Antibody that binds to human programmed death ligand 1 (PD-L1)

Abstract

The present invention provides isolated antibodies that specifically bind to human programmed death ligand 1 (PD-L1), antigen-binding fragments of such antibodies, and a kit comprising the anti-PD-L1 antibody or binding fragment and a set of reagents of a complex for detecting the antibody or its antigen-binding fragment that binds to human PD-L1. The antibodies and antigen-binding fragments of the present invention are suitable for immunohistochemical detection of human PD-L1 expression in tissue samples. Also provided are nucleic acid molecules encoding the antibodies and antigen-binding fragments of the present invention, and expression vectors and host cells for their expression.

Description

[0001] This application is a divisional application of Chinese Patent Application CN201380065289.9, "Antibodies that Bind to Human Programmed Death Ligand 1 (PD-L1)", with a filing date of December 18, 2013. Technical Field

[0002] The present invention relates to antibodies having specific sequences that bind to human programmed death ligand 1 (PD-L1) and are suitable for detecting PD-L1 expression in human tissue samples by immunohistochemical (IHC) analysis. The present invention also relates to specific IHC assays employing these anti-human-PD-L1 antibodies. Background Art

[0003] PD-L1 is a cell surface glycoprotein that is one of two known ligands for programmed death 1 (PD-1) and is thought to play an important role in immune regulation and the maintenance of peripheral tolerance. Expression of PD-L1 has been observed on the surface of many immune cells, including untreated lymphocytes and activated B and T cells, monocytes, and dendritic cells (ibid.). In addition, PD-L1 mRNA is expressed by non-lymphoid tissues, including vascular endothelial cells, epithelial cells, and myocytes, as well as in tonsil and placenta tissues. See, for example, Keir, M.E. et al., Annu Rev Immunol. 26:677-704 (2008); Sharp A.H. et al., Nature Immunol. 8:239-245 (2007); Okazaki T and HonjoT, Internat. immunol. 19:813-824 (2007).

[0004] PD-L1 expression has also been observed in many human cancers, and the interaction of PD-L1 expressed by tumor cells with PD-1 can induce suppression or apoptosis of tumor-specific T cells. In a large number of samples of, for example, ovarian cancer, renal cancer, colorectal cancer, pancreatic cancer, liver cancer, and melanoma, PD-L1 expression has been confirmed to be associated with poor prognosis and short overall survival, regardless of subsequent treatment. Anti-PD-1 monoclonal antibodies that block the binding of PD-L1 to PD-1 have been shown to have anti-tumor activity against many tumor types, and early human clinical data indicate that patients with tumors expressing PD-L1 are more likely to respond to anti-PD-1 therapy. See, for example, Iwai et al., PNAS 99:12293-12297 (2002); Ohigashi et al., Clin Cancer Res 11:2947-2953 (2005); Ghebeh et al., Neoplasia8:190 - 198 (2006); Hamanishi, J et al., PNAS 104:3360 - 3365 (2007); Yang et al., Invest Ophthalmol Vis Sci . 49(6):2518 - 2525 (2008); Gao et al., Clin Cancer Res 15:971 - 979 (2009); Brahmer J.R. et al., J Clin Oncol. 28:3167 - 3175 (2010).

[0005] Recent reports have described a comparison of the utility of 15 anti - human PD - L1 antibodies for detecting hPD - L1 expression in formalin - fixed paraffin - embedded (FFPE) human melanoma samples (Gadiot, J., et al., Cancer 117(10):2192 - 2201(2011)). The utility criteria evaluated in this comparison were: (1) the ability to stain paraffin - embedded tissue, (2) to produce low background staining and (3) to block binding to PD - L1 by pre - incubation with a PD - L1 fusion protein. The authors inferred that Ab #4059, a rabbit anti - human polyclonal antibody (obtained from ProSci, Poway, CA USA), was the only anti - human PD - L1 antibody among the 15 tested antibodies that acceptably met all of these criteria (ibid 2195, second column). SUMMARY OF THE INVENTION

[0006] The present invention relates to a monoclonal anti - human PD - L1 antibody that produces an IHC staining pattern in FFPE tonsil tissue that the inventors herein believe is more immunologically relevant compared to the staining pattern produced by ProSci Ab #4059. As described in the Examples below, the inventors found that this ProSci antibody (PRS4059, Sigma - Aldrich lot number 40590604) stained all hematopoietic lineages in the tonsil with equal intensity, while the two antibodies of the present invention, 22C3 and 20C3, selectively stained tonsillar crypt epithelial cells and follicular CD68+ myeloid cells (morphologically consistent with macrophages). In addition, 22C3 and 20C3 showed a consistent intensity difference between these two individual cell populations, with the staining intensity in crypt epithelial cells being significantly greater than that in follicular macrophages. All three antibodies (PRS4059, 22C3, and 20C3) were neutralized by pre - incubation with the PD - L1 antigen, indicating that the reactivity was mediated by the antigen - binding domain (CDR). Accordingly, the present invention also relates to the use of the antibodies of the present invention in an IHC assay for detecting PD - L1 expression on the surface of human cells, including detection of PD - L1 in FFPE tissue sections.

[0007] In one aspect, the present invention provides an isolated antibody or antigen-binding fragment thereof that specifically binds to human PD-L1. The isolated monoclonal antibody or antigen-binding fragment thereof comprises three light chain CDRs of CDRL1, CDRL2, and CDRL3 and three heavy chain CDRs of CDRH1, CDRH2, and CDRH3.

[0008] CDRL1 is selected from: SEQ ID NO: 1, SEQ ID NO: 9, SEQ ID NO: 21, variants of SEQ ID NO: 9, and variants of SEQ ID NO: 21. CDRL2 is selected from SEQ ID NO: 2 and variants of SEQ ID NO: 2. CDRL3 is selected from: SEQ ID NO: 3, SEQ ID NO: 10, SEQ ID NO: 22, variants of SEQ ID NO: 10, and variants of SEQ ID NO: 22.

[0009] CDRH1 is selected from: SEQ ID NO: 5, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 26, SEQ ID NO: 27, variants of SEQ ID NO: 14, variants of SEQ ID NO: 15, variants of SEQ ID NO: 26, and variants of SEQ ID NO: 27. CDRH2 is selected from: SEQ ID NO: 6, SEQ ID NO: 16, SEQ ID NO: 28, variants of SEQ ID NO: 16, and variants of SEQ ID NO: 28. CDRH3 is selected from: SEQ ID NO: 7, SEQ ID NO: 17, SEQ ID NO: 29, variants of SEQ ID NO: 17, and variants of SEQ ID NO: 29.

[0010] In the antibodies and antigen-binding fragments of the present invention, the variant CDR sequences (light or heavy chain) are identical to the reference sequences except having one or two conservative amino acid substitutions in the reference sequences, and preferably having only one conservative amino acid substitution in the reference sequences. In a preferred embodiment, at most two of the three light chain CDRs are variant sequences, and at most two of the three heavy chain CDRs are variant sequences. In a more preferred embodiment, only three, two, or one of the six CDRs are variant sequences.

[0011] In a preferred embodiment, the three light chain CDRs are SEQ ID NO: 1, SEQ ID NO: 2, and SEQ ID NO: 3, and the three heavy chain CDRs are SEQ ID NO: 5, SEQ ID NO: 6, and SEQ ID NO: 7.

[0012] In another preferred embodiment, the three light chain CDRs are SEQ ID NO: 9, SEQ ID NO: 2 and SEQ ID NO: 10 and the three heavy chain CDRs are SEQ ID NO: 14, SEQ ID NO: 16 and SEQ ID NO: 17.

[0013] In yet another preferred embodiment, the three light chain CDRs are SEQ ID NO: 21, SEQ ID NO: 2 and SEQ ID NO: 22 and the three heavy chain CDRs are SEQ ID NO: 26, SEQ ID NO: 28 and SEQ ID NO: 29.

[0014] Some antibodies and antigen-binding fragments of the present invention comprise a light chain variable region and a heavy chain variable region. The light chain variable region is selected from: SEQ ID NO: 4, SEQ ID NO: 13, variants of SEQ ID NO: 13, SEQ ID NO: 25 and variants of SEQ ID NO: 25, and the heavy chain variable region is selected from: SEQ ID NO: 8, SEQ ID NO: 20, variants of SEQ ID NO: 20, SEQ ID NO: 32 and variants of SEQ ID NO: 32. In such embodiments, the variant light chain variable region sequence is identical to the reference sequence except having up to five conservative amino acid substitutions in the framework regions (i.e., outside the CDRs), and preferably having less than four, three or two conservative amino acid substitutions in the framework regions. Similarly, the variant heavy chain variable region sequence is identical to the reference sequence except having up to 17 conservative amino acid substitutions in the framework regions (i.e., outside the CDRs), and preferably having less than ten, nine, eight, seven, six or five conservative amino acid substitutions in the framework regions.

[0015] In a preferred antibody or antigen-binding fragment of the present invention, the light chain variable region is SEQ ID NO: 13 and the heavy chain variable region is SEQ ID NO: 20.

[0016] Another preferred antibody or antigen-binding fragment of the present invention comprises a light chain variable region of SEQ ID NO: 25 and a heavy chain variable region of SEQ ID NO: 32.

[0017] In yet another embodiment, the antibody or binding fragment of the present invention comprises a light chain variable region of SEQ ID NO: 25 and a heavy chain variable region of SEQ ID NO: 32, wherein X in SEQ ID NO: 32 is pE.

[0018] In another embodiment, the antibody or binding fragment of the invention comprises the light chain variable region of SEQ ID NO: 25 and the heavy chain variable region of SEQ ID NO: 32, wherein X in SEQ ID NO: 32 is Q.

