Anti-PD-1 antibody antigen-binding domain and immunoconjugate

Abstract

The present invention relates to immunoconjugates comprising an anti-PD-1 antibody binding domain that does not disrupt signaling between PD1 and PD-L1, and an immunomodulatory domain targeting the same to PD-1-expressing cells.

Description

[Technical Field]

[0001] The present invention relates to anti-PD-1 antibody antigen-binding domains that are compatible with checkpoint inhibitors configured to disrupt signaling between PD-1 and its ligands. The invention further relates to immunoconjugates comprising said anti-PD-1 binding domains, and methods of using them to treat cancer patients, optionally in combination with PD-1 or PD-L1 antagonist antibodies. [Background technology]

[0002] The efficacy of antitumor activity of immunotherapeutic molecules, such as recombinant cytokines, can be improved by including a domain that targets the molecule to where it is needed within the tumor, or to the tumor microenvironment. For example, IL-2 constructs with tumor-specific antibody binding domains have been shown to have synergistic effects when used in combination with checkpoint inhibitor antibodies targeting PD-L1 to treat cancer models (Klein C. 2017 Oncoimmunology 6:e1277306). Alternatively, IL-2 molecules can be targeted to specific immune cells with antitumor properties, for example, by using targeting antibodies that bind to CD8, PD-1, or other antigens expressed on the surface of CD8 T cells. Because these latter constructs simultaneously present IL-2 and bind to the surface of T cells, they are often referred to as "cis-signaling" IL-2 bispecifics or immunoconjugates. A "cis-signaling" anti-PD-1 / IL-2 bispecific fusion protein (WO2018184964A1, Deak LC et al. 2022, Nature 610:161) was shown to have superior efficacy compared to untargeted IL-2 in mouse preclinical models.

[0003] In known anti-PD-1 / IL-2 constructs, the targeting arm consists of a blocking anti-PD-1 antibody (blocking the PD-1 / PD-1L interaction). Such a targeting arm has the potential advantage of alleviating PD-L1-induced T cell suppression and delivering a proliferative IL-2 signal to the same effector cells. However, there are inherent pharmacological challenges in administering blocking therapeutics (requiring dosing that provides full target coverage) and agonistic therapeutics (requiring dosing and frequency sufficient for safe and agonistic effects). Particularly in the case of PD-1 and IL-2 therapeutics, anti-PD-1 antibodies are administered at mg / kg doses to ensure full target coverage, whereas cytokines such as IL-2 are administered at microgram / kg doses to avoid overstimulation and on-target toxicity. This discrepancy has been addressed by mutating the IL-2 portion of the molecule to reduce its affinity for its receptor, thereby enabling higher doses of the bispecific. However, mutations in the IL-2 structure can lead to the generation of anti-drug antibodies, leading to treatment failure and toxicity. Furthermore, in patients receiving PD-1 checkpoint inhibitor (CPi) therapy, the epitope targeted by the PD-1 / IL-2 fusion protein is already occupied by the CPi, preventing the fusion protein from binding to target cells, making the drug incompatible with standard checkpoint inhibitor treatment regimes as an adjuvant. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] WO 2018 / 184964(A1) [Non-patent literature]

[0005] [Non-Patent Document 1] Klein C.2017 Oncoimmunology 6:e1277306 [Non-patent document 2] Deak LC et al. 2022, Nature 610:161 Summary of the Invention [Problem to be solved by the invention]

[0006] Based on the above prior art, it is an object of the present invention to provide a PD-1 targeting domain that is compatible with existing therapeutic agents that interrupt signaling between PD-1 and its ligands. [Means for solving the problem]

[0007] This object is achieved by the subject matter of the independent claims herein, with further advantageous embodiments described in the dependent claims herein, the examples, the figures and the general description. Summary of the Invention

[0008] The present inventors have developed a PD-1-targeting antibody antigen-binding domain (also referred to herein as an immunoglobulin variable domain) that does not inhibit the binding or antagonism of conventional PD-1- or PD-L1-targeting checkpoint inhibitor antibodies. The anti-PD-1 binding domain of a non-blocking antibody can be used as a targeting moiety for immunoconjugates to facilitate targeted delivery of immunomodulatory polypeptides, delivering IL-2 signals to appropriate exhausted CD8 T cells. This results in immunoconjugate compounds that effectively deliver payloads to PD-1-expressing cells, allowing for the controlled administration of potent active agents to achieve safe and effective cancer treatment.

[0009] A first aspect of the present invention relates to immunoglobulin (Ig) variable domains capable of binding to PD-1 in the presence of at least equimolar amounts of PD-1-specific checkpoint inhibitory antibodies. The interaction of an antibody or immunoconjugate comprising an Ig variable domain according to the invention with PD-1 is compatible with the simultaneous binding of an anti-PD-1 checkpoint inhibitory antibody and maintains their immunostimulatory effects. In certain embodiments, the binding of an antibody characterized by an immunoglobulin variable domain according to the invention is reduced by no more than 20% in the presence of a 100-fold molar excess of an anti-PD-1 checkpoint inhibitory antibody (e.g., pembrolizumab or nivolumab). In certain embodiments, such an antibody or immunoconjugate comprising an Ig variable domain according to the invention is capable of binding to PD-1 in an amount of 1.0×10 -9 Affinity constant (K D ) are characterized by high affinity for PD-1.

[0010] The PD-1-specific non-blocking Ig variable domain is composed of both an antibody heavy chain variable domain polypeptide (PD1-VH) and an antibody light chain variable domain polypeptide (PD1-VL), which associate as a heterodimer to provide a functional antibody antigen-binding domain.

[0011] In some embodiments, the sequence of PD1-VH comprises the following heavy chain complementarity-determining regions (HCDRs) 1-3: HCDR1 having the sequence of SEQ ID NO: 118, HCDR2 having the sequence of SEQ ID NO: 119, and HCDR3 having the sequence of SEQ ID NO: 120, and PD1-VL comprises light chain complementarity-determining regions (LCDRs): LCDR1 having the sequence of SEQ ID NO: 121, LCDR2 having the sequence of SEQ ID NO: 122, and LCDR3 having the sequence of SEQ ID NO: 123.

[0012] In some embodiments, the PD1-VH comprises an HCDR1 having the sequence of SEQ ID NO: 127, an HCDR2 having the sequence of SEQ ID NO: 165, and an HCDR3 having the sequence of SEQ ID NO: 129, and the PD1-VL is characterized by an LCDR1 having the sequence of SEQ ID NO: 130, an LCDR2 having the sequence of SEQ ID NO: 166, and an LCDR3 having the sequence of SEQ ID NO: 132.

[0013] In some embodiments, PD1-VH and PD1-VL have amino acid sequences identical to or similar to SEQ ID NOs: 085, 086, 087, or 091; or SEQ ID NOs: 071 and 072; or SEQ ID NOs: 073 and 074; or SEQ ID NOs: 075 and 076; or SEQ ID NOs: 077 and 078; or SEQ ID NOs: 079 and 080; or SEQ ID NOs: 081 and 082; or SEQ ID NOs: 083 and 084; or SEQ ID NOs: 061 and 062; or SEQ ID NOs: 067 and 068, and share the same biological function.

[0014] A next aspect of the invention relates to an immunoconjugate comprising a non-blocking anti-PD-1 antigen-binding variable domain as defined in the above aspect of the invention, which further comprises an immunoactive polypeptide component that delivers a signaling effect to a target other than PD-1, such as an interleukin, or an interleukin receptor, particularly one that is also expressed on PD-1+ cells, such as T cells or natural killer cells.

[0015] In some embodiments, the immunoconjugate comprises a non-blocking PD-1 antibody variable domain and an interleukin, particularly a non-blocking PD-1 specific domain, linked to an Ig antibody binding domain specific for said interleukin, and an IL-2 polypeptide linked to an anti-IL-2 antibody binding domain as shown in the Examples, in various functional formats.

[0016] A further aspect of the invention pertains to isolated nucleic acid sequences, expression vectors, or cells that express or encode the anti-PD-1 antibody variable domains and / or immunoconjugates described herein.

[0017] The present invention further relates to a method for treating cancer patients by administering an effective amount of an immunoconjugate according to the invention, as well as to pharmaceutical compositions comprising the immunoconjugates according to the invention and their use in the treatment of cancer. Terms and Definitions

[0018] For the purposes of interpreting this specification, the following definitions shall apply, and where appropriate, terms used in the singular shall also include the plural and vice versa. In the event that a definition set forth below conflicts with any document incorporated herein by reference, the definition set forth herein shall control.

[0019] As used herein, the terms "comprising," "having," "containing," "including," and other similar forms, and their grammatical equivalents, are intended to be equivalent in meaning and to be open-ended in that the listing of one or more items following any one of these words does not imply an exhaustive listing of such one or more items, or that it is limited to only the listed item or items. For example, an item "comprising" components A, B, and C can consist of components A, B, and C (i.e., contain only components A, B, and C), or it can include not only components A, B, and C, but also one or more other components. Thus, "comprises" and its similar forms, and its grammatical equivalents, are intended and understood to include disclosure of "consisting essentially of" or "consisting of" embodiments.

[0020] Where a range of values ​​is provided, unless the context clearly dictates otherwise, it is understood that each intervening value, to the tenth of the unit of the lower limit, between the upper and lower limit of that range, and any other stated or intervening value within that stated range, is encompassed within the disclosure, subject to any specifically excluded limit in the stated range. Where one or both of the limits are included in the stated range, ranges excluding either or both of those included limits are also included in the disclosure.

[0021] As used herein, reference to "about" a value or parameter includes (and describes) a variation on the value or parameter itself. For example, a statement referring to "about X" also includes the statement "X."

[0022] As used in this specification, including the appended claims, the singular forms "a," "or," and "the" include plural references unless the context clearly dictates otherwise.

[0023] As used herein, "and / or" is considered to specifically describe each of the two specified features or components with or without the other features or components. Thus, the term "and / or" used in phrases such as "A and / or B" is intended to include "A and B," "A or B," "A alone," and "B alone." Similarly, the term "and / or" used in phrases such as "A, B, and / or C" is intended to encompass each of the following aspects: A, B, and C; A, B, or C; A or C; A or B; B or C; A and C; A and B; B and C; A alone; B alone; and C alone.

[0024] 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 (e.g., cell culture, molecular genetics, nucleic acid chemistry, hybridization techniques and biochemistry, organic synthesis). Standard techniques are used for molecular, genetic, and biochemical techniques (see generally, Sambrook et al., Molecular Cloning: A Laboratory Manual, 4th Ed (2012) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, and Ausubel et al., Short Protocols in Molecular Biology (2002) 5th Ed, John Wiley & Sons, Inc.), and chemical techniques.

[0025] The term "PD-1" in the context of the present specification relates to the human PD-1 protein, which is sometimes also called CD279 (Uniprot Q15116), encoded by the "PDCD1" gene.

[0026] The term "PD-1" in the context of the present specification relates to the human PD-L1 protein, which is sometimes also called CD274 (Uniprot Q9NZQ7), encoded by the gene "CD274".

[0027] In the context of this specification, the term "immunoconjugate" refers to a recombinant polypeptide molecule comprising at least one anti-PD-1 antibody binding domain disclosed herein, covalently linked to at least one additional moiety capable of binding to a polypeptide ligand expressed on the cell surface of an immune cell and initiating a downstream effect. That is, the anti-PD-1 binding domain of the "immunoconjugate" functions as a targeting domain, and binding of the second moiety to the cell initiates a downstream function in the target immune cell. Examples of such second moieties that initiate a downstream function via a polypeptide expressed on an immune cell include an interleukin, the extracellular domain of a costimulatory ligand, or an agonist / antagonist antibody (or antigen-binding fragment of an antibody, or antibody-like molecule). Examples of targets to which the second moiety can bind on the cell surface include, for example, cytokine or chemokine receptors, integrins, antigen receptors, and / or costimulatory molecules. In certain embodiments, the second moiety capable of binding to a polypeptide ligand present on the surface of an immune cell comprises an interleukin. A downstream function elicited by a second moiety capable of binding to and signaling an immune cell refers to a function other than Fc receptor binding or PD-1 binding provided by a monoclonal antibody having an Ig variable domain according to the invention. However, an immunoconjugate as used herein may further comprise an Fc portion, such as a bispecific antibody in which one antibody arm is an anti-PD-1 variable domain according to the first aspect of the invention and a second arm has a second specificity, and such an immunoconjugate may further have an Fc receptor-mediated effect.

[0028] The term "IL-2" in the context of this specification relates to human IL-2 and functional variants thereof, such as the wild-type human amino acid sequence (Uniprot P60568), and variant proteins as set out in Table 2.

[0029] The term "IL2CP" or "IL2-CP" in the context of this specification relates to a circularly permuted IL-2 polypeptide created by "opening up" the IL-2 polypeptide sequence and fusing the native N- and C-terminal amino acid residues to create new N'- and C'-termini. This generates a rearranged IL2CP polypeptide while retaining the critical tertiary structure of the cytokine that enables it to signal through its dedicated receptor.

[0030] The term "dimeric IL-2 receptor" in the present context relates to a heterodimeric receptor comprising two receptor chains CD122 and CD132.

[0031] Any patent documents cited herein are deemed to be incorporated herein by reference in their entirety.

[0032] array Sequences similar or identical (e.g., at least about 70% sequence identity) to the sequences disclosed herein are also part of the present invention. In some embodiments, sequence identity at the amino acid level can be about 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or more. At the nucleic acid level, sequence identity can be about 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or more. Alternatively, substantial identity exists when the nucleic acid segment hybridizes under selective hybridization conditions (e.g., under very high stringency hybridization conditions), to the complement of the strand. The nucleic acid can be present in whole cells, in a cell lysate, or in a partially purified or substantially pure form.

[0033] In the present context, the terms "sequence identity" and "percentage of sequence identity" refer to a quantitative parameter that represents the results of sequence comparison, which is determined by comparing the two aligned sequences position by position. Methods for aligning sequences for comparison are well known to those skilled in the art. Alignment of sequences for comparison can be performed using the local homology algorithm of Smith and Waterman, Adv. Appl. Math. 2:482 (1981), the global alignment algorithm of Needleman and Wunsch, J. Mol. Biol. 48:443 (1970), the similarity search method of Pearson and Lipman, Proc. Nat. Acad. Sci. 85:2444 (1988), or computerized implementations of these algorithms, including, but not limited to, CLUSTAL, GAP, BESTFIT, BLAST, FASTA, and TFASTA. Software for performing BLAST analysis is publicly available from the National Center for Biotechnology Information (http: / / blast.ncbi.nlm.nih.gov / ) and other sources.

[0034] An example of a comparison of amino acid sequences is the BLASTP algorithm using default settings: expected threshold: 10; word size: 3; maximum match within query range: 0; matrix: BLOSUM62; gap cost; existence 11, extension 1; composition adjustment: conditional composition score matrix adjustment. An example of such a comparison of nucleic acid sequences is the BLASTN algorithm using default settings: expected threshold: 10; word size: 28; maximum match within query range: 0; match / mismatch score: 1-2; gap cost: linear. Unless otherwise specified, the sequence identity values ​​provided herein refer to the values ​​obtained using the BLAST series of programs (Altschul et al., J. Mol. Biol. 215: 403-410 (1990)) using the default parameters identified above for protein and nucleic acid comparisons, respectively.

[0035] Reference to identical sequences without specifying a percentage value means 100% identical sequences (ie, the same sequence).

[0036] General biochemistry: peptides, amino acid sequences The term "polypeptide" in the context of this specification refers to a molecule consisting of 50 or more amino acids forming a linear chain, where the amino acids are connected by peptide bonds. The amino acid sequence of a polypeptide may represent the amino acid sequence of an entire protein (as found physiologically) or a fragment thereof. The terms "polypeptide" and "protein" are used interchangeably herein and include proteins and fragments thereof. Polypeptides are disclosed herein as amino acid residue sequences.

[0037] The term "peptide" in the present context relates to a molecule consisting of up to 50 amino acids, in particular 8 to 30 amino acids, more in particular 8 to 15 amino acids, which form a linear chain in which the amino acids are connected by peptide bonds.

[0038] The sequence of amino acid residues is written from the amino terminus to the carboxyl terminus. Capital letters at sequence positions refer to L-amino acids using the single-letter code (Stryer, Biochemistry, vol. 3, p. 21). Lowercase letters at amino acid sequence positions refer to the corresponding D- or (2R)-amino acids. The sequence is written from left to right from the amino terminus to the carboxyl terminus. According to standard nomenclature, amino acid residue sequences are represented by either three-letter or one-letter codes as follows: alanine (Ala, A), arginine (Arg, R), asparagine (Asn, N), aspartic acid (Asp, D), cysteine ​​(Cys, C), glutamine (Gln, Q), glutamic acid (Glu, E), glycine (Gly, G), histidine (His, H), isoleucine (Ile, I), leucine (Leu, L), lysine (Lys, K), methionine (Met, M), phenylalanine (Phe, F), proline (Pro, P), serine (Ser, S), threonine (Thr, T), tryptophan (Trp, W), tyrosine (Tyr, Y), and valine (Val, V).

[0039] The term "variant" refers to a polypeptide that differs from a reference polypeptide but retains essential properties. A typical variant of a polypeptide differs from another, reference polypeptide in its primary amino acid sequence. Generally, differences are limited so that the sequences of the reference polypeptide and the variant are similar overall and, in many regions, identical. A variant and reference polypeptide may differ in amino acid sequence by one or more modifications (e.g., substitutions, additions, and / or deletions). A substituted or inserted amino acid residue may or may not be one encoded by the genetic code. A polypeptide variant may be naturally occurring, such as an allelic variant, or it may be a variant that is not known to occur naturally.

[0040] In the present context, the term "dimer" refers to a unit consisting of two subunits. In the present context, the term "heterodimer" refers to a dimer composed of two non-identical subunits. The term "heterotetramer" refers to a compound containing four polypeptides, none of which are identical.

[0041] The term "heterotetrameric Ig" or "heterotetrameric Ig" refers to a recombinant immunoglobulin-like molecule comprising a first heavy and light chain pair characterized by a first antibody binding domain specific for a first antigen, and a second heavy and light chain pair characterized by a second antibody binding domain specific for a second associated antigen. These are sometimes referred to as bispecific antibodies, and include heterotetrameric kappa / lambda formats and crossmab formats (WO2009080253).

[0042] In the present context, the term "amino acid linker" or "peptide linker" refers to a polypeptide of variable length that is used to join two polypeptides to produce a single polypeptide chain. Exemplary embodiments of linkers useful in practicing the invention defined herein are oligopeptide chains of 1, 2, 3, 4, 5, 10, 20, 30, 40, or 50 amino acids.

[0043] There are no restrictions on the amino acid composition of the linker. In one embodiment, the linker is composed of amino acids selected from the group G, S, A, and D. Important features of the conjugate peptide linkers defined herein are low immunogenicity and a peptide length that allows the domains connected by the linker to interact to form a functional entity, such as an immunoconjugate as disclosed herein. In particularly preferred embodiments of the above-defined domain peptide linkers, the sequence is composed primarily of stretches of small polar amino acids, such as glycine (G) and serine (S).

[0044] In one embodiment, the peptide linker is at least (≧) 15 amino acids in length, particularly 15-30 amino acids in length, and the amino acids are selected from G, S, A and D.

[0045] Non-limiting examples of amino acid linkers are monomers, dimers, trimers or tetramers of a peptide motif consisting of three or four glycines and one serine.

[0046] Any of the embodiments of peptide linkers disclosed herein encompass structures in which amino acids with similar properties can be substituted, e.g., amino acids V, L, I, P, S, C, or M can be substituted with G, S, or S, and D can be substituted with E.

[0047] Specific non-limiting examples of linkers are provided by SEQ ID NOs: 014, 015, 016, 017, 018, 019, 020, 021, 022, 023, and 024.

[0048] General molecular biology: nucleic acid sequence, expression The term "recombinant" in the context of the present specification relates to a nucleic acid that is the product of one or more steps of cloning, restriction and / or ligation and that differs from a naturally occurring nucleic acid. A recombinant viral particle comprises a recombinant nucleic acid.

