline sequence from which the antigen-binding domain originally originated. In other embodiments, only certain residues mutate back to the original germline sequence, for example, mutated residues present only in the first 8 amino acids of FR1 or the last 8 amino acids of FR4, or mutated residues present only in CDR1, CDR2, or CDR3. In other embodiments, one or more frame and / or CDR residues mutate to corresponding residues of a different germline sequence (i.e., a germline sequence different from the original germline sequence from which the antigen-binding domain originally originated). Furthermore, the antigen-binding domain may contain any combination of two or more germline mutations within the frame and / or CDR regions, for example, where certain individual residues mutate to corresponding residues of a specific germline sequence, while certain other residues different from the original germline sequence are maintained or mutated to corresponding residues of a different germline sequence. Once obtained, one or more desired properties of the antigen-binding domain containing one or more germline mutations can be readily tested, such as improved binding specificity, increased binding affinity, improved or enhanced antagonistic or agonistic biological properties (as the case may be), reduced immunogenicity, etc. Bispecific antigen-binding molecules containing one or more antigen-binding domains obtained in this general manner are covered in this disclosure.

[0154] This disclosure 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 conserved substitutions. For example, this disclosure includes antigen-binding molecules comprising antigen-binding domains having HCVR, LCVR, and / or CDR amino acid sequences having, for example, 10 or fewer, 8 or fewer, 6 or fewer, 4 or fewer, etc., conserved amino acid substitutions relative to any of the HCVR, LCVR, and / or CDR amino acid sequences disclosed herein. A “conserved amino acid substitution” is an amino acid substitution in which an amino acid residue is replaced by another amino acid residue having a side chain (R group) with similar chemical properties (e.g., charge or hydrophobicity). Generally, conserved amino acid substitutions do not substantially alter the functional properties of the protein. Examples of amino acid groups with side chains having 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, conservative substitutions are any changes with positive values ​​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.

[0155] This disclosure also includes antigen-binding molecules comprising an antigen-binding domain that is substantially identical to any of the HCVR, LCVR, and / or CDR amino acid sequences disclosed herein. When referring to amino acid sequences, the terms “significant identity” or “substantially identical” mean that, when optimally aligned, for example, using the GAP or BESTFIT procedure with default vacancy weights, two amino acid sequences share at least 95% sequence identity, and more preferably at least 98% or 99% sequence identity. Preferably, dissimilar residue sites differ due to conserved amino acid substitutions. In cases where two or more amino acid sequences differ from each other due to conserved substitutions, the percentage of sequence identity or similarity can be upregulated to correct for the conservatism of the substitutions. Means for making such adjustments are well known to those skilled in the art. See, for example, Pearson (1994) Methods Mol. BioI. 24: 307-331.

[0156] Sequence similarity of peptides, also known as sequence identity, is typically determined using sequence analysis software. Protein analysis software uses similarity measures specified for various substitutions, deletions, and other modifications, including conserved amino acid substitutions, to match similar sequences. For example, GCG software includes programs such as Gap and Bestfit, which can use default parameters to determine sequence homology or sequence identity between closely related peptides, such as homologous peptides from different biological species, or between wild-type proteins and their mutant proteins. See, for example, GCG version 6.1. Peptide sequences can also be compared using FASTA (the program in GCG version 6.1) with default or recommended parameters. FASTA (e.g., FASTA2 and FASTA3) provides alignment of the best overlapping region between the query sequence and the retrieved sequence and a percentage of sequence identity (Pearson (2000) Methods Mol. BioI. 132: 185-219). When comparing sequences of this disclosure with databases containing large numbers of sequences from different organisms, another preferred algorithm is the computer program BLAST, particularly BLASTP or TBLASTN, using default parameters. See, for example, Altschul et al., (1990) J. Mol. BioI. 215:403-410; Altschul et al., (1997) Nucleic Acids Res. 25:3389-402.

[0157] Antibodies containing Fc variants

[0158] According to certain embodiments of this disclosure, an anti-PD-L1X / anti-4-1BB bispecific antigen-binding molecule comprising an Fc domain is provided, wherein the Fc domain contains one or more mutations that enhance or weaken antibody binding to the FcRn receptor, for example, at acidic pH compared to neutral pH. For example, this disclosure includes CFc in the Fc domain. H 2 or C H Region 3 contains a mutated antibody and antigen-binding molecule, wherein the mutation increases the affinity of the Fc domain for FcRn in an acidic environment (e.g., in endosomes with a pH range of about 5.5 to about 6.0). When administered to animals, such a mutation can lead to a prolonged serum half-life of the antibody. Some non-limiting examples of such Fc modifications include, for example, modifications at the following positions:

[0159] 250 (e.g., E or Q);

[0160] 250 and 428 (e.g., L or F);

[0161] 252 (e.g., L / Y / F / W or T),

[0162] 254 (e.g., S or T), and / or

[0163] 256 (e.g., S / R / Q / E / D or T);

[0164] Or, as an embellishment in the following locations:

[0165] 428 and / or 433 (e.g., H / L / R / S / P / Q or K) and / or

[0166] 434 (e.g., H / F or Y);

[0167] Or, as an embellishment in the following locations:

[0168] 250 and / or 428;

[0169] Or, as an embellishment in the following locations:

[0170] 307 or 308 (e.g., 308F, V308F), and / or

[0171] 434.

[0172] In one implementation, the modification includes:

[0173] Modifications include 428L (e.g., M428L) and 434S (e.g., N434S);

[0174] Modifications include 428L, 259I (e.g., V259I) and 308F (e.g., V308F);

[0175] Modified with 433K (e.g., H433K) and 434 (e.g., 434Y);

[0176] Modifications 252, 254, and 256 (e.g., 252Y, 254T, and 256E);

[0177] 250Q and 428L modifications (e.g., T250Q and M428L); and / or

[0178] 307 and / or 308 modifications (e.g., 308F or 308P).

[0179] For example, this disclosure includes a PD-L1x4-1BB bispecific antigen-binding molecule comprising an Fc domain, wherein the Fc domain comprises one or more pairs or one or more groups of mutations selected from:

[0180] 250Q and 248L (e.g., T250Q and M248L);

[0181] 252Y, 254T and 256E (e.g. M252Y, S254T and T256E);

[0182] 428L and 434S (e.g., M428L and N434S); and

[0183] 433K and 434F (e.g., H433K and N434F).

[0184] This disclosure also includes the first C H 3 structural domains and second Ig C H A bispecific antigen-binding molecule with three domains, wherein the first and second IgC HThe three domains differ from each other by at least one amino acid, and wherein at least one amino acid difference reduces the binding of the bispecific antibody to protein A compared to a bispecific antibody lacking amino acid difference. In one embodiment, the first IgC H 3-domain binding to protein A and second IgC H Domain 3 contains mutations that reduce or eliminate protein A binding, such as H95R modification (IMGT exon number; H435R (EU number)). Second C H 3 may also include Y96F modification (IMGT; Y436F (EU)). See, for example, U.S. Patent No. 8,586,713. Second C H Other modifications that may exist in 3 include: in the case of IgG1 antibody, D16E, L18M, N44S, K52N, V57M and V82I (IMGT; D356E, L358M, N384S, K392N, V397M and V422I (EU)); in the case of IgG2 antibody, N44S, K52N and V82I (IMGT; N384S, K392N and V422I (EU)); and in the case of IgG4 antibody, Q15R, N44S, K52N, V57M, R69K, E79Q and V82I (IMGT; Q355R, N384S, K392N, V397M, R409K, E419Q and V422I (EU)).

[0185] All possible combinations of the aforementioned Fc domain mutations and other mutations within the antibody variable domains disclosed herein are contemplated within the scope of this disclosure.

[0186] Biological characteristics of bispecific antibodies and antigen-binding molecules

[0187] This disclosure includes antibodies and antigen-binding fragments thereof that bind to human 4-1BB and PD-L1 with high affinity. This disclosure also includes antibodies and antigen-binding fragments thereof that bind to human 4-1BB and / or PD-L1 with intermediate or low affinity, depending on the therapeutic context and the specific targeting characteristics desired. For example, in the case of a bispecific antigen-binding molecule where one arm binds to 4-1BB and the other arm binds to a target antigen (e.g., PD-L1), it is desirable for the target antigen-binding arm to bind to the target antigen with high affinity, while the anti-4-1BB arm binds to 4-1BB only with intermediate or low affinity. In this way, the antigen-binding molecule can preferentially target cells expressing the target antigen, while avoiding general / untargeted 4-1BB binding and the associated adverse side effects.

[0188] According to certain implementations, this disclosure includes K with a density of less than about 200 nM. DAntibodies binding to human 4-1BB (e.g., at 25°C) and antigen-binding fragments of the antibodies, as determined by surface plasmon resonance, are used, for example, in the assay format defined in Example 2 herein. In some embodiments, the antibodies or antigen-binding fragments of this disclosure are expressed in Kc values ​​of less than about 200 nM, less than about 150 nM, less than about 100 nM, less than about 90 nM, less than about 80 nM, less than about 60 nM, less than about 40 nM, less than about 30 nM, less than 20 nM, less than 10 nM, or less than 5 nM. D Combined with 4-1BB, as measured by surface plasmon resonance, for example, using an assay format defined in Example 2 herein or a substantially similar assay. In some embodiments, the antibody or antigen-binding fragment of this disclosure is expressed at a Kc of about 10 nM to about 200 nM. D Combined with 4-1BB.

[0189] This disclosure also includes antibodies and antigen-binding fragments of the present disclosure that bind to 4-1BB with a dissociation half-life (t½) greater than about 0.1 minutes, as measured by surface plasmon resonance at 25°C or 37°C, for example, using the assay format defined in Example 2 herein or a substantially similar assay. In some embodiments, the antibodies or antigen-binding fragments of the present disclosure bind to 4-1BB with a t½ greater than about 0.5 minutes, greater than about 1 minute, greater than about 3 minutes, greater than about 5 minutes, greater than about 10 minutes, greater than about 15 minutes, greater than about 20 minutes, greater than about 30 minutes, greater than about 40 minutes, or greater than about 50 minutes, as measured by surface plasmon resonance at 25°C or 37°C, for example, using the assay format defined in Example 2 herein or a substantially similar assay.

[0190] According to certain implementations, this disclosure includes K with a value less than about 1 nM. D Antibodies binding to human PD-L1 (e.g., at 25°C) and antigen-binding fragments of the antibodies are measured, for example, by surface plasmon resonance, using the assay format defined in Example 2 herein. In some embodiments, the antibodies or antigen-binding fragments of this disclosure are expressed in Kc values ​​less than about 1 nM, less than about 0.9 nM, less than about 0.8 nM, less than about 0.6 nM, less than about 0.4 nM, less than about 0.3 nM, less than 0.2 nM, less than 0.1 nM, or less than 0.05 nM. D Combined with PD-L1, such as that measured by surface plasmon resonance, for example, using an assay format defined as in Example 2 herein or a substantially similar assay. In some embodiments, the antibody or antigen-binding fragment of this disclosure is expressed at a Kc of about 0.05 nM to about 0.2 nM. D Combined with PD-L1.

[0191] This disclosure also includes antibodies and antigen-binding fragments of the present disclosure that bind to PD-L1 with a dissociation half-life (t½) greater than about 30 minutes, as measured by surface plasmon resonance at 25°C or 37°C, for example, using the assay format defined in the examples herein or a substantially similar assay. In some embodiments, the antibodies or antigen-binding fragments of the present disclosure bind to PD-L1 with a t½ greater than about 30 minutes, greater than about 60 minutes, greater than about 120 minutes, greater than about 150 minutes, greater than about 180 minutes, or greater than about 200 minutes, as measured by surface plasmon resonance at 25°C or 37°C, for example, using the assay format defined in Example 2 herein or a substantially similar assay.

[0192] This disclosure includes bispecific antigen-binding molecules (e.g., bispecific antibodies) capable of binding simultaneously to human 4-1BB and human PD-L1. According to certain embodiments, the bispecific antigen-binding molecules of this disclosure specifically interact with cells expressing 4-1BB and / or PD-L1. The extent to which the bispecific antigen-binding molecules bind to cells expressing 4-1BB and / or PD-L1 can be assessed by fluorescence-activated cell sorting (FACS), as illustrated in Example 3 herein. For example, this disclosure includes bispecific antigen-binding molecules that specifically bind to human cell lines expressing 4-1BB but not PD-L1 (e.g., Jurkat cells genetically modified to express 4-1BB). In some embodiments, the bispecific antigen-binding molecules are expressed at a concentration of less than 1 × 10⁻⁶. -5 M's EC 50 The value binds to human or cynomolgus monkey T cells expressing 4-1BB. In some embodiments, the bispecific antigen-binding molecule is expressed at a concentration of 1 × 10⁻⁶. -12 M to 1×10 -5 M's EC 50 The value binds to human or cynomolgus monkey T cells expressing 4-1BB. In some embodiments, the bispecific antigen-binding molecule is expressed at a concentration of 1 × 10⁻⁶. -12 M to 1×10 -9 M's EC 50 The value binds to human or cynomolgus monkey T cells expressing 4-1BB. In some embodiments, the bispecific antigen-binding molecule is expressed at a concentration of 1 × 10⁻⁶. -12 M to 1×10 -10 M's EC 50 The value binds to human or cynomolgus monkey T cells expressing 4-1BB. In some embodiments, the bispecific antigen-binding molecule is expressed at a value less than approximately 4 × 10⁻⁶. -9 M's EC 50It binds to the surface of cell lines expressing PD-L1. The binding of the bispecific antigen-binding molecule to the surface of cells or cell lines can be measured by in vitro FACS binding assay, as described in Example 3.

[0193] This disclosure includes a PD-L1x4-1BB bispecific antigen-binding molecule capable of depleting tumor cells in a subject. For example, according to certain embodiments, a PD-L1x4-1BB bispecific antigen-binding molecule is provided, wherein a single administration of the antigen-binding molecule to a subject at a therapeutically effective dose results in a reduction in the number of tumor cells in the subject.

[0194] This disclosure includes an anti-PD-L1 x anti-CD28 bispecific antigen-binding molecule capable of activating T cells by binding to PD-L1 on target cells and 4-1BB on T cells (see Example 3). For example, binding of the anti-PD-L1 x anti-4-1BB bispecific antigen-binding molecule to T cells can lead to increased IL-2 release (Example 4). Therefore, the bispecific antigen-binding molecule of this disclosure may be useful in promoting T cell-mediated immune responses.

[0195] This disclosure includes an anti-PD-L1X anti-4-1BB bispecific antigen-binding molecule capable of blocking the interaction between PD-L1 and PD-1 (see Example 5). Therefore, the bispecific antigen-binding molecule of this disclosure can be used to inhibit the immune checkpoint pathway and reduce T cell exhaustion, thereby promoting T cell-mediated immune responses.

[0196] This disclosure includes an anti-PD-L1 X anti-4-1BB bispecific antigen-binding molecule capable of binding to PD-L1 expressed on cell surfaces. A variety of tumor cells express PD-L1, including breast cancer cells (e.g., HeLa, MCF-7, and MDA-MB-231), melanoma cells (e.g., A375), lung cancer cells (e.g., HCC44), ovarian cancer cells (e.g., ES-2, SNU-8, MCAS), pancreatic cancer cells (e.g., SNU-324), and prostate cancer cells (e.g., DU145). Therefore, the bispecific antibody of this disclosure may prove useful in treating a variety of cancer indications.

[0197] This disclosure includes an anti-PD-L1 x anti-4-1BB bispecific antigen-binding molecule that can enhance the cytotoxic efficacy of anti-tumor-associated antigen (TAA) x anti-CD3 bispecific antibodies against a variety of cell lines (see Example 6). In some embodiments, TAA is selected from AFP, ALK, BAGE protein, BCMA, BIRC5 (survival protein), BIRC7, β-linkin, brc-abl, BRCA1, BORIS, CA9, carbonic anhydrase IX, caspase-8, CALR, CCR5, CD19, CD20 (MS4A1), CD22, CD30, CD40, CDK4, CEA, CTLA4, cyclin-B1, CYP1B1, EGFR, EGFRvIII, ErbB2 / Her2, ErbB3, ErbB4, ETV6-AML, EpCAM, EphA2, Fra-1, FOLR1, GAGE ​​protein (e.g., GAGE-1, -2), GD2, GD3, GloboH, phosphatidylinositol proteoglycan-3, GM3, gp100, Her2, HLA / B-raf, HLA / k-ras, HLA / MAG E-A3, hTERT, LMP2, MAGE proteins (e.g., MAGE-1, -2, -3, -4, -6, and -12), MART-1, mesothelin, ML-IAP, Muc1, Muc2, Muc3, Muc4, Muc5, Muc16 (CA-125), MUM1, NA17, NY-BR1, NY-BR62, NY-BR85, NY-ESO1, OX40, p15, p53, PAP, PAX3, PAX5, PCTA-1, PLAC1, PRLR, PRAME, PSMA (FOLH1), RAGE protein, Ras, RGS5, Rho, SART-1, SART-3, STEAP1, STEAP2, TAG-72, TGF-β, TMPRSS2, Thompson-nouvelle antigen (Tn), TRP-1, TRP-2, tyrosinase, and urinary plaque protein-3. The anti-PD-L1 X anti-4-1BB antibody can also be useful when combined with checkpoint inhibitors (e.g., antibodies against PD-1) or any other checkpoint inhibitor.