[0019] In all of the above antibody embodiments, the isolated antibody can be a full-length antibody of any type of immunoglobulin, including IgM, IgG, IgD, IgA, and IgE. The antibody is preferably an IgG antibody, such as IgG 1 、IgG 2 、IgG 3 or IgG 4 。In one embodiment, the antibody comprises a murine IgG 1 constant region.

[0020] Particularly preferred antibodies are monoclonal antibodies 20C3 and 22C3, which are IgG1 antibodies expressed by hybridomas MEB037.20C3 and MEB037.22C3, respectively.

[0021] The invention also provides an isolated monoclonal antibody or its antigen-binding fragment that specifically binds to human PD-L1 and blocks the binding of 20C3 or 22C3 or a reference antibody comprising SEQ ID NO: 25 and SEQ ID NO: 32 to human PD-L1. In a preferred embodiment, the antibody or antigen-binding fragment of the invention blocks the binding of each of 20C3 and 22C3 or each of the following to human PD-L1: (a) a reference antibody comprising SEQ ID NO: 13 and SEQ ID NO: 20 and (b) a reference antibody comprising SEQ ID NO: 25 and SEQ ID NO: 32.

[0022] The invention also provides an antibody composition that comprises any one of the above antibodies or antibody fragments in a formulation. A suitable formulation comprises 20 mM sodium acetate and 9% sucrose (pH 5.0). In a preferred embodiment, the composition comprises a mixture of antibody molecules, wherein most (i.e., any one of greater than 60%, 65%, 70%, 80%, 85%, 90%, or 95%) of the antibody molecules in the mixture comprise SEQ ID NO: 25 and SEQ ID NO: 32, wherein X in SEQ ID NO: 32 is pE, and the remaining antibody molecules in the mixture comprise SEQ ID NO: 25 and SEQ ID NO: 32, wherein X in SEQ ID NO: 32 is Q.

[0023] In any of the above embodiments, the antigen-binding fragment is a Fab fragment, a Fab' fragment, a (Fab') 2 fragment.

[0024] In any of the above embodiments, the antibody or antigen-binding fragment may further comprise a detectable label.

[0025] The present invention also provides an isolated nucleic acid encoding an antibody variable region disclosed above. In a preferred embodiment, the nucleic acid comprises one or both of SEQ ID NO: 33 and SEQ ID NO: 34. In another preferred embodiment, the nucleic acid comprises one or both of SEQ ID NO: 35 and SEQ ID NO: 36. In any of these embodiments, the isolated nucleic acid is preferably an expression vector.

[0026] The present invention also relates to a host cell comprising an expression vector encoding an antibody variable region disclosed above. The expression vector preferably comprises SEQ ID NO: 33 and SEQ ID NO: 34 or comprises SEQ ID NO: 35 and SEQ ID NO: 36.

[0027] The present invention also provides a method for analyzing PD-L1 expression in a human tissue sample removed from a human. The analysis method includes contacting the tissue sample with a PD-L1 binding reagent under conditions that allow the PD-L1 binding reagent to specifically bind to human PD-L1, removing the unbound PD-L1 binding reagent, and detecting the presence or absence of the bound PD-L1 binding agent. In a preferred embodiment, the method further includes quantifying the amount of the bound binding reagent. The binding reagent is any one of the monoclonal antibodies or antigen-binding fragments described above. The binding reagent is preferably an antibody comprising SEQ ID NO: 13 and SEQ ID NO: 20 or comprising SEQ ID NO: 25 and SEQ ID NO: 32. In a preferred embodiment, the binding reagent is an antibody composition containing a mixture of antibody molecules comprising SEQ ID NO: 25 and SEQ ID NO: 32, wherein most of the molecules (i.e., any one of greater than 60%, 65%, 70%, 80%, 85%, 90% or 95%) have pE at position X in SEQ ID NO: 32, and the remaining molecules have Q at position X in SEQ ID NO: 32.

[0028] In another aspect, the present invention provides a kit for analyzing the expression of PD-L1 in a human tissue sample. The kit comprises a PD-L1 binder and a set of reagents for detecting a complex comprising a binder that binds to human PD-L1. The PD-L1 binder is any of the monoclonal antibodies or antigen-binding fragments described above that specifically bind to human PD-L1. The antibody or binding fragment preferably comprises SEQ ID NO: 13 and SEQ ID NO: 20, or comprises SEQ ID NO: 25 and SEQ ID NO: 32. In a preferred embodiment, the binding reagent is an antibody composition that contains a mixture of antibody molecules comprising SEQ ID NO: 25 and SEQ ID NO: 32, wherein most of the molecules (i.e., any one of greater than 60%, 65%, 70%, 80%, 85%, 90% or 95%) have pE at position X in SEQ ID NO: 32, and the remaining molecules have Q at position X in SEQ ID NO: 32. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Shows the nucleotide sequences of the antibody variable light and heavy chain cDNAs prepared from total RNA isolated from the hybridoma MEB037.20C3 and the predicted amino acid sequences (bold) encoded thereby, with the brackets indicating the nucleotide and amino acid sequences of the leader peptide and the underlines indicating the CDR sequences.

[0030] Figure 2 Shows the nucleotide sequences of the antibody variable light and heavy chain cDNAs prepared from total RNA isolated from the hybridoma MEB037.22C3 and the predicted amino acid sequences (bold) encoded thereby, wherein the brackets indicate the nucleotide and amino acid sequences of the leader peptide and the underlines indicate the CDR sequences.

[0031] Figure 3 Shows the aligned amino acid sequences of the mature variable regions of the light and heavy chains of antibodies 20C3 and 22C3, wherein the positions where the sequences vary are indicated in bold, the CDR sequences as defined by the Kabat numbering system are underlined, and the heavy chain CDR1 as defined by the Chothia numbering system is indicated in brackets.

[0032] Figure 4 shows the staining of tonsil sections produced by immunohistochemical analysis using a commercially available antibody PRS4059 ( Figure 4A ) or the 22C3 antibody of the present invention ( Figure 4B ), wherein Figure 4A and Figure 4B the sections on the right show the results after pre-incubation with a PD-L1-IgG1 fusion protein (R&D Systems) that competes with the anti-human PD-L1 antibody for binding to human PD-L1.

[0033] Figure 5 shows photographs of adjacent normal FFPE tonsil tissue sections, where human PD-L1 protein and in situ hybridization (ISH) mRNA expression were analyzed by IHC using antibody 22C33( Figure 5A ) and in situ hybridization (ISH)( Figure 5B ), respectively, and show differential staining between two distinct cell populations: crypt epithelial cells( Figure 5A , left magnified view and Figure 5B , upper panel) and follicular macrophages( Figure 5A , right magnified view and Figure 5B , lower panel).

[0034] Figure 6 illustrates the results of flow cytometric evaluation of the binding of multiple anti-human PD-L1 antibodies and isotype control antibodies to HT144 cells (known to be negative for hPD-L1 expression by mRNA analysis (qPCR))( Figure 6A ) and LOX melanoma cells (known to highly express hPD-L1 mRNA (qPCR))( Figure 6B ).

[0035] Figure 7 shows IHC staining generated by antibody 22C3 on FFPE cell pellets of engineered CHO cell lines( Figure 7A ) and human cell lines( Figure 7B , upper part of the figure), and shows that the staining intensity is significantly correlated with the hPD-L1 mRNA expression levels measured in the same human cell lines( Figure 7B , lower part of the figure).

[0036] Figure 8 illustrates the selective binding and relative affinity of antibodies 22C3 and 20C3 for hPD-L1, where the graph shows the results of a cell-based ELISA experiment in which cells not expressing hPD-L1( Figure 8A ), expressing hPDL-1( Figure 8B and Figure 8C ) or expressing human PD-L2( Figure 8D ) were incubated with the indicated primary antibodies at the indicated concentrations and then the binding of the primary antibodies was detected by a secondary goat anti-human IgG antibody as described in the examples.

[0037] Figure 9A , Figure 9B and Figure 9C show the results of antibody-binding competition assays for antibodies 22C3, 20C3, and 5F9, respectively. The data show that antibodies 22C3 ( Figure 9A ) and 20C3 ( Figure 9B ) bind to different but overlapping epitopes.

[0038] Figure 10Describe the results of semi - quantitative gestalt scoring of 22C3 IHC staining intensity in FFPE samples from a specified tumor type, where the degree of staining increases with increasing score value.

[0039] Figure 11 shows that the expression of human PD - L1 detected by antibody 22C3 in the IHC assay is correlated with the response of melanoma patients to treatment with an anti - human PD - 1 antibody (MK - 3475), Figure 11A showing representative images of 22C3 - IHC staining described as positive, negative, or of indeterminate significance for hPD - L1 expression, and Figure 11B showing the number of patients with positive or negative responses and scored as positive or negative for hPD - L1 expression (including patients scored as of indeterminate significance) by the IHC assay. Detailed implementation

[0040] Abbreviations. The following abbreviations will be used throughout the specification and examples of the present invention:

[0041] ADCC Antibody - dependent cell - mediated cytotoxicity

[0042] CDC Complement - dependent cytotoxicity

[0043] CDR Complementary determining region in the immunoglobulin variable region defined using the Kabat numbering system, unless otherwise indicated

[0044] CHO Chinese hamster ovary

[0045] Clothia Al - Lazikani et al., JMB the antibody numbering system described in 273:927 - 948 (1997)

[0046] EC50 Concentration that produces 50% efficacy or binding

[0047] ELISA Enzyme - linked immunosorbent assay

[0048] FFPE Formalin - fixed paraffin - embedded

[0049] FR Antibody framework region: the immunoglobulin variable region excluding the CDR regions.

