Bispecific checkpoint inhibitor antibodies

Abstract

The present invention relates to bispecific heterodimeric checkpoint antibodies.

Description

[0001] This application is a divisional application, the parent application of which has the application number 201780045861.3, the application date of which is June 14, 2017, and the invention title is "Bispecific checkpoint inhibitor antibody".

[0002] Cross-reference of related applications

[0003] This application claims priority to U.S. Provisional Patent Application No. 62 / 350,145, filed June 14, 2016; U.S. Provisional Patent Application No. 62 / 353,511, filed June 22, 2016; and U.S. Provisional Patent Application No. 62 / 420,500, filed November 10, 2016, the contents of which are expressly and completely incorporated herein by reference in their entirety.

[0004] sequence list

[0005] This application contains a sequence list submitted electronically in ASCII format and incorporated herein by reference in its entirety. The ASCII text, created on June 9, 2017, is named 067461_5191WO_SL.txt and has a size of 32,442,145 kilobytes. Background Technology

[0006] Checkpoint receptors such as CTLA-4, PD-1 (programmed cell death 1), TIM-3 (T cell immunoglobulin and mucin domain 3), LAG-3 (lymphocyte-activating gene 3), and TIGIT (a T cell immune receptor with Ig and ITIM domains) inhibit the activation, proliferation, and / or effector activity of T cells and other cell types. Based on the hypothesis that checkpoint receptors suppress endogenous T cell responses against tumor cells, preclinical and clinical studies of anti-CTLA4 and anti-PD1 antibodies (including nivolumab, pembrolizumab, ipilimumab, and tremelimumab) have indeed demonstrated that checkpoint blockade produces memorable anti-tumor responses, stimulating endogenous T cells to attack tumor cells, resulting in long-term cancer remission in a small percentage of patients with various malignancies. Unfortunately, only a small percentage of patients respond to these therapies, with response rates varying generally from 10% to 30% and sometimes higher, depending on indications and other factors. Combinations of these drugs, such as ipilimumab plus nivolumab, have produced even higher response rates, approaching 60% in some cases. Preclinical studies have shown additional synergy between anti-PD-1 and / or anti-CTLA-4 antibodies, blocking more recently identified checkpoint receptors, including LAG-3, TIM-3, BTLA, and TIGIT. While the potential for multiple checkpoint blockade is very high, combination therapies using such drugs are expected to incur a high financial burden. Moreover, combination therapies, such as nivolumab plus ipilimumab, have significantly increased autoimmune toxicity compared to monotherapy, thus discontinuing treatment for many patients.

[0007] Multiple studies examining tumor-infiltrating lymphocytes (TILs) (Ahmadzadeh et al., *Blood* 114:1537 (2009), Matsuzaki et al., *Proceedings of the National Academy of Sciences (PNAS)* 107(17):7875-7880 (2010), Fourcade et al., *Cancer Res.* 72(4):887-896 (2012), and Gros et al., *Journal of Clinical Invest.* 124(5):2246 (2014)) have shown that TILs co-express multiple checkpoint receptors. Moreover, TILs expressing multiple checkpoints are likely to be the most tumor-reactive. In contrast, peripheral non-tumor-reactive T cells are more likely to express a single checkpoint. For tumor-reactive TILs that promote autoimmune toxicity under the de-repression hypothesis, checkpoint blockade using monospecific full-length antibodies may be indistinguishable from autoantigen-reactive mono-expressing T cells.

[0008] Therefore, the present invention relates to bispecific antibodies that bind to two different checkpoint inhibitory proteins. Summary of the Invention

[0009] This invention provides bispecific heterodimeric antibodies that bind to two different checkpoint cell surface receptors, such as human PD-1, human CTLA-4, human TIM-3, human LAG-3, and human TIGIT. Therefore, in some aspects, suitable bispecific antibodies bind to PD-1 and CTLA-4, PD-1 and TIM-3, PD-1 and LAG-3, PD-1 and TIGIT, PD-1 and BTLA, CTLA-4 and TIM-3, CTLA-4 and LAG-3, CTLA-4 and TIGIT, CTLA-4 and BTLA, TIM-3 and LAG-3, TIM-3 and TIGIT, TIM-3 and BTLA, LAG-3 and TIGIT, LAG-3 and BTLA, and TIGIT and BTLA.

[0010] In one aspect, the present invention provides a bottle opener format comprising: a) a first monomer (“scFv monomer”, sometimes referred to as “scFv heavy chain”) comprising a scFv having a variable heavy structure domain and a variable light structure domain connected by charged scFv connectors (the +H sequence of FIG7 is preferred in some embodiments), an Fc structure domain including the tilted variant S364K / E357Q and the ablation variant E233P / L234V / L235A / G236del / S267K, and as described herein... The Fv that binds to the checkpoint receptor as described herein; b) a second monomer (“Fab monomer” or “heavy chain”) comprising an Fc domain having tilted variants L368D / K370S, pI variants N208D / Q295E / N384D / Q418E / N421D, and ablation variants E233P / L234V / L235A / G236del / S267K, and a variable heavy domain, the variable heavy domain and the variable light domain constituting an Fv that binds to the second checkpoint inhibitor as outlined herein; and c) a light chain. In this particular embodiment, suitable monomer Fv pairs comprise (Fab is listed first, scFv is listed second) PD-1 and C1LA-4, CTLA-4 and PD-1, PD-1 and TIM-3, TIM-3 and PD-1, PD-1 and LAG-3, LAG-3 X PD1, PD-1 and TIGIT, TIGIT and PD-1, PD-1 and BTLA, BTLA and PD-1, CTLA-4 and TIM-3, TIM-3 and CTLA-4, CTLA-4 and LAG-3, LAG-3 and CTLA-4, CTLA-4 and TIGIT, TIGIT and CTLA-4, CTLA-4 and BTLA, BTLA and CTLA-4, TIM-3 and LAG-3, LAG-3 and TIM-3, TIM-3 and TIGIT, TIGIT and TIM-3, TIM-3 and BTLA, BTLA and TIM-3, LAG-3 and TIGIT, TIGIT and LAG-3, LAG-3 and BTLA, BTLA and LAG-3, BTLA and TIGIT, and TIGIT and BTLA.

[0011] Other aspects of the invention are provided herein. Attached Figure Description

[0012] Figures 1A to 1ISeveral formats of the present invention are described. The first is a "bottle opener" format having first and second anti-antigen binding domains. Additionally, mAb-Fv, mAb-scFv, central scFv, central Fv, single-arm central scFv, single scFv-mAb, scFv-mAb, and double scFv formats are shown. For all the depicted scFv domains, they can be N-terminally to C-terminally variable heavy-(optional linker)-variable light, or vice versa. Furthermore, for all single-arm scFv-mAbs, the scFv can be linked to the N-terminus of the heavy chain monomer or the N-terminus of the light chain.

[0013] Figure 2 Figure 2A , Figure 2B , Figure 2C and Figure 2D The antigen sequences of several antigens used in this invention are described, in many cases including human and cynomolgus monkey, to facilitate the development of antigen-binding domains that bind to both human and cynomolgus monkey for clinical development.

[0014] Figures 3A to 3F A set of useful heterodimerization variants (including tilt and pI variants) is described. Figure 3E Among them, there are variants that do not have a corresponding "monomer 2" variant; these are pI variants that can be used alone on the monomer or, for example, included on the Fab side of the bottle opener, and appropriately charged scFv linkers can be used on a second monomer in which scFv is used as a second antigen-binding domain. Suitable charged linkers are shown in Figure 7.

[0015] Figure 4 A list depicting the constant regions of isotropic variant antibodies and their corresponding substitutions is provided, with pI_(-) indicating lower pI variants and pI_(+) indicating higher pI variants.

[0016] These can be optionally and independently combined with other heterodimerization variants of the invention (and other variant types, as outlined herein).

[0017] Figure 5 Useful ablation variants (sometimes referred to as "knockout" or "KO" variants) that ablate FcγR binding are described. Typically, ablation variants are found on both monomers, but in some cases, ablation variants may be found on only one monomer.

[0018] Figure 6 illustrates how the present invention can be used. Figure 1A or Figure 1F Two particularly useful embodiments of the format. For Figure 1A The format, in this embodiment, shows the "non-Fv" component. Figure 37AHowever, other formats can also be used (as well as the format in Figure 38).

[0019] Figure 7 depicts multiple charged scFv linkers used to increase or decrease the pI of heterodimeric antibodies utilizing one or more scFvs as components. (+H) positive linkers are specifically used herein, and specifically, anti-CD3vl and vh sequences are shown here. A single prior art scFv linker with a single charge is referred to as "Whitlow," from Whitlow et al., *Protein Engineering* 6(8): 989-995 (1993). It should be noted that this linker is used to reduce polymerization and enhance the proteolytic stability of scFvs.

[0020] Figure 8 A list of engineered heterodimer tilted Fc variants with heterodimer yields (determined by HPLC-CIEX) and thermal stability (determined by DSC) is presented. Undetermined thermal stability is indicated by "nd".

[0021] Figures 9A to 9ESelected numbers of PD-1 ABDs are depicted, with additional anti-PD-1 ABDs listed as SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394, and SEQ ID NO: 36127 to 36146. CDRs are underlined, scFv connectors are double-underlined (in the sequence, the scFv connector is the positively charged scFv(GKPGS)4 connector (SEQ ID NO: 37755), but as those skilled in the art will understand, this connector can be replaced by other connectors, including uncharged or negatively charged connectors, some of which are depicted in Figure 7), and slashes indicate one or more boundaries of the variable structural domains. Additionally, the naming convention shows the orientation of the scFv from the N-terminus to the C-terminus. That is, “H1.279_L1.194” shows an orientation of vh-scFv connector-vl (from N-end to C-end, with optional domain connectors on one or both sides, depending on the format used), but these sequences can also be used with the reverse orientation (from N-end to C-end) vl-connector-vh. Similarly, “L1.194_H1.279” shows an orientation of vl-scFv connector-vh (from N-end to C-end, again with optional domain connectors), and the reverse orientation is also included in this invention. As noted herein and as is the case with each sequence containing a CDR herein, the precise identification of the CDR position can vary slightly depending on the numbering used as shown in Table 1, and therefore, this document includes not only underlined CDRs, but also CDRs contained within vh and vl domains using other numbering systems. Furthermore, as with all sequences in the figures, these vh and vl sequences can be used in scFv or Fab format.

[0022] Figures 10A to 10PPMultiple CTLA-4 ABDs are depicted, with additional anti-CTLA-4 ABDs listed as SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818, and SEQ ID NO: 35395 to 35416. CDRs are underlined, scFv connectors are double-underlined (in the sequence, the scFv connector is the positively charged scFv(GKPGS)4 connector (SEQ ID NO: 37755), but as those skilled in the art will understand, this connector can be replaced by other connectors, including uncharged or negatively charged connectors, some of which are depicted in Figure 7), and slashes indicate one or more boundaries of variable structural domains. As shown above, the naming conventions indicate the orientation of scFv from the N-end to the C-end; in the sequences listed in this figure, they are all oriented vh-scFv connector-vl (from the N-end to the C-end), but these sequences can also be used with the opposite orientation (from the N-end to the C-end) vl-connector-vh; additionally, some sequences in SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818, and SEQ ID NO: 35395 to 35416 are in the opposite orientation. As noted herein and as is the case with each sequence containing a CDR herein, the precise identification of the CDR position can vary slightly depending on the numbering used as shown in Table 1, and therefore this document includes not only underlined CDRs, but also CDRs contained within the vh and vl domains using other numbering systems. Furthermore, as with all sequences in the figure, these vh and vl sequences can be used in either scFv or Fab format. Specifically, many of the XENP identifiers in the accompanying figures include both scFv and Fab formats; see, for example, an illustration showing XENP19235 as a molecule using the Fab format and XENP19769 as an scFv molecule. Figure 10A .

[0023] Figures 11A to 11NMultiple LAG-3 ABDs are depicted, with additional anti-LAG-3 ABDs listed as SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793, and SEQ ID NO: 32794 to 33002. CDRs are underlined, scFv connectors are double-underlined (in the sequence, the scFv connector is a positively charged scFv(GKPGS)4 connector, but as those skilled in the art will understand, this connector can be replaced by other connectors, including uncharged or negatively charged connectors, some of which are depicted in Figure 7), and slashes indicate one or more boundaries of variable structural domains. As shown above, the naming convention indicates the orientation of scFv from the N-end to the C-end; in the sequences listed in this figure, they are all oriented vh-scFv connector-vl (from the N-end to the C-end), but these sequences can also be used with the opposite orientation (from the N-end to the C-end) vl-connector-vh; in addition, some sequences in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793, and SEQ ID NO: 32794 to 33002 are in the opposite orientation. As noted herein and as is the case with each sequence containing a CDR herein, the precise identification of the CDR position can vary slightly depending on the numbering used as shown in Table 1, and therefore this document includes not only underlined CDRs, but also CDRs contained within the vh and vl domains using other numbering systems. Furthermore, as with all the sequences in the attached figure, these vh and vl sequences can be used in either scFv or Fab format.

[0024] Figures 12A to 12CMultiple BTLAABDs are depicted, with additional anti-BTLAABDs listed as SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738. CDRs are underlined, scFv connectors are double-underlined (in the sequence, the scFv connector is a positively charged scFv(GKPGS)4 connector, but as those skilled in the art will understand, this connector can be replaced by other connectors, including uncharged or negatively charged connectors, some of which are depicted in Figure 7), and slashes indicate one or more boundaries of variable structural domains. As shown above, the naming convention indicates the orientation of scFv from the N-end to the C-end; in the sequences listed in this figure, they are all oriented as vh-scFv connector-vl (from the N-end to the C-end), but these sequences can also be used with the opposite orientation (from the N-end to the C-end) vl-connector-vh; additionally, some sequences in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738 are in the opposite orientation. As noted herein and as is the case with each sequence containing a CDR herein, the precise identification of the CDR position can vary slightly depending on the numbering used as shown in Table 1, and therefore this document includes not only underlined CDRs, but also CDRs contained within the vh and vl domains using other numbering systems. Furthermore, as with all sequences in the figure, these vh and vl sequences can be used in either scFv or Fab format.

[0025] Figures 13A to 13IMultiple TIM-3 ABDs are depicted, with additional anti-TIM-3 ABDs listed as SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698, and SEQ ID NO: 36347 to 36706. CDRs are underlined, scFv connectors are double-underlined (in the sequence, the scFv connector is a positively charged scFv(GKPGS)4 connector, but as those skilled in the art will understand, this connector can be replaced by other connectors, including uncharged or negatively charged connectors, some of which are depicted in Figure 7), and slashes indicate one or more boundaries of variable structural domains. As shown above, the naming conventions indicate the orientation of scFv from the N-end to the C-end; in the sequences listed in this figure, they are all oriented vh-scFv connector-vl (from the N-end to the C-end), but these sequences can also be used with the opposite orientation (from the N-end to the C-end) vl-connector-vh; additionally, some sequences in SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698, and SEQ ID NO: 36347 to 36706 are in the opposite orientation. As noted herein and as is the case with each sequence containing a CDR herein, the precise identification of the CDR position can vary slightly depending on the numbering used as shown in Table 1, and therefore this document includes not only underlined CDRs, but also CDRs contained within the vh and vl domains using other numbering systems. Furthermore, as with all sequences in the figure, these vh and vl sequences can be used in either scFv or Fab format.

[0026] Figures 14A to 14I The amino acid sequence of a specific anti-CTLA-4 X anti-PD-1 antibody in an opener format (Fab-scFv-Fc) is described. Antibodies are named using the following sequence: first the Fab variable region and second the scFv variable region separated by dashes, followed by the chain name (Fab-Fc heavy chain, scFv-Fc heavy chain, or light chain). The CDR is underlined, and slashes indicate one or more boundaries of the variable region. The scFv domain has different orientations (N-terminus to C-terminus) as shown (vh-linker-vl or vl-linker-vh), but this can be reversed. Additionally, each sequence outlined herein may or may not contain one or preferably both M428L / N434S variants of the Fc domains, resulting in a longer serum half-life.

[0027] Figures 15A to 15KThe amino acid sequence of the specific anti-LAG-3 X anti-PD-1 Fab-scFv-Fc bispecific antibody was described. The antibody was named using the following sequence: first the Fab variable region and second the scFv variable region separated by dashes, followed by the chain name (Fab-Fc heavy chain, scFv-Fc heavy chain, or light chain). The CDR is underlined, and slashes indicate one or more boundaries of the variable region. The scFv domain has a vl-linker oriented (N-terminus to C-terminus) towards the vh-linker, but this can be reversed. Additionally, each sequence outlined herein may or may not contain one or preferably one of the M428L / N434S variants in both Fc domains, resulting in a longer serum half-life.

[0028] Figure 16 depicts the amino acid sequence of the specific anti-BTLA X anti-PD-1 Fab-scFv-Fc bispecific antibody. Antibodies are named using the following sequence: first the Fab variable region and second the scFv variable region separated by dashes, followed by the chain name (Fab-Fc heavy chain, scFv-Fc heavy chain, or light chain). CDRs are underlined, and slashes indicate one or more boundaries of the variable regions. The scFv domain has a vl-linker oriented (N-terminus to C-terminus) towards the vl-linker-vh, but this can be reversed. Additionally, each sequence outlined herein may or may not contain one or preferably one of the M428L / N434S variants in both Fc domains, resulting in a longer serum half-life.

[0029] Figure 17 depicts the amino acid sequence of the specific anti-LAG-3 X anti-CTLA-4 Fab-scFv-Fc bispecific antibody. Antibodies are named using the following sequence: first the Fab variable region and second the scFv variable region separated by dashes, followed by the chain name (Fab-Fc heavy chain, scFv-Fc heavy chain, or light chain). The CDR is underlined, and slashes indicate one or more boundaries of the variable regions. The scFv domain has a vh-linker-vl oriented (N-terminus to C-terminus), but this can be reversed. Additionally, each sequence outlined herein may or may not contain one or preferably one of the M428L / N434S variants in both Fc domains, resulting in a longer serum half-life.

[0030] Figure 18Some results of anti-LAG-3 hybridoma screening are shown. 1 pg of human LAG-3-hIg in 10 μL was mixed with 50 μL of hybridoma supernatant (diluted 8 times 2-fold with 10% FBS in RPMI medium) at room temperature for 20 min. 40 μL of Daudi or Ramos cells (endogenously expressing MHC-II) were added and incubated at 4°C for 30 min. The cells were then washed and incubated with an anti-human-Fc-Alexa647 secondary antibody for 30 min. The cells were then washed again and analyzed against Alexa647 by FACS.

[0031] Figure 19A and Figure 19B The assay depicts the release of cytokines (A: IL-2, B: IFNγ) after SEB stimulation of human PBMCs and treatment with anti-CTLA-4X anti-PD-1 bispecific antibody.

[0032] Figures 20A to 20C The CD45+ and CD8+ events were described on day 14 after human PBMCs were transplanted into NGS mice on day 0 and subsequently administered the test sample as shown on day 1.

[0033] Figure 21A and Figure 21B The T-cell binding in SEB-stimulated PBMC assays was described via chimeric antibodies generated from anti-TIM-3 hybridomas.

[0034] Figure 22 This paper describes some engineered anti-TIM-3 antigen-binding domain data from three experiments. It depicts the XENP encoding of the bivalent embodiment, the derived clones, the names of the vh and vl engineered domains, the KD binding constant, and the association and dissociation constants of human TIM-3 as measured by the octet.

[0035] Figures 23A to 23N Some engineered data for anti-PD-1 antigen-binding domains are depicted. This depicts the XENP encoding of the bivalent and scFv embodiments, the names of the vh and vl engineered domains, the scFv orientation (N-terminus to C-terminus), the KD binding constant of human PD-1 measured by octet, and the Tm of the scFv.

[0036] Figures 24A to 24GResults of some CTLA-4 Fab screenings were depicted. This included the XENP coding of Fab and scFv embodiments, the naming of the vh and vl engineered domains, the KD binding constants of human and cynomolgus monkey CTLA-4 measured by octet, and the Tm of scFv and Fab. Additionally, multiple sequence 9-mers precisely matched at least one human VH or VL germline were depicted as measures of the human origin of the variable regions of Fab and scFv.

[0037] Figure 25 A mixed lymphocyte response was described that enhances IL-2 release by: nivolumab alone (an anti-PD-1 monoclonal antibody, as...) (sold separately), ipilimumab (anti-CTLA-4 monoclonal antibody, as) (Sales), prototype anti-CTLA-4 x anti-PD-1 bispecific antibody based on nivolumab and ipilimumab arms, and "single-arm" combination control.

[0038] Figure 26 Mixed lymphocyte responses that enhance IL-2 release were characterized by: anti-CTLA-4 x anti-PD-1 bispecific antibody with a variant anti-CTLA-4 Fab arm and a variant anti-PD-1 scFv arm, as well as nivolumab alone, ipilimumab alone, and the prototype anti-CTLA-4 x anti-PD-1 bispecific antibody based on nivolumab and ipilimumab as controls.

[0039] Figure 27 The enhancement observed by the anti-CTLA-4 x anti-PD-1 bispecific antibody in NSG mice transplanted with human PBMCs (as measured by human CD45 counting) is shown. This enhancement is greater than the enhancement seen using nivolumab (XENP16432) alone (dashed line).

[0040] Figure 28 The correlation between body weight and CD45 cell count in graft-versus-host disease was depicted, thus demonstrating that CD45 cell levels predict disease.

[0041] Figure 29 Depicting Figure 27 The study depicted the correlation between CD45 cell count and IFNγ release.

[0042] Figure 30 The enhancement observed by the anti-CTLA-4 x anti-PD-1 bispecific antibody in NSG mice transplanted with human PBMCs (as measured by human CD45 counting) is shown. This enhancement is greater than the enhancement seen using nivolumab (XENP 16432) alone (dashed line).

[0043] Figure 31 Depicting Figure 30 The study depicted the correlation between CD45 cell count and IFNγ release.

[0044] Figure 32 It shows Figure 27 and Figure 30 The comparison of test results between the studies described demonstrates a consistent advantage of the anti-PD-1x and anti-CTLA-4 bispecific checkpoint antibodies over nivolumab alone.

[0045] Figure 33A and Figure 33B Results of mixed lymphocyte responses are shown to evaluate anti-CTLA-4 x anti-PD-1, anti-LAG-3 x anti-PD-1, and anti-LAG-3 x anti-CTLA-4 bispecific antibodies. Analyte levels were normalized to those induced by nivolumab alone (values ​​greater than one indicate enhancement relative to nivolumab).

[0046] Figure 34 The SEB reaction was shown to evaluate the anti-LAG-3 x anti-CTLA-4 bispecific antibody. Compared to the control, the anti-LAG-3 x anti-CTLA-4 bispecific antibody alone enhanced the IL-2 response, but not as much as nivolumab alone. However, the anti-LAG-3 x anti-CTLA-4 bispecific antibody in combination with nivolumab produced a significantly higher IL-2 response than either antibody alone.

[0047] Figure 35 The study demonstrated that the bispecific antibodies against CTLA-4 x anti-PD-1, anti-LAG-3 x anti-PD-1, anti-BTLA x anti-PD-1, and anti-LAG-3 x anti-CTLA-4 enhanced transplantation in NSG mice transplanted with human PBMCs (as measured by human CD45 counting). This enhancement was greater than that seen using nivolumab (XENP 16432) alone. Moreover, the combination of the anti-LAG-3 x anti-CTLA-4 bispecific antibody and nivolumab produced the highest level of transplantation.

[0048] Figure 36A and Figure 36B This demonstrates that anti-BTLA x anti-PD-1 bispecific antibodies require disruption of the HVEM / BTLA interaction to possess depressant activity equivalent to nivolumab.

[0049] Figures 37A to 37ESequences of several useful opener-format backbones based on human IgG1 are shown, without Fv sequences (e.g., scFV and vh and vl on the Fab side). Opener backbone 1 is based on human IgG1 (356E / 358M allotype) and includes the S364K / E357Q:L368D / K370S tilted variant, the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side, and the E233P / L234V / L235A / G236del / S267K ablation variant on both strands. The bottle opener backbone 2 is based on human IgG1 (356E / 358M allotype) and contains different tilted variants, N208D / Q295E / N384D / Q418E / N421D pI variants on the Fab side, and E233P / L234V / L235A / G236del / S267K ablation variants on both chains. The bottle opener backbone 3 is based on human IgG1 (356E / 358M allotype) and contains different tilted variants, N208D / Q295E / N384D / Q418E / N421D pI variants on the Fab side, and E233P / L234V / L235A / G236del / S267K ablation variants on both chains. The bottle opener backbone 4 is based on human IgG1 (356E / 358M allotype) and contains various tilted variants, N208D / Q295E / N384D / Q418E / N421D pI variants on the Fab side, and E233P / L234V / L235A / G236del / S267K ablation variants on both chains. The bottle opener backbone 5 is based on human IgG1 (356D / 358L allotype) and contains S364K / E357Q:L368D / K370S tilted variants, N208D / Q295E / N384D / Q418E / N421D pI variants on the Fab side, and E233P / L234V / L235A / G236del / S267K ablation variants on both chains. Bottle opener backbone 6 is based on human IgG1 (356E / 358M allotype) and includes the S364K / E357Q:L368D / K370S tilted variant, the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side, and the E233P / L234V / L235A / G236del / S267K ablation variant on both strands, as well as the N297A variant on both strands. Bottle opener backbone 7 is identical to bottle opener backbone 6 except that the mutation is N297S. Alternative formats for bottle opener backbone 6 and bottle opener backbone 7 may exclude the ablation variant E233P / L234V / L235A / G236del / S267K from both strands.Main chain 8 is based on human IgG4 and contains the S364K / E357Q:L368D / K370S tilted variant, the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains, as well as the S228P (EU designation, which is S241P in Kabat) variant ablated at Fab arm exchange on both chains, as is known in the art. Alternative formats for the opener main chain 8 may omit the ablation variant E233P / L234V / L235A / G236del / S267K from both chains. Main chain 9 is based on human IgG2 and contains the S364K / E357Q:L368D / K370S tilted variant, the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side. Main chain 10 is based on human IgG2 and contains the S364K / E357Q:L368D / K370S tilted variant, the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side, and the S267K variant on both chains.

[0050] As those skilled in the art will understand and as outlined below, these sequences can be used with any vh and vh pairs outlined herein, one monomer containing scFv (optionally containing charged scFv linkers) and the other monomer containing a Fab sequence (e.g., vh linked to a “Fab-heavy chain” and vl linked to a “constant light chain”). That is, any Fv sequence outlined herein for anti-CTLA-4, anti-PD-1, anti-LAG-3, anti-TIM-3, anti-TIGIT, and anti-BTLA, whether as scFv (again, optionally having charged scFv linkers) or as Fab, can be incorporated into these main chains of Figure 37 in any combination. Figure 37A The depicted constant light chain can be used for all constructs in this diagram, but the κ constant light chain can also be replaced.

[0051] It should be noted that these bottle opener main chains are used for Figure 1F The central scFv format, wherein a second Fab (vh-CH1 and vl-constant light) with the same antigen binding as the first Fab is added to the N-terminus of the scFv on the "opener side".

[0052] Each of these backbones contains (as defined herein) 90%, 95%, 98%, and 99% identical sequences to the listed sequences, and / or contains 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 additional amino acid substitutions (as will be understood by those skilled in the art compared to the “parent” in the figures, such parent as parent IgG1 (or IgG2 or IgG4, depending on the backbone) already contains multiple amino acid modifications). That is, in addition to the tilted variants, pI variants, and ablation variants contained in the backbones of this figure, the listed backbones may contain additional amino acid modifications (typically amino acid substitutions).

[0053] Figures 38A to 38DThe sequence of the mAb-scFv backbone used in this invention is shown, to which the Fv sequence of this invention is incorporated. The mAb-scFv backbone 1 is based on human IgG1 (356E / 358M allotype) and includes the S364K / E357Q:L368D / K370S tilted variant, the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side, and the E233P / L234V / L235A / G236del / S267K ablation variant on both strands. Main chain 2 is based on human IgG1 (356D / 358L allotypes) and includes the S364K / E357Q:L368D / K370S tilted variant, the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains. Main chain 3 is based on human IgG1 (356E / 358M allotype) and includes the S364K / E357Q:L368D / K370S tilted variant, the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains, as well as the N297A variant on both chains. Main chain 4 is identical to main chain 3 except that the mutation is N297S. Alternative formats for mAb-scFv main chain 3 and mAb-scFv main chain 4 may exclude the ablation variant E233P / L234V / L235A / G236del / S267K from both chains. Main chain 5 is based on human IgG4 and contains the S364K / E357Q:L368D / K370S tilted variant, the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side, and the E233P / L234V / L235A / G236del / S267K ablation variant on both chains, as well as the S228P (EU designation, which is S241P in Kabat) variant ablated at the Fab arm exchange on both chains, as is known in the art. Main chain 6 is based on human IgG2 and contains the S364K / E357Q:L368D / K370S tilted variant and the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side. Main chain 7 is based on human IgG2 and contains the S364K / E357Q:L368D / K370S tilted variant, the N208D / Q295E / N384D / Q418E / N421D pI variant on the Fab side, and the S267K variant on both chains.

[0054] As those skilled in the art will understand and as outlined below, these sequences can be used with any vh and vh pairs outlined herein, one monomer comprising Fab and scFv (optionally comprising charged scFv linkers) and another monomer comprising a Fab sequence (e.g., vh linked to a “Fab-heavy chain” and vl linked to a “constant light chain”). That is, any Fv sequence outlined herein for anti-CTLA-4, anti-PD-1, anti-LAG-3, anti-TIM-3, anti-TIGIT, and anti-BTLA, whether as scFv (again, optionally having charged scFv linkers) or as Fab, can be incorporated into this main chain of Figure 38 in any combination. Monomer 1 is the negative side of the Fab-scFv pI and contains heterodimerized variants L368D / K370S, isotropic pI variants N208D / Q295E / N384D / Q418E / N421D, and ablation variants E233P / L234V / L235A / G236del / S267K (all relative to IgG1). Monomer 2 is the positive side of the scFv pI and contains heterodimerized variants 364K / E357Q. However, other tilted variant pairs can be substituted, especially [S364K / E357Q:L368D / K370S], [L368D / K370S:S364K], [L368E / K370S:S364K], [T411T / E360E / Q362E:D401K], [L368D / K370S:S364K / E357L], [K370S:S364K / E357Q], [T366S / L368A / Y407V:T366W] and [T366S / L368A / Y407V / Y394C:T366W / S354C].

[0055] Figure 38A The depicted constant light chain can be used for all constructs in this diagram, but the κ constant light chain can also be replaced.

[0056] It should be noted that these mAb-scFv mainchains are used for Figure 1H The mAb-Fv format (where one monomer includes vl at the C-terminus and the other includes vh at the C-terminus) and Figure 1E The scFv-mAb format (scFv domains are added to the C-terminus of one of the monomers).

[0057] Each of these backbones contains (as defined herein) 90%, 95%, 98%, and 99% of the sequence identical to the listed sequences, and / or contains 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 additional amino acid substitutions (as will be understood by those skilled in the art compared to the “parent” in the figures, such parent as parent IgG1 (or IgG2 or IgG4, depending on the backbone) already contains multiple amino acid modifications). That is, in addition to the tilted variants, pI variants, and ablation variants contained in the backbones of this figure, the listed backbones may contain additional amino acid modifications (typically amino acid substitutions).