[0198] Epitope Mapping and Related Technologies

[0199] The epitopes on 4-1BB or PD-L1 that the antigen-binding molecules of this disclosure bind to may consist of a single continuous sequence of three or more amino acids (e.g., 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or more) of the 4-1BB or PD-L1 protein. Alternatively, the epitope may consist of multiple non-continuous amino acids (or amino acid sequences) of 4-1BB or PD-L1. The antibodies of this disclosure may interact with amino acids contained in a 4-1BB monomer or with amino acids on two different 4-1BB chains of a 4-1BB dimer. As used herein, the term "epitope" refers to an antigenic determinant that interacts with a specific antigen-binding site in the variable region of an antibody molecule, called a complementary site. A single antigen may have more than one epitope. Therefore, different antibodies may bind to different regions on an antigen and may 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 epitopes generated from adjacent amino acid residues in the polypeptide chain. In some cases, epitopes may contain portions of sugars, phosphoryl groups, or sulfonyl groups from the antigen.

[0200] Various techniques known to those skilled in the art can be used to determine whether an antibody's antigen-binding domain interacts with "one or more amino acids" in a peptide or protein. Exemplary techniques that can be used to determine epitopes or binding domains of a particular antibody or antigen-binding domain include, for example, conventional cross-blocking assays, such as those performed in... AntibodiesAs described in Harlow and Lane (Cold Spring Harbor Press, Cold Spring Harb., NY), methods include assays, site mutagenesis (e.g., alanine scan mutagenesis, arginine scan mutagenesis, etc.), Western blot analysis (Reineke, 2004, Methods Mol Biol 248:443-463), protease protection, and peptide cleavage analysis. Alternatively, methods such as epitope cleavage, epitope extraction, and antigen chemical modification can be used (Tomer, 2000, Protein Science 9:487-496). Another method for identifying amino acids in peptides to which antibodies interact is hydrogen / deuterium exchange detected by mass spectrometry. Generally, hydrogen / deuterium exchange methods involve deuterating the target protein, followed by binding the antibody to the deuterated protein. The protein / antibody complex is then transferred to water to allow hydrogen-deuterium exchange to occur at all residues except those protected by the antibody (which retains the deuterium label). After antibody dissociation, the target protein is subjected to protease cleavage and mass spectrometry analysis to reveal 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. Alternatively, in some embodiments, the target protein binds to the antibody, followed by hydrogen-deuterium exchange. After antibody dissociation, the target protein is subjected to protease cleavage and mass spectrometry analysis to reveal non-deuterium-labeled residues corresponding to the specific amino acids with which the antibody interacts. X-ray crystal structure analysis can also be used to identify amino acids in the peptide with which the antibody interacts.

[0201] This disclosure also includes anti-4-1BB and anti-PD-L1 antibodies that bind to the same epitopes as any particular exemplary antibody described herein (e.g., antibodies containing any amino acid sequences shown in Tables 1, 3, 6, and 9). Similarly, this disclosure also includes anti-4-1BB and / or anti-PD-L1 antibodies that compete with any particular exemplary antibody described herein (e.g., antibodies containing any amino acid sequences shown in Tables 1, 3, 6, and 9) for binding to 4-1BB and / or PD-L1.

[0202] This disclosure also includes bispecific antigen-binding molecules comprising a first antigen-binding domain that specifically binds to human 4-1BB and a second antigen-binding fragment that specifically binds to human PD-L1, wherein the first antigen-binding domain binds to the same epitope on 4-1BB as any particular exemplary 4-1BB-specific antigen-binding domain described herein, and / or wherein the second antigen-binding domain binds to the same epitope on PD-L1 as any particular exemplary PD-L1-specific antigen-binding domain described herein.

[0203] Similarly, this disclosure also includes bispecific antigen-binding molecules comprising a first antigen-binding domain that specifically binds to human 4-1BB and a second antigen-binding fragment that specifically binds to human PD-L1, wherein the first antigen-binding domain competes with any particular exemplary 4-1BB-specific antigen-binding domain described herein for binding to 4-1BB, and / or wherein the second antigen-binding domain competes with any particular exemplary PD-L1-specific antigen-binding domain described herein for binding to PD-L1.

[0204] By using conventional methods known in the art, it is readily possible to determine 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 this disclosure, or whether it competes with the reference antigen-binding molecule of this disclosure for binding. For example, to determine whether a test antibody binds to the same epitope on 4-1BB (or PD-L1) as a reference bispecific antigen-binding molecule of this disclosure, the reference bispecific molecule is first allowed to bind to the 4-1BB protein (or PD-L1 protein). Next, the ability of the test antibody to bind to the 4-1BB (or PD-L1) molecule is evaluated. After saturation binding with the reference bispecific antigen-binding molecule, if the test antibody is able to bind to 4-1BB (or PD-L1), it can be concluded that the test antibody does not compete with the reference bispecific antigen-binding molecule for binding to 4-1BB (or PD-L1), and / or there is steric interference between antibodies at different sites on the binding antigen. On the other hand, after saturation binding with the reference bispecific antigen-binding molecule, if the test antibody fails to bind to the 4-1BB (or PD-L1) molecule, the test antibody competes with the reference bispecific antigen-binding molecule of this disclosure for binding to 4-1BB (or PD-L1). Further routine experiments (e.g., peptide mutation and binding assays) can then be performed to determine whether the observed lack of test antibody binding is actually due to binding to the same epitope as the reference bispecific antigen-binding molecule, or whether steric hindrance (or other phenomena) is the cause of the observed lack of binding. Such experiments 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 this disclosure, if, for example, an excess of 1, 5, 10, 20, or 100 times, one antigen-binding protein inhibits the binding of another antigen-binding protein by at least 50%, but preferably 75%, 90%, or even 99% (as measured in a competitive binding assay), then the two antigen-binding proteins compete for binding to the antigen (see, for example, Junghans et al., Cancer Res. 1990:50:1495-1502). Alternatively, if substantially all amino acid mutations in the antigen that reduce or eliminate the binding of one antigen-binding protein also reduce or eliminate the binding of the other antigen-binding protein, then the two antigen-binding proteins may bind to the same epitope. If only a subset of amino acid mutations that reduce or eliminate the binding of one antigen-binding protein reduces or eliminates the binding of the other antigen-binding protein, then the two antigen-binding proteins may have “overlapping epitopes.”

[0205] To determine whether an antibody or its antigen-binding domain competes with a reference antigen-binding molecule for binding, the binding method described above is performed in two orientations: In the first orientation, the reference antigen-binding molecule is allowed to bind to the 4-1BB protein (or PD-L1 protein) under saturation conditions, and then the binding of the test antibody to the 4-1BB (or PD-L1) molecule is evaluated. In the second orientation, the test antibody is allowed to bind to the 4-1BB (or PD-L1) molecule under saturation conditions, and then the binding of the reference antigen-binding molecule to the 4-1BB (or PD-L1) molecule is evaluated. If, in both orientations, only the first (saturated) antigen-binding molecule is able to bind to the 4-1BB (or PD-L1) molecule, the following conclusion is drawn: the test antibody and the reference antigen-binding molecule compete for binding to 4-1BB (or PD-L1). As will be understood by those skilled in the art, the antibody competing for binding with the reference antigen-binding molecule may not necessarily bind to the same epitope as the reference antibody, but may spatially block the binding of the reference antibody by binding to overlapping or adjacent epitopes.

[0206] Preparation of antigen-binding domains and construction of bispecific molecules

[0207] Antigen-binding domains specific to a particular antigen can be prepared using any antibody production technique known in the art. Once obtained, two different antigen-binding domains specific to two different antigens (e.g., 4-1BB and PD-L1) can be appropriately arranged relative to each other using conventional methods to produce the bispecific antigen-binding molecule of this disclosure. (Exemplary forms of bispecific antibodies that can be used to construct the bispecific antigen-binding molecule of this disclosure are discussed elsewhere herein.) In some embodiments, individual components (e.g., heavy and light chains) of one or more multispecific antigen-binding molecules of this disclosure are derived from chimeric antibodies, humanized antibodies, or fully human antibodies. Methods for preparing such antibodies are known in the art. For example, VELOCIMMUNE can be used. TM The technique described herein involves preparing one or more of the heavy chain and / or light chain of the bispecific antigen-binding molecule. Using VELOCIMMUNE... TM The technique (or any other human antibody production technique) initially isolates a high-affinity chimeric antibody against a specific antigen (e.g., 4-1BB or PD-L1) having a human variable region and a mouse constant region. The antibody is characterized and selected for desired characteristics, including affinity, selectivity, epitopes, etc. The mouse constant region is replaced with the desired human constant region to produce a fully human heavy chain and / or light chain in a bispecific antigen-binding molecule that can be incorporated into this disclosure.

[0208] Genetically modified animals can be used to prepare human bispecific antigen-binding molecules. For example, genetically modified mice that cannot rearrange and express endogenous mouse immunoglobulin light chain variable sequences can be used, wherein the mice express only one or two human light chain variable domains encoded by human immunoglobulin sequences operatively linked to a mouse κ constant gene at an endogenous mouse κ locus. Such genetically modified mice can be used to generate fully human bispecific antigen-binding molecules comprising two distinct heavy chains associated with the same light chain containing a variable domain derived from one of the two distinct human light chain variable region gene segments. (See, for example, US 2011 / 0195454, which discusses in detail such modified mice and their use in generating bispecific antigen-binding molecules).

[0209] bioequivalent

[0210] This disclosure covers antigen-binding molecules having an amino acid sequence different from that of the antibodies described herein but retaining the ability to bind 4-1BB and PD-L1. When compared to the parental sequence, such variant molecules contain the addition, deletion, or substitution of one or more amino acids but exhibit substantially equivalent biological activity to the antigen-binding molecule described herein. Similarly, the DNA sequences encoding antigen-binding molecules in this disclosure cover sequences that contain one or more nucleotide additions, deletions, or substitutions when compared to the disclosed sequences but encode antigen-binding molecules substantially bioequivalent to those of the antigen-binding molecules described herein. Some examples of such variant amino acid and DNA sequences are discussed above.

[0211] This disclosure includes antigen-binding molecules that are bioequivalent to any of the exemplary antigen-binding molecules described herein. If, for example, two antigen-binding proteins or antibodies are administered at the same molar dose under similar experimental conditions, and their absorption rates and extents do not show significant differences, they are considered bioequivalent. If some antibodies are equivalent in extent of absorption but not in rate of absorption, said antibodies will be considered equivalents or drug substitutes and may still be considered bioequivalent because such differences in absorption rates are intentional and reflected in labeling, are not necessary to achieve effective in vivo drug concentrations (e.g., for long-term use), and are considered medically insignificant for the specific pharmaceutical product under investigation.

[0212] In one implementation, two antigen-binding proteins are bioequivalent if there are no clinically significant differences in their safety, purity, and potency.

[0213] In one implementation, the two antigen-binding proteins are bioequivalent if a patient can switch between the reference product and the biological product once or more, and there is no increased risk of anticipated adverse effects (including clinically significant alterations in immunogenicity or reduced efficacy) compared to continuous treatment without such switching.

[0214] In one implementation, if two antigen-binding proteins function for one or more conditions through a common mechanism of action, to the extent that such mechanism is known.

[0215] Bioequivalence can be demonstrated through in vivo and in vitro methods. Bioequivalence measurements include, for example, (a) in vivo testing in humans or other mammals, wherein the concentration of the antibody or its metabolites as a function of time is measured in blood, plasma, serum, or other biological fluids; (b) in vitro testing that correlates with and reasonably predicts bioavailability data in humans; (c) in vivo testing in humans or other mammals, wherein appropriate acute pharmacological effects of the antibody (or its target) as a function of time are measured; and (d) in clinical trials in which well-controlled trials have been established to demonstrate the safety, potency, or bioavailability or bioequivalence of the antibody.

[0216] Bioequivalent variants of the exemplary bispecific antigen-binding molecules shown herein can be constructed, for example, by making multiple substitutions to 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 cases, bioequivalent antibodies may comprise exemplary bispecific antigen-binding molecules shown herein that include amino acid changes that alter the glycosylation characteristics of the antibody, such as by eliminating or removing glycosylation mutations.

[0217] Species selectivity and species cross-reactivity

[0218] According to certain embodiments, this disclosure provides antigen-binding molecules that bind to human 4-1BB but not to 4-1BB from other species. This disclosure also provides antigen-binding molecules that bind to human PD-L1 but not to PD-L1 from other species. This disclosure further includes antigen-binding molecules that bind to both human 4-1BB and 4-1BB from one or more non-human species; and / or antigen-binding molecules that bind to both human PD-L1 and PD-L1 from one or more non-human species.

[0219] According to certain exemplary embodiments of this disclosure, an antigen-binding molecule is provided that binds to human 4-1BB and / or human PD-L1, and may or may not bind to (as the case may) one or more of mouse, rat, guinea pig, hamster, gerbil, pig, cat, dog, rabbit, goat, sheep, cattle, horse, camel, cynomolgus monkey, marmoset, rhesus monkey, or chimpanzee 4-1BB and / or PD-L1. For example, in one specific exemplary embodiment of this disclosure, a bispecific antigen-binding molecule is provided that comprises a first antigen-binding domain binding to human 4-1BB and cynomolgus monkey 4-1BB and a second antigen-binding domain specifically binding to human PD-L1.

[0220] Immunoconjugates

[0221] This disclosure covers PD-L1x4-1BB antigen-binding proteins, such as antibodies or antigen-binding fragments, such as REGN6191 (“immunoconjugates”), that are conjugated to another portion (e.g., a therapeutic portion). In one embodiment of this disclosure, a PD-L1x4-1BB antigen-binding protein (e.g., an antibody or antigen-binding fragment) is conjugated to any other therapeutic agent described herein. As used herein, the term “immunoconjugate” refers to an antigen-binding protein, such as an antibody or antigen-binding fragment, that is chemically or biologically linked to another antigen-binding protein, drug, radioactive agent, reporter portion, enzyme, peptide, protein, or therapeutic agent.

[0222] In some embodiments, the therapeutic component may be a cytotoxic agent, a chemotherapeutic agent, an immunosuppressant, or a radioactive isotope. Cytotoxic agents include any agent that is harmful to cells. Examples of suitable cytotoxic agents and chemotherapeutic agents for forming immune conjugates are known in the art (see, for example, WO 05 / 103081).

[0223] Therapeutic uses of antigen-binding molecules

[0224] The bispecific antibodies and antigen-binding molecules (and therapeutic compositions comprising them) of this disclosure are particularly useful for treating any disease or condition in which stimulation, activation, and / or targeting of an immune response would be beneficial. Specifically, the PD-L1x4-1BB bispecific antigen-binding molecule of this disclosure can be used to treat, prevent, and / or improve hyperproliferative diseases such as cancer. In some embodiments, this disclosure provides a method for treating cancer in a subject comprising administering a therapeutically effective dose of a PD-L1x4-1BB antigen-binding molecule, such as REGN6191.

[0225] For the purposes of this document, hyperproliferative disorders are diseases characterized by abnormal, excessive, and / or uncontrolled cell growth, such as cells expressing PD-L1. Hyperproliferative disorders include, for example, cancers. Exemplary cancers include, but are not limited to, esophageal cancer, squamous cell carcinoma of the lung, lung adenocarcinoma, squamous cell carcinoma of the cervix, glioma, thyroid cancer, lung cancer (e.g., non-small cell lung cancer), colorectal cancer, colon cancer, bladder cancer, rectal cancer, head and neck cancer, gastric cancer, liver cancer, pancreatic cancer, kidney cancer, urothelial carcinoma, prostate cancer, testicular cancer, breast cancer, cervical cancer, endometrial cancer, ovarian cancer, gastroesophageal cancer (e.g., gastroesophageal adenocarcinoma), and melanoma. Therefore, the antibodies and bispecific antigen-binding molecules of this disclosure can be used to treat a wide range of cancers.

[0226] Cancers characterized by solid tumor cells or cancerous blood cells can be PD-L1-expressing cancers, for example, in which PD-L1 expression has been identified in the cells of a specific subject to be treated, including esophageal cancer, lung squamous cell carcinoma, lung adenocarcinoma, cervical squamous cell carcinoma, endometrial adenocarcinoma, bladder urothelial carcinoma, lung cancer (e.g., non-small cell lung cancer), colorectal cancer, rectal cancer, endometrial cancer, skin cancer (e.g., head and neck squamous cell carcinoma), brain cancer (e.g., glioblastoma multiforme), breast cancer, gastroesophageal cancer (e.g., gastroesophageal adenocarcinoma), prostate cancer, and / or ovarian cancer.