[0050] HRP Horseradish peroxidase

[0051] IFN Interferon

[0052] IC50 Concentration that produces 50% inhibition

[0053] IgG Immunoglobulin G

[0054] The Kabat immunoglobulin alignment and numbering system created by Elvin A. Kabat ((1991) Sequences of Proteins of Immunological Interest, 5th ed. Public Health Service, National Institutes of Health, Bethesda, Md.)

[0055] mAb or Mab or MAb Monoclonal antibody

[0056] MES 2-(N-Morpholino)ethanesulfonic acid

[0057] MOA Mechanism of action

[0058] NHS Normal human serum

[0059] PCR Polymerase chain reaction

[0060] pE Pyroglutamate

[0061] PK Pharmacokinetics

[0062] SEB Staphylococcus Enterotoxin B

[0063] TT Tetanus toxoid

[0064] V region A segment of the IgG chain, the sequence of which can vary between different antibodies. It extends to Kabat residue 109 in the light chain and Kabat residue 113 in the heavy chain.

[0065] VH Variable region of the immunoglobulin heavy chain

[0066] VK Variable region of the immunoglobulin κ light chain.

[0067] Definitions

[0068] For a better understanding of the present invention, certain technical and scientific terms are specifically defined below. Unless otherwise specifically defined elsewhere in this document, all other technical and scientific terms used herein have the meanings commonly understood by those of ordinary skill in the art to which this invention pertains.

[0069] Unless the context clearly dictates otherwise, the singular forms of words as used herein (including the appended claims), such as "a, an" and "the", include their corresponding plural referents.

[0070] Unless the context or an explicit indication otherwise dictates, "activation", when used in reference to a cell or receptor, means activation or treatment of the cell or receptor with a ligand. "Ligand" encompasses natural and synthetic ligands such as cytokines, cytokine variants, analogs, mutant proteins, and binding compounds derived from antibodies. "Ligand" also encompasses small molecules such as peptidomimetics of cytokines and peptidomimetics of antibodies. "Activation" can refer to cell activation as regulated by internal mechanisms as well as external or environmental factors.

[0071] The "activity" of a molecule can be described or can refer to the binding of the molecule to a ligand or receptor, catalytic activity; the ability to stimulate gene expression or cell signaling, differentiation, or maturation; antigenic activity, regulation of the activity of other molecules, etc. The "activity" of a molecule can also refer to the activity of regulating or maintaining cell - cell interactions (such as adhesion) or the activity of maintaining cell structure (such as the cell membrane or cytoskeleton). "Activity" can also mean specific activity (e.g., [catalytic activity] / [milligrams of protein] or [immunological activity] / [milligrams of protein]), concentration in a biological compartment, etc. "Activity" can refer to the regulation of components of the innate or adaptive immune system.

[0072] "Administration" and "treatment", when used in reference to an animal, human, experimental subject, cell, tissue, organ, or biological fluid, mean the contact of an exogenous pharmaceutical agent, therapeutic agent, diagnostic agent, or composition with the animal, human, subject, cell, tissue, organ, or biological fluid. Treatment of a cell encompasses contacting the reagent with the cell as well as contacting the reagent with a fluid that is in contact with the cell. "Administration" and "treatment" also mean in vitro and ex vivo treatment of one cell by a reagent, diagnostic binding compound, or by another cell. The term "subject" includes any living organism, preferably an animal, more preferably a mammal (such as a rat, mouse, dog, cat, rabbit), and most preferably a human.

[0073] "Treat / treating" means administering a therapeutic agent, such as a composition comprising any of the antibodies or antigen-binding fragments of the invention, internally or externally to an individual or patient suffering from or suspected of suffering from one or more disease symptoms, to which the agent is therapeutically active. Generally, the agent is administered in an amount effective to induce regression of the symptoms or to inhibit progression of the symptoms at any clinically measurable level to thereby alleviate one or more disease symptoms in the treated individual or population. The amount of the therapeutic agent effective to alleviate any particular disease symptom (also referred to as a "therapeutically effective amount") can vary depending on factors such as the disease condition, age, and weight of the patient, as well as the ability of the drug to elicit the desired response in the individual. Whether a disease symptom has been alleviated can be evaluated by any clinical measurement commonly used by a physician or other skilled healthcare provider to assess the severity level or progression status of the disease symptom. Although the embodiments of the invention (e.g., methods of treatment or articles) may not be effective in alleviating the target disease symptoms in every individual, they should alleviate the target disease symptoms in a statistically significant amount of individuals as determined by any statistical test known in the art, such as Student's t-test, chi 2 -test, the U-test according to Mann and Whitney, the Kruskal-Wallis test (H-test), the Jonckheere-Terpstra test, and the Wilcoxon test.

[0074] "Treatment" when used in reference to a human, veterinary, or research individual refers to therapeutic treatment as well as research and diagnostic applications. "Treatment" when used in reference to a human, veterinary, or research individual, or cell, tissue, or organ encompasses contact of the antibodies or antigen-binding fragments of the invention with the human or animal individual, cell, tissue, physiological compartment, or physiological fluid.

[0075] Anti-PD-L1 antibody

[0076] Antibody 20C3 is an antibody produced by hybridoma subclone MEB037.20C3.116.

[0077] Antibody 22C3 is an antibody produced by hybridoma subclone MEB037.22C3.138 and corresponds to the isotype S414R of murine IgG1. The N-terminal residue of the mature heavy chain of 22C3 is glutamine or pyroglutamate (pE), which is a common post-translational modification typically observed in monoclonal antibodies when the gene sequence encodes an N-terminal glutamine in the mature heavy or light chain.

[0078] The antibodies and antigen-binding fragments of the present invention bind to the mature form of human PD-L1 (lacking the pre-pro leader sequence, also known as the leader peptide) expressed on the surface of certain human cells. The terms "PD-L1" and "mature PD-L1" are used interchangeably herein and, unless otherwise indicated or apparent from the context, are understood to refer to the same molecule. The mature human PD-L1 molecule consists of amino acids 19-290 of the following sequence (SEQ ID NO: 37):

[0079]

[0080] .

[0081] The extracellular domain of mature human PD-L1 consists of the following sequence (SEQ ID NO: 38):

[0082]

[0083] .

[0084] As used herein, an anti-human PD-L1 antibody or an anti-hPD-L1 antibody refers to an antibody that specifically binds to human PD-L1. An antibody that "specifically binds to human PD-L1" or an antibody that "specifically binds to a polypeptide comprising the amino acid sequence of human PD-L1" is an antibody that exhibits preferential binding to human PD-L1 compared to other antigens, but this specificity does not require absolute binding specificity. An anti-hPD-L1 antibody is considered to have "specificity" for human PD-L1 if the binding of the anti-hPD-L1 antibody can, for example, determine the presence or absence of human PD-L1 in a sample without producing undue results such as false positives in an IHC diagnostic assay. The degree of specificity required for the anti-hPD-L1 antibodies of the present invention may depend on the intended use of the antibody and is defined at least by its suitability for the intended purpose. The antibodies or their binding fragments of the present invention bind to human PD-L1 with an affinity that is at least two-fold, preferably at least ten-fold, more preferably at least 20-fold, and most preferably at least 100-fold greater than the affinity for any non-PD-L1 protein. As used herein, an antibody is considered to specifically bind to a polypeptide comprising a given sequence if the antibody binds to a polypeptide comprising that sequence but does not bind to a protein lacking that sequence, such as mature human PD-L1 (in this case amino acids 19-290 of SEQ ID NO: 37). For example, an antibody that specifically binds to a polypeptide comprising amino acids 19-290 of SEQ ID NO: 37 may bind to the FLAG ® tagged form of amino acids 19-290 of SEQ ID NO: 37, but does not bind to other FLAG ® tagged proteins.

[0085] As used herein, the term "antibody" refers to any form of antibody that exhibits the desired biological activity. Thus, it is used in the broadest sense and specifically encompasses (but is not limited to) monoclonal antibodies (including full-length monoclonal antibodies), polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), humanized and fully human antibodies, chimeric antibodies, and camelid single-domain antibodies. A "parent antibody" is an antibody obtained by exposing the immune system to an antigen prior to modifying the antibody for a predetermined use, such as humanizing the antibody for use as a human therapeutic antibody.

[0086] As used herein, unless otherwise indicated, the term "antibody fragment" or "antigen-binding fragment" refers to an antigen-binding fragment of an antibody, i.e., an antibody fragment that retains the ability to specifically bind the antigen bound by the full-length antibody, such as a fragment that retains one or more CDR regions. Examples of antibody-binding fragments include (but are not limited to) Fab, Fab', F(ab') 2 and Fv fragments; bispecific antibodies; linear antibodies; single-chain antibody molecules, such as sc-Fv; nanobodies; and multispecific antibodies formed from antibody fragments.

[0087] A "Fab fragment" consists of a light chain and the C H 1 and variable regions of a heavy chain. The heavy chain of a Fab molecule cannot form a disulfide bond with another heavy chain molecule. A "Fab fragment" can be the product of papain cleavage of an antibody.

[0088] The "Fc" region contains two heavy chain fragments that comprise the C H 1 and C H 2 domains of the antibody. The two heavy chain fragments are held together by two or more disulfide bonds and hydrophobic interactions of the C H 3 domain.

[0089] A "Fab' fragment" contains a light chain and a portion or fragment of a heavy chain that contains the V H domain and the C H 1 domain, as well as the region between the C H 1 and C H 2 domains, such that an interchain disulfide bond can be formed between the two heavy chains of two Fab' fragments to form an F(ab') 2 molecule.

[0090] An "F(ab') 2 fragment" contains two light chains and two heavy chains that contain a portion of the constant region between the C H 1 and C H 2 domains, such that an interchain disulfide bond is formed between the two heavy chains. Thus, F(ab') 2The fragment consists of two Fab' fragments held together by a disulfide bond between the two heavy chains. "F(ab') 2 fragment" can be the product of pepsin cleavage of an antibody.