[0058] Figure 39A and Figure 39B A matrix depicting possible combinations of the bispecific checkpoint antibodies of the present invention is provided. Figure 39AIn this context, combinations are not restricted by format and can use any format shown in Figure 1. Box “A” indicates that a CDR from the first ABD (listed on the X-axis) can be combined with a CDR from the second ABD (listed on the Y-axis). Box “B” indicates that the vh and vl chains from the first ABD can be combined with the vh and vl chains from the second ABD. Box “C” indicates that a CDR from the first ABD can be combined with the vh and vl chains from the second ABD. Box “D” indicates that the vh and vl chains from the first ABD can be combined with a CDR from the second ABD. The “E” in the box indicates that PD-1 ABD is selected from the following groups: 1G6_H1.279_L1.194; 1G6_H1.280_L1.224; 1G6_L1.194_H1.279; 1G6_L1.210_H1.288; and 2E9 H1L1.The "F" in the box indicates that CTLA-4ABD is selected from the following groups: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H2_L0; [CTLA-4]_H3. 21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA-4]_H3.23_L0.129;[CTLA-4]_ H3.23_L0.132;[CTLA-4]_H3.25_L0.124;[CTLA-4]_H3.25_L0.129;[CTLA-4]_H3.25_L0.132;[CTLA-4]_H3.4_L0.118;[CTLA-4] _H3.4_L0.119;[CTLA-4]_H3.4_L0.12;[CTLA-4]_H3.4_L0.121;[CTLA-4]_H3.4_L0.122;[CTLA-4]_H3.4_L0.123;[CTLA-4]_H3 .4_L0.124;[CTLA-4]_H3.4_L0.125;[CTLA-4]_H3.4_L0.126;[CTLA-4]_H3.4_L0.127;[CTLA-4]_H3.4_L0.128;[CTLA-4]_H3.4_ L0.129; [CTLA-4]_H34_L0.130; [CTLA-4]_H3.4_L0.131; [CTLA-4]_H3.4_L0.132; [CTLA-4]_H3.5_L2.1; [CTLA-4]_H3.5_L2.2; [CTLA-4]_H3.5_L2.3; [CTLA-4]_H3_L0; [CTLA-4]_H3_L0.22; [CTLA-4]_H3_L0.44; [CTLA-4]_H3_L0.67; and [CTLA-4]_H3_L0.74. The “G” in the box indicates that TIM-3ABD is selected from the following groups: 1D10H0L0; 1D12H0L0; 3H3_H1_L2.1; 6C8_H0L0; 6D9_H0_1D12_L0; 7A9_H0L0; 7B11_H0L0; 7B11var_H0L0; and 7C2_H0L0.The "H" in the box indicates that LAG-3 ABD is selected from the group of the following identifiers: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.93; 2A11_H1_L2.97; 2A11_H1L 1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_L1.11;7G8_ H3.23_L1.11; 7G8_H3.28_L1; 7G8_H3.28_L1.11; 7G8_H3.28_L1.13; 7G8_H3.30_L1.34; 7G8_H3.30_L1.34; The "I" in the box refers to the "J" in the box, which refers to BTLA ABD selected from groups: 9C6_H0L0; 9C6_H1.1_L1; and 9C6_H1.11_L1. Except for... Figure 39B This is for formats other than bottle openers. Figure 39B and Figure 39A Same. Figure 39B In this process, when the first ABD binds to PD-1, the first ABD is an scFv monomer, and the other ABDs (CTLA-4, LAG-3, TIGIT, TIM-3, and BTLA) are located within the Fab monomer. Figure 39B In the process, when the first ABD is combined with CTLA-4, it is located in the scFv monomer (except when it is on the Fab side and combined with PD-1), while the other ABDs (CTLA-4, LAG-3, TIGIT, TIM-3 and BTLA) are located in the Fab monomer.

[0059] Figure 40A matrix depicting possible combinations of bottle opener formats is shown. The “Q” in the box indicates (again, listed on the X-axis) that the first ABD domain is scF and (again, listed on the Y-axis) the second ABD is Fab-side. The “R” in the box indicates that the first ABD is Fab-side and the second ABD is scFv. The “S” in the box indicates that the first ABD is PD-1 resistant and is scFv-side. The “T” in the box indicates that the first ABD is CTLA-4 resistant and is scFv-side. The “U” in the box indicates that the first ABD is TIM-3 resistant and is scFv-side. The “V” in the box indicates that the first ABD is LAG-3 resistant and is scFv-side. The “W” in the box indicates that the first ABD is TIGIT resistant and is scFv-side. The “X” in the box indicates that the first ABD is BTLA resistant and is scFv-side. Additionally, each combination in Figure 39 can use the CDR, scFv, and vh and vl combinations of Figure 38. Additionally, a specific embodiment of the bottle opener main chain in Figure 39 is a sequence of Figure 36.

[0060] Figure 41A and Figure 41B A schematic diagram illustrates the associated benefits that bispecific checkpoint antibodies can provide compared to combination therapies using two different antibodies or drugs.

[0061] Figure 42 A similar schematic diagram illustrates that because tumor TILs co-express multiple checkpoints, bivalent binding increases affinity, thereby enhancing antitumor activity and avoiding peripheral toxicity.

[0062] Figure 43 This demonstrates that the bispecific checkpoint antibody of the present invention (e.g., anti-LAG-3 x anti-CTLA-4) can be combined with other monospecific checkpoint antibodies (e.g., nivolumab, pembrolizumab).

[0063] Figure 44 The study showed that PD-1 and CTLA-4 are co-expressed in various tumor types, including bladder cancer, breast cancer, colorectal cancer, prostate cancer, lung cancer, melanoma, and ovarian cancer.

[0064] Figures 45A to 45C The comparisons of IL-2 enhancement were described: in SEB-stimulated PBMC assays, B) comparison of anti-PD-1 bivalent antibody with anti-CTLA-4 x anti-PD-1 and C) comparison of single-arm anti-PD-1 + single-arm anti-CTLA-4 with anti-CTLA-4 x anti-PD-1, and C) control experiment in the absence of SEB stimulation.

[0065] Figure 46A and Figure 46BThe blockade of PD-1 ligands PD-L1 and PD-L2 by an exemplary anti-CTLA-4 x anti-PD-1 bispecific antibody, compared to single-arm anti-PD-1 antibodies and single-arm anti-CTLA-4 antibodies, is described.

[0066] Figure 47 T cell binding via an exemplary anti-CTLA-4 x anti-PD-1 bispecific antibody was depicted in a SEB-stimulated PBMC assay.

[0067] Figure 48 The anti-CTLA-4 x anti-PD-1 bispecific antibody was shown to enhance transplantation in NSG mice with human PBMCs (as measured by human CD45 count). This enhancement was greater than that seen with nivolumab (XENP 16432) alone.

[0068] Figure 49 The study demonstrated that the anti-BTLA x anti-PD-1 bispecific candidate bound T cells more readily than the "single-arm" control in SEB-stimulated PBMC assays.

[0069] Figure 50A and Figure 50B This study demonstrates how an anti-BTLA x anti-PD-1 chimeric bispecific antibody promotes IL-2 secretion from SEB-stimulated PBMCs. PBMCs were stimulated with 10 ng / mL SEB for 3 days using the indicated test sample. Cell supernatants were collected, and the analytes shown in the supernatants were determined by MSD. A: 20 μg / mL test sample; B: 5 μg / mL test sample.

[0070] Figure 51A and Figure 51B This study demonstrates how an anti-BTLA x anti-PD-1 chimeric bispecific antibody promotes FNγ secretion from SEB-stimulated PBMCs. PBMCs were stimulated with 10 ng / mL SEB for 3 days using the indicated test sample. Cell supernatants were collected, and the analytes shown in the supernatants were determined by MSD. A: 20 μg / mL test sample; B: 5 μg / mL test sample.

[0071] Figure 52A and Figure 52B The anti-BTLA x anti-PD-1 bispecific antibody (which is chimeric and has a humanized / optimized anti-BTLA Fab arm) was shown to promote the secretion of IL-2 and IFN-γ from SEB-stimulated PBMCs. Both figures are from PBMCs stimulated for 3 days with 10 ng / mL SEB using the 20 μg / mL test sample shown. Cell supernatants were collected after 72 hours and the analytes shown in the cell supernatants were determined.

[0072] Figures 53A to 53FThe timeline of CD45 cell count enhancement and IFNγ secretion (days 10, 14, and 22) in a GVHD study using an exemplary anti-BTLA x anti-PD-1 bispecific antibody is shown.

[0073] Figure 54 Some engineered data for the 9C6 anti-BTLA antigen-binding domains are described. This describes the names of the XENP-encoded, vh, and vl engineered domains of the bivalent embodiment, and the KD binding constant of human BTLA as measured by octet.

[0074] Figures 55A to 55E Some engineering data for the 2A11 anti-LAG-3 antigen-binding domains are depicted. This depicts the XENP encoding of the Fab embodiment, the names of the vh and vl engineered domains, the KD binding constant of human LAG-3 as measured by octet, and the Tm of the Fab.

[0075] Figures 56A to 56K Some engineering data for the 7G8 anti-LAG-3 antigen-binding domains are described. This describes the XENP encoding of the Fab embodiment, the names of the vh and vl engineered domains, the KD binding constant of human LAG-3 as measured by octet, and the Tm of the Fab.

[0076] Figure 57A and Figure 57B Kd is depicted as an anti-LAG-3 X anti-CTLA-4 bispecific heterodimer opener format based on optimized 2A11 or 7G8 anti-LAG-3 Fab arms, as measured by octet.

[0077] Figure 58 The bispecific antibodies against LAG-3(7G8)x anti-CTLA-4 and anti-LAG-3(2A11)x anti-CTLA-4 showed greater affinity binding than the single-arm anti-LAG-3 control. PBMCs were stimulated with 100 ng / mL SEB for 3 days. Cells were then treated with the test samples shown at 4°C for 30 min and washed twice. Cells were then treated with anti-CD3-FITC and anti-human-Fc-APC antibodies. Cells were then washed twice and analyzed by flow cytometry.

[0078] Figure 59A and Figure 59BAs shown by IL-2 enhancement and IFNγ release, the 7G8-based anti-LAG-3 x anti-CTLA-4 bispecific antibody exhibited greater selectivity against PBMCs than the 2A11-based anti-LAG-3 x anti-CTLA-4 bispecific antibody. PBMCs were stimulated with 500 ng / mL SEB for 2 days. Cells were then washed twice in culture medium and stimulated with 500 ng / mL SEB in the indicated amounts of test sample. The analytes (IL-2 or IFN-γ) were measured in cells 24 hours post-treatment. Each spot represents a single donor tested in the technical singlet state.

[0079] Figure 60A and Figure 60B Mixed lymphocyte reactions (MLRs) with the anti-LAG-3 and anti-CTLA-4 bispecific antibodies were characterized. Forty unique MLR reactions were performed in the presence of 20 μg / mL of the assay sample. The A: IL-2 and B: IFNγ in the cell supernatant were then determined by MSD after 6 days of treatment.

[0080] Figure 61A and Figure 61B The enhancement of IL-2 and IFN-γ release by additional anti-LAG-3 X anti-CTLA-4 candidates is shown in the SEB assay. PBMCs were stimulated with 500 ng / mL SEB for 2 days. Cells were then washed twice in culture medium and stimulated with 500 ng / mL SEB in the indicated amounts of test sample. The analytes (IL-2 or IFN-γ) were measured in cells 24 hours after treatment. Each spot represents a single donor tested in the technical singlet state.

[0081] Figure 62A and Figure 62B Kd is depicted as an anti-LAG-3 X anti-PD-1 bispecific heterodimer opener format based on optimized 2A11 or 7G8 anti-LAG-3 Fab arms, as measured by octet.

[0082] Figure 63A and Figure 63B The ability of humanized / optimized 7G8 and 2A11 anti-LAG-3 clones to block the binding of LAG-3 to cells homologously expressing MHC-II was demonstrated.

[0083] Figure 64A and Figure 64BThe function of anti-LAG-3 x anti-PD-1 candidates in response to SEB stimulation of T cells was characterized. PBMCs were stimulated with 500 ng / mL SEB for 2 days. Cells were then washed twice in culture medium and stimulated with 500 ng / mL SEB in the indicated amounts of test sample. The analytes were measured in cells 24 hours post-treatment. Each spot represents a unique donor tested in the singlet state.

[0084] Figure 65 This diagram illustrates the co-expression of multiple checkpoint receptors in various tumors by tumor-infiltrating lymphocytes (TILs). Specifically, it shows the co-expression of PD-1 and CTLA-4, PD-1 and BTLA, PD-1 and LAG-3, and LAG-3 and CTLA-4 in various tumors. The results shown are based on data generated through the TCGA Research network.

[0085] http: / / cancergenome.nih.gov /

[0086] Figure 66 The subject-specific antibody presented herein selectively targets dual checkpoint positive T cells. As compared to a negative control, the bispecific PD-1 x LAG-3 antibody was used to demonstrate PD-1 and LAG-3 receptor occupancy in CD3+ T cells stimulated by staphylococcal enterotoxin B (SEB).

[0087] Figures 67A to 67F This is a diagram illustrating how the component antibody domains of the subject antibody presented in this paper can block checkpoint receptor / ligand interactions.

[0088] Specifically, the bispecific antibody including the 1G6 anti-PD-1 scFv arm can block PD-1 / PD-L1 interaction and PD-1 / PD-L2 interaction; the 7G8 anti-LAG-3 single arm can block LAG-3 / MHC II interaction; the bispecific antibody including the exemplary anti-PD-1 Fab arm can block CTLA-4 / CD80 interaction and CTLA-4 / CD86 interaction; and the bispecific antibody including the 9C6 anti-BTLA Fab arm can block BTLA / HVEM interaction.

[0089] Figure 68 Enhanced IL-2 release was compared using exemplary anti-CTLA-4 x anti-PD-1 bispecific antibodies and nivolumab.

[0090] Figure 69 Enhanced IL-2 release was compared using exemplary anti-LAG-3 x anti-CTLA-4 bispecific antibodies, the same bispecific antibodies combined with nivolumab, and nivolumab alone.

[0091] Figure 70 Enhanced IL-2 release was compared using exemplary anti-LAG-3 x anti-PD-1 bispecific antibodies and nivolumab.

[0092] Figure 71 Enhanced IL-2 release was compared using exemplary anti-BTLA x anti-PD-1 bispecific antibodies and nivolumab.

[0093] Figure 72 GVHD enhancement (as shown by CD45 cell counts) was compared using exemplary anti-PD-1 x anti-CTLA-4 bispecific antibodies, nivolumab alone, and nivolumab combined with ipilimumab.

[0094] Figure 73 GVHD enhancement (as shown by CD45 cell counts) was compared using exemplary anti-BTLA x anti-PD-1 bispecific antibodies and nivolumab.

[0095] Figure 74 GVHD enhancement (as shown by CD45 cell counts) was compared using exemplary anti-LAG-3 x anti-CTLA-4 bispecific antibodies, the same bispecific antibodies combined with nivolumab, and nivolumab alone.

[0096] Figure 75 GVHD enhancement (as shown by CD45 cell counts) was compared using exemplary anti-LAG-3 x anti-PD-1 bispecific antibodies and nivolumab.

[0097] Figures 76A to 76B Two studies were described, demonstrating that anti-CTLA-4 x anti-PD-1 bispecific antibodies can promote T-cell-mediated antitumor efficacy in vivo. KGla-luc cancer cells were transplanted into mice. Twenty-one days later, huPMCs were transplanted into the same mice and treated weekly with antibody therapy (anti-CTLA-4 x anti-PD-1 bispecific antibody; anti-PD-1 bivalent antibody; or anti-PD-1 bivalent antibody + anti-CTLA-4 bivalent antibody). In Vivisometry (IVIS) imaging of cancer cells was performed in the mice to estimate cancer cell size, as determined by changes in tumor flux. Detailed Implementation

[0098] A. Material Consolidation

[0099] 1. Attached Figures and Legends

[0100] All figures and illustrations in USSN 62,350,145, 62 / 353,511 and 62 / 420,500 are expressly and independently incorporated herein by reference in their entirety, particularly for the amino acid sequences depicted therein.

[0101] 2. Sequence

[0102] Referring to the attached sequence listing, the anti-PD-1 sequences suitable for use as ABD include SEQ ID NO: 6209 to 11464 (PD-1 scFv sequences, but the Fv sequences therein can be formatted as Fab), SEQ ID NO: 11465 to 17134 (PD-1 Fab sequences, but the Fv sequences therein can be formatted as scFv), SEQ ID NO: 33003 to 33072 (additional PD-1 Fab sequences, but the Fv sequences therein can be formatted as scFv), SEQ ID NO: 33073 to 35394 (additional PD-1 scFv sequences, but the Fv sequences therein can be formatted as Fab) and SEQ ID NO: 36127 to 36146 (PD-1 bivalent constructs, which can be formatted as scFv or Fab). Suitable anti-CTLA-4 sequences for use as ABD include SEQ ID NO: 21 to 2918 (CTLA-4 scFv sequences, but the Fv sequences therein can be formatted as Fab), SEQ ID NO: 2919 to 6208 (CTLA-4 Fab sequences, but the Fv sequences therein can be formatted as scFv), SEQ ID NO: 36739 to 36818 (additional CTLA-4 Fab sequences, but the Fv sequences therein can be formatted as scFv) and SEQ ID NO: 35395 to 35416 (CTLA-4 single-arm constructs, which can be formatted as scFv or Fab). Suitable anti-LAG-3 sequences for use as ABD include SEQ ID NO: 17135 to 20764 (LAG-3 Fab, but the Fv sequence therein can be formatted as scFv), SEQ ID NO: 36819 to 36962 (additional LAG-3 Fab, but the Fv sequence therein can be formatted as scFv), SEQ ID NO: 35417 to 35606 (additional LAG-3 Fab, but the Fv sequence therein can be formatted as scFv), SEQ ID NO: 25194 to 32793 (additional LAG-3 Fab, but the Fv sequence therein can be formatted as scFv) and SEQ ID NO: 32794 to 33002 (single-arm LAG-3 constructs, which can be formatted as Fab or scFv).Suitable anti-TIM-3 sequences for use as ABD include SEQ ID NO: 20765 to 20884 (TIM-3 Fab, but the Fv sequence therein can be formatted as scFv), SEQ ID NO: 37587 to 37698 (additional TIM-3 Fab, but the Fv sequence therein can be formatted as scFv), and SEQ ID NO: 36347 to 36706 (bivalent TIM-3 constructs, which can be formatted as Fab or scFv). Suitable anti-BTLA sequences for use as ABD include SEQ ID NO: 20885 to 21503 (BTLA Fab, but the Fv sequence therein can be formatted as scFv) and SEQ ID NO: 36707 to 36738 (additional BTLA Fab, but the Fv sequence therein can be formatted as scFv). Suitable anti-TIGIT sequences for use as ABD include SEQ ID NO: 21504 to 21523 (TIGIT Fab, but the Fv sequence therein can be formatted as scFv) and SEQ ID NO: 37435 to 37586 (additional TIGIT Fab, but the Fv sequence therein can be formatted as scFv).

[0103] The bispecific antibodies of the present invention comprise the LAG3 x CTLA4 constructs of SEQ ID NO: 35607 to 35866 and SEQ ID NO: 21524 to 22620. The PD-1 x CTLA4 construct comprises those listed as SEQ ID NO: 36167 to 36346 and SEQ ID NO: 23316 to 23735. The PD-1 x TIM3 construct comprises those listed as SEQ ID NO: 25174 to 25193. The PD-1 x LAG3 construct comprises those listed as SEQ ID NO: 35867 to 36126 and SEQ ID NO: 23736 to 25133. The PD-1 x TIGIT construct comprises those listed as SEQ ID NO: 25134 to 25173. The PD-1 XBTLA construct includes those listed as SEQ ID NO: 22724 to 23315 and SEQ ID NO: 36147 to 36166. The CTLA4X BTLA construct includes those listed as SEQ ID NO: 22624 to 22723. Ultimately, the following names, which were unintentionally omitted, should have included the symbol "M428L / N434S" in the title: XENP23552, XENP22841, XENP22842, XENP22843, XENP22844, XENP22845, XENP22846, XENP22847, XENP22848, XENP22849, XENP22850, XENP22851, XENP22852, XENP22858, XENP22854, XENP22855.

[0104] B. Overview

[0105] Therapeutic antibodies targeting immune checkpoint inhibitors such as PD-1 have shown great promise in clinical practice, in limited cases, and in the treatment of cancer. Cancer can be viewed as a patient's inability to recognize and eliminate cancer cells. In many cases, these transformed (e.g., cancerous) cells counteract immune surveillance. There are natural control mechanisms that limit T cell activation in the body to prevent unrestrained T cell activity, which can be used by cancer cells to evade or suppress the immune response. Restoring the ability of immune effector cells—especially T cells—to recognize and eliminate cancer is the goal of immunotherapy. The immuno-oncology domain, sometimes referred to as "immunotherapy," is rapidly evolving, and several T cell checkpoint inhibitory antibodies, such as Yervoy, Keytruda, and Opdivo, have recently been approved. These antibodies are often called "checkpoint inhibitors" because they block the normal negative regulators of T cell immunity. It should generally be understood that various co-stimulatory and co-inhibitory immune regulatory signals can be used to coordinate optimal antigen-specific immune responses.

[0106] Typically, these monoclonal antibodies bind to checkpoint inhibitory proteins such as CTLA-4 and PD-1, which normally prevent or suppress cytotoxic T cell (CTL) activation. By inhibiting checkpoint proteins, for example by using antibodies that bind to those proteins, an increased T-cell response against tumors can be achieved. That is, these cancer checkpoint proteins suppress the immune response; when the proteins are blocked, for example by using antibodies targeting checkpoint proteins, the immune system is activated, generating immune stimulation that leads to treatment for conditions such as cancer and infectious diseases.

[0107] However, as discussed above, studies have shown that TILs co-express multiple checkpoint receptors; this suggests that blockade of a single checkpoint may not be sufficient to promote a full T-cell response. Furthermore, TILs expressing multiple checkpoints may actually be the most tumor-responsive, suggesting that therapies involving more than one checkpoint antigen may be very useful.

[0108] Therefore, the present invention provides a method for binding bispecific checkpoint antibodies to cells expressing two antigens and for activating T cells and / or NK cells to treat diseases such as cancer and infectious diseases, as well as other conditions in which the increased immune activity causes treatment.

[0109] Therefore, in some cases, the present invention addresses the toxicity and cost problems associated with administering multiple antibodies by providing bispecific antibodies that bind to two different checkpoint inhibitor molecules on a single cell, and advantageously requiring the administration of only one therapeutic substance.

[0110] Bispecific antibodies that can simultaneously bind to two different targets offer the potential to improve selectivity for targeting TILs and peripheral T cells while also reducing therapy costs. The bivalent interaction between an antibody and two targets on the cell surface should—in some cases—produce a higher binding affinity than a single monovalent interaction with a single target. Because of this, normal bivalent antibodies tend to have high affinity for their targets on the cell surface. With bispecific antibodies, it is possible to create higher selectivity for cells simultaneously expressing two different targets, thereby utilizing the higher affinity provided by binding to both targets simultaneously.

[0111] Therefore, this invention relates to novel constructs for providing heterodimeric antibodies that allow binding to more than one checkpoint antigen or ligand, such as allowing bispecific binding. Thus, for example, an anti-PD1 x anti-CTLA4 (PD1 x CTLA4) bispecific antibody shows promise for greater selectivity against PD1+CTLA4+ double-positive TILs compared to single-positive PD1-only or CTLA4-only T cells. The selective blocking of double-positive TILs compared to single-positive T cells thus promises to improve the therapeutic index of combined checkpoint blockade. Other possible combinations as outlined herein are similar. Therefore, suitable bispecific antibodies of the present invention bind to PD-1 and CTLA-4, PD-1 and TIM-3, PD-1 and LAG-3, PD-1 and TIGIT, PD-1 and BTLA, CTLA-4 and TIM-3, CTLA-4 and LAG-3, CTLA-4 and TIGIT, CTLA-4 and BTLA, TIM-3 and LAG-3, TIM-3 and TIGIT, TIM-3 and BTLA, LAG-3 and TIGIT, LAG-3 and BTLA, and TIGIT and BTLA. It should be noted that for each pair, these bispecific antibodies are generally named “anti-PD-1 x anti-CTLA-4” or are generally simplified or, for convenience (and therefore interchangeably) named “PD-1 x CTLA-4”, etc.

[0112] The heterodimeric bispecific checkpoint antibody of the present invention can be used to treat various types of cancer. As those skilled in the art will recognize, compared to conventional monoclonal antibodies that bind to tumor antigens or newer classes of bispecific antibodies that bind to, for example, CD3 and tumor antigens (as described, for example, USSN 15 / 141,350), checkpoint antibodies are used to enhance immune responses but generally do not possess tumor specificity in their action. That is, the bispecific checkpoint antibody of the present invention inhibits the suppression of the immune system, typically generating T cell activation, which in turn produces a greater immune response against cancer cells and thus a therapeutic effect. Such antibodies may therefore hold promise for treating various types of tumors. For example, the FDA recently approved... Anti-PD-1 monospecific antibodies based on genetic characteristics rather than tumor type.

[0113] As discussed below, there are various ways to measure T cell activation. The functional effects of bispecific checkpoint antibodies on NK and T cells can be estimated in vitro (and in some cases in vivo, as described more fully below) by measuring changes in the following parameters: proliferation, cytokine release, and cell surface markers. For NK cells, increased cell proliferation, cytotoxicity (the ability to kill target cells, as measured by increased expression of CD107a, granzyme, and perforin, or by directly measuring target cell killing), cytokine production (e.g., IFN-γ and TNF), and cell surface receptor expression (e.g., CD25) indicate immune regulation, such as enhanced cancer cell killing. For T cells, increased proliferation, increased expression of activated cell surface markers (e.g., CD25, CD69, CD137, and PD1), cytotoxicity (the ability to kill target cells), and cytokine production (e.g., IL-2, IL-4, IL-6, IFN-γ, TNF-α, IL-10, IL-17A) indicate immune regulation, such as enhanced cancer cell killing. Therefore, treatment can be evaluated using one or more of the following assays: (i) increased immune response; (ii) increased activation of αβ and / or γδ T cells; (iii) increased cytotoxic T cell activity; (iv) increased NK and / or NKT cell activity; (v) remission of αβ and / or γδ T cell suppression; (vi) increased secretion of pro-inflammatory cytokines; (vii) increased IL-2 secretion; (viii) increased interferon-γ production; (ix) increased Th1 response; (x) decreased Th2 response; (xi) decreased number and / or activity of at least one of regulatory T cells and other cells; (xii) increased tumor immune infiltration.

[0114] Therefore, in some embodiments, the present invention provides the use of bispecific checkpoint antibodies to perform one or more of the following on a subject of need: (a) upregulating pro-inflammatory cytokines; (b) increasing T cell proliferation, expansion, or tumor infiltration; (c) increasing the production of interferon-γ, TNF-α, and other cytokines via T cells; (d) increasing IL-2 secretion; (e) stimulating antibody responses; (f) inhibiting cancer cell growth; (g) promoting antigen-specific T cell immunity; (h) promoting CD4+ and / or CD8+ T cell activation; (i) alleviating T cell suppression; (j) promoting NK cell activity; (k) promoting apoptosis or lysis of cancer cells; and / or (l) cytotoxic effects or cell growth inhibition effects on cancer cells.

[0115] Therefore, this invention provides bispecific heterodimeric checkpoint antibodies. Heterodimeric antibody constructs are based on the self-assembly of two Fc domains of the antibody's heavy chain, for example, two "monomers" assembling into a "dimer". Heterodimeric antibodies are formed by altering the amino acid sequence of each monomer, as discussed more fully below. Thus, this invention generally relates to the creation of heterodimeric antibodies that can co-bind checkpoint antigens in various ways, depending on differences on each chain to promote heterodimer formation and / or allow for easier purification of amino acid variants in the constant region of the heterodimer compared to homodimers.

[0116] Therefore, this invention provides bispecific checkpoint antibodies. A persistent problem in antibody technology is the desire for “bispecific” antibodies that bind simultaneously to two (or more) different antigens, typically allowing for the accessibility of different antigens and the generation of novel functions and therapies. Typically, these antibodies are formed by including the genes for each heavy and light chain in the host cell (in this invention, this generally refers to the genes for the two heavy chain monomers and the light chain as outlined herein). This typically results in the formation of the desired heterodimer (AB) and two homodimers (AA and BB). However, a major obstacle to the formation of bispecific antibodies is the difficulty in purifying heterodimer antibodies away from homodimer antibodies and / or biasing the formation of heterodimers compared to homodimer formation.

[0117] To address this issue, various mechanisms exist for generating the heterodimers of this invention. Furthermore, as those skilled in the art will appreciate, these mechanisms can be combined to ensure high heterodimerization. Therefore, amino acid variants leading to the production of heterodimeric antibodies are referred to as "heterodimerized variants." As discussed below, heterodimerized variants may contain...

[0118] Spatial variants (e.g., the “knobs and holes” or “tilted” variants and the “charge pairs” variants described below) and the “pI variants” allow for the purification of homodimers away from heterodimers.

[0119] Alternatively, a mechanism may be used that creates spatial and / or electrostatic effects that favor heterodimer formation and discourage homodimer formation, commonly referred to in the art as “bulge-hole” (“KIH”) or sometimes referred to herein as the “tilt” variant, as described below: Ridgway et al., *Protein Engineering* 9(7): 617(1996); Atwell et al., *Journal of Molecular Biology* 1997 270: 26; U.S. Patent No. 8,216,805; US 2012 / 0149876, all of which are incorporated herein by reference in their entirety. The figures identify multiple pairs of “monomer A-monomer B” containing “bulge-hole” amino acid substitutions. Furthermore, as Merchant et al. described in *Nature Biotech.* 16:677 (1998), these “bulge and hole” mutations can combine with disulfide bonds to tilt the formation toward heterodimerization. The present invention utilizes T366S / L368A / Y407V paired with T366W, this variant paired with a bridging disulfide, and T366S / L368A / Y407V / Y349C paired with T366W / S354C, particularly in combination with other heterodimerization variants containing pI variants as outlined below.

[0120] Additional mechanisms used to generate heterodimeric antibodies are sometimes referred to as “electrostatic steering” or “charge pairs,” as described by Gunasekaran et al., Journal of Biochemistry, 285(25): 19637(2010), cited in its entirety hereincorporated. This is sometimes referred to herein as a “charge pair.” In this embodiment, electrostatics are used to tilt the formation toward heterodimerization. As those skilled in the art will appreciate, these charge pairs can also act on pIs and thus on purification, and therefore in some cases can also be considered pI variants. However, since these charge pairs are generated to force heterodimerization and are not used as purification tools, they are classified as “spatial variants.” These charge pairs include, but are not limited to, D221E / P228E / L368E paired with D221R / P228R / K409R (e.g., these are “monomer counterparts”) and C220E / P228E / 368E paired with C220R / E224R / P228R / K409R, as well as other charge pairs shown in the figures.

[0121] In some embodiments of the present invention, pI variants are used to change the pI of one or both of the monomers and thus allow isoelectric separation of AA, AB, and BB dimer proteins.

[0122] In this invention, several basic mechanisms exist that lead to the easy purification of heterodimerized proteins; one mechanism relies on the use of pI variants, such that each monomer has a different pI, thereby allowing isoelectric purification of AA, AB, and BB dimer proteins. Alternatively, some scaffold formats, such as the “triple F” format, also allow for size-based separation. As further outlined below, it is also possible to “tilt” heterodimers compared to homodimers. Therefore, combinations of spatial heterodimerization variants with pI or charge-pair variants are particularly useful in this invention. Additionally, as more fully outlined below, scaffolds utilizing one or more scFvs, such as the triple F format, can include charged scFv linkers (positively or negatively charged) that provide an additional pI boost for purification purposes. As those skilled in the art will appreciate, some triple F formats can be used with only charged scFv linkers without additional pI adjustments, but this invention also provides for the use of tilt variants with charged scFv linkers (and combinations of the Fc, FcRn, and KO variants discussed herein).

[0123] In this invention, which uses pI as a separation mechanism to allow the purification of heterodimeric proteins, amino acid variants can be introduced into one or both of the monomeric polypeptides; that is, the pI of one of the monomers (referred to herein as "monomer A" for simplicity) can be engineered away from monomer B, or both monomer A and monomer B can be altered, wherein the pI of monomer A increases and the pI of monomer B decreases. As more fully outlined below, changes in the pI of any one or both monomers can be accomplished by: removing or adding charged residues (e.g., replacing a neutral amino acid with a positively or negatively charged amino acid residue, such as glycine to glutamic acid), changing charged residues from positively or negatively charged to the opposite charge (e.g., aspartic acid to lysine), or changing charged residues to neutral residues (e.g., charge loss; lysine to serine). Several of these variants are shown in the accompanying drawings. Additionally, suitable pI variants used in this paper to create heterodimeric antibodies are those belonging to the same IgG isotype, such as pIs introduced from different IgG isotypes, thereby altering the pI without introducing significant immunogenicity; see US Publication No. 20140288275, which is incorporated herein by reference in its entirety. Figure 29 .

[0124] Therefore, this embodiment of the invention provides for a sufficient pI variation in at least one of the monomers, thereby allowing the heterodimer to be separated from the homodimer. As those skilled in the art will understand and further discussed below, this can be accomplished by using a constant region of the "wild-type" heavy chain and a variant region engineered to increase or decrease its pI (wt A-+B or wt A--B) or increasing one region and decreasing the other (A+-B-0rA-B+).

[0125] Therefore, typically, components of some embodiments of the present invention are amino acid variants in the constant region of an antibody, said amino acid variant involving altering the isoelectric point (pI) of at least one (if not both) of the monomers by incorporating amino acid substitutions (“pI variants” or “pI substitutions”) into one or both of the monomers to form “pI heterodimers” (when the protein is an antibody, these pI heterodimers are referred to as “pI antibodies”). As shown herein, the separation of the heterodimer from the two homodimers can be accomplished when the pIs of the two monomers differ by as little as 0.1 pH units, with 0.2, 0.3, 0.4, and 0.5 or more pH units used in the present invention.