[0227] The antigen-binding molecules disclosed herein may also be used to treat primary and / or metastatic tumors (or cancers discussed herein) arising in, for example, the colon, lung, breast, ovary, kidney, and bladder.

[0228] The antigen-binding molecules disclosed herein can be used for residual cancer in subjects. As used herein, the term "residual cancer" means one or more types of cancer cells present or persistent in a subject after treatment with anticancer therapies.

[0229] As used herein, the term "subject" refers to a mammal (e.g., rat, mouse, cat, dog, cow, sheep, horse, goat, rabbit), preferably a human, for example, who needs to prevent and / or treat cancer. The subject may have cancer, may be susceptible to such a condition, and / or will benefit from the administration of the bispecific antibody or its antigen-binding fragment of the present disclosure. In one embodiment, the subject may have a hyperproliferative disease or be at risk of developing a hyperproliferative disease.

[0230] The method for treating or preventing cancer (e.g., PD-L1-expressing cancer) in a subject requiring treatment or prevention by administering a therapeutically effective dose of the PD-L1x4-1BB antigen-binding protein associated with an additional therapeutic agent is part of this disclosure. Other therapeutic agents are disclosed elsewhere herein.

[0231] An “effective” or “therapeuticly effective” dose of PD-L1x4-1BB antigen-binding protein (e.g., an antibody or antigen-binding fragment) for the treatment or prevention of hyperproliferative diseases (e.g., cancers expressing PD-L1) is an amount of antigen-binding protein sufficient to alleviate one or more signs and / or symptoms of the disease in the treated subject, whether by inducing the regression or elimination of such signs and / or symptoms or by inhibiting the progression of such signs and / or symptoms. In one embodiment of this disclosure, the therapeutically effective dose of PD-L1x4-1BB antigen-binding protein is from 0.1 to 2000 mg. The amount of dose may vary depending on the age and size of the subject to be administered, the target disease, the condition, the route of administration, etc. In some embodiments, a second or more subsequent doses of antigen-binding protein may be administered after the initial dose in an amount substantially the same as, less than, or more than the initial dose, wherein the subsequent doses may be spaced 1 to 8 weeks apart.

[0232] The dosage of antigen-binding molecules administered to a patient can vary depending on the patient's age and body size, target disease, symptoms, route of administration, etc. The preferred dosage is usually calculated based on body weight or body surface area. The frequency and duration of treatment can be adjusted according to the severity of the condition. Effective dosages and timing of bispecific antigen-binding molecules can be determined empirically; for example, patient progression can be monitored through regular assessments, and the dosage adjusted accordingly.

[0233] Combination therapy

[0234] The bispecific antigen-binding molecule of this disclosure can be used in combination with one or more pharmaceutical agents, for example, for treating cancer in a subject. In some embodiments, the bispecific antigen-binding molecule can be administered in combination with one or more pharmaceutical agents (e.g., corticosteroids) to reduce or improve one or more adverse side effects, such as cytokine storms. In some embodiments, the bispecific antigen-binding molecule can be administered in combination with one or more therapeutic agents or treatments to enhance the efficacy of cancer treatment. Exemplary additional therapeutic agents or treatments that can be administered in combination with or in combination with the antigen-binding molecules of this disclosure include, for example, chemotherapy (e.g., anticancer chemotherapy, such as paclitaxel, docetaxel, vincristine, cisplatin, carboplatin, or oxaliplatin), radiation therapy, surgery, checkpoint inhibitors, PD-1 inhibitors (e.g., anti-PD-1 antibodies, such as pembrolizumab, nivolumab, or cimipril), CTLA-4 inhibitors, LAG3 inhibitors, TIM3 inhibitors, GITR agonists, OX40 agonists, and 4-1B. B agonists, oncolytic viruses, cancer vaccines, CAR-T cells, nucleic acid therapeutics, stem cell transplantation, modified IL2, modified IL12, IL15, IL6 inhibitors (e.g., sarilumab or tocilizumab), IL4R inhibitors (e.g., dupilumab), EGFR inhibitors, Ang2 inhibitors, VEGF inhibitors, corticosteroids, bispecific antibodies binding to CD3 and tumor-associated antigens (TAAs) (e.g., MUC16, PSMA, STEAP2, or any TAA disclosed herein), or antigen-binding fragments thereof. Exemplary bispecific antibodies containing an antigen-binding domain that binds to CD3 include, but are not limited to, those described in, for example, WO2017 / 053856A1, WO2014 / 047231A1, WO2018 / 067331A1, and WO2018 / 058001A1. PD-L1 is expressed in a wide range of cancers. Therefore, the bispecific anti-PD-L1xCD28 antibody of this disclosure can be combined with a wide range of bispecific antibodies containing an antigen-binding domain that binds to CD3 for the treatment of various cancers.

[0235] Other agents may be administered shortly before, simultaneously with, or shortly after the administration of the antigen-binding molecule of this disclosure (for the purposes of this disclosure, such administration regimens are considered to be administration of the antigen-binding molecule in combination with other agents or therapeutics or treatments).

[0236] Pharmaceutical preparation and administration

[0237] This disclosure provides compositions comprising PD-L1x4-1BB antigen-binding protein and one or more components; and methods of using such compositions and methods of preparing such compositions. Pharmaceutical formulations (e.g., aqueous pharmaceutical formulations comprising water) of the PD-L1x4-1BB antigen-binding protein and pharmaceutically acceptable carriers or excipients contained in this disclosure are part of this disclosure.

[0238] The pharmaceutical compositions disclosed herein can be formulated with suitable carriers, excipients, and other agents that provide improved transfer, delivery, tolerability, etc. A variety of suitable formulations are available in all formulation sets known to medicinal chemists: Remington's Pharmaceutical Sciences, Mack Publishing Company, Easton, PA. These formulations include, for example, powders, pastes, ointments, gels, waxes, oils, lipids, and lipid-containing vesicles (e.g., LlPOFECTIN). TM Life Technologies, Carlsbad, CA), DNA conjugates, anhydrous absorbent pastes, oil-in-water and water-in-oil emulsions, carbowax emulsions (polyethylene glycol in various molecular weights), semi-solid gels, and semi-solid mixtures containing carbowax. See also Powell et al., "Compendium of excipients for parenteral formulations" PDA (1998) J Pharm Sci Technol 52:238-311.

[0239] To prepare pharmaceutical formulations of PD-L1x4-1BB antigen-binding proteins, such as antibodies and their antigen-binding fragments (e.g., REGN6191), the antigen-binding protein is mixed with a pharmaceutically acceptable carrier or excipient. See, e.g., Remington's Pharmaceutical Sciences and US Pharmacopeia: National Formulary, Mack Publishing Company, Easton, Pa. (1984); Hardman et al., (2001) Goodman and Gilman's The Pharmacological Basis of Therapeutics, McGraw-Hill, NY; Gennaro (2000) Remington: The Science and Practice of Pharmacy, Lippincott, Williams, and Wilkins, NY; Avis et al., (eds.) (1993) Pharmaceutical Dosage Forms: Parenteral Medications, Marcel Dekker, NY; Lieberman et al., (eds.) (1990) Pharmaceutical Dosage Forms: Tablets, Marcel Dekker, NY; Lieberman et al., (eds.) (1990) Pharmaceutical Dosage Forms: Disperse Systems, Marcel Dekker, NY; Weiner and Kotkoskie (2000) Excipient Toxicity and Safety, Marcel Dekker, NY. In one embodiment of this disclosure, the pharmaceutical formulation is sterile. Such compositions are part of this disclosure.

[0240] The pharmaceutical formulations disclosed herein contain PD-L1x4-1BB antigen-binding protein and pharmaceutically acceptable carriers, including, for example, water, buffers, preservatives, and / or detergents.

[0241] The scope of this disclosure includes dried, such as lyophilized, compositions containing PD-L1x4-1BB antigen-binding protein (e.g., antibody or antigen-binding fragment thereof), or pharmaceutical formulations containing a pharmaceutically acceptable carrier but substantially lacking water.

[0242] Various delivery systems are known and can be used to administer the pharmaceutical compositions disclosed herein, such as encapsulation in liposomes, microparticles, microcapsules, recombinant cells capable of expressing mutant viruses, and receptor-mediated endocytosis (see, for example, Wu et al., 1987, J. BioI. Chem. 262:4429-4432). Delivery methods include, but are not limited to, intradermal, intramuscular, intraperitoneal, intravenous, subcutaneous, intranasal, rectal, intestinal, epidural, and oral routes. The composition can be administered via any convenient route, such as by infusion or bolus injection, by absorption through an epithelial or mucocutaneous lining (e.g., oral mucosa, rectal and intestinal mucosa), and can be administered with other bioactive agents. Administration can be systemic or local.

[0243] As discussed herein, this disclosure provides containers (e.g., plastic or glass vials) or injection devices (e.g., syringes, pre-filled syringes, or autoinjectors) that contain any of the PD-L1x4-1BB antigen-binding proteins described herein, such as antibodies or antigen-binding fragments thereof, or pharmaceutical formulations that contain pharmaceutically acceptable carriers or excipients.

[0244] The pharmaceutical compositions of this disclosure can be delivered subcutaneously or intravenously using standard needles and syringes. Alternatively, regarding subcutaneous delivery, pen-type delivery devices known in the art can be used to deliver the pharmaceutical compositions of this disclosure. Such pen-type delivery devices can be reusable or disposable.

[0245] Many reusable and disposable pen-type and auto-injector-type delivery devices are used for subcutaneous delivery of the pharmaceutical compositions disclosed herein. See, for example, AUTOPEN TM (Owen Mumford, Inc., Woodstock, UK) or HUMIRA TM Pen (Abbott Labs, Abbott Park, IL).

[0246] In some cases, the pharmaceutical composition can be delivered using a controlled-release system. In one embodiment, a pump can be used (see Langer, ibid.; Sefton, 1987, CRC Crit. Ref. Biomed. Eng. 14:201). In another embodiment, polymeric materials can be used; see Medical Applications of Controlled Release, Langer and Wise (ed.), 1974, CRC Pres., Boca Raton, Florida. In yet another embodiment, the controlled-release system can be placed near the target of the composition, thus requiring only a portion of the systemic dose (see, for example, Goodson, 1984, in Medical Applications of Controlled Release, ibid., Vol. 2, pp. 115–138). Other controlled-release systems are discussed in a review in Langer, 1990, Science 249:1527–1533.

[0247] Injectable formulations may include dosage forms for intravenous, subcutaneous, intradermal, and intramuscular injection, infusion, etc. These injectable formulations can be prepared by known methods. For example, they can be prepared, for instance, by dissolving, suspending, or emulsifying the aforementioned antibodies or their salts in a sterile aqueous or oily medium conventionally used for injection. Aqueous media for injection include, for example, physiological saline and other isotonic solutions that can be used in combination with suitable solubilizers. Injectable oily media are also part of this disclosure. Such oily media can be combined with solubilizers.

[0248] Advantageously, the above-described pharmaceutical compositions for oral or parenteral use are prepared into dosage forms suitable for matching the dosage of the active ingredient. Such dosage forms include, for example, tablets, pills, capsules, injections (ampoules), suppositories, etc. The amount of the antibody contained is typically from about 0.1 to about 2000 mg per unit dose dosage form; particularly in the form of injections.

[0249] Diagnostic uses

[0250] The bispecific antibodies of this disclosure can also be used to detect and / or measure 4-1BB or PD-L1 in samples, or cells expressing 4-1BB or PD-L1, for example, for diagnostic purposes. For example, a PD-L1x4-1BB antibody or its antigen-binding fragment can be used to diagnose a condition or disease characterized by abnormal expression of 4-1BB or PD-L1 (e.g., overexpression, low expression, lack of expression, etc.). Exemplary diagnostic assays for 4-1BB or PD-L1 may include, for example, contacting a sample obtained from a patient with an antibody of this disclosure, wherein the antibody is labeled with a detectable marker or reporter molecule. Alternatively, an unlabeled antibody may be combined with a secondary antibody that is itself a detectable marker for diagnostic applications. The detectable marker or reporter molecule may be a radioisotope, for example... 3 H, 14 C 32 p、 35 S or 125 1; a fluorescent or chemiluminescent component, 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 4-1BB or PD-L1 in a sample include enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), and fluorescence-activated cell sorting (FACS). Samples that can be used for 4-1BB or PD-L1 diagnostic assays according to this disclosure include any tissue or fluid sample obtainable from a patient that contains detectable amounts of 4-1BB or PD-L1 protein or fragments thereof under normal or pathological conditions. Typically, the level of 4-1BB or PD-L1 in a specific sample obtained from a healthy patient (e.g., a patient without a disease or condition associated with abnormal 4-1BB or PD-L1 levels or activity) will be measured to initially establish a baseline or standard level of 4-1BB or PD-L1. The baseline 4-1BB or PD-L1 level can then be compared with the 4-1BB or PD-L1 level measured in samples obtained from individuals suspected of having 4-1BB or PD-L1-related diseases or conditions.

[0251] Example

[0252] The following embodiments are provided to provide those skilled in the art with a complete disclosure and description of how to prepare and use the compositions and methods of this disclosure, and are not intended to limit the scope of what the inventors consider to be their invention.

[0253] Example 1: Construction of anti-PD-L1x4-1BB antibody

[0254] Production of anti-PD-L1 antibodies

[0255] Anti-PD-L1 antibodies were obtained by immunizing genetically modified mice containing DNA encoding the variable regions of the heavy and κ light chains of human immunoglobulins with human PD-L1 antigen (e.g., SEQ ID NO: 79).

[0256] Following immunization, spleen cells were harvested from each mouse and B cells were sorted using a human PD-L1 fragment as a sorting agent to bind to and identify reactive antibodies (antigen-positive B cells) (as described in US 2007 / 0280945).

[0257] Antibodies are characterized and selected for desired characteristics, including affinity and selectivity. Antibodies may have desired constant regions, such as wild-type or modified hIgG1 or hIgG4 constant regions. As those skilled in the art will understand, antibodies with a specific constant region (e.g., modified hIgG1) can be converted into antibodies with different constant regions (e.g., modified hIgG4). While constant regions may vary depending on the specific application, high-affinity antigen binding and target-specific characteristics are present in variable regions.

[0258] Table 1 shows the amino acid sequence identifiers of the variable regions and CDRs of the heavy and light chains of the selected anti-PD-L1 antibodies of this disclosure. The corresponding nucleic acid sequence identifiers are shown in Table 2.

[0259] Table 1: Amino acid sequence identifiers of the selected parental PD-L1 monoclonal antibodies

[0260]

[0261] Table 2: Nucleic acid sequence identifiers of the selected parental PD-L1 monoclonal antibodies

[0262]

[0263] Production of anti-4-1BB antibodies

[0264] Anti-4-1BB antibodies were obtained by immunizing VELOCIMMUNE® mice (i.e., modified mice containing DNA encoding the variable regions of the heavy and universal light chains of human immunoglobulins) with human 4-1BB protein fused with the Fc portion of mouse IgG2a, or with DNA encoding 4-1BB.

[0265] Antibody immune responses are monitored using a 4-1BB-specific immunoassay. When the desired immune response is achieved, anti-4-1BB antibodies are isolated directly from antigen-positive B cells, as described in US7,582,298.

[0266] Antibodies are characterized and selected for desired characteristics, including affinity and selectivity. Antibodies may have desired constant regions, such as wild-type or modified hIgG1 or hIgG4 constant regions. As those skilled in the art will understand, antibodies with a specific constant region (e.g., modified hIgG1) can be converted into antibodies with different constant regions (e.g., modified hIgG4). While constant regions may vary depending on the specific application, high-affinity antigen binding and target-specific characteristics are present in variable regions.

[0267] Table 3 shows the amino acid sequence identifiers of the variable regions and CDRs of the heavy and light chains of the selected anti-4-1BB antibodies of this disclosure. The corresponding nucleic acid sequence identifiers are shown in Table 4.

[0268] Table 3: Amino acid sequence identifiers of the selected parental 4-1BB monoclonal antibodies

[0269]

[0270] Table 4: Nucleic acid sequence identifiers of the selected parental 4-1BB antibodies

[0271]

[0272] Production of bispecific antibodies (bsAb) binding to 4-1BB and PD-L1

[0273] Bispecific antibodies containing both an anti-PD-L1 specific binding domain and an anti-4-1BB specific binding domain are constructed using standard methods. Each of the anti-PD-L1 and anti-4-1BB antigen-binding domains contains a distinct HCVR that pairs with a common LCVR. In some cases, bispecific antibodies are constructed using the heavy chain from the anti-4-1BB antibody, the heavy chain from the anti-PD-L1 antibody, and a common light chain. Table 5 summarizes the components of the selected bispecific PD-L1x4-1BB antibodies (parental antibody names). Tables 6 and 7 show the amino acid and nucleic acid identifiers of the selected bispecific antibodies, respectively. Table 8 shows the full-length heavy and light chain sequences of the selected bispecific antibodies. Additional bispecific antibodies binding to PD-L1 and 4-1BB can be prepared using parental monoclonal antibodies with the names shown in Table 9.