[0091] The "Fv region" contains the variable regions from the heavy and light chains, but lacks the constant regions.

[0092] The term "single-chain Fv" or "scFv" antibody refers to an antibody fragment that contains the V H and V L domains of an antibody, where these domains are present in a single polypeptide chain. Generally, the Fv polypeptide further contains a polypeptide linker between the V H and V L domains, which enables the scFv to form the required structure for antigen binding. For a review of scFv, see Pluckthun (1994) The Pharmacology of Monoclonal Antibodies, vol. 113, edited by Rosenberg and Moore. Springer-Verlag, New York, pp. 269-315. See also International Patent Application Publication No. WO 88 / 01649 and U.S. Patent Nos. 4,946,778 and 5,260,203.

[0093] A "domain antibody" is an immunofunctional immunoglobulin fragment that contains only the variable region of the heavy chain or the variable region of the light chain. In some cases, two or more V H regions are covalently joined to a peptide linker to produce a bivalent domain antibody. The two V H regions of the bivalent domain antibody can target the same or different antigens.

[0094] A "bivalent antibody" contains two antigen-binding sites. In some cases, the two binding sites have the same antigen specificity. However, a bivalent antibody can be bispecific (see below).

[0095] In certain embodiments, monoclonal antibodies also include camelid single-domain antibodies herein. See, for example, Muyldermans et al. (2001) Trends Biochem. Sci . 26:230; Reichmann et al. (1999) J. Immunol. Methods 231:25; WO 94 / 04678; WO 94 / 25591; U.S. Patent No. 6,005,079). In one embodiment, the present invention provides a single-domain antibody that contains two modified V H domains so as to form a single-domain antibody.

[0096] As used herein, the term "bispecific antibody" refers to a small antibody fragment having two antigen-binding sites, the fragment comprising a heavy-chain variable domain (V H -V L or V L -V H ) linked to a light-chain variable domain (V L ) in the same polypeptide chain (V H ). By using a linker that is too short to allow pairing between the two domains on the same chain, the domains are forced to pair with the complementary domain on the other chain and create two antigen-binding sites. Bispecific antibodies are described more fully, for example, in EP 404,097; WO 93 / 11161; and Holliger et al. . (1993) Proc. Natl. Acad. Sci. USA 90: 6444-6448. For a review of engineered antibody variants, see generally Holliger and Hudson (2005) Nat. Biotechnol . 23:1126-1136.

[0097] Generally, an antibody or antigen-binding fragment of the invention retains at least 10% of its human PD-L1 binding activity (when compared to the parental antibody) when its human PD-L1 binding activity is expressed in molar concentration. The antibody or antigen-binding fragment of the invention preferably retains at least 20%, 50%, 70%, 80%, 90%, 95%, or 100% or more of the human PD-L1 binding affinity of the parental antibody. The antibody or antigen-binding fragment of the invention is also intended to optionally include conservative or non-conservative amino acid substitutions that do not substantially alter its biological activity (referred to as "conservative variants" or "functionally conservative variants" of the antibody).

[0098] "Isolated antibody" refers to a purified state and in this context means that the molecule is substantially free of other biomolecules, such as nucleic acids, proteins, lipids, carbohydrates, or other substances (such as cell debris and growth medium). Generally, the term "isolated" is not intended to mean the complete absence of such substances or the absence of water, buffer, or salt, unless present in an amount that substantially interferes with the experimental or therapeutic use of the binding compound as described herein.

[0099] As used herein, the term "monoclonal antibody" refers to a substantially homogeneous population of antibodies, i.e., the amino acid sequences of the antibody molecules comprising the population are identical except for possible naturally occurring mutations present in minor amounts. In contrast, conventional (polyclonal) antibody preparations typically include a large number of different antibodies that differ in their variable domains, particularly in their CDRs, and that usually have specificities for different epitopes. The modifier "monoclonal" indicates the character of the antibody as being obtained from a substantially homogeneous population of antibodies and should not be construed as requiring that the antibody be made by any particular method. For example, monoclonal antibodies used in accordance with the present invention may be made by the hybridoma method first described by Kohler et al., (1975) Nature 256:495, or may be made by recombinant DNA methods (see, e.g., U.S. Patent No. 4,816,567). "Monoclonal antibodies" may also be isolated using, for example, the techniques described in Clackson et al. . (1991) Nature 352: 624-628 and Marks et al. . (1991) J. Mol. Biol . 222:581-597 and isolated from phage antibody libraries. See also Presta (2005) J. Allergy Clin. Immunol . 116:731.

[0100] Generally, the basic antibody structural unit comprises a tetramer. Each tetramer includes two identical pairs of polypeptide chains, each pair having one "light" chain (about 25 kDa) and one "heavy" chain (about 50-70 kDa). The amino-terminal portion of each chain includes a variable region of about 100 to 110 or more amino acids that is primarily responsible for antigen recognition. The carboxyl-terminal portion of the heavy chain may define the constant region that is primarily responsible for effector functions. Human light chains are usually classified as kappa and lambda light chains. In addition, human heavy chains are usually classified as mu, delta, gamma, alpha, or epsilon, and the isotypes of antibodies are defined as IgM, IgD, IgG, IgA, and IgE, respectively. Within light and heavy chains, the variable and constant regions are joined by a "J" region of about 12 or more amino acids, and the heavy chain also includes a "D" region of about 10 or more amino acids. See generally Fundamental Immunology Chapter 7 (Paul, W., ed., 2nd ed. Raven Press, N.Y. (1989).

[0101] The variable regions of each light chain / heavy chain pair form the antibody binding site. Thus, generally, an intact antibody has two binding sites. Unless in a bifunctional or bispecific antibody, the two binding sites are generally the same.

[0102] Generally, the variable domains of the heavy and light chains contain three hypervariable regions, also known as complementarity-determining regions (CDRs), located within relatively conserved framework regions (FRs). The CDRs are typically aligned by the framework regions, enabling binding to specific epitopes. Generally, from the N-terminus to the C-terminus, the variable domains of the light and heavy chains both contain FR1, CDR1, FR2, CDR2, FR3, CDR3, and FR4. Generally according to Sequences of Proteins of Immunological Interest , Kabat, et al.; National Institutes of Health, Bethesda, Md.; 5th Edition; NIH Publ. No. 91-3242 (1991); Kabat (1978) Adv. Prot. Chem. 32:1-75; Kabat, et al., (1977) J. Biol. Chem. 252:6609-6616; Chothia, et al., (1987) J Mol. Biol. 196:901-917 or Chothia, et al., (1989) Nature 342:878-883, amino acids are classified into each domain.

[0103] As used herein, the term "hypervariable region" refers to the amino acid residues of an antibody responsible for antigen binding. The hypervariable region contains the amino acid residues from the "complementarity-determining region" or "CDR" (i.e., CDRL1, CDRL2, and CDRL3 in the light chain variable domain and CDRH1, CDRH2, and CDRH3 in the heavy chain variable domain). See Kabat et al. . (1991) Sequences of Proteins of Immunological Interest, 5th Edition. Public Health Service, National Institutes of Health, Bethesda, Md. (defining the CDR regions of antibodies by sequence); see also Chothia and Lesk (1987) J. Mol. Biol . 196: 901-917 (defining the CDR regions of antibodies by structure). As used herein, the term "framework" or "FR" residues refers to those variable domain residues other than the hypervariable region residues defined herein as CDR residues.

[0104] "Homology" refers to sequence similarity between two polynucleotide sequences or between two polypeptide sequences upon optimal alignment. When positions in the two compared sequences are occupied by the same base or amino acid monomer subunit, e.g., if the position in each of two DNA molecules is occupied by adenine, then the molecules are homologous at that position. The percentage of homology is the number of homologous positions shared by the two sequences divided by the total number of comparison positions × 100. For example, if 8 of 10 positions in two sequences match or are homologous upon optimal alignment of the sequences, then the two sequences are 80% homologous. In general, comparisons are made in sequence alignment to obtain the maximum percentage of homology. For example, comparisons can be made by the BLAST algorithm, where the parameters of the algorithm are selected to obtain the maximum match between the individual sequences over the entire length of the respective reference sequences.

[0105] "Isolated nucleic acid molecule" means genomic DNA or RNA, mRNA, cDNA, or its synthetic source or some combination thereof that is not associated with all or a portion of the polynucleotide, where the isolated polynucleotide is found in nature or linked to polynucleotides not linked in nature. For the purposes of this invention, it is understood that a "nucleic acid molecule comprising a specific nucleotide sequence" does not encompass an entire chromosome. An isolated nucleic acid molecule "comprising" the designated nucleic acid sequence may further include coding sequences for up to 10 or even up to 20 or more other proteins or portions or fragments thereof, or may include operably linked regulatory sequences that control the expression of the coding region of the nucleic acid sequence, and / or may include vector sequences.

[0106] The phrase "control sequence" means a DNA sequence required for the expression of an operably linked coding sequence in a particular host organism. Control sequences suitable for prokaryotes include, for example, a promoter, an optional operator sequence, and a ribosome binding site. It is known that eukaryotic cells use promoters, polyadenylation signals, and enhancers.

[0107] A nucleic acid is "operably linked" when it is in a functional relationship with another nucleic acid sequence. For example, if the DNA of a presequence or secretory leader sequence is expressed as a precursor protein involved in the secretion of a polypeptide, then it is operably linked to the DNA of that polypeptide; if a promoter or enhancer affects the transcription of a coding sequence, then it is operably linked to that sequence; or if a ribosome binding site is positioned so as to facilitate translation, then it is operably linked to the coding sequence. In general, "operably linked" means that the DNA sequences being linked are contiguous and, in the case of a secretory leader sequence, contiguous and in the reading phase. However, enhancers do not have to be contiguous. The ligation is accomplished by ligation at suitable restriction sites. If such sites do not exist, then synthetic oligonucleotide adaptors or linkers are used by convention.