[0126] As those skilled in the art will appreciate, the number of pI variants to be included on each or both monomers for achieving good separation will depend in part on the starting pI of the scFv and Fab of interest. That is, to determine which monomer to engineer or in which “direction” (e.g., more positive or more negative), the Fv sequences of the two target antigens are calculated and a decision is made accordingly. As is known in the art, different Fvs will have different starting pIs, which are utilized in this invention. Typically, as outlined herein, pIs are engineered so that the total pI difference between the individual monomers is at least about 0.1 logS, preferably 0.2 logS to 0.5 logS as outlined herein.

[0127] Furthermore, as those skilled in the art will understand and as outlined herein, in some cases (depending on the format), heterodimers can be separated from homodimers based on size (e.g., molecular weight). For example, as shown in some embodiments of Figure 1, some formats produce homodimers and heterodimers of different sizes (e.g., for a bottle opener, one homodimer is a "double scFv" format, one homodimer is a standard antibody, and the heterodimer has one Fab and one scFv).

[0128] Furthermore, as depicted in Figure 1, it will be recognized that some antigens may bind bivalently (e.g., two antigen-binding sites of a single antigen). As will be understood, any combination of Fab and scFv can be utilized to achieve the desired results and combinations.

[0129] In cases where pI variants are used to achieve heterodimers optimized relative to homodimers, a more modular approach to designing and purifying multispecific proteins (including antibodies) is provided by using one or more constant regions of one or more heavy chains. Therefore, in some embodiments, heterodimerized variants (including tilted and purified heterodimerized variants) are not included in the variable region, making each individual antibody engineered. Additionally, in some embodiments, by introducing pI variants from different IgG isotypes, the likelihood of immunogenicity caused by the pI variant is significantly reduced, allowing pI modification without introducing significant immunogenicity. Therefore, another problem to be addressed is elucidating low pI constant domains with high human sequence content, e.g., minimizing or avoiding non-human residues at any particular position.

[0130] A potential negative effect accompanying this pI engineering is also the prolongation of serum half-life and increased FcRn binding. That is, as described in USSN 13 / 194,904 (incorporated in its entirety by reference), pIs that reduce the antibody constant domains (those found in antibody-Fc fusions) may result in longer serum retention in vivo. These pI variants with increased serum half-life also facilitate pI changes for purification.

[0131] Additionally, it should be noted that pI variants of heterodimerized proteins offer extra benefits to the analysis and quality control processes of bispecific antibodies, as their ability to eliminate, minimize, and distinguish homodimers in their presence is significantly enhanced. Similarly, the ability to reliably test the reproducibility of heterodimer protein production is important.

[0132] As those skilled in the art will appreciate and will be discussed more fully below, the heterodimeric fusion protein of the present invention can take various configurations, as generally depicted in Figure 1. Some figures depict a “single-end” configuration, wherein one “arm” of the molecule has one type of specificity and the other “arm” has a different type of specificity. Other figures depict a “double-end” configuration, wherein the “top” of the molecule has at least one type of specificity and the “bottom” of the molecule has one or more different types of specificity. Therefore, the present invention relates to novel immunoglobulin compositions that co-conjugate a first antigen and a second antigen. The first antigen and the second antigen of the present invention are referred to herein as antigen 1 and antigen 2 (or “checkpoint 1” and “checkpoint 2”), respectively.

[0133] A heterodimer scaffold specifically used in this invention is as follows: Figure 1AThe depicted “triple F” or “bottle opener” scaffold format. In this embodiment, one heavy chain of the antibody contains a single-chain Fv (“scFv”, as defined below) and the other heavy chain is a “conventional” FAb format comprising a variable heavy chain and a light chain. This structure is sometimes referred to herein as a “triple F” format (scFv-FAb-Fc) or a “bottle opener” format due to its general visual similarity to a bottle opener (see [link to documentation]). Figure 1A The two chains are brought together by using amino acid variants in the constant region (e.g., the Fc domain and / or hinge region) that promote the formation of heterodimer antibodies, as described more fully below.

[0134] The "triple F" format of this invention offers several distinct advantages. As is known in the art, antibody analogs relying on two scFv constructs often suffer from stability and polymerization problems, which can be mitigated in this invention by adding a "conventional" heavy chain and light chain pairing. Furthermore, unlike formats relying on two heavy chains and two light chains, there is no problem of incorrect heavy chain and light chain pairing (e.g., heavy chain 1 paired with light chain 2, etc.).

[0135] Furthermore, as outlined herein, additional amino acid variants can be introduced into the bispecific antibodies of the present invention to add additional functionality. For example, amino acid changes within the Fc region (to one or both monomers) can be added to promote increased ADCC or CDC (e.g., binding with altered Fcγ receptors) and increased binding to FcRn and / or increased serum half-life of the resulting molecule. As further described herein and as those skilled in the art will appreciate, any and all variants outlined herein can be optionally and independently combined with other variants.

[0136] Similarly, another category of functional variants is the “Fcγ ablation variant” or “Fc knockout (FcKO or KO) variant.” In these embodiments, for some therapeutic applications, it is desirable to reduce or eliminate the normal binding of the Fc domain to one or all of the Fcγ receptors (e.g., FcγR1, FcγRIIa, FcγRIIb, FcγRIIIa, etc.) to avoid additional action mechanisms. That is, for example, it is generally desirable to ablate FcγRIIIa binding to eliminate or significantly reduce ADCC activity. Suitable ablation variants are shown in Figure 5 middle.

[0137] C. Nomenclature

[0138] The bispecific antibodies of this invention are listed in several different formats. Each polypeptide is given a unique “XENP” number, but as will be understood in the art, longer sequences may contain shorter XENP numbers. For example, the heavy chain of the scFv-side monomer in the opener format of a given sequence will have a first XENP number, while the scFv domain will have a different XENP number. Some molecules have three polypeptides, and therefore, in the case of components, the XENP number is used as the name. Thus, the molecule XENP20717 in the opener format comprises three sequences, commonly referred to as “XENP20717HC-Fab,” “XENP20717HC-scFv,” and “XENP20717LC” or equivalents, but those skilled in the art can likely identify these readily by sequence alignment. These XENP numbers are in the sequence listing and identifiers, and are used in the figures. Additionally, a molecule comprising three components generates multiple sequence identifiers. For example, the list of Fab monomers has a full-length sequence, a variable heavy sequence, and three CDRs for the variable heavy sequence; the light chain has a full-length sequence, a variable light sequence, and three CDRs for the variable light sequence; and the scFv-Fc domain has a full-length sequence, an scFv sequence, a variable light sequence, three light CDRs, an scFv structure, a variable heavy sequence, and three heavy CDRs. It should be noted that all molecules with scFv domains in this paper use a single charged scFv linker (+H), but others may also be used. Furthermore, the nomenclature for specific variable domains uses the format "Hx.xx_Ly.yy", where the number is a unique identifier for the specific variable chain sequence. Therefore, the variable domain on the Fab side of XENP22841 is "7G8_H3.30_L1.34", indicating a combination of the variable heavy domain H3.30 and the light domain L1.34. In the case of these sequences used as scFv, the name "7G8_H3.30_L1.34" indicates that the variable heavy domain H3.30 is combined with the light domain L1.34 and is oriented from the N-terminus to the C-terminus vh-linker -vl. A molecule with the same sequence as the heavy and light variable domains but in reverse order can be called "7G8_L1.34_H3.30". Similarly, different constructs can be "mixed and matched" with the heavy and light chains, as will be apparent from the sequence listing and figures.

[0139] D. Definition

[0140] To provide a more thorough understanding of this application, several definitions are set forth below. These definitions are intended to cover syntactic equivalents.

[0141] In this article, "ablation" means reducing or removing activity. Therefore, for example, "FcγR ablation binding" means that the Fc region amino acid variant has less than 50% of the initial binding compared to the Fc region without the specific variant, preferably more than 70% to 80% to 90% to 95% to 98% of the activity loss, and typically, the activity is below the level detectable binding in Biacore, SPR, or BLI assays. Specifically used for ablation of FcyR binding is... Figure 5 The variants shown are typically added to both monomers.

[0142] As used herein, “ADCC” or “antibody-dependent cell-mediated cytotoxicity” refers to a cell-mediated reaction in which nonspecific cytotoxic cells expressing FcyR recognize binding antibodies on target cells and subsequently cause target cell lysis. ADCC is associated with binding to FcyRIIIa; increased binding to FcγRIIIa leads to increased ADCC activity.

[0143] As used in this article, “ADCP” or antibody-dependent cell-mediated phagocytosis refers to a cell-mediated response in which nonspecific phagocytes expressing FcγR recognize binding antibodies on target cells and subsequently induce phagocytosis of the target cells.

[0144] In this document, "antigen-binding domain" or "ABD" refers to a set of six complementarity-determining regions (CDRs) that, when present as part of a polypeptide sequence, specifically bind to a target antigen as discussed herein. Thus, "checkpoint antigen-binding domain" refers to the binding of a target checkpoint antigen as outlined herein. As is known in the art, these CDRs typically exist as a first set of variable heavy CDRs (vhCDRs or VHCDRs) and a second set of variable light CDRs (vlCDRs or VLCDRs), each comprising three CDRs: vhCDR1, vhCDR2, and vhCDR3 of the heavy chain and vlCDR1, vlCDR2, and vlCDR3 of the light chain. The CDRs are located in the variable heavy and variable light domains, respectively, and together form the Fv region. (See Table 1 and the related discussion above for the CDR numbering scheme). Therefore, in some cases, the six CDRs of the antigen-binding domain originate from the variable heavy and variable light domains. In the “Fab” format, a set of six CDRs originates from two distinct polypeptide sequences, a variable heavy domain (vh or VH; containing vhCDR1, vhCDR2, and vhCDR3) and a variable light domain (vl or VL; containing vlCDR1, vlCDR2, and vlCDR3). The C-terminus of the vh domain is linked to the N-terminus of the CH1 domain of the heavy chain, and the C-terminus of the vl domain is linked to the N-terminus of the constant light domain (thus forming the light chain). In the scFv format, the vh and vl domains are typically bivalently linked to a single polypeptide sequence using a linker (“scFv linker”) as outlined herein. This can be (starting from the N-terminus) vh-linker-vl or vl-linker-vh, with the former generally preferred (containing optional domain linkers on each side, depending on the format used (e.g., according to Figure 1)). Typically, the C-terminus of the scFv domain is linked to the N-terminus of the hinge in the second monomer.

[0145] In this article, "modification" refers to amino acid substitutions, insertions, and / or deletions in a polypeptide sequence, or alterations to the portion of the polypeptide chemically linked to the protein. For example, a modification can be an altered carbohydrate or PEG structure linked to the protein. "Amino acid modification" in this article refers to amino acid substitutions, insertions, and / or deletions in a polypeptide sequence. For clarity, unless otherwise specified, amino acid modifications always refer to amino acids encoded by DNA, such as the 20 amino acids that have a codon in DNA and RNA.

[0146] As used herein, “amino acid substitution” or “replacement” means replacing an amino acid at a specific position in the parent polypeptide sequence with a different amino acid. Specifically, in some embodiments, substitution is used for an amino acid that is not naturally occurring at a specific position or in any organism. For example, substitution of E272Y refers to a variant polypeptide, in this case, the Fc variant, where the glutamic acid at position 272 is replaced with tyrosine. For clarity, a protein that has been engineered to alter its nucleic acid coding sequence but not its starting amino acid (e.g., replacing CGG (encoding arginine) with CGA (still encoding arginine) to increase expression levels in a host organism) is not “amino acid substitution”; that is, although a new gene encoding the same protein is generated, if the protein has the same amino acid at its starting specific position, then the protein is not an amino acid substitution.

[0147] As used herein, “amino acid insertion” or “insertion” means the addition of an amino acid sequence at a specific position in the parent polypeptide sequence. For example, -233E or 233E indicates the insertion of glutamic acid after position 233 and before position 234. Similarly, -233ADE or A233ADE indicates the insertion of AlaAspGlu after position 233 and before position 234.

[0148] As used herein, “amino acid deletion” or “deletion” means the removal of an amino acid sequence at a specific position in the parent polypeptide sequence. For example, E233- or E233#, E233(), or E233del refers to the deletion of glutamic acid at position 233. Additionally, EDA233- or EDA233# refers to the deletion of the sequence GluAspAla starting at position 233.

[0149] As used herein, "variant protein" or "protein variant" or "variant" means a protein that differs from its parent protein by at least one amino acid modification. Compared to the parent protein, a protein variant has at least one amino acid modification.

[0150] However, there is not much difference that variant proteins will not align with parent proteins when using the alignment procedures described below. Typically, using alignment procedures such as BLAST described below, variant proteins (such as the variant Fc domains outlined herein) are at least 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identical to parent proteins.

[0151] As described below, in some embodiments, the parental polypeptide, such as the Fc parental polypeptide, is a human wild-type sequence, such as a recurrent constant domain or Fc region derived from IgG1, IgG2, IgG3, or IgG4. However, human sequences with variants can also be used as "parental polypeptides," for example, they may comprise the IgG1 / 2 hybrid of US Publication 2006 / 0134105. Protein variants as used herein will preferably have at least about 80% identity with the parental protein sequence, and most preferably at least about 90% identity, more preferably at least about 95% to 98% to 99% identity. Therefore, as used herein, "antibody variant" or "variant antibody" means an antibody that differs from the parental antibody by at least one amino acid modification; as used herein, "IgG variant" or "variant IgG" means an antibody that differs from the parental IgG (again, in some cases, from a human IgG sequence) by at least one amino acid modification; and as used herein, "immunoglobulin variant" or "variant immunoglobulin" means an immunoglobulin sequence that differs from the parental immunoglobulin sequence by at least one amino acid modification. As used herein, “Fc variant” or “variant Fc” means a protein that includes amino acid modifications of the Fc domain compared to the Fc domain of human IgG1, IgG2, or IgG4.

[0152] The Fc variants of this invention are defined according to the amino acid modifications constituting them. Thus, for example, N434S or 434S is an Fc variant with a serine substitution at position 434 relative to the parental Fc polypeptide, where the numbering is based on the EU index. Similarly, M428L / N434S defines an Fc variant with substitutions for both M428L and N434S relative to the parental Fc polypeptide. Identification of WT amino acids may not be specified; in this case, the aforementioned variants are referred to as 428L / 434S. It should be noted that the order of substitutions is arbitrary, that is, for example, N434S / M428L is the same Fc variant as M428L / N434S, etc. For all positions relating to antibodies discussed in this invention, unless otherwise indicated, the amino acid position numbering is based on the EU index. The EU index, or the EU index in a scheme such as Kabat or the EU numbering scheme, refers to the number of the EU antibody.

[0153] Kabat et al. collected multiple primary sequences of the variable regions of the heavy and light chains. Based on the conservation of the sequences, they classified the individual primary sequences into CDRs and frames and made a list of them (see *Sequences of Immunological Interest*, 5th edition, NIH Publication, No. 91-3242, E.A. Kabat et al., incorporated herein by reference in full). See also Edelman et al., 1969, *Proceedings of the National Academy of Sciences of the United States of America*, 63:78-85, incorporated herein by reference in full. Modifications can be additions, deletions, or substitutions.

[0154] In this document, "protein" refers to at least two covalently linked amino acids, including proteins, polypeptides, oligopeptides, and peptides. Furthermore, the polypeptide constituting the antibody of this invention may include synthetic derivatization, glycosylation, PEGylation, cyclization, cyclization, linkers of other molecules, fusion with a protein or protein domain, and the addition of a peptide tag or label.

[0155] As used herein, “residue” refers to a position in a protein or its associated amino acid identity. For example, asparagine 297 (also known as Asn297 or N297) is the residue at position 297 in the human antibody IgG1.

[0156] As used herein, “Fab” or “Fab region” means a polypeptide that includes the VH, CH1, VL, and CL immunoglobulin domains, which are typically located on two different polypeptide chains (e.g., VH-CH1 on one chain and VL-CL on another). Fab can refer to this region in the case of separation or in the context of the bispecific antibody of the present invention. In the context of Fab, in addition to the CH1 and CL structures, Fab also includes the Fv region.

[0157] As used herein, “Fv” or “Fv fragment” refers to a polypeptide that includes the VL and VH domains of ABD. The Fv region can be formatted as both Fab (as discussed above, typically two distinct polypeptides that also contain constant regions as outlined above) and scFv, where the combination of the vl and vh domains (typically combined with linkers as discussed herein) forms the scFv.

[0158] In this document, "single-chain Fv" or "scFv" refers to a variable heavy domain that is typically covalently linked to a variable light domain using scFv connectors as discussed herein to form an scFv or scFv domain. The scFv domain can be oriented in either direction (vh-connector-vl or vl-connector-vh) from the N-terminus to the C-terminus. In the sequences depicted in the sequence listing and figures, the order of the vh and vl domains is indicated by name; for example, HX LY indicates vh-connector-vl from the N-terminus to the C-terminus, and LY HX indicates vl-connector-vh.

[0159] As used herein, “IgG subclass modification” or “isotype modification” refers to an amino acid modification that converts one amino acid of an IgG isotype into the corresponding amino acid in a different alignment IgG isotype. For example, because IgG1 includes tyrosine and IgG2 includes phenylalanine at position 296 EU, the F296Y substitution of IgG2 is considered an IgG subclass modification.

[0160] As used herein, “non-naturally occurring modification” means an amino acid modification that is not of the same type. For example, since human IgG does not include serine at position 434, the substitution of 434S in IgG1, IgG2, IgG3, or IgG4 (or their hybrids) is considered a non-naturally occurring modification.

[0161] As used in this article, "amino acid" and "amino acid identity" refer to one of the 20 naturally occurring amino acids encoded by DNA and RNA.

[0162] As used herein, "effective function" refers to the biochemical event that causes the antibody Fc region to interact with the FC receptor or configuration. Effector functions include, but are not limited to, ADCC, ADCP, and CDC.

[0163] As used herein, “IgG Fc ligand” means any molecule, especially a polypeptide, from any organism that binds to the Fc region of an IgG antibody to form an Fc / Fc ligand complex. Fc ligands include, but are not limited to, FcγRI, FcγRII, FcyRIII, FcRn, Clq, C3, mannan-binding lectin, mannose receptor, staphylococcal protein A, streptococcal protein G, and viral FcγR. Fc ligands also include Fc receptor homologs (FcRH), which are Fc receptor families homologous to FcyR (Davis et al., 2002, *Immunological Reviews* 190: 123-136). Fc ligands may include undiscovered molecules that bind to Fc. Specific IgG Fc ligands are FcRn and Fcγ receptors. As used herein, “Fc ligand” means any molecule, especially a polypeptide, from any organism that binds to the Fc region of an antibody to form an Fc / Fc ligand complex.

[0164] As used herein, “Fcγ receptor” or “FcγR (FcgammaR)” refers to any member of the protein family that binds to the Fc region of IgG antibodies and is encoded by the FcγR gene. In humans, this family includes, but is not limited to: FcγRI (CD64), including isoforms FcγRIa, FcγRIb, and FcγRIc; FcγRII (CD32), including isoforms FcγRIIa (including alloforms H131 and R131), FcγRIIb (including FcγRIIb-1 and FcγRIIb-2), and FcγRIIc; and FcγRIII (CD16), including isoforms FcγRIIIa (including V158 and F158) and FcγRIIIb (including alloforms FcγRIIb-NA1 and FcγRIIb-NA2) (Jefferis et al., 2002, *Immunol Letters*). Lett. 82:57-65, and any undiscovered human FcγR or FcγR allotypes or allotypes. FcγR can be derived from any organism, including but not limited to humans, mice, rats, rabbits, and monkeys. Mouse FcγR includes, but is not limited to, FcγRI (CD64), FcγRII (CD32), FcγRIII (CD16), and FcγRIII-2 (CD16-2), and any undiscovered mouse FcγR or FcγR allotypes or allotypes.

[0165] As used herein, “FcRn” or “nascent Fc receptor” means a protein that binds to the Fc region of an IgG antibody and is at least partially encoded by the FcRn gene. FcRn can originate from any organism, including but not limited to humans, mice, rats, rabbits, and monkeys. As is known in the art, functional FcRn proteins comprise two peptides, often referred to as the heavy chain and the light chain. The light chain is β-2-microglobulin, and the heavy chain is encoded by the FcRn gene. Unless otherwise indicated herein, FcRn or FcRn protein refers to the complex of the FcRn heavy chain and β-2-microglobulin. Various FcRn variants are used to increase binding to the FcRn receptor and, in some cases, to increase serum half-life. An “FcRn variant” is a variant that increases binding to the FcRn receptor, and suitable FcRn variants are shown below.

[0166] As used herein, “parental polypeptide” means the starting polypeptide that is subsequently modified to generate a variant. A parental polypeptide can be a naturally occurring polypeptide or a variant or engineered version of a naturally occurring polypeptide. Therefore, as used herein, “parental immunoglobulin” means an unmodified immunoglobulin polypeptide that has been modified to generate a variant, and as used herein, “parental antibody” means an unmodified antibody that has been modified to generate a variant antibody. It should be noted that “parental antibody” includes known commercially available recombinant antibodies as outlined below. In this context, “parental Fc domain” will be relative to the listed variants; thus, “variant human IgG1 Fc domain” will be compared to the parental Fc domain of human IgG1, “variant human IgG4 Fc domain” will be compared to the parental Fc domain of human IgG4, and so on.

[0167] As used herein, “Fc” or “Fc region” or “Fc domain” refers to a polypeptide that includes the CH2-CH3 domain of an IgG molecule and, in some cases, contains a hinge. When human IgG1 is numbered in EU, the CH2-CH3 domain comprises amino acids 231 to 447, and the hinge is 216 to 230. Therefore, the definition of an “Fc domain” includes amino acids 231 to 447 (CH2-CH3) or 216 to 447 (hinge-CH2-CH3) or fragments thereof. In this context, an “Fc fragment” can be derived from either or both of the N-terminus and C-terminus, containing fewer amino acids but still capable of forming a dimer with another Fc domain or Fc fragment, as can be detected using standard methods typically based on size (e.g., non-denaturing chromatography, size exclusion chromatography, etc.). Human IgG Fc domains are particularly used in this invention and can be Fc domains derived from human IgG1, IgG2, or IgG4.

[0168] A “variant Fc domain” contains amino acid modifications compared to the parental Fc domain. Therefore, a “variant human IgG Fc domain” is a variant Fc domain that contains amino acid modifications (usually amino acid substitutions, but in the case of ablation variants, amino acid deletions) compared to the human IgG1 Fc domain. Typically, the variant Fc domain has at least about 80%, 85%, 90%, 95%, 97%, 98%, or 99% identity with the corresponding parental IgG Fc domain (using the identity algorithm discussed below; one embodiment utilizes the BLAST algorithm as known in the art, using default parameters). Alternatively, the variant Fc domain may have 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acid modifications compared to the parental Fc domain. In addition, as discussed herein, the variant Fc domains in this paper retain the ability to form dimers with another Fc domain, as measured using known techniques described herein, such as non-denaturing gel electrophoresis.

[0169] In this article, “heavy chain constant region” refers to the CH1-hinge-CH2-CH3 portion of an antibody (or a fragment thereof) that does not contain a variable restructure domain; in the EU numbering of human IgG1, this is amino acids 118 to 447. In this article, “heavy chain constant fragment” refers to a heavy chain constant region containing fewer amino acids from either or both of the N-terminus and C-terminus but still capable of forming a dimer with another heavy chain constant region.

[0170] As used in this article, "position" refers to a location within the protein sequence. Positions can be numbered sequentially or according to established formats such as antibody numbering (EU index).

[0171] As used herein, “target antigen” means a molecule that binds specifically through an antigen-binding domain comprising a variable region or a given antibody. In this case, as discussed below, the target antigen is a checkpoint inhibitor protein.

[0172] In this document, in the context of the monomers of the heterodimeric antibodies of the present invention, "strandedness" means, similar to the "matching" two strands of DNA, where heterodimerized variants are incorporated into each monomer to preserve the ability to "match" to form a heterodimer. For example, if some pI variants are engineered for monomer A (e.g., to make the pI higher), spatial variants belonging to the also usable "charge pair" do not interfere with the pI variants; for example, charge variants that make the pI higher are placed on the same "strand" or "monomer" to preserve the function of both. Similarly, for "tilted" variants that appear in pairs as more fully outlined below, those skilled in the art will consider that the pI determines which strand or monomer one of the pairs will enter, thereby maximizing pI separation by also using pIs of tilted variants.

[0173] As used in this article, "target cell" refers to a cell that expresses the target antigen.

[0174] In this document, in the context of generating bispecific antibodies according to the invention, "host cell" means a cell containing exogenous nucleic acid encoding components of a bispecific antibody and capable of expressing the bispecific antibody under suitable conditions. Suitable host cells are discussed below.

[0175] As used herein, a “variable region” or “variable domain” means an immunoglobulin comprising one or more Ig domains encoded by the Vκ, Vλ, and / or VH genes, which respectively constitute the gene sites of the κ, λ, and heavy chain immunoglobulins, and containing a CDR that confers antigen specificity. Therefore, a “variable heavy domain” pairs with a “variable light domain” to form an antigen-binding domain (“ABD”). Additionally, each variable domain comprises three hypervariable regions (“complementarity-determining regions”, “CDRs”) arranged from the amino-terminus to the carboxyl-terminus in the following order: (vhCDR1, vhCDR2, and vhCDR3 of the variable heavy domain and vlCDR1, vlCDR2, and vlCDR3 of the variable light domain) and four frame regions (FRs): FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4.

[0176] In this article, "wild-type or WT" refers to an amino acid or nucleotide sequence found in nature, including allelic variations. WT proteins have an amino acid or nucleotide sequence that has not been intentionally modified.

[0177] This invention provides multiple antibody domains that share sequence identity with human antibody domains. Sequence identity between two similar sequences (e.g., antibody variable domains) can be measured using algorithms such as those found in: Smith, TF, and Waterman, MS (1981), “Comparison of Biosequences,” Adv. Appl. Math. 2:482 [Local Homology Algorithm]; Needleman, SB, and Wunsch, CD. (1970), “A General Method Applicable To The Search For Similarities In The Amino Acid Sequence Of Two Proteins,” J. Mol. Biol. 48:443 [Homology Algorithm]; Pearson, WR., and Lipman, DJ (1988), “Improved Tools For Biological Sequence.” "Comparison" Proceedings of the National Academy of Sciences (Proc. Natl. Acad. Sci. (USA) 85:2444 [Similarity Search Methods]; or Altschul, SF et al., (1990) "Basic Local Alignment Search Tool" Journal of Molecular Biology 215:403-10, "BLAST" algorithm, see https: / / blast.nchi.nlm.nih.gov / Blast.cgi. When using any of the above algorithms, use the default parameters (for window length, space penalty, etc.). In one embodiment, the BLAST algorithm is used to perform sequence identity verification using the default parameters.

[0178] The antibodies of this invention are typically isolated or recombinant. In describing the various polypeptides disclosed herein, “isolated” means a polypeptide that has been identified and isolated from and / or recovered from the cells or cell cultures expressing it. Generally, isolated polypeptides are prepared by at least one purification step. “Isolated antibody” refers to an antibody that is substantially free of other antibodies with different antigen specificities. “Recombinant” means the generation of antibodies in exogenous host cells using recombinant nucleic acid technology, and antibodies can also be isolated.

[0179] "Specific binding," "specifically binding," or "specific to" a particular antigen or epitope means a binding that is measurably different from nonspecific interactions. Specific binding can be measured, for example, by determining the binding of a molecule against a control molecule, typically a molecule with a similar structure and no binding activity. For instance, specific binding can be determined by competing with a control molecule that resembles a target.

[0180] Specific binding to a particular antigen or epitope can be demonstrated, for example, by antibodies having a KD of at least about 10 for antigens or epitopes. -4 M, at least about 10 -5 M, at least about 10 -6 M, at least about 10 -7 M, at least about 10 -8 M, at least about 10 -9 M, alternatively at least about 10 -10 M, at least about 10 -11 M, at least about 10 -12 M or greater, where KD refers to the dissociation rate of a specific antibody-antigen interaction. Typically, antibodies that specifically bind to an antigen or epitope will have a KD that is 20, 50, 100, 500, 1000, 5,000, 10,000, or more times larger than that of a control molecule. Furthermore, the specificity of a particular antigen or epitope can be demonstrated, for example, by antibodies having KA or Ka that is at least 20, 50, 100, 500, 1000, 5,000, 10,000, or more times larger than the control, where KA or Ka refers to the association rate of a specific antibody-antigen interaction. Binding affinity is typically measured using Biacore, SPR, or BLI assays.

[0181] E. Antibody

[0182] As discussed herein, the present invention relates to the generation of bispecific checkpoint antibodies that bind two different checkpoint antigens. The term "antibody" is generally used as discussed below. Antibodies used in the present invention can take the form of various formats as described herein, including conventional antibodies described herein and depicted in the figures, as well as antibody derivatives, fragments, and mimics.

[0183] Traditional antibody structural units typically comprise tetramers. Each tetramer usually consists of two pairs of identical polypeptide chains, each pair having a "light" chain (typically with a molecular weight of about 25 kDa) and a "heavy" chain (typically with a molecular weight of about 50 kDa to 70 kDa). Human light chains are classified as κ and λ light chains. This invention relates to bispecific antibodies typically based on the IgG class, which has multiple subclasses, including but not limited to IgG1, IgG2, IgG3, and IgG4. Typically, IgG1, IgG2, and IgG4 are used more frequently than IgG3. It should be noted that IgG1 has different allotypes with polymorphism at 356 (D or E) and 358 (L or M). The sequences depicted herein use the 356E / 358M allotype; however, other allotypes are also included herein. That is, any sequence containing the IgG1 Fc domain as described herein may have 356D / 358L instead of the 356E / 358M allotype.

[0184] In addition, many of the antibodies described herein have at least one cysteine ​​residue at position 220 that has been replaced by a serine residue; typically, this is on the “scFv monomer” side of most sequences depicted herein, but it may also be on the “Fab monomer” side or both to reduce disulfide formation. In this document, one or both of these substituted cysteine ​​residues (C220S) are specifically included in the sequence.

[0185] Therefore, as used herein, “isotype” means any subclass of immunoglobulin defined by the chemical and antigenic properties of its constant region. It should be understood that therapeutic antibodies may also include hybrids of isotypes and / or subclasses. For example, as illustrated by reference to U.S. Publication 2009 / 0163699, the present invention uses a human IgG1 / G2 hybrid.

[0186] Hypervariable regions typically encompass approximately amino acid residues 24 to 34 (LCDR1; "L" indicates light chain), 50 to 56 (LCDR2), and 89 to 97 (LCDR3) in the light chain variable region, and approximately amino acid residues around 31 to 35B (HCDR1; "H" indicates heavy chain), 50 to 65 (HCDR2), and 95 to 102 (HCDR3) in the heavy chain variable region; Rabat et al., *SEQUENCES OF PROTEINS OF IMMUNOLOGICAL INTEREST*, 5th edition, Public Health Service, National Institutes of Health. Health), Bethesda, Maryland (Md.) (1991) and / or those residues forming the hypervariable ring (e.g., residues 26 to 32 (LCDR1), 50 to 52 (LCDR2), and 91 to 96 (LCDR3) in the light chain variable region and 26 to 32 (HCDR1), 53 to 55 (HCDR2), and 96 to 101 (HCDR3) in the heavy chain variable region; Chothia and Lesk (1987) Journal of Molecular Biology 196: 901-917. The specific CDRs of the present invention are described below.

[0187] As those skilled in the art will appreciate, the precise numbering and placement of CDRs can vary in different numbering systems. However, it should be understood that the disclosure of variable weight and / or variable light sequences includes the disclosure of the associated (inherent) CDR. Thus, the disclosure of each variable weight region is the disclosure of the vhCDR (e.g., vhCDR1, vhCDR2, and vhCDR3), and the disclosure of each variable light region is the disclosure of the vlCDR (e.g., vlCDR1, vlCDR2, and vlCDR3). A useful comparison of CDR numbering is provided below, see Lafranc et al., Developmental Immunology and Comparative Immunology (Dev.Comp.Immunol.) 27(1): 55-77 (2003):

[0188] Table 1

[0189]

[0190] Throughout this specification, the Kabat numbering system is generally used when referring to residues in the variable domains (generally residues 1 to 107 in the light chain variable region and residues 1 to 113 in the heavy chain variable region), and the EU numbering system is used for the Fc region (e.g., Kabat et al., ibid. (1991)).

[0191] Another type of Ig domain in heavy chains is the hinge region. In this document, "hinge," "hinge region," or "antibody hinge region" refers to a flexible polypeptide comprising the amino acids between the first and second constant domains of the antibody. Structurally, the IgG CH1 domain terminates at EU position 215, and the IgG CH2 domain begins at residue EU position 231. Therefore, for IgG, the antibody hinge is defined herein as encompassing positions 216 (E216 in IgG1) to 230 (p230 in IgG1), where the numbering is based on an EU index such as Kabat. In some cases, a "hinge fragment" is used, which contains fewer amino acids at either or both of the N-terminus and C-terminus of the hinge domain. As noted herein, pI variants may also be located within the hinge region.