[0274] Table 5: Summary of the components of the selected anti-PD-L1 x anti-4-1BB bispecific antibodies

[0275]

[0276] Table 6: Amino acid sequence identifiers of the selected anti-PD-L1 x anti-4-1BB bispecific antibodies

[0277]

[0278] Table 7: Nucleic acid sequence identifiers of selected anti-PD-L1 x anti-4-1BB bispecific antibodies

[0279]

[0280] Table 8. Amino acid and nucleotide sequences of the full-length immunoglobulin chains of bispecific antibodies REGN6188, REGN6189, REGN6190, and REGN6191

[0281]

[0282] D = nucleotide sequence of DNA encoding the specified sequence

[0283] P = amino acids of the polypeptide with the specified sequence.

[0284] The number refers to the SEQ ID NO of the specified sequence.

[0285] HC is the full-length heavy chain of the specified antibody.

[0286] LC refers to the full-length light chain of a specified antibody.

[0287] Additional bispecific antibodies comprising one HCVR arm from a parental PD-L1 antibody and another HCVR arm from a parental 4-1BB antibody can be prepared using the techniques described herein. The parental PD-L1 antibody used to generate these additional anti-PD-L1 X anti-4-1BB bispecific antibodies has the HCVR sequence described in Table 1 above. The 4-1BB parental antibody used to generate these additional anti-PD-L1 X anti-4-1BB bispecific antibodies has the amino acid sequence described in Table 3 above. These anti-PD-L1 and anti-4-1BB binding domains (paired) are shown in Table 9 below.

[0288] Table 9: Summary of parental antibody names for the HCVR arm of other PD-L1 x anti-4-1BB bispecific antibodies

[0289] Anti-PD-L1 antigen-binding domain (parental antibody names and sequences are shown in Table 1) Anti-4-1BB antigen-binding domain (parental antibody names and sequences are shown in Table 3 of this article). mAb9364 mAb25894 mAb9364 mAb25898 mAb9364 mAb25907 mAb9364 mAb25921 mAb9373 mAb25894 mAb9373 mAb25898 mAb9373 mAb25907 mAb9373 mAb25921

[0290] The bispecific antibodies described in the following examples consist of antigen-binding arms that bind to human h4-1BB protein and human PD-L1 (see Biacore binding data below). Exemplary bispecific antibodies comprise a modified (chimeric) IgG4 Fc domain as shown in U.S. Patent No. 9,359,437.

[0291] The bispecific antibody generated according to this embodiment comprises two separate antigen-binding domains (i.e., binding arms). The first antigen-binding domain comprises a heavy chain variable region (“4-1BB-VH”) derived from the anti-4-1BB antibody, and the second antigen-binding domain comprises a heavy chain variable region (“PD-L1-VH”) derived from the anti-PD-L1 antibody. Both anti-PD-L1 and anti-4-1BB share a common light chain. The 4-1BB-VH / PD-L1-VH pairing produces antigen-binding domains that specifically recognize 4-1BB (e.g., on T cells) and PD-L1 (e.g., on tumor cells or antigen-presenting cells).

[0292] The control used in the following examples

[0293] For comparative purposes, three control constructs were included in the following experiments: Comparative object 1 According to US 2012 / 0237498 (Pfizer), V contains the antibody "PF-05082566". H / V L Human monoclonal antibody against 4-1BB sequence; Compare Object 2 According to US 2014 / 0193422 (BMS), V contains antibody "20H4.9". H / V L Human monoclonal antibody against 4-1BB sequence; and " Comparative object 3 According to WO 2019 / 025545 (GenMab), V contains antibodies “CD137-009-HC7LC2” and “PD-L1-547”. H / V L A bispecific PD-L1 x 4-1BB antibody with a specific sequence.

[0294] Example 2: Characterization of bispecific antibodies binding to 4-1BB and PD-L1 by surface plasmon resonance

[0295] PD-L1 kinetics: Using real-time surface plasmon resonance (SPR) biosensor technology and a Biacore S200 instrument, the equilibrium dissociation constant (K0) for the binding of human PD-L1 (hPD-L1.mmH, SEQ ID NO: 79) expressed with the C-terminal myc-myc-hexahistine tag to purified anti-PD-L1x4-1BB antibody was determined. DThe CM5 Biacore sensor surface was derivatized by amine conjugation with a monoclonal mouse anti-human Fc antibody. All Biacore binding studies were performed in a buffer (HBS-EP run buffer) consisting of 0.01 M HEPES pH 7.4, 0.15 M NaCl, 3 mM EDTA, and 0.05% v / v surfactant P20. Different concentrations of hPD-L1.mmH (0.37 nM to 30 nM in 3-fold serial dilutions), prepared in the HBS-EP run buffer, were injected onto the captured antibody at a flow rate of 50 µL / min. Antibody-reagent association was monitored for 5 min, while dissociation in the HBS-EP run buffer was monitored for 10 min. At the end of each cycle, the antibody capture surface was regenerated by a 12-second injection of 20 mM phosphate. All binding kinetics experiments were performed at 25 °C.

[0296] 4-1BB Kinetics: Using real-time surface plasmon resonance biosensor technology and a Biacore T200 instrument, the equilibrium dissociation constant (K0) for the binding of human 4-1BB (h4-1BB.mmH, SEQ ID NO: 80) expressed with the C-terminal myc-myc-hexahistine tag to purified anti-PD-L1x4-1BB antibody was determined. D The CM5 Biacore sensor surface was derivatized by amine conjugation with a monoclonal mouse anti-human Fc antibody. All Biacore binding studies were performed in a buffer (HBS-EP run buffer) consisting of 0.01 M HEPES pH 7.4, 0.15 M NaCl, 3 mM EDTA, and 0.05% v / v surfactant P20. Different concentrations of h4-1BB.mmH (30 nM to 270 nM in 3-fold serial dilutions) prepared in the HBS-EP run buffer were injected onto the captured antibody at a flow rate of 50 µL / min. Antibody-reagent association was monitored for 5 min, while dissociation in the HBS-EP run buffer was monitored for 10 min. At the end of each cycle, the antibody capture surface was regenerated by a 10-second injection of 20 mM phosphate. All binding kinetics experiments were performed at 25 °C.

[0297] Data Analysis: Specific SPR-Biacore sensor maps were obtained using a double-reference procedure. Double referencing was performed by subtracting the signal from the reference surface (anti-hFc or anti-mFc) for each injection from the signal on the experimental surface (anti-hFc-captured anti-PD-L1x4-1BB antibody) to remove the influence of refractive index variations. Additionally, run buffer injections were performed to allow subtraction of signal changes caused by the dissociation of the captured antibody or antigen from the conjugated anti-hFc or anti-mFc surface. Kinetic association rate constant (ka) and dissociation rate constant (k) were also analyzed. d The real-time sensor image was fitted to a 1:1 model using Scrubber v2.0c curve fitting software to determine the model. This was combined with the dissociation equilibrium constant (K). D The dissociation half-life (t½) and the dissociation half-life are calculated from the kinetic rate constant as follows:

[0298]

[0299] Biacore analysis showed that PD-L1x4-1BB had a Kc ratio of approximately 4.5E-11 to 1.2E-10. D Combined with hPD-L1, and with K at approximately 1.7E-07 to a.2E-08 D Combine with h4-1BB (Tables 10 and 11).

[0300] Table 10: Kinetic and equilibrium binding parameters of hPD-L1 with surface-captured anti-PD-L1x4-1BB antibody at 25 °C

[0301]

[0302] NB: No binding

[0303] Table 11: Kinetic and equilibrium binding parameters of h4-1BB with surface-captured anti-PD-L1x4-1BB antibody at 25℃

[0304]

[0305]

[0306] NB: No binding

[0307] Example 3: Characterization of the bispecific PD-L1x4-1BB antibody bound to cells

[0308] The binding of the hPD-L1x4-1BB bispecific antibody to cells was characterized using flow cytometry. To assess the specificity of antibody binding to PD-L1 and 41BB, binding was evaluated using OVCAR3 (low endogenous PD-L1 expression), OVCAR3 / hPD-L1 (modified to overexpress human PD-L1), and Jurkat / NFkB-Luc / h41BB (low PD-L1 and modified to express human 41BB) cell lines. Binding was detected by using labeled secondary antibodies and measuring fluorescence on a flow cytometer. The results are shown in Table 12.

[0309] Cell lines: Jurkat / NFkB-Luc / h41BB (ACL12578) are Jurkat cells stably transduced and modified to express human 4-1BB using a nuclear factor kappa-light-chain enhancer of activated B cells (NF-κB)-luciferase reporter construct; the cell line was maintained in RPMI-1640 + 10% FBS + L-Glu / PS + 500 μg / mL G418 + 1 μg / mL puromycin. NIH:OVCAR-3 (HCT728) is an epithelial cell line isolated from malignant ascites of patients with progressive ovarian adenocarcinoma. Cells were maintained in RPMI-1640 + 20% FBS + 10 mM HEPES + L-glut + P / S + NaPyr + 10 μg / mL bovine insulin. OVCAR-3 / hPD-L1 (ACL14719) is a NIH:OVCAR-3 cell line modified to express human PD-L1 (accession number NP_054862.1, amino acid M1-T290). Cells were maintained in RPMI-1640 + 20% FBS + 10 mM HEPES + L-glutamate + P / S + NaPyr + 10 μg / mL bovine insulin + 0.25 μg / mL puromycin.

[0310] Assay Setup: OVCAR-3 + / - PD-L1 cells were lifted with trypsin, washed, and resuspended in staining buffer (2% FBS in PBS). OVCAR-3 and Jurkat / NFkB-Luc / h41BB cells were counted and added to wells at 200,000 cells / well. Cells in 96-well round-bottom plates were centrifuged and resuspended in 150 μL staining buffer (PBS containing 2% FBS). The plates were centrifuged again, and then 100 μL of staining buffer alone or staining buffer containing antibody (titrated from 200 nM to 12 pM in a 9-point 1:4 titration) was added to the cells. Cells and antibody were incubated on ice for 30 min, followed by washing in staining buffer. Cells were resuspended in 2 μg / ml allophycocyanin (APC) conjugated goat anti-human secondary antibody. A separate secondary antibody control was included. Cells and secondary antibody were incubated on ice for 30 min and then washed in staining buffer. After subsequent washing with PBS (FBS-free), cells were resuspended in viability dye (reconstituted in DMSO according to the manufacturer's protocol and diluted 1:1000 in PBS) and incubated on ice for 20 min. They were then washed with staining buffer and resuspended in 2% PFA at 4C for 30 min. After washing, the cells were filtered and analyzed by flow cytometry. EC50 of the antibody was determined by a 4-parameter logistic equation on a 9-point dose-response curve (including secondary antibody control only) using GraphPad Prism software. 50 value.

[0311] Controls: In addition to the isotype control and the comparative control, the following controls were used: Control 1: Bispecific antibody with the following: one arm that binds to PD-L1 (derived from parental antibody mAb9373) and the other arm that binds to unrelated antigens; Control 2: Bispecific antibody with the following: one arm that binds to PD-L1 (derived from parental antibody mAb9364) and the other arm that binds to unrelated antigens.

[0312] Binding on Jurkat / NFkB-Luc / h41BB cells: Dose-dependent binding of PD-L1x41BB antibodies (REGN6188, REGN6189, REGN6190, REGN6191) was observed in the presence of Jurkat / NFkB-Luc / h41BB cells, with REGN6191 exhibiting the strongest potency and highest maximum binding. In contrast, the 41BB bivalent parents of REGN6188, REGN6189, and REGN6190 (mAb25894, mAb25898, and mAb25907, respectively) showed larger maximum binding, while the bivalent parent of REGN6191 (mAb25921) showed a similar maximum binding but was more potent. Comparative antibody 2 showed dose-dependent binding to Jurkat / NFkB-Luc / h41BB cells, with potency and maximum binding within the range of other 41BB bivalent antibodies. Additionally, bispecific controls or bivalent PD-L1 (controls 1 and 2, and mAb9373 and mAb9364, respectively) showed low maximum binding on Jurkat / NFkB-Luc / h41BB cells, but this binding was dose-dependent, suggesting that endogenous PD-L1 expression may be low on these cells. Related isotype controls did not bind on Jurkat / NFkB-Luc / h41BB cells.

[0313] Binding on OVCAR3 and OVCAR3 / PD-L1 cells: Low-dose-dependent binding of PD-L1x41BB, PD-L1 bivalent, or bispecific controls was observed on OVCAR3 cells, indicating low levels of endogenous PD-L1 expression. Notably, the maximum binding was not significantly higher than that observed in the isotype control. Conversely, for OVCAR3 / hPD-L1 cells, both PD-L1x41BB and the PD-L1 control antibody showed high maximum-dose-dependent binding in the low to sub-nanomolar range compared to the isotype control.

[0314] Table 12: Maximum binding and EC50 values ​​of antibody binding

[0315]

[0316] Mean fluorescence intensity (MFI) values ​​were plotted using GraphPad Prism, and the EC50 values ​​of the antibody were determined on a 9-point dose-response curve (with a 1:4 titration range of 200 nM to 12 pM for the antibody) and under antibody-free conditions using a four-parameter, variable slope, nonlinear regression equation.

[0317] Maximum MFI is the highest MFI along the dose-response curve, and the binding fold is the maximum MFI divided by the MFI value from the secondary antibody alone.

[0318] Abbreviations: ND: Undetermined because no dose-dependent response was observed; NC: Not calculated because the data do not conform to the 4-parameter logistic equation; N / A: Unavailable because the conditions were not tested.

[0319] Example 4: Activation of T cells by PD-L1x4-1BB antibody

[0320] Two signals, “Signal 1” and “Signal 2”, are essential for proper T cell activation. “Signal 1” is induced by the binding of the T cell receptor (TCR) on T cells to a peptide on antigen-presenting cells (APCs) via the major histocompatibility complex (MHC) molecule. “Signal 2” is provided by the binding of a co-stimulatory receptor expressed on T cells (e.g., the 4-1BB receptor) to a ligand expressed on APCs (e.g., 4-1BBL). Therefore, activation of 4-1BB signaling provides a targeted approach to enhance existing TCR signaling.

[0321] The PD-L1x4-1BB bispecific antibody was designed to work by using PD-L1 + Target cells and 4-1BB + T cell bridging mimics the natural ligand of 4-1BB to provide “signal 2” to enhance T cell activation in the presence of “signal 1” provided by the tumor-associated antigen (TAA) x CD3 bispecific antibody.

[0322] T cell activation was characterized using reported assays.

[0323] In this modified reporter assay, the ability of a PD-L1-targeting antibody to interact with and activate 4-1BB, a co-stimulatory receptor present on T cells as “signal 2” for proper T cell activation, was measured. In this assay, Jurkat cells were modified to express the reporter gene luciferase (NFkB-Luc) and the co-stimulatory receptor 4-1BB (Jurkat / NFkB-Luc / h41BB) under the control of the transcription factor NF-κB. Target cells were modified HEK293 cells expressing CD20 and PSMA alone or in combination with PD-L1. Reporter cells were incubated with target cells and either a bispecific PD-L1x41BB or control antibody. The ability of the bispecific PD-L1-targeting formulation to specifically bind to PD-L1+ target cells and subsequently bind to and activate 4-1BB on Jurkat / NFkB-Luc / h41BB reporter cells was evaluated by luminescent readout.

[0324] Experimental procedure: One day prior to the experiment, Jurkat reporter cells were divided into 5 × 10⁶ cells in RPMI + 10% FBS + penicillin / streptomycin / L-glutamine (P / S / G) + 1 μg / ml puromycin + 500 μg / ml G418 growth medium. 5 Cells / ml. On the day of the experiment, Jurkat / NFkB-luc / h4-1BB cells were resuspended in assay medium (RPMI supplemented with 10% FBS + P / S / G) and cultured at 5 × 10⁻⁶ cells / ml. 4 The final concentration of cells / well was added to 96-well white plates. HEK293 / hCD20 / hPSMA and HEK293 / hCD20 / hPSMA / hPD-L1 were detached with trypsin, washed, and resuspended in assay medium. Cells were added at 1×10⁻⁶ cells / well. 4 The final concentration of cells / well was added to the wells of a 96-well white plate. Subsequently, PD-L1x4-1BB, 4-1BB divalent, Comparative 2, PD-L1 divalent, and isotype control antibodies were titrated from 15 pM to 100 nM at a 1:3 dilution, with the endpoint of the 10-point dilutions being antibody-free. All titrations were performed in duplicate and added to the appropriate wells. Antibody dilutions were generated in the assay medium. The plates were incubated at 37°C and 5% CO2 for 5 hours, and then ONE-Glo luciferase substrate was added to each well according to the manufacturer's instructions. Luciferase activity was recorded as a luminescent signal using an ENVISION plate reader and expressed as a relative light unit (RLU). GraphPad Prism was used. TM EC is determined using a 4-parameter logic equation on the 10-point response curve. 50Value. The signal recorded at point 10 on the dilution curve at 5 pM (antibody-free condition). The maximum RLU is given as the average maximum response detected within the reagent volume range.