[0108] As used herein, the expressions "cell", "cell line" and "cell culture" are used interchangeably and all such names include progeny. Thus, the terms "transformants" and "transformed cells" include the primary target cell and cultures derived therefrom (regardless of the number of transfers). It is also understood that not all progeny have exactly the same DNA content, due to intentional or unintentional mutations. Included are variant progeny that have the same function or biological activity as screened in the originally transformed cell. When a different name is intended, it should be distinguished from the context.

[0109] As used herein, "polymerase chain reaction" or "PCR" refers to a procedure or technique in which a specific nucleic acid sequence, RNA and / or DNA, is amplified, such as described in, for example, U.S. Patent No. 4,683,195. In general, oligonucleotide primers are designed using sequence information from the ends of the region of interest or beyond. These primers are identical or similar to the sequence of the opposite strand of the template to be amplified. The 5' terminal nucleotides of the two primers may be consistent with the ends of the amplified material. PCR can be used to amplify specific RNA sequences, specific DNA sequences from total genomic DNA, and cDNA transcribed from total cellular RNA, phage or plasmid sequences, etc. See generally Mullis et al. (1987) Cold Spring Harbor Symp. Quant. Biol. 51:263; Erlich, ed., (1989) PCR Technology (Stockton Press, NY). As used herein, PCR is considered to be an example, but not the only example, of a nucleic acid polymerase reaction for amplifying a nucleic acid test sample, which includes using known nucleic acids as primers and using a nucleic acid polymerase to amplify or generate a specific segment of nucleic acid.

[0110] As used herein, "germline sequence" refers to the sequence of a non-recombinant immunoglobulin DNA sequence. Any suitable source of non-rearranged immunoglobulin sequences can be used. For example, the JOINSOLVER sequence can be obtained from the website of the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health. ® Human germline sequences can be obtained from germline databases. For example, Giudicelli et al. (2005) Nucleic Acids Res. Mouse germline sequences were obtained as described in 33:D256-D261.

[0111] Physical and functional properties of exemplary anti-PD-L1 antibodies

[0112] The present invention provides isolated anti-PD-L1 antibodies and methods of using the antibodies or antigen-binding fragments thereof for detecting PD-L1 expression on the surface of cells. Examples of the anti-PD-L1 antibodies of the present invention include (but are not limited to): antibody 20C3 and 22C3 (see Figure 1 and Figure 2 ). The 20C3 and 22C3 antibodies bind to non-identical but adjacent epitopes (see Example 2 and Figure 9), indicating that the CDRs of the two antibodies can be mixed to derive other antibodies that specifically bind to PD-L1 at one or both of these epitopes. Thus, an isolated antibody or antigen-binding fragment that binds to human PD-L1 may comprise three of the light chain complementarity determining regions shown in Tables 1 to 3 and three of the heavy chain CDRs.

[0113]

[0114]

[0115]

[0116] "Conservative modified variants" or "conservative substitutions" refer to the replacement of an amino acid in a protein with another amino acid having similar characteristics (such as charge, side chain size, hydrophobicity / hydrophilicity, backbone conformation, and rigidity, etc.), so that the change can generally be made without altering the biological activity of the protein. Those skilled in the art believe that a single amino acid substitution in a non-essential region of a polypeptide generally does not substantially alter the biological activity (see, for example, Watson et al. (1987) Molecular Biology of the Gene , The Benjamin / Cummings Pub. Co., page 224 (4th edition)). Additionally, substitutions of amino acids that are structurally or functionally similar are less likely to disrupt the biological activity. Exemplary conservative substitutions are set forth in Table 4.

[0117]

[0118] The present invention also encompasses functionally conservative variants of the antibodies of the present invention. As used herein, "functionally conservative variants" refer to antibodies or fragments in which one or more amino acid residues are altered without changing the desired properties (such as antigen affinity and / or specificity). Such variants include (but are not limited to) replacing an amino acid with an amino acid having similar properties, such as the conservative amino acid substitutions in Table 4.

[0119] In another embodiment, the present invention includes an antibody or antigen-binding fragment thereof that specifically binds to PD-L1 and has a V L domain and a V Ha domain that shares 100% sequence homology with the light and heavy chain CDRs of Table 1 or Table 2 and at least 90%, 92%, 94%, 96%, 98% or 99% sequence homology with the light and heavy chain mature variable regions of Table 1 or Table 2.

[0120] Nucleic acid

[0121] The present invention also provides nucleic acids encoding immunoglobulin chains of anti-PD-L1 antibodies and antigen-binding fragments disclosed herein. For example, the present invention includes nucleic acids encoding the amino acids described in Tables 1, 2 and 3 and nucleic acids that hybridize thereto.

[0122] Generally, nucleic acids hybridize under low, medium or high stringency conditions and encode antibodies that maintain the ability to specifically bind PD-L1. A first nucleic acid molecule "hybridizes" to a second nucleic acid molecule when the single-stranded form of the first nucleic acid molecule can anneal to the second nucleic acid molecule under appropriate temperature conditions and solution ionic strength (see Sambrook et al., supra). The temperature and ionic strength conditions determine the "stringency" of the hybridization. Typical low stringency hybridization conditions include 55°C, 5X SSC, 0.1% SDS and no formamide; or 30% formamide, 5X SSC, 0.5% SDS, at 42°C. Typical medium stringency hybridization conditions are 40% formamide, 5X or 6X SSC and 0.1% SDS, at 42°C. High stringency hybridization conditions are 50% formamide, 5X or 6X SSC, at 42°C or optionally at a higher temperature (e.g., 57°C, 59°C, 60°C, 62°C, 63°C, 65°C or 68°C). Generally, SSC is 0.15M NaCl and 0.015M sodium citrate. Hybridization requires that the two nucleic acids contain complementary sequences, but depending on the stringency of the hybridization, there may be mismatches between bases. The appropriate stringency of nucleic acid hybridization depends on variables well known in the art, such as the length and degree of complementarity of the nucleic acids. The greater the degree of similarity or homology between two nucleotide sequences, the higher the stringency at which the nucleic acids can hybridize. For hybrids greater than 100 nucleotides in length, equations for calculating the melting temperature have been derived (see Sambrook et al., supra, 9.50-9.51). For hybridization to shorter nucleic acids (e.g., oligonucleotides), the position of the mismatch is more important and the length of the oligonucleotide determines its specificity (see Sambrook et al., supra, 11.7-11.8).

[0123] The following references are related to the BLAST algorithms commonly used in sequence analysis: BLAST ALGORITHMS: BLASTALGORITHMS: Altschul, S.F., et al., (1990) J. Mol. Biol. 215:403-410; Gish, W., et al., (1993) Nature Genet. 3:266-272; Madden, T.L., et al., (1996) Meth. Enzymol. 266:131-141; Altschul, S.F., et al., (1997) Nucleic Acids Res. 25:3389-3402; Zhang, J., et al., (1997) Genome Res. 7:649-656; Wootton, J.C., et al., (1993) Comput. Chem. 17:149-163; Hancock, J.M. et al., (1994) Comput. Appl. Biosci. 10:67-70; ALIGNMENT SCORING SYSTEMS: Dayhoff, M.O., et al., "A model of evolutionary change in proteins." in Atlas of Protein Sequence and Structure, (1978) vol. 5, suppl. 3. M.O. Dayhoff (ed.), pp. 345-352, Natl. Biomed. Res. Found., Washington, DC; Schwartz, R.M., et al., "Matrices for detecting distant relationships." in Atlas of Protein Sequence and Structure, (1978) vol. 5, suppl. 3. " M.O. Dayhoff (ed.), pp. 353-358, Natl. Biomed. Res. Found., Washington, DC; Altschul, S.F., (1991) J. Mol. Biol. 219:555-565; States, D.J., et al., (1991) Methods 3:66-70; Henikoff, S., et al., (1992) Proc. Natl. Acad. Sci.USA 89:10915-10919; Altschul, S.F., et al., (1993) J.Mol. Evol. 36:290-300; ALIGNMENT STATISTICS: Karlin, S., et al., (1990) Proc. Natl. Acad.Sci. USA 87:2264-2268; Karlin, S., et al., (1993) Proc. Natl. Acad. Sci. USA90:5873-5877; Dembo, A., et al., (1994) Ann. Prob. 22:2022-2039; and Altschul,S.F. "Evaluating the statistical significance of multiple distinctlocalalignments." in Theoretical and Computational Methods in Genome Research(S. Suhai, ed.), (1997) pp. 1-14, Plenum, New York。。.

[0124] In another embodiment, the invention provides a isolated nucleic acid encoding at least one of the polypeptide chains of the isolated anti-PD-L1 antibody or antigen-binding fragment described herein, such as DNA. In some embodiments, the isolated nucleic acid encodes a light chain and a heavy chain on a single nucleic acid molecule, and in other embodiments, encodes the light chain and the heavy chain on separate nucleic acid molecules. In another embodiment, the nucleic acid further encodes a signal sequence.

[0125] The invention also provides an expression vector comprising the isolated nucleic acid of the invention, wherein the nucleic acid is operably linked to a control sequence recognized by a host cell when the vector is transfected into the host cell. Also provided are host cells comprising the expression vector of the invention and a method of producing the antibody or antigen-binding fragment disclosed herein, the method comprising culturing a host cell having an expression vector encoding the antibody or antigen-binding fragment in a medium, and isolating the antigen or antigen-binding fragment from the host cell or the medium.