[0049] The terms "gene expression" or "expression," or "gene product," may refer to either or both the process—and product—of producing a nucleic acid (RNA) or producing a peptide or polypeptide, also referred to as transcription and translation, respectively, or any intermediate process that regulates the processing of genetic information to produce a polypeptide product. The term "gene expression" can also apply to the transcription and processing of an RNA gene product, such as regulatory RNA or structural (e.g., ribosomal) RNA. When the expressed polynucleotide is derived from genomic DNA, expression can include splicing of mRNA in eukaryotic cells. Expression can be assessed both at the level of transcription and translation, i.e., mRNA and / or protein product.

[0050] The term "nucleic acid expression vector" in the context of this specification refers to a plasmid, viral genome, or RNA, which is used to transfect (in the case of a plasmid or RNA) or transduce (in the case of a viral genome) a gene of interest into a target cell, or, in the case of a transfected RNA construct, to translate the corresponding protein of interest from the transfected mRNA. In vectors that operate at the level of transcription and subsequent translation, the gene of interest is under the control of a promoter sequence that is operable in the target cell, and therefore the gene of interest is transcribed constitutively, in response to a stimulus, or depending on the state of the cell. In certain embodiments, a viral genome is packaged into a capsid, resulting in a viral vector that can transduce a target cell.

[0051] Binding; binders, ligands, antibodies: The term "specific binding" in the context of the present invention refers to the property of a ligand to bind to a target with a particular affinity and target specificity. The affinity of such a ligand is indicated by the dissociation constant of the ligand. A specifically reactive ligand has a dissociation constant of 10 or less when bound to a target. -7 mol / L or less (especially 10 -9 mol / L or less), but when interacting with a molecule that has nearly the same chemical composition as the target but a different conformation, the dissociation constant is at least three orders of magnitude higher.

[0052] In the context of this specification, the term "dissociation constant (K D The term "equilibrium constant" is used in the sense known in the arts of chemistry and physics; it refers to the equilibrium constant that measures the tendency of a complex consisting of [often two] different components to reversibly dissociate into its constituent components. The complex may be, for example, an antibody-antigen complex AbAg consisting of an antibody Ab and an antigen Ag. Dis expressed in molar concentration [mol / L] and corresponds to the concentration of [Ab] at which half of the binding sites for [Ag] are occupied, i.e., the concentration of unbound [Ab] is equal to the concentration of the [AbAg] complex. The dissociation constant can be calculated using the following formula:

number

[0053] In the context of this specification, "off-rate (K off ;[1 / s])" and "on-rate (K on ;[L / (sec * The terms "antibody dissociation (K mol)" and "antibody dissociation (K mol)" are used in the sense known in the art of chemistry and physics; they refer to the dissociation of an antibody from its target antigen (K off ) or meeting (K on ) refers to the rate constant that measures the off and K. on can be determined experimentally using well-established methods by those skilled in the art. off and K. on Surface plasmon resonance is employed as a method for measuring the dissociation constant K, which is the principle behind biosensor systems such as the Biacore® or ProteOn® systems. These also measure the dissociation constant K using the following equation: D can be used to determine:

number

[0054] Binding rate K on The natural upper limit of 9 L / (seconds * mol).

[0055] In the context of this specification, the term "antibody" relates to whole antibodies, including but not limited to immunoglobulin G (IgG), A (IgA), D (IgD), E (IgE) or M (IgM), any antigen-binding fragment thereof, or single chains thereof, and related or derived constructs. A whole antibody is a glycoprotein comprising at least two heavy (H) chains and two light (L) chains interconnected by disulfide bonds. Each heavy chain is composed of a heavy chain variable region (VH) and a heavy chain constant region (CH). The heavy chain constant region of IgG is C H 1. C H 2 and C H Each light chain is composed of three domains: a light chain variable region (abbreviated herein as VL) and a light chain constant region (CL). The light chain constant region is composed of one domain, CL. The variable regions of the heavy and light chains associate to create an antibody-binding domain that interacts with an antigen. The binding specificity of an antibody is primarily determined by its complementarity-determining regions (CDRs). CDRs are sometimes referred to by their position on the light chain (LCDR, CDR-L) or heavy chain (HCDR, CDR-H) and their number 1, 2, or 3, which designates their position in the variable domain region. The numbering system used herein to identify amino acid residues in CDR regions is shown in Figure 1. The constant region of an antibody may contain various cells of the immune system (e.g., effector cells) and the first component of the classical complement system, mediating the binding of the immunoglobulin to host tissues or factors. The term antibody encompasses so-called nanobodies or single domain antibodies, antibody fragments consisting of a single monomeric variable antibody domain.

[0056] In the context of this specification, the term "fragment crystallizable (Fc) region" is used in the sense known in the art of cell biology and immunology; when applied to IgG, it refers to the C region covalently linked by disulfide bonds. H 2 and C H The term "Fab fragment" refers to the portion of an antibody that contains two identical heavy chain fragments consisting of three domains. The term "Fab fragment" refers to the region of an antibody that binds to an antigen, and contains the constant and variable domains of a pair of heavy and light chains.

[0057] The term "immunoglobulin variable domain" in the context of this specification means the combination of a heavy chain variable domain sequence and a light chain variable domain polypeptide sequence which, when assembled, provides a functional antibody antigen-binding domain specific for a non-blocking PD-1 epitope of the invention.

[0058] In the present context, the term "single-chain variable fragment (scFv)" refers to a fusion protein in which the variable region of the light chain and the variable region of the heavy chain variable domain of a monoclonal antibody are linked by a flexible peptide linker. Multimeric "scFv" can be produced by linking two or more pairs of heavy and light chain domains with a peptide linker.

[0059] In the context of this specification, the term "humanized antibody" refers to an antibody originally produced by immune cells of a non-human species whose protein sequence has been modified to increase its similarity to antibody variants naturally produced in humans. As used herein, the term "humanized antibody" includes antibodies in which CDR sequences derived from the germline of another mammalian species, such as a mouse, have been grafted onto human framework sequences. Additional framework region modifications may be made within the human framework sequences as well as within the CDR sequences derived from the germline of another mammalian species.

[0060] (Cancer) Immunotherapy In the context of this specification, the term "anti-PD-1 antagonist antibody" is intended to encompass agents, in particular antibodies (or antibody-like molecules), that are capable of interrupting the signal cascade that leads to T cell inhibition after T cell activation as part of an immune checkpoint mechanism known to those skilled in the art. Non-limiting examples include clinically available antibody drugs such as nivolumab (Bristol-Myers Squibb; CAS No. 946414-94-4), pembrolizumab (Merck; CAS No. 1374853-91-4), dostallimab (Tesaro; CAS No. 2022215-59-2), and sintilimab (Eli Lilly, InnoVenture). and antibodies against PD-1 (Uniprot Q15116), exemplified by tislelizumab (BeiGene; CAS No. 1858168-59-8), cemiplimab (CAS No. 1801342-60-8), cetrelimab (CAS No. 2050478-92-5), southernlimab (CAS No. 2206792-50-7), toripalimab (CAS No. 1924598-82-2), zelvalimab (CAS No. 2315361-37-4), or ezabenlimab (CAS No. 2249882-54-8).

[0061] As used herein, the term "pharmaceutical composition" refers to a compound of the present invention or a pharmaceutically acceptable salt thereof, together with at least one pharmaceutically acceptable carrier. In certain embodiments, a pharmaceutical composition according to the present invention is provided in a form suitable for topical, parenteral, or injectable administration.

[0062] As used herein, the term "pharmaceutically acceptable carrier" includes any solvents, dispersion media, coatings, surfactants, antioxidants, preservatives (e.g., antibacterial, antifungal), isotonic agents, absorption delaying agents, salts, preservatives, drugs, drug stabilizers, binders, excipients, disintegrants, lubricants, sweeteners, flavoring agents, dyes, and the like, and combinations thereof, as known to those of skill in the art (see, e.g., Remington: The Science and Practice of Pharmacy, ISBN 0857110624).

[0063] The term "cancer" as used in the context of this specification relates to malignant neoplastic diseases; the terms "cancer" and "malignant neoplastic diseases" are used interchangeably herein. Specific examples include carcinomas (cancers of epithelial origin), sarcomas (cancers of connective tissue origin), lymphomas and leukemias, germ cell-derived tumors, and blastomas. Certain alternatives of any of the aspects and embodiments disclosed herein are directed to the use of the compounds and compositions of the present invention in the treatment of solid tumors. Other alternatives of any of the aspects and embodiments disclosed herein are directed to the use of the combinations of the present invention in the treatment of liquid cancers such as myeloid or granulocytic leukemias, particularly AML, lymphocytic, lymphocytic, or lymphoblastic leukemias and lymphomas, polycythemia vera, or erythema.

[0064] As used herein, the term "treating" or "treatment" of any disease or disorder (e.g., cancer) refers, in one embodiment, to ameliorating the disease or disorder (e.g., slowing, halting, or reducing the progression of the disease or at least one of its clinical symptoms). In another embodiment, "treating" or "treatment" refers to alleviating or improving at least one physical parameter, including those that may not be discernible to the patient. In yet another embodiment, "treating" or "treatment" refers to modulating the disease or disorder, either physically (e.g., stabilization of a discernible symptom), physiologically (e.g., stabilization of a physical parameter), or both. Methods for assessing the treatment and / or prevention of a disease are generally known to those of skill in the art, unless otherwise specified herein below.

[0065] Detailed Description of the Invention Non-blocking immunoglobulin antigen-binding domain specific for PD-1 A first aspect of the invention is an immunoglobulin (Ig) variable domain capable of binding to PD-1 in the presence of at least an equimolar amount of a PD-1-specific checkpoint inhibitory antibody, as the interaction of an antibody or immunoconjugate comprising an Ig variable domain according to the invention with PD-1 is compatible with co-administration of anti-PD-1 checkpoint inhibitory antibodies in vitro and in vivo without significantly inhibiting their immunostimulatory function, as shown, for example, in the in vitro assays summarized in Table 7.

[0066] The non-blocking Ig variable domain capable of binding to PD-1 according to the present invention comprises both an antibody heavy chain variable domain polypeptide (PD1-VH) and an antibody light chain variable domain polypeptide (PD1-VL), which associate as a heterodimer to form a functional antibody antigen-binding domain.

[0067] In certain embodiments, the non-blocking properties of an Ig variable domain that binds to PD-1 in the presence of a PD-1 checkpoint inhibitor antibody are characterized by assessing the binding of an antibody having two heavy chains and two light chains forming two antibody PD-1 binding domains according to the invention. According to these embodiments, the binding of such an antibody to PD-1 expressed on the surface of a cell is reduced by 20% or less in the presence of a 100-fold molar excess of a PD-1 agonist selected from pembrolizumab or nivolumab.

[0068] The non-blocking properties of PD-1 checkpoint inhibitors of the PD-1-binding Ig variable domains of the present invention can be characterized by assessing their binding to PD-1 expressed by mammalian cells (e.g., Jurkat cells expressing PD-1 as shown in Example 3). Briefly, PD-1-expressing cells are incubated with serial dilutions of pembrolizumab or nivolumab at 4°C for 30 minutes. Without washing the cells, an antibody characterized by an immunoglobulin variable domain of the present invention is added at a fixed concentration of 100 nMol / L and labeled with a detectable label, such as biotin, or a fluorescent dye. Bound antibody is then detected by flow cytometry. To determine the percentage of binding inhibition by pembrolizumab or nivolumab, the mean fluorescence intensity (MFI) of the bound antibody is compared to that of a control sample containing no competitor, and the percentage of binding inhibition at the desired molar excess level of checkpoint inhibitor is identified.

[0069] In some embodiments of the Ig variable domain capable of binding to PD-1 according to the invention, it binds to PD-1 without significantly inhibiting the interaction of PD-1 with pembrolizumab, while in other embodiments it binds to PD-1 without significantly inhibiting the interaction of PD-1 with nivolumab.

[0070] The PD-1-binding Ig variable domains according to the present invention, or antibodies characterized by said binding domains, specifically bind to the human protein PD-1 expressed on the surface of cells. Preferably, to facilitate preclinical testing, the antibodies are also highly specific for primate PD-1. In certain embodiments, the affinity constant (K ) for PD-1 of a classical antibody structure (i.e., the association of two heavy and light chains to provide two of the immunoglobulin variable domains) characterized by the Ig variable domains according to the present invention is D ) is 1.0 x 10 as measured using surface plasmon resonance as described in the "Binding; Binders, Ligands, Antibodies" section above. -9 mol / L or less.

[0071] In some embodiments of an immunoglobulin variable domain capable of binding to PD-1 according to the invention, it is characterized by a PD1-VH comprising a heavy chain CDR (HCDR) 1 having the sequence GFTFSINAMT (SEQ ID NO: 118), an HCDR2 having the sequence TISGSGFSTYYADSLKGR (SEQ ID NO: 119), and an HCDR3 having the sequence EVYGDY (SEQ ID NO: 120). 1 SSNIGSX 2 X 3 VF(X 1 is N, S, Q, or A, and X 2 X 3 LCDR1 having the sequence SNNQRPS (SEQ ID NO: 122), and LCDR3 having the sequence AAWDDSLSIWVF (SEQ ID NO: 123).

[0072] In certain embodiments, LCDR1 is X 1 is S, Q, or A, and X 2 X 3 These CDRs correspond to derivatives of antibody clone 21A08, which has favorable non-blocking properties and high affinity for PD-1 (both human PD-1 and macaque). 1 Any modification of the site removes the NS deamidation site in this clone without compromising PD-1 binding, so that all alternatives produced maintained hPD-1 and cPD-1 binding (see Table 15). Alternatively, or in addition, X 2 X 3 The modification in may remove the N-glycosylation site present in LCDR1 of 21A08. All alternatives tested maintained favorable binding to PD-1 compared to the parent sequence (see Table 14).

[0073] In certain embodiments of the Ig variable domain according to the preceding paragraph, the binding affinity K for PD-1 of an antibody characterized by said Ig variable domain D is 1.0 × 10 as measured by SPR -9~1.5×10 -11 In certain embodiments, K D is 1.0 x 10 -10 ~1.5×10 -11 In more particular embodiments, K D is 5.0 x 10 -10 ~1.5×10 -11 This high affinity for PD-1 is in the K D is 1.28 x 10 -8 nM (Table 10) or 1.6 x 10 -8 This distinguishes the binding domains of the present invention from known PD-1 non-blocking clones such as XVT458, whose binding domains were in the range of 0.1 nM (Table 8).

[0074] In certain embodiments of variable domains according to the invention, related to the generated clone designated herein as 21A08P1, the PD1-VL comprises an LCDR1 having the sequence SGASSNIGSQSVF (SEQ ID NO: 124). In particular embodiments of the 21A08P1Ig variable domain, the PD1-VH comprises or consists of a polypeptide at least (≧) 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 085, and the PD1-VL comprises or consists of a polypeptide at least 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 086. In more particular embodiments, the PD1-VH comprises a polypeptide having the sequence of SEQ ID NO: 085, and the PD1-VL comprises a polypeptide having the sequence of SEQ ID NO: 086. In even more particular embodiments, the PD1-VH consists of the polypeptide of SEQ ID NO: 085, and the PD1-VL consists of the polypeptide of SEQ ID NO: 086.

[0075] A further embodiment of a 21A08Ig variable domain according to the invention relates to the clone designated herein as 21A08P2, wherein the PD1-VL comprises an LCDR1 having the sequence SGASSNIGSSSVF (SEQ ID NO: 125). In particular embodiments of the 21A08P2Ig variable domain, the PD1-VH comprises or consists of a polypeptide at least (≧) 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 085, and the PD1-VL comprises or consists of a polypeptide at least 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 087. In more particular embodiments, the PD1-VH comprises a polypeptide having the sequence of SEQ ID NO: 085, and the PD1-VL comprises a polypeptide having the sequence of SEQ ID NO: 087. In even more particular embodiments, the PD1-VH consists of the polypeptide of SEQ ID NO: 085, and the PD1-VL consists of the polypeptide of SEQ ID NO: 087.

[0076] A further embodiment of a 21A08Ig variable domain according to the invention relates to a derived clone designated herein as 21A08P3, wherein the PD1-VL comprises an LCDR1 having the sequence SGASSNIGSNAVF (SEQ ID NO: 126). In particular embodiments of the variable domain, the PD1-VH comprises or consists of a polypeptide at least (≧) 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 085, and the PD1-VL comprises or consists of a polypeptide at least 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 091. In more particular embodiments, the PD1-VH comprises a polypeptide having the sequence of SEQ ID NO: 085, and the PD1-VL comprises a polypeptide having the sequence of SEQ ID NO: 091. In even more particular embodiments, the PD1-VH consists of the polypeptide of SEQ ID NO: 085, and the PD1-VL consists of the polypeptide of SEQ ID NO: 091.

[0077] Another aspect of the present invention relates to Ig variable domains capable of binding to PD-1, wherein the PD1-VH comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 071, and the PD1-VL comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 072. In a specific embodiment, the polypeptide at least 95% similar to SEQ ID NO: 071 has retained HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 071, and the LCDR1, LCDR2, and LCDR3 of the polypeptide at least 95% similar to SEQ ID NO: 072 are identical to those annotated in SEQ ID NO: 072. Such similar PD1-VH and PD1-VL retain the binding epitope of clone 21A08, which has high specificity for PD-1 and does not inhibit PD-1 checkpoint inhibitor antibody binding, by not introducing any amino acid substitutions in the CDR regions (Figure 1).

[0078] Another aspect of the invention relates to Ig variable domains capable of binding to PD-1, wherein the PD1-VH comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 073, and the PD1-VL comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 074. In a specific embodiment, the polypeptide at least 95% similar to SEQ ID NO: 073 has retained HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 073, and the LCDR1, LCDR2, and LCDR3 of the polypeptide at least 95% similar to SEQ ID NO: 074 are identical to those annotated in SEQ ID NO: 074. Such similar PD1-VH and PD1-VL polypeptides retain the binding epitope of clone 22F13, which has high specificity for PD-1, and do not inhibit the binding of PD-1 checkpoint inhibitor antibodies.

[0079] Another aspect of the invention relates to Ig variable domains capable of binding to PD-1, wherein the PD1-VH comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 075, and the PD1-VL comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 076. In a specific embodiment, the polypeptide at least 95% similar to SEQ ID NO: 075 has retained HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 075, and the LCDR1, LCDR2, and LCDR3 of the polypeptide at least 95% similar to SEQ ID NO: 076 are identical to those annotated in SEQ ID NO: 076. Such similar PD1-VH and PD1-VL retain the binding epitope of clone 25I20, which has high specificity for PD-1 and does not inhibit PD-1 checkpoint inhibitor antibody binding.

[0080] Another aspect of the invention relates to Ig variable domains capable of binding to PD-1, wherein the PD1-VH comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 077, and the PD1-VL comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 078. In a specific embodiment, the polypeptide at least 95% similar to SEQ ID NO: 077 has conserved HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 077, and the LCDR1, LCDR2, and LCDR3 of the polypeptide at least 95% similar to SEQ ID NO: 078 are identical to those annotated in SEQ ID NO: 078. Such similar PD1-VH and PD1-VL retain the binding epitope of clone 20H02, which has high specificity for PD-1.

[0081] Another aspect of the invention relates to Ig variable domains capable of binding to PD-1, wherein the PD1-VH comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 079, and the PD1-VL comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 080. In a specific embodiment, the polypeptide at least 95% similar to SEQ ID NO: 079 has conserved HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 079, and the LCDR1, LCDR2, and LCDR3 of the polypeptide at least 95% similar to SEQ ID NO: 080 are identical to those annotated in SEQ ID NO: 080. Such similar PD1-VH and PD1-VL retain the binding epitope of clone 39F23, which has high specificity for PD-1 and does not inhibit PD-1 checkpoint inhibitor antibody binding.

[0082] Another aspect of the invention relates to Ig variable domains capable of binding to PD-1, wherein the PD1-VH comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 081, and the PD1-VL comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 082. In a specific embodiment, the polypeptide at least 95% similar to SEQ ID NO: 081 has conserved HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 081, and the LCDR1, LCDR2, and LCDR3 of the polypeptide at least 95% similar to SEQ ID NO: 082 are identical to those annotated in SEQ ID NO: 082. Such similar PD1-VH and PD1-VL retain the binding epitope of clone 40B20, which has high specificity for PD-1 and does not inhibit PD-1 checkpoint inhibitor antibody binding.