[0325] Results: In the presence of HEK293 / hCD20 / hPSMA / hPD-L1 target cells, PD-L1x4-1BB molecules (REGN6188, REGN6189, REGN6190, and REGN6191) and Comparative 2 resulted in a dose-dependent increase in NF-κB activity, with REGN6191 causing the most potent observed response. PD-L1 bivalent, 41BB bivalent, and isotype controls did not increase NF-κB activity in a dose-dependent manner.

[0326] In the absence of PD-L1 expression, comparator 2 resulted in a dose-dependent increase in NF-κB activity. At the highest tested concentrations, REGN6189 and REGN6190 resulted in a slight increase in NF-κB activity; however, due to the lack of a signal plateau, EC50 values ​​were not generated. REGN6188 and REGN6191 did not produce dose-response curves, and their maximum responses were comparable to those of the isotype control.

[0327] The luciferase activity and potency values ​​of the antibodies are shown in Table 13.

[0328] Table 13: Maximum luciferase activity and potency of antibodies

[0329]

[0330] Abbreviations: ND: Undetermined because no dose-dependent response was observed; NC: Not calculated because the data do not conform to the 4-parameter logistic equation.

[0331] Characterization of T cell activation using human T cells

[0332] In the presence of a human embryonic renal cell carcinoma line (HEK293 / HCD20 / HPSMA / hPD-L1) modified to express hCD20, hPSMA, and hPD-L1, this study evaluated the ability of a PD-L1x4-1BB bispecific antibody to activate primary human T cells by conjugating PD-L1 on APCs to 4-1BB on T cells to deliver “signal 2,” as determined by IL-2 release. In this system, a bispecific CD20xCD3 antibody was used to provide “signal 1.” HEK293 cells expressing only hCD20 and hPSMA were included as a control to measure the activity that would occur in the absence of PD-L1 on APCs.

[0333] Experimental Procedure: Human peripheral blood mononuclear cells (PBMCs) were isolated from healthy donor leukocyte packs from Precision for Medicine (donor 555075) using the following protocol: 15 mL of density gradient medium (FicollPaque Plus) was added to an empty 50 mL tube. The leukocyte pack was diluted 1:1 with PBS + 2% FBS. The leukocyte pack was allowed to separate into layers on top of the density gradient medium. Centrifuged at 400 G for 30 minutes at room temperature, then braked off. The upper plasma layer was removed and discarded. The mononuclear cell layer at the plasma-density gradient medium interface was removed and retained. PBS + 2% FBS was added to the collected mononuclear cell layer, and centrifuged at 300 G for 8 minutes. The supernatant was discarded, and the resulting PBMCs were resuspended in an appropriate culture medium. CD3 was used. + Use the T-cell isolation kit (Stemcell) and follow the manufacturer's recommended instructions to isolate CD3 cells. + T cells.

[0334] IL2 release assay: CD3-enriched cells resuspended in stimulation medium were... + T cells at 1×10 5 A concentration of cells / well was added to 96-well round-bottom plates. Growth-inhibiting (arresting) HEK293 / hCD20 / hPSMA or HEK293 / hCD20 / hPSMA / hPD-L1 was added at a concentration of 1×10⁻⁶ cells / well. 4 The final concentration per cell / well added to CD3 +T cells. After cell addition, a constant 0.1 nM of CD20xCD3 antibody or its matched isotype control was added to wells containing HEK293 / hCD20 / hPSMA or HEK293 / hCD20 / hPSMA / hPD-L1. Subsequently, PD-L1x4-1BB, 4-1BB bivalent, Comparative 2, PD-L1 bivalent, and isotype control antibodies were titrated at 1:3 dilutions from 15 pM to 100 nM and added to the wells. The endpoint of the 10-point dilutions was free of titrating antibody. The plates were incubated at 37°C and 5% CO2 for 72 hours, and 5 μL of total supernatant was removed for IL2 measurement. The amount of cytokines in the supernatant was determined according to the manufacturer's protocol using the AlphaLisa kit from PerkinElmer. Cytokine measurements were obtained on the PerkinElmer Envision multilabel plate reader and reported as pg / mL. All serial dilutions were tested in duplicate. Using GraphPad Prism TM The software determines the antibody's EC50 using a four-parameter logistic equation on a 10-point dose-response curve. 50 Values, where point 10 (antibody without titration) is represented by 5.1 pM. Maximum IL2 is given as the average maximum response detected within the range of reagent titrations.

[0335] Results: In the presence of the target and “signal 1” provided by CD20xCD3, treatment with PD-L1x4-1BB antibodies (REGN6188, REGN6189, REGN6190, REGN6191) and Comparative 2 resulted in a dose-dependent increase in IL-2 release compared to 41BB bivalent, PD-L1 bivalent, and isotype control antibodies, with REGN6191 causing the largest increase. In the absence of “signal 1”, no tested antibody enhanced cytokine release from T cells, while in the presence of CD20xCD3 and HEK293 / hCD20 / hPSMA cells (lacking PD-L1), only Comparative 2 resulted in a dose-dependent increase in IL-2 release.

[0336] The luciferase activity and potency values ​​of the antibodies are shown in Table 14.

[0337] Table 14: Maximum IL2 Release and Potential Value of Antibodies

[0338]

[0339] Abbreviations: ND: Undetermined because no dose-dependent response was observed; NC: Not calculated because the data do not conform to the 4-parameter logistic equation.

[0340] Example 5: Blocking of PD-L1-PD1 binding by PD-L1x4-1BB antibody

[0341] Characterization of PD-L1 and 4-1BB blockade by ELISA

[0342] An ELISA-based blocking assay was developed to determine the ability of the PD-L1x4-1BB bispecific antibody to block the binding of human programmed death-ligand 1 (hPD-L1) to human programmed cell death protein 1 (hPD1) or human tumor necrosis factor receptor superfamily member 9 (h4-1BB) to human tumor necrosis factor ligand superfamily member 9 (h4-1BBL).

[0343] The recombinant human PD-L1 protein used in the experiment contains the hPD-L1 extracellular domain (amino acids F19 to T239), which is expressed at the C-terminus along with the Fc portion of mouse IgG2a (amino acids E98 to K330) (hPD-L1-mFc, accession number NP_054862.1). The human PD1 protein contains the hPD1 extracellular domain (amino acids L25 to V170; C93S), which is expressed at the C-terminus along with the Fc portion of human IgG1 (amino acids D104 to K330) (hPD1-hFc, accession number NP_005009.2). The human 4-1BB protein contains the h4-1BB extracellular domain (amino acids L24 to Q186), which is expressed along with the C-terminal myc-myc-6 histidine tag (h4-1BB-mmh, accession number Q07011-1). The human 4-1BB ligand protein was purchased from AcroBiosystems and contains the h4-1BBL extracellular domain (amino acids R71 to E254), which is expressed together with the N-terminal 6-histidine-Flag tag (6His-Flag-h4-1BBL, accession number P41273-1).

[0344] In the PD-L1 blocking assay, hPD1-hFc protein (2 g / ml in PBS) was coated onto 96-well microtiter plates overnight at 4°C. Non-specific binding sites were then blocked using 0.5% (w / v) BSA in PBS. In separate 96-well microtiter plates, a fixed amount of 0.7 nM hPD-L1-mFc was bound for 1 hour to PD-L1x4-1BB bispecific antibody, its parental bivalent anti-PD-L1, or anti-4-1BB antibody, PD-L1x4-1BB comparative, or human IgG1, IgG4, or IgG4s isotype antibody diluted in PBS + 0.5% BSA in the range of 1.7 pM to 100 nM. The fixed concentration of hPD-L1-mFc was selected to be close to the 50% maximum binding (EC50) of the hPD1 plate. 50 The concentration of hPD-L1-mFc antibody complex was then transferred to an hPD1-coated plate. After incubation at room temperature for 1 hour, the plate was washed and the plate-bound hPD-L1-mFc protein was detected using a horseradish peroxidase (HRP)-conjugated goat anti-mouse Fcγ fragment-specific antibody. The plate was then developed using TMB substrate solution (BD Biosciences) according to the manufacturer's recommendations, and the absorbance was measured at 450 nm on a SpectraMax I3x plate reader.

[0345] Similar to the PD-L blocking assay, a 4-1BB blocking assay was performed. 6His-Flag-h4-1BBL protein (2 g / ml in PBS) was coated onto 96-well microtiter plates overnight at 4°C. In separate 96-well microtiter plates, a fixed amount of 0.25 nM h4-1BB-mmh was bound for 1 hour to PD-L1x4-1BB bispecific antibody, parental bivalent anti-PD-L1 or anti-4-1BB antibody, comparative antibody 1, 2, or 3, or an associated human isotype control antibody diluted in PBS + 0.5% BSA in the range of 1.7 pM to 100 nM. The fixed concentration of h4-1BB-mmh was selected to be close to 50% of the maximum binding (EC50) that produced the 6His-Flag-h4-1BBL plate. 50 The concentration of h4-1BB-mmh antibody complex was then determined. The complex was then transferred to a 6His-Flag-h4-1BBL-coated plate. After incubation at room temperature for 1 hour, the plate was washed, and the plate-bound h4-1BB-mmh protein was detected using an HRP-conjugated anti-cMyc antibody. The plate was developed using TMB substrate solution (BD Biosciences), and absorbance was measured at 450 nm on a SpectraMax I3x plate reader.

[0346] Combining data with an S-shaped (four-parameter logic) dose-response model using GraphPad Prism TM The software performs the analysis. The calculated IC... 50 The value (defined as the antibody concentration required to block 50% binding of hPD-L1-mFc to plate-coated hPD1-hFc or h4-1BB-mmh to plate-coated 6His-Flag-h4-1BBL) is used as an indicator of blocking efficacy. The percentage of blockade by the test antibody at the highest tested concentration of 100 nM is calculated based on the following formula:

[0347]

[0348] Antibodies that block more than 50% binding at the highest tested concentration are classified as blocking agents, and their IC50 is reported. 50 value.

[0349] Results: Four PD-L1x4-1BB bispecific antibodies (REGN6188, REGN6189, REGN6190, and REGN6191) exhibited concentration-dependent blocking of hPD-L1-hPD1 binding for blocking PD-L1-PD1 interaction, with blocking rates ranging from 97% to 99% at the highest tested antibody concentration (100 nM). The IC50 values ​​for these bispecific antibodies were... 50 The values ​​ranged from 0.2 nM to 1.2 nM (Table 15). Their parental bivalent PD-L1 antibodies (mAb9373 and mAb9364) showed similar 99% blockade percentages and IC50 values ​​of 0.49 nM and 0.52 nM, respectively. 50 The antibody inhibited the binding of hPD-L1 to hPD1. All parental bivalent 4-1BB antibodies (mAb25894, mAb25898, mAb25907, and mAb25921) and human isotype control antibodies showed no or low blocking effect on hPD-L1 binding to hPD1, with blocking percentages ranging from -1% to 19%. These antibodies were classified as non-blocking agents. Comparative antibody 3 showed 99% blocking effect and an IC50 of 0.46 nM. 50 Inhibits the binding of hD-L1 to hPD1.

[0350] Table 15: Summary of PD-L1x4-1BB bispecific Abs and their parental Abs blocking the binding of human PD-L1 to human PD1

[0351]

[0352] NBL: Non-blocking, %blocking less than or equal to 50%.

[0353] NA: Unavailable

[0354] Regarding the blocking of the interaction between 4-1BB and 4-1BBL, three of the four evaluated PD-L1x4-1BB bispecific antibodies (REGN6189, REGN6190, and REGN6191) showed concentration-dependent blocking of the binding of h4-1BB to h4-1BBL, with blocking percentages ranging from 64% to 99% at the highest antibody concentrations tested. Furthermore, REGN6191 showed a high IC50 value. 50 The value was 1.9 nM. Their parental bivalent 4-1BB antibodies (mAb25898, mAb25907, and mAb25921) showed similar blocking percentages ranging from 70% to 94%, and mAb25921 (the parental 4-1BB of REGN6191) had an IC50 value of 1.9 nM. 50 The value was 0.75 nM. One of the four PD-L1x4-1BB bispecific antibodies (REGN6188) showed less than 50% blockade and was classified as a non-blocker, while its parental bivalent 4-1BB antibody (mAb25894) inhibited h4-1BB binding to h4-1BBL with 61% blockade. Comparative antibodies Comparative 1 and Comparative 3 inhibited h4-1BB binding to h4-1BBL with 96% and 72% blockade, respectively, and IC50 was 0.75 nM. 50 The values ​​were 0.33 nM and undetermined, respectively. Another comparative antibody, Comparative Antibody 2, was a non-blocking agent with 12% blocking activity. All parental bivalent PD-L1 antibodies (mAb9373 and mAb9364) and human isotype control antibodies showed no blocking activity, as expected.

[0355] Table 16: Summary of PD-L1x4-1BB Bispecific Abs and Their Parental Abs Blocking the Binding of Human 4-1BB to Human 4-1BBL

[0356]

[0357] NBL: Non-blocking, %blocking less than or equal to 50%.

[0358] ND: Undetermined due to insufficient S-curve for IC50 calculations.

[0359] NA: Unavailable

[0360] ( In separate experiments, the blocking effect of Comparative Antibody 3 and the isotype control antibody on the binding of h4-1BB to h4-1BBL was tested.

[0361] Characterization of PD-L1 blockade in cells

[0362] Characterization of the PD-L1x4-1BB bispecific antibody was performed using WSU-DLCL2, WSU-DLCL2 / hPD-L1, and Jurkat / AP1-luc / hPD1 cells in a PD-L1 blocking assay. This study investigated whether bivalent PD-L1, PD-L1x4-1BB, bispecific control antibodies, or related isotype controls could block the interaction between PD-L1 and PD-1. The CD20xCD3 bispecific antibody provided "signal 1" by binding CD20 (endogenously expressed) on WSU-DLCL2 target cells to CD3 on Jurkat / AP1-luc / hPD1 cells, leading to increased AP-1-Luc activity, as observed by the enhanced luminescent signal. WSU-DLCL2 cells modified to express PD-L1 resulted in decreased AP-1 activity. The ability of the PD-L1 antibody to block the PD-L1 interaction and subsequently restore luciferase signaling was evaluated.

[0363] Cell lines: Jurkat / AP1-Luc / hPD1 (ACL8709) were generated from Jurkat cells stably transduced using the human programmed cell death protein 1 construct; the cell line was maintained in RPMI-1640 + 10% FBS + L-Glu / PS + 1ug / mL puromycin. WSU-DLCL2 / hPD-L1 (ACL17386) were generated from WSU-DLCL2 cells stably transduced using the human programmed cell death ligand-1 construct; the cell line was maintained in RPMI-1640 + 10% FBS + L-Glu / PS + 1ug / mL puromycin. WSU-DLCL2 cells (HCT883) were maintained in RPMI-1640 + 10% FBS + L-Glu / PS.

[0364] Experimental setup: One day prior to the experiment, Jurkat reporter cells were divided into 5 × 10⁶ cells in RPMI + 10% FBS + penicillin / streptomycin / L-glutamine (P / S / G) + 1 μg / ml puromycin growth medium. 5 Cells / ml. Jurkat / AP1-Luc / PD1 was resuspended in assay medium (RPMI supplemented with 10% FBS + P / S / G) and incubated at 5 × 10⁻⁶ cells / ml. 4 A concentration of 10 cells / well was added to 96-well white plates. Antigen-presenting cells with or without PD-L1 expression (WSU-DLCL2 / PD-L1 or WSU-DLCL2, respectively) were also resuspended in assay medium at 2.5 × 10⁻⁶ cells / well. 4Cell / well concentrations were added to the plate. Bispecific CD20xCD3 antibody was added to all wells at a constant concentration of 1 nM. Subsequently, PD-L1x4-1BB, or associated monospecific or bispecific control or isotype control antibody, was titrated from 7.6 pM to 500 nM at a 1:4 dilution, with the endpoint of 10-point dilutions being free of titrated antibody (constant CD20xCD3 only). All titrations were performed in duplicate and added to the appropriate wells. Antibody dilutions were generated in the assay medium. The plate was incubated at 37°C and 5% CO2 for 5 hours, and then ONE-Glo luciferase substrate was added to each well according to the manufacturer's instructions. Luciferase activity was recorded as a luminescent signal using an ENVISION plate reader and expressed as relative light units (RLU). GraphPad Prism was used. TM EC is determined using a 4-parameter logic equation on the 10-point response curve. 50 Values. The signal recorded at point 10 was plotted on a dilution curve at 1.9 pM. The maximum RLU is given as the average maximum response detected within the range of reagent doses. Controls 1 and 2 (as described above) were used in the assay. Anti-PD1 antibody (REGN2810; cimipril) was also included in the experiment.

[0365] Results: In the presence of CD20xCD3 antibody, PD-L1-expressing WSU-DLCL2 cells showed reduced AP-1 activity compared to PD-L1-negative WSU-DLCL2 cells.