[0126] Epitope binding

[0127] The present invention further provides an antibody or an antigen-binding fragment thereof that blocks the binding of antibody 20C3 or 22C3 to human PD-L1 by binding to the same epitope as 20C3 or 22C3, respectively. Such antibodies and binding fragments can be identified by using any cross-blocking or competition assay known in the art, including the octamer competition assay described in Example 2, followed by identification of the epitope on human PD-L1 to which the cross-blocking antibody binds. If the target is pre-bound to saturation with the first antibody such that the concentration of the second antibody required to achieve half-maximal binding of the target is increased by 2, 3, 4, 5, 10, 20, 50, 100, 200-fold or more than 200-fold, then the first antibody is considered to cross-block the binding of the second antibody. Techniques well known in the art can be used to identify the binding epitope for the cross-blocking antibody.

[0128] One such epitope mapping technique is hydrogen / deuterium exchange coupled to proteolysis and mass spectrometry (HDX-MS). This method relies on the accurate measurement and comparison of the extent of deuterium incorporation by the antigen when incubated alone in heavy water (D 2 2O) or in the presence of its antibody at various time intervals. Deuterium exchanges with the hydrogen on the amide backbone of the protein in the exposed regions, while the regions of the antigen that are bound to the antibody will be protected and show little or no exchange after analysis by liquid chromatography of proteolytic fragments combined with mass spectrometry (LC-MS / MS).

[0129] Based on the HDX-MS epitope mapping described in Example 3, the proposed epitope on mature human PD-L1 for antibody 22C3 comprises residues in two non-contiguous amino acid segments in the extracellular domain (SEQ ID NO: 38): 156 to 178 and 196 to 206. Other epitope residues may be present in the following segments of the extracellular domain (SEQ ID NO: 38): 3 to 9; 10 to 13; 88 to 93 and 135 to 147.

[0130] Thus, in one embodiment, an antibody that blocks the binding of antibody 22C3 to human PD-L1 by binding to the same epitope as 22C3 binds to the residues in the first amino acid segment 156 to 178 of SEQ ID NO: 38 and the residues in the second amino acid segment 196 to 206 of SEQ ID NO: 38, and in some embodiments also binds to the residues in any one, two, three, or all four of the following segments of SEQ ID NO: 38: amino acids 3 to 9; amino acids 10 to 13; amino acids 88 to 93 and amino acids 135 to 147.

[0131] Methods for preparing antibodies and their antigen-binding fragments

[0132] Hybridoma cells that produce parental (e.g., rodent) monoclonal anti-X antibodies can be produced by methods commonly known in the art. These methods include (but are not limited to) the hybridoma technique originally developed by Kohler et al., (1975) (Nature 256:495-497) and the trioma technique (Hering et al., (1988) Biomed. Biochim. Acta. 47:211-216 and Hagiwara et al., (1993) Hum. Antibod. Hybridomas 4:15), human B cell hybridoma technique (Kozbor et al., (1983) Immunology Today 4:72 and Cote et al., (1983) Proc. Natl. Acad. Sci. U. S. A 80:2026-2030), EBV hybridoma technique (Cole et al., Monoclonal Antibodies and Cancer Therapy , Alan R. Liss, Inc., pp. 77-96, 1985), and electric field-based electrofusion using a Cyto Pulse large chamber knockout fusion electroporator (Cyto Pulse Sciences, Inc., Glen Burnie, MD). Preferably, mouse spleen cells are isolated and fused with mouse myeloma cell lines by electrofusion using PEG or based on standard protocols.

[0133] The resulting hybridomas can then be screened for the production of antigen-specific antibodies. For example, a single cell suspension of spleen lymphocytes from immunized mice can be fused with one-sixth the number of P3X63-Ag8.653 non-secreting mouse myeloma cells (ATCC, CRL 1580) using 50% PEG. The cells can be at about 2 x 10 5Cells / ml were inoculated into flat-bottom microtiter plates and then incubated for two weeks in selective medium containing: 20% fetal clone serum, 18% "653" conditioned medium, 5% origen (IGEN), 4 mM L-glutamine, 1 mM L-glutamine, 1 mM sodium pyruvate, 5 mM HEPES, 0.055 mM 2-mercaptoethanol, 50 units / ml penicillin, 50 mg / ml streptomycin, 50 mg / ml gentamycin, and 1X HAT (Sigma; HAT was added 24 hours after fusion). After two weeks, the cells can be cultured in medium with HT replacing HAT. Then individual wells can be screened by ELISA for anti-X monoclonal IgG antibodies. Once extensive hybridoma growth occurs, the medium can usually be observed after 10 - 14 days. The antibody-secreting hybridomas are replated, rescreened, and if still positive for human IgG (anti-X monoclonal antibody), they can be subcloned at least twice by limiting dilution.

[0134] Then the stable subclones can be cultured in vitro to produce small amounts of antibody in tissue culture medium for characterization. For example, the following procedure can be used to produce approximately 1 gram of 22C3 antibody from the mouse hybridoma cell line MEB037.22C3.138 and purify it. The frozen MEB037.22C3.138 cells are thawed and adhered to a shake flask using medium without hybridoma serum containing 2 mM additional L-glutamine and with or without 0.18% Pluronic F-68. The presence of Pluronic F-68 can improve the viability of the shake flask culture. Once the cells are completely adhered to the shake flask, a 20-liter production culture is carried out in serum-free medium in a WAVE bioreactor (GE Healthcare Life Sciences) with the addition of 10% CHO CD Efficient Feed B (Invitrogen, catalog number A10240-01). For cell expansion, a 1-liter culture is started in a small WAVE bag and then the 1-liter WAVE culture is expanded to a 20-liter culture in the WAVE bioreactor. The 20-liter culture can be started at a cell density of 0.5 x 10 6 viable cells / ml, fed 10% CHO CD Efficient Feed B on day 1, and the pH is adjusted daily with 1N Na 2 CO 3 . The cells are harvested after four days. Small samples can be collected daily for NOVA analysis.

[0135] The anti-hPD-L1 antibody of the present invention can be purified from hybridoma cultures by the following method. The hybridoma cultures are clarified by depth filtration using 1.2 micron fiberglass and 0.2 micron cellulose acetate filters. An equal volume of 2X ProSepA buffer (100 mM boric acid, 5 M NaCl, pH 8.5) is added to the clarified harvest and the diluted harvest is loaded onto a 170 mL bed volume protein-A column. The column is washed with 5 column volumes (CV) of 1X ProSepA buffer (50 mM boric acid, 2.5 M NaCl, pH 8.5), then with 2 CV of 1X PBS, and the anti-hPD-L1 antibody is eluted with 5 CV of elution buffer (0.1 M glycine, pH 3.0). The fractions containing IgG are pooled and the pH is neutralized by adding 1 / 10 volume of 1.0 M Tris (pH buffer). The neutralized antibody composition is then sterile filtered using a 10 kDa disposable TFF cassette. The antibody can be formulated for storage by diafiltration against 10 liters of formulation buffer (20 mM sodium acetate, 9% sucrose, pH 5.0) and using 20 volume changes. Using this protocol, antibody 22C3 at a concentration of approximately 5.0 mg / ml can be prepared and has a purity of at least 98% as measured by SDS-PAGE, SEC HPLC, and C8RP-HPLC, and an endotoxin level of less than 0.1 EU / ml and less than 0.02 EU / mg.

[0136] The anti-PD-L1 antibodies disclosed herein can also be produced recombinantly (e.g., in an E. coli / T7 expression system as discussed above). In this embodiment, nucleic acids encoding the antibody molecules of the present invention (e.g., V H or V L ) can be inserted into a pET-based plasmid and expressed in an E. coli / T7 system. Several methods for producing recombinant antibodies are known in the art. An example of a method for producing antibodies recombinantly is disclosed in U.S. Patent No. 4,816,567. Transformation can be carried out by any known method for introducing polynucleotides into host cells. Methods for introducing heterologous polynucleotides into mammalian cells are well known in the art and include dextran-mediated transfection, calcium phosphate precipitation, polybrene-mediated transfection, protoplast fusion, electroporation, encapsulation of polynucleotides in liposomes, biolistic injection, and direct microinjection of DNA into the nucleus. Additionally, nucleic acid molecules can be introduced into mammalian cells by viral vectors. Methods for cell transformation are well known in the art. See, e.g., U.S. Patent Nos. 4,399,216; 4,912,040; 4,740,461; and 4,959,455.

[0137] The anti-PD-L1 antibody can also be synthesized by any of the methods described in U.S. Patent No. 6,331,415.

[0138] Mammalian cell lines that can be used as hosts for expressing the antibodies or fragments disclosed herein are well known in the art and include many immortalized cell lines available from the American Type Culture Collection (ATCC). These cell lines include, in particular, Chinese hamster ovary (CHO) cells, NSO, SP2 cells, HeLa cells, baby hamster kidney (BHK) cells, monkey kidney cells (COS), human hepatocellular carcinoma cells (such as Hep G2), A549 cells, 3T3 cells, HEK-293 cells, and many other cell lines. Mammalian host cells include human, mouse, rat, dog, monkey, pig, goat, cow, horse, and hamster cells. Particularly preferred cell lines are selected by determining cell lines with high expression levels. Other cell lines that can be used are insect cell lines, such as Sf9 cells, amphibian cells, bacterial cells, plant cells, and fungal cells. When a recombinant expression vector encoding a heavy chain or its antigen-binding portion or fragment, a light chain, and / or its antigen-binding fragment is introduced into a mammalian host cell, the antibody is produced by culturing the host cell for a time sufficient to express the antibody in the host cell or, more preferably, to secrete the antibody into the culture medium in which the host cell is growing.