[0083] Another aspect of the invention relates to Ig variable domains capable of binding to PD-1, wherein the PD1-VH comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 083, and the PD1-VL comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 084. In a specific embodiment, the polypeptide at least 95% similar to SEQ ID NO: 083 has conserved HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 083, and the LCDR1, LCDR2, and LCDR3 of the polypeptide at least 95% similar to SEQ ID NO: 084 are identical to those annotated in SEQ ID NO: 084. Such similar PD1-VH and PD1-VL retain the binding epitope of clone 56H02, which has high specificity for PD-1 and does not inhibit PD-1 checkpoint inhibitor antibody binding.

[0084] Another aspect of the invention relates to binding domains of antibody clones such as XVT458-ZA-M3 and XVT458-ZA-M3, which have preferred high affinity for human and macaque PD-1, a non-blocking binding epitope for checkpoint inhibitors, and low human tissue cross-reactivity as shown in the Examples. The Ig variable domain capable of binding to PD-1 according to this aspect of the invention comprises a PD1-VH and a PD1-VL characterized by a PD1-VH having an HCDR1 with the sequence NFYIH (SEQ ID NO: 127), an HCDR2 with the sequence XIYPNYGITAYNQKFKD (where X is R or S, SEQ ID NO: 165), and an HCDR3 with the sequence GYSYAMDY (SEQ ID NO: 129). PD1-VL also contains an LCDR1 having the sequence SASQGISGDLN (SEQ ID NO: 130), an LCDR2 having the sequence HTSQLHS (where X is L or R, SEQ ID NO: 166), and an LCDR3 having the sequence QGYSKDLLT (SEQ ID NO: 132).

[0085] In certain embodiments of the Ig variable domain according to one aspect of the invention related to the preceding paragraph, the K D(measured by the protocol provided in Example 2) is 1.0 x 10 -9 ~1.0×10 -11 In a more specific embodiment, the K of binding to PD-1 is in the range of 100 mol / L, for example, including any one of the binding domains of clones XVT458-z2-m1 to -6. D is 1.0 x 10 -10 ~1.0×10 -11 In a more particular embodiment, the K D is 5.0 x 10 -10 ~1.0×10 -11 mol / L range. This is, for example, D 4.17×10 -10 nM) and XT458-z2-m6(K D 4.39×10 -10 The binding domains confer functional PD-1 binding properties (p < 0.05, p < 0.05) (Table 10).

[0086] A further embodiment of this aspect of the invention relates to sequences derived from antibody clone XVT458-ZA-M3, wherein the Ig variable domain comprises a PD1-VH comprising an HCDR2 having the sequence RIYPNYGITAYNQKFKD (SEQ ID NO: 128), and the PD1-VL comprises an LCDR2 having the sequence HTSQRHS (SEQ ID NO: 131). In a specific embodiment, the PD1-VH comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 061, and the PD1-VL comprises or consists of a polypeptide at least 95%, at least 98%, or at least 99% similar to SEQ ID NO: 062. In a more specific embodiment, the PD1-VH comprises or consists of a polypeptide having the sequence of SEQ ID NO: 061, and the PD1-VL comprises or consists of a polypeptide having the sequence of SEQ ID NO: 062.

[0087] A further embodiment of this aspect of the invention relates to sequences from antibody clone XVT458-ZA-M6, wherein the Ig variable domain comprises a PD1-VH comprising an HCDR2 having the sequence SIYPNYGITAYNQKFKD (SEQ ID NO: 133), and the PD1-VL comprises an LCDR2 having the sequence HTSQLHS (SEQ ID NO: 134). In a specific embodiment, the PD1-VH comprises or consists of a polypeptide 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 067, and the PD1-VL comprises or consists of a polypeptide 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 068. In a more specific embodiment, the PD1-VH comprises or consists of a polypeptide having the sequence of SEQ ID NO: 067, and the PD1-VL comprises or consists of a polypeptide having the sequence of SEQ ID NO: 068.

[0088] Another aspect of the invention is an isolated immunoglobulin having first and second heavy chain polypeptides, each associated with a light chain polypeptide, providing a first and second binding domain, both of which are characterized by having an Ig variable domain of any one of the non-blocking antibody clones described in this section. Such antibodies may be useful, for example, in mixed cell stimulation assays, conjugated to a fluorescent dye to allow detection of PD-1 without interfering with blockade of PD-1 binding to PD-L1 or PD-L2, or as an isotype control for immunoconjugates of the invention in functional assays.

[0089] Non-blocking immunoconjugates that bind to PD-1 An immunoconjugate is an antibody, antibody fragment, or antibody-like molecule, such as an ScFV, conjugated or attached to a further functional moiety. A further aspect of the present invention is an immunoconjugate comprising a targeting moiety corresponding to any one of the immunoglobulin variable domains described above in the section "Non-Blocking Immunoglobulin Antigen-Binding Domains Specific for PD-1." The targeting moiety is conjugated to an immunomodulatory domain capable of delivering a signaling effect to PD-1-expressing cells. Due to the unique PD-1 epitopes targeted by the binding domains according to the present invention, co-administration of the immunoconjugate to a subject with (or direct delivery before or after) a checkpoint inhibitor antibody, such as an anti-PD-1 antagonist antibody, does not significantly inhibit the immunostimulatory function of the anti-PD-1 antagonist antibody. In certain embodiments, the immunoconjugate is suitable for co-administration (i.e., does not interfere with the immunostimulatory properties of) an anti-PD-1 antagonist antibody selected from the list consisting of nivolumab, pembrolizumab, dostallimab, sintilimab, tislelizumab, cemiplimab, cetrelimab, southernlimab, toripalimab, zelvalimab, or ezabenlimab. In more particular embodiments, the immunoconjugate is suitable for co-administration with cemiplimab, dostallimab, zelvalimab, tislelizumab, ezabenlimab, toripalimab, cetrelimab, nivolumab, or pembrolizumab (see Example 14) within a medically relevant dosing therapeutic window.

[0090] A further embodiment of the present invention relates to immunoconjugates that, in addition to the Ig variable domain-binding PD-1, comprise an additional immunoactive polypeptide ligand capable of binding to a cell surface molecule expressed by immune cells. In certain embodiments, the immunoactive polypeptide ligand specifically binds to a cell surface molecule expressed by T cells or natural killer cells. These cells express PD-1 and are desirable targets for PD-1-targeted immunomodulatory activators or inhibitors, such as IL-2, due to their known anti-tumor effector properties, although other cells can also express PD-1.

[0091] In certain embodiments, the immunoactive polypeptide ligand specifically binds to a cell surface molecule expressed by CD8 T cells, such as an inhibitory co-receptor such as LAG-3 or CTLA-4, a cell surface receptor such as a T cell receptor component or interleukin receptor, or an adhesion molecule such as an integrin.

[0092] Further embodiments of immunoconjugates according to the invention comprise an immunoactive polypeptide ligand comprising an interleukin. In certain embodiments, the immunoactive polypeptide ligand portion of the immunoconjugate consists of an interleukin. In particular embodiments of immunoconjugates according to the invention, the immunoactive polypeptide ligand portion is or comprises an interleukin-2 (IL-2) polypeptide.

[0093] In some embodiments of immunoconjugates according to the invention, the immunoactive polypeptide ligand comprises or consists of an immunoglobulin variable domain reactive to, i.e., specifically binds to, an interleukin. In different embodiments of immunoconjugates according to the invention, the immunoactive polypeptide ligand comprises or consists of an immunoglobulin variable domain reactive to an interleukin receptor.

[0094] One embodiment of an immunoconjugate according to the invention relates to an immunologically active polypeptide ligand comprising both an interleukin and an immunoglobulin variable domain reactive with said interleukin. Optionally, a peptide linker connects the interleukin to either the light chain or the heavy chain of the immunoglobulin variable domain reactive with said interleukin. Such antibody domain-interleukin fusions can tailor the interleukin to enhance specific signaling functions, for example, biasing signaling of IL-2 to the IL-2 receptor on effector CD8+ T cells expressing PD-1, thereby limiting bystander activation of T regulatory T cells.

[0095] IL-2 polypeptides present in the context of anti-IL-2 antibody binding domains Some embodiments of immunoconjugates according to the invention comprise a PD-1 binding domain linked to an immunoactive polypeptide ligand comprising both an anti-IL-2 binding domain (heavy and light chain polypeptides derived from an anti-IL-2 antibody capable of specifically binding to human IL-2 protein) and an IL-2 polypeptide covalently linked to amino acid residues of the anti-IL-2 binding domain. This linkage, optionally in the form of one or two peptide linkers, provides a single, contiguous recombinant polypeptide that enhances biased signaling toward the dimeric IL-2 receptor compared to an unlinked IL-2 polypeptide, such as Proleukin.

[0096] Embodiments of the IL-2 polypeptide portion of an immunoconjugate according to the invention include human IL-2, an artificial IL-2 variant polypeptide such as Proleukin, or an IL-2 mutein such as those provided in Table 2. The IL-2 polypeptide according to the invention lacks the signal peptide M1 to S21 of native IL-2 of SEQ ID NO: 005.

[0097] Circularly permuted IL-2 polypeptide In particular embodiments of immunoconjugates according to the invention, the IL-2 polypeptide is a circularly permuted IL-2 (IL2CP) having functional domains rearranged relative to their domain positions in the wild-type sequence of SEQ ID NO: 005, as in the examples provided in Table 3. Particular embodiments of IL-2 polypeptide starting sequences to which circular permutation is applicable are wild-type SEQ ID NO: 005, SEQ ID NO: 006, or a mutein as set forth in Table 2.

[0098] Fusion of an anti-IL-2 binding domain and an IL-2 polypeptide In the immunoconjugates of the present invention, one of the antibody variable domains in the heavy or light chain of the anti-IL-2 antigen-binding domain is optionally linked to an IL-2 polypeptide by one or two peptide linkers. The anti-IL-2 binding domain linked to the IL-2 polypeptide delivers a biased IL-2 signal to the dimeric IL-2 receptor expressed on exhausted CD8+ T cells. In some embodiments, the N-terminus of IL2CP or the C-terminus of L2CP is optionally covalently linked to an IL2-VL or IL2-VH by one or two peptide linkers. In some embodiments, the N-terminus of IL2CP or the C-terminus of L2CP is optionally covalently linked to an IL2-VL and an IL2-VH by one or two peptide linkers, i.e., IL2CP is incorporated into an IL2-VL or IL2-VH.

[0099] In certain embodiments of the immunoconjugate according to the invention, it comprises IL2CP incorporated into IL-2-VH or IL-2-VL, flanked on both sides by or connected to short peptide linkers. In more specific embodiments, both the first and second peptide linkers are 1 to 20 amino acids in length. In even more specific embodiments, both the first and second peptide linkers are between 2 and 7 amino acids in length. In even more specific embodiments of the immunoconjugate according to the invention, any peptide linker flanking IL2CP is composed of glycine (G), or G and serine (S) residues. In more specific embodiments, the peptide linker connecting the C-terminal or N-terminal residue of the IL2 polypeptide to the amino acids of IL-2VH or IL-VL has a sequence selected from the sequences assigned as SEQ ID NOs: 014-024.

[0100] Anti-IL-2 binding domain In one embodiment of an immunoconjugate according to the invention, the immunoactive polypeptide ligand comprises a polypeptide having the sequence of SEQ ID NO: 167 (IL2 incorporated into the LC variable domain used in QTY065) and a polypeptide having the sequence of SEQ ID NO: 043 (the corresponding variable domain of the HC used in QTY065).

[0101] Fc part In certain embodiments of the immunoconjugates according to the invention, the immunoconjugate comprises an Fc portion, which increases its half-life in vivo. In a specific embodiment, the Ig Fc portion is an IgG Fc portion.

[0102] In certain embodiments of the immunoconjugate, the IgG Fc portion is characterized by the presence of one or more modifications to the constant region of the heavy chain to enhance correct heavy chain pairing. In certain embodiments, the modifications are selected from the following knob and hole pair mutations to enhance heavy chain pairing: -Knob: S354C, T366W & Hole:: Y349C, T366S, L368A, Y407V; -Knob: T366Y, and hole Y407T; -Knob: Y349C T366W, and hole: S354C, T366S, L368A, Y407V; -Knob: T366W, and holes: Y407A, T366S, L368A.

[0103] In certain embodiments of the immunoconjugate according to the invention, the IgG Fc portion is characterized by the presence of one or more modifications to the constant region of the heavy chain that reduce the effector function of the Fc portion. In certain embodiments, the modification or modifications are selected from L234A, L235A (LALA), L234A, L235A, P329G (LALA-PG), L234A, L235A, P329A (LALA-PA), N297A, N297Q, N297G, and D265A, N297G (DANG). In certain embodiments, the immunoconjugate comprises P329A (LALA-PA).

[0104] Immunoconjugate Format In some embodiments of the immunoconjugates of the present invention, the immunoconjugate is a heterotetrameric IgG. Heterotetrameric IgGs contain four different polypeptides and are also known as bispecific antibodies. According to such embodiments, the immunoconjugate is composed of a first pair of heavy and light chain heterodimers characterized by a non-blocking anti-PD-1 antibody binding domain of the present invention and a second Ig heavy and light chain heterodimer providing a second antigen-binding domain characterized by specific binding to an immune molecule other than PD-1. The second heterodimer comprises or consists of a second antibody heavy and light chain heterodimer containing an immunoglobulin variable domain reactive with a cell surface molecule expressed by an immune cell. In some embodiments, the second antigen-binding domain is fused to a cytokine, e.g., an anti-IL2 binding domain is fused to an IL-2 polypeptide. This format is shown in Figure 4 and includes formats such as the heterotetrameric kappa / lambda IgG format and the crossmab format (Figure 4C, WO2009080253).

[0105] In certain embodiments of immunoconjugates according to the invention, the immunoconjugate is a heterotetrameric kappa / lambda IgG. In certain embodiments, the antibody heavy and light chain heterodimers forming the anti-PD-1 Ig variable domain are characterized by a heavy chain and a kappa light chain paired with a lambda light chain that form an antibody binding domain specific for a second target, such as an interleukin, or a cell surface molecule expressed by T cells or NK cells, e.g., the 21A08 clone family described above.

[0106] In a specific embodiment, the immunoconjugate is a heterotetrameric IgG comprising a heterodimer of a first antibody heavy and light chain comprising an immunoglobulin variable domain capable of binding to PD-1, characterized in that the light chain is a lambda light chain. Additionally, the immunologically active polypeptide ligand portion of the immunoconjugate comprises or consists of a heterodimer of a second antibody heavy and light chain comprising an immunoglobulin variable domain reactive to a cell surface molecule expressed by an immune cell, characterized in that the light chain is a kappa light chain.

[0107] In a further embodiment of an immunoconjugate according to the invention, the immunoconjugate is in the single chain variable fragment (ScFv) format of immunoglobulin, in which both arms are antibodies specific for PD-1, and further comprises a conjugated ScFv specific for a different antigen attached to the N-terminus of the PD-1 antibody heavy or light chain, as shown in Figure 4. Such an immunoconjugate comprises: an anti-PD1 antibody comprising a heterodimer of first and second antibody heavy and light chains, each of which comprises an anti-PD-1 immunoglobulin variable domain, as defined in the section "Non-Blocking Immunoglobulin Antigen-Binding Domains Specific for PD-1"; and The antibody comprises an interleukin and an immunoglobulin scFv domain reactive to the interleukin, which is linked via a peptide linker to the N- or C-terminal residue of the heavy chain or light chain of an anti-PD-1 antibody.

[0108] In a specific embodiment, the C-terminal residue of the scFv domain reactive to the interleukin is linked to the N-terminal residue of the heavy or light chain of the anti-PD1 antibody via a peptide linker.

[0109] In certain embodiments of the immunoconjugates according to the invention, the immunoactive polypeptide ligands comprise a polypeptide having the sequence of SEQ ID NO:167 and a polypeptide having the sequence of SEQ ID NO:043.

[0110] In a specific embodiment of one of the preferred formats assigned to NZA596, the immunoconjugate comprises or consists of polypeptides having the sequences of SEQ ID NO:111, SEQ ID NO:095, SEQ ID NO:112, and SEQ ID NO:052.

[0111] In a specific embodiment of one of the preferred formats assigned to XWY176, the immunoconjugate comprises or consists of polypeptides having the sequences of SEQ ID NO:094, SEQ ID NO:095, SEQ ID NO:100, and SEQ ID NO:052.

[0112] In a specific embodiment of one of the preferred formats assigned to GQM289, the immunoconjugate comprises or consists of polypeptides having the sequences of SEQ ID NO:097, SEQ ID NO:098, SEQ ID NO:100, and SEQ ID NO:052.

[0113] In a particular embodiment of one of the preferred formats assigned to LTJ498, the immunoconjugate comprises or consists of polypeptides having the sequences of SEQ ID NO:094, SEQ ID NO:095, and SEQ ID NO:96.

[0114] In a particular embodiment of one of the preferred formats assigned to JLI141, the immunoconjugate comprises or consists of polypeptides having the sequences of SEQ ID NO:097, SEQ ID NO:098, and SEQ ID NO:99.

[0115] In certain embodiments of immunoconjugates according to the invention, the immunoconjugate comprises or consists of polypeptides having the sequences of SEQ ID NOs: 105, 106, and 107. In certain embodiments of immunoconjugates according to the invention, the immunoconjugate comprises or consists of polypeptides having the sequences of SEQ ID NOs: 108, 109, and 110. In certain embodiments of immunoconjugates according to the invention, the immunoconjugate comprises or consists of polypeptides having the sequences of SEQ ID NOs: 113, 106, and 114. In certain embodiments of immunoconjugates according to the invention, the immunoconjugate comprises or consists of polypeptides having the sequences of SEQ ID NOs: 115, 109, and 116.

[0116] Vectors, cells, expression domains or immunoconjugates according to the invention Another aspect of the invention relates to an isolated nucleic acid encoding an immunoconjugate according to any one of the immunoconjugate aspects of the invention. Another aspect of the invention relates to an expression vector comprising the isolated nucleic acid. The invention further encompasses host cells comprising the nucleic acid or the expression vector. In a specific embodiment, the isolated nucleic acid is comprised in a mammalian expression vector under the control of a promoter operable in mammalian cells.

[0117] Another aspect of the present invention is a method for producing a semiconductor device comprising: i) an immunoconjugate as defined in any one of the aspects or embodiments of the invention herein, and ii) Anti-PD-1 antagonist antibody The present invention relates to a combination medicine comprising:

[0118] In certain embodiments, the anti-PD1 antagonist antibody is selected from the list consisting of nivolumab, pembrolizumab, dostallimab, sintilimab, tislelizumab, cemiplimab, cetrelimab, southernlimab, toripalimab, zelvalimab, and ezabenlimab. In more particular embodiments, the antibody is cemiplimab, dostallimab, zelvalimab, tislelizumab, ezabenlimab, toripalimab, cetrelimab, nivolumab, or pembrolizumab.

[0119] Therapeutic and medical uses Also within the scope of the present invention is a method of treating cancer in a patient in need thereof, comprising administering to the patient an effective amount of an immunoconjugate according to the above aspects and embodiments.

[0120] In some embodiments, the immunoconjugate is a heterotetrameric immunoglobulin characterized by two heavy chains and two light chains, one pair of which forms an IL-2-binding ligand and is fused to an IL-2 polypeptide (as in Figure 4C or E). In some embodiments, the immunoconjugate consists of an anti-PD-1 antibody that is a heterodimer of an associated pair of heavy and light chains, linked by a peptide linker to an scFv comprising the heavy and light chain variable domains of an anti-IL-2 antibody, one of which anti-IL-2 domains is linked (optionally by a peptide linker) to an IL-2 polypeptide (as in Figure 4D or E).

[0121] Another aspect of the invention is an immunoconjugate comprising a non-blocking anti-PD-1 binding domain according to the invention for use in a patient receiving an anti-PD1 antagonist antibody selected from the list consisting of nivolumab, pembrolizumab, dostallimab, sintilimab, tislelizumab, cemiplimab, cetrelimab, southernlimab, toripalimab, zelvalimab, and ezabenlimab. In a more particular embodiment, the antibody is cemiplimab, dostallimab, zelvalimab, tislelizumab, ezabenlimab, toripalimab, cetrelimab, nivolumab, or pembrolizumab.