[0192] Light chains typically consist of two domains: a variable light structure domain (containing a light chain CDR and forming an Fv region together with a variable heavy structure domain) and a constant light chain region (often referred to as CL or Cκ).

[0193] The additional area of ​​interest, which is outlined below, is the Fc area.

[0194] This invention provides a large number of different CDR sets. In this case, a “complete CDR set” includes three variable light CDRs and three variable heavy CDRs, such as vlCDR1, vlCDR2, vlCDR3, vhCDR1, vhCDR2, and vhCDR3. Importantly, these can be part of a larger variable light domain or a variable heavy domain. Furthermore, as more fully outlined herein, when using heavy and light chains (e.g., when using Fab), the variable heavy and variable light domains can be on their respective polypeptide chains, or, in the case of scFv sequences, on a single polypeptide chain.

[0195] CDRs facilitate the formation of antigen-binding sites, or more specifically, epitope-binding sites of antibodies. An "epitaph" is a determinant that interacts with a specific antigen-binding site in a variable light of an antibody molecule, known as a complementary site. Epitopes are groups of molecules such as amino acids or sugar side chains and typically possess specific structural and charge properties. A single antigen can have more than one epitope.

[0196] Epitopes can include amino acid residues that directly participate in binding (also known as the immunodominant component of epitopes) as well as other amino acid residues that do not directly participate in binding, such as amino acids that are effectively blocked by peptides that specifically bind to antigens; in other words, the amino acid residues are within the footprint of peptides that specifically bind to antigens.

[0197] Epitopes can be conformational or linear. Conformational epitopes are generated from spatially aligned amino acids from different segments of a linear polypeptide chain. Linear epitopes are generated from adjacent amino acid residues in the polypeptide chain. The difference between conformational and non-conformational epitopes lies in the fact that, in the presence of denaturing solvents, conformational epitopes lose their binding to the former but not the latter.

[0198] In a unique spatial conformation, an epitope typically comprises at least three, and more commonly at least five or eight to ten amino acids. Antibodies recognizing the same epitope can be verified using a simple immunoassay demonstrating the ability of one antibody to block the binding of another antibody to a target antigen, such as through "compartmentalization." As outlined below, the present invention includes not only the antigen-binding domains and antibodies enumerated herein, but also those that competitively bind to epitopes bound by the enumerated antigen-binding domains.

[0199] Therefore, this invention provides different antibody domains. As described herein and known in the art, the heterodimeric antibodies of this invention comprise different domains within the heavy and light chains, and these domains may overlap. These domains include, but are not limited to, Fc domains, CH1 domains, CH2 domains, CH3 domains, hinge domains, heavy constant domains (CH1-hinge-Fc domains or CH1-hinge-CH2-CH3 domains), variable heavy domains, variable light domains, light constant domains, Fab domains, and scFv domains.

[0200] Therefore, the “Fc domain” includes a -CH2-CH3 domain and an optional hinge domain (-H-CH2-CH3). In the embodiments described herein, when the scFv is connected to the Fc domain, the C-end of the scFv construct is connected to all or part of the hinge of the Fc domain; for example, it is typically connected to the beginning of the hinge, i.e., the sequence EPKS. The heavy chain includes a variable heavy domain and a constant domain, the constant domain including a CH1-optional hinge-Fc domain comprising CH2-CH3. The light chain includes a variable light chain and a light constant domain. The scFv includes a variable heavy chain, an scFv connector, and a variable light domain. In most of the constructs and sequences outlined herein, the C-end of the variable heavy chain is connected to the N-end of the scFv connector, and the C-end of the scFv connector is connected to the N-end of the variable light chain (N-vh-connector-vl-C), but this can be switched (N-vl-connector-vh-C).

[0201] Some embodiments of the present invention include at least one scFv domain, which, while not naturally occurring, typically comprises variable heavy and variable light domains linked together by scFv connectors. As outlined herein, while scFv domains typically oriented from the N-end to the C-end as vh-scFv connector-vl, this can be reversed to vl-scFv connector-vh for any scFv domain (or those constructed using the vh and vl sequences of Fab), with optional connectors at one or both ends depending on the format (generally see Figure 1).

[0202] As illustrated herein, several suitable linkers exist (for use as domain linkers or scFv linkers) that can be used for covalently linking the listed domains, comprising conventional peptide bonds generated through recombination techniques. In some embodiments, the linker peptide may primarily comprise the following amino acid residues: Gly, Ser, Ala, or Thr. The linker peptide should have a length sufficient to link two molecules in such a way that the assumed correct conformation of the molecules relative to each other allows them to retain the desired activity. In one embodiment, the linker is about 1 to 50 amino acids in length, preferably about 1 to 30 amino acids in length. In one embodiment, linkers of 1 to 20 amino acids in length can be used, and in some embodiments, about 5 to about 10 amino acids are used. Useful linkers include glycine-serine polymers, glycine-alanine polymers, alanine-serine polymers, and other flexible linkers, wherein the glycine-serine polymers include, for example, (GS)n, (GSGGS)n (SEQ ID NO: 37756), (GGGGS)n (SEQ ID NO: 37757), and (GGGS)n (SEQ ID NO: 37758), where n is an integer of at least one (and typically 3 to 4). Alternatively, various non-protein polymers can be used as linkers, including, but not limited to, polyethylene glycol (PEG), polypropylene glycol, polyoxyethylene, or copolymers of polyethylene glycol and polypropylene glycol.

[0203] Other linker sequences can contain any sequence with a CL / CH1 domain of any length, but not all residues of the CL / CH1 domain; for example, the first 5 to 12 amino acid residues of the CL / CH1 domain. Linkers can be derived from immunoglobulin light chains, such as Cκ or Cλ. Linkers can be derived from any type of immunoglobulin heavy chain, including, for example, Cγ1, Cγ2, Cγ3, Cγ4, Cα1, Cα2, Cδ, Cε, and Cμ. Linker sequences can also be derived from other proteins, such as Ig-like proteins (e.g., TCR, FcR, KIR), hinge region-derived sequences, and other natural sequences from other proteins.

[0204] In some embodiments, a connector is a "domain connector" used to join any two domains together as outlined herein. For example, in Figure 1F In this embodiment, there may be a domain linker that connects the C-terminus of the CH1 domain of Fab to the N-terminus of scFv, and another optional domain linker that connects the C-terminus of scFv to the CH2 domain (however, in many embodiments, a hinge is used as this domain linker). While any suitable linker can be used, many embodiments utilize glycine-serine polymers as domain linkers and any peptide sequence that allows recombination of the two domains to a length and flexibility sufficient to allow each domain to retain its biological function, said glycine-serine polymers comprising, for example, (GS)n, (GSGGS)n (SEQ ID NO: 37756), (GGGGS)n (SEQ ID NO: 37757), and (GGGS)n (SEQ ID NO: 37758), where n is an integer of at least one (and typically 3 to 4 to 5). In some cases, and when noting the term "chain-like," as outlined below, charged domain linkers, such as those used in some embodiments of the scFv linker, may be used.

[0205] In some embodiments, the connector is an scFv connector for covalently connecting vh and vl domains as discussed herein. In some cases, the scFv connector is a charged scFv connector, and multiple charged scFv connectors are shown in...

[0206] As shown in Figure 7, the present invention further provides charged scFv linkers for facilitating pI separation between the first and second monomers. That is, by incorporating charged scFv linkers that are positively or negatively charged (or both in the case of scFvs on different monomers), this allows monomers including charged linkers to change their pI without further altering the Fc domain. These charged linkers can be substituted into any scFv containing a standard linker. Again, as those skilled in the art will appreciate, the charged scFv linkers are used on the correct “chain” or monomer depending on the desired pI change. For example, as discussed herein, to form a triple F-format heterodimeric antibody, the original pI of the Fv region for each desired antigen-binding domain is calculated, and one is selected to form the scFv, and a positive or negative linker is selected based on the pI.

[0207] Charged structural domain linkers can also be used to increase the pI separation of the monomers of the present invention, and therefore those included in FIG7

[0208] This can be used in any embodiment that utilizes connectors as described herein.

[0209] Specifically, the format depicted in Figure 1 is an antibody, often referred to as a "heterodimeric antibody," meaning a protein having at least two associated Fc sequences that self-assemble into a heterodimeric Fc domain and at least two Fv regions, whether as Fab or as scFv.

[0210] F. Chimeric antibodies and humanized antibodies

[0211] In some embodiments, the antibodies of the present invention comprise heavy chain variable regions from a specific germline heavy chain immunoglobulin gene and / or light chain variable regions from a specific germline light chain immunoglobulin gene. For example, such antibodies may comprise, or be composed of, human antibodies comprising, a heavy chain region or light chain region that is a "product" of, or "derived from", a specific germline sequence. Human antibodies that are a "product" of, or "derived from" a human germline immunoglobulin sequence can be identified by comparing the amino acid sequence of the human antibody with the amino acid sequence of a human germline immunoglobulin and selecting the human germline immunoglobulin sequence that is most closely sequenced (i.e., has the highest percentage of identity) with the human antibody. Human antibodies that are a "product" of, or "derived from" a specific human germline immunoglobulin sequence may contain amino acid differences compared to the germline sequence due to, for example, naturally occurring somatic mutations or intentionally introduced site-directed mutations. However, humanized antibodies typically share at least 90% amino acid sequence identity with the amino acid sequence encoded by human germline immunoglobulin genes and, when compared with the amino acid sequences of germline immunoglobulins from other species (e.g., murine germline sequences), contain amino acid residues that identify the antibody as derived from the human sequence. In some cases, humanized antibodies may be at least 95%, 96%, 97%, 98%, or 99% identical in amino acid sequence to the amino acid sequence encoded by germline immunoglobulin genes, or even at least 96%, 97%, 98%, or 99%. Typically, humanized antibodies derived from a particular human germline sequence will show no more than 10 to 20 amino acid differences from the amino acid sequence encoded by human germline immunoglobulin genes (previously, the number of variants was generally low before the introduction of any tilt variants, pI variants, and ablation variants herein; i.e., before the introduction of the variants of the present invention). In some cases, humanized antibodies may show differences of no more than 5 or even no more than 4, 3, 2 or 1 amino acids from the amino acid sequence encoded by germline immunoglobulin genes (again, prior to the introduction of any tilt variants, pI variants and ablation variants herein; i.e. prior to the introduction of the variants of the present invention, the number of variants is generally low).

[0212] In some embodiments, the parent antibody has been affinity-matured, as is known in the art. Humanization and affinity maturation can be performed using structure-based methods, such as those described in USSN 11 / 004,590. Selection-based approaches can be used to humanize and / or mature the affinity of antibody variable regions, including but not limited to the following methods, all of which are incorporated herein by reference in their entirety: Wu et al., 1999, J.Mol.Biol. 294: 151-162; Baca et al., 1997, J.Biol.Chem. 272(16): 10678-10684; Rosok et al., 1996, J.Biol.Chem. 271(37): 22611-22618; Rader et al., 1998, Proceedings of the National Academy of Sciences of the United States of America 95: 8910-8915; Krauss et al., 2003, Protein Engineering 16(10): 753-759. Other humanization methods may involve the transplantation of only a portion of the CDR, including but not limited to the methods described below, all of which are incorporated herein by reference in their entirety: USSN 09 / 810,510; Tan et al., 2002, Journal of Immunology 169:1119-1125; De Pascalis et al., 2002, Journal of Immunology 169:3076-3084.

[0213] IV. Heterodimeric antibodies

[0214] Therefore, in some embodiments, the present invention provides heterodimeric checkpoint antibodies that rely on the use of two different heavy chain variants of the Fc sequence that will self-assemble to form a heterodimeric Fc domain and a heterodimeric antibody.

[0215] This invention relates to novel constructs for providing heterodimeric antibodies that allow binding to more than one checkpoint antigen or ligand, such as allowing bispecific binding. The heterodimeric antibody construct is based on the self-assembly of two Fc domains of the antibody's heavy chain, for example, two "monomers" assembling into a "dimer". Heterodimeric antibodies are formed by altering the amino acid sequence of each monomer, as discussed more fully below. Therefore, this invention generally relates to the creation of heterodimeric checkpoint antibodies that can co-bind antigens in various ways, depending on differences on each chain to promote heterodimer formation and / or allow amino acid variants in the constant region of the heterodimer that are easier to purify compared to homodimers.

[0216] Therefore, this invention provides bispecific antibodies. A persistent problem in antibody technology is the desire for “bispecific” antibodies that bind simultaneously to two different antigens, typically allowing for the accessibility of different antigens and the generation of novel functions and therapies. Typically, these antibodies are formed in host cells by including genes for each heavy and light chain. This usually results in the formation of the desired heterodimer (AB) as well as two homodimers (AA and BB (excluding the light chain heterodimer issue)). However, a major obstacle to the formation of bispecific antibodies is the difficulty in purifying heterodimer antibodies away from homodimer antibodies and / or biasing heterodimer formation compared to homodimer formation.

[0217] Multiple mechanisms exist for generating the heterodimers of this invention. Furthermore, as those skilled in the art will appreciate, these mechanisms can be combined to ensure high heterodimerization. Therefore, the amino acid variants that result in the production of heterodimers are referred to as “heterodimerization variants.” As discussed below, heterodimerization variants may include spatial variants (e.g., “bulge-hole” or “tilt” variants and “charge pairs” variants as described below) and “pI variants,” which allow for the purification of homodimers away from heterodimers. Useful mechanisms for heterodimerization, as generally described in WO 2014 / 145806 which is incorporated herein by reference in its entirety and particularly in WO 2014 / 145806 for the discussion of “heterodimerization variants”, include “bulge-hole” (“KIH”; sometimes referred to herein as the “tilt” variant (see discussion in WO 2014 / 145806)), “isoelectric turning” or “charge pairing” as described in WO 2014 / 145806, pI variants as described in WO 2014 / 145806, and general additional Fc variants as outlined in WO 2014 / 145806 and below.

[0218] In this invention, several basic mechanisms exist that lead to easy purification of heterodimerized antibodies; one mechanism relies on the use of pI variants, such that each monomer has a different pI, thereby allowing isoelectric purification of AA, AB, and BB dimer proteins. Alternatively, some scaffold formats, such as the “triple F” format, also allow for size-based separation. As further outlined below, the formation of “tilted” heterodimers may also be possible compared to homodimers. Therefore, combinations of spatial heterodimerization variants with pI or charge-pair variants are particularly useful in this invention.

[0219] Typically, embodiments particularly applicable to this invention rely on a group of variants containing tilted variants that encourage heterodimerization formation compared to homodimerization formation, coupled with pI variants that increase the pI difference between the two monomers to facilitate the purification of heterodimers away from homodimers.

[0220] Furthermore, as more fully outlined below, depending on the format of the heterodimeric antibody, the monomer's constant and / or Fc domains may contain pI variants, or charged linkers, domain linkers, or scFv linkers may be used. That is, a scaffold utilizing one or more scFvs, such as a triple F format, may contain charged scFv linkers (positively or negatively charged) that provide additional pI boosts for purification purposes. As those skilled in the art will appreciate, some triple F formats can be used with only charged scFv linkers without additional pI adjustments, but the present invention also provides pI variants and / or charged domain linkers on one or both of the monomers. Additionally, additional amino acid engineering for alternative functions can also confer pI variations, such as Fc, FcRn, and KO variants.

[0221] In this invention, which uses pI as a separation mechanism to allow the purification of heterodimeric proteins, amino acid variants can be introduced into one or both monomeric polypeptides; that is, the pI of one monomer (referred to herein as "monomer A") can be engineered away from monomer B, or both monomer A and monomer B can be altered, wherein the pI of monomer A increases and the pI of monomer B decreases. As discussed, the pI change of any one or both monomers can be accomplished by removing or adding charged residues (e.g., replacing a neutral amino acid with a positively or negatively charged amino acid residue, such as glycine to glutamic acid), changing charged residues from positively or negatively charged to the opposite charge (e.g., aspartic acid to lysine), or changing charged residues to neutral residues (e.g., charge loss; lysine to serine). Several of these variants are shown in the accompanying drawings.

[0222] Therefore, this embodiment of the invention provides a sufficient pI variation to produce at least one of the monomers, thereby allowing the heterodimer to be separated from the homodimer. As those skilled in the art will understand and further discussed below, this can be accomplished by using a constant region of the "wild-type" heavy chain and a variant region engineered to increase or decrease its pI (wt A-+B or wt A--B) or increasing one region and decreasing the other (A+-B- or A-B+).

[0223] Therefore, typically, components of some embodiments of the present invention are amino acid variants in the constant region of an antibody, said amino acid variant involving altering the isoelectric point (pI) of at least one (if not both) of the monomers to form a “pI antibody” by incorporating an amino acid substitution (“pI variant” or “pI substitution”) into one or both of the monomers. As shown herein, the separation of heterodimers from two homodimers can be accomplished when the pIs of the two monomers differ by as little as 0.1 pH units, with 0.2, 0.3, 0.4, and 0.5 or more pH units used in the present invention.

[0224] As those skilled in the art will appreciate, the number of pI variants to be included on each or both monomers for achieving good separation will depend in part on the starting pI of the component, such as the starting pI of the scFv and Fab of interest in the triple F format. That is, to determine which monomer to engineer or in which “direction” (e.g., more positive or more negative), the Fv sequences of the two target antigens are calculated and the decision is made accordingly. As is known in the art, different Fvs will have different starting pIs, and these starting pIs are utilized in this invention. Typically, as outlined herein, pIs are engineered so that the total pI difference between the individual monomers is at least about 0.1 logS, preferably 0.2 logS to 0.5 logS as outlined herein.

[0225] Furthermore, as those skilled in the art will understand and as outlined herein, in some embodiments, heterodimers can be separated from homodimers based on size. As shown in Figure 1, for example, several formats allow for the separation of heterodimers and homodimers based on size.

[0226] A. Heterodimerization variant

[0227] The present invention provides heterodimeric proteins comprising various forms of heterodimeric antibodies, said heterodimeric antibodies utilizing heterodimeric variants to allow heterodimer formation and / or purification away from homodimers.

[0228] Suitable sets of multiple heterodimerization tilt variants exist. These variants appear as “paired” “sets.” That is, one set of the pair is incorporated into the first monomer and the other set of the pair is incorporated into the second monomer. It should be noted that these sets do not necessarily exhibit “jackknife-cavitation” variants, where residues on one monomer correspond one-to-one with residues on the other monomer; that is, these pairs form an interaction between the two monomers that encourages heterodimer formation and inhibits homodimer formation, thereby allowing the percentage of spontaneously formed heterodimers under biological conditions to exceed 90%, rather than the expected 50% (25% homodimer A / A: 50% heterodimer A / B: 25% homodimer B / B).

[0229] B. Spatial Variations

[0230] In some embodiments, heterodimer formation can be promoted by adding spatial variants. That is, by changing the amino acids in each heavy chain, different heavy chains are more likely to associate to form heterodimer structures than homodimers with the same Fc amino acid sequence. Suitable spatial variants are included in the accompanying drawings.

[0231] One mechanism commonly referred to in the art as “bulge-cavitation” refers to the optional use of amino acid engineering that creates spatial effects that favor heterodimer formation and discourage homodimer formation; this is sometimes referred to as “bulge-cavitation” as described below: USSN 61 / 596,846; Ridgway et al., *Protein Engineering* 9(7):617(1996); Atwell et al., *Journal of Molecular Biology* 1997 270:26; US Patent No. 8,216,805, all of which are incorporated herein by reference in their entirety. The figures identify multiple pairs of “monomer A-monomer B” dependent on “bulge-cavitation”. In addition, as described by Merchant et al., *Nature Biotech* 16:677(1998), these “bulge-cavitation” mutations can combine with disulfide bonds to tilt the formation toward heterodimerization.

[0232] The additional mechanism used to generate heterodimers is sometimes referred to as “electrostatic shift,” as described by Gunasekaran et al., Journal of Biochemistry, 285(25): 19637(2010), which is incorporated herein by reference in its entirety. This is sometimes referred to herein as a “charge pair.” In this embodiment, electrostatics are used to tilt the formation toward heterodimerization. As those skilled in the art will appreciate, these charge pairs can also act on pI and thus on purification, and therefore can be considered pI variants in some cases. However, since these charge pairs are generated to force heterodimerization and are not used as purification tools, they are classified as “spatial variants.” These charge pairs include, but are not limited to, D221E / P228E / L368E paired with D221R / P228R / K409R (e.g., these are “monomer counterparts”) and C220E / P228E / 368E paired with C220R / E224R / P228R / K409R.

[0233] Additional monomer A and monomer B variants may be optionally and independently combined in any number with other variants such as the pI variants outlined herein or other spatial variants shown in Figure 37 of US 2012 / 0149876, the figures and illustrations of US2012 / 0149876, and SEQ ID NO, which are expressly incorporated herein by reference.

[0234] In some embodiments, the spatial variants outlined herein may optionally and independently be incorporated into one or two monomers with any pI variant (or other variants, such as Fc variants, FcRn variants, etc.) and may be included independently and optionally in or not included in the proteins of the present invention.

[0235] A list of suitable tilted variants is shown in Figure 3 and Figure 8 This illustrates specific uses of some pairs in some embodiments. Pairs including, but not limited to, the following are particularly used in some embodiments: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L; K370S: S364K / E357Q and T366S / L368A / Y407V: T366W (optionally including bridging disulfide T366S / L368A / Y407V / Y349C: T366W / S354C). In terms of nomenclature, the “S364K / E357Q:L368D / K370S” pair means that one of the monomers has a divariate group S364K / E357Q and the other has a divariate group L368D / K370S; as above, these “chain” pairs depend on the starting pI.

[0236] C. pI (isoelectric point) of heterodimers

[0237] Generally, as those skilled in the art will understand, there are two general categories of pI variants: those that increase the pI of a protein (basic change) and those that decrease the pI of a protein (acidic change). As described herein, all combinations of these variants can be accomplished: one monomer may be of the wild type or a variant that does not exhibit a pI significantly different from the wild type, and another may be more basic or more acidic. Alternatively, the individual monomers can be modified, one to be made more basic and the other more acidic.

[0238] Preferred combinations of pI variants are shown in Figure 4As outlined herein and illustrated in the figures, these changes are shown relative to IgG1, but all isotypes and isotype hybrids can be altered in this manner. R133E and R133Q can also be used in cases where the heavy chain constant domain originates from IgG2 to IgG4.

[0239] In one embodiment, for example in Figure 1A , Figure 1E , Figure 1F , Figure 1G , Figure 1H and Figure 1I In the format, a preferred combination of pI variants has a monomer (negative Fab side) comprising a 208D / 295E / 384D / 418E / 421D variant (N208D / Q295E / N384D / Q418E / N421D relative to human IgG1) and a second monomer (positive scFv side) comprising a positively charged scFv linker comprising (GKPGS)4 (SEQ ID NO: 37755). However, as those skilled in the art will understand, the first monomer comprises a CH1 domain containing position 208. Therefore, in constructs that do not contain a CH1 domain (e.g., for antibodies that utilize a CH1 domain not on one of the domains, e.g., in a dual scFv format or a “single-arm” format, such as... Figure 1B , Figure 1C or Figure 1D Among those formats described, the preferred negative pI variant Fc group contains the 295E / 384D / 418E / 421D variant (Q295E / N384D / Q418E / N421D in relation to human IgG1).

[0240] Therefore, in some embodiments, a monomer has from Figure 4 One set of substitutions and another monomer has a charged connector (in the form of a charged scFv, since that monomer includes scFv or charged structural domain connectors as indicated in the format, which may be selected from those depicted in FIG7).

[0241] 1. Isotype variant

[0242] In addition, many embodiments of the present invention rely on the “input” of pI amino acids at specific locations from one IgG isotype to another, thereby reducing or eliminating the possibility of introducing unwanted immunogenicity into variants.

[0243] Several of these are illustrated in Figure 21 of U.S. Publication 2014 / 0370013, which is incorporated herein by reference. Specifically, IgG1 is a commonly used isotype of therapeutic antibodies for various reasons, containing high-effects functions. However, the recurrent constant region of IgG1 has a higher pI than IgG2 (8.10 vs. 7.31). By introducing IgG2 at specific positions into the IgG1 backbone, the resulting monomer exhibits a lower (or higher) pI and, additionally, a longer serum half-life. For example, IgG1 has glycine at position 137 (pI 5.97), and IgG2 has glutamate (pI 3.22); the introduction of glutamate will affect the pI of the resulting protein. As described below, multiple amino acid substitutions are typically required to significantly affect the pI of variant antibodies. However, it should be noted that, as discussed below, even changes in the IgG2 molecule can allow for an increase in serum half-life.

[0244] In other embodiments, different types of amino acids were modified to reduce the total charge state of the resulting protein (e.g., by replacing higher pI amino acids with lower pI amino acids) or to allow for structural regulation for stability, as further described below.

[0245] Furthermore, significant variations in the pI of each monomer of the heterodimer can be observed by engineering heavy and light constant structural domains. As discussed in this paper, ensuring a pI difference of at least 0.5 between the two monomers allows for separation by ion exchange chromatography, isoelectric focusing, or other methods sensitive to the isoelectric point.

[0246] D. Calculate pI

[0247] The pI of each monomer can depend on the pI of the variable heavy chain constant domain and the pI of the total monomer, which comprises the variable heavy chain constant domain and the fusion mating body. Therefore, in some embodiments, the pI variation is calculated based on the variable heavy chain constant domain using the graph in Figure 19 of U.S. Publication 2014 / 0370013. As discussed herein, which monomer to engineer is typically determined by the Fv and the inherent pI of the scaffold region. Alternatively, the pIs of the individual monomers can be compared.

[0248] E. also endowed the pI variant with better in vivo binding of FcRn.

[0249] When the pI variant reduces the pI of the monomer, it can have the additional benefit of improving serum fixation in vivo.

[0250] Although still under investigation, it is believed that the Fc region has a longer in vivo half-life because the binding of FcRn in the endosome chelates Fc at pH 6 (Ghetie and Ward, 1997, Immunol Today, 18(12): 592-598, all cited in whole). The endosome septum then allows Fc to circulate to the cell surface. Once the septum opens to the extracellular space, a higher pH of about 7.4 induces the release of Fc back into the bloodstream. In mice, Dall'Acqua et al. showed that Fc mutants with increased FcRn binding at pH 6 and pH 7.4 actually had reduced serum concentrations and the same half-life as wild-type Fc (Dall'Acqua et al., 2002, J. Immunol., 169: 5171-5180, all cited in whole).

[0251] It is believed that the increased affinity of Fc to Fc at pH 7.4 prevents the release of Fc back into the bloodstream. Therefore, Fc mutations that increase the in vivo half-life of Fc would ideally increase FcRn binding at lower pH levels while still allowing Fc release at higher pH levels. The amino acid histidine alters its charge state across a pH range of 6.0 to 7.4. Therefore, the presence of His residues at key positions in the Fc / FcRn complex is not surprising.

[0252] Recently, it has been proposed that antibodies with variable regions having lower isoelectric points can also have longer serum half-lives (Igawa et al., 2010 PEDS. 23(5): 385-392, all incorporated herein by reference). However, the mechanism remains poorly understood. Moreover, the variable regions vary from antibody to antibody. Constant regions with lower pI and longer half-lives would provide a more modular approach to improving the pharmacokinetic properties of antibodies, as described herein.

[0253] F. Additional Fc variants for extra functionality

[0254] In addition to pI amino acid variants, there are many useful Fc amino acid modifications that can be made for various reasons, including but not limited to altering the binding to one or more FcγR receptors, altering the binding to FcRn receptors, etc.

[0255] Therefore, the proteins of the present invention may contain amino acid modifications, including heterodimerization variants as outlined herein, which include pI variants and spatial variants. Each set of variants may be included independently and optionally not in any particular heterodimer protein.

[0256] G.FcγR variant

[0257] Therefore, there are several useful Fc substitutions that can be made to alter the binding to one or more of the FcγR receptors. Substitutions that result in both increased and decreased binding can be useful. For example, increased binding to FcγRIIIa is known to result in increased ADCC (antibody-dependent cell-mediated cytotoxicity; a cell-mediated reaction in which nonspecific toxic cells expressing FcγR recognize a binding antibody on a target cell and subsequently cause lysis of the target cell). Similarly, decreased binding to FcγRIIb (inhibitory receptor) can also be beneficial in some cases. The amino acid substitutions used in this invention include those listed in USSN 11 / 124,620 (especially Figure 41), 11 / 174,287, 11 / 396,495, and 11 / 538,406, all of which are expressly incorporated herein by reference in their entirety and particularly for the variants disclosed herein. The specific variants used include, but are not limited to, 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D, 332E / 330L, 243A, 243L, 264A, 264V, and 299T.

[0258] Additionally, as specifically disclosed by reference in USSN 12 / 341,769, which is incorporated herein by reference in its entirety, there are additional Fc substitutions for increased binding to FcRn and increased serum half-life, including, but not limited to, 434S, 434A, 428L, 308F, 2591, 428L / 434S, 259I / 308F, 436I / 428L, 4361 or V / 434S, 436V / 428L and 259I / 308F / 428L.

[0259] H. Ablation variant

[0260] Similarly, another category of functional variants is the “FcγR ablation variant” or “Fc knockout (FcKO or KO) variant.” In these embodiments, for some therapeutic applications, it is desirable to reduce or eliminate the normal binding of the Fc domain to one or all of the Fcγ receptors (e.g., FcγR1, FcγRIIa, FcγRIIb, FcγRIIIa, etc.) to avoid additional mechanisms of action. That is, for example, in many embodiments, especially when using bispecific checkpoint antibodies, it is desirable to ablate FcγRIIIa binding to eliminate or significantly reduce ADCC activity such that one of the Fc domains includes one or more Fcγ receptor ablation variants. Figure 5These ablation variants are described herein, and each may be included independently and optionally, with preferred aspects utilizing ablation variants selected from the group consisting of: G236R / L328R, E233P / L234V / L235A / G236del / S239K, E233P / L234V / L235A / G236del / S267K, E233P / L234V / L235A / G236del / S239K / A327G, E233P / L234V / L235A / G236del / S267K / A327G, and E233P / L234V / L235A / G236del. It should be noted that the ablation variants cited herein ablate FcyR bindings but generally do not ablate FcRn bindings.

[0261] As is known in the art, the Fc domain of human IgG1 has the highest binding affinity to the Fcγ receptor, and therefore, when the constant domain (or Fc domain) in the backbone of a heterodimeric antibody is IgG1, an ablation variant can be used. For example, alternatively or in addition to an ablation variant in the IgG1 background, a mutation at glycosylation position 297 (typically mutated to A or S) can significantly ablate binding to FcγRIIIa. Human IgG2 and IgG4 have naturally reduced binding affinity to the Fcγ receptor, and therefore those backbones can be used with or without an ablation variant.

[0262] I. Combination of heterodimer variants and Fc variants

[0263] As those skilled in the art will appreciate, all the listed heterodimerization variants (including tilt variants and / or pI variants) can be combined in any optional and independent manner, provided that their retainers are “chain-like” or “monomer-separated”. Furthermore, all these variants can be combined into any heterodimerization format.

[0264] In the case of pI variants, although particularly useful examples are shown in the figures, other combinations can be generated by following the basic rule of changing the pI difference between the two monomers to facilitate purification.

[0265] In addition, as generally outlined in this article, any of the heterodimerization variants, tilt variants, and pI variants are also independent of and optionally combined with the Fc ablation variant, Fc variant, and FcRn variant.

[0266] V. Useful Format of the Invention

[0267] As those skilled in the art will appreciate and as discussed more fully below, the bispecific heterodimeric antibody of the present invention can take various configurations, as generally depicted in Figure 1. Some figures depict a “single-end” configuration, wherein one “arm” of the molecule has one type of specificity and the other “arm” has a different type of specificity. Other figures depict a “double-end” configuration, wherein the “top” of the molecule has at least one type of specificity and the “bottom” of the molecule has one or more different types of specificity. Therefore, the present invention relates to novel immunoglobulin compositions that co-conjugate different first and second antigens.

[0268] As those skilled in the art will appreciate, the heterodimeric format of the present invention can have different valences and can be bispecific. That is, the heterodimeric antibody of the present invention can be divalent and bispecific, wherein one checkpoint target is bound by one ABD and the other checkpoint target is bound by a second ABD. The heterodimeric antibody can also be trivalent and bispecific, wherein the first antibody is bound by two ABDs and the second antibody is bound by a second ABD.