[0139] The antibody can be recovered from the culture medium using standard protein purification methods. In addition, a variety of known techniques can be used to enhance the expression of the antibodies of the present invention (or other portions thereof) from production cell lines. For example, the glutamine synthetase gene expression system (GS system) is a common method for enhancing expression under certain conditions. The GS system is discussed in its entirety or in part in European Patent Nos. 0 216 846, 0 256 055, and 0 323 997 and European Patent Application No. 89303964.4.

[0140] A polyclonal antibody is an antibody produced in or in the presence of one or more other different antibodies. Generally, polyclonal antibodies are produced by a collection of different B-lymphocytes, such as the B-lymphocytes of an animal treated with a related immunogen, thereby producing a population of different antibodies all directed against the immunogen. Usually, polyclonal antibodies are obtained directly from the immunized animal, such as the spleen, serum, or ascitic fluid.

[0141] The present invention further includes antibody fragments of the anti-PD-L1 antibodies disclosed herein. Antibody fragments include F(ab) 2 fragments, which can be produced by enzymatic cleavage of IgG, for example, with pepsin. Fab fragments can be obtained, for example, by reducing F(ab) with dithiothreitol or mercaptoethylamine2 for production. The Fab fragment is attached to the V H -C H1 chain via a disulfide bridge. The V L -C L chain. The F(ab) 2 fragment is two Fab fragments, which are in turn attached by two disulfide bridges. The Fab portion of the F(ab) 2 molecule includes the portion in the F c region with a disulfide bridge disposed therein. The F V fragment is the V L or V H region.

[0142] Immunoglobulins can be classified into different classes depending on the amino acid sequence of the constant domain of the heavy chain of the visual immunoglobulin. There are at least five major immunoglobulin classes: IgA, IgD, IgE, IgG, and IgM, and several of these classes can be further divided into subclasses (isotypes), such as IgG-1, IgG-2, IgG-3, and IgG-4; IgA-1 and IgA-2. The present invention encompasses antibodies and antigen-binding fragments in any of these antibody classes or subclasses.

[0143] In one embodiment, the antibody or antigen-binding fragment comprises a heavy chain constant region, such as a human constant region, such as the γ1, γ2, γ3, or γ4 human heavy chain constant region or a variant thereof. In another embodiment, the antibody or antigen-binding fragment comprises a light chain constant region, such as a human light chain constant region, such as the λ or κ human light chain region or a variant thereof. For example (but not limited to), the human heavy chain constant region can be γ1 and the human light chain constant region can be κ. In an alternative embodiment, the Fc region of the antibody is γ4 with a Ser228Pro mutation (Schuurman, J et al., Mol. Immunol. 38:1-8, 2001).

[0144] In some embodiments, different constant domains can be attached to the humanized V L and V H regions derived from the CDRs provided herein. For example, if a particular intended use of the antibody (or fragment) of the present invention requires an alteration in effector function, a heavy chain constant domain other than human IgG 1 can be used, or a hybrid IgG 1 / IgG 4 can be utilized.

[0145] Antibody engineering

[0146] In certain embodiments, certain amino acids containing exposed side chains are converted to another amino acid residue as described below to provide greater chemical stability of the final antibody. Deamidation of asparagine can occur on N-G or D-G sequences and results in the production of an isoaspartic acid residue, which introduces a kink into the polypeptide chain and reduces its stability (the isoaspartic acid effect). In certain embodiments, the antibodies of the invention do not contain asparagine isomerization sites.

[0147] For example, asparagine (Asn) residues can be changed to Gln or Ala to reduce the likelihood of forming isoaspartic acid at any Asn-Gly sequence, particularly within the CDRs. Similar problems can exist at Asp-Gly sequences. Reissner and Aswad (2003) Cell. Mol. Life Sci. 60:1281. Isoaspartic acid formation can weaken or completely eliminate the binding of the antibody to its target antigen. See Presta (2005) J. Allergy Clin. Immunol. 116:731-734. In one embodiment, asparagine is changed to glutamine (Gln). It may also be necessary to change the amino acids adjacent to asparagine (Asn) or glutamine (Gln) residues to reduce the likelihood of deamidation, which occurs with greater probability when a small amino acid is adjacent to asparagine or glutamine. See Bischoff and Kolbe (1994) J. Chromatog. 662:261. Additionally, any methionine residue in the CDR (usually solvent-exposed Met) can be changed to Lys, Leu, Ala, or Phe to reduce the likelihood of methionine sulfoxidation, which can reduce antigen-binding affinity and also contribute to molecular heterogeneity in the final antibody formulation. Ibid. In one embodiment, methionine is changed to alanine (Ala). Additionally, to prevent or minimize potentially labile Asn-Pro peptide bonds, any Asn-Pro combinations seen in the CDR may need to be changed to Gln-Pro, Ala-Pro, or Asn-Ala. Antibodies with such substitutions are then screened to ensure that the substitutions do not reduce the affinity or specificity of the antibody against human PD-L1, or other desired biological activities, to an unacceptable level.

[0148]

[0149] Antibody conjugates

[0150] The anti-PD-L1 antibody molecules disclosed herein can also be conjugated to chemical moieties such as radionuclides or other detectable labels. Radionuclides include 99 Tc, 90 Y, 111 In, 32 P, 14 C, 125I, 3 H, 131 I, 11 C, 15 O, 13 N, 18 F, 35 S, 51 Cr, 57 To, 226 Ra, 60 Co, 59 Fe, 57 Se, 152 Eu, 67 CU, 217 Ci, 211 At, 212 Pb, 47 Sc, 109 Pd, 234 Th, and 40 K, 157 Gd, 55 Mn, 52 Tr and 56 Fe. Fluorescent or chemiluminescent labels include fluorophores such as rare earth chelates, fluorescein and its derivatives, rhodamine and its derivatives, isothiocyanates, phycoerythrin, allophycocyanin, o-phthalaldehyde, fluorescamine, 152 Eu, dansyl, umbelliferone, fluorescein, luminal label, isoluminal label, aromatic acridinium ester label, imidazole label, acridinium salt label, oxalate ester label, aequorin label, 2,3-dihydrophthalazinedione, biotin / avidin, spin labels, and stable free radicals.

[0151] Any method known in the art for conjugating antibody molecules to a variety of moieties can be employed, including those described in the following references: Hunter, et al., (1962) Nature 144:945; David, et al., (1974) Biochemistry 13:1014; Pain, et al., (1981) J. Immunol. Meth. 40:219; and Nygren, J., (1982) Histochem. and Cytochem. 30:407. Methods for conjugating antibodies are routine and well known in the art.

[0152] Experimental and diagnostic uses

[0153] The anti-PD-L1 antibodies and antibody fragments disclosed herein can be used to specifically detect human PD-L1 expressed on the cell surface. The cells can be present in tissue or serum samples obtained from a human individual, and detection of PD-L1 expression is performed using any of a variety of in vitro assay methods known in the art.

[0154] For example, certain embodiments include an ELISA assay (enzyme-linked immunosorbent assay), which generally comprises the following steps:

[0155] (a) Coating a substrate (e.g., the surface of a microtiter plate well (e.g., a plastic plate)) with an anti-PD-L1 antibody or an antigen-binding fragment thereof;

[0156] (b) Coating a sample to be tested for the presence of human PD-L1 onto the substrate;

[0157] (c) Washing the plate, thereby removing unbound substances in the sample;

[0158] (d) Coating an antibody that is also specific for human PD-L1 and that is detectably labeled (e.g., an antibody linked to an enzyme);

[0159] (e) Washing the substrate, thereby removing unbound labeled antibody;

[0160] (f) If the labeled antibody is linked to an enzyme, coating a chemical that can be converted by the enzyme into a fluorescent signal; and

[0161] (g) Detecting the presence of the labeled antibody.

[0162] In another embodiment, the labeled antibody is labeled with a peroxidase that can react with ABTS (e.g., 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)) or 3,3',5,5'-tetramethylbenzidine to produce a detectable color change. Alternatively, the labeled antibody is labeled with a detectable radioisotope (e.g., 3 H) that can be detected by a scintillation counter in the presence of a scintillant.

[0163] The anti-PD-L1 antibodies of the present invention and antigen-binding fragments thereof can be used in Western blotting or immunoblotting procedures. This procedure forms part of the present invention and includes, for example:

[0164] (1) Contacting a membrane or other solid substrate to be tested for the presence of human PD-L1 with an antibody of the present invention or an antigen-binding fragment thereof. Such membranes can be in the form of membranes based on nitrocellulose or vinyl (e.g., polyvinylidene difluoride (PVDF)), where testing for the presence is performed in a non-denaturing PAGE (polyacrylamide gel electrophoresis) gel or an SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel electrophoresis) gelX The protein has been transferred to the membrane (e.g., after electrophoretic separation in a gel). Prior to contacting the membrane with the anti-PD-L1 antibody or fragment, the membrane is optionally blocked, e.g., with non-fat dry milk powder, etc., to bind non-specific protein binding sites on the membrane.

[0165] (2) Wash the membrane one or more times to remove unbound anti-PD-L1 antibody or fragment and other unbound substances; and

[0166] (3) Detect the bound anti-PD-L1 antibody or fragment.

[0167] The bound antibody or fragment can be detected as follows: incubate the bound antibody or fragment with a secondary antibody (anti-immunoglobulin antibody) that has been detectably labeled, and then detect the presence of the secondary antibody.

[0168] The anti-PD-L1 antibodies and antigen-binding fragments thereof disclosed herein can also be used in immunohistochemistry (IHC) assays, which can be performed using a variety of IHC formats known in the art and constitute embodiments of the present invention. A typical IHC assay uses FFPE tissue sections that are approximately 3-4 mm and preferably 4 microns, mounted on microscope slides and dried, and includes, for example, (1) subjecting the tissue section to dewaxing and hydration, contacting the rehydrated tissue section with the anti-PD-L1 antibody or antigen-binding fragment of the present invention; and (2) detecting the anti-PD-L1 antibody or antigen-binding fragment on the surface of one or more cells in the tissue. If the antibody or fragment itself is detectably labeled, it can be detected directly. Alternatively, the antibody or fragment can be bound by a detectably labeled secondary antibody that has been detected.