[0122] Another aspect of the invention is an anti-PD-1 antagonist antibody selected from the list consisting of nivolumab, pembrolizumab, dostallimab, sintilimab, tislelizumab, cemiplimab, cetrelimab, southernlimab, toripalimab, zelvalimab, and ezabenlimab for use in patients who have received an immunoconjugate according to the invention. This includes use in patients who will soon receive treatment with the immunoconjugate within a medically relevant therapeutic window, particularly within six weeks of anti-PD-1 administration. In one embodiment, the anti-PD-1 antagonist antibody is provided for use in patients who have received an immunoconjugate according to the invention in the previous month.

[0123] The invention further includes nivolumab for use in a patient receiving an immunoconjugate according to the invention.The invention further includes pembrolizumab for use in a patient receiving an immunoconjugate according to the invention.

[0124] Similarly, the present invention encompasses a method of treating a patient diagnosed with cancer, which method involves administering to the patient an effective amount of an immunoconjugate as specified herein, or a pharmaceutically acceptable salt thereof as defined in detail herein, optionally in combination with an anti-PD1 antagonist antibody, as defined in detail herein.

[0125] In certain embodiments, the same subject is co-administered, within a medically relevant therapeutic window, an anti-PD1 antagonist antibody selected from the list consisting of nivolumab, pembrolizumab, dostallimab, sintilimab, tislelizumab, cemiplimab, cetrelimab, southernlimab, toripalimab, zelvalimab, or ezabenlimab. In more particular embodiments, the antibody is cemiplimab, dostallimab, zelvalimab, tislelizumab, ezabenlimab, toripalimab, cetrelimab, nivolumab, or pembrolizumab.

[0126] Pharmaceutical Compositions, Dosage Forms / Dosage Forms and Salts According to one aspect of the compounds according to the invention, the immunoconjugate according to the invention is provided as a pharmaceutical composition, pharmaceutical dosage form or pharmaceutical dosage form, said pharmaceutical composition, pharmaceutical dosage form or pharmaceutical dosage form comprising the immunoconjugate of the invention and at least one pharmaceutically acceptable carrier, diluent or excipient.

[0127] In certain embodiments of the invention, the immunoconjugates of the invention are typically formulated into pharmaceutical formulations to provide easily controlled dosage of the drug and to provide the patient with a clear, easy-to-use product.

[0128] The present invention further encompasses pharmaceutical compositions comprising an immunoconjugate of the invention and a pharmaceutically acceptable carrier, hi a further embodiment, the composition comprises at least two pharmaceutically acceptable carriers, such as those described herein.

[0129] Certain embodiments of the present invention relate to dosage forms for parenteral administration, such as subcutaneous, intravenous, intrahepatic or intramuscular injection forms. Optionally, pharmaceutically acceptable carriers and / or excipients may be present.

[0130] The pharmaceutical compositions of the present invention may be subjected to conventional pharmaceutical procedures such as sterilization, and / or may contain conventional inert diluents, lubricants, or buffers, as well as auxiliary substances such as preservatives, stabilizers, wetting agents, emulsifiers, and buffers. They may be manufactured by standard processes, such as conventional mixing, granulation, dissolution, or lyophilization processes. Many such procedures and methods for preparing pharmaceutical compositions are known to those skilled in the art; see, for example, L. Lachman et al., The Theory and Practice of Industrial Pharmacy, 4th Ed., 2013 (ISBN 8123922892).

[0131] Manufacturing and treatment methods according to the present invention The present invention further encompasses, as a further aspect, the use of an immunoconjugate as defined in detail above for use in a method for the manufacture of a medicament for the treatment or prevention of cancer.

[0132] For example, wherever alternatives of a single separable feature, such as an isotype protein or coding sequence, or cancer, are positioned herein as "embodiments," it will be understood that such alternatives may be freely combined to form individual embodiments of the invention disclosed herein. Thus, any of the alternative embodiments of an immunoconjugate may be combined with any of the alternative embodiments of an anti-PD-1 antagonist compound, and these combinations may be combined with any of the medical indications described herein.

[0133] The present invention further encompasses the following: A. An immunoglobulin variable domain capable of binding to PD-1 in the presence of at least an equimolar amount of a PD-1-specific antibody selected from the group consisting of pembrolizumab and nivolumab, The immunoglobulin variable domain - an antibody heavy chain variable domain polypeptide (PD1-VH), and - Antibody light chain variable domain polypeptide (PD1-VL) An immunoglobulin variable domain comprising: B. The immunoglobulin variable domain of paragraph A, wherein the binding of an antibody comprising the immunoglobulin variable domain defined in claim 1 to PD-1 is reduced by no more than 20% in the presence of a 100-fold molar excess of a PD-1-specific antibody selected from the group consisting of pembrolizumab and nivolumab. C. K for PD-1 of antibodies characterized by the immunoglobulin variable domain D is 1.0 × 10 measured using surface plasmon resonance. -9 mol / L or less. D. Immunoglobulin variable domains capable of binding to PD-1, including PD1-VH and PD1-VL, - the PD1-VH comprises a heavy chain CDR (HCDR) 1 having the sequence GFTFSINAMT (SEQ ID NO: 118), an HCDR2 having the sequence TISGSGFSTYYADSLKGR (SEQ ID NO: 119), and an HCDR3 having the sequence EVYGDY (SEQ ID NO: 120); and - said PD1-VL has the sequence SGX 1 SSNIGSX 2 X 3 VF (SEQ ID NO: 121), an LCDR2 having the sequence SNNQRPS (SEQ ID NO: 122), and an LCDR3 having the sequence AAWDDSLSIWVF (SEQ ID NO: 123); where X 1 is N, S, Q or A, in particular X 1 is S, Q, or A; and X 2 X 3 is NS, QS, SS, or NA, and in particular X 2 X 3 is QS, SS, or NA The immunoglobulin variable domain, characterized in that E. PD1-VL contains the sequence SGASSNIGS QS VF (underlined in bold) (SEQ ID NO: 124), In particular, PD1-VH comprises or consists of a polypeptide at least (≧) 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 085, and PD1-VL comprises or consists of a polypeptide at least (≧) 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 086; More particularly, the immunoglobulin variable domain according to paragraph D, wherein the PD1-VH comprises or consists of a polypeptide having the sequence of SEQ ID NO: 085, and the PD1-VL comprises or consists of a polypeptide having the sequence of SEQ ID NO: 086. F. PD1-VL is a sequence SGASSNIGS SS comprising LCDR1 with VF (underlined in bold) (SEQ ID NO: 125); In particular, PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 085, and PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 087; More particularly, the immunoglobulin variable domain according to paragraph D, wherein the PD1-VH comprises or consists of a polypeptide having the sequence of SEQ ID NO: 085, and the PD1-VL comprises or consists of a polypeptide having the sequence of SEQ ID NO: 087. G. PD1-VL is a sequence SGASSNIGS NA comprising LCDR1 with VF (underlined in bold) (SEQ ID NO: 126); In particular, PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 085; and PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 091; More particularly, the immunoglobulin variable domain according to paragraph D, wherein the PD1-VH comprises or consists of a polypeptide having the sequence of SEQ ID NO: 085, and the PD1-VL comprises or consists of a polypeptide having the sequence of SEQ ID NO: 91. H. Immunoglobulin variable domains capable of binding to PD-1, including PD1-VH and PD1-VL, - the PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 071, in particular, the polypeptide has HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 071; and - the PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 072, and in particular, the polypeptide has LCDR1, LCDR2, and LCDR3 identical to the CDRs annotated in SEQ ID NO: 072. The immunoglobulin variable domain, characterized in that I. Immunoglobulin variable domains capable of binding to PD-1, including PD1-VH and PD1-VL, - the PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 073, in particular, the polypeptide has HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 073; and - the PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 074, and in particular, the polypeptide has LCDR1, LCDR2, and LCDR3 identical to the CDRs annotated in SEQ ID NO: 074. The immunoglobulin variable domain, characterized in that J. Immunoglobulin variable domains capable of binding to PD-1, including PD1-VH and PD1-VL, - the PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 075, in particular, the polypeptide has HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 075; and - the PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 076, and in particular, the polypeptide has LCDR1, LCDR2, and LCDR3 identical to the CDRs annotated in SEQ ID NO: 076. The immunoglobulin variable domain, characterized in that K. Immunoglobulin variable domains capable of binding to PD-1, including PD1-VH and PD1-VL, - the PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 077, in particular, the polypeptide has HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 077; and - the PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 078, and in particular, the polypeptide has LCDR1, LCDR2, and LCDR3 identical to the CDRs annotated in SEQ ID NO: 078. The immunoglobulin variable domain, characterized in that L. Immunoglobulin variable domains capable of binding to PD-1, including PD1-VH and PD1-VL, - the PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 079, in particular, the polypeptide has HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 079; and - the PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 080, and in particular, the polypeptide has LCDR1, LCDR2, and LCDR3 identical to the CDRs annotated in SEQ ID NO: 080. The immunoglobulin variable domain, characterized in that M. Immunoglobulin variable domains capable of binding to PD-1, including PD1-VH and PD1-VL, - the PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 081, in particular, the polypeptide has HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 081; and - the PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 082, and in particular, the polypeptide has LCDR1, LCDR2, and LCDR3 identical to the CDRs annotated in SEQ ID NO: 082. The immunoglobulin variable domain, characterized in that N. Immunoglobulin variable domains capable of binding to PD-1, including PD1-VH and PD1-VL, - the PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 083, in particular, the polypeptide has HCDR1, HCDR2, and HCDR3 identical to the CDRs annotated in SEQ ID NO: 083; and - the PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, or 99% or more similar to SEQ ID NO: 084, and in particular, the polypeptide has LCDR1, LCDR2, and LCDR3 identical to the CDRs annotated in SEQ ID NO: 084. An immunoglobulin variable domain characterized by: O. The affinity constant (K) of an antibody characterized by the immunoglobulin variable domain for PD-1 D ) is 1.0 × 10 as measured by SPR -9 ~1.5×10 -11 mol / L, especially 1.0 × 10 -10 ~1.5×10 -11 mol / L, more specifically 5.0 × 10 -1 0 to 1.5 × 10 -11 The immunoglobulin variable domain according to any one of items A to N, wherein the IgG ranges from 0.1 to 1.0 mol / L. P. Immunoglobulin variable domains capable of binding to PD-1, including PD1-VH and PD1-VL, - the PD1-VH has an HCDR1 with the sequence NFYIH (SEQ ID NO: 127), wherein X is R or S, and the sequence XIYPNYG I comprising an HCDR2 having the sequence TAYNQKFKD (underlined in bold) (SEQ ID NO: 165), and an HCDR3 having the sequence GYSYAMDY (SEQ ID NO: 129); and - the PD1-VL has an LCDR1 having the sequence SASQGISGDLN (SEQ ID NO: 130), X is L, or R, Q XHS (underlined in bold) (SEQ ID NO: 166), and LCDR2 with sequence Q G Contains LCDR3 with YSKDLLT (underlined in bold) (SEQ ID NO: 0132) The immunoglobulin variable domain, characterized in that Q. The immunoglobulin variable domain according to paragraph P, - PD1-VH is the sequence R IYPNYGITAYNQKFKD (underlined in bold) (SEQ ID NO: 128); and - PD1-VL is sequence HTS QR containing LCDR2 with HS (underlined in bold) (SEQ ID NO: 131); In particular, wherein the PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 061, and the PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 062; More particularly, the immunoglobulin variable domains wherein the PD1-VH comprises or consists of a polypeptide having the sequence of SEQ ID NO: 061, and the PD1-VL comprises or consists of a polypeptide having the sequence of SEQ ID NO: 062. R. An immunoglobulin variable domain according to paragraph P, - PD1-VH comprises an HCDR2 having the sequence SIYPNYGITAYNQKFKD (SEQ ID NO: 133), and - PD1-VL is sequence HTS Q comprising LCDR2 with LHS (underlined in bold) (SEQ ID NO: 134); particularly wherein PD1-VH comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 067, and PD1-VL comprises or consists of a polypeptide that is 95% or more, 98% or more, 99% or more similar to SEQ ID NO: 068; More particularly, the immunoglobulin variable domains wherein the PD1-VH comprises or consists of a polypeptide having the sequence of SEQ ID NO: 067, and the PD1-VL comprises or consists of a polypeptide having the sequence of SEQ ID NO: 068. S. The immunoglobulin variable domain of any one of paragraphs P to R, wherein the K of binding to PD-1 of an antibody characterized by said immunoglobulin variable domain is Dis 1.0 x 10 as measured by the protocol provided in Example 2 -9 ~1.0×10 -11 mol / L, especially 1.0 × 10 -10 ~1.0×10 -11 mol / L, more specifically 5.0 × 10 -10 ~1.0×10 -11 The immunoglobulin variable domain is in the range of mol / L. T. The immunoglobulin variable domain of any one of paragraphs D-S, wherein the immunoglobulin variable domain further meets the functional specifications of any one of paragraphs AC. U. An immunoconjugate comprising an immunoglobulin variable domain according to any one of paragraphs A-T. V. The immunoconjugate of paragraph U, characterized in that when the immunoconjugate is co-administered with an anti-PD-1 antagonist antibody, it does not significantly inhibit the immunostimulatory function of the anti-PD-1 antagonist antibody, particularly when the anti-PD-1 antagonist antibody is selected from the list consisting of nivolumab, pembrolizumab, dostalizumab, sintilimab, tislelizumab, cemiplimab, cetrelimab, southernlimab, toripalimab, zelvalimab, or ezabenlimab, more particularly nivolumab or pembrolizumab. W. The immunoconjugate of paragraphs U or V, further comprising an immunoactive polypeptide ligand capable of binding to a cell surface molecule expressed by an immune cell, particularly a cell surface molecule expressed by a T cell or a natural killer cell. X. The immunoconjugate of paragraph W, wherein the immunoactive polypeptide ligand comprises or consists of an interleukin. Y. The immunoconjugate of paragraphs W or X, wherein the immunoactive polypeptide ligand comprises or consists of an interleukin or an immunoglobulin variable domain reactive to an interleukin receptor. Z. The immunoconjugate of any one of paragraphs WY, wherein the immunoactive polypeptide ligand comprises or consists of an interleukin and an immunoglobulin variable domain reactive to said interleukin, and optionally a peptide linker connecting the interleukin and the immunoglobulin variable domain reactive to said interleukin. AA. The immunoconjugate of any one of paragraphs X-Z, wherein the interleukin is an IL-2 polypeptide. BB. The immunoconjugate of paragraph AA, wherein the IL-2 polypeptide is a circularly permuted IL-2 (IL2CP) polypeptide. CC. The immunoconjugate of any one of paragraphs U through BB, wherein the immunoconjugate comprises a fragment crystallizable (Fc) immunoglobulin domain. DD. The immunoconjugate of paragraph CC, wherein the Fc domain is an immunoglobulin gamma (IgG) Fc domain. EE. The IgG Fc domain may contain one or more modifications to the heavy chain constant region to enhance correct heavy chain pairing, in particular: - Knob: S354C, T366W & Hole: Y349C, T366S, L368A, Y407V; - Knob: T366Y, and hole Y407T; - Knob: Y349C, T366W, and hole: S354C, T366S, L368A, Y407V; or - Knob T366W and holes: Y407A, T366S, L368A The immunoconjugate of paragraph DD, characterized by the presence of a set of knob and hole modifications selected from: FF. IgG Fc domains containing one or more modifications to the constant region of the heavy chain to reduce the effector function of the Fc portion; Modifications to the constant region of the heavy chain selected from L234A, L235A (LALA), L234A, L235A, P329G (LALA-PG), L234A, L235A, P329A (LALA-PA), N297A, N297Q, N297G and D265A, N297G (LALA-DANG), more particularly P329A (LALA-PA). The immunoconjugate of paragraphs DD or EE, characterized by the presence of: GG. Immunoconjugates - a first antibody heavy chain and light chain heterodimer comprising an anti-PD-1 immunoglobulin variable domain as defined in any one of paragraphs A to T; and a heterodimer of a second antibody heavy and light chain comprising an immunoglobulin variable domain reactive to a cell surface molecule expressed by an immune cell, in particular an interleukin or an interleukin receptor; The immunoconjugate of any one of paragraphs U to FF, which is a heterotetrameric IgG comprising: HH. The immunoconjugate of paragraph GG, wherein the second antibody heavy and light chain heterodimer comprises an immunoglobulin variable domain reactive to an interleukin, and the interleukin is covalently bound to the variable domain reactive to the interleukin. II. The immunoconjugate of paragraphs GG or HH, wherein the heterotetrameric IgG format is selected from a heterotetrameric kappa / lambda IgG format or a CrossMab format. JJ. The immunoconjugate of any one of paragraphs GG-II, wherein the heterotetrameric IgG is a heterotetrameric kappa / lambda IgG; in particular, the heterodimer of antibody heavy and light chains comprising an anti-PD-1 immunoglobulin variable domain is characterized by a lambda light chain. KK. The immunoconjugate comprises: - an anti-PD1 antibody comprising a heterodimer of first and second antibody heavy and light chains, each of which comprises an anti-PD-1 immunoglobulin variable domain as defined in any one of paragraphs A to T; and - an interleukin, and an immunoglobulin scFv domain reactive with said interleukin, wherein said immunoglobulin scFv domain is linked to the N-terminal or C-terminal residue of the heavy chain or light chain of an anti-PD-1 antibody via a peptide linker, particularly said peptide linker being 10-30 amino acids in length, and even more particularly said peptide linker having the sequence of (G4S)2 or SEQ ID NO: 024; Including, In particular, the immunoconjugate according to any one of paragraphs U to FF, wherein the C-terminal residue of the interleukin-reactive scFv domain is linked to the N-terminal residue of the heavy or light chain of the anti-PD1 antibody via a peptide linker. LL. The immunoconjugate of any one of paragraphs U to KK, wherein the immunoactive polypeptide ligand comprises a polypeptide having the sequence of SEQ ID NO: 167 (incorporated QTY065 LC IL2) and a polypeptide having the sequence of SEQ ID NO: 043. MM. An isolated nucleic acid encoding an immunoglobulin variable domain capable of binding to PD-1 according to any one of paragraphs A-T. NN. An isolated nucleic acid encoding an immunoconjugate described in any one of paragraphs U through LL. OO. An expression vector comprising an isolated nucleic acid according to paragraph NN. PP. A host cell comprising a nucleic acid according to Section 00, or an expression vector according to Section 00. QQ. - an immunoconjugate as defined in any one of paragraphs U to LL, and - an anti-PD1 antagonist antibody, in particular selected from the list consisting of nivolumab, pembrolizumab, dostallimab, sintilimab, tislelizumab, cemiplimab, cetrelimab, southernlimab, toripalimab, zelvalimab or ezabenlimab, more in particular nivolumab or pembrolizumab; A combination medicine comprising: RR. A pharmaceutical composition comprising an immunoconjugate described in any one of Items U to LL, an isolated nucleic acid described in Item NN, an expression vector described in Item OO, a host cell described in Item PP, or a pharmaceutical combination described in Item QQ, and a pharmaceutically acceptable excipient. SS. An immunoconjugate according to any one of items U-LL, an isolated nucleic acid according to item NN, an expression vector according to item OO, a host cell according to item PP, a pharmaceutical combination according to item QQ, or a pharmaceutical composition according to item RR for use as a pharmaceutical. TT. An agent selected from an immunoconjugate described in any one of items U to LL, an isolated nucleic acid described in item NN, an expression vector described in item OO, a host cell described in item PP, a pharmaceutical combination described in item QQ, or a pharmaceutical composition described in item RR for use in the treatment of cancer. UU: The agent described in item TT, wherein the agent is administered within 6 weeks before or after treatment with an anti-PD-1 antagonist antibody. VV. An anti-PD-1 antagonist antibody for use in the treatment of cancer, wherein the anti-PD-1 antagonist antibody is administered within 6 weeks before or after treatment with an agent specified in section TT. WW. An anti-PD-1 antagonist antibody for use according to paragraph VV, wherein the anti-PD-1 antagonist antibody is selected from the list consisting of nivolumab, pembrolizumab, dostallimab, sintilimab, tislelizumab, cemiplimab, southernlimab, zelvalimab, ezavenlimab, toripalimab, or cetrelimab, in particular the anti-PD-1 antagonist antibody is selected from the list consisting of nivolumab, pembrolizumab, dostallimab, tislelizumab, cemiplimab, zelvalimab, ezavenlimab, toripalimab, or cetrelimab. XX. i) selecting a suitable patient; and ii) administering a therapeutically effective amount of an immunoconjugate according to any one of items U to LL, an isolated nucleic acid according to item NN, an expression vector according to item OO, a host cell according to item PP, a pharmaceutical combination according to item QQ, or a pharmaceutical composition according to item RR. A method of treatment comprising:

[0134] The present invention is further explained by the following examples and figures, from which further embodiments and advantages can be derived, which are intended to illustrate the invention without limiting its scope. [Brief explanation of the drawings]

[0135] [Figure 1-1] FIG. 1 shows the labeling of antibody residues in a recognized format using the standard numbering and Kabat numbering systems. [Figure 1-2] FIG. 1 shows the labeling of antibody residues in a recognized format using the standard numbering and Kabat numbering systems. [Figure 2] Figure 2 shows how the IL2-CP format can be altered to suit incorporation into various variable domain sites: (Left) Diagram of IL-2 in the correct orientation for binding to antibody A or B; (Center) Diagram of LCDR1 with CDAB IL-2CP incorporated into antibody A, B, or C; (Right) HCDR3 with BCDA IL-2CP incorporated. [Figure 3] SPR sensorgrams (response units, RU per time (sec)) of PD-1 premixed with antibody or buffer performed on immobilized PD-L1 as analyte. [Figure 4] Figure 4 shows testing of major formats with various PD-1 binding moiety configurations, including A. dual-scFv fusion, B. Fab-dual-scFv, C. heterotetrameric IgG (IgG CrossMab), D-E. bivalent PD-1 IgG antibody with two binding sites of anti-hIL-2 ScFv, and F. heterotetrameric IgG (kappa / gamma). [Figure 5] Figure 5 shows cell proliferation analyzed by flow cytometry, detecting Ki67+ cells in CD8 T cells, NK cells, and Treg cells. 0.2 mg / kg of the bispecific compound was administered intravenously to wild-type (WT) C57BL / 6 mice. Blood was collected before injection and on days 3 and 6 after compound administration. [Figure 6-1] A) The top shows a representative image of the NZA596 compound digested to obtain two Fab fragments and the Fc domain. The bottom shows the deconvoluted MS profile of the detected Fab fragment, showing a single species with the expected molecular weight. [Figure 6-2] B) The top shows a representative image of the BGY642 compound digested to obtain two Fab fragments and the Fc domain. The bottom shows the deconvoluted MS profile of the detected Fab fragment, showing a single species with the expected molecular weight. [Figure 7] Top) Representative images of cis-signaling on unblocked CTV+ cells and blocked CSFE+ cells. Bottom) Percentage of pSTAT5+ cells plotted against NZA596 compound concentration in nM. CFSE+ cells were pre-incubated with 21A08Ap1 antibody, pembrolizumab, or nivolumab (n=3). [Figure 8] Figure 8 shows the number of unique TCR rearrangements in TILs from mice treated intravenously with vehicle, NZA594, or QTY065 at 0.2 mg / kg on days 0 and 3. Genomic DNA was extracted from tumors of mice sacrificed on study day 5 (n=8). Statistical analysis was performed by unpaired Wilcox test. [Example]

[0136] Example 1: Design and production of anti-IL-2 antibody IL-2 cytokine fusion protein Antibodies that bind to IL-2 were derived, isolated, and structurally analyzed using methods well known to those skilled in the art. Antibody A (HC SEQ ID NO: 001, LC SEQ ID NO: 002) is a high-affinity anti-IL-2 antibody, and antibody B (HC SEQ ID NO: 003, LC SEQ ID NO: 004) is a low-affinity anti-IL-2 antibody. Antibody C has no affinity for IL-2 and is included as a control. Genes encoding VL-CL (light chain) and VH-CH1-CH2-CH3 (heavy chain) were cloned into separate plasmids in the mammalian expression vector pcDNA3.4. Antibodies were produced using transient gene expression in Expi293 cells (Gibco, A14527) according to the standard protocol provided by the vendor. Plasmid DNA (HC / LC ratio 1:2 w / w) was transfected using ExpiFectamine™ 293 reagent. Cells were maintained at 37°C, 8% CO2, and on an orbital shaker (150 rpm) for 6 days. Antibodies were purified to homogeneity from the supernatant by Protein A Chromatography (MabSelect™ SuRe™, GE17-5438-01). Protein quality control was performed by SDS-PAGE (NuPAGE™ 4-12% Bis-Tris Protein Gels, ThermoFisher) and analytical size exclusion chromatography (SEC) (GE lifesciences, Superdex 200 increase 10 / 300). Yields and purity are reported in Table 1. All antibodies were obtained as pure protein products and eluted as a single peak in SEC-HPLC analysis.

[0137] Binding to IL-2 was first tested by ELISA using serial dilutions of the antibody on coated IL-2. Recombinant human IL-2 was coated onto Maxisorp ELISA plates (Invitrogen, 44-2404-21) at 5 μg / ml overnight at 4°C. After 2 hours of blocking, antibodies A, B, and C were incubated in serial dilutions starting at 10 μg / ml in assay buffer. Antibodies were detected with anti-human IgG-peroxidase (Sigma, A0170). After addition of chemiluminescent substrate and blocking, absorbance was read (A on a Spectramax iD3 plate reader).450 ~A 570 ). EC 50 Values ​​were obtained by plotting absorbance versus log(concentration) and performing sigmoidal 4-PL fitting using GraphPad Prism (Table 1). Only antibody A showed binding to IL-2 by ELISA, which could be detected on the Fc portion with an anti-human IgG antibody conjugated with HRP. Next, the binding affinity of antibodies A, B, and C of the present invention to recombinant human IL-2 (rhIL-2) was measured using the more sensitive biolayer interferometry (BLI) method. The antibodies were immobilized on an amine-reactive (second-generation) sensor (ForteBio, 18-5092). The association (600 s) and dissociation (900 s) of a dilution series of recombinant IL-2 (Acro Biosystems, IL2-H4113) were measured using the antibody-coated biosensor on the Octet-System (Octet RED, ForteBio). K D Values ​​were obtained by fitting the kinetic data with ForteBio data analysis software (8.2). This assay confirmed that antibody A binds to rhIL-2 with high affinity. Binding of antibody B could not be measured by ELISA, but a weak affinity constant could be measured by BLI. Antibody C showed no binding to rhIL-2 up to concentrations above 1000 nM (Table 1).

[0138] Table 1. Production yield by transient expression, purity and binding affinity to recombinant hIL-2 as measured by ELISA or BLI. [Table 1]

[0139] Fusion proteins comprising IL-2 polypeptides or IL-2 muteins containing amino acid substitutions with favorable pharmacokinetic properties (Table 2) have previously been demonstrated to provide IL-2 signaling when the cytokine is conjugated to an antibody chain by C-terminal attachment to an antibody variable chain domain (see, e.g., WO2018184964A1; Deak LC. et al., 2022, Nature 610:161; Gutbrodt KL. 2013 Sci. Trans. Med. 5:201; Gillies SD. (1992) PNAS 89(4):1428) or by N-terminal attachment to an anti-IL-2 antibody (see, e.g., WO2017122130A1).

[0140] Table 2. IL-2 Polypeptides [Table 2]

[0141] We analyzed the structure of IL-2 in complex with the antibody NARA1 (RCSB Protein Structure Data Bank 5LQB), from which humanized antibodies A and B were derived, and evaluated alternative formats for guiding IL-2 incorporation or fusion to maintain the cytokine's orientation relative to the antibody and similar receptor signaling qualities (Arena-Ramires et al., 2016 Sci. Trans. Med. 8:367). Based on the crystal structure of the NARA1 / IL-2 complex, we identified potential incorporation sites by examining the linker chain between different alpha-helical domains (here designated as AD) and their proximity to CDRs or framework regions on the antibody's VH or VL. We designed various combinations of circularly permuted IL-2 (IL-2CP) to preserve the cytokine's tertiary structure and orientation relative to its natural binding to the antibody when fused to the indicated heavy or light chain variable regions (Table 3, Figure 2). To maintain the same orientation on the antibody, IL-2 must be permuted (rearranged) differently depending on the junction site. To create a circular permutation in the IL-2 helical domain CDAB format, the crystal structure suggests that the region V89-D104 (see the Proleukin sequence) between the B and C helices should be opened to maintain the essential tertiary structure. To create a circular permutation in the IL-2 helical domain BCDA format, the crystal structure suggests that the region G47-E72 between the A and B helices could be opened. To create a circular permutation in the IL-2 helical domain DABC format, the crystal structure suggests that the region G118-I134 between the C and D helices could be targeted.

[0142] Table 3. Circularly permuted IL-2 polypeptides and optimal binding sites [Table 3]

[0143] Representative IL-2 fusion proteins were then designed according to several embodiments of the invention comprising antibody A, B, or C linked to IL-2 directly or via linkers of various lengths as shown in Table 4.

[0144] The LCDR1 of IL-2 antibody A, residues Y27d-D30 (or Y31-D34 structural numbering) and the region connecting IL-2 helix B and helix C between residues S95 and N97, were identified as promising regions for further engineering. The LCDR1 of each antibody A was opened between Y27d and D30 according to the Kabat definition, providing a new C-terminal residue for LCDR1 Y27d and a new N-terminal residue for LCDR1 D30. LCDR1 residue Y27d was attached to the N-terminus of IL-2CP, opened between K96 and N97 (SEQ ID NO: 10). The C-terminus of IL-2CP was attached to LCDR1 residue 30 according to the Kabat definition, and residues Q28 and G29 of the antibody LCDR1 were replaced with cyclized IL-2CP residues SEQ ID NO: 10, either directly or via peptide linkers of various lengths. This incorporation was repeated for the equivalent amino acid residues with the LCDR1 of the lower affinity IL-2 antibody B and antibody C, which has no affinity for IL-2 (Table 4).

[0145] Additional constructs were designed to enable the insertion of alternative IL-2CPs in the HDCR3 or HCDR2 regions. VBE401 was developed with IL-2 of SEQ ID NO: 011 opened between N97 and K96, and the cytokine inserted (N97 was removed, leaving F96 as the new N-terminus). SEQ ID NO: 011 was inserted between residues E98 and G99 in the HCDR3 of antibody A, following and preceding the linker GGG. Additionally, insertion of IL-2CP into HDRC2 was tested. A circularly permuted IL-2 was created by opening the sequence between K63 and F64, fusing the original N- and C-termini to provide SEQ ID NO: 012. SEQ ID NO: 012 was inserted between residues G53 and S54 in the HDCR2 of antibody A, following and preceding the linker GGG, as described above (LIZ707).

[0146] For each construct (Table 4), genes encoding VL-(IL-2)-CL, VL-CL, VH-CH1-CH2-CH3, and VH-(IL-2)-CH1-CH2-CH3 were cloned into separate plasmids in the mammalian expression vector pcDNA3.4. Antibody-IL-2 fusion proteins were produced using transient gene expression in Expi293 cells as described above for antibodies A, B, and C. All constructs were highly pure (>95%) and eluted as a single peak in SEC-HPLC analysis (Table 5).

[0147] Interaction of immunoconjugates with the CD132-CD122 heterodimeric receptor. The anti-IL-2 antibody clone 5344, which binds to the CD122 binding site on IL-2, was used to determine the correct folding of the IL-2 portion of the fusion protein. 50 The values ​​are reported in Table 6. The IL-2 portion of all constructs bound well to an anti-IL-2 antibody (clone 5344) in ELISA and could be detected on the Fc portion with an anti-human IgG antibody conjugated to HRP (Table 5).

[0148] Table 4. Antibody-IL-2 fusion proteins with an IL-2CP fusion site on each CDR of Antibody A, B, or C. The IL-2CP sequence is incorporated into the variable domain of the heavy or light chain of the antibody as indicated. All heavy chain constant regions were SEQ ID NO:048 and all light chain constant regions were SEQ ID NO:049. [Table 4]

[0149] The binding of one representative anti-IL-2 antibody / IL-2 fusion (QTY065) to the CD122 / CD132 complex (His-tagged, Acro Biosystem, catalog no. ILG-H5283) was further evaluated by SPR (BIAcore 3000, v4.1.2; GE Healthcare) and compared with Proleukin. CD122 / CD132 was captured on an NTA sensor chip (GE Healthcare). Proleukin or QTY065 Fab was used as the analyte. The calculated affinity constant for QTY065 Fab was 0.32 nM, comparable to the 0.42 nM obtained with Proleukin. The intermediate-affinity binding to the IL-2R complex (CD122 / CD132) remained unchanged for the IL-2CP antibody fusion protein.

[0150] Table 5. Purity of final protein by SEC-HPLC, functionality assessed by ELISA, and CD25 binding by SPR. [Table 5]

[0151] Biological effects of IL-2 tested by HEK Blue IL-2 reporter assay The functionality of the IL-2 fusion proteins was assessed using HEK-Blue™ IL-2 reporter cells, which express the trimeric IL-2 receptor (CD25 / CD122 / CD132) and were engineered to secrete secreted embryonic alkaline phosphatase activity (SEAP) upon IL-2-induced STAT5 activation. HEK-Blue™ IL-2 reporter cells (Invivogen) were incubated in growth medium with a dilution series of antibody-IL-2 fusion proteins (1:3 dilutions starting from 0.56 nM) for 20 hours at 37°C in 5% CO2. HEK-Blue™ IL-2 cell supernatants were diluted 1:10 in QUANTI-Blue™ solution for colorimetric quantification of SEAP and incubated for 3 hours at 37°C. Absorbance was read at 620 nm and plotted against protein concentration. E 50Values ​​were interpolated using GraphPad Prism. EC determined by colorimetric SEAP quantification for each concentration of IL-2 antibody fusion protein. 50 The values ​​showed that the antibody-IL-2 fusion proteins induced STAT5 signaling with comparable potency on HEK Blue IL-2R reporter cells (Table 6).

[0152] Table 6. EC50 values ​​for immunoconjugate HEK Blue IL-2 reporter assay [Table 6]

[0153] Binding affinity to CD25 measured by surface plasmon resonance (SPR) analysis To provide optimally biased signaling to CD8 T cells, CD25 binding of the IL-2 fusion protein is preferably minimal. To understand whether the CD25-binding site on the antibody-IL-2 fusion protein is accessible, binding to CD25 was assessed by surface plasmon resonance (SPR) analysis (BIAcore 3000, GE Healthcare, 33-1140587-3682). His-tagged recombinant CD25 was captured via TrisNTA-biotin on an SA chip, and kinetic titrations of the IL-2 fusion protein of the present invention were performed at concentrations up to 500 nM. The IL-2 fusion protein against Antibody A showed no binding to CD25 at concentrations below 500 nM. Antibody A binds IL-2 with high affinity at the CD25-binding site, thereby blocking IL-2 binding to its receptor, CD25. Antibody B binds to IL-2 with low affinity and allows IL-2 to bind to CD25 only when fused to IL-2 with a long linker (13 or 14 amino acids or more). Antibody C has no affinity for IL-2 and allows IL-2 fused to it with any linker length to bind to CD25 (Tables 4 and 5).

[0154] pSTAT5 in mouse splenocytes (EC50 for Treg, CD8, and NK) STAT5 phosphorylation was analyzed in mouse splenocytes as a downstream signaling pathway of IL-2R activation. To assess the in vitro selectivity of antibody-IL-2 fusion proteins, pSTAT5 was measured in different cell populations after stimulation with Proleukin or compounds of the present invention. Freshly isolated mouse splenocytes from C57BL / 6 mice were incubated with a dilution series of Proleukin or IL-2 antibody fusion proteins starting at 100 nM. Cells were immediately fixed and stained for surface markers (i.e., CD25, CD3, NK1.1, CD4, CD8). After permeabilization (Perm III buffer, BD Biosciences), intracellular staining was performed (FoxP3, pSTAT5) before acquisition by flow cytometry. CD8 + T cells, NK cells, CD4 + CD25 + FoxP3 + The percentage of pSTAT5+ Treg cells was plotted against the molar concentration of the IL-2-antibody fusion protein for each compound. EC50 values ​​were calculated using GraphPad Prism (Table 7). The potency of IL-2 fusion proteins with antibody A on Treg cells was significantly reduced compared to Proleukin, but the EC50 values ​​on NK and CD8 T cells were comparable. Fusion proteins with antibody B also showed reduced potency on Treg cells compared to Proleukin, but the bias effect decreased with increasing linker length (compound BFC885). IL-2 fused to antibody C signaled Treg cells with high activity comparable to Proleukin. The affinity of the antibody for the substituted IL-2 was required to effectively provide steric hindrance, thereby removing CD25 from the signaling complex (Table 7).

[0155] Table 7. EC50 of STAT5 activation in mouse splenocytes of IL-2-antibody fusion proteins on NK cells, CD8 cells, and Treg cells. [Table 7]

[0156] Applying the representative incorporation process to IL-2-specific antibody A or B yielded fusion proteins with equally favorable IL-2 CD25 binding (Table 5), stimulation of HEK Blue IL-2 reporter cells (Table 6), and reduced potency in STAT5 phosphorylation on Treg cells compared to Proleukin. This confirms that fusion to heavy or light chain variable regions is possible by adjusting the orientation of the circulating IL-2 protein to maintain the natural binding orientation of the anti-IL-2 antibody and cytokine. Antibody B's low affinity for IL-2 (269 nM K D ) was sufficient to confer selective function to the resulting fusion protein. Peptide linkers up to 20 amino acids long retained the desired signaling qualities of IL-2 fusion proteins with antibody A, but shorter peptides were preferred in combination with lower affinity antibodies. Fusion proteins incorporating IL-2CP with antibody C did not eliminate CD25 binding (even with 0-1 amino acid linkers), indicating that an antibody affinity for IL-2 at least equivalent to that of antibody B is required.

[0157] Example 2: Generation of high-affinity, non-blocking anti-PD-1 antibodies Identification of PD-1 agonist and non-competitive anti-hPD-1 antibodies Three anti-human PD-1 antibodies were identified that were described as non-competitive with PD-1 antagonists or non-PD-L1 blocking (Table 8). These antibodies were tested by flow cytometry for competition with commercially available PD-1 antagonist antibodies using a 20-fold molar excess of competitor. Raji PD-1-expressing cells (Invitrogen) were incubated with serial dilutions of pembrolizumab and nivolumab starting at 81 μg / ml for 30 minutes at 4°C (1:3 serial dilutions). After washing the cells, biotin-labeled antibodies listed in column 1 of Table 8 were added at a constant concentration of 2 μg / ml. Bound antibodies were detected with Streptavidin-PE (Biolegend), and the MFI levels of bound compounds were compared to the MFI of samples without competing compounds to determine percent signal inhibition. The background of samples incubated with Streptavidin-PE alone was subtracted from all samples. Antibody XVT458 showed no competition with any of the PD-1 antagonists tested. Antibody ZJN296 showed partial competition with pembrolizumab, exhibiting a 47.4% reduced mean fluorescence intensity (MFI) compared to samples without competitor. Nevertheless, the signal did not show a concentration-dependent decrease, so the decrease in MFI in samples with a 20-fold higher concentration of competitor may be unrelated to the presence of pembrolizumab. Nivolumab did not significantly reduce the signal of antibody ZJN296. Antibody OVL714 completely competed with both pembrolizumab and nivolumab, indicating a possible shared epitope on the antigen.