[0269] A. Bottle opener format

[0270] A heterodimer scaffold specifically used in this invention is as follows: Figure 1A The “triple F” or “bottle opener” scaffold format is shown. In this embodiment, one heavy chain of the antibody contains a single-chain Fv (“scFv”, as defined below) and the other heavy chain is a “conventional” Fab format comprising a variable heavy chain and a light chain. This structure is sometimes referred to herein as a “triple F” format (scFv-Fab-Fc) or a “bottle opener” format due to its general visual similarity to a bottle opener (see [link to documentation]). Figure 1A The two chains are brought together using amino acid variants that promote the formation of heterodimer antibodies, as more fully described below, in constant regions (e.g., Fc domain, CH1 domain, and / or hinge region).

[0271] The "triple F" format of this invention offers several distinct advantages. As is known in the art, antibody analogs relying on two scFv constructs often suffer from stability and polymerization problems, which can be mitigated in this invention by adding a "conventional" heavy chain and light chain pairing. Furthermore, unlike formats relying on two heavy chains and two light chains, there is no problem of incorrect heavy chain and light chain pairing (e.g., heavy chain 1 paired with light chain 2, etc.).

[0272] Many of the embodiments outlined herein typically rely on a bottle opener format comprising a first monomer including an scFv, the scFv comprising a variable heavy structural domain and a variable light structural domain covalently linked using scFv connectors (in many, but not all, charged), wherein the scFv is often covalently linked to the N-terminus of the first Fc structural domain via structural domain connectors (as outlined herein, these structural domain connectors may be uncharged or charged and may be exogenous or endogenous (e.g., all or part of a natural hinged structural domain)). The second monomer of the bottle opener format is a heavy chain, and the composition further includes a light chain.

[0273] Additionally, the Fc domain in the opener format typically includes tilted variants (e.g., as shown in Figure 3 and...). Figure 8 The group of amino acid substitutions shown includes particularly useful tilt variants selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L, K370S: S364K / E357Q, T366S / L368A / Y407V: T366W and T366S / L368A / Y407V / Y349C: T366W / S354C), and optionally ablation variants (containing...). Figure 5 Those shown), optional zone-charged scFv linkers (including those shown in Figure 7), and heavy chains including pI variants (including Figure 4 Those shown).

[0274] In some embodiments, the opener format includes tilt variants, pI variants, and ablation variants. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilted variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, and an Fv that binds to a checkpoint receptor as outlined herein; b) a second monomer (“Fab monomer”) comprising a tilted variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, an ablation variant E233P / L234V / L235A / G236del / S267K, and a variable restructure domain, the variable restructure domain and the variable light structure domain constituting an Fv that binds to a second checkpoint receptor as outlined herein; and c) a light chain. In this particular embodiment, suitable monomer Fv pairs include (Fab is listed first, scFv is listed second) PD-1 and CTLA-4, CTLA-4 and PD-1, PD-1 and TIM-3, TIM-3 and PD-1, PD-1 and LAG-3, LAG-3 X PD1, PD-1 and TIGIT, TIGIT and PD-1, PD-1 and BTLA, BTLA and PD-1, CTLA-4 and TIM-3, TIM-3 and CTLA-4, CTLA-4 and LAG-3, LAG-3 and CTLA-4, CTLA-4 and TIGIT, TIGIT and CTLA-4, CTLA-4 and BTLA, BTLA and CTLA-4, TIM-3 and LAG-3, LAG-3 and TIM-3, TIM-3 and TIGIT, TIGIT and TIM-3, TIM-3 and BTLA, BTLA and TIM-3, LAG-3 and TIGIT, TIGIT and LAG-3, LAG-3 and BTLA, BTLA and LAG-3, BTLA and TIGIT, and TIGIT and BTLA. In this particular embodiment, the bottle opener having these variants has an scFv side including ABD 1G6_L1.194_H1.279 in combination with PD-1 for a special purpose. In this particular embodiment, the bottle opener having these variants has an scFv side including [CTLA-4]_H3.23_L0.129 ABD in combination with CTLA-4 for a special purpose.

[0275] In particular, the following are used in some embodiments, especially in the bottle opener format: CTLA-4 X PD-1, LAG-3 X PD-1, BTLA X PD-1, TIM-3 X PD-1, and LAG-3 X CTLA-4.

[0276] These combined ABD sequences can be as disclosed in the sequence listing or as shown in Figures 9 to 13, and are in the position shown in Figure 39 and Figure 40 Any combination shown.

[0277] In some embodiments, the opener format includes tilt variants, pI variants, ablation variants, and FcRn variants. Thus, some embodiments include an opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilt variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, an FcRn variant M428L / N434S, and an Fv bound to a checkpoint inhibitor as outlined herein; b) a second The monomer (“Fab monomer”) comprises tilt variant L368D / K370S, pI variant N208D / Q295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a variable restructure domain, which, together with a variable light structure domain, constitutes an Fv that binds to a second checkpoint inhibitor as outlined herein; and c) a light chain. In this particular embodiment, suitable Fv pairs comprise (Fab is listed first, scFv is listed second) PD-1 and CTLA-4, CTLA-4 and PD-1, PD-1 and TIM-3, TIM-3 and PD-1, PD-1 and LAG-3, LAG-3 X PD1, PD-1 and TIGIT, TIGIT and PD-1, PD-1 and BTLA, BTLA and PD-1, CTLA-4 and TIM-3, TIM-3 and CTLA-4, CTLA-4 and LAG-3, LAG-3 and CTLA-4, CTLA-4 and TIGIT, TIGIT and CTLA-4, CTLA-4 and BTLA, BTLA and CTLA-4, TM-3 and LAG-3, LAG-3 and TIM-3, TIM-3 and TIGIT, TIGIT and TIM-3, TIM-3 and BTLA, BTLA and TIM-3, LAG-3 and TIGIT, TIGIT and LAG-3, LAG-3 and BTLA, BTLA and LAG-3, BTLA and TIGIT, and TIGIT and BTLA. In this particular embodiment, the bottle opener having these variants has an scFv side including ABD 1G6_L1.194_H1.279 in combination with PD-1 for a special purpose. In this particular embodiment, the bottle opener having these variants has an scFv side including [CTLA-4]_H3.23_L0.129 ABD in combination with CTLA-4 for a special purpose.

[0278] In particular, the following are used in some embodiments, especially in the bottle opener format: CTLA-4 X PD-1, LAG-3 X PD-1, BTLA X PD-1, TIM-3 X PD-1, and LAG-3 X CTLA-4.

[0279] Specifically, Figure 37 illustrates some opener “main chain” sequences lacking the Fv sequences that can be used in this invention. That is, Fv sequences from any combination of the scFv and Fab portions can be used: PD-1 and CTLA-4, PD-1 and TIM-3, PD-1 and LAG-3, PD-1 and TIGIT, PD-1 and BTLA, CTLA-4 and TIM-3, CTLA-4 and LAG-3, CTLA-4 and TIGIT, CTLA-4 and BTLA, TIM-3 and LAG-3, TIM-3 and TIGIT, TIM-3 and BTLA, LAG-3 and TIGIT, LAG-3 and BTLA, and TIGIT and BTLA. The sequences can be any of those sequences disclosed herein in the sequence listing and / or Figures 9 through 13.

[0280] For the bottle opener main chain 1 of Figure 37, the specific Fv combinations used in this invention include PD-1 and CTLA-4, PD-1 and TIM-3, PD-1 and LAG-3, PD-1 and TIGIT, PD-1 and BTLA, CTLA-4 and TIM-3, CTLA-4 and LAG-3, CTLA-4 and TIGIT, CTLA-4 and BTLA, TIM-3 and LAG-3, TIM-3 and TIGIT, TIM-3 and BTLA, LAG-3 and TIGIT, LAG-3 and BTLA, and TIGIT and BTLA. The sequences can be any of those sequences disclosed herein in the sequence listing and / or Figures 9 through 13.

[0281] For the main chain 1 of the bottle opener in Figure 37, the specific Fv combinations used in this invention include CTLA-4(Fab)X PD-1(scFv), PD-1(Fab)XCTLA-4(scFv), LAG-3(Fab)X PD-1(scFv), BILA(Fab)X PD-1(scFv) and LAG-3(Fab)X CTLA-4(scFv).

[0282] For the bottle opener main chain 1 of Figure 37 (optionally including the 428L / 434S variant), the specific ABDs associated with human PD-1 include, but are not limited to, 1G6_H1.279_L1.194, 1G6_H1.280_L1.224, 1G6_L1.194_H1.279, 1G6_L1.210_H1.288 and 2E9H1L1, as well as those listed in SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146.

[0283] For the bottle opener main chain 1 of Figure 37 (optionally including the 428L / 434S variant), the specific ABD of combined human CTLA-4 includes, but is not limited to: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L0.22; [CTLA-4 ]_H2_L0;[CTLA-4]_H3.21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA-4]_H3.23_ L0.129; [CTLA-4]_H3.23_L0.132; [CTLA-4]_H3.25_L0.124; [CTLA-4]_H3.25_L0.129; [CTLA-4]_H3.25_L0.132; [CTLA-4]_H3.4_L0. 118;[CTLA-4]_H3.4_L0.119;[CTLA-4]_H3.4_L0.12;[CTLA-4]_H3.4_L0.121;[CTLA-4]_H3.4_L0.122;[CTLA-4]_H3.4_L0.123;[CT LA-4]_H3.4_L0.124;[CTLA-4]_H3.4_L0.125;[CTLA-4]_H3.4_L0.126;[CTLA-4]_H3.4_L0.127;[CTLA-4]_H3.4_L0.128;[CTLA-4]_H 3.4_L0.129;[CTLA-4]_H3.4_L0.130;[CTLA-4]_H3.4_L0.131;[CTLA-4]_H3.4_L0.132;[CTLA-4]_H3.5_L2.1;[CTLA-4]_H3.5_L2.2 ; [CTLA-4]_H3.5_L2.3; [CTLA-4]_H3_L0; [CTLA-4]_H3_L0.22; [CTLA-4]_H3_L0.44; [CTLA-4]_H3_L0.67 and [CTLA-4]_H3_L0.74; and SEQ. Those listed in SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416.

[0284] For the bottle opener main chain 1 of Figure 37 (optionally including the 428L / 434S variant), the specific ABDs of human LAG-3 include, but are not limited to: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.93; 2A11_ H1_L2.97;2A11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_L1. 11;7G8_H3.23_L1.11;7G8_H3.28L1;7G8_H3.28_L1.11;7G8_H3.28_L1.13; 7G8_H3.30_L1.34; Those listed in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002.

[0285] For the bottle opener main chain 1 of Figure 37 (optionally including 428L / 434S variants), the specific ABDs of combined human BTLA include, but are not limited to: 9C6_H0L0; 9C6_H1.1_L1; and 9C6_H1.11_L1; as well as those listed in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738.

[0286] For the main chain 1 of the bottle opener in Figure 37 (optionally including the 428L / 434S variant), the specific ABDs of combined human TIM-3 include, but are not limited to: 1D10_H0L0; 1D12_H0L0; 3H3_H1_L2.1; 6C8_H0L0; 6D9_H0_1D12_L0; 7A9_H0L0; 7B11_H0L0; 7B11var_H0L0 and 7C2_H0L0; and those listed in SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698 and SEQ ID NO: 36347 to 36706.

[0287] The following section outlines specific examples of bottle openers.

[0288] B.mAb-Fv format

[0289] A heterodimer scaffold specifically used in this invention is as follows: Figure 1H The mAb-Fv format is shown. In this embodiment, the format relies on the connection of the C-terminus of an “additional” variable heavy domain to one monomer and the C-terminus of an “additional” variable light domain to another monomer to form a third antigen-binding domain, wherein the Fab portions of the two monomers bind to one checkpoint target and the “additional” scFv domain binds to a different checkpoint target.

[0290] In this embodiment, the first monomer includes a first heavy chain comprising a first variable heavy structure domain and a first constant heavy structure domain. The first constant heavy structure domain includes a first Fc structure domain. The first variable light structure domain is covalently connected to the C-terminus of the first Fc structure domain using a domain connector (vh1-CH1-hinge-CH2-CH3-[optional connector]-vl2). The second monomer includes a second variable heavy structure domain and a third variable heavy structure domain. The second variable heavy structure domain belongs to a second constant heavy structure domain including the second Fc structure domain. The third variable heavy structure domain is covalently connected to the C-terminus of the second Fc structure domain using a domain connector (vh1-CH1-hinge-CH2-CH3-[optional connector]-vh2). The two C-terminus-connected variable structure domains constitute an scFv. This embodiment further utilizes a light chain comprising a variable light structure domain and a constant light structure domain, which is associated with the heavy chain to form two identical Fabs. For many embodiments herein, as desired and described herein, these constructs include tilt variants, pI variants, ablation variants, additional Fc variants, etc. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) PD-1 and CTLA-4, CTLA-4 and PD-1, PD-1 and TIM-3, TIM-3 and PD-1, PD-1 and LAG-3, LAG-3 X PD1, PD-1 and TIGIT, TIGIT and PD-1, PD-1 and BTLA, BTLA and PD-1, CTLA-4 and TIM-3, TIM-3 and CTLA-4, CTLA-4 and LAG-3, LAG-3 and CTLA-4, CTLA-4 and TIGIT, TIGIT and CTLA-4, CTLA-4 and BTLA, BTLA and CTLA-4, TIM-3 and LAG-3, LAG-3 and TIM-3, TIM-3 and TIGIT, TIGIT and TIM-3, TIM-3 and BTLA, BTLA and TIM-3, LAG-3 and TIGIT, TIGIT and LAG-3, LAG-3 and BTLA, BTLA and LAG-3, BTLA and TIGIT, and TIGIT and BTLA.

[0291] These combined ABD sequences can be as disclosed in the sequence listing or as shown in Figures 9 to 13, and are in the position shown in Figure 39 and Figure 40 Any combination shown.

[0292] Additionally, the Fc domain in the mAb-Fv format includes tilted variants (e.g., as shown in Figure 3 and...). Figure 8The group of amino acid substitutions shown includes particularly useful tilt variants selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L, K370S: S364K / E357Q, T366S / L368A / Y407V: T366W, and T366S / L368A / Y407V / Y349C: T366W / S354C), and optionally ablation variants (including...). Figure 5 Those shown), optional zone-charged scFv linkers (including those shown in Figure 7), and heavy chains including pI variants (including Figure 4 (Those shown).

[0293] In some embodiments, the mAb-Fv format includes a tilted variant, a pI variant, and an ablation variant. Therefore, some embodiments include an opener format comprising: a) a first monomer comprising the tilted variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and a first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising the tilted variant L368D / K370S and the pI variant N208D / Q. 295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable heavy structure domain and a second variable light chain, the first variable heavy structure domain and the first variable light structure domain constituting an Fv that binds to a first checkpoint inhibitor as outlined herein, the second variable light chain together with the second variable heavy chain forming an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. Particularly used in some embodiments under this format are (Fab-scFv sequence) CTLA-4X PD-1, LAG-3 X PD-1, BTLA X PD-1, and LAG-3 X CTLA-4.

[0294] In some embodiments, the mAb-Fv format includes tilt variants, pI variants, ablation variants, and FcRn variants. Therefore, some embodiments include an opener format comprising: a) a first monomer comprising tilt variant S364K / E357Q, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and a first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising tilt variant L368D / K370S and pI variant N208D / Q. 295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a first variable heavy structure domain and a second variable light chain, wherein the first variable heavy structure domain and the first variable light structure domain constitute an Fv that binds to a first checkpoint inhibitor as outlined herein, and the second variable light chain together with the second variable heavy chain forms an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. Particularly used in some embodiments under this format are (Fab-scFv sequence) CTLA-4 x PD-1, LAG-3 x PD-1, BTLA x PD-1, and LAG-3 x CTLA-4.

[0295] For mAb-Fv sequences similar to the mAb-scFv main chain 1 of Figure 38 (optionally containing M428L / N434S), the specific ABDs binding to human PD-1 include, but are not limited to: 1G6_H1.279_L1.194, 1G6_H1.280_L1.224, 1G6_L1.194_H1.279, 1G6_L1.210_H1.288 and 2E9_H1L1, as well as those listed in SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146.

[0296] For mAb-Fv sequences similar to the mAb-scFv main chain 1 in Figure 38 (optionally containing M428L / N434S), specific ABDs associated with human CTLA-4 include, but are not limited to: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L 0.22;[CTLA-4]_H2_L0;[CTLA-4]_H3.21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA -4]_H3.23_L0.129; [CTLA-4]_H3.23_L0.132; [CTLA-4]_H3.25_L0.124: [CTLA-4]_H3.25_L0.129; [CTLA-4]_H3.25_L0.132; [CILA-4]_ H3.4_L0.118; [CTLA-4]_H3.4_L0.119; [CTLA-4]_H3.4_L0.12; [CTLA-4]_H3.4_L0.121; [CTLA-4]_H3.4_L0.122; [CTLA-4]_H3.4_L0.1 23;[CTLA-4]_H3.4_L0.124;[CTLA-4]_H3.4_L0.125;[CTLA-4]_H3.4_L0.126;[CTLA-4]_H3.4_L0.127;[CTLA-4]_H3.4_L0.128;[CTLA- 4]_H3.4_L0.129;[CTLA-4]_H3.4_L0.130;[CTLA-4]_H3.4_L0.131;[CTLA-4]_H3.4_L0.132;[CTLA-4]_H3.5_L2.1;[CTLA-4]_H3.5_L2 .2; [CTLA-4]_H3.5_L2.3; [CTLA-4]_H3_L0; [CTLA-4]_H3_L0.22; [CTLA-4]_H3_L0.44; [CTLA-4]_H3_L0.67 and [CTLA-4]_H3_L0.74; and SEQ. Those listed in SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416.

[0297] For mAb-Fv sequences similar to the mAb-scFv main chain 1 in Figure 38 (optionally containing M428L / N434S), the specific ABDs associated with human LAG-3 include, but are not limited to: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.9 3;2A11_H1_L2.97;2A11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_ and SEQ. Those listed in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002.

[0298] For mAb-Fv sequences similar to the mAb-scFv main chain 1 of Figure 38 (optionally containing M428L / N434S), the specific ABDs binding to human BTLA include, but are not limited to: 9C6_H0L0; 9C6_H1.1_L1; and 9C6_H1.11_L1; as well as those listed in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738.

[0299] For mAb-Fv sequences similar to the mAb-scFv main chain 1 of Figure 38 (optionally containing M428L / N434S), the specific ABDs of human TIM-3 include, but are not limited to: 1D10_H0L0; 1D12_H0L0; 3H3_H1_L2.1; 6C8_H0L0; 6D9_H0_1D12_L0; 7A9_H0L0; 7B11_H0L0; 7B11var_H0L0 and 7C2_H0L0; and those listed in SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698 and SEQ ID NO: 36347 to 36706.

[0300] C.mAb-scFv

[0301] A heterodimer scaffold specifically used in this invention is as follows: Figure 1I The mAb-scFv format is shown. In this embodiment, the format depends on the connection of the C-terminus of the scFv to one of the monomers to form a third antigen-binding domain, wherein the Fab portions of the two monomers bind to one checkpoint target and the “additional” scFv domain binds to a different checkpoint target.

[0302] In this embodiment, the first monomer includes a first heavy chain (comprising a variable heavy structure domain and a constant structure domain), wherein the C-terminal covalently linked scFv includes a scFv variable light structure domain, a scFv connector, and a scFv variable heavy structure domain in either orientation (vh1-CH1-hinge-CH2-CH3-[optional linker]-vh2-scFv linker-vl2 or vhl-CH1-hinge-CH2-CH3-[optional linker]-vl2-scFv linker-vh2). This embodiment further utilizes a shared light chain comprising a variable light structure domain and a constant light structure domain, said shared light chain being associated with the heavy chain to form two identical Fabs that bind to one of the target antigens. For many embodiments herein, as desired and described herein, these constructs comprise tilted variants, pI variants, ablation variants, additional Fc variants, etc. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) PD-1 and CTLA-4, CTLA-4 and PD-1, PD-1 and TIM-3, TIM-3 and PD-1, PD-1 and LAG-3, LAG-3 X PD1, PD-1 and TIGIT, TIGIT and PD-1, PD-1 and BTLA, BTLA and PD-1, CTLA-4 and TIM-3, TIM-3 and CTLA-4, CTLA-4 and LAG-3, LAG-3 and CTLA-4, CTLA-4 and TIGIT, TIGIT and CTLA-4, CTLA-4 and BTLA, BTLA and CTLA-4, TIM-3 and LAG-3, LAG-3 and TIM-3, TIM-3 and TIGIT, TIGIT and TIM-3, TIM-3 and BTLA, BTLA and TIM-3, LAG-3 and TIGIT, TIGIT and LAG-3, LAG-3 and BTLA, BTLA and LAG-3, BTLA and TIGIT, and TIGIT and BTLA.

[0303] These combined ABD sequences can be as disclosed in the sequence listing or as shown in Figures 9 to 13, and are in the position shown in Figure 39 and Figure 40 Any combination shown.

[0304] In addition, the Fc domain of the mAb-scFv format typically includes tilted variants (e.g., as shown in Figure 3 and...). Figure 8The group of amino acid substitutions shown includes particularly useful tilted variants selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L, K370S: S364K / E357Q, T366S / L368A / Y407V: T366W, and T366S / L368A / Y407V / Y349C: T366W / S354C), and optionally ablation variants (including...). Figure 5 Those shown), optional zone-charged scFv linkers (including those shown in Figure 7), and heavy chains including pI variants (including Figure 4 Those shown).

[0305] In some embodiments, the mAb-scFv format includes tilt variants, pI variants, and ablation variants. Therefore, some embodiments include an opener format comprising: a) a first monomer comprising the tilt variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and a first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising the tilt variant L368D / K370S and the pI variant N208D / Q. 295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable heavy structure domain and a second variable light chain, the first variable heavy structure domain and the first variable light structure domain constituting an Fv that binds to a first checkpoint inhibitor as outlined herein, the second variable light chain together with the second variable heavy chain forming an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. Particularly used in some embodiments under this format are (Fab-scFv order) CTLA-4 x PD-1, LAG-3 x PD-1, BTLA x PID-1, and LAG-3 x CTLA-4.

[0306] In some embodiments, the mAb-scFv format includes tilt variants, pI variants, ablation variants, and FcRn variants. Therefore, some embodiments include an opener format comprising: a) a first monomer comprising tilt variant S364K / E357Q, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and a first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising tilt variant L368D / K370S and pI variant N208D / Q. 295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a first variable heavy structure domain and a second variable light chain, wherein the first variable heavy structure domain and the first variable light structure domain constitute an Fv that binds to a first checkpoint inhibitor as outlined herein, and the second variable light chain together with the second variable heavy chain form an Fv(ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. In the mAb-scFv format, specific Fv combinations used in this invention include CTLA-4(Fab)X PD-1(scFv), PD-1(Fab)X CTLA-4(scFv), LAG-3(Fab)X PD-1(scFv), BTLA(Fab)X PD-1(scFv), and LAG-3(Fab)X CTLA-4(scFv).

[0307] In the mAb-scFv main chain 1 of Figure 38 (optionally including M428L / N434S), the specific ABD binding human PD-1 includes, but is not limited to: 1G6_H1.279_L1.194, 1G6_H1.280_L1.224, 1G6_L1.194_H1.279, 1G6_L1.210_H1.288 and 2E9_H1L1, and those listed in SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146.

[0308] In the mAb-scFv main chain 1 of Figure 38 (optionally containing M428L / N434S), the specific ABDs binding human CTLA-4 include, but are not limited to: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L0.22; [CTL A-4]_H2_L0;[CTLA-4]_H3.21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA-4]_H3.2 3_L0.129;[CTLA-4]_H3.23_L0.132;[CTLA-4]_H3.25_L0.124;[CTLA-4]_H3.25_L0.129;[CTLA-4]_H3.25_L0.132;[CTLA-4]_H3.4_L 0.118;[CTLA-4]_H3.4_L0.119;[CTLA-4]_H3.4_L0.12;[CTLA-4]_H3.4_L0.121;[CTLA-4]_H3.4_L0.122;[CTLA-4]_H3.4_L0.123;[C TLA-4]_H3.4_L0.124;[CTLA-4]_H3.4_L0.125;[CTLA-4]_H3.4_L0.126;[CTLA-4]_H3.4_L0.127;[CTLA-4]_H3.4_L0.128;[CTLA-4]_ H3.4_L0.129; [CTLA-4]_H3.4_L0.130; [CTLA-4]_H3.4_L0.131; [CTLA-4]_H3.4_L0.132; [CTLA-4]_H3.5_L2.1; [CTLA-4]_H3.5_L2.2 ; [CTLA-4]_H3.5_L2.3; [CTLA-4]_H3_L0; [CTLA-4]_H3_L0.22; [CTLA-4]_H3_L0.44; [CTLA-4]_H3_L0.67 and [CTLA-4]_H3_L0.74; and SEQ. Those listed in SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416.

[0309] In the mAb-scFv main chain 1 of Figure 38 (optionally containing M428L / N434S), the specific ABDs binding to human LAG-3 include, but are not limited to: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.93; 2A1 1_H1_L2.97;2A11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_L1 .11;7G8_H3.23_L1.11;7G8_H3.28_L1;7G8_H3.28_L1.11;7G8_H3.28_L1.13; 7G8_H3.30_L1.34; Those listed in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002.

[0310] In the mAb-scFv main chain 1 of Figure 38 (optionally containing M428L / N434S), the specific ABDs binding human BTLA include, but are not limited to: 9C6_H0L0; 9C6_H1.1_L1; and 9C6_H1.11_L1; as well as those listed in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738.

[0311] In the mAb-scFv main chain 1 of Figure 38 (optionally including M428L / N434S), the specific ABDs binding human TIM-3 include, but are not limited to: 1D10_H0L0; 1D12_H0L0; 3H3_H1_L2.1; 6C8_H0L0; 6D9_H0_1D12_L0; 7A9_H0L0; 7B11_H0L0; 7B11var_H0L0 and 7C2_H0L0; and those listed in SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698 and SEQ ID NO: 36347 to 36706.

[0312] D. Central scFv

[0313] A heterodimer scaffold specifically used in this invention is as follows: Figure 1F The central-scFv format is shown. In this embodiment, the format depends on the use of an inserted scFv domain to form a third antigen-binding domain, wherein the Fab portions of the two monomers bind to one checkpoint target, and the "additional" scFv domain binds to another checkpoint target. The scFv domain is inserted between the Fc domain and the CH1-Fv region of one of the monomers to provide the third antigen-binding domain.

[0314] In this embodiment, a monomer includes a first heavy chain comprising a first variable heavy structural domain, a CH1 structural domain (and optional hinge), and an Fc structural domain, wherein the scFv includes a scFv variable light structural domain, an scFv connector, and a scFv variable heavy structural domain. The scFv is covalently connected between the C-terminus of the CH1 structural domain of the heavy constant structural domain and the N-terminus of the first Fc structural domain using optional structural domain connectors (vhl-CH1-[optional connector]-vh2-scFv connector-vl2-[optional connector with hinge]-CH2-CH3 or the opposite orientation of the scFv, vhl-CH1-[optional connector]-vl2-scFv connector-vh2-[optional connector with hinge]-CH2-CH3). The other monomers are standard Fab sides. This embodiment further utilizes a shared light chain comprising a variable light structural domain and a constant light structural domain, which is associated with the heavy chain to form two identical Fabs binding checkpoint inhibitors. For many embodiments herein, as desired and described herein, these constructs include tilt variants, pI variants, ablation variants, additional Fc variants, etc. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) PD-1 and CTLA-4, CTLA-4 and PD-1, PD-1 and TIM-3, TIM-3 and PD-1, PD-1 and LAG-3, LAG-3 X PD1, PD-1 and TIGIT, TIGIT and PD-1, PD-1 and BTLA, BTLA and PD-1, CTLA-4 and TIM-3, TIM-3 and CTLA-4, CTLA-4 and LAG-3, LAG-3 and CTLA-4, CTLA-4 and TIGIT, TIGIT and CTLA-4, CTLA-4 and BTLA, BTLA and CTLA-4, TIM-3 and LAG-3, LAG-3 and TIM-3, TIM-3 and TIGIT, TIGIT and TIM-3, TIM-3 and BTLA, BTLA and TIM-3, LAG-3 and TIGIT, TIGIT and LAG-3, LAG-3 and BTLA, BTLA and LAG-3, BTLA and TIGIT, and TIGIT and BTLA.

[0315] These combined ABD sequences can be as disclosed in the sequence listing or as shown in Figures 9 to 13, and are in the position shown in Figure 39 and Figure 40 Any combination shown.

[0316] Furthermore, the Fc structure domain of the central scFv format typically includes tilted variants (e.g., as shown in Figure 3 and...). Figure 8The group of amino acid substitutions shown includes particularly useful tilt variants selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L, K370S: S364K / E357Q, T366S / L368A / Y407V: T366W, and T366S / L368A / Y407V / Y349C: T366W / S354C), and optionally ablation variants (including such as...). Figure 5 The variants shown), optional band-charged scFv linkers (including those shown in Figure 7), and heavy chains including pI variants (including those shown in Figure 7) Figure 4 (Those variants shown).

[0317] In some embodiments, the central scFv format includes a tilted variant, a pI variant, and an ablation variant. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer comprising the tilted variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and the first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising the tilted variant L368D / K370S and the pI variant N208D / Q. The Fv consists of 295E / N384D / Q418E / N421D, ablation variants E233P / L234V / L235A / G236del / S267K, and a first variable heavy structure domain and a second variable light chain, wherein the first variable heavy structure domain and the first variable light structure domain constitute an Fv that binds to a first checkpoint inhibitor as outlined herein, and the second variable light chain together with the second variable heavy chain forms an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) CTLA-4 x PD-1, PD-1 x CTLA-4, LAG-3 x PD-1, BTLA x PD-1, and LAG-3 x CTLA-4.

[0318] In some embodiments, the central scFv format includes tilted variants, pI variants, ablation variants, and FcRn variants. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer comprising tilted variant S364K / E357Q, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and a first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising tilted variant L368D / K370S and pI variant N208D / Q. The FcRn variants are 295E / N384D / Q418E / N421D, E233P / L234V / L235A / G236del / S267K, M428L / N434S, and a first variable heavy structure domain and a second variable light chain, wherein the first variable heavy structure domain and the first variable light structure domain constitute an Fv that binds to a first checkpoint inhibitor as outlined herein, and the second variable light chain together with the second variable heavy chain forms an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) CTLA-4 x PD-1, PD-1 x CTLA-4, LAG-3 x PD-1, BTLA x PD-1, and LAG-3 x CTLA-4.

[0319] For a central-scFv sequence (optionally including M428L / N434S) similar to / utilizing the bottle opener main chain 1 of FIG37, specific Fv combinations used in the present invention include CTLA-4(Fab)X PD-1(scFv), PD-1(Fab)X CTLA-4(scFv), LAG-3(Fab)X PD-1(scFv), BTLA(Fab)X PD-1(scFv), and LAG-3(Fab)X CTLA-4(scFv).

[0320] For a central-scFv sequence (optionally including M428L / N434S) similar to / utilizing the opener main chain 1 of Figure 37, the specific ABD that binds to human PD-1 includes, but is not limited to: 1G6_H1.279_L1.194, 1G6_H1.280_L1.224; 1G6_L1.194_H1.279, 1G6_L1.210_H1.288 and 2E9_H1L1, and those listed in SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146.

[0321] For a central-scFv sequence (optionally containing M428L / N434S) similar to / utilizing the opener main chain 1 of Figure 37, the specific ABD binding to human CTLA-4 includes, but is not limited to: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L 0.22;[CTLA-4]_H2_L0;[CTLA-4]_H3.21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA -4]_H3.23_L0.129;[CTLA-4]_H3.23_L0.132;[CTLA-4]_H3.25_L0.124;[CTLA-4]_H3.25_L0.129;[CTLA-4]_H3.25_L0.132;[CTLA-4]_ H3.4_L0.118; [CTLA-4]_H3.4_L0.119; [CTLA-4]_H3.4_L0.12; [CTLA-4]_H3.4_L0.121; [CTLA-4]_H3.4_L0.122; [CTLA-4]_H3.4_L0.1 23;[CTLA-4]_H3.4_L0.124;[CTLA-4]_H3.4_L0.125;[CTLA-4]_H3.4_L0.126;[CTLA-4]_H3.4_L0.127;[CTLA-4]_H3.4_L0.128;[CTLA- 4]_H3.4_L0.129;[CTLA-4]_H3.4_L0.130;[CTLA-4]_H3.4_L0.131;[CTLA-4]_H3.4_L0.132;[CTLA-4]_H3.5_L2.1;[CTLA-4]_H3.5_L2 .2; [CTLA-4]_H3.5_L2.3; [CTLA-4]_H3_L0; [CTLA-4]_H3_L0.22; [CTLA-4]_H3_L0.44; [CTLA-4]_H3_L0.67 and [CTLA-4]_H3_L0.74; and SEQ. Those listed in SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416.