[0366] In the absence of PD-L1 expression in WSU-DLCL2 cells, PD-L1 bispecific antibodies and bivalent antibodies and controls did not affect AP-1 activity.

[0367] In the presence of PD-L1 expression on WSU-DLCL2 cells, PD-L1 bispecific, PD-L1 divalent, REGN2810, and bispecific control (Control 1 and Control 2) antibodies resulted in dose-dependent recovery of the luminescent signal, which was inhibited by the interaction between PD1 and PD-L1. The isotype control did not result in increased AP-1 activity.

[0368] Table 17: Maximum RLU and Potential Value of Antibodies (AP-1 Activity)

[0369]

[0370] Abbreviations: ND: Undetermined because no dose-dependent response was observed; NC: Not calculated because the data do not conform to the 4-parameter logistic equation.

[0371] Example 6: Enhancement of MUC16+ cell killing by combination of bispecific PD-L1x4-1BB antibody and Muc16xCD3 antibody

[0372] The ability of co-stimulatory PD-L1x4-1BB bsAb REGN6190 or REGN6191 bispecific antibodies to enhance the cytotoxic potency of TAAxCD3 bsAb REGN4019 (WO 2018 / 067331) targeting MUC16, a tumor antigen expressed on the surface of OVCAR-3 tumor cells modified to overexpress hPD-L1, was tested. Comparative compound 3 was included in the experiment.

[0373] To monitor the killing of MUC16+ cells by flow cytometry in the presence of a combination of hPBMCs and a bispecific antibody against MUC16xCD3 and PD-L1x4-1BB (bsAb), OVCAR-3 / hPD-L1 cells were labeled with the 1 μM fluorescent tracking dye Violet cell tracker. After labeling, cells were plated overnight at 37°C. Human PBMCs were separately labeled at 1 × 10⁻⁶. 6 Cells / mL plates were seeded in supplemented RPMI medium and incubated overnight at 37°C to enrich lymphocytes by depleting adherent macrophages, dendritic cells, and some monocytes. The next day, target cells were co-incubated at 37°C for 72 hours with adherent cell-depleted initial PBMCs (effect cell / target cell ratio 4:1), serially diluted MUC16xCD3 bispecific antibody REGN4019 or isotype control (concentration range: 66.7 nM to 150 pM), and a fixed concentration of PD-L1x4-1BB co-stimulatory bispecific antibody (2.5 ug / ml (16.7 nM)). Cells were removed from the culture plates using trypsin-EDTA dissociation buffer and analyzed by flow cytometry on a BD Celesta cell counter. For flow cytometry analysis, cells were stained with dead / live Near IR Reactive (Invitrogen) dye. Immediately before FACS analysis, 2E04 counting beads were added to each well. 1E04 beads were collected for each sample. To assess the specificity of killing, cells were gated to a viable, Violet-tagged population. The percentage of the viable population was recorded and used to calculate survival.

[0374] T cell activation and upregulation of PD1 markers were assessed by incubating cells with antibodies directly conjugated to CD2, CD4, CD8, CD25, and PD1, and by reporting the percentage of activated (CD25+ / CD8+, CD25+ / CD4+) T cells and PD1+ / CD4+, PD1+ / CD8+ T cells as a percentage of total T cells (CD2+).

[0375] The release of Th1 / Th2 cytokines from the test well supernatant of human PBMC assays was evaluated using the BD Cell Counting Bead Array Human Kit and according to the manufacturer’s protocol.

[0376] Results: In the absence of a bispecific agent targeting xCD3 (MUC16xCD3 REGN4019), REGN6190, REGN6191, and Comparative 3 were all inert. However, in the presence of human PBMCs, fixed concentrations of REGN6190 and REGN6191 PD-L1x4-1BB bsAb successfully enhanced the cytotoxic potency of REGN4019 against OVCAR-3 / hPD-L1 cells, with cells being killed in a dose-dependent manner. The potency of REGN6190 and REGN6191 was comparable to that of Comparative 3. In addition to the observed enhanced target cell lysis, T cell activation was also increased, with upregulated expression of CD25 and PD1 on CD4+ and CD8+ T cells (Table 18).

[0377] Table 18: T cell activation and killing of ECGs via a combination of bispecific PD-L1x4-1BB antibody and Muc16xCD3 antibody (REGN4019) 50

[0378]

[0379]

[0380] N / C: Not calculable

[0381] The release of cytokines from the supernatant of the cytotoxicity assay was evaluated. The cytotoxic potency of REGN4019 was correlated with the accumulation of IFNg in the culture medium, and the accumulation of IFNg in the culture medium was enhanced when REGN4019 was combined with PD-L1x4-1BB.

[0382] Table 19: Cytokine release induced by the combination of bispecific PD-L1x4-1BB antibody and Muc16xCD3 antibody (REGN4019)

[0383]

[0384] N / C: Not calculable

[0385] Example 7: Robust antitumor efficacy of four bispecific PD-L1x4-1BB antibodies in an MC38 / hPDL1 tumor model

[0386] This embodiment relates to an in vivo study demonstrating the efficacy of a bispecific PD-L1x4-1BB antibody in treating human PD-L1-expressing tumors in a mouse model of colon cancer compared to treatment with single or combined anti-PD-L1 or anti-4-1BB antibodies. Mice were humanized against PD1, PD-L1, and 4-1BB by knocking out mouse genes and replacing them with their human homologs (hPD1 / hPD-L1 / h4-1BB mice), and tumors were generated from MC38 colon cancer cells (M38-hPDL1-mPDL1KO cells) modified to knock out mouse PD-L1 and overexpress human PD-L1. The following antibodies were used in this study: PD-L1x4-1BB (REGN6188, REGN6189, REGN6190, and REGN6191); PD-L1xBetV1 (a bispecific control antibody with one arm binding PD-L1 and the other arm binding the unrelated antigen BetV1); anti-PD-L1 (mAb9373, the parent antibody of REGN6191); anti-4-1BB (mAb25921, the parent antibody of REGN6191); and isotype controls. Figure 1 In this context, mAb9373 is referred to as “a-PD-L1”, mAb25921 as “a-4-1BB”, and PD-L1xBetV1 as “aPD-L1xBetV1”.

[0387] On day 0, M38-hPDL1-mPDL1KO cells were implanted into 6 to 7 hPD1 / hPD-L1 / h4-1BB mice in each group. Tumors reached approximately 100 mm on day 8. 3 The average tumor volume was measured, and antibodies were administered at a dose of 5 mg / kg on days 8, 11, 15, 19, and 23. Tumors were measured approximately twice weekly, and tumor-free mice were identified at the end of the experiment on day 59.

[0388] Results: Compared with administration of mAb9373 (a-PD-L1 monospecific agent), mAb25921 (a-4-1BB monospecific agent), PD-L1xBetV1 bispecific agent, or isotype control, administration of any one of REGN6188, REGN6189, REGN6190, or REGN6191 (PD-L1x4-1BB bispecific agent) strongly controlled tumor growth. Figure 1 For example, on day 24, although the average tumor volume in mice treated with the isotype control was approximately 2600 mm...3 However, mice treated with any one of REGN6188, REGN6189, REGN6190, or REGN6191 showed a significant reduction in mean tumor volume (e.g., approximately 170 mm for REGN6190). 3 , Figure 1 In contrast, on the same day 24, the average tumor volume in mice treated with the a-PD-L1 monospecific antibody mAb9373 was approximately 2000 mm. 3 The average tumor volume in mice treated with PD-L1xBetV1 was approximately 1700 mm. 3 The average tumor volume in mice treated with the α-4-1BB monospecific antibody mAb25921 was approximately 1400 mm. 3 ( Figure 1 Administration of REGN6190 was as effective as combination therapy with the α-4-1BB antibody mAb9373 and the α-PD-L1 monospecific antibody mAb25921 in inhibiting tumor growth. Monotherapy with REGN6188, REGN6189, REGN6190, or REGN6191 was also superior to monotherapy with any one of mAb9373, PD-L1xBetV1, or mAb25921 in inducing complete tumor regression. In mice treated with any one of REGN6188, REGN6189, REGN6190, or REGN6191, the percentage of tumor-free mice at the end of day 59 was between 2 / 7 and 4 / 7 (Table 20). In contrast, at day 59, no mice were tumor-free in mice treated with mAb9373 (α-4-1BB), PD-L1xBetV1, mAb25921 (α-PD-L1), or the isotype control. At the end of day 59, the ability of REGN6188, REGN6189, REGN6190, or REGN6191 monotherapy to induce complete tumor regression (tumor-free in 2 / 7 to 4 / 7 mice) was equivalent to that of mAb9373 (a-PD-L1) and mAb25921 (a-4-1BB) combination therapy to induce complete tumor regression (tumor-free in 3 / 7 mice). The superior antitumor efficacy of the PD-L1x4-1BB bispecific antibody compared to antibodies binding to PD-L1 or 4-1BB was an unexpected result.

[0389] Table 20: Percentage of tumor-free mice 59 days after implantation

[0390]

[0391] Example 8: Strong tumor growth control and survival with bispecific PD-L1x4-1BB antibody monotherapy in an MC38 / hPDL1 tumor model

[0392] This embodiment relates to an in vivo study demonstrating the efficacy of a bispecific PD-L1x4-1BB antibody in inhibiting tumor growth and inducing complete and durable regression of M38 tumors expressing human PD-L1. Mice were humanized against PD1, PD-L1, and 4-1BB by knocking out the mouse gene and replacing it with its human homolog (hPD1 / hPD-L1 / h4-1BB mice), and tumors were generated from MC38 colon cancer cells (M38-hPDL1-mPDL1KO cells) modified to knock out mouse PD-L1 and overexpress human PD-L1. The following antibodies were used in this study: PD-L1x4-1BB (REGN6191); anti-PD-L1 (H4H8314N, US 2022 / 0184241); anti-4-1BB (mAb25921, the parental antibody of REGN6191); and an isotype control. Figures 2A to 2B In this context, H4H8314N is referred to as “a-PD-L1”, mAb25921 as “a-4-1BB”, and REGN6191 as a-PD-L1x4-1BB.

[0393] On day 0, M38-hPDL1-mPDL1KO cells were implanted into 4 to 7 hPD1 / hPD-L1 / h4-1BB mice per group. Tumors reached approximately 100 mm on day 7. 3 The mean tumor volume was measured, and antibodies were administered on days 7, 11, 14, 18, and 21. All antibodies were administered at a dose of 5 mg / kg, except for the second REGN6191 group which was given 10 mg / kg. Tumors were measured approximately twice weekly until day 38, and survival was recorded at the end of the experiment on day 60.

[0394] Results: Compared with H4H8314N or mAb25921 alone or in combination, REGN6191 significantly controlled tumor growth. Figure 2A For example, on day 21, although the average tumor volume of mice treated with the isotype control was approximately 1500 mm, 3 However, mice treated with REGN6191 had significantly lower average tumor volume (i.e., approximately 80 mm). 3 In contrast, on the same day 21, the average tumor volume in mice treated with H4H8314N was approximately 1200 mm. 3 The average tumor volume in mice treated with mAb25921 was approximately 800 mm. 3 The average tumor volume in mice treated with the combination of H4H8314N and mAb25921 was approximately 300 mm. 3Furthermore, mice administered REGN6191 had significantly better survival rates than mice administered H4H8314N or mAb25921, and significantly better survival rates than mice administered a combination of H4H8314N and mAb25921. Figure 2B For example, by day 35, none of the mice treated with H4H8314N or mAb25921 alone survived, and by the end of day 60, 43% of the mice treated with the combination of H4H8314N and mAb25921 survived. In contrast, by day 60, 73% of the mice treated with 5 mg / kg REGN6191 survived, and 75% of the mice treated with 10 mg / kg REGN6191 survived. By the end of day 60, REGN6191 monotherapy was also superior to the combination of H4H8314N and mAb25921 in inducing complete tumor regression. The proportion of tumor-free mice was 5 / 7 among those treated with 5 mg / kg REGN6191, and 3 / 4 among those treated with 10 mg / kg REGN6191. In contrast, at the end of day 60, 3 out of 7 mice treated with the combination of H4H8314N and mAb25921 were tumor-free. In summary, PD-L1x4-1BB monotherapy demonstrated unexpected improvements over anti-PD-L1 monotherapy, 4-1BB monotherapy, or the combination of anti-PD-L1 and 4-1BB in inhibiting tumor growth, improving survival, and inducing complete and durable tumor regression.

[0395] Example 9: In an MC38 / hPDL1 tumor model, REGN6191 monotherapy demonstrated superior tumor growth control and survival compared to the comparative drug PD-L1x4-1BB bispecific antibody.

[0396] This embodiment relates to an in vivo study demonstrating the superior ability of REGN6191 to inhibit tumor growth and induce complete and durable regression compared to the comparative PD-L1x4-1BB antibody. Mice were humanized against PD1, PD-L1, and 4-1BB by knocking out mouse genes and replacing them with their human homologs (hPD1 / hPD-L1 / h4-1BB mice), and tumors were generated from MC38 colon cancer cells (M38-hPDL1-mPDL1KO cells) modified to knock out mouse PD-L1 and overexpress human PD-L1. The following antibodies were used in this study: five PD-L1x4-1BB antibodies (REGN6188, REGN6189, REGN6190, REGN6191 and comparative 3 (according to WO 2019 / 025545 (GenMab), V with antibodies “CD137-009-HC7LC2” and “PD-L1-547”). H / VL Bispecific PD-L1x4-1BB antibody (sequence); PD-L1xBetV1 (a bispecific control antibody with one arm binding PD-L1 and the other arm binding the unrelated antigen BetV1); anti-PD-L1 (mAb9373, i.e., the parent antibody of REGN6191); anti-4-1BB (mAb25921, i.e., the parent antibody of REGN619); and isotype control. Figures 3A to 3B In this context, mAb9373 is referred to as “a-PD-L1”, mAb25921 is referred to as “a-4-1BB” or “a-41BB”, and PD-L1xBetV1 is referred to as “aPD-L1xBetV1”.

[0397] On day 0, M38-hPDL1-mPDL1KO cells were implanted into seven hPD1 / hPD-L1 / h4-1BB mice in each group. Tumors reached approximately 115 mm on day 8. 3 The average tumor volume was measured, and antibodies were administered at a dose of 10 mg / kg on days 9, 13, 15, and 19. Tumors were measured approximately twice weekly until day 39, and tumor-free mice were identified and their survival was recorded at the end of the experiment on day 43.

[0398] Results: REGN6191 monotherapy was more effective at controlling tumor growth than any other therapy (including monotherapy with Comparator 3 and combination therapy with mAb9373 and mAb25921). Figure 3A For example, on day 19, although the average tumor volume in mice treated with the isotype control was approximately 1600 mm. 3 However, the average tumor volume in mice treated with REGN6191 was approximately 200 mm. 3 In contrast, also on day 19, the average tumor volume in mice treated with mAb9373 was approximately 900 mm. 3 The average tumor volume in mice treated with PD-L1xBetV1 was approximately 750 mm. 3 The average tumor volume in mice treated with either Comparative Agent 3 or mAb25921 was approximately 350 mm. 3 The average tumor volume in mice treated with a combination of mAb9373 and mAb25921 was approximately 250 mm. 3 Furthermore, mice treated with REGN6191 monotherapy had better survival rates than mice treated with any other monotherapy, including Comparative Agent 3 monotherapy (…). Figure 3BREGN6191 monotherapy was also superior to any other monotherapy in inducing complete tumor regression, including Comparative 3 monotherapy (Table 21). At the end of the experiment on day 39, 4 / 7 of the mice administered REGN6191 were tumor-free, but only 3 / 7 of the mice administered Comparative 3 were tumor-free.

[0399] Table 21: Percentage of tumor-free mice 60 days after implantation

[0400]

[0401]

[0402] In summary, REGN6191 demonstrated greater antitumor efficacy than H4H8314N or mAb25921 alone or in combination. Furthermore, REGN6191 monotherapy showed superior ability to inhibit tumor growth, improve survival, and induce complete and durable tumor regression compared to comparative drug 3 monotherapy.

[0403] Example 10: Robust ability of PD-L1x4-1BB to enhance the efficacy of MUC16xCD3 in controlling OVCAR-3 / hPD-L1 tumor growth

[0404] This embodiment relates to an in vivo study demonstrating the ability of a bispecific PD-L1x4-1BB antibody to enhance the efficacy of MUC16xCD3 in treating human PD-L1-overexpressing OVCAR-3 tumors. Mice were a highly immunodeficient SRG mouse strain transplanted with human peripheral blood mononuclear cells (hPBMCs), which are derived from human immune cells. Tumors were generated from OVCAR-3 tumor cells (OVCAR-3 / Luc / PD-L1 cells) that endogenously expressed MUC16 and were modified to overexpress human PD-L1 and carry the luciferase reporter gene. The following antibodies were used in this study: PD-L1x4-1BB (REGN6191); MUC16xCD3 (REGN4018, WO 2018 / 067331); anti-PD-L1 (H4H8314N, US2022 / 0184241); anti-4-1BB (mAb25921, the parental antibody of REGN6191); and isotype controls. Figure 4 In this context, H4H8314N is referred to as “a-PD-L1”, mAb25921 as “a-4-1BB”, and REGN4018 as “MUC16xCD3”.