[0158] K D Decision: Binding of each antibody to PD-1 was assessed by SPR (Biacore 8K, GE Healthcare). Anti-human Fc IgG (Jackson) was immobilized on an amine-reactive CM5 chip (GE Healthcare) using 1x HBS-EP + running buffer (GE Healthcare). Test antibodies were captured at a flow rate of 10 μl / min with a contact time of 30 seconds. Anti-PD-1 against human PD-1 (hPD-1) (AcroBiosystems, PD1-H82E4) and cynomolgus monkey (cyno) PD-1 (cynoPD-1) (AcroBiosystems, PD1-C5223) were run as analytes, diluted in 1x HBS-EP + running buffer (GE Healthcare) and subjected to a 240 s association and 600 s dissociation run at a flow rate of 30 μl / min. Regeneration was performed in 10 mM Glycine-HCl, pH 1.5, at a flow rate of 10 μL / min. This protocol is consistent with the K values ​​of the antibodies defined herein unless otherwise noted. D In all cases used to define K, adapted as necessary with respect to the nature of the interaction partners. D is used to determine the

[0159] Table 8. Antibodies and sources listed as PD-1 antagonists or non-competitive with PD-L1. Percent inhibition of binding signal to cells expressing human PD-1 upon pre-coating with a 20-fold molar excess of the indicated commercially available PD-1 antagonist antibodies. [Table 8]

[0160] Rehumanization and affinity maturation of antibody XVT458 The humanized VH sequence of XVT458, composed of frameworks shuffled from germline genes of different families, was rehumanized to obtain a VH with a matched germline (GH) sequence, resulting in z0-XVT458 (SEQ ID NOs: 053, 054). The rehumanization of the VH of XVT458 resulted in the human IGHV1-18 VH, which showed the highest homology with the framework regions of the parent construct. *01 IGHJ6 * This was achieved by CDR-grafting onto the 01 framework. Mutations of Vernier zone residues were introduced to observe the effect on affinity for the hPD-1 antigen. Germline (GH wild-type or with point mutations) and parental humanized (uVH) were combined with humanized VL (uVL) in a full hIgG1 format (Table 9).

[0161] Binding of antibody candidates to cell surface-expressed antigens was confirmed by flow cytometry. CHO-S hPD-1 cells were incubated with 5-fold serial dilutions of the indicated humanized antibodies. Antibody binding was detected with goat anti-human IgG-PE. MF was plotted against antibody concentration to obtain EC50 values. SPR binding kinetics, FACS EC50 binding values, and maximum MFI values ​​indicated that the antibodies bound to hPD-1-expressing cells, with the exception of z8-uIgGKV326 (Tables 9 and 10).

[0162] Table 9. Anti-PD-1 antibody XVT458 and newly humanized VH derivatives and binding kinetics measured by SPR. [Table 9]

[0163] The humanized antibody z2-XVT458 (SEQ ID NOs: 055 and 056) was affinity matured by inserting random mutations into the CDRs (parsimonious mutagenesis method). Each position in the antibody VH and VL-CDRs was mutated by PCR using mutagenesis primers containing the degenerate codon NNS at specific CDR positions, introducing all 20 amino acid combinations. The single mutant library was screened by scFv capture ELISA with a redundancy factor of 4 (96-well plate / position). Clones showing ELISA signals ≥2-fold higher than those of the wild-type scFv were sequenced, and clones with unique sequences were re-propagated, rescreened by capture ELISA, and ranked by dose-dependent ELISA against human and cyno antigens and FACS using human antigen-expressing cells. Mutants with improved affinity were used to design a combinatorial library using the Kunkel method. The library was screened by scFv capture ELISA with a redundancy factor of 4 (4x the number of combinations). Clones showing ELISA signals ≥2-fold higher than those of wild-type scFv were sequenced. Clones with unique sequences were re-grown, re-screened by capture ELISA, and ranked by dose-dependent ELISA for human and cynoantigens, as well as FACS using human antigen-expressing cells. Primary screening by saturation mutagenesis of antibody CDRs identified 25 mutants at 10 CDR amino acid positions that showed capture ELISA signals for hPD-1 at least 2-fold higher than those of wild-type. Nine mutants showed improved binding to both recombinant hPD-1 and cynoPD-1, as well as to hPD-1 on CHO-S cells. A combinatorial library was created with these nine mutants, yielding 16 combinatorial mutants with improved affinity for recombinant and cell-expressed antigens. Based on the sequence of z2-XVT458 (variable domain SEQ ID NOs: 055 and 056), six subclones, z2-XVT458 m1 to m6 (SEQ ID NOs: 055 and 056), were selected for IgG conversion.In addition to the CDR mutations, the sequences of z2-XVT458-m1, z2-XVT458-m3, and z2-XVT458-m6 contain a G57D mutation immediately after CDRL2, which may contribute to the increased affinity of these clones. Clones selected in the IgG format showed significantly improved affinity for recombinant hPD-1, cynoPD-1, and hPD-1 on cells. Compared to the parent clone z2-XVT458, these clones (z2-XVT458m1-m6) showed improved affinity for both hPD-1 and cynoPD-1 as measured by SPR. off They also showed a 10-84 fold improvement in efficiency (Table 11; corresponding SEQ ID NOs: 057-068). For SPR, the same method as above was used.

[0164] Table 10. Maximum MFI and EC50 values ​​of anti-PD-1 antibodies binding to hPD-1 expressed on CHO-S cells. [Table 10]

[0165] Table 11. Parental XVT458, z2-XVT458 and subclonal affinity matured variants in IgG format. Binding affinities measured by flow cytometry, ELISA and SPR. [Table 11] JPEG2026501338000014.jpg109153

[0166] Humanization of antibody ZJN296 ZJN296 was cloned into the germline vector with the highest identity to the mouse framework (IGKV1-33 * 01 IGKJ2 and IGHV1-18 * 01 IGHJ6 *The VL and VL genes were humanized by CDR grafting using the IgG1A-specific CDR gene (Table 1). To increase the likelihood of retaining binding affinity to the target antigen, eight additional humanized VL genes and two humanized VH genes were designed by mutating human amino acids back to mouse. Such mutations maintain the original structure of the CDR loops of VH and VL, and can maintain antigen binding when these loops contact the antigen (Table 12).

[0167] Table 12. Humanized ZJN296-0 Antibody variants of humanized ZJN296 with mutations inserted into VH (GH, SEQ ID NO: 069) and VL (SEQ ID NO: 070). [Table 12]

[0168] A recombinant humanized antibody variant of ZJN296 was expressed as human IgG1 in HEK293 cells. For SPR analysis (Biacore 8K, GE Healthcare), anti-human Fc IgG (Bethyl, A80-304P) was immobilized on a CM5 chip (Cytiva) using 1x HBS-EP + running buffer. Humanized anti-PD-1 antibodies were captured on the chip. hPD-1 and cynoPD-1 were run as analytes for a 180-second association phase followed by a 400-second dissociation phase using 1x HBS-EP + running buffer at a flow rate of 30 μL / min. 10 mM glycine (pH 1.5) was used as a regeneration buffer and injected into the flow cell following each dissociation phase.

[0169] Based on the SPR binding kinetics of human and cyno antigens and the binding data to hPD-1 expressed in cells tested in the supernatants, the five best candidates were recombinantly expressed and subjected to full SPR kinetics (Table 13). ZJN296-0 and ZJN296-6 were selected as candidates because ZJN296-0 has minimal back mutations and comparable binding avidity compared to the chimeric antibody, and ZJN296-6 has the best binding properties.

[0170] Table 13. SPR kinetic results of humanized purified ZJN296-derived antibodies [Table 13]

[0171] Screening for non-blocking antibody hybridomas A high-affinity binding antibody specific for human PD-1 (hPD-1), cross-reactive with cynomolgus monkey PD-1 (cynoPD-1), but not blocking PD-L1 or PD-1 antagonists, was generated by immunization of humanized mice (AlivaMab® mice) and hybridoma technology. Two separate immunization rounds (10 kappa-lambda mice in the first round, and 4 kappa and 4 lambda mice in the second round) were performed by immunization with the hPD-1 / cynoPD-1 combination. Primary functional screening was performed using HEK293 cells transfected with hPD-1 and cyPD-1. Binding was assessed by flow cytometry. Thirty-one 384-well plates were screened in the first round, and 15 384-well plates were screened in the second round. In the second round of screening, selected clones were first tested for competition with nivolumab and pembrolizumab from supernatants and then tested for binding to antigen-expressing cells by flow cytometry in the presence of competing antibodies and detected with an anti-mouse IgG antibody. Non-competitors were determined as MFI (+competition) / MFI (uncompetition) > 0.6. Only non-competing clones that showed high affinity (KD < 1 nM) on BLI were expanded and purified. Of all screened clones, seven clones were identified that fit the desired binding profile (Table 14, CH 1,2,3The VH and VL of the antibodies used were SEQ ID NOs: 071-084, with the corresponding CL (kappa) or CL (lambda). Table 14 summarizes the binding characteristics to human and cyno PD-1. EC50 binding to HEK293 cells transfected with hPD-1 and cyno PD-1 was measured by flow cytometry. Binding kinetics to recombinant antigens was measured by BLI. The association (220 seconds) and dissociation (480 seconds) of a dilution series of recombinant hPD-1 and cyno PD-1 (100, 25, 6.25 nM, AcroBio) were measured using an antibody-coated biosensor (anti-human-IgG CH1 Biosensor, ForteBio) on an Octet system (Octet RED, ForteBio). D Values ​​were obtained by fitting the kinetic data with ForteBio data analysis software (8.2). PBS was used as the assay buffer (PBS 10 mM phosphate, 150 mM sodium chloride, pH 7.4). For dissociation, the biosensor was immersed in kinetic assay buffer (PBS 10 mM phosphate, 150 mM sodium chloride, 0.1% BSA, 0.02% Tween, pH 7.4). Sequencing of the final clones revealed that clones 21A08, 22F13, and 25I20 share the same HC / LC V-region and have the same CDR3. Clones 20H02, 39F23, 40B20, and 56H02 showed multiple developmental liabilities in the CDR regions. Clone 21A08 was selected as the best clone based on affinity and purity measured by SEC-HPLC compared with other sibling clones.

[0172] Table 14. FACS EC50 for binding to hPD-1 or cynoPD-1 expressing HEK293 cells, SPR binding kinetics for hPD-1 and cynoPD-1. [Table 14]

[0173] Mutations were introduced into the CDR-L1 of 21A08 to remove the N-glycosylation site present in CDR-L1. The first asparagine (Asp, N) in CDR-L1 was replaced with serine (Ser, S), glutamine (Gln, Q), or alanine (Ala, A). Binding of the 21A08 clone variants (21A08S, 21A08Q, and 21A08A) was tested by BLI and compared with the wild-type clone using supernatants from transiently transfected HEK293 cells and a positive control (pembrolizumab).

[0174] Table 15. Binding affinity (KD) and dissociation constant (Koff) of antibody clone 21A08 and deglycosylated variants. [Table 15]

[0175] Additionally, the deamidation site present in CDR-L1 was removed by inserting the mutations shown in Table 16. The binding kinetics of 21A08A and the deamidation site-removed mutants were measured by SPR. For SPR analysis (Biacore 8K, GE Healthcare), anti-human Fc IgG (Bethyl, A80-304P) was immobilized on a CM5 chip (Cytiva), and the test anti-PD-1 antibodies were captured using 1x HBS-EP + running buffer at a flow rate of 10 μL / min. hPD-1 and cynoPD-1 (His-tagged) were run as analytes diluted in 1x HBS-EP + running buffer at a flow rate of 30 μL / min for a 250-second association phase followed by a 3600-second dissociation phase. 10 mM glycine (pH 1.5) was used as a regeneration buffer and injected into the flow cell following each dissociation phase. All clones lacking the N-glycosylation site retained binding to PD-1 (Table 15). None of the mutations in CDR-L1 significantly impaired the binding kinetics to hPD-1 or cynoPD-1.

[0176] Table 16. Antibody clone 21A08A and variants with removed deamidation site: Binding affinity and dissociation constants [Table 16]

[0177] Example 3: Binding Profiles of Non-Blocking Anti-PD-1 Antibodies Next, the binding profiles of non-blocking anti-PD-1 antibodies were evaluated using a human membrane protein cell array. The Retrogenix Cell Microarray Technology platform screened anti-PD-1 antibody candidates for cross-reactive binding to non-target proteins. Test antibodies were individually screened for binding to human HEK293 cells expressing 6018 full-length human membrane proteins, secreted and cell surface-associated human secreted proteins, and an additional 397 human heterodimers. For prescreening, 2 μg / mL of each test antibody or PBS alone was added to slides of fixed, non-transfected HEK293 cells. Binding to non-transfected cells was assessed using an AlexaFluor 647-labeled anti-human IgG Fc detection antibody (AF647 anti-hIgG Fc), followed by fluorescence imaging. Prescreening with antibody XVT458-z2-m5 detected high background. A second prescreening run at 0.5 mg / mL resulted in reduced background. For library screening, 6018 expression vectors encoding both ZsGreen1 and full-length human plasma membrane proteins or cell surface-tethered human secreted proteins were individually arrayed in duplicate on 17 microarray slides ("slide sets"). Additionally, vectors encoding an additional 397 human heterodimers were co-arrayed on additional microarray slides. Human HEK293 cells were used for reverse transfection / expression. Test antibody XVT458-z2-m5 was added to each of the 18 slide sets at a final concentration of 0.5 μg / mL after cell fixation, while the remaining nine test antibodies were added at a final concentration of 2 μg / mL. Binding detection was performed using the same fluorescent secondary antibody (AF647 anti-hIgG Fc) used in prescreening. Fluorescent images were analyzed and quantified (for transfection) using ImageQuant software (GE Healthcare, version 8.2). Protein "hits" were defined as duplicate spots showing elevated signal compared to background levels.This was achieved by visual inspection using gridded images in ImageQuant software. Hits were classified as "strong, moderate, weak, or very weak" based on the intensity of the duplicate spots. 29 libraries were hit in the screening. After removing 12 interactions observed with the test antibodies and control treatments designated as nonspecific, and an additional 7 low-confidence interactions, 10 specific interactions for the test antibodies were identified (Table 17). All test antibodies showed a single specific interaction with PD-1 (PDCD1). Test antibodies ZJN296-0, ZJN296-6, XVT458-z2-m1, XVT458-z2-m2, XVT458-z2-m4, and XVT458-z2-m5 all showed moderate to weak interactions with other target proteins. Clones 21A08Ap1 (sequence numbers VH 085, VL 086), 21A08Ap2 (sequence numbers VH 085, VL 087), XVT458-z2-m3 (sequence numbers 061, 062), and XVT458-z2-m6 (sequence numbers 067, 068) showed a single specific interaction with the target of interest.

[0178] Table 17. Hits from the Retrogenix cell microarray screen. Protein type: plasma membrane (PM), secreted (S), tethered secreted (TS), heterodimer (HD), or evidence of plasma membrane (M). Hit number: very weak excess intensity. [Table 17]

[0179] Example 4: Non-competitive binding of anti-PD-1 antibodies to PD-L1 antagonists and PD-L1 Competition in hPD-1-expressing cells Antibodies 21A08Ap1 and 21A08Ap2 were tested for competition with commercially available PD-1 antagonist antibodies by flow cytometry. Jurkat PD-1-expressing cells were incubated with serial dilutions of pembrolizumab and nivolumab starting at 10 μM (1:3 serial dilutions) at 4°C for 30 minutes. Without washing the cells, biotin-labeled antibodies 21A08Ap1 and 21A08Ap2 were added at a constant concentration of 100 nM. Bound antibodies were detected with Streptavidin-PE (Biolegend), and the MFI levels of bound compounds were compared to the MFI of samples without competing compounds to determine percent signal inhibition. The background of samples incubated with Streptavidin-PE alone was subtracted from all samples. Inhibition was minimal at less than 20%, and neither pembrolizumab nor nivolumab significantly reduced the signal of antibodies 21A08Ap1 and 21A08Ap2 (Table 18).

[0180] Table 18. Competition of cell-expressed hPD-1 with commercially available PD-1 antagonists pembrolizumab and nivolumab. Percent flow cytometry signal inhibition of test antibodies using 100-fold molar excess of competitor. [Table 18]

[0181] Simultaneous binding of hPD-1 to 21A08Ap1 and PD-L1 Simultaneous binding of PD-1 (Fc-tagged) to anti-PD-1 antibodies and PD-L1 (His-tagged, Acrobiosystems, #H52H3) was assessed by SPR analysis using a Biacore T200 (Cytiva, #28975001). PD-L1 was captured on an anti-His-CM5 chip (Cytiva, chip: #29104988, His capture kit: #28995056). 1 μM PD-1 was premixed with 5 μM antibody (pembrolizumab, IgG1 isotype control antibody, or 21A08p1) or running buffer (Xantec, HBSTE:B HBSTE10) for at least 30 minutes before binding analysis (60 s association and dissociation). Data were analyzed using Biacore Insight Evaluation Software (Cytiva, V4.0.8 #29310606). Double reference subtraction was performed (surface without ligand and running buffer injection). Figure 3 shows the Biacore sensorgram. No binding (as a change in RU) was observed for PD-1 premixed with pembrolizumab, confirming that pembrolizumab blocks the PD-L1 binding site on PD-1. In contrast, the complex of PD-1 and antibody 21A08Ap1 was still able to bind to PD-L1, giving a signal of approximately 60 RU. Controls of PD-1 alone and PD-1 with an IgG1 isotype control showed a relative signal of approximately 30 RU.

[0182] Example 5: Rational design of a non-blocking PD-1-targeted IL-2 fusion protein To determine the most suitable format for adding a non-blocking PD-1 targeting moiety to a CD122 / CD132 dimeric receptor-biased IL-2 anti-IL-2 antibody fusion moiety, proof-of-concept compounds were designed and characterized. The anti-PD-1 antibody XVT458 or ZJN296 (Table 8) was combined with the antibody A-IL-2 fusion protein (QTY065, SEQ ID NOS: 051 and 052) to generate a bispecific compound that specifically delivers CD122-CD132-biased IL-2 to PD-1-expressing cells. Five formats with different sizes and valencies to the PD-1 antigen were designed (Figure 4, Table 19). The dual scFv fusion contained the anti-PD-1 antibody XVT458 in scFv format, i.e., the VH and VL domains fused by a 15-amino acid glycine (G)-serine (S) linker, fused to the QTY065 IL-2 antibody fusion scFv fused by a G4S linker (Figure 4A). The Fab-duplex scFv consisted of the QTY065 antibody IL-2 fusion as a Fab fragment (IL-2-VL-CL and VH-CH1) fused at their respective C-termini to two anti-PD-1 scFv fragments of the XVT458 antibody (via the N-terminal VH-(G4S)3-VL (Figure 4B)). The IgG CrossMab format was a heterotetrameric human IgG1 with HC1 and LC1 from the anti-PD-1 antibody XVT458 or ZJN296, and HC2 and LC2 from the antibody-IL-2 fusion QTY065 (Figure 4C). Knob-into-hole mutations (Y407T on HC1 and T366Y on HC2) were used to ensure correct heavy chain pairing. The CL and CH1 domains of HC1 and LC1 were swapped to enhance correct light chain pairing (WO2009080253A1). For Fc silencing of the IgG1 Fc region, the mutations L234A, L235A, and P329G (WO2012130831A1) were inserted into both HC1 and HC2. The IgG-scFv fusion protein contains antibody XVT458 (consisting of HC1, identical LCs, and HC2) as a fully human IgG1 with HC2 fused at either the N-terminus (4E) or C-terminus (4D) to the IL-2 antibody fusion QTY065 as an scFv (e.g., IL-2-VL 抗体A -(G4S)3-VH抗体A -(G4S)2-HC2 XVT458 or HC2 XVT458 -(G4S) 2 / 4 -IL-2-VL 抗体A -(G4S)3-VH 抗体A ) Pairing to HC1 was improved by knob-and-hole mutations (as in CrossMab) and Fc silencing with L234A, L235A, and P329G mutations. Bispecific antibodies were produced in Expi297 cells as described for the IL-2 fusion proteins above.

[0183] The functionality of the bispecific antibodies was assessed by a sandwich ELISA, which relies on binding to the target antigen (coated hPD-1) and the integrity of fusion IL-2 via a secondary antibody (anti-IL-2 clone 5344). 60 nM hPD-1 (ECD-His, in-house) was coated onto Maxisorp plates (Nunc) overnight at 4°C and blocked with 5% BSA in PBS. The bispecific antibodies were serially diluted in assay buffer and detected with biotinylated 5344 and streptavidin-HRP (BD Pharmingen, 554066). After the addition of TMB, the absorbance signal was read at 450 nm on a plate reader (Spectramax ID3). EC50 values ​​were determined by blotting absorbance versus concentration (Graphpad Prism, sigmoidal curve fit, 4PL, logarithmic scale). All compounds demonstrated binding to the bispecific site; the bivalent binding formats to hPD-1 exhibited lower EC50 values ​​compared to the monovalent formats (Table 19). Binding to hPD-1 expressed on the surface of Jurkat-PD-1 cells was confirmed by flow cytometry (Table 19). Binding to hPD-1 was maintained in all formats. As expected, compounds with two PD-1-binding domains exhibited increased binding to hPD-1 due to an avidity effect.