[0322] For a central-scFv sequence (optionally containing M428L / N434S) similar to / utilizing the bottle opener main chain 1 of Figure 37, the specific ABD binding to human LAG-3 includes, but is not limited to: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2. 93;2A11_H1_L2.97;2A11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18 7G8_H3.23_L1.11; 7G8_H3.28_L1; 7G8_H3.28_L1.11; 7G8_H3.28_L1.13; 7G8_H3.30_L1.34; Those listed in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002.

[0323] For a central-scFv sequence (optionally containing M428L / N434S) similar to / utilizing the opener main chain 1 of Figure 37, the specific ABD that binds to human BTLA includes, but is not limited to: 9C6_H0L0; 9C6_H1.1_L1; and 9C6_H1.11_L1; as well as those listed in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738.

[0324] For a central-scFv sequence (optionally including M428L / N434S) similar to / utilizing the opener main chain 1 of Figure 37, the specific ABD that binds to human TIM-3 includes, but is not limited to: 1D10_H0L0; 1D12_H0L0; 3H3_H1_L2.1; 6C8_H0L0; 6D9_H0_1D12_L0; 7A9_H0L0; 7B11_H0L0; 7B11var_H0L0 and 7C2_H0L0; and those listed in SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698 and SEQ ID NO: 36347 to 36706.

[0325] E. Central-Fv format

[0326] A heterodimer scaffold specifically used in this invention is as follows: Figure 1G The central-Fv format is shown. In this embodiment, the format depends on the use of an inserted scFv domain to form a third antigen-binding domain, wherein the Fab portions of two monomers bind to one checkpoint target, and an “additional” scFv domain binds to another checkpoint target. The scFv domain is inserted between the Fc domain and the CH1-Fv region of the monomer to provide the third antigen-binding domain, wherein each monomer contains a component of scFv (e.g., one monomer includes a variable heavy domain and the other includes a variable light domain).

[0327] In this embodiment, one monomer includes a first heavy chain comprising a first variable heavy structural domain, a CH1 structural domain, an Fc structural domain, and an additional variable light structural domain. The light structural domain is covalently connected between the C-terminus of the CH1 structural domain of the heavy constant structural domain and the N-terminus of the first Fc structural domain using a domain connector (vhl-CH1-[optional connector]-vl2-hinge-CH2-CH3). Another monomer includes a first heavy chain comprising a first variable heavy structural domain, a CH1 structural domain, an Fc structural domain, and an additional variable heavy structural domain (vhl-CH1-[optional connector]-vh2-hinge-CH2-CH3). The light structural domain is covalently connected between the C-terminus of the CH1 structural domain of the heavy constant structural domain and the N-terminus of the first Fc structural domain using a domain connector. This embodiment further utilizes a shared light chain comprising variable light structural domains and constant light structural domains, which is associated with the heavy chain to form two identical Fabs that bind a TTA. For many embodiments herein, as desired and described herein, these constructs include tilt variants, pI variants, ablation variants, additional Fc variants, etc. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) PD-1 and CTLA-4, CTLA-4 and PD-1, PD-1 and TIM-3, TIM-3 and PD-1, PD-1 and LAG-3, LAG-3 X PD1, PD-1 and TIGIT, TIGIT and PD-1, PD-1 and BTLA, BTLA and PD-1, CTLA-4 and TIM-3, TIM-3 and CTLA-4, CTLA-4 and LAG-3, LAG-3 and CTLA-4, CTLA-4 and TIGIT, TIGIT and CTLA-4, CTLA-4 and BTLA, BTLA and CTLA-4, TIM-3 and LAG-3, LAG-3 and TIM-3, TIM-3 and TIGIT, TIGIT and TIM-3, TIM-3 and BTLA, BTLA and TIM-3, LAG-3 and TIGIT, TIGIT and LAG-3, LAG-3 and BTLA, BTLA and LAG-3, BTLA and TIGIT, and TIGIT and BTLA.

[0328] These combined ABD sequences can be as disclosed in the sequence listing or as shown in Figures 9 to 13, and are in the position shown in Figure 39 and Figure 40 Any combination shown.

[0329] In the central-scFv format, specific Fv combinations used in this invention include CTLA-4(Fab)X PD-1(scFv), PD-1(Fab)X CTLA-4(scFv), LAG-3(Fab)X PD-1(scFv), BTLA(Fab)X PD-1(scFv), and LAG-3(Fab)X CTLA-4(scFv).

[0330] In the central-scFv format, specific ABDs associated with human PD-1 include, but are not limited to: 1G6_H1.279_L1.194, 1G6_H1.280_L1.224; 1G6_L1.194_H1.279, 1G6_L1.210_H1.288 and 2E9_H1L1, and those listed in SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146.

[0331] In the central-scFv format, specific ABDs associated with human CTLA-4 include, but are not limited to: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0;

[0332] [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H2_L0; [CTLA-4]_H3.21_L0.124; [CTLA-4]_H3.21_L 0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA-4]_H3.23_L0.129;[CTLA-4]_H3.23_L0 .132;[CTLA-4]_H3.25_L0.124;[CTLA-4]_H3.25_L0.129;[CTLA-4]_H3.25_L0.132;[CTLA-4]_H3.4_L0.1 18;[CTLA-4]_H3.4_L0.119;[CTLA-4]_H3.4_L0.12;[CTLA-4]_H3.4_L0.121;[CTLA-4]_H3.4_L0.122;[CT LA-4]_H3.4_L0.123;[CTLA-4]_H3.4_L0.124;[CTLA-4]_H3.4_L0.125;[CTLA-4]_H3.4_L0.126;[CTLA_4] _H3.4_L0.127;[CTLA_4]_H3.4_L0.128;[CTLA-4]_H3.4_L0.129;[CTLA-4]_H3.4_L0.130;[CTLA-4]_H3.4 _L0.131; [CTLA-4]_H3.4_L0.132; [CTLA-4]_H3.5_L2.1; [CTLA-4]_H3.5_L2.2; [CTLA-4]_H3.5_L2.3; [CTLA-4]_H3_L0; [CTLA-4]_H3_L0.22; [CTLA-4]_H3_L0.44; [CTLA-4]_H3_L0.67 and [CTLA-4]_H3_L0.74; and those listed in SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416.

[0333] In the central-scFv format, specific ABDs associated with human LAG-3 include, but are not limited to: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.93; 2A11_H1_L 2.97;2A11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_ and 7G8 H3L1, and those listed in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002.

[0334] In the central-scFv format, the specific ABDs associated with human BTLA include, but are not limited to: 9C6_H0L0; 9C6_H1.1_L1; and 9C6_H1.11_L1; as well as those listed in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738.

[0335] In the central-scFv format, the specific ABD associated with human TIM-3 includes, but is not limited to: 1D10_H0L0; 1D12_H0L0; 3H3_H1L2.1; 6C8_H0L0; 6D9_H0_1D12_L0; 7A9_H0L0; 7B11_H0L0; 7B11var_H0L0 and 7C2_H0L0; and those listed in SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698 and SEQ ID NO: 36347 to 36706.

[0336] F Single Arm Center - scFv

[0337] A heterodimer scaffold specifically used in this invention is as follows: Figure 1C The single-arm central-scFv format is shown. In this embodiment, one monomer includes only one Fc domain, while the other monomer uses an inserted scFv domain to form a second antigen-binding domain. In this format, the Fab portion binds to one checkpoint target, and the scFv binds to another checkpoint target. The scFv domain is inserted between the Fc domain and the CH1-Fv region of one of the monomers.

[0338] In this embodiment, a monomer includes a first heavy chain comprising a first variable heavy structure domain, a CH1 structure domain, and an Fc structure domain, wherein the scFv includes an scFv variable light structure domain, an scFv linker, and an scFv variable heavy structure domain. The scFv is covalently linked between the C-terminus of the CH1 structure domain of the heavy constant structure domain and the N-terminus of the first Fc structure domain using a domain linker. A second monomer includes an Fc structure domain. This embodiment further utilizes a light chain comprising a variable light structure domain and a constant light structure domain, the light chain being associated with the heavy chain to form a Fab. For many embodiments herein, as desired and described herein, these constructs comprise tilted variants, pI variants, ablation variants, additional Fc variants, etc. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv second) PD-1 and CTLA-4, CTLA-4 and PD-1, PD-1 and TIM-3, TIM-3 and PD-1, PD-1 and LAG-3, LAG-3 X PD1, PD-1 and TIGIT, TIGIT and PD-1, PD-1 and BTLA, BTLA and PD-1, CTLA-4 and TIM-3, TIM-3 and CTLA-4, CTLA-4 and LAG-3, LAG-3 and CTLA-4, CTLA-4 and TIGIT, TIGIT and CTLA-4, CTLA-4 and BTLA, BTLA and CTLA-4, TIM-3 and LAG-3, LAG-3 and TIM-3, TIM-3 and TIGIT, TIGIT and TIM-3, TIM-3 and BTLA, BTLA and TIM-3, LAG-3 and TIGIT, TIGIT and LAG-3, LAG-3 and BTLA, BTLA and LAG-3, BTLA and TIGIT, and TIGIT and BTLA.

[0339] These combined ABD sequences can be as disclosed in the sequence listing or as shown in Figures 9 to 13, and are in the position shown in Figure 39 and Figure 40 Any combination shown.

[0340] Furthermore, the Fc structural domain of the single-arm central-scFv format typically includes tilted variants (e.g., as shown in Figure 3 and...). Figure 8The group of amino acid substitutions shown includes particularly useful tilt variants selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L, K370S: S364K / E357Q, T366S / L368A / Y407V: T366W, and T366S / L368A / Y407V / Y349C: T366W / S354C), and optionally ablation variants (including such as...). Figure 5 The variants shown), optional band-charged scFv linkers (including those shown in Figure 7), and heavy chains including pI variants (including those shown in Figure 7) Figure 4 (Those variants shown).

[0341] In some embodiments, the single-arm central-scFv format includes a tilted variant, a pI variant, and an ablation variant. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer comprising the tilted variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and the first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising the tilted variant L368D / K370S and the pI variant N208D / Q. The ablation variants are 295E / N384D / Q418E / N421D, E233P / L234V / L235A / G236del / S267K, and a first variable heavy structure domain and a second variable light chain, wherein the first variable heavy structure domain and the first variable light structure domain constitute an Fv that binds to a first checkpoint inhibitor as outlined herein, and the second variable light chain together with the second variable heavy chain forms an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) CTLA-4 x PD-1, PD-1 x CTLA-4, LAG-3 x PD-1, BTLAX PD-1, and LAG-3 x CTLA-4.

[0342] In some embodiments, the single-arm central-scFv format includes tilted variants, pI variants, ablation variants, and FcRn variants. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer comprising tilted variant S364K / E357Q, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a first and second variable restructure domains, wherein the first variable restructure domain and a first variable light domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising tilted variant L368D / K370S and pI variant N208D / Q. The FcRn variants are 295E / N384D / Q418E / N421D, E233P / L234V / L235A / G236del / S267K, M428L / N434S, and a first variable heavy structure domain and a second variable light chain, wherein the first variable heavy structure domain and the first variable light structure domain constitute an Fv that binds to a first checkpoint inhibitor as outlined herein, and the second variable light chain together with the second variable heavy chain forms an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) CTLA-4 x PD-1, PD-1 x CTLA-4, LAG-3 x PD-1, BTLA x PD-1, and LAG-3 x CTLA-4.

[0343] In the single-arm central-scFv format, specific ABDs associated with human PD-1 include, but are not limited to: 1G6_H1.279_L1.194, 1G6_H1.280_L1.224; 1G6_L1.194_H1.279, 1G6_L1.210_H1.288 and 2E9H1L1, and those listed in SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146.

[0344] In the single-arm central-scFv format, the specific ABD associated with human CTLA-4 includes, but is not limited to: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H2_L0; [CTLA-4]_H2_L0; [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26 ... A-4]_H3.21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA-4]_H3.23_L0.129;[CT LA-4]_H3.23_L0.132;[CTLA-4]_H3.25_L0.124;[CTLA-4]_H3.25_L0.129;[CTLA-4]_H3.25_L0.132;[CTLA-4]_H3.4_L0.118;[CTL A-4]_H3.4_L0.119;[CTLA-4]_H3.4_L0.12;[CTLA-4]_H3.4_L0.121;[CTLA-4]_H3.4_L0.122;[CTLA-4]_H3.4_L0.123;[CTLA-4]_ H3.4_L0.124; [CTLA-4]_H3.4_L0.125; [CTLA-4]_H3.4_L0.126; [CTLA-4]_H3.4_L0.127; [CTLA-4]_H3.4_L0.128; [CTLA-4]_H3.4_ L0.129;[CTLA-4]_H3.4_L0.130;[CTLA-4]_H3.4_L0.131;[CTLA-4]_H3.4_L0.132;[CTLA-4]_H3.5_L2.1;[CTLA-4]_H3.5_L2.2;[ and SEQ. Those listed in SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416.

[0345] In the single-arm central-scFv format, the specific ABD associated with human LAG-3 includes, but is not limited to: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.93; 2A11_H1_L2.97 ;2A11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_L1.11;7 and SEQ. Those listed in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002.

[0346] In the single-arm central-scFv format, the specific ABD associated with human BTLA includes, but is not limited to: 9C6_H0L0; 9C6_H1.1_L1; and 9C6_H1.11_L1; as well as those listed in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738.

[0347] In the single-arm central-scFv format, the specific ABD associated with human TIM-3 includes, but is not limited to: 1D10_H0L0; 1D12_H0L0; 3H3_H1L2.1; 6C8_H0L0; 6D9_H0_1D12_L0; 7A9_H0L0; 7B11_H0L0; 7B11var_H0L0 and 7C2_H0L0; and those listed in SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698 and SEQ ID NO: 36347 to 36706.

[0348] G. Single-arm scFv-mAb

[0349] A heterodimer scaffold specifically used in this invention is as follows: Figure 1D The single-arm scFv-mAb format is shown. In this embodiment, one monomer includes only one Fc domain, while another monomer typically uses an scFv domain attached to the N-terminus of the heavy chain using the following connectors: vh-scFv connector-vl-[optional domain connector]-CH1-hinge-CH2-CH3 or (in the opposite direction) vl-scFv connector-vh-[optional domain connector]-CH1-hinge-CH2-CH3. In this format, the Fab portion is coupled to one checkpoint target, and the scFv is coupled to another checkpoint target. This embodiment further utilizes light chains comprising variable light domains and constant light domains, which are associated with heavy chains to form the Fab. For many embodiments herein, as desired and described herein, these constructs include tilted variants, pI variants, ablation variants, additional Fc variants, etc. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) PD-1 and CTLA-4, CTLA-4 and PD-1, PD-1 and TIM-3, TIM-3 and PD-1, PD-1 and LAG-3, LAG-3 X PD1, PD-1 and TIGIT, TIGIT and PD-1, PD-1 and BTLA, BTLA and PD-1, CTLA-4 and TIM-3, TIM-3 and CTLA-4, CTLA-4 and LAG-3, LAG-3 and CTLA-4, CTLA-4 and TIGIT, TIGIT and CTLA-4, CTLA-4 and BTLA, BTLA and CTLA-4, TIM-3 and LAG-3, LAG-3 and TIM-3, TIM-3 and TIGIT, TIGIT and TIM-3, TIM-3 and BTLA, BTLA and TIM-3, LAG-3 and TIGIT, TIGIT and LAG-3, LAG-3 and BTLA, BTLA and LAG-3, BTLA and TIGIT, and TIGIT and BTLA.

[0350] These combined ABD sequences can be as disclosed in the sequence listing or as shown in Figures 9 to 13, and are in the position shown in Figure 39 and Figure 40 Any combination shown.

[0351] Furthermore, the Fc domain typically includes tilted variants (e.g., as shown in Figure 3 and...). Figure 8The group of amino acid substitutions shown includes particularly useful tilt variants selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L, K370S: S364K / E357Q, T366S / L368A / Y407V: T366W, and T366S / L368A / Y407V / Y349C: T366W / S354C), and optionally ablation variants (including such as...). Figure 5 The variants shown), optional band-charged scFv linkers (including those shown in Figure 7), and heavy chains including pI variants (including those shown in Figure 7) Figure 4 (Those variants shown).

[0352] In some embodiments, the single-arm scFv-mAb format includes a tilted variant, a pI variant, and an ablation variant. Therefore, some embodiments include an opener format comprising: a) a first monomer comprising the tilted variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and the first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising the tilted variant L368D / K370S and the pI variant N208D / Q. The ablation variants are 295E / N384D / Q418E / N421D, E233P / L234V / L235A / G236del / S267K, and a first variable heavy structure domain and a second variable light chain, wherein the first variable heavy structure domain and the first variable light structure domain constitute an Fv that binds to a first checkpoint inhibitor as outlined herein, and the second variable light chain together with the second variable heavy chain forms an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) CTLA-4 x PD-1, PD-1 x CTLA-4, LAG-3 x PD-1, BTLA x PD-1, and LAG-3 x CTLA-4.

[0353] In some embodiments, the single-arm scFv-mAb format includes tilted variants, pI variants, ablation variants, and FcRn variants. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer comprising tilted variant S364K / E357Q, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and a first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising tilted variant L368D / K370S and pI variant N208D / Q. The FcRn variants are 295E / N384D / Q418E / N421D, E233P / L234V / L235A / G236del / S267K, M428L / N434S, and a first variable heavy structure domain and a second variable light chain, wherein the first variable heavy structure domain and the first variable light structure domain constitute an Fv that binds to a first checkpoint inhibitor as outlined herein, and the second variable light chain together with the second variable heavy chain forms an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. In this embodiment, suitable Fv pairs include (Fab is listed first, scFv is listed second) CTLA-4 x PD-1, PD-1 x CTLA-4, LAG-3 x PD-1, BTLA x PD-1, and LAG-3 x CTLA-4.

[0354] In the single-arm scFv-mAb format, specific ABDs that bind to human PD-1 include, but are not limited to: 1G6_H1.279_L1.194, 1G6_H1.280_L1.224; 1G6_L1.194_H1.279, 1G6_L1.210_H1.288 and 2E9H1L1, and those listed in SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146.

[0355] In the single-arm scFv-mAb format, the specific ABDs associated with human CTLA-4 include, but are not limited to: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H2_L0; [CTLA-4]_H2_L0; [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26 ... A-4]_H3.21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA-4]_H3.23_L0.129;[CT LA-4]_H3.23_L0.132;[CTLA-4]_H3.25_L0.124;[CTLA-4]_H3.25_L0.129;[CTLA-4]_H3.25_L0.132;[CTLA-4]_H3.4_L0.118;[CTL A-4]_H3.4_L0.119;[CTLA-4]_H3.4_L0.12;[CTLA-4]_H3.4_L0.121;[CTLA-4]_H3.4_L0.122;[CTLA-4]_H3.4_L0.123;[CTLA-4]_ H3.4_L0.124; [CTLA-4]_H3.4_L0.125; [CTLA-4]_H3.4_L0.126; [CTLA-4]_H3.4_L0.127; [CTLA-4]_H3.4_L0.128; [CTLA-4]_H3.4_ L0.129;[CTLA-4]_H3.4_L0.130;[CTLA-4]_H3.4_L0.131;[CTLA-4]_H3.4_L0.132;[CTLA-4]_H3.5_L2.1;[CTLA-4]_H3.5_L2.2;[ and SEQ. Those listed in SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416.

[0356] In the single-arm scFv-mAb format, specific ABDs associated with human LAG-3 include, but are not limited to: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.93; 2A11_H1_L2.97 ;2A11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_L1.11;7 and SEQ. Those listed in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002.

[0357] In the single-arm scFv-mAb format, the specific ABDs that bind to human BTLA include, but are not limited to: 9C6_H0L0; 9C6_H1.1_L1; and 9C6_H1.11_L1; as well as those listed in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738.

[0358] In the single-arm scFv-mAb format, the specific ABDs that bind to human TIM-3 include, but are not limited to: 1D10_H0L0; 1D12_H0L0; 3H3_H1_L2.1; 6C8_H0L0; 6D9_H0_1D12_L0; 7A9_H0L0; 7B11_H0L0; 7B11var_H0L0 and 7C2_H0L0; and those listed in SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698 and SEQ ID NO: 36347 to 36706.

[0359] H.scFv-mAb format

[0360] A heterodimer scaffold specifically used in this invention is as follows: Figure 1E The mAb-scFv format is shown. In this embodiment, the format depends on linking the N-terminus of the scFv to one of the monomers to form a third antigen-binding domain, wherein the Fab portions of the two monomers bind to a checkpoint target, and the “additional” scFv domain binds to a different checkpoint target.

[0361] In this embodiment, the first monomer includes a first heavy chain (including a variable heavy structure domain and a constant structure domain), wherein the N-terminal covalently linked scFv includes a scFv variable light structure domain, a scFv connector, and a scFv variable heavy structure domain in either orientation (((vh1-scFv connector-vl1-[optional structure domain connector]-vh2-CH1-hinge-CH2-CH3) or (scFv in the opposite orientation) ((vl1-scFv connector-vh1-[optional structure domain connector]-vh2-CH1-hinge-CH2-CH3)). This embodiment further utilizes A shared light chain comprising variable and constant light domains is used, which is associated with a heavy chain to form two identical Fabs binding one of the target antigens. For many embodiments herein, as desired and described herein, these constructs comprise tilted variants, pI variants, ablation variants, additional Fc variants, etc. In this embodiment, suitable Fv pairs comprise (Fabs are listed first, scFvs are listed second) PD-1 and CTLA-4, CTLA-4 and PD-1, PD-1 and TIM-3, TIM-3 and PD-1, PD-1 and LAG-3, LAG-3 X PD1, PD-1 and TIGIT, TIGIT and PD-1, PD-1 and BTLA, BTLA and PD-1, CTLA-4 and TIM-3, TIM-3 and CTLA-4, CTLA-4 and LAG-3, LAG-3 and CTLA-4, CTLA-4 and TIGIT, TIGIT and CTLA-4, CTLA-4 and BTLA, BTLA and CTLA-4, TIM-3 and LAG-3, LAG-3 and TIM-3, TIM-3 and TIGIT, TIGIT and TIM-3, TIM-3 and BTLA, BTLA and TIM-3, LAG-3 and TIGIT, TIGIT and LAG-3, LAG-3 and BTLA, BTLA and LAG-3, BTLA and TIGIT, and TIGIT and BTLA.

[0362] These combined ABD sequences can be as disclosed in the sequence listing or as shown in Figures 9 to 13, and are in the position shown in Figure 39 and Figure 40 Any combination shown.

[0363] Furthermore, the Fc domain in the scFv-mAb format typically includes tilted variants (e.g., as shown in Figure 3 and...). Figure 8 The group of amino acid substitutions shown includes particularly useful tilt variants selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L, K370S: S364K / E357Q, T366S / L368A / Y407V: T366W, and T366S / L368A / Y407V / Y349C: T366W / S354C), and optionally ablation variants (including such as...). Figure 5 The variants shown), optional band-charged scFv linkers (including those shown in Figure 7), and heavy chains including pI variants (including those shown in Figure 7) Figure 4 (Those variants shown).

[0364] In some embodiments, the mAb-scFv format includes tilt variants, pI variants, and ablation variants. Therefore, some embodiments include an opener format comprising: a) a first monomer comprising the tilt variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and a first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising the tilt variant L368D / K370S and the pI variant N208D / Q. 295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable heavy structure domain and a second variable light chain, the first variable heavy structure domain and the first variable light structure domain constituting an Fv that binds to a first checkpoint inhibitor as outlined herein, the second variable light chain together with the second variable heavy chain forming an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. Particularly used in some embodiments under this format are (Fab-scFv sequence) CTLA-4 x PD-1, LAG-3 x PD-1, BTLA x PD-1, and LAG-3 x CTLA-4.

[0365] In some embodiments, the mAb-scFv format includes tilt variants, pI variants, ablation variants, and FcRn variants. Therefore, some embodiments include an opener format comprising: a) a first monomer comprising tilt variant S364K / E357Q, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and a first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising tilt variant L368D / K370S and pI variant N208D / Q. 295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a first variable heavy structure domain and a second variable light chain, wherein the first variable heavy structure domain and the first variable light structure domain constitute an Fv that binds to a first checkpoint inhibitor as outlined herein, and the second variable light chain together with the second variable heavy chain forms an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. Particularly used in some embodiments under this format are (Fab-scFv sequence) CTLA-4 XPD-1, LAG-3 X PD-1, BTLA X PD-1, and LAG-3 X CTLA-4.

[0366] For the mAb-scFv format master chain 1 from Figure 38 (optionally containing M428L / N434S), the specific ABDs that bind to human PD-1 include, but are not limited to: 1G6_H1.279_L1.194, 1G6_H1.280_L1.224; 1G6_L1.194_H1.279, 1G6_L1.210_H1.288 and 2E9_H1L1, and those listed in SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146.

[0367] For the mAb-scFv format master chain 1 from Figure 38 (optionally containing M428L / N434S), the specific ABD associated with human CTLA-4 includes, but is not limited to: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.25 ... _H0_L0.22;[CTLA-4]_H2_L0;[CTLA-4]_H3.21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L 0.124;[CTLA-4]_H3.23_L0.129;[CTLA-4]_H3.23_L0.132;[CTLA-4]_H3.25_L0.124;[CTLA-4]_H3.25_L0.129;[CTLA-4]_H3.25 _L0.132;[CTLA-4]_H3.4_L0.118;[CTLA-4]_H3.4_L0.119;[CTLA-4]_H3.4_L0.12;[CTLA-4]_H3.4_L0.121;[CTLA-4]_H3.4_L0 .122;[CTLA-4]_H3.4_L0.123;[CTLA-4]_H3.4_L0.124;[CTLA-4]_H3.4_L0.125;[CTLA-4]_H3.4_L0.126;[CTLA-4]_H3.4_L0.12 7; [CTLA-4]_H3.4_L0.128; [CTLA-4]_H3.4_L0.129; [CTLA-4]_H3.4_L0.130; [CTLA-4]_H3.4_L0.131; [CTLA-4]_H3.4_L0.132; [CTLA-4]_H3.5_L2.1; [CTLA-4]_H3.5_L2.2; [CTLA-4]_H3.5_L2.3; [CTLA-4]_H3_L0; [CTLA-4]_H3_L0.22; [CTLA-4]_H3_L0.44;

[0368] [CTLA-4]_H3_L0.67 and [CTLA-4]_H3_L0.74; and those listed in SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416.

[0369] For the mAb-scFv format main chain 1 from Figure 38 (optionally containing M428L / N434S), the specific ABD associated with human LAG-3 includes, but is not limited to: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.93; 2A11_H1_L2.97;2A11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_ and 7G8_H3.30L1.34; and 7G8_H3L1; and SEQ. Those listed in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002.

[0370] For the mAb-scFv format main chain 1 from Figure 38 (optionally containing M428L / N434S), the specific ABD that binds to human BTLA includes, but is not limited to: 9C6_H0L0; 9C6_H1.1_L11 and 9C6_H1.11_L1; and those listed in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738.

[0371] For the mAb-scFv format master chain 1 from Figure 38 (optionally containing M428L / N434S), the specific ABD associated with human TIM-3 includes, but is not limited to: 1D10_H0L0; 1D12_H0L0; 3H3_H1L2.1; 6C8_H0L0; 6D9 H01D12 L0; 7A9_H0L0; 7B11_H0L0; 7B11var_H0L0 and 7C2_H0L0; and those listed in SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698 and SEQ ID NO: 36347 to 36706.

[0372] I. Dual SCFV format

[0373] The present invention also provides, as is known in the art and Figure 1B The dual scFv format is shown in the figure. In this embodiment, the heterodimeric bispecific antibody is composed of two scFv-Fc monomers (the two monomers are in the format of (vhscFv linker-vl-[optional domain linker]-CH2-CH3) or (vl-scFv linker-vh-[optional domain linker]-CH2-CH3), or one monomer is in one orientation and the other monomer is in the other orientation).

[0374] In this case, all ABDs are in scFv format, where any combination of PD-1 and CTLA-4, PD-1 and TIM-3, PD-1 and LAG-3, PD-1 and TIGIT, PD-1 and BTLA, CTLA-4 and TIM-3, CTLA-4 and LAG-3, CTLA-4 and TIGIT, CTLA-4 and BTLA, TIM-3 and LAG-3, TIM-3 and TIGIT, TIM-3 and BTLA, LAG-3 and TIGIT, LAG-3 and BTLA, and TIGIT and BTLA are useful. The ABD sequences for these combinations can be as disclosed in the sequence listing or as shown in Figures 9 to 13, and are in the format shown in Figure 39 and... Figure 40 Any combination shown.

[0375] Furthermore, the Fc domain of the double scFv format typically includes tilted variants (e.g., as shown in Figure 3 and...). Figure 8The group of amino acid substitutions shown includes particularly useful tilt variants selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L, K370S: S364K / E357Q, T366S / L368A / Y407V: T366W, and T366S / L368A / Y407V / Y349C: T366W / S354C), and optionally ablation variants (including such as...). Figure 5 The variants shown), optional band-charged scFv linkers (including those shown in Figure 7), and heavy chains including pI variants (including those shown in Figure 7) Figure 4 (Those variants shown).

[0376] In some embodiments, the dual scFv format includes a tilt variant, a pI variant, and an ablation variant. Therefore, some embodiments include an opener format comprising: a) a first monomer comprising the tilt variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and the first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising the tilt variant L368D / K370S and the pI variant N208D / Q. 295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, and a first variable heavy structure domain and a second variable light chain, the first variable heavy structure domain and the first variable light structure domain constituting an Fv that binds to a first checkpoint inhibitor as outlined herein, the second variable light chain together with the second variable heavy chain forming an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. Particularly used in some embodiments under this format are (Fab-scFv sequence) CTLA-4 x PD-1, LAG-3 x PD-1, BTLA x PD-1, and LAG-3 x CTLA-4.

[0377] In some embodiments, the dual scFv format includes a tilted variant, a pI variant, an ablation variant, and an FcRn variant. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer comprising the tilted variant S364K / E357Q, the ablation variant E233P / L234V / L235A / G236del / S267K, the FcRn variant M428L / N434S, and a first variable restructure domain and a second variable restructure domain, wherein the first variable restructure domain and the first variable light structure domain of the light chain constitute an Fv that binds to a first checkpoint inhibitor; b) a second monomer comprising the tilted variant L368D / K370S and the pI variant N208D / Q. 295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a first variable heavy structure domain and a second variable light chain, wherein the first variable heavy structure domain and the first variable light structure domain constitute an Fv that binds to a first checkpoint inhibitor as outlined herein, and the second variable light chain together with the second variable heavy chain forms an Fv (ABD) that binds to a second checkpoint inhibitor; and c) a light chain comprising a first variable light structure domain and a constant light structure domain. Particularly used in some embodiments under this format are (Fab-scFv sequence) CTLA-4 x PD-1, LAG-3 x PD-1, BTLA x PD-1, and LAG-3 x CTLA-4.

[0378] J. Non-heterodimeric bispecific antibody

[0379] As those skilled in the art will appreciate, the Fv sequences outlined herein can also be used for both monospecific antibodies (e.g., “conventional monoclonal antibodies”) and non-heterodimeric bispecific formats.

[0380] Suitable non-heterodimer bispecific formats are known in the art and include many different formats generally described in Spiess et al., Molecular Immunology (67): 95-106 (2015) and Kontermann, mAb 4:2, 182-197 (2012), the formats of which are explicitly incorporated into this paper by reference, and particularly for figures, legends and citations.

[0381] K. Monospecific monoclonal antibodies

[0382] As those skilled in the art will appreciate, the novel Fv sequences outlined herein can also be used for both monospecific antibodies (e.g., “conventional monoclonal antibodies”) and non-heterodimeric bispecific formats. Therefore, the present invention provides monoclonal (monospecific) antibodies comprising six CDRs and / or vh and vl sequences from the accompanying drawings, typically having a constant region of IgG1, IgG2, IgG3, or IgG4, wherein IgG1, IgG2, and IgG4 (including an IgG4 constant region comprising an S228P amino acid substitution) are particularly used in some embodiments. That is, any sequence with the designation “H_L” herein can be linked to a constant region of a human IgG1 antibody.

[0383] VI. Antigen-binding domain targeting the target antigen

[0384] The bispecific antibody of the present invention has two distinct antigen-binding domains (ABDs) that bind to two distinct target checkpoint antigens (“target pairs”) in a generally bivalent or trivalent bispecific format as shown in Figure 1. Suitable target checkpoint antigens include human (and sometimes cynomolgus monkey) PD-1, CTLA-4, TIM-3, LAG-3, TIGIT, and BTLA, the sequences of which are shown in Figure 2. Therefore, suitable bispecific antibodies bind to PD-1 and CTLA-4, PD-1 and TIM-3, PD-1 and LAG-3, PD-1 and TIGIT, PD-1 and BTLA, CTLA-4 and TIM-3, CTLA-4 and LAG-3, CTLA-4 and TIGIT, CTLA-4 and BTLA, TIM-3 and LAG-3, TIM-3 and TIGIT, TIM-3 and BTLA, LAG-3 and TIGIT, LAG-3 and BTLA, and TIGIT and BTLA. It should be noted that, for each pair, these bispecific antibodies are typically named “anti-PD-1 x anti-CTLA-4” or, more simply or for convenience (and therefore interchangeably), “PD-1 x CTLA-4”, etc. It should be noted that, unless otherwise stated herein, the order of the antigens listed by name does not confer a structure; that is, a PD-1 x CTLA-4 opener antibody may have an scFv that binds to either PD-1 or CTLA-4, although in some cases, the order specifies a structure as indicated.