[0169] Preferred IHC assays employ commercially available Dako EnVision TMThe FLEX detection system is intended to be used with the DakoAutostainer instrument (Dako, Agilent Technologies Company, Glostrup, Denmark). When this system is used with the 22C3 antibody or an antibody comprising the variable regions of the heavy and light chains of the 22C3 antibody, IHC analysis can be performed as described below. A four-micron thick FFPE tissue section mounted on a slide is air-dried overnight, baked at 60 °C for 45 minutes, deparaffinized and rehydrated. After deparaffinization, the FFPE slides are subjected to heat-induced epitope retrieval by using EnVision™ FLEX high pH target retrieval solution at 97 °C and then for 20 minutes at room temperature. The slides are then washed, stained with 2 μg / mL of 22C3 for 60 minutes, and then detected using Dako EnVision™ FLEX reagents as follows: EnVision™ FLEX+ MS linker (15 minutes), EnVision™ FLEX / HRP (20 minutes), EnVision™ FLEX DAB (10 minutes) and DAB enhancer (7 minutes), with wash steps interspersed.

[0170] Certain anti-PD-L1 antibodies and antigen-binding fragments thereof disclosed herein can also be used for in vivo tumor imaging. Such methods can include injecting a radio-labeled anti-PD-L1 antibody or antigen-binding fragment thereof into a human patient to be tested for the presence of a tumor associated with PD-L1 expression, followed by nuclear imaging of the patient's body to detect the presence of the labeled antibody or fragment, such as at loci containing high concentrations of the antibody or fragment bound to the tumor.

[0171] Imaging techniques include SPECT imaging (single photon emission computed tomography) or PET imaging (positron emission tomography). Labels include, for example, iodine-123 ( 123 I) and technetium-99m ( 99m Tc) (e.g., for use in conjunction with SPECT imaging), or 11 C, 13 N, 15 O or 18 F (e.g., for use in conjunction with PET imaging), or indium-111 (see, for example, Gordon et al., (2005) International Rev. Neurobiol. 67: 385-440).

[0172] Detection kits and therapeutic kits

[0173] For convenience, the antibodies or specific binding agents disclosed herein may be provided in a kit, i.e., a packaged combination of a pre-determined amount of the reagent and instructions for performing a diagnostic or detection assay. When the antibody is enzyme-labeled, the kit will include the substrate and cofactor required for the enzyme (e.g., a substrate precursor that provides a detectable chromophore or fluorophore). In addition, other additives may be included, such as stabilizers, buffers (e.g., blocking buffer or lysis buffer), and the like. The relative amounts of the various reagents can vary widely to provide reagent concentrations in solution that substantially optimize the sensitivity of the assay. Specifically, the reagents may be provided in dry powder form (usually lyophilized), which includes excipients that, upon dissolution, provide a reagent solution of appropriate concentration.

[0174] Diagnostic or detection reagents and kits containing one or more such reagents are also provided for use in a variety of detection assays (including, for example, immunoassays such as ELISA (sandwich or competitive formats)). The components of the kit may be pre-attached to a solid support or may be coated onto the surface of a solid support when using the kit. In some embodiments, the signal generating means may be pre-associated with the antibodies of the present invention or may need to be combined with one or more components (e.g., buffer, antibody-enzyme conjugate, enzyme substrate, etc.) prior to use. The kit may also include other reagents, such as blocking reagents, washing reagents, enzyme substrates, etc., for reducing non-specific binding to the solid phase surface. The solid phase surface may be in the form of a tube, bead, microtiter plate, microsphere, or other material suitable for immobilizing proteins, peptides, or polypeptides. In certain aspects, the enzyme that catalyzes the formation of a chemiluminescent or chromogenic product or the reduction of a chemiluminescent or chromogenic substrate is a component of the signal generating means. Such enzymes are well known in the art. The kit may contain any of the capture reagents and detection reagents described herein. The kit may optionally also include instructions for practicing the methods of the present invention.

[0175] Detection kits may also be prepared that contain at least one of the antibodies or antigen-binding fragments disclosed herein and instructions for using the composition as a detection reagent. The container for such a kit generally may comprise at least one vial, test tube, flask, bottle, syringe, or other suitable container into which one or more detection compositions may be placed and preferably appropriately aliquoted. The kits disclosed herein generally also include means for containing vials that are prohibited from commercial sale, such as an injection or blow molded plastic container that holds the desired vial therein. When a radioactive label, chromogenic, fluorescent, or other type of detectable label or detection means is included within the kit, the labeling agent may be provided in the same container as the detection composition itself, or may alternatively be placed in a second different container means that can hold the second composition and be appropriately aliquoted. Alternatively, the detection reagent may be prepared in a single container means, and in most cases, the kit generally also includes means for containing vials that are prohibited from commercial sale and / or for facilitating packaging and delivery.

[0176] Devices or apparatuses for performing the detection or monitoring methods described herein are also provided. Such devices may include a chamber or tube into which a sample can be introduced, a fluid handling system optionally including valves or pumps to direct the flow of the sample through the device, an optional filter for separating plasma or serum from blood, a mixing chamber for adding capture or detection reagents, and an optional detection device for detecting the amount of a detectable label bound to an immunocomplex with a capture reagent. The flow of the sample can be passive (e.g., by capillary force, hydrostatic force, or other forces that do not require further manipulation of the device once the sample is applied) or active (e.g., by applying a force generated by a mechanical pump, electroosmotic pump, centrifugal force, or increased air pressure) or achieved by a combination of active and passive forces.

[0177] Other embodiments also provide a processor, a computer-readable memory, and a program stored on the computer-readable memory and adapted to be executed on the processor to implement any of the methods described herein. Examples of suitable computing systems, environments, and / or configurations include personal computers, server computers, hand-held or laptop devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments (including any of the above systems or devices), or any other systems known in the art.

[0178] General methods

[0179] Standard methods in molecular biology are described in Sambrook, Fritsch, and Maniatis (1982 and 1989, 2nd edition, 2001, 3rd edition) Molecular Cloning, A Laboratory Manual , Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY; Sambrook and Russell (2001) Molecular Cloning, 3rd Edition , Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY; Wu (1993) Recombinant DNA , Volume 217, Academic Press, San Diego, CA). Standard methods are also published in Ausbel et al., (2001) Current Protocols in Molecular Biology, in Volumes 1 - 4, John Wiley and Sons, Inc., New York, NY, this literature describes cloning in bacterial cells and DNA mutagenesis (Volume 1), cloning in mammalian cells and yeast (Volume 2), glycoconjugates and protein expression (Volume 3), and bioinformatics (Volume 4).

[0180] Describes protein purification methods, which include immunoprecipitation, chromatography, electrophoresis, centrifugation, and crystallization (Coligan et al., (2000) Current Protocols in Protein Science , Volume 1, John Wiley and Sons, Inc., New York). Describes chemical analysis, chemical modification, post - translational modification, production of fusion proteins, glycosylation of proteins (see, for example, Coligan et al., (2000) Current Protocols in Protein Science , Volume 2, John Wiley and Sons, Inc., New York; Ausubel et al., (2001) Current Protocols in Molecular Biology , Volume 3, John Wiley and Sons, Inc., NY, NY, pages 16.0.5 - 16.22.17; Sigma - Aldrich, Co. (2001) Products for Life Science Research , St. Louis, MO; pages 45 - 89; Amersham Pharmacia Biotech (2001) BioDirectory , Piscataway, N.J., pages 384 - 391). Describes the production, purification, and fragmentation of polyclonal and monoclonal antibodies (Coligan et al., (2001) Current Protocols in Immunology , Volume 1, John Wiley and Sons, Inc., New York; Harlow and Lane (1999) Using Antibodies , Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY; Harlow and Lane, ibid). Standard techniques are available for characterizing ligand / receptor interactions (see, for example, Coligan et al., (2001) Current Protocols in Immunology , Volume 4, John Wiley, Inc., New York).

[0181] Monoclonal, polyclonal, and humanized antibodies can be prepared (see, for example, Shepherd and Dean (eds.) (2000) Monoclonal Antibodies, Oxford Univ. Press, New York, NY; Kontermann and Dubel (eds.) (2001) Antibody Engineering , Springer-Verlag, New York; Harlow and Lane (1988) Antibodies A Laboratory Manual , Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 139 - 243; Carpenter, et al. (2000) J. Immunol . 165:6205; He, et al. (1998) J. Immunol . 160:1029; Tang et al. (1999) J.Biol. Chem.274:27371 - 27378; Baca et al. (1997) J. Biol. Chem . 272:10678 - 10684; Chothia et al. (1989) Nature 342:877 - 883; Foote and Winter (1992) J. Mol. Biol. 224:487 - 499; U.S. Patent No. 6,329,511).

[0182] Alternative methods for humanization are to use human antibody libraries displayed on phage or human antibody libraries in transgenic gene mice (Vaughan et al. (1996) Nature Biotechnol. 14:309 - 314; Barbas (1995) Nature Medicine 1:837 - 839; Mendez et al. (1997) Nature Genetics 15:146 - 156; Hoogenboom and Chames (2000) Immunol. Today 21:371 - 377; Barbas et al. (2001) Phage Display: A Laboratory Manual , Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York; Kay et al. (1996) Phage Display of Peptides and Proteins: A Laboratory Manual , Academic Press, San Diego, CA; de Bruin et al. (1999) Nature Biotechnol. 17:397 - 399).

[0183] Antigen purification is not