[0405] Eight days prior to tumor cell implantation, hPBMCs were implanted into NSG mice. On day 0, OVCAR-3 / Luc / PD-L1 cells were implanted into the mice. On day 5, bioluminescent imaging (BLI) was performed, and mice were randomly assigned to groups with equal mean radiation. Antibodies were administered at a dose of 5 mg / kg on days 6, 9, and 12, except for REGN4018, which was administered at a suboptimal dose of 0.1 mg / kg. BLI was performed on days 5, 9, 12, and 15.

[0406] Results: REGN6191 significantly improved the ability of REGN4018 to control OVCAR3 / hPD-L1 tumor growth. In contrast, H4H8314N and mAb25921 did not significantly improve the antitumor activity of REGN4018. Figure 4 For example, on day 12, the average radiation in mice administered the combination of REGN4018 and isotype control was approximately 18,000 p / s / cm. 2 / sr, but the mean radiation in mice given the combination of REGN4018 and REGN6191 was significantly reduced (i.e., approximately 900 p / s / cm). 2 / sr). In contrast, also on day 12, mice administered the combination of REGN4018 and mAb25921 showed an increase in mean radiation (i.e., approximately 70,000 p / s / cm). 2 The average radiation dose in mice treated with the combination of REGN4018 and H4H8314N was approximately 9000. The superior antitumor efficacy of the combination of MUC16xCD3 and PD-L1 x 4-1BB was an unexpected result compared to either the combination of MUC16xCD3 and anti-PD-L1 or the combination of MUC16xCD3 and anti-4-1BB.

[0407] Example 11: Growth control of tumors that do not express human PD-L1 by combination therapy with PD-L1x4-1BB and anti-human PD-1

[0408] This embodiment relates to an in vivo study demonstrating that the antitumor efficacy of the bispecific PD-L1x4-1BB antibody is increased when combined with anti-human PD-1, even against tumors that do not express human PD-L1. Mice were humanized against PD1, PD-L1, and 4-1BB by knocking out the mouse gene and replacing it with its human homolog (hPD1 / hPD-L1 / h4-1BB mice). Tumors were generated from parental MC38 cells expressing mouse PD-L1, not human PD-L1. The following antibodies were used in this study: PD-L1x4-1BB (REGN6191), anti-PD1 (REGN2810; cimipril), and an isotype control. Figures 5A to 5E In this study, cimipril malaboxetine was referred to as "a-PD-1". Given that (i) the host mice were humanized against both PD-L1 and PD-1, (ii) the tumor cells did not express human PD-L1, and (iii) REGN6191 did not recognize mouse PD-L1, the ability of cimipril malaboxetine to improve the efficacy of REGN6191 suggests a non-tumor PD-L1 response. + The potential role of cells in inhibiting tumor growth.

[0409] On day 0, MC38 parental tumor cells were implanted into 7 to 8 hPD1 / hPD-L1 / h4-1BB mice in each group. The tumors reached approximately 100 mm in size by day 10. 3 The average tumor volume was measured, and antibodies were administered at a dose of 10 mg / kg on days 10, 14, 17, and 20. Tumors were measured approximately twice weekly, and tumor-free mice were identified at the end of the experiment on day 31.

[0410] Results: REGN6191, when administered in combination with cimipril, was more effective than either antibody alone in inhibiting parental MC38 tumor growth and inducing complete tumor regression. Figures 5A to 5E For example, on day 20, although the average tumor volume of parental MC38 tumors treated with isotype controls was approximately 800 mm. 3 However, the mean tumor volume of parental MC38 tumors treated with a combination of REGN6191 and cimipril was significantly reduced (i.e., approximately 180 mm). 3 , Figure 5A In contrast, also on day 20, the mean tumor volume of parental MC38 tumors treated with REGN6191 alone and the mean tumor volume of parental MC38 tumors treated with cimiprimab alone were both approximately 500 mm. 3 ( Figure 5AFurthermore, the combination therapy of REGN6191 and cimiprimab induced complete tumor regression (i.e., 3 out of 8 mice were tumor-free (TF) at the end of day 31), while anti-PD1 monotherapy and REGN6191 monotherapy did not (i.e., 0 out of 7 mice were tumor-free at the end of day 31). Given that the parental MC38 tumor does not express human PD-L1, the ability of the anti-PD1 and PD-L1x4-1BB combination therapy to robustly inhibit tumor growth and induce complete tumor regression was unexpected. This suggests that non-tumor PD-L1... + Cells can help control tumor growth.

[0411] Example 12: Evaluation of the safety spectrum of bispecific PD-L1x4-1BB

[0412] This embodiment relates to an in vivo study demonstrating that administration of the anti-PD-L1x4-1BB bispecific antibody REGN6191 advantageously resulted in the absence of induction of NEFA, ALT, IFNg, and IL-2, compared to administration of Comparative 2 and a combination of anti-PD-L1 and anti-4-1BB monospecific antibodies, in mice with gene knockout and replacement with their human homologs (hPD1 / hPD-L1 / h4-1BB mice). The following antibodies were used in this study: PD-L1x4-1BB (REGN6191), anti-PD-L1 (mAb9373, the parent antibody of REGN6191); two anti-4-1BB antibodies (mAb25921, the parent antibody of REGN6191, and Comparative 2), and an isotype control. Figures 6A to 6D In this context, mAb9373 is referred to as “a-PD-L1”, and mAb25921 is referred to as “a-4-1BB”.

[0413] On days 0, 3, 6, and 7, antibodies were administered to four hPD1 / hPD-L1 / h4-1BB mice in each group according to Table 22. Blood samples were collected on day 0 (4 hours after antibody administration) and on day 7 (24 hours after the third dose). Serum concentrations of ALT, AST, LDL, HDL, NEFA, IL-1β, IL-2, IL-5, IL-6, KC / GRO, IL-10, and TNF-α were measured.

[0414] Table 22: Antibody dosage for each group of mice

[0415]

[0416] Results: Compared with REGN6191, administration of Urelumab induced significantly more IL-2 release on day 0 and significantly more IFNg and NEFA release on day 7. Figure 6A , 6B And 6D). Compared with REGN6191, the combination of mAb9373 and mAb25921 induced significantly greater release of IFNg, ALT and NEFA on day 7. Figures 6B to 6D ).

[0417] Example 13: In the A431 / PBMC xenograft model, the antitumor efficacy of low-dose EGFRxCD3 was significantly improved when combined with PD-L1x4-1BB.

[0418] This embodiment relates to an in vivo study demonstrating the ability of PD-L1x4-1BB to synergize with an anti-tumor-associated antigen (TAA) x CD3 bispecific antibody to inhibit tumor growth in an A431 / PBMC xenograft model of human skin cancer. The exemplary TAAxCD3 antibody used in this study is EGFRxCD3. The mice were a highly immunodeficient SRG mouse strain transplanted with human peripheral blood mononuclear cells (hPBMCs) as a source of human immune cells, and the tumors were generated by A431 cells. The following antibodies were used in this study: PD-L1x4-1BB (REGN6191), PD-L1xBetV1 (a bispecific control antibody with one arm binding PD-L1 and another arm binding the unassociated antigen BetV1), and the exemplary EGFRxCD3 bispecific antibody.

[0419] On day 0, 2×10⁻⁶ SRG mice were implanted into each group of 5 SRG mice. 6 A431 cells and 1×10 6 hPBMCs were collected. Antibody was administered on days 0, 3, 7, and 11. All groups were administered EGFRxCD3 at a dose of 0.01 mg / kg. In addition, each group was administered one of the following: a load cell control, 0.1 mg / kg of PD-L1xBetV1, or 0.1 mg / kg of REGN6191. Tumors were measured approximately twice weekly, and tumor-free mice were identified at the end of the experiment on day 34.

[0420] Results: REGN6191 significantly enhanced the ability of the second-best dose of EGFRxCD3 to inhibit tumor growth. Figures 7A to 7C For example, on day 34, although mice given a suboptimal dose of EGFRxCD3 alone showed 800 mm 3 Up to 3000 mm 3The tumor volume was significantly reduced, but mice treated with the second-optimal dose of EGFRxCD3 combined with REGN6191 showed a significant improvement in tumor growth inhibition, i.e., no tumor was detected at any time point. In contrast, mice treated with the second-optimal dose of EGFRxCD3 combined with PD-L1xBetV1 also showed tumors of approximately 800 to 1400 mm in size at day 34. 3 The tumor volume. In summary, compared with the combination of PD-L1xBetV1 and a suboptimal dose of EGFRxCD3, the combination of PD-L1x4-1BB and a suboptimal dose of the exemplary TAAxCD3 antibody EGFRxCD3 significantly inhibited tumor growth. This is a remarkable and unexpected result.

[0421] This disclosure is not limited by the specific embodiments described herein. In fact, based on the preceding description and accompanying drawings, various modifications to this disclosure, other than those described herein, will become apparent to those skilled in the art. Such modifications are intended to fall within the scope of the appended claims. All patents, applications, and non-patent publications referenced in this specification are incorporated herein by reference in their entirety.

Claims

1. An isolated bispecific antigen-binding molecule, comprising: (a) The first antigen-binding domain, which has a size of less than about 2 × 10 -7 M of K D Specific binding to human 4-1BB was measured by surface plasmon resonance at 25°C; and (b) The second antigen-binding domain, which has a capacitance of less than about 2 × 10⁻⁶. -10 M of K D It specifically binds to human programmed death ligand 1 (PD-L1) and is measured at 25°C by surface plasmon resonance.

2. The isolated bispecific antigen-binding molecule of claim 1, wherein the bispecific antigen-binding molecule has a concentration of less than about 4 × 10⁻⁶. -9 M's EC 50 It binds to the surface of human T cells and is detected by in vitro FACS binding assay.

3. The isolated bispecific antigen-binding molecule according to claim 1 or 2, wherein the bispecific antigen-binding molecule has a concentration of less than about 4 × 10⁻⁶. -9 M's EC 50 It binds to the surface of cells expressing PD-L1 and is detected by in vitro FACS binding assay.

4. The isolated bispecific antigen-binding molecule according to any one of claims 1 to 3, wherein the bispecific antigen-binding molecule is at an IC50 concentration of less than about 1.3 nM. 50 Blocking the binding of PD-L1 to PD-1 was detected by an ELISA-based blocking assay.

5. The isolated bispecific antigen-binding molecule according to any one of claims 1 to 4, wherein the bispecific antigen-binding molecule is combined with a bispecific MUC16xCD3 antibody at a concentration of less than about 10 -10 M's EC 50 Mediates in vitro T cell killing of OVCAR-3 cells expressing PD-L1.

6. The isolated bispecific antigen-binding molecule according to any one of claims 1 to 5, wherein the first antigen-binding domain comprises: (a) Three heavy chain complementarity-determining regions (HCDR1, HCDR2, and HCDR3) contained within a heavy chain variable region (HCVR), the heavy chain variable region (HCVR) comprising an amino acid sequence selected from SEQ ID NO: 58, 42, 32, and 10 or variants thereof; and (b) Three light chain complementarity-determining regions (LCDR1, LCDR2 and LCDR3) contained within a light chain variable region (LCVR), the light chain variable region (LCVR) containing an amino acid sequence selected from SEQ ID NO: 66 and 18 or a variant thereof.

7. The isolated bispecific antigen-binding molecule of claim 6, comprising: HCDR1, comprising an amino acid sequence selected from SEQ ID NO: 60, 34 and 12; HCDR2, comprising an amino acid sequence selected from SEQ ID NO: 62, 44, 36 and 14; and HCDR3, comprising an amino acid sequence selected from SEQ ID NO: 64, 46, 38 and 16.

8. The isolated bispecific antigen-binding molecule of claim 6 or 7, comprising: LCDR1, comprising an amino acid sequence selected from SEQ ID NO: 68 and 20; LCDR2, comprising an amino acid sequence selected from SEQ ID NO: 70 and 22; and LCDR3, comprising an amino acid sequence selected from SEQ ID NO: 72 and 24.

9. The isolated bispecific antigen-binding molecule of claim 7 or 8, wherein the first antigen-binding domain comprises HCVR and LCVR, wherein the HCVR comprises the amino acid sequence of SEQ ID NO: 58 or a variant thereof, and the LCVR comprises the amino acid sequence of SEQ ID NO: 66 or a variant thereof.

10. The isolated bispecific antigen-binding molecule of claim 7 or 8, wherein the first antigen-binding domain comprises HCVR and LCVR, wherein the HCVR comprises the amino acid sequence of SEQ ID NO: 42 or a variant thereof, and the LCVR comprises the amino acid sequence of SEQ ID NO: 18 or a variant thereof.

11. The isolated bispecific antigen-binding molecule of claim 7 or 8, wherein the first antigen-binding domain comprises HCVR and LCVR, wherein the HCVR comprises the amino acid sequence of SEQ ID NO: 32 or a variant thereof, and the LCVR comprises the amino acid sequence of SEQ ID NO: 18 or a variant thereof.

12. The isolated bispecific antigen-binding molecule of claim 7 or 8, wherein the first antigen-binding domain comprises HCVR and LCVR, wherein the HCVR comprises the amino acid sequence of SEQ ID NO: 10 or a variant thereof, and the LCVR comprises the amino acid sequence of SEQ ID NO: 18 or a variant thereof.

13. The isolated bispecific antigen-binding molecule of any one of claims 1 to 12, wherein the second antigen-binding domain comprises: (a) Three heavy chain complementarity-determining regions (HCDR1, HCDR2, and HCDR3) contained within a heavy chain variable region (HCVR), the heavy chain variable region (HCVR) comprising an amino acid sequence selected from SEQ ID NO: 50 and 2 or variants thereof; and (b) Three light chain complementarity-determining regions (LCDR1, LCDR2 and LCDR3) contained within a light chain variable region (LCVR), the light chain variable region (LCVR) containing an amino acid sequence selected from SEQ ID NO: 66 and 18 or a variant thereof.

14. The isolated bispecific antigen-binding molecule of claim 13, wherein the second antigen-binding domain comprises: (a) HCDR1, which contains the amino acid sequence of SEQ ID NO: 52 or SEQ ID NO: 4; (b) HCDR2, comprising the amino acid sequence of SEQ ID NO: 54 or SEQ ID NO: 6; and (c) HCDR3, which contains the amino acid sequence of SEQ ID NO: 58 or SEQ ID NO:

8.

15. The isolated bispecific antigen-binding molecule of claim 13 or 14, wherein the second antigen-binding domain comprises: LCDR1, comprising an amino acid sequence selected from SEQ ID NO: 68 and 20; LCDR2, comprising an amino acid sequence selected from SEQ ID NO: 70 and 22; and LCDR3, comprising an amino acid sequence selected from SEQ ID NO: 72 and 24.

16. The isolated bispecific antigen-binding molecule of claim 15, wherein the second antigen-binding domain comprises: (a) HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 52, 54, and 56, respectively; and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 68, 70, and 72, respectively; or (b) HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 4, 6, and 8, respectively; and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, and 24, respectively.

17. The isolated bispecific antigen-binding molecule of claim 16, wherein the second antigen-binding domain comprises: (a) HCVR containing the amino acid sequence of SEQ ID NO: 50 or a variant thereof, and LCVR containing the amino acid sequence of SEQ ID NO: 66 or a variant thereof; or (b) HCVR containing the amino acid sequence of SEQ ID NO: 2 or a variant thereof, and LCVR containing the amino acid sequence of SEQ ID NO: 18 or a variant thereof.

18. An isolated bispecific antigen-binding molecule comprising: (a) A first antigen-binding domain specifically binding to human 4-1BB, wherein the first antigen-binding domain comprises: HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 60, 62, and 64, respectively, and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 68, 70, and 72, respectively; and (b) A second antigen-binding domain that specifically binds to human PD-L1, wherein the second antigen-binding domain comprises: HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 52, 54, and 56, respectively, and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 68, 70, and 72, respectively.

19. An isolated bispecific antigen-binding molecule comprising: (a) A first antigen-binding domain specifically binding to human 4-1BB, wherein the first antigen-binding domain comprises: HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 12, 44, and 46, respectively, and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, and 24, respectively; and (b) A second antigen-binding domain that specifically binds to human PD-L1, wherein the second antigen-binding domain comprises: HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 4, 6, and 8, respectively, and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, and 24, respectively.

20. An isolated bispecific antigen-binding molecule comprising: (a) A first antigen-binding domain specifically binding to human 4-1BB, wherein the first antigen-binding domain comprises: HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 34, 36, and 38, respectively, and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, and 24, respectively; and (b) A second antigen-binding domain that specifically binds to human PD-L1, wherein the second antigen-binding domain comprises: HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 4, 6, and 8, respectively, and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, and 24, respectively.