[0184] Table 19. Anti-PD-1, anti-hIL-2-IL-2 bispecific antibodies in various formats with: valency, molecular weight (Mw), EC50 against hPD-1 measured by ELISA, binding to hPD-1 expressed on Jurkat-hPD-1 cells measured by flow cytometry (fold MFI over background). [Table 19]

[0185] Example 6: Cell selectivity in vitro and in vivo To evaluate the functionality of the anti-IL-2 / IL-2 fusion protein arm of the bispecific compounds, STAT5 phosphorylation was analyzed in mouse splenocytes as a downstream signaling pathway of IL-2R activation. pSTAT5 was measured in different cell populations after stimulation with Proleukin or one of the bispecific compounds. Mouse splenocytes were incubated with serial dilutions of the bispecific compounds diluted in RPMI + 10% FBS. Starting at a concentration of 100 nM, the dilutions were 1:5, resulting in six total concentrations. Samples were incubated at 37°C for 15 minutes, and an equal volume of cytofix buffer (BD Biosciences, catalog no. 554655) was added per sample. The samples were then immediately fixed after 10 minutes of incubation at 37°C. After fixation, cells were stained with BV421 rat anti-mouse CD25 (clone PC61, BD Biosciences, 0.5 μL / sample), BV650 hamster anti-mouse CD3e (clone 145-2C11, BD Biosciences, 1 μL / sample), and BV711 mouse anti-mouse NK1.1 (clone PC136, BD Biosciences, 0.33 μL / sample) for 30 minutes at room temperature, followed by permeabilization with Perm Buffer III (BD Biosciences, catalog no. 558050) for 10 minutes on ice. The secondary staining included PE-CF594 rat anti-mouse CD4 (clone RM4-5, BD Biosciences, 0.25 μL / sample), APC-780 rat anti-mouse CD8b (clone H35-17.2, eBioscience, 0.167 μL / sample), AF488 rat anti-mouse FoxP3 (clone FJK-16, eBioscience, 0.5 μL / sample), and AF647 mouse anti-mouse pSTAT5 (clone pY694, BD Bioscience, 20 μL / sample). All samples were acquired using Cytek® Aurora's SpectroFlo® software. ".fcs" files were analyzed using FlowJo_v10.6.2. EC50 values ​​for the percentage of pSTAT5+ cells were obtained by sigmoidal 4PL fitting. The percentage of pSTAT5+ cells was plotted against the logarithmic concentration (M) for each cell population and each test compound.

[0186] The obtained EC50 values ​​are shown in Table 20. All formats except the IgG C-terminal scFv bispecific showed comparable (or up to 10-fold lower) potency than Proleukin in inducing STAT5 phosphorylation in NK and CD8 T cells (Table 20). EC50 values ​​for QWT744 and CUM013 could not be determined because the percentage of pSTAT5-positive cells did not reach maximal levels at the highest concentration of test compound (100 nM). All compounds showed reduced potency in inducing pSTAT5 in Treg cells compared to Proleukin, indicating that signaling through the high-affinity trimeric IL-2R is inhibited in vitro (Table 20).

[0187] Table 20. EC50 values ​​of bispecific compounds, QTY065, and Proleukin against pSTAT5 positive parental NK, CD8, and Treg cells measured by flow cytometry in mouse splenocytes. [Table 20]

[0188] To confirm in vivo selectivity for CD122-CD132-expressing cells, the bispecific compound was administered intravenously at 0.2 mg / kg to wild-type C57BL / 6 mice. Blood samples were collected before injection and on days 3 and 6 after compound administration. Cell proliferation was analyzed by flow cytometry, and Ki67+ cells were detected in CD8 T cells, NK cells, and Treg cells (Figure 5). C57BL / 6 mice received a single intravenous injection of 0.2 mg / kg of the compound. Blood samples collected before compound administration and on days 3 and 6 after compound administration were processed and analyzed by flow cytometry. Cells were initially stained for CD25, CD3, NK1.1, CD4, CD8, and Zombie aqua fixable viability dye (Biolegend). After fixation and permeabilization (eBioscience), intracellular staining for Ki67 and FoxP3 was performed. All compounds showed only a slight increase in Ki67 expression in Treg cells on day 3, with levels decreasing by day 6 after compound administration. The proliferation (Ki67 expression) of CD8 T cells and NK cells in the blood of animals treated with bispecific compounds in IgG format (FQQ111 and ONG682) or IgG-N-ter scFv (YIM345) format was comparable to that of QTY065, reaching maximal levels of Ki67+ cells (>75%) by day 6 after administration. Compounds with low molecular weights (MW) and lacking the Fc domain, such as HRL470 and VNP090, showed Ki67+ levels of 22–25% in CD8 T cells and 33–40% in NK cells. The bispecific format of C-terminal fusion of anti-hIL-2 / IL-2 scFv on anti-PD-1 IgG (CUM013) showed lower potency compared to other Fc domain-bearing compounds with 47% Ki67+ NK cells and 25% Ki67+ CD8 T cells, which surprisingly suggests that IL-2 signaling is less potent when the cytokine is C-terminally linked to the target antibody domain (Figure 5).

[0189] Example 7: Antitumor effect in vivo To examine the additional protection conferred by the non-blocking PD-1 targeting moiety compared to anti-IL-2 / IL-2 fusion proteins lacking the PD-1 target, we chose the B16F10 melanoma model, an aggressive tumor model resistant to immune checkpoint inhibitors (CPi). Immune cell proliferation and immunophenotyping of B16F10 subcutaneous tumors were performed in humanized PD-1 mice (C57BL / 6N-Pdcd1tm1 (huPDCD1-ICP11; Geno)), allowing assessment of both PD-1 targeting and IL-2 activity. Transgenic hPD-1 mice were inoculated with 1 × 10 6 B16F10 cells were injected subcutaneously. 3 When the tumors reached an average size of 100 mg / kg, the compound was administered intravenously at a dose of 0.2 mg / kg. Tumor volume was measured daily with a caliper and calculated using the formula (length x width x depth) / 2. At the end of the study, 6 days after the start of administration, the difference in tumor volume (%) relative to the vehicle was calculated. Tumor-infiltrating cells were analyzed by flow cytometry as described above using the antibodies listed in Table 21.

[0190] Table 21. Antibodies used for intracellular (IC) and extracellular (surface) staining for immunophenotyping of B16F10 tumors. [Table 21]

[0191] Administration of the nontargeted anti-IL-2 / IL-2 fusion protein QTY065 primarily led to an increase in intratumoral NK cells (8.2-fold compared to vehicle) in tumors of treated mice, while all PD-1-targeted compounds promoted a significant increase in CD8 T cells in B16F10 tumors. The smaller format HRL470 with a dual scFv fusion had the least effect (2.7-fold compared to vehicle), while IgG and IgG-scFv fusions led to a 4.6- to 8.1-fold increase in CD8 T cells compared to vehicle. Notably, the effect was enhanced within the target cell population, CD8+PD-1+ cells (Table 22). The C-terminal scFv compound CUM013, while not as potent in previous assays, demonstrated comparable efficacy to other bispecific formats in tumor-infiltrating lymphocyte (TIL) activity. Nevertheless, CUM013 was less effective in suppressing tumor growth than the other bispecifics (Table 23). Based on the results obtained, it was concluded that the heterotetrameric IgG (CrossMab format) and IgG-N'-scFv-IL-2 fusion format bispecific molecules are particularly effective in vivo, and the VNP090 and CUM013 formats were not pursued further and were shelved as potential back-ups.

[0192] Table 22. TILs from transgenic hPD-1 mice treated with non-targeting or PD-1-targeting IL-2 antibody fusion proteins as fold increase in cells / gram tumor relative to vehicle. Tumors analyzed on day 6 (n=4). [Table 22]

[0193] Table 23. Difference in tumor volume (%) compared to vehicle-treated mice on day 6. [Table 23]

[0194] In an additional study, tumor growth in hPD-1 transgenic mice was monitored after two doses of the bispecific compound compared to vehicle, a non-blocking PD-1 antibody lacking IL-2 binding (antibody XVT458 as an Fc-silenced human IgG1), or the non-targeting bivalent antibody-IL-2 fusion QTY065. On day 0, tumors were 70-100 mm 3 This corresponds to the start of the study, when tumor volume reached 1000 mg / kg. On days 0 and 3, 1.25 nmole / kg of each compound was administered intravenously, and tumor growth was monitored. At the end of the study, the difference in tumor volume was compared to vehicle-treated mice (Table 24). Five days after treatment, tumors were treated and TILs were analyzed by flow cytometry for intracellular staining, as shown in Table 21. We determined the doubling of cell numbers per gram of tumor compared to vehicle for Treg cells, NK cells, CD8 T cells, and CD8 T cell subfamilies, including stem-like pre-exhausted CD8 T cells (CD8+PD-1+TCF1+), better effector-exhausted T cells (CD8+TCF1-CX3CR1+), and terminally exhausted T cells (CD8+TCF1-CX3CR1-). The control antibody XVT458 induced only a minimal increase in tumor cell infiltration. Bispecific compounds particularly induced an expansion of CD8+PD-1+ T cells and their derived subfamilies, with heterotetrameric IgG CrossMab and IgG-scFv N-terminal formats being the most effective compounds in terms of target cell targeting and antitumor potency (Tables 24 and 25).

[0195] Table 24. Difference (%) in tumor volume compared to vehicle-treated mice on day 5 (n=5-6). [Table 24]

[0196] Table 25. TILs from transgenic hPD-1 mice treated with non-targeted or PD-1-targeted IL-2 antibody fusion proteins as fold increase in cells per gram tumor relative to vehicle. Tumors analyzed 5 days after initiation of treatment. [Table 25]

[0197] Example 8: Increased in vitro potency of PD-1-targeted anti-hIL-2 / IL-2 fusion proteins Selected bispecific formats were then combined with fully functional non-blocking anti-hPD-1 antibodies (Table 26). Antibodies targeting the unrelated antigens MDC982 and KVC110 were used as non-targeting control antibodies. The anti-hIL-2 / IL-2 arm of QTY065 possesses a light chain of the kappa subfamily, which can be combined with a second arm of the lambda subfamily to form heterotetrameric compounds (bispecific antibodies with different specificities in each antigen-binding domain), eliminating the need for genetic engineering strategies such as CrossMab to ensure correct light chain pairing. The IL-2-anti-IL-2 fusion portion of QTY065 was combined with two representative non-blocking high-affinity PD-1 antibodies, 21A08Ap1 and 21A08Ap2 (both possessing lambda light chains), to design a bispecific in a heterotetrameric IgG kappa / lambda (Figure 4F) format. Constructs encoding the sequences disclosed in Table 26 were transfected into Expi293 cells and produced and purified as described above. The knob-in-hole mutations were Y407T in HC1 and T366Y in HC2, and the Fc-silencing mutations L234A, L235A, and P329A were incorporated.

[0198] Table 26. Heterodimeric combinations of QTY065 VH, VL(IL-2), and a second anti-hPD1 (or control) antibody in three different bispecific antibody formats. [Table 26]

[0199] Bispecific compounds were tested for their potency in inducing STAT5 phosphorylation in PD-1+ Jurkat cells expressing IL-2R CD122-CD132. Jurkat-PD1+ CD122+ cells were activated with serial dilutions of bispecific compounds for 15 minutes at 37°C and fixed with an equal volume of Cytofix buffer (BD Biosciences) for 10 minutes at 37°C. For intracellular antigen staining, cells were permeabilized with ice-cold Perm buffer III (BD Biosciences) for 15 minutes on ice. Phosphorylated STAT5 was stained using anti-p-STAT5 pY694 antibody (clone 47 / Stat5, BD Biosciences). Flow cytometry was performed as previously described. EC50 values ​​were calculated using Graphpad Prism v9.3.1 using the equation Y = Bottom + (X^Hillslope). * Calculations were made using (Top-Bottom) / (X^HillSlope + EC50^HillSlope) ([antagonist] vs. response-variable slope (4 parameters)). PD-1-targeted bispecific compounds, with the exception of YPW986 and PXU588, showed increased potency and lower EC50 values ​​compared to the bivalent QTY065 and the non-targeted compound TSQ225 (Table 27). Binding ELISA assays demonstrated that YPW986 and PXU588 (i.e., heterotetramers in a CrossMab bispecific format, with one arm derived from QTY065 and one arm from either 2108Ap1 or 21A08Ap2) lost binding affinity to PD-1, in contrast to heterotetrameric kappa / lambda IgG formats using the same non-blocking anti-PD-1 clones XWY176 and GQM289, which retained the ability to stimulate the dimeric IL-2 receptor. In general, PD-1 anchoring to cells increased the IL-2 signaling potency of the bispecific compounds. The heterotetrameric CrossMab bispecific IgG FQQ111 with a non-affinity-matured anti-PD-1 arm (XVT458) induced a smaller increase in potency upon targeting to PD-1, indicating that higher affinity binding is desirable.

[0200] Table 27. EC50 values ​​for STAT5 phosphorylation in Jurkat-PD-1+ CD122+ cells. [Table 27]

[0201] An additional functional effect of the fusion proteins of the present invention is the reduction of cell surface PD-1 upon binding to IL-2R. Signaling through IL-2R leads to the internalization of dimeric or trimeric receptor complexes (Robb RJ. et al., J Exp. Med. (1987) doi:10.1084 / jem.165.4.1201). The reduction of cell surface hPD-1 and CD122 by the immunoconjugates of the present invention was tested by incubating stimulated PBMCs (i.e., PD-1+) with three different concentrations of the bispecific antibody for 16 hours. As controls, the bivalent anti-hIL-2 / IL-2 compound QTY065 and a non-targeting IL-2 compound, IgG CrossMab, were used. Buffy coats or whole blood from healthy volunteers were received from the Blutspendezentrum SRK beider Basel or Aarau in accordance with the Swiss ethical committee, the Swiss Human Research Act (HRA; May 2021), and other applicable ethical regulations. PBMCs were isolated by density gradient centrifugation using Ficoll Plaque Plus (GE Healthcare). Frozen PBMCs were thaw-activated with plate-coated anti-CD3 (clone OKT3, BioLegend) and soluble anti-CD28 (clone CD28.2, BioLegend) for 3 days. After activation, PBMCs were incubated with the indicated bispecific compounds overnight at 37°C. Immediately after incubation, an equal volume of Cytofix buffer (BD Biosciences) was added, and cells were fixed for 10 minutes at 37°C. Surface markers were then stained using the antibodies listed in Table 28. Cells were acquired by flow cytometry as described above.

[0202] Table 28. Antibodies used for extracellular staining for cell surface expression experiments. [Table 28]

[0203] The difference in MFI of detected surface CD122 and PD-1 was calculated as the percentage reduction compared to PBMCs incubated with medium alone. Control compounds induced a reduction in surface CD122, but did not change PD-1 levels. PD-1-targeting compounds reduced both CD122 and PD-1 levels. This may indicate that binding to IL-2R is required for the reduction of cell surface PD-1 (Table 29).

[0204] Table 29. Percent reduction in CD122 and PD-1 MFI of CD8 T cells after incubation with control compounds or bispecific compounds compared to CD8 T cells incubated with media alone (n=2 donors). [Table 29]

[0205] Example 9: Improved PD-1 binders: in vivo antitumor efficacy and subcutaneous tumor immunophenotyping The improved bispecific compounds in Table 26 were tested in hPD-1 transgenic mice bearing subcutaneous B16F10 tumors. Transgenic hPD-1 mice (C57BL / 6N-Pdcd1tm1(huPDCD1-ICP11)Geno) were inoculated with 1x10 6 B16F10 cells were injected subcutaneously. 3When tumors reached an average size of 1000 nmoles (day 0), the compound was administered intravenously at 1.25 nmoles / kg. A second administration was administered on day 3. Five days after the start of the study, mice were sacrificed and tumors were excised. Tumors were treated with a GentleMACS Octo Dissociator (Milteny), and cells were stained with NIR live dead stain (ThermoFisher, L10119). Cells were incubated in Fc block (TruStain FcX™, Biolegend, 101320) for 10 minutes before surface staining (see Table 21). Cells were fixed and permeabilized with FoxP3 Staining Buffer Set (ebioscience, #00-5523-00), followed by intracellular staining according to Table 21. All PD-1-targeting anti-IL-2 / IL-2 bispecific antibodies significantly increased intratumoral CD8 T cell numbers, especially PD-1+ stem-like T cell and derivative numbers, compared with non-targeting compounds or vehicle (Table 30). The CD8 T cell / Treg ratio was increased in tumors in mice treated with the bispecific compound compared to vehicle or non-targeting bispecific.

[0206] Table 30. TILs from transgenic hPD-1 mice administered non-targeting or PD-1-targeting IL-2 antibody fusion proteins as fold increase in cells per gram tumor relative to vehicle. Tumors analyzed on day 5. [Table 30]

[0207] In efficacy studies, transgenic hPD-1 mice (C57BL / 6N-Pdcd1tm1(huPDCD1-ICP11)Geno) were injected with 1 × 10 6 B16F10 cells were injected subcutaneously. 3When tumors reached an average size of 1.25 nmoles / kg (day 0), compounds were administered intravenously at 1.25 nmoles / kg. Booster doses were administered on days 3 and 7. Tumors were measured daily, and the volume was calculated using the formula (length x width x depth) / 2. The difference in tumor volume between mice administered vehicle or each compound was reported as a percentage reduction compared to vehicle. All PD-1-targeted IL-2 fusion protein compounds effectively reduced tumor volume (Table 31).

[0208] Table 31. Day 7 tumor volume difference from vehicle (%). [Table 31]

[0209] Example 10: Stress test with anti-PD-1, anti-IL-2 / IL-2 bispecific antibody The heterotetrameric bispecific antibodies XWY176, TMU471, QAB373, and MDS446 were engineered with modified knob-in-hole mutations (S354C, T366W / S354C, T366S, L368A, Y407V) and Fc silencing (Table 32).

[0210] Table 32. Bispecific compounds with knob-in-hole mutations and Fc silencing [Table 32]

[0211] The four bispecific antibodies in Table 32 were exposed to heat, pH, oxidation, and freeze-thaw stress conditions. The compounds were then tested by SEC-HPLC (Table 33), iCEIF (Table 34), and functional ELISA (Table 35). Post-translational modifications were analyzed by mass spectrometry. The bispecific antibodies XWY176, TMU471, QAB373, and MDS446 were exposed to different stress conditions: i) incubation at 40°C for 1 or 2 weeks, ii) three or four freeze-thaw cycles, iii) 0.1% (v / v) HO for 4 or 24 hours, iv) low pH 3.5 for 24 or 48 hours, and v) high pH 9.0 for 24 or 48 hours. To evaluate protein changes after the stress conditions, the compounds were tested by SEC-HPLC, iCEIF, and LC-MS. The functionality of the bispecific antibodies was assessed by a sandwich ELISA, which relies on binding to the target antigen (hPD-1) and the integrity of fusion IL-2 via a secondary antibody (anti-IL-2 clone 5344). 60 nM hPD-1 (ECD-His, in-house) was coated onto Maxisorp plates (Nunc) overnight at 4°C and blocked with 5% BSA in PBS. The bispecific antibodies were serially diluted in assay buffer and detected with biotinylated 5344 and streptavidin-HRP (BD Pharmingen, 554066). The end-of-absorbance signal after the addition of TMB was read at 450 nm on a plate reader (Spectramax ID3). EC50 values ​​were determined by blotting absorbance versus concentration (Graphpad Prism, sigmoidal curve fit, 4PL, logarithmic scale). The change in EC50 value of unstressed samples (TO) was compared to stressed samples.

[0212] All constructs showed high potential for further development based on the criteria evaluated in this example. SEC-HPLC and iCIEF indicated that the conditions that induced the most protein changes were heat stress and high pH. The construct that showed a smaller increase in high molecular weight (HMW) species by SEC-HPLC was NZA596. The same construct also showed a smaller change in charged species by iCIEF. ELISA functionality assessment showed that all proteins retained at least 40% functionality after exposure to different stress conditions. Constructs NZA596 and CIT348 retained greater than 60% functionality under all conditions tested. PTM analysis showed oxidation of all four methionines under heat stress and oxidative conditions, with NZA596 performing slightly better.