[0385] As outlined more fully in this article, these combinations of ABDs can take the form of a wide variety of formats, as summarized below, typically a combination of one ABD in Fab format and another in scFv format. As discussed in this article and shown in Figure 1, some formats use a single Fab and a single scFv (… Figure 1A , Figure 1C and Figure 1D ), and some formats use two Fabs and a single scFv ( Figure 1E , Figure 1F , Figure 1G , Figure 1H and Figure 1I ).

[0386] A. Antigen-binding domain

[0387] As discussed herein, the bispecific checkpoint heterodimeric antibodies of the present invention comprise two antigen-binding domains (ABDs), each of which binds to a different checkpoint protein. As outlined herein, these heterodimeric antibodies can be bispecific and bivalent (each antigen binds to, for example, a checkpoint protein). Figure 1A The format described herein is either a single ABD binding or a bispecific and trivalent binding (one antigen binds to a single ABD, and another antigen binds to, for example, a single ABD binding). Figure 1F The two ABD combinations described in the text).

[0388] In addition, typically, one of the ABDs includes the scFv, as outlined herein, oriented from the N-terminus to the C-terminus of the vh-scFv linker-vl or the vl-scFv linker-vh. Depending on the format, one or both of the other ABDs typically include a Fab consisting of a vh domain on one protein chain (typically a component of the heavy chain) and a vl domain on another protein chain (typically a component of the light chain).

[0389] This invention also provides numerous ABDs that bind to a number of different checkpoint proteins as outlined below. As those skilled in the art will appreciate, any set of six CDRs or vh and vl domains can be in scFv or Fab format, which is then added to heavy constant and light constant domains, wherein the heavy constant domains include variants (contained within the CH1 domain and the Fc domain). The scFv sequences contained in the sequence listing utilize specific charged linkers, while uncharged or other charged linkers, including those described in Figure 7, can be used as outlined herein.

[0390] Furthermore, as discussed above, the number used in the sequence listing to identify CDRs is Kabat; however, different numbers can be used, which will change the amino acid sequence of the CDRs as shown in Table 1.

[0391] Further variations can be made for all the variable heavy and light domains listed herein. As outlined herein, in some embodiments, a set of six CDRs can have 0, 1, 2, 3, 4, or 5 amino acid modifications (where amino acid substitution is particularly useful) and variations in the frame regions of the variable heavy and light domains, provided that the frame (other than the CDRs) maintains at least about 80%, 85%, or 90% identity with the phylogenetic sequences selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1, the entire contents of which are incorporated herein by reference. Thus, for example, the same CDRs as described herein can be combined with different frame sequences from phylogenetic sequences, provided that the frame regions maintain at least 80%, 85%, or 90% identity with the phylogenetic sequences selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1. Alternatively, CDRs may have amino acid modifications (e.g., one, two, three, four, or five amino acid modifications in a set of CDRs; that is, a CDR may be modified as long as the total number of variations in a set of six CDRs is less than six amino acid modifications, wherein any combination of CDRs is altered; for example, one change may exist in vlCDR1, two changes may exist in vhCDR2, no change may exist in vhCDR3, etc.)), and frame region modifications, provided that the frame region maintains at least 80%, 85%, or 90% identity with the phylogenetic sequences selected from those sequences listed in U.S. Patent No. 7,657,380 in Figure 1.

[0392] B. PD-1 antigen-binding domain

[0393] In some embodiments, one of the ABDs is combined with PD-1. Suitable sets of six CDRs and / or vh and vl domains, along with scFv sequences, are depicted in SEQ ID NOs: 6209 to 11464, SEQ ID NOs: 11465 to 17134, SEQ ID NOs: 33003 to 33072, SEQ ID NOs: 33073 to 35394, and SEQ ID NOs: 36127 to 36146. Figure 9 illustrates ABD sequences of particular interest in some embodiments, and these ABD sequences include those in the sequence listing with the following identifiers: 1G6_H1.279_L1.194; 1G6_H1.280_L1.224; 1G6_L1.194_H1.279; 1G6_L1.210_H1.288; and 2E9_H1L1.

[0394] As those skilled in the art will appreciate, a suitable anti-PD-1 ABD may include a set of six CDRs as described in these sequences and figures, as underlined, or, in cases using different numbering schemes as described herein and shown in Table 1, other CDRs identified by alignment within the vh and vl sequences SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394, and SEQ ID NO: 36127 to 36146. A suitable ABD may also include the entire vh and vl sequences as depicted in these sequences and figures as scFv or Fab. In many embodiments containing Fv to PD-1 herein, the scFv monomer is bound to PD-1. As discussed herein, when PD-1 is one of the antigens, the other target pair is selected from CTLA-4 (SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818, and SEQ ID NO: 35395 to 35416), which depict suitable sequences (the sequences may be scFv sequences, CDR sequence groups, or vh and vl sequences)), TIM-3 (SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698, and SEQ ID NO: 36347 to 36706), which depict suitable sequences (the sequences may be scFv sequences, CDR sequence groups, or vh and vl sequences)), LAG-3 (SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, ... Suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002, respectively. BTLA (suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738, respectively. BTLA (suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 21504 to 21523 and SEQ ID NO: 37435 to 37586, respectively. BTLA (suitable sequences (which may be scFv sequences, CDR sequences, or vh and vl sequences) are depicted in SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002. BTLA (suitable sequences (which may be scFv sequences, CDR sequences, or vh and vl sequences) are depicted in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738 ...1504 to 21523 and SEQ ID NO: 37435 to 37586. BTLA (suitable sequences (which may

[0395] Particularly useful ABDs that bind to human PD-1 include, but are not limited to: 1G6_H1.279_L1.194, 1G6_H1.280_L1.224; 1G6_L1.194_H1.279, 1G6_L1.210_H1.288 and 2E9_H1L1.

[0396] In addition to the parental CDR sets that form ABD to PD-1 disclosed in the sequence listing, the present invention provides variant CDR sets. In one embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), a set of six CDRs may have one, two, three, four, or five amino acid changes from the parental CDRs, provided that ABD is still able to bind to the target antigen, wherein the BLI is particularly used in many embodiments.

[0397] In addition to the parental variable heavy and variable light domains for forming ABD to PD-1 disclosed herein, this invention provides variant vh and vl domains. In one embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), the variant vh and vl domains may each have 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid changes from the parental vh and vl domains, wherein the BLI is particularly used in many embodiments. In another embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), variants vh and vl have at least 90%, 95%, 97%, 98%, or 99% identity with the corresponding parent vh or vl, provided that the ABD is still able to bind to the target antigen, wherein the BLI is particularly used in many embodiments.

[0398] A particularly preferred embodiment includes 1G6_L1.194_H1.279 anti-PD-1Fv in scFv format contained within any of the bottle opener format main chains of FIG37.

[0399] A particularly preferred embodiment includes 1G6_L1.194_H1.279 anti-PD-1Fv in scFv format contained within any of the mAb-scFv format main chains of FIG38.

[0400] C.CTLA-4 antigen-binding domain

[0401] In some embodiments, one of the ABDs is combined with CTLA-4. Suitable multiple sets of six CDRs and / or vh and vl domains, as well as scFv sequences, are depicted in SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416.Figure 10 illustrates ABD sequences of particular interest in some embodiments, and the ABD sequences also include those sequences in the sequence listing with the following identifiers: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H0_L0.22; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H0.25_L0 ... -4]_H2_L0;[CTLA-4]_H3.21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA-4]_H3.2 3_L0.129;[CTLA-4]_H3.23_L0.132;[CTLA-4]_H13.25_L0.124;[CTLA-4]_H3.25_L0.129;[CTLA-4]_H3.25_L0.132;[CTLA-4]_H3.4_ L0.118;[CTLA-4]_H3.4_L0.119;[CTLA-4]_H3.4_L0.12;[CTLA-4]_H3.4_L0.121;[CTLA-4]_H3.4_L0.122;[CTLA-4]_H3.4_L0.123; [CTLA-4]_H3.4_L0.124; [CTLA-4]_H3.4_L0.125; [CTLA-4]_H3.4_L0.126; [CTLA-4]_H3.4_L0.127; [CTLA-4]_H3.4_L0.128; [CTLA-4 ]_H3.4_L0.129;[CTLA-4]_H3.4_L0.130;[CTLA-4]_H3.4_L0.131;[CTLA-4]_H3.4_L0.132;[CTLA-4]_H3.5_L2.1;[CTLA-4]_H3.5_L 2.2; [CTLA-4]_H3.5_L2.3; [CTLA-4]_H3_L0; [CTLA-4]_H3_L0.22; [CTLA-4]_H3_L0.44; [CTLA-4]_H3_L0.67; and [CTLA-4]_H3_L0.74.

[0402] As those skilled in the art will appreciate, suitable anti-CTLA-4 ABDs may include a set of six CDRs as described in these sequences and figures, as underlined, or, in cases using different numbering schemes as described herein and shown in Table 1, other CDRs identified by alignment within the vh and vl sequences SEQ ID NOs: 21 to 2918, SEQ ID NOs: 2919 to 6208, SEQ ID NOs: 36739 to 36818, and SEQ ID NOs: 35395 to 35416. Suitable ABDs may also include the entire vh and vl sequences as depicted in these sequences and figures, used as scFv or Fab. In many embodiments containing Fv to CTLA-4 herein, the scFv monomer is bound to CTLA-4. As discussed herein, when CTLA-4 is one of the antigens, the other target pair is selected from PD-1 (SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146), which depicts suitable sequences (the sequences may be scFv sequences, CDR sequence groups, or vh and vl sequences)), TIM-3 (SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698, and SEQ ID NO: 36347 to 36706), which depict suitable sequences (the sequences may be scFv sequences, CDR sequence groups, or vh and vl sequences)), LAG-3 (SEQ ID NO: 17135 to 20764, SEQ ID NO: 36127 to 36146), which depict suitable sequences (the sequences may be scFv sequences, CDR sequence groups, or vh and vl sequences)). Suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002), BTLA (suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738), and TIGIT (suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 21504 to 21523 and SEQ ID NO: 37435 to 37586).

[0403] In addition to the parental CDR sets forming ABD to CTLA-4 disclosed in the sequence listing, the present invention provides variant CDR sets. In one embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), a set of six CDRs may have one, two, three, four, or five amino acid alterations from the parental CDRs, provided that ABD is still able to bind to the target antigen, wherein the BLI is particularly used in many embodiments.

[0404] In addition to the parental variable heavy and variable light domains for forming ABD to CTLA-4 disclosed herein, this invention provides variant vh and vl domains. In one embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), the variant vh and vl domains may each have 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid changes from the parental vh and vl domains, provided that ABD remains able to bind to the target antigen.

[0405] The BLI described therein is particularly used in many embodiments. In another embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), variants vh and vl have at least 90%, 95%, 97%, 98%, or 99% identity with the corresponding parent vh or vl, provided that the ABD is still able to bind to the target antigen, and the BLI described therein is particularly used in many embodiments.

[0406] A particularly preferred embodiment includes a Fab-formatted [CTLA-4]_H3_L0.22 anti-CTLA-4 Fv contained within any of the bottle opener format main chains of FIG37.

[0407] A particularly preferred embodiment includes [CTLA-4]_H3_L0.22 anti-CTLA-4 Fv in scFv format contained within any of the bottle opener format main chains of FIG37.

[0408] A particularly preferred embodiment includes [CTLA-4]_H3_L0.22 anti-CTLA-4 Fv in scFv format contained within any of the mAb-scFv format main chains of FIG38.

[0409] A particularly preferred embodiment includes a [CTLA-4]_H3_L0.22 anti-CTLA-4 Fv in Fab format contained within any of the mAb-scFv format main chains of FIG38.

[0410] D.TIM-3 antigen-binding domain

[0411] In some embodiments, one of the ABDs is combined with TIM-3. Suitable sets of six CDRs and / or vh and vl domains, as well as scFv sequences, are depicted in SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698, and SEQ ID NO: 36347 to 36706. ABD sequences of particular interest in some embodiments include those in the sequence listing that have the following identifiers: 1D10_H0L0; 1D12_H0L0; 3H3_H1_L2.1; 6C8_H0L0; 6D9_H0_1D12_L0; 7A9_H0L0; 7B11_H0L0; 7B11var_H0L0; and 7C2_H0L0.

[0412] As those skilled in the art will appreciate, a suitable anti-TIM-3 ABD may include a set of six CDRs as described in these sequences and figures, as underlined, or, in cases using different numbering schemes as described herein and shown in Table 1, other CDRs identified by alignment within the vh and vl sequences SEQ ID NO: 20765 to 20884, SEQ ID NO: 37587 to 37698, and SEQ ID NO: 36347 to 36706. A suitable ABD may also include the entire vh and vl sequences as depicted in these sequences and figures as scFv or Fab. In many embodiments containing Fv to TIM-3 herein, the Fab monomer is combined with TIM-3. As discussed herein, when TIM-3 is one of the antigens, the other target pair is selected from PD-1 (SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)), CTLA-4 (SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)), LAG-3 (SEQ ID NO: 17135 to 20764, SEQ ID NO: 36127 to 36146), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)), PD-1 (SEQ ID NO: 17135 to 20764, SEQ ID NO: 36127 to 36146), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)). Suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002), BTLA (suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738), and TIGIT (suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 21504 to 21523 and SEQ ID NO: 37435 to 37586).

[0413] In addition to the parental CDR sets that form ABD to TIM-3 disclosed in the sequence listing, the present invention provides variant CDR sets. In one embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), a set of six CDRs may have one, two, three, four, or five amino acid changes from the parental CDRs, provided that ABD is still able to bind to the target antigen, wherein the BLI is particularly used in many embodiments.

[0414] In addition to the parental variable heavy and variable light domains for forming ABD to TIM-3 disclosed herein, the present invention provides variant vh and vl domains. In one embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), the variant vh and vl domains may each have 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid changes from the parental vh and vl domains, wherein the BLI is particularly used in many embodiments. In another embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), variants vh and vl have at least 90%, 95%, 97%, 98%, or 99% identity with the corresponding parent vh or vl, provided that the ABD is still able to bind to the target antigen, wherein the BLI is particularly used in many embodiments.

[0415] LAG-3 antigen-binding domain

[0416] In some embodiments, one of the ABDs is combined with LAG-3. Suitable multiple sets of six CDRs and / or vh and vl domains and scFv sequences are depicted in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002. Figure 11 illustrates ABD sequences of particular interest in some embodiments, and the ABD sequences also include those sequences in the sequence listing with the following identifiers: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.93; 2A 11_H1_L2.97;2A11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_ L1.11; 7G8_H3.23_L1.11; 7G8_H3.28_L1; 7G8_H3.28_L1.11; 7G8_H3.28_L1.13; 7G8_H3.30_L1.34; 7G8_H3.30_L1.34; and 7G8_H3L1.

[0417] As those skilled in the art will appreciate, suitable anti-LAG-3 ABDs may include a set of six CDRs as described in these sequences and figures, as underlined, or, in cases using different numbering schemes as described herein and shown in Table 1, other CDRs identified by alignment within the vh and vl sequences SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793, and SEQ ID NO: 32794 to 33002. Suitable ABDs may also comprise the entire vh and vl sequences as depicted in these sequences and figures as scFv or Fab. In many embodiments containing Fv to LAG-3 herein, the Fab monomer is bound to LAG-3. As discussed herein, when LAG-3 is one of the antigens, the other target pair is selected from PD-1 (SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)), CTLA-4 (SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)), TIM-3 (SEQ ID NO: 20765 to 20884, SEQ ID NO: 36127 to 36146), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)). Suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NOs 37587 to 37698 and SEQ ID NOs 36347 to 36706, respectively. BTLA (suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NOs 20885 to 21503 and SEQ ID NOs 36707 to 36738, respectively.) and TIGIT (suitable sequences (which may be scFv sequences, CDR sequences, or vh and vl sequences) are depicted in SEQ ID NOs 21504 to 21523 and SEQ ID NOs 37435 to 37586, respectively.) are depicted in SEQ ID NOs 37587 to 37698 and SEQ ID NOs 36347 to 36706.

[0418] In addition to the parental CDR sets forming ABD to LAG-3 disclosed in the sequence listing, the present invention provides variant CDR sets. In one embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), a set of six CDRs may have one, two, three, four, or five amino acid changes from the parental CDRs, provided that ABD is still able to bind to the target antigen, wherein the BLI is particularly used in many embodiments.

[0419] In addition to the parental variable heavy and variable light domains for forming ABD to LAG-3 disclosed herein, the present invention provides variant vh and vl domains. In one embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), the variant vh and vl domains may each have 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid changes from the parental vh and vl domains, wherein the BLI is particularly used in many embodiments. In another embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), variants vh and vl have at least 90%, 95%, 97%, 98%, or 99% identity with the corresponding parent vh or vl, provided that the ABD is still able to bind to the target antigen, wherein the BLI is particularly used in many embodiments.

[0420] A particularly preferred embodiment includes a Fab-formatted 7G8_H3.30_L1.34 anti-LAG-3 Fv contained within any of the bottle opener format main chains of FIG37.

[0421] A particularly preferred embodiment includes any of the bottle opener format main chains of FIG37 containing 7G8_H3.30_L1.34 anti-LAG-3 Fv in scFv format.

[0422] E.BTLA antigen-binding domain

[0423] In some embodiments, one of the ABDs is combined with BTLA. Suitable sets of six CDRs and / or vh and vl domains, as well as scFv sequences, are depicted in SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738. Figure 12 illustrates ABD sequences of particular interest in some embodiments, and these ABD sequences also include those sequences in the sequence listing with the following identifiers: 9C6_H0L0; 9C6_H1.1_L1; and 9C6_H1.11_L1.

[0424] As those skilled in the art will appreciate, a suitable anti-BTLA ABD may include a set of six CDRs as described in these sequences and figures, as underlined, or other CDRs identified by alignment within the vh and vl sequences SEQ ID NO: 20885 to 21503 and SEQ ID NO: 36707 to 36738, in cases using different numbering schemes as described herein and shown in Table 1. A suitable ABD may also include the entire vh and vl sequences as depicted in these sequences and figures as scFv or Fab. In many embodiments containing Fv to BTLA herein, the Fab monomer is bound to BTLA. As discussed herein, when LAG-3 is one of the antigens, the other target pair is selected from PD-1 (SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)), CTLA-4 (SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)), TIM-3 (SEQ ID NO: 20765 to 20884, SEQ ID NO: 36127 to 36146), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)). Suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NOs 37587 to 37698 and SEQ ID NOs 36347 to 36706, respectively. LAG-3 (suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NOs 17135 to 20764, SEQ ID NOs 36819 to 36962, SEQ ID NOs 35417 to 35606, SEQ ID NOs 25194 to 32793 and SEQ ID NOs 32794 to 33002.) and TIGIT (suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NOs 21504 to 21523 and SEQ ID NOs 37435 to 37586.) are depicted in SEQ ID NOs 21504 to 21523.

[0425] In addition to the parental CDR sets that form ABD to BTLA disclosed in the sequence listing, the present invention provides variant CDR sets. In one embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), a set of six CDRs may have one, two, three, four, or five amino acid changes from the parental CDRs, provided that ABD is still able to bind to the target antigen, wherein the BLI is particularly used in many embodiments.

[0426] In addition to the parental variable heavy and variable light domains for forming ABD to BTLA disclosed herein, the present invention provides variant vh and vl domains. In one embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), the variant vh and vl domains may each have 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid changes from the parental vh and vl domains, wherein the BLI is particularly used in many embodiments. In another embodiment, as measured by at least one of Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), variants vh and vl have at least 90%, 95%, 97%, 98%, or 99% identity with the corresponding parent vh or vl, provided that the ABD is still able to bind to the target antigen, wherein the BLI is particularly used in many embodiments.

[0427] A particularly preferred embodiment includes 9C6H1.1L1 anti-LAG-3 Fv in Fab format contained within any of the bottle opener format main chains of FIG37.

[0428] A particularly preferred embodiment includes any of the bottle opener format main chains of FIG37 containing 7G8_H3.30_L1.34 anti-LAG-3 Fv in scFv format.

[0429] F. TIGIT antigen-binding domain

[0430] In some embodiments, one of the ABDs is combined with TIGIT. Suitable multiple sets of six CDRs and / or vh and vl domains, as well as scFv sequences, are depicted in SEQ ID NO: 21504 to 21523 and SEQ ID NO: 37435 to 37586.

[0431] As those skilled in the art will appreciate, a suitable anti-TIGIT ABD may include a set of six CDRs as described in these sequences and figures, as underlined, or other CDRs identified by alignment within the vh and vl sequences SEQ ID NO: 21504 to 21523 and SEQ ID NO: 37435 to 37586, in cases using different numbering schemes as described herein and shown in Table 1. A suitable ABD may also include the entire vh and vl sequences as depicted in these sequences and figures as scFv or Fab. In many embodiments containing Fv to TIGIT herein, the Fab monomer is combined with TIGIT. As discussed herein, when LAG-3 is one of the antigens, the other target pair is selected from PD-1 (SEQ ID NO: 6209 to 11464, SEQ ID NO: 11465 to 17134, SEQ ID NO: 33003 to 33072, SEQ ID NO: 33073 to 35394 and SEQ ID NO: 36127 to 36146), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)), CTLA-4 (SEQ ID NO: 21 to 2918, SEQ ID NO: 2919 to 6208, SEQ ID NO: 36739 to 36818 and SEQ ID NO: 35395 to 35416), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)), TIM-3 (SEQ ID NO: 20765 to 20884, SEQ ID NO: 36127 to 36146), which depicts a suitable sequence (the sequence may be an scFv sequence, a CDR sequence group, or a vh and vl sequence)). Suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 37587 to 37698 and SEQ ID NO: 36347 to 36706, respectively. LAG-3 (suitable sequences (which may be scFv sequences, CDR sequence groups, or vh and vl sequences) are depicted in SEQ ID NO: 17135 to 20764, SEQ ID NO: 36819 to 36962, SEQ ID NO: 35417 to 35606, SEQ ID NO: 25194 to 32793 and SEQ ID NO: 32794 to 33002) are depicted in SEQ ID NO: 33002.

[0432] G. Specific Bispecific Examples

[0433] This invention provides a number of specific bispecific antibodies as outlined below.

[0434] 1. LAG-3 X CTLA-4

[0435] In some embodiments, the present invention provides a bispecific heterodimeric antibody comprising a first ABD binding to human LAG-3 and a second ABD binding to human CTLA-4, wherein the first ABD and the second ABD may be in any format shown in FIG. 1. Much of this disclosure relates to a bottle opener format in which the Fab is the LAG-3 side and the CTLA-4 side is the scFv side, but this may be reversed for all embodiments herein.

[0436] In one embodiment, the LAG-3 XCTLA-4 bispecific antibody presents... Figure 1A The bottle opener format, where CTLA-4ABD is scFv. In another embodiment, the LAG-3 X CTLA-4 bispecific antibody presents as... Figure 1F The central-scFv format, where LAG-3ABD is the Fab component. In another embodiment, the LAG-3 X CTLA-4 bispecific antibody presents as... Figure 1F The central-scFv format, where CTLA-4ABD is scFv.

[0437] LAG-3 X CTLA-4 bispecific antibodies (in opener or center-scFv format) typically contain tilted variants, pI variants, and ablation variants as outlined herein. That is, in either format, the Fc domains of the two monomers can include tilted variants (e.g., a set as shown in Figure 3 and...). Figure 8 The amino acid substitutions shown), optionally ablation variants (containing Figure 5 Those variants shown in the diagram), and monomers including the Fab side (e.g., heavy-chain constant domains) include pI variants (containing... Figure 4 (Those variants shown in the image).

[0438] In some embodiments, the LAG-3 XCTLA-4 bispecific antibody includes an Fc domain having tilted variants, wherein particularly useful tilted variants are selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L; K370S: S364K / E357Q; T366S / L368A / Y407V: T366W; and T366S / L368A / Y407V / Y349C: T366W / S354C.

[0439] In some embodiments, the LAG-3 X CTLA-4 antibody comprises a tilt variant, a pI variant, and an ablation variant. Therefore, some embodiments include an opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilt variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, and an Fv bound to a checkpoint inhibitor as outlined herein; b) a second monomer (“Fab monomer”). The Fv comprises the tilt variant L368D / K370S, the pI variant N208D / Q295E / N384D / Q418E / N421D, the ablation variant E233P / L234V / L235A / G236del / S267K, and a variable restructure domain, which, together with a variable light structure domain, constitutes an Fv that binds to a second checkpoint inhibitor as outlined herein; and c) a light chain. Specific examples of this embodiment utilize LAG-3 Fab 7G8_H3.30_L1.34 and CTLA-4scFv[CTLA-4]_H3.23_L0.129, although either CTLA-4 or LAG-3 Fv in the sequence listing can be paired and used in any combination.

[0440] In some embodiments, the LAG-3 XCTLA-4 antibody comprises a tilted variant, a pI variant, an ablation variant, and an FcRn variant. Therefore, some embodiments include an opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilted variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, an FcRn variant M428L / N434S, and an Fv that binds to a checkpoint inhibitor as outlined herein; b) a second monomer. The body (“Fab monomer”) comprises tilt variant L368D / K370S, pI variant N208D / Q295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a variable restructure domain, which, together with a variable light structure domain, constitutes an Fv that binds to a second checkpoint inhibitor as outlined herein; and c) a light chain. A specific example of this embodiment utilizes LAG-3 Fab7G8_H3.30_L1.34 and CTLA-4 scFv[CTLA-4]_H3.23_L0.129, although either CTLA-4 or LAG-3 Fv in the sequence listing can be paired and used in any combination.

[0441] Additional embodiments include either of the main chains from FIG37 having LAG-3 Fab 7G8_H3.30_L1.34 and CTLA-4scFv[CTLA-4]_H3.23_L0.129.

[0442] Additional embodiments include either of the main chains from FIG38 having LAG-3 Fab 7G8_H3.30_L1.34 and CTLA-4scFv[CTLA-4]_H3.23_L0.129.

[0443] In some embodiments, for the LAG-3 XCLTA-4 bispecific antibody, the Fv on the LAG-3 Fab side is selected from those sequences in the sequence listing that have the following identifiers: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.93; 2A11_H1_L2.97; 2A 11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_L1.11;7 G8_H3.23_L1.11; 7G8_H3.28_L1; 7G8_H3.28_L1.11; 7G8_H3.28_L1.13; 7G8_H3.30_L1.34; 7G8_H3.30_L1.34;The Fv on the CTLA-4scFv side is selected from those sequences in the sequence listing that have the following identifiers: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H2_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H2_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H2_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H2_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H0 ... -4]_H3.21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA-4]_H3.23_L0.129;[CTL A-4]_H3.23_L0.132;[CTLA-4]_H3.25_L0.124;[CTLA-4]_H3.25_L0.129;[CTLA-4]_H3.25_L0.132;[CTLA-4]_H3.4_L0.118;[CTL A-4]_H3.4_L0.119;[CTLA-4]_H3.4_L0.12;[CTLA-4]_H3.4_L0.121;[CTLA-4]_H3.4_L0.122;[CTLA-4]_H3.4_L0.123;[CTLA-4] _H3.4_L0.124;[CTLA-4]_H3.4_L0.125;[CTLA-4]_H3.4_L0.126;[CTLA-4]_H3.4_L0.127;[CTLA-4]_H3.4_L0.128;[CTLA-4]_H3. 4_L0.129; [CTLA-4]_H3.4_L0.130; [CTLA-4]_H3.4_L0.131; [CTLA-4]_H3.4_L0.132; [CTLA-4]_H3.5_L2.1; [CTLA-4]_H3.5_L2.2 ; [CTLA-4]_H3.5_L2.3; [CTLA-4]_H3_L0; [CTLA-4]_H3_L0.22; [CTLA-4]-H3-L0.44; [CTLA-4]_H3_L0.67; and [CTLA-4]_H3_L0.74.

[0444] In some embodiments, the LAG-3 X CTLA-4 bispecific antibody is selected from those constructs listed in SEQ ID NO: 35607 to 35866 and SEQ ID NO: 21524 to 22620.

[0445] In some embodiments, the LAG-3 X CTLA-4 bispecific antibody is selected from: XENP20206, XENP21582, XENP21584, XENP21588, XENP22123, XENP22124, XENP22125, XENP22604, XENP22672, XENP22847, XENP22847, XENP22841, and XENP22849.

[0446] 2. BTLAX PD-1

[0447] In some embodiments, the present invention provides a bispecific heterodimeric antibody comprising a first ABD that binds to human BTLA and a second ABD that binds to human PD-1, the first ABD and the second ABD being in any format shown in FIG1. ​​Much of this disclosure relates to a bottle opener format in which the Fab is the BTLA side and the PD-1 side is the scFv side, but this may be reversed for all embodiments herein.

[0448] In one embodiment, the BTLAX PD-1 bispecific antibody presents... Figure 1A The bottle opener format, where PD-1 ABD is scFv. In another embodiment, the BTLA X PD-1 bispecific antibody presents as... Figure 1F The central-scFv format, where BTLAABD is the Fab component. In another embodiment, the BTLA X PD-1 bispecific antibody presents as... Figure 1F The central-scFv format, where PD-1ABD is scFv.

[0449] BTLA x PD-1 bispecific antibodies (in opener or center-scFv format) typically contain tilted variants, pI variants, and ablation variants as outlined herein. That is, in either format, the Fc domains of the two monomers can include tilted variants (e.g., a set as shown in Figure 3 and...). Figure 8 The amino acid substitutions shown), optionally ablation variants (containing Figure 5 Those variants shown in the diagram), and monomers including the Fab side (e.g., heavy-chain constant domains) include pI variants (containing... Figure 4 (Those variants shown in the image).

[0450] In some embodiments, the BTLAX PD-1 bispecific antibody includes an Fc domain having tilted variants, wherein particularly useful tilted variants are selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L; K370S: S364K / E357Q; T366S / L368A / Y407V: T366W; and T366S / L368A / Y407V / Y349C: T366W / S354C.

[0451] In some embodiments, the BTLAX PD-1 antibody comprises a tilt variant, a pI variant, and an ablation variant. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilted variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, and an Fv bound to a checkpoint inhibitor as outlined herein; b) a second monomer (“Fab monomer”) comprising a tilted variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, an ablation variant E233P / L234V / L235A / G236del / S267K, and a variable restructure domain, the variable restructure domain and the variable light structure domain constituting an Fv bound to a second checkpoint inhibitor as outlined herein; and c) a light chain. A specific instance of this embodiment utilizes BTLAFab 9C6_H1.1_L1 and PD-1 scFv 1G6_L1.194_H1.279, although either BTLA or PD-1Fv in the sequence list can be paired and used in any combination.

[0452] In some embodiments, the BTLA X PD-1 antibody comprises a tilted variant, a pI variant, an ablation variant, and an FcRn variant. Therefore, some embodiments include an opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilted variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, an FcRn variant M428L / N434S, and an Fv that binds to a checkpoint inhibitor as outlined herein; b) a second... The two monomers (“Fab monomers”) comprise tilt variants L368D / K370S, pI variants N208D / Q295E / N384D / Q418E / N421D, ablation variants E233P / L234V / L235A / G236del / S267K, FcRn variants M428L / N434S, and a variable restructure domain, which, together with a variable light structure domain, constitutes an Fv that binds to a second checkpoint inhibitor as outlined herein; and c) a light chain. A specific example of this embodiment utilizes BTLA Fab 9C6_H1.1_L1 and PD-1 scFv 1G6_L1.194_H1.279, although either BTLA or PD-1 Fv in the sequence listing can be paired and used in any combination.

[0453] Additional embodiments include either of the main chains from FIG37 having BTLA Fab 9C6 H1.1L1 and PD-1 scFv 1G6_L1.194_H1.279.

[0454] Additional embodiments include either of the main chains from FIG38 having BTLA Fab 9C6 H1.1L1 and PD-1scFv 1G6_L1.194_H1.279.

[0455] In some embodiments, for the BTLA x PD-1 bispecific antibody, the Fv on the BTLA Fab side is selected from those sequences in the sequence listing that have the following identifiers: 9C6_H0L0, 9C6_H1.1_L1, and 9C6_H1.11_L1. The Fv on the PD-1 scFv side is selected from those sequences in the sequence listing that have the following identifiers: 1G6_H1.279_L1.194; 1G6_H1.280_L1.224; 1G6_L1.194_H1.279; 1G6_L1.210_H1.288; and 2E9_H1L1.