21. An isolated bispecific antigen-binding molecule, comprising: (a) A first antigen-binding domain specifically binding to human 4-1BB, wherein the first antigen-binding domain comprises: HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 12, 14, and 16, respectively, and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, and 24, respectively; and (b) A second antigen-binding domain that specifically binds to human PD-L1, wherein the second antigen-binding domain comprises: HCDR1, HCDR2, and HCDR3 domains containing the amino acid sequences of SEQ ID NO: 4, 6, and 8, respectively, and LCDR1, LCDR2, and LCDR3 domains containing the amino acid sequences of SEQ ID NO: 20, 22, and 24, respectively.

22. The isolated bispecific antigen-binding molecule of claim 18, comprising: (a) A first antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 58 and an LCVR containing the amino acid sequence of SEQ ID NO: 66; and (b) A second antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 50 and an LCVR containing the amino acid sequence of SEQ ID NO:

66.

23. The isolated bispecific antigen-binding molecule of claim 19, comprising: (a) A first antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 42 and an LCVR containing the amino acid sequence of SEQ ID NO: 18; and (b) A second antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 2 and an LCVR containing the amino acid sequence of SEQ ID NO:

18.

24. The isolated bispecific antigen-binding molecule of claim 20, comprising: (a) A first antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 32 and an LCVR containing the amino acid sequence of SEQ ID NO: 18; and (b) A second antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 2 and an LCVR containing the amino acid sequence of SEQ ID NO:

18.

25. The isolated bispecific antigen-binding molecule of claim 21, comprising: (a) A first antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 10 and an LCVR containing the amino acid sequence of SEQ ID NO: 18; and (b) A second antigen-binding domain comprising an HCVR containing the amino acid sequence of SEQ ID NO: 2 and an LCVR containing the amino acid sequence of SEQ ID NO:

18.

26. An isolated bispecific antigen-binding molecule that competes with a reference antibody for binding to PD-L1 or to the same epitope on PD-L1, wherein the reference antibody comprises a first antigen-binding domain and a second antigen-binding domain, the first antigen-binding domain comprising an HCVR / LCVR pair containing an amino acid sequence selected from SEQ ID NO: 58 / 66, 42 / 18, 32 / 18 and 10 / 18, and the second antigen-binding domain comprising an HCVR / LCVR pair containing an amino acid sequence selected from SEQ ID NO: 50 / 66 and 2 / 18.

27. An isolated bispecific antigen-binding molecule that competes with a reference antibody for binding to human 4-1BB or to the same epitope on human 4-1BB, wherein the reference antibody comprises a first antigen-binding domain and a second antigen-binding domain, the first antigen-binding domain comprising an HCVR / LCVR pair containing an amino acid sequence selected from SEQ ID NO: 58 / 66, 42 / 18, 32 / 18 and 10 / 18, and the second antigen-binding domain comprising an HCVR / LCVR pair containing an amino acid sequence selected from SEQ ID NO: 50 / 66 and 2 / 18.

28. The isolated bispecific antigen-binding molecule according to any one of claims 1 to 27, wherein it is a human bispecific antigen-binding molecule.

29. The isolated bispecific antigen-binding molecule according to any one of claims 1 to 28, which is a bispecific antibody. 30.29 The isolated bispecific antigen-binding molecule, wherein the antibody comprises a human IgG heavy chain constant region linked to HCVR of each of the first antigen-binding domain and the second antigen-binding domain, respectively.

31. The isolated bispecific antigen-binding molecule of claim 30, wherein the heavy chain constant region is isotype IgG1.

32. The isolated bispecific antigen-binding molecule of claim 30, wherein the heavy chain constant region is isotype IgG4.

33. The isolated bispecific antigen-binding molecule of any one of claims 30 to 32, wherein the heavy chain constant region connected to the HCVR of the first antigen-binding domain or the heavy chain constant region connected to the HCVR of the second antigen-binding domain contains an amino acid modification that reduces protein A binding, but does not simultaneously contain, the reduction in protein A binding being relative to the same isotype of heavy chain without said modification.

34. The isolated bispecific antigen-binding molecule of claim 33, wherein the modification comprises an H435R substitution (EU number) in the heavy chain of isotype IgG1 or IgG4.

35. The isolated bispecific antigen-binding molecule of claim 33, wherein the modification comprises an H435R substitution and a Y436F substitution (EU number) in the heavy chain of isotype IgG1 or IgG4.

36. The isolated bispecific antigen-binding molecule of any one of claims 29 to 35, wherein the bispecific antibody comprises a chimeric hinge that reduces the binding of the Fcγ receptor relative to a wild-type hinge of the same isotype.

37. The isolated bispecific antigen-binding molecule of any one of claims 29 to 36, wherein the antibody comprises a first heavy chain of HCVR containing the first antigen-binding domain and a second heavy chain of HCVR containing the second antigen-binding domain, wherein the first heavy chain comprises an amino acid sequence selected from SEQ ID NO: 76, 48, 40 and 28; and the second heavy chain comprises an amino acid sequence selected from SEQ ID NO: 74 and 26.

38. The isolated bispecific antigen-binding molecule of claim 37, wherein the antibody comprises a common light chain of LCVRs containing the first antigen-binding domain and the second antigen-binding domain, wherein the common light chain comprises an amino acid sequence selected from SEQ ID NO: 78 and 30.

39. The isolated bispecific antigen-binding molecule of any one of claims 27 to 36, wherein the antibody comprises a first heavy chain of HCVR containing the first antigen-binding domain and a second heavy chain of HCVR containing the second antigen-binding domain, wherein the first heavy chain comprises the amino acid sequence of SEQ ID NO: 76 and the second heavy chain comprises the amino acid sequence of SEQ ID NO:

74.

40. The isolated bispecific antigen-binding molecule of claim 39, wherein the antibody comprises a common light chain of LCVR containing the first antigen-binding domain and the second antigen-binding domain, wherein the common light chain comprises the amino acid sequence of SEQ ID NO:

78.

41. A bispecific antibody comprising a first antigen-binding domain specifically binding to human 4-1BB and a second antigen-binding domain specifically binding to human PD-L1, wherein the bispecific antibody comprises a first heavy chain containing the amino acid sequence of SEQ ID NO: 76, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 78; and a second heavy chain containing the amino acid sequence of SEQ ID NO: 74, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO:

78.

42. A bispecific antibody comprising a first antigen-binding domain specifically binding to human 4-1BB and a second antigen-binding domain specifically binding to human PD-L1, wherein the bispecific antibody comprises a first heavy chain containing the amino acid sequence of SEQ ID NO: 48, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 30; and a second heavy chain containing the amino acid sequence of SEQ ID NO: 26, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO:

30.

43. A bispecific antibody comprising a first antigen-binding domain specifically binding to human 4-1BB and a second antigen-binding domain specifically binding to human PD-L1, wherein the bispecific antibody comprises a first heavy chain containing the amino acid sequence of SEQ ID NO: 40, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 30; and a second heavy chain containing the amino acid sequence of SEQ ID NO: 26, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO:

30.

44. A bispecific antibody comprising a first antigen-binding domain specifically binding to human 4-1BB and a second antigen-binding domain specifically binding to human PD-L1, wherein the bispecific antibody comprises a first heavy chain containing the amino acid sequence of SEQ ID NO: 28, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO: 30; and a second heavy chain containing the amino acid sequence of SEQ ID NO: 26, which is paired with a common light chain containing the amino acid sequence of SEQ ID NO:

30.

45. The bispecific antibody according to any one of claims 41 to 44, wherein it is a human antibody.

46. ​​A pharmaceutical composition comprising the bispecific antigen-binding molecule of any one of claims 1 to 40 and a pharmaceutically acceptable carrier or diluent.

47. A pharmaceutical composition comprising the bispecific antibody of any one of claims 41 to 45 and a pharmaceutically acceptable carrier or diluent.

48. A method for preparing the bispecific antigen-binding molecule according to any one of claims 1 to 40, comprising: (a) Introducing one or more nucleic acid molecules comprising a nucleic acid sequence encoding an immunoglobulin chain of the bispecific antigen-binding molecule into a host cell; (b) culturing the host cell under conditions favorable for expression of the nucleic acid molecule; and (c) optionally, isolating the bispecific antigen-binding molecule or immunoglobulin chain from the host cell and / or the culture medium in which the host cell is cultured.

49. The method of claim 48, wherein the host cell is a Chinese hamster ovary (CHO) cell.

50. The method of claim 48 or 49, further comprising formulating the antigen-binding molecule into a pharmaceutical composition comprising an acceptable carrier.

51. An antigen-binding molecule or immunoglobulin chain, which is the product of the method of claim 48 or 49.

52. A nucleic acid molecule comprising a nucleotide sequence encoding a bispecific antigen-binding molecule as described in any one of claims 1 to 40; or a group of nucleic acid molecules comprising a nucleotide sequence encoding a first antigen-binding domain specifically binding to human 4-1BB, a second antigen-binding domain specifically binding to human PD-L1, and an LCVR as described in any one of claims 1 to 40.

53. An expression vector comprising the nucleic acid molecule of claim 52; or a set of expression vectors comprising the set of nucleic acid molecules of claim 52.

54. A host cell comprising the expression vector or group of expression vectors as described in claim 53.

55. The host cell of claim 54, wherein the host cell is a Chinese hamster ovary (CHO) cell.

56. A method for generating a bispecific antigen-binding molecule that binds to PD-L1 and 4-1BB, comprising: (a) The host cells of claim 54 are cultured under conditions conducive to the production of the bispecific antigen-binding molecules; (b) Optionally, the antigen-binding molecule or immunoglobulin chain is isolated from the host cell and / or the culture medium in which the host cell is cultured.

57. The method of claim 56, wherein the host is a CHO cell.

58. The method of claim 56 or 57, further comprising formulating the antigen-binding molecule into a pharmaceutical composition comprising an acceptable carrier.

59. An antigen-binding molecule or immunoglobulin chain, which is the product of the method of claim 56 or 57.

60. A nucleic acid molecule comprising a nucleotide sequence encoding a bispecific antibody as described in any one of claims 41 to 45; or a group of nucleic acid molecules comprising a nucleotide sequence encoding a heavy chain of a first antigen-binding domain specifically binding to human 4-1BB, a heavy chain of a second antigen-binding domain specifically binding to human PD-L1, and a light chain as described in any one of claims 41 to 45.

61. An expression vector comprising the nucleic acid molecule of claim 60; or a set of expression vectors comprising the set of nucleic acid molecules of claim 60.

62. A host cell comprising the expression vector or group of expression vectors as described in claim 61.

63. The host cell of claim 62, wherein the host cell is a Chinese hamster ovary (CHO) cell.

64. A method for generating bispecific antibodies that bind to PD-L1 and 4-1BB, comprising: (a) Culture the host cells of claim 62 under conditions favorable to the production of the bispecific antibodies; (b) Optionally, the bispecific antibody is isolated from the host cell and / or the culture medium in which the host cell is cultured.

65. The method of claim 64, wherein the host is a CHO cell.

66. The method of claim 64 or 65, further comprising formulating the bispecific antibody into a pharmaceutical composition comprising an acceptable carrier.

67. An antibody, which is the product of the method of claim 64 or 65.

68. A method for inhibiting tumor growth in a subject, comprising administering to the subject an isolated bispecific antigen-binding molecule of any one of claims 1 to 40, or a bispecific antibody of any one of claims 41 to 45, or a pharmaceutical composition of claim 46 or 47.

69. The method of claim 68, wherein the tumor is esophageal cancer, squamous cell carcinoma of the lung, adenocarcinoma of the lung, squamous cell carcinoma of the cervix, adenocarcinoma of the endometrium, urothelial carcinoma of the bladder, lung cancer, non-small cell lung cancer, colorectal cancer, rectal cancer, endometrial cancer, skin cancer, squamous cell carcinoma of the head and neck, brain cancer, glioblastoma multiforme, breast cancer, gastroesophageal cancer, gastroesophageal adenocarcinoma, hepatocellular carcinoma, prostate cancer, ovarian cancer, B-cell cancer, T-cell cancer, leukemia, pancreatic cancer, colon cancer, melanoma, basal cell carcinoma, cervical cancer, diffuse large B-cell lymphoma, or multiple myeloma.

70. The method of claim 68 or 69, wherein the tumor expresses PD-L1.

71. The method of any one of claims 68 to 70, further comprising administering a second therapeutic agent or treatment regimen.

72. The method of claim 71, wherein the second therapeutic agent or treatment regimen comprises a chemotherapeutic agent, a DNA alkylating agent, an immunomodulator, a proteasome inhibitor, a histone deacetylase inhibitor, radiotherapy, surgery, stem cell transplantation, a bispecific antibody interacting with tumor-associated antigens (TAAs) and T-cell or immune cell antigens, an antibody-drug conjugate, an oncolytic virus, a bispecific antibody conjugated to an antitumor agent, a VEGF inhibitor, a checkpoint inhibitor, a GITR agonist, a CD27 agonist, a 4-1BB activator, a PD-1 inhibitor, a CTLA-4 inhibitor, an EGFR inhibitor, an Ang2 inhibitor, a MUC16 inhibitor, a cancer vaccine, a cytokine, modified IL2, modified IL12, an IL4 inhibitor, an IL6 inhibitor, a corticosteroid, or a combination thereof.

73. The method of claim 72, wherein the T cell or immune cell antigen is CD3.

74. The method of claim 72 or 73, wherein the TAA is selected from AFP, ALK, BAGE protein, BCMA, BIRC5 (survival protein), BIRC7, β-linkin, brc-abl, BRCA1, BORIS, CA9, carbonic anhydrase IX, caspase-8, CALR, CCR5, CD19, CD20 (MS4A1), CD22, CD30, CD40, CDK4, CEA, CTLA4, cyclin-B1, CYP1B1, EGFR, EGFRvIII, ErbB2 / Her2, ErbB3, ErbB4, ETV6-AML, EpCAM, EphA2, Fra-1, FOLR1, GAGE ​​protein (e.g., GAGE-1, -2), GD2, GD3, GloboH, phosphatidylinositol proteoglycan-3, GM3, gp100, Her2, HLA / B-raf, HLA / k-ras, H LA / MAGE-A3, hTERT, LMP2, MAGE proteins (e.g., MAGE-1, -2, -3, -4, -6, and -12), MART-1, mesothelin, ML-IAP, Muc1, Muc2, Muc3, Muc4, Muc5, Muc16 (CA-125), MUM1, NA17, NY-BR1, NY-BR62, NY-BR85, NY-ESO1, OX40, p15, p53, P AP, PAX3, PAX5, PCTA-1, PLAC1, PRLR, PRAME, PSMA (FOLH1), RAGE protein, Ras, RGS5, Rho, SART-1, SART-3, STEAP1, STEAP2, TAG-72, TGF-β, TMPRSS2, Thompson-nouvelle antigen (Tn), TRP-1, TRP-2, tyrosinase, and urinary plaque protein-3.

75. Use of the bispecific antigen-binding molecule of any one of claims 1 to 40, or the bispecific antibody of any one of claims 41 to 45, or the pharmaceutical composition of claim 46 or 47 in the treatment of tumors.

76. The use according to claim 75, wherein the tumor is esophageal cancer, squamous cell carcinoma of the lung, adenocarcinoma of the lung, squamous cell carcinoma of the cervix, adenocarcinoma of the endometrium, urothelial carcinoma of the bladder, lung cancer, non-small cell lung cancer, colorectal cancer, rectal cancer, endometrial cancer, skin cancer, squamous cell carcinoma of the head and neck, brain cancer, glioblastoma multiforme, breast cancer, gastroesophageal cancer, gastroesophageal adenocarcinoma, hepatocellular carcinoma, prostate cancer, ovarian cancer, B-cell cancer, T-cell cancer, leukemia, pancreatic cancer, colon cancer, melanoma, basal cell carcinoma, cervical cancer, diffuse large B-cell lymphoma, or multiple myeloma.

77. The use as described in claim 75 or 76, wherein the tumor expresses PD-L1.

78. The use according to any one of claims 75 to 77, wherein the antigen-binding molecule or pharmaceutical composition is used in combination with a second therapeutic agent or treatment regimen, the second therapeutic agent or treatment regimen comprising a chemotherapeutic agent, a DNA alkylating agent, an immunomodulator, a proteasome inhibitor, a histone deacetylase inhibitor, radiotherapy, surgery, stem cell transplantation, a bispecific antibody interacting with a tumor-associated antigen (TAA) and a T-cell or immune cell antigen, an antibody-drug conjugate, an oncolytic virus, a bispecific antibody conjugated to an antitumor agent, a VEGF inhibitor, a checkpoint inhibitor, a GITR agonist, a CD27 agonist, a 4-1BB activator, a PD-1 inhibitor, a CTLA-4 inhibitor, an EGFR inhibitor, an Ang2 inhibitor, a MUC16 inhibitor, a cancer vaccine, a cytokine, a modified IL2, a modified IL12, an IL4 inhibitor, an IL6 inhibitor, a corticosteroid, or a combination thereof.

Images