[0213] Example 11: In vitro immunogenicity of anti-PD-1, anti-IL-2 / IL-2 bispecific antibodies The in vitro immunogenicity of NZA596, CIT348, BGY642, and KTX917 was tested by MHC-associated peptide proteomics (MAPPS) performed at Lonza. Monocyte-derived dendritic cells (DCs) from 10 different healthy donors (carrying mutations in HLA-DRB alleles) were incubated with 50 μg / ml of the protein of interest and then matured with LPS for 24 hours. After maturation, DCs were lysed, and the membrane fraction containing the HLA:peptide complexes was solubilized and incubated overnight at 4°C with protein A mag 63 epharose beads (Cytiva) coated with anti-HLA-DR antibody (Lonza). The following morning, the beads were washed with TBS, and peptides were eluted from the HLA-DR complexes with 0.1% TFA. Finally, peptides were purified through a 10 kDa molecular weight cutoff spin column and analyzed by MS. Peptide analysis revealed "immunogenic hotspots" that corresponded to HLA-presented peptides from two or more donors and were not present in the antibody's natural framework. Table 36 summarizes the immunogenic hotspots identified for each compound and the location of the peptides within the protein. The two non-blocking anti-PD-1 binding domains tested produced similar hotspot profiles. While all formats had acceptably low immunogenicity, those with fewer mutation regions and peptide linkers, most notably the kappa / lambda format, had the most favorable immunogenicity profiles.

[0214] Table 33. SEC-HPLC of bispecific compounds at time 0 (T0) or after different stress conditions: 40°C for 1 or 2 weeks, freeze-thaw cycles, oxidative conditions (0.1% H2O2) for 4 or 24 hours, low pH (pH 3.5) and high pH (pH 9.0) for 4 or 24 hours. HMW: high molecular weight; LMW: low molecular weight. [Table 33]

[0215] Table 34. SEC-HPLC of bispecific compounds at time point (T0) or after different stress conditions: 1 week (1W) or 2 weeks (2W) at 40°C, freeze-thaw cycles, oxidative conditions (0.1% H2O2) for 4 or 24 hours, low pH (pH 3.5) and high pH (pH 9.0) for 4 or 24 hours. [Table 34]

[0216] Table 35. Functional ELISA of bispecific compounds at time point 0 (T0) or after different stress conditions: full functionality at T0 set as 100% and EC50 values ​​(%) of stressed samples with EC50 at T0, 2 weeks at 40°C, 24 hours in oxidative conditions (0.1% H2O2), 24 hours at low pH (pH 3.5) and high pH (pH 9.0). [Table 35]

[0217] Table 36. Immunogenicity hot spots of bispecific compounds [Table 36]

[0218] Example 12: Mass spectrometry demonstrating correct light chain pairing To test for correct light chain pairing in bispecific constructs purified in the IgG CrossMab or IgG kappa / lambda formats, IgdE enzyme digestion was performed and the Fab fragments were analyzed by mass spectrometry. For each assayed Fab, a single peak corresponding to the theoretical molecular weight of the correctly assembled light and heavy chains was observed (Figure 6). No incorrect light chain pairing was detected.

[0219] Example 13: Cis-signaling in combination with PD-1 blocking antibodies An assay was established to assess whether exemplary bispecific constructs as disclosed herein transmit IL-2 to CD122-CD132 on PD-1-expressing cells bound by their anti-PD-1 arms (cis signaling) or to neighboring cells (trans signaling). Jurkat-PD-1+ CD122+ cells were labeled with either CFSE (Invitrogen, C34557) or CTV (Invitrogen, C34554). CFSE-labeled cells were exposed to 700 nm of the non-competitive parent antibodies against PD-1, pembrolizumab or nivolumab, to block the PD-1 epitope for 30 minutes at room temperature. After two washes, CTV-labeled and CFSE-labeled cells were mixed at a 1:1 ratio and activated with the bispecific compound NZA596 (1 nM) for 15 minutes at 37°C. Immediately after the stimulation period, cells were fixed with an equal volume of Cytofix buffer (BD Biosciences 554655) at 37°C for 10 minutes. For intracellular antigen staining, cells were permeabilized with ice-cold Perm buffer III (BD Biosciences) for 15 minutes on ice. Phosphorylated STAT5 was stained using anti-p-STAT5 pY694 antibody (clone 47 / Stat5, BD Biosciences). Immunoconjugate-mediated induction of STAT5 phosphorylation in these cells was analyzed. Cells preincubated with the parent antibody significantly reduced the efficacy of the immunoconjugate. This interference was not observed in cells pre-exposed to pembrolizumab or nivolumab (Table 37, Figure 7). Furthermore, the potency of the immunoconjugate on cells not previously exposed to any PD-1-binding antibody was the same for all samples (incubated with CFSE+ pre-blocked cells), indicating that the immunoconjugate signals in a cis manner on the same cells where PD-1 binding occurs (Table 37, Figure 7).

[0220] Table 37. Percentage of pSTAT5 compared to unblocked (non-blocked) cells after stimulation with 1 nM NZA596: [Table 37]

[0221] Example 14: Antitumor efficacy of a bispecific immunoconjugate in combination with pembrolizumab and nivolumab in mouse models of two cancers The efficacy of an exemplary bispecific immunoconjugate (NZA596) as disclosed herein was tested as monotherapy or in combination with the commercially available PD-1 blockers pembrolizumab and nivolumab. hPD-1 transgenic C57BL / 6 mice were injected subcutaneously with B16F10 melanoma cells or MC38 colon tumor cells. Tumors were 70-100 mm 3 Mice were randomized when tumors reached an average size of 100 μg / mL, and treatment was initiated as shown in Table 38. Mice were dosed on day 0 (the day of randomization) and day 3. Pembrolizumab or nivolumab was also administered on day 7. Tumor growth inhibition was calculated relative to tumor volume in vehicle-treated mice on day 13 for mice bearing MC38 tumors and day 14 for mice bearing B16F10 tumors. In both tumor models, the bispecific immunoconjugate in combination with either of the checkpoint inhibitors induced significant tumor growth retardation compared to vehicle, even at low doses. The combination therapy resulted in more potent tumor growth inhibition compared to either monotherapy.

[0222] Table 38. Dosing schedule and doses for in vivo efficacy studies in hPD-1 transgenic mice bearing B16F10 or MC38 tumors. Tumor growth inhibition (TGI) was calculated as the percentage reduction in volume compared to the mean volume of the vehicle-treated group. * Dose 1: MC38 model: 0.1 mg / kg, B16F10 model: 0.2 mg / kg ** Dose 2: MC38 model: 0.2 mg / kg, B16F10 model: 2 mg / kg: 0.4 mg / kg. (n) = number of animals. [Table 38]

[0223] Example 15: Efficacy on pSTAT5 signaling in the presence of various existing PD-1 blockers After 15 minutes of incubation with a dilution series of an exemplary bispecific immunoconjugate (NZA596), STAT5 phosphorylation in Jurkat PD-1+ CD122+ cells was analyzed by flow cytometry. Prior to stimulation, cells were incubated for 30–60 minutes with either growth medium alone, a pre-existing PD-1 blocker, or the parental anti-PD-1 antibody 21A08Ap1, all at saturating concentrations (200 nM). NZA596 was added to the cells without washing out the anti-PD-1 IgG, and pSTAT5+ cells were detected by flow cytometry. The EC50 of the bispecific immunoconjugate NZA596 on stimulated Jurkat PD-1+ CD122+ cells was calculated to be 0.85 nM by plotting the concentration against the percentage of pSTAT5+ cells. This value significantly increased when the parental anti-PD-1 antibody 21A08Ap1 was used for pre-blocking of PD-1 in the cells. Preincubation of cells with pembrolizumab, nivolumab, cemiplamab, dostallimab, zelvalimab, tislelizumab, ezabenlimab, toripalimab, or cetrelimab did not alter the potency of NZA596 in inducing pSTAT5, demonstrating the ability of these antibodies to block the PD-L1-PD-1 axis without altering the activity of the bispecific immunoconjugates described herein (Table 39). The distinct binding epitopes of PD-1 were also demonstrated by SPR, which demonstrated simultaneous binding of NZA596 and a commercially available PD-1 blocking antibody to immobilized recombinant PD-1 (data not shown).

[0224] Table 39. List of anti-PD-1 antibodies pre-incubated with Jurkat PD-1+ CD122+ cells prior to stimulation with the bispecific immunoconjugate NZA596. Potency (EC50) values ​​in nM and fold reduction based on pre-incubation with medium. [Table 39]

[0225] Example 16: Bispecific immunoconjugates induce expansion of tumor-specific subsets of T cells in the tumor microenvironment In the context of tumors, PD-1 marks T cells chronically exposed to tumor antigens and thus marks a subset of T cells with relevant anti-tumor specificity. To gain a better understanding of the T cell repertoire expanded by the bispecific immunoconjugates of the present invention, we performed T cell receptor beta (TCRβ) chain sequencing of TILs from hPD-1 mice bearing B16F10 tumors to assess the diversity of the TCR repertoire. Transgenic hPD-1 mice were injected subcutaneously with B16F10 cells. Tumors were 50-100 mm. 3 On study day 0, when tumors reached a size of 100 μg, mice were randomized and intravenously administered 0.2 mg / kg of the bispecific immunoconjugate (targeting PD-1), 0.2 mg / kg of the non-PD-1-targeting immunoconjugate (QTY065), or vehicle, followed by a second dose on day 3. Mice were sacrificed on day 5, and genomic DNA was extracted from tumors for sequencing of the TCRβ repertoire. Mice treated with the PD-1-targeting bispecific immunoconjugate showed a reduced TCR repertoire, resulting from the expansion of a subset of T cell clones derived from PD-1+ cells present before treatment that may be focused and tumor antigen-specific (Figure 8). In contrast, the non-PD-1-targeting immunoconjugate induced an increased enrichment of the TCR repertoire compared to vehicle, indicating that this non-targeting IL-2Rβ / γ antagonist expanded a more diverse set of T cell clones, including many specificities not associated with the tumor.

[0226] To quantify the effect of bispecific immunoconjugates on CD8+ T cells with defined tumor-neoantigen specificity, the kinetics of antigen-specific T cell responses in TILs from mice bearing MC38 or B16F10-OVA were analyzed by MHC dextramer and flow cytometry. Wild-type C57BL / 6 mice bearing B16F10-OVA or MC38 tumors were injected with 0.2 mg / kg of the murine surrogate bispecific immunoconjugate BGY642 (designated DSQ964, which targets mouse PD-1 with the antigen-binding domain of clone RMP1-30 and does not interfere with PD-1 binding to inhibitory ligands), a non-targeting immunoconjugate (QTY065), or vehicle on day 0 post-randomization (tumor volumes 50–100 mm). 3 ) and 3 days after administration. TILs were isolated on day 5 and analyzed by flow cytometry using dextramer, which specifically stains CD8+ T cells that recognize the ovalbumin neoantigen expressed by the B16F10-OVA tumor cell line and the p15E retroviral antigen expressed by the MC38 tumor cell line. In both tumor models, tumor neoantigen-specific T cells were increased in mice treated with DSQ964 compared with vehicle or a non-targeting IL-2 / anti-IL-2 fusion protein (Table 40). Furthermore, all tumor antigen-specific CD8 T cells were found to be PD-1+. Combined with the TCR sequencing data, these experiments confirmed that the PD-1-targeting bispecific immunoconjugate increases a specific subset of CD8+ T cells that recognize tumor cells in a mouse model of cancer.

[0227] Table 40. Fold increase in intratumoral Dextramer+CD8+PD-1+ T cells in mice treated with DSQ964 or QTY065 relative to mice administered vehicle in B16F10-OVA (n=6 for vehicle, n=8 for DQS964, QTY065) or MC38 tumors (n=9). [Table 40]

[0228] JPEG2026501338000044.jpg207153JPEG2026501338000045.jpg189153JPEG2026501338000046.jpg224153JPEG2026501338000047.jpg201153JPEG2026501338000048.jpg201153JPEG2026501338000049.jpg208153JPEG2026501338000050.jpg199153JPEG2026501338000051.jpg216153JPEG2026501338000052.jpg163153JPEG2026501338000053.jpg193153JPEG2026501338000054.jpg203153JPEG2026501338000055.jpg190153JPEG2026501338000056.jpg200153JPEG2026501338000057.jpg193153JPEG2026501338000058.jpg189153JPEG2026501338000059.jpg195153JPEG2026501338000060.jpg175153JPEG2026501338000061.jpg167153JPEG2026501338000062.jpg209153JPEG2026501338000063.jpg196153JPEG2026501338000064.jpg141153JPEG2026501338000065.jpg220153JPEG2026501338000066.jpg159153JPEG2026501338000067.jpg159153JPEG2026501338000068.jpg127153JPEG2026501338000069.jpg222153JPEG2026501338000070.jpg195153JPEG2026501338000071.jpg160153JPEG2026501338000072.jpg219153JPEG2026501338000073.jpg219153JPEG2026501338000074.jpg209153JPEG2026501338000075.jpg227153JPEG2026501338000076.jpg223153JPEG2026501338000077.jpg168153.

Claims

1. An immunoglobulin variable domain capable of binding to PD-1, an antibody heavy chain variable domain polypeptide (PD1-VH), and - antibody light chain variable domain polypeptide (PD1-VL), Including, - PD1-VH comprises an HCDR1 having the sequence GFTFSINAMT (SEQ ID NO:118), an HCDR2 having the sequence TISGSGFSTYYADSLKGR (SEQ ID NO:119), and an HCDR3 having the sequence EVYGDY (SEQ ID NO:120); and - PD1-VL has the sequence SGX １ SSNIGSX ２ X ３ VF (SEQ ID NO: 121), LCDR1 having the sequence SNNQRPS (SEQ ID NO: 122), and LCDR3 having the sequence AAWDDSLSIWVF (SEQ ID NO: 123); It is characterized by the fact that and where X １ is N, S, Q, or A, and X ２ X ３ is NS, QS, SS, or NA.

2. 2. The immunoglobulin variable domain of claim 1, wherein the binding of an antibody comprising the immunoglobulin variable domain defined in claim 1 to PD-1 is reduced by no more than 20% in the presence of a 100-fold molar excess of a PD-1-specific antibody selected from the group consisting of pembrolizumab and nivolumab.

3. 3. The immunoglobulin variable domain of claim 1 or 2, wherein PD1-VL comprises an LCDR1 having the sequence SGASSNIGSQSVF (bold underlined) (SEQ ID NO: 124), SGASSNIGSSSVF (bold underlined) (SEQ ID NO: 125), or SGASSNIGSNAVF (bold underlined) (SEQ ID NO: 126).

4. 4. The immunoglobulin variable domain of any one of claims 1-3, wherein PD1-VH comprises or consists of a polypeptide at least (≧) 95% or more, 98% or more, or 99% or more similar to SEQ ID NO:085, and PD1-VL comprises or consists of a polypeptide at least (≧) 95% or more, 98% or more, or 99% or more similar to a sequence selected from SEQ ID NO:086, SEQ ID NO:087, or SEQ ID NO:

091.

5. 5. The immunoglobulin variable domain of any one of claims 1 to 4, wherein the PD1-VH comprises or consists of a polypeptide having the sequence of SEQ ID NO:085, and the PD1-VL comprises or consists of a polypeptide having a sequence selected from the list consisting of SEQ ID NO:086, SEQ ID NO:087, or SEQ ID NO:

091.

6. The affinity constant (K) of an antibody characterized by the immunoglobulin variable domain for PD-1 Ｄ ) is measured by surface plasmon resonance and is found to be 1.0 × 10 －９ ~1.5 x 10 －１１ mol / L, especially 1.0 x 10 －１０ ~1.5 x 10 －１１ mol / L, more particularly 5.0×10 －１０ ~1.5 x 10 －１１ The immunoglobulin variable domain according to any one of claims 1 to 5, wherein the immunoglobulin variable domain has a binding affinity of 0.01 to 0.1 mol / L.

7. An immunoconjugate comprising an immunoglobulin variable domain capable of binding to PD-1 as defined in any one of claims 1 to 6.

8. The immunoconjugate of claim 7, further comprising an immunoactive polypeptide ligand capable of binding to a cell surface molecule expressed by T cells or natural killer cells.

9. The immunoconjugate of claim 8 , wherein the immunoactive polypeptide ligand comprises an interleukin and / or an immunoglobulin variable domain reactive to an interleukin.

10. 10. The immunoconjugate of claim 8 or 9, wherein the immunoactive polypeptide ligand comprises or consists of an interleukin and an immunoglobulin variable domain reactive with said interleukin.

11. The immunoconjugate of claim 10, wherein one or two peptide linkers link the interleukin to the immunoglobulin variable domain reactive with said interleukin.

12. The immunoconjugates comprise a fragment crystallizable (Fc) immunoglobulin domain, in particular an IgG Fc domain, more particularly an IgG Fc characterized by the presence of one or more modifications to the constant region of the heavy chain to enhance correct heavy chain pairing, even more particularly: - Knob: S354C, T366W, and holes: Y349C, T366S, L368A, Y407V; - Knob: T366Y, and hole: Y407T; - knob: Y349C, T366W and hole: S354C, T366S, L368A, Y407V; or - Knob: T366W, and holes: Y407A, T366S, L368A IgG Fc characterized by the presence of a set of knob and hole modifications selected from The immunoconjugate of any one of claims 7 to 11, comprising:

13. The immunoconjugate is a heterotetrameric IgG comprising: - a heterodimer of a first antibody heavy chain and light chain comprising an immunoglobulin variable domain capable of binding to PD-1 as defined in any one of claims 1 to 6, characterized in that the light chain is a lambda light chain; and wherein the immunologically active polypeptide ligand comprises or consists of: a heterodimer of a second antibody heavy and light chain comprising an immunoglobulin variable domain reactive against a cell surface molecule expressed by an immune cell, characterized in that the light chain is a kappa light chain; The immunoconjugate according to any one of claims 8 to 12.

14. The immunoconjugate is an immunoglobulin single chain variable fragment (scFv) format comprising: an anti-PD1 antibody comprising a heterodimer of a first antibody heavy chain and a light chain, and a heterodimer of a second antibody heavy chain and a light chain, each comprising an anti-PD-1 immunoglobulin variable domain as defined in any one of claims 1 to 4; and wherein the immunologically active polypeptide ligand comprises or consists of: an interleukin and an immunoglobulin scFv domain reactive against said interleukin, wherein the interleukin is covalently linked to the immunoglobulin scFv domain reactive against said interleukin to provide a single contiguous recombinant polypeptide; The immunoconjugate according to any one of claims 8 to 12.

15. The immunoconjugate of any one of claims 9 to 14, wherein the interleukin is an IL-2 polypeptide or a circularly permuted IL-2 (IL2CP) polypeptide.

16. The immunoconjugate of any one of claims 7 to 15, wherein the immunoactive polypeptide ligand comprises a polypeptide having the sequence of SEQ ID NO: 167 and a polypeptide having the sequence of SEQ ID NO:

043.

17. The immunoconjugate: - SEQ ID NO: 111, SEQ ID NO: 095, SEQ ID NO: 112 and SEQ ID NO: 052; - SEQ ID NO: 094, SEQ ID NO: 095, SEQ ID NO: 100, and SEQ ID NO: 052; - SEQ ID NO: 097, SEQ ID NO: 098, SEQ ID NO: 100, and SEQ ID NO: 052; SEQ ID NO: 094, SEQ ID NO: 095, and SEQ ID NO: 96; or - SEQ ID NO: 097, SEQ ID NO: 098, and SEQ ID NO: 99 An immunoconjugate according to any one of claims 7 to 16, comprising or consisting of a polypeptide having the sequence:

18. An isolated nucleic acid encoding the immunoconjugate of any one of claims 7 to 17; in particular, said isolated nucleic acid being contained in a mammalian expression vector under the control of a promoter operable in mammalian cells.

19. 18. The immunoconjugate defined in any one of claims 7 to 17 for use in the treatment of cancer patients receiving concurrent treatment with an anti-PD-1 antagonist antibody, in particular an anti-PD-1 antagonist antibody selected from the list consisting of nivolumab, pembrolizumab, dostarlimab, sintilimab, tislelizumab, cemiplimab, cetrelimab, or southernlimab, in particular nivolumab or pembrolizumab.

Images