[0456] In some embodiments, the BTLA X PD-1 bispecific antibody is selected from constructs comprising those listed in SEQ ID NO: 22724 to 23315 and SEQ ID NO: 36147 to 36166.

[0457] In some embodiments, the BTLA X PD-1 bispecific antibody is selected from XENP20895, XENP21220, XENP21221 and XENP22858.

[0458] 3. CTLA-4 X PD-1

[0459] In some embodiments, the present invention provides a bispecific heterodimeric antibody comprising a first ABD binding to human CTLA-4 and a second ABD binding to human PD-1, wherein the first ABD and the second ABD may be in any format shown in FIG1. ​​Much of this disclosure relates to a bottle opener format in which the Fab is the CTLA-4 side and the PD-1 side is the scFv side, but this may be reversed for all embodiments herein.

[0460] In one embodiment, the CTLA-4X PD-1 bispecific antibody presents Figure 1A The bottle opener format, where PD-1ABD is scFv. In another embodiment, the CTLA-4X PD-1 bispecific antibody is presented as... Figure 1F The central-scFv format, where CTLA-4ABD is the Fab component. In another embodiment, the CTLA-4 X PD-1 bispecific antibody presents as... Figure 1F The central-scFv format, where PD-1 ABD is scFv.

[0461] CTLA-4 x PD-1 bispecific antibodies (in opener or center-scFv format) typically contain tilted variants, pI variants, and ablation variants as outlined herein. That is, in either format, the Fc domains of the two monomers can include tilted variants (e.g., a set as shown in Figure 3 and...). Figure 8 The amino acid substitutions shown), optionally ablation variants (containing Figure 5 Those variants shown in the diagram), and monomers including the Fab side (e.g., heavy-chain constant domains) include pI variants (containing... Figure 4 (Those variants shown in the image).

[0462] In some embodiments, the CTLA-4 X PD-1 bispecific antibody includes an Fc domain having tilted variants, wherein particularly useful tilted variants are selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364KT411T / E360E / Q; 362E: D401K; L368D / K370S: S364K / E357L; K370S: S364K / E357Q; T366S / L368A / Y407V T366W; and T366S / L368A / Y407V / Y349C: T366W / S354C.

[0463] In some embodiments, the CTLA-4 X PD-1 antibody includes tilt variants, pI variants, and ablation variants. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilted variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, and an Fv bound to a checkpoint inhibitor as outlined herein; b) a second monomer (“Fab monomer”) comprising a tilted variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, an ablation variant E233P / L234V / L235A / G236del / S267K, and a variable restructure domain, the variable restructure domain and the variable light structure domain constituting an Fv bound to a second checkpoint inhibitor as outlined herein; and c) a light chain. A specific instance of this embodiment utilizes CTLA-4 Fab[CTLA-4]_H3_L0.22 and PD-1scFv 1G6_L1.194_H1.279, although either CTLA-4 or PD-1 Fv in the sequence list can be paired and used in any combination.

[0464] In some embodiments, the CTLA-4 X PD-1 antibody comprises a tilted variant, a pI variant, an ablation variant, and an FcRn variant. Therefore, some embodiments include an opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilted variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, an FcRn variant M428L / N434S, and an Fv that binds to a checkpoint inhibitor as outlined herein; b) a second The monomer (“Fab monomer”) includes tilt variant L368D / K370S, pI variant N208D / Q295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a variable restructure domain, which, together with a variable light structure domain, constitutes an Fv that binds to a second checkpoint inhibitor as outlined herein; and c) a light chain. A specific example of this embodiment utilizes CTLA-4 Fab[CTLA-4]_H3_L0.22 and PD-1 scFv 1G6_L1.194_H1.279, although either CTLA-4 or PD-1 Fv in the sequence listing can be paired and used in any combination.

[0465] Additional embodiments include either of the main chains from FIG37 having CTLA-4 Fab[CTLA-4]_H3_L0.22 and PD-1 scFv1G6_L1.194_H1.279.

[0466] Additional embodiments include either of the main chains from FIG38 having CTLA-4 Fab[CTLA-4]_H3_L0.22 and PD-1 scFv1G6_L1.194_H1.279.

[0467] In some embodiments, for the CTLA-4 X PD-1 bispecific antibody, the Fv on the CTLA-4 Fab side is selected from those sequences in the sequence listing that have the following identifiers: [CTLA-4]_H0.25_L0; [CTLA-4]_H0.26_L0; [CTLA-4]_H0.27_L0; [CTLA-4]_H0.29_L0; [CTLA-4]_H0.38_L0; [CTLA-4]_H0.39_L0; [CTLA-4]_H0.40_L0; [CTLA-4]_H0.70_L0; [CTLA-4]_H0_L0.22; [CTLA-4]_H2_L0; [CTLA-4]_H 3.21_L0.124;[CTLA-4]_H3.21_L0.129;[CTLA-4]_H3.21_L0.132;[CTLA-4]_H3.23_L0.124;[CTLA-4]_H3.23_L0.129;[CTLA-4] _H3.23_L0.132;[CTLA-4]_H3.25_L0.124;[CTLA-4]_H3.25_L0.129;[CTLA-4]_H3.25_L0.132;[CTLA-4]_H3.4_L0.118;[CTLA-4 ]_H3.4_L0.119;[CTLA-4]_H3.4_L0.12;[CTLA-4]_H3.4_L0.121;[CTLA-4]_H3.4_L0.122;[CTLA-4]_H3.4_L0.123;[CTLA-4]_H3 .4_L0.124;[CTLA-4]_H3.4_L0.125;[CTLA-4]_H3.4_L0.126;[CTLA-4]_H3.4_L0.127;[CTLA-4]_H3.4_L0.128;[CTLA-4]_H3.4_ L0.129; [CTLA-4]_H3.4_L0.130; [CTLA-4]_H3.4_L0.131; [CTLA-4]_H3.4_L0.132; [CTLA-4]_H3.5_L2.1; [CTLA-4]_H3.5_L2.2; [CTLA-4]_H3.5_L2.3; [CTLA-4]_H3_L0; [CTLA-4]_H3_L0.22; [CTLA-4]_H3_L0.44; [CTLA-4]_H3_L0.67; and [CTLA-4]_H3_L0.74.The Fv on the PD-1scFv side is selected from those sequences in the sequence list that have the following identifiers: 1G6_H1.279_L1.194; 1G6_H1.280_L1.224; 1G6_L1.194_H1.279; 1G6_L1.210_H1.288; and 2E9H1L1.

[0468] In some embodiments, the CTLA-4 X PD-1 bispecific antibody is selected from those listed in SEQ ID NO: 36167 to 36346 and SEQ ID NO: 23316 to 23735.

[0469] In some embodiments, the CTLA-4 X PD-1 bispecific antibody is selected from XENP19738, XENP19739, XENP19741, XENP20053, XENP20066, XENP20130, XENP20146, XENP20717 and XENP22836.

[0470] 4. LAG-3 X PD-1

[0471] In some embodiments, the present invention provides a bispecific heterodimeric antibody comprising a first ABD binding to human LAG-3 and a second ABD binding to human PD-1, wherein the first ABD and the second ABD may be in any format as shown in FIG. 1. Much of this disclosure relates to a bottle opener format in which the Fab is the LAG-3 side and the PD-1 side is the scFv side, but this may be reversed for all embodiments herein.

[0472] In one embodiment, the LAG-3 X PD-1 bispecific antibody presents Figure 1A The bottle opener format, where PD-1ABD is scFv. In another embodiment, the LAG-3 X PD-1 bispecific antibody is presented as... Figure 1F The central-scFv format, where LAG-3 ABD is the Fab component. In another embodiment, the LAG-3 X PD-1 bispecific antibody presents as... Figure 1F The central-scFv format, where PD-1 ABD is scFv.

[0473] LAG-3 X PD-1 bispecific antibodies (in opener or center-scFv format) typically contain tilted variants, pI variants, and ablation variants as outlined herein. That is, in either format, the Fc domains of the two monomers can include tilted variants (e.g., a set as shown in Figure 3 and...). Figure 8 The amino acid substitutions shown), optionally ablation variants (containing Figure 5Those variants shown in the diagram), and monomers including the Fab side (e.g., heavy-chain constant domains) include pI variants (containing... Figure 4 (Those variants shown in the image).

[0474] In some embodiments, the LAG-3 X PD-1 bispecific antibody includes an Fc domain having tilted variants, wherein particularly useful tilted variants are selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364K; T411T / E360E / Q362E: D401K; L368D / K370S: S364K / E357L; K370S: S364K / E357Q; T366S / L368A / Y407V: T366W; and T366S / L368A / Y407V / Y349C: T366W / S354C.

[0475] In some embodiments, the LAG-3 X PD-1 antibody comprises a tilt variant, a pI variant, and an ablation variant. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilted variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, and an Fv bound to a checkpoint inhibitor as outlined herein; b) a second monomer (“Fab monomer”) comprising a tilted variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, an ablation variant E233P / L234V / L235A / G236del / S267K, and a variable restructure domain, the variable restructure domain and the variable light structure domain constituting an Fv bound to a second checkpoint inhibitor as outlined herein; and c) a light chain. A specific instance of this embodiment utilizes LAG-3 Fab 7G8_H3.30_L1.34 and PD-1scFv 1G6_L1.194_H1.279, although either LAG-3 or PD-1Fv in the sequence list can be paired and used in any combination.

[0476] In some embodiments, the LAG-3 X PD-1 antibody comprises a tilted variant, a pI variant, an ablation variant, and an FcRn variant. Therefore, some embodiments include an opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilted variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, an FcRn variant M428L / N434S, and an Fv that binds to a checkpoint inhibitor as outlined herein; b) a second monomer. The monomer (“Fab monomer”) comprises tilt variant L368D / K370S, pI variant N208D / Q295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a variable restructure domain, which, together with a variable light structure domain, constitutes an Fv that binds to a second checkpoint inhibitor as outlined herein; and c) a light chain. A specific example of this embodiment utilizes LAG-3 Fab7G8_H3.30_L1.34 and PD-1 scFv 1G6_L1.194_H1.279, although either LAG-3 or PD-1 Fv in the sequence listing can be paired and used in any combination.

[0477] Additional embodiments include either of the main chains from Figure 37 having LAG-3 Fab 7G8_H3.30_L1.34 and PD-1scFv 1G6_L1.194_H1.279.

[0478] Additional embodiments include either of the main chains from FIG38 having LAG-3 Fab 7G8_H3.30_L1.34 and PD-1scFv 1G6_L1.194_H1.279.

[0479] In some embodiments, for the LAG-3 X PD-1 bispecific antibody, the Fv on the LAG-3 Fab side is selected from those sequences in the sequence listing that have the following identifiers: 2A11_H0L0; 2A11_H1.125_L2.113; 2A11_H1.144_L2.142; 2A11_H1_L2.122; 2A11_H1_L2.123; 2A11_H1_L2.124; 2A11_H1_L2.25; 2A11_H1_L2.47; 2A11_H1_L2.50; 2A11_H1_L2.91; 2A11_H1_L2.93; 2A11_H1_L2.97; 2A 11_H1L1;2A11_H1L2;2A11_H2L2;2A11_H3L1;2A11_H3L2;2A11_H4L1;2A11_H4L2;7G8_H0L0;7G8_H1L1;7G8_H3.18_L1.11;7 G8_H3.23_L1.11; 7G8_H3.28_L1; 7G8_H3.28_L1.11; 7G8_H3.28_L1.13; 7G8_H3.30_L1.34; 7G8_H3.30_L1.34; The Fv on the PD-1scFv side is selected from those sequences in the sequence list that have the following identifiers: 1G6_H1.279_L1.194; 1G6_H1.280_L1.224; 1G6_L1.194_H1.279; 1G6_L1.210_H1.288; and 2E9 H1L1.

[0480] In some embodiments, the LAG-3 X PD-1 bispecific antibody is selected from constructs comprising those listed in SEQ ID NO: 35867 to 36126 and SEQ ID NO: 23736 to 25133.

[0481] In some embodiments, the LAG-3 (X PI)-1 bispecific antibody is selected from: XENP20206, XENP21582, XENP21584, XENP21588, XENP22123, XENP22124, XENP22125, XENP22604, XENP22672, XENP22847, XENP22847, and XENP22849.

[0482] 5TIGIT X PD-1

[0483] In some embodiments, the TIGIT X PD-1 bispecific antibody is selected from those constructs listed in SEQ ID NO: 25134 to 25173.

[0484] 6. TIM-3 X PD-1

[0485] In some embodiments, the present invention provides a bispecific heterodimeric antibody comprising a first ABD binding to human TIM-3 and a second ABD binding to human PD-1, wherein the first ABD and the second ABD may be in any format shown in FIG1. ​​Much of this disclosure relates to a bottle opener format in which the Fab is the TIM-3 side and the PD-1 side is the scFv side, but this may be reversed for all embodiments herein.

[0486] In one embodiment, the TIM-3 X PD-1 bispecific antibody presents Figure 1A The bottle opener format, where PD-1ABD is scFv. In another embodiment, the TIM-3 X PD-1 bispecific antibody is presented as... Figure 1F The central-scFv format, where TIM-3 ABD is the Fab component. In another embodiment, the TIM-3 X PD-1 bispecific antibody presents as... Figure 1F The central-scFv format, where PD-1 ABD is scFv.

[0487] TIM-3 X PD-1 bispecific antibodies (in opener or center-scFv format) typically contain tilted variants, pI variants, and ablation variants as outlined herein. That is, in either format, the Fc domains of the two monomers can include tilted variants (e.g., a set as shown in Figure 3 and...). Figure 8 The amino acid substitutions shown), optionally ablation variants (containing Figure 5 Those variants shown in the diagram), and monomers including the Fab side (e.g., heavy-chain constant domains) include pI variants (containing... Figure 4 (Those variants shown in the image).

[0488] In some embodiments, the TIM-3 X PD-1 bispecific antibody includes an fc domain having tilted variants, wherein particularly useful tilted variants are selected from the group consisting of: S364K / E357Q: L368D / K370S; L368D / K370S: S364K; L368E / K370S: S364KT411T / E360E / Q; 362E: D401K; L368D / K370S: S364K / E357L; K370S: S364K / E357Q; T366S / L368A / Y407V: T366W; and T366S / L368A / Y407V / Y349C: T366W / S354C.

[0489] In some embodiments, the TIM-3 X PD-1 antibody comprises a tilt variant, a pI variant, and an ablation variant. Therefore, some embodiments include a bottle opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilted variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, and an Fv bound to a checkpoint inhibitor as outlined herein; b) a second monomer (“Fab monomer”) comprising a tilted variant L368D / K370S, a pI variant N208D / Q295E / N384D / Q418E / N421D, an ablation variant E233P / L234V / L235A / G236del / S267K, and a variable restructure domain, the variable restructure domain forming a variable light structure domain.

[0490] Fv bound to a second checkpoint inhibitor as outlined herein; and c) a light chain. A specific example of this embodiment utilizes PD-1 scFv 1G6_L1.194_H1.279, although either TIM-3 or PD-1 Fv in the sequence listing can be paired and used in any combination.

[0491] In some embodiments, the TIM-3 X PD-1 antibody comprises a tilted variant, a pI variant, an ablation variant, and an FcRn variant. Therefore, some embodiments include an opener format comprising: a) a first monomer (“scFv monomer”) comprising a charged scFv linker (the +H sequence of FIG7 is preferred in some embodiments), a tilted variant S364K / E357Q, an ablation variant E233P / L234V / L235A / G236del / S267K, an FcRn variant M428L / N434S, and an Fv that binds to a checkpoint inhibitor as outlined herein; b) a second... The two monomers (“Fab monomers”) comprise tilt variant L368D / K370S, pI variant N208D / Q295E / N384D / Q418E / N421D, ablation variant E233P / L234V / L235A / G236del / S267K, FcRn variant M428L / N434S, and a variable restructure domain, which, together with a variable light structure domain, constitutes an Fv and c) light chain that binds to a second checkpoint inhibitor as outlined herein. A specific example of this embodiment utilizes PD-1 scFv1G6_L1.194_H1.279, although either TIM-3 or PD-1 Fv in the sequence listing can be paired and used in any combination.

[0492] Additional embodiments include any one of the main chains from FIG37 having a TIM-3 Fab side and PD-1 scFv1G6_L1.194_H1.279.

[0493] Additional embodiments include any one of the main chains from FIG38 having a TIM-3 Fab side and PD-1 scFv1G6_L1.194_H1.279.

[0494] In some embodiments, for the TIM-3 Fab-side X PD-1 bispecific antibody, the Fv on the TIM-3 Fab side is selected from those sequences in the sequence listing that have the following identifiers: 1D10_H0L0; 1D12_H0L0; 3H3_H1_L2.1; 6C8_H0L0; 6D9_H0_1D12_L0; 7A9 H0L0; 7B11_H0L0; 7B11var_H0L0; and 7C2_H0L0. The Fv on the PD-1 scFv side is selected from those sequences in the sequence listing that have the following identifiers: 1G6_H1.279L_1.194; 1G6_H1.280_L1.224; 1G6_L1.194_H1.279; 1G6_L1.210_H1.288; and 2E9 H1L1.

[0495] Furthermore, the antibodies of the present invention comprise antibodies that bind to or compete with the antigen-binding domains outlined herein, which are the same as those outlined herein. In some embodiments, a bispecific checkpoint antibody may contain one of the ABDs outlined herein and a second ABD that competes with one of the ABDs outlined herein. In some embodiments, the two ABDs compete with the respective ABDs outlined herein. Binding competition is typically determined using Biacore, surface plasmon resonance (SPR), and / or BLI (biological bilayer interference, e.g., octet body assay), wherein the BLI is particularly used in many embodiments.

[0496] VII. Useful Examples

[0497] In one embodiment, a specific combination of the tilted variant and pI variant used in this invention is T366S / L368A / Y407V:T366W (optionally containing a bridging disulfide, T366S / L368A / Y407V / Y349C:T366W / S354C), where one monomer comprises Q295E / N384D / Q418E / N481D and the other monomer comprises a positively charged scFv linker (when the format includes an scFv domain). As will be appreciated in the art, the "mortar and pestle structure" variant does not change the pI and can therefore be used on either monomer.

[0498] VIII. Nucleic Acids of the Invention

[0499] The present invention further provides nucleic acid compositions encoding the bispecific antibodies of the present invention (or, in the case of "monospecific" antibodies, also encoding those monospecific antibodies).

[0500] As those skilled in the art will understand, nucleic acid compositions will depend on the format and scaffold of the heterodimeric protein. Thus, for example, when the format requires three amino acid sequences, as for all formats described in Figure 1 except for the dual scFv format, the three nucleic acid sequences can be incorporated into one or more expression vectors for expression. Similarly, some formats require only two nucleic acids (e.g., the dual scFv format, as disclosed in Figure 1); again, the two nucleic acids are placed into one or more expression vectors.

[0501] As is known in the art, nucleic acids encoding the compositions of the present invention can be incorporated into expression vectors as is known in the art and depend on the host cells used to generate the heterodimeric antibodies of the present invention. Typically, the nucleic acid is operatively linked to any number of regulatory elements (promoters, origins of replication, selectable markers, ribosome binding sites, inducers, etc.). The expression vector can be an extrachromosomal vector or an integration vector.

[0502] The nucleic acid and / or expression vectors of the present invention can then be converted into any number of different types of host cells as is well known in the art, including mammalian cells, bacterial cells, yeast cells, insect cells and / or fungal cells, wherein mammalian cells (e.g., CHO cells) are used in many embodiments.

[0503] In some embodiments, the nucleic acid encoding each monomer and the optional nucleic acid encoding the light chain are typically contained in a single expression vector under different or the same promoter control conditions, depending on the applicable format. In embodiments particularly used in this invention, each of the two or three nucleic acids is contained in a different expression vector. As shown herein and as illustrated by reference to 62 / 025,931, which is hereby incorporated herein, different vector ratios can be used to drive heterodimer formation. That is, surprisingly, although the protein comprises a first monomer: second monomer: light chain ratio of 1:1:2 (in the case of many embodiments herein with three polypeptides comprising a heterodimer antibody), these are not ratios that yield optimal results.

[0504] The heterodimeric antibodies of the present invention are prepared by culturing host cells including one or more expression vectors as well known in the art. Once produced, conventional antibody purification steps are performed, including ion exchange chromatography. As discussed herein, making the pIs of the two monomers differ by at least 0.5 allows separation by ion exchange chromatography, isoelectric focusing, or other methods sensitive to the isoelectric point. That is, pI substitutions that alter the isoelectric point (pI) of each monomer result in each monomer having a different pI and the heterodimer also having a different pI, thereby facilitating isoelectric purification of the “triple F” heterodimer (e.g., anion exchange column, cation exchange column). These substitutions also help identify and monitor any contamination of the double scFv-Fc and purified mAh homodimers (e.g., IEF gel, cIEF, and analytical IEX column).

[0505] IX. Biological and Biochemical Functions of Heterodimeric Checkpoint Antibodies Generally, the bispecific checkpoint antibodies of the present invention are administered to patients with cancer and their efficacy is evaluated in a variety of ways as described herein. Therefore, immuno-oncology therapy can be estimated based on immune status assessments, where standard efficacy assays such as cancer burden, tumor size, assessment of presence or metastasis can be performed. This can be done in a variety of ways, including both in vitro and in vivo assays. For example, assessments of altered immune status (e.g., presence of ICOS+CD4+ T cells after IPI treatment) and "old-fashioned" measurements such as tumor burden, size, invasiveness, LN involvement, metastasis, etc., can be performed. Therefore, any one or all of the following can be evaluated: the inhibitory effect of checkpoints on CD4+ T cell activation or proliferation, CD8+ T (CTL) cell activation or proliferation, CD8+ T cell-mediated cytotoxic activity and / or CTL-mediated cell exhaustion, NK cell activity and NK-mediated cell exhaustion, the enhancing effect of checkpoints on Treg cell differentiation and proliferation and Treg-derived or myeloid-derived suppressor cell (MDSC)-mediated immunosuppression or immune tolerance, and / or the effect of checkpoints on pro-inflammatory cytokines produced by immune cells, such as IL-2, IFN-γ or TNF-α produced by T cells or other immune cells.

[0506] In some embodiments, treatment estimation is performed by assessing immune cell proliferation using methods such as CFSE dilution, Ki67 intracellular staining of immune effector cells, and 3H-thymidine incorporation assays. In some embodiments, treatment estimation is performed by assessing increased protein levels of gene expression or activation-related markers, including one or more of the following: CD25, CD69, CD137, ICOS, PD1, GITR, OX40, and cell degranulation as measured by surface expression of CD107A.

[0507] Typically, gene expression assays are performed, as is known in the art.

[0508] Typically, as is known in the art, protein expression measurements are also performed similarly.

[0509] In some embodiments, therapeutic estimation is performed by assessing various cellular parameters such as enzyme activity (including protease activity), cell membrane permeability, cell adhesion, ATP production, coenzyme production, and nucleotide uptake activity to estimate cytotoxic activity as measured by target cell viability assays. Specific examples of these assays include, but are not limited to, Trypan Blue or PI staining, 51Cr or 35S release methods, LDH activity, MTT and / or WST assays, calcein-AM assays, luminescent matrix assays, and other assays.

[0510] In some embodiments, treatment estimation is performed using well-known techniques by estimating T cell activity measured by cytokine production, using intracellular measurements in culture supernatants with cytokines including, but not limited to, IFNγ, TNFα, GM-CSF, IL2, IL6, IL4, IL5, IL10, and IL13.

[0511] Therefore, treatment can be evaluated using one or more of the following assays: (i) increased immune response; (ii) increased activation of αβ and / or γδ T cells; (iii) increased cytotoxic T cell activity; (iv) increased NK and / or NKT cell activity; (v) remission of αβ and / or γδ T cell suppression; (vi) increased secretion of pro-inflammatory cytokines; (vii) increased IL-2 secretion; (viii) increased interferon-γ production; (ix) increased Th1 response; (x) decreased Th2 response; (xi) decreased number and / or activity of at least one of regulatory T cells (Tregs).

[0512] Determination of efficacy

[0513] In some embodiments, mixed lymphocyte response (MLR) assays, as known in the art, are used to estimate T cell activation. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are outlined below.

[0514] In one embodiment, signal transduction pathway assays measure increases or decreases in immune responses, such as by phosphorylation or dephosphorylation of different factors or by other post-translational modifications. Increased activity indicates immune-stimulating activity. Appropriate increases in activity are outlined below.

[0515] In one embodiment, signal transduction pathway assays measure the activation of αβ and / or γδ T cells by means of increased or decreased activity, such as through cytokine secretion, proliferation, or altered expression of activation markers such as CD137, CD107a, PD1, etc. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are outlined below.

[0516] In one embodiment, signal transduction pathway assays measure an increase or decrease in cytotoxic T cell activity, such as by direct killing of target cells (e.g., cancer cells), by cytokine secretion, by proliferation, or by altering the expression of activation markers such as CD137, CD107a, PD1, etc. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are outlined below.

[0517] In one embodiment, signal transduction pathway assays measure an increase or decrease in NK and / or NKT cell activity, such as by direct killing of target cells, such as cancer cells, or by cytokine secretion, or by altering the expression of activation markers, such as CD107a. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are outlined below.

[0518] In one embodiment, signal transduction pathway assays measure an increase or decrease in αβ and / or γδ T cell suppression, such as through cytokine secretion, proliferation, or alteration of the expression of activation markers such as CD137, CD107a, PD1, etc. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are summarized below.

[0519] In one embodiment, signal transduction pathway assays measure an increase or decrease in the secretion of pro-inflammatory cytokines, such as by ELISA, Luminex, bead-based methods, intracellular staining and FACS analysis, or methods like Alispot. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are outlined below.

[0520] In one embodiment, signal transduction pathway assays measure an increase or decrease in IL-2 secretion, such as by ELISA, Luminex, bead-based methods, intracellular staining and FACS analysis, or methods like Alispot. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are outlined below.

[0521] In one embodiment, signal transduction pathway assays measure an increase or decrease in interferon-γ production, such as by ELISA, Luminex, bead-based methods, intracellular staining and FACS analysis, or measurements by Alispot, etc. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are outlined below.

[0522] In one embodiment, signal transduction pathway assays measure an increase or decrease in Th1 response, such as through cytokine secretion or by altering the expression of activation markers. Increased activity indicates immune stimulation activity. Appropriate increases in activity are outlined below.

[0523] In one embodiment, signal transduction pathway assays measure an increase or decrease in the Th2 response, such as through cytokine secretion or by altering the expression of activation markers. Increased activity indicates immune stimulation activity. Appropriate increases in activity are outlined below.

[0524] In one embodiment, signal transduction pathway assays measure an increase or decrease in the number and / or activity of one of the regulatory T cells (Tregs), such as by flow cytometry or by IHC. A decrease in response indicates immune stimulation activity. Appropriate decreases, as outlined below, are the same as increases.

[0525] In one embodiment, signal transduction pathway assays measure an increase or decrease in the number of M2 macrophages, such as by flow cytometry or by IHC. A decrease in response indicates immune stimulation activity. Appropriate decreases, as outlined below, are equivalent to increases.

[0526] In one embodiment, signal transduction pathway assays measure increases or decreases in M2 macrophage oncogene activity, such as through cytokine secretion or by altering the expression of activation markers. A decrease in response indicates immune stimulation activity. Appropriate decreases, as outlined below, are equivalent to increases.

[0527] In one embodiment, signal transduction pathway assays measure an increase or decrease, such as an increase in N2 neutrophils as measured, for example, by flow cytometry or by IHC. A decrease in response indicates immune stimulation activity. Appropriate decreases are the same as increases, as outlined below.

[0528] In one embodiment, signal transduction pathway assays measure increases or decreases in N2 neutrophil-promoting oncogene activity, such as through cytokine secretion or by altering the expression of activation markers. A decrease in response indicates immune stimulation activity. Appropriate decreases, as outlined below, are equivalent to increases.

[0529] In one embodiment, signal transduction pathway assays measure an increase or decrease in inhibition of T cell activation, such as through cytokine secretion, proliferation, or alteration of the expression of activation markers such as CD137, CD107a, PD1, etc. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are summarized below.

[0530] In one embodiment, signal transduction pathway assays measure an increase or decrease in inhibition of CTL activation, such as by direct killing of target cells, such as cancer cells, or by cytokine secretion, or by proliferation, or by altering the expression of activation markers such as CD137, CD107a, PD1, etc. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are summarized below.

[0531] In one embodiment, signal transduction pathway determination measures an increase or decrease, such as by altering the expression of an activation marker, measuring αβ and / or [other parameters]. γ δT cell depletion. A reduced response indicates immune stimulation activity. Appropriate reductions are equivalent to increases, as outlined below.

[0532] In one embodiment, signal transduction pathway assays measure an increase or decrease in αβ and / or γδ T cell responses, such as through cytokine secretion, proliferation, or alteration of the expression of activation markers such as CD137, CD107a, PD1, etc. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are outlined below.

[0533] In one embodiment, signal transduction pathway assays measure the stimulation of an antigen-specific memory response by means of, for example, cytokine secretion, proliferation, or alteration of the expression of activation markers such as, for example, CD45RA, CCR7, etc. Increased activity indicates immunostimulatory activity. Appropriate increases in activity are outlined below.

[0534] In one embodiment, signal transduction pathway assays measure an increase or decrease in apoptosis or lysis of cancer cells, such as by cytotoxicity assays such as MTT, Cr release, calcein AM, or by flow cytometry-based assays such as CFSE dilution or ...

Claims

1. A heterodimeric antibody, comprising: a) The first chain, which includes: i) The first variant Fc structural domain; and ii) a single-chain Fv region (scFv) that binds to the first antigen, wherein the scFv region includes a first variable heavy structure domain, a first variable light structure domain, and a charged scFv linker, wherein the charged scFv linker covalently connects the first variable heavy structure domain and the variable light structure domain; and b) A second heavy chain comprising a VH-CH1-hinge-CH2-CH3 monomer, wherein VH is a second variable heavy structure domain and CH2-CH3 is a second variant Fc structure domain; and c) A light chain comprising a second variable light structural domain and a light constant structural domain; The second variant's Fc domain includes amino acid substitutions of N208D / Q295E / N384D / Q418E / N241D, wherein... The first variant Fc domain and the second variant Fc domain each comprise an amino acid substitution E233P / L234V / L235A / G236del / S267K; wherein the first variant Fc domain comprises an amino acid substitution S364K / E357Q, and the second variant Fc domain comprises an amino acid substitution L368D / K370S, wherein the first variable heavy domain and the first variable light domain are selected from the group consisting of: SEQ ID NO: 11376 and SEQ ID NO: 11377, SEQ ID NO: 22970 and SEQ ID NO: 22971, SEQ ID NO: 11394 and SEQ ID NO: 11395, SEQ ID NO: 11367 and SEQ ID NO: 11368, and SEQ ID NO: 11412 and SEQ ID NO: 11413, wherein the numbering is based on the EU index as in Kabat.

2. The heterodimeric antibody according to claim 1, wherein the CH1-hinge-CH2-CH3 component of the second heavy chain has SEQ ID NO: 37725, the first variant Fc domain has SEQ ID NO: 37726, and the constant light domain has SEQ ID NO: 37727.

3. The heterodimeric antibody according to claim 1 or 2, wherein the second variable heavy domain and the second variable light domain form an antigen-binding domain, the antigen-binding domain binding to human checkpoint receptors from the group of human CTLA-4, human LAG-3, human TIM-3 and human TIGIT.

4. The heterodimeric antibody according to claim 1, 2 or 3, wherein the first variable heavy structure domain has SEQ ID NO: 11394, and the first variable light structure domain has SEQ ID NO: 11395.

5. The heterodimeric antibody according to claim 1, 2, 3 or 4, wherein the first heavy chain has SEQ ID NO: 23581, the second heavy chain has SEQ ID NO: 23576, and the light chain has SEQ ID NO: 23591.

6. A nucleic acid composition, correspondingly comprising: a) A first nucleic acid that encodes the first heavy strand as described in claims 1 to 5; b) A second nucleic acid that encodes the second heavy strand as described in claims 1 to 5; and c) A third nucleic acid that encodes the light chain as described in claims 1 to 5.

7. An expression vector composition comprising: a) A first expression vector comprising the first nucleic acid as described in claim 6; b) A second expression vector comprising the second nucleic acid as described in claim 6; and c) A third expression vector comprising the third nucleic acid as described in claim 6.

8. A host cell comprising the expression vector composition of claim 7.

9. A method for producing a heterodimeric antibody according to claims 1 to 5, comprising culturing a host cell according to claim 8 under conditions expressing the antibody, and recovering the antibody.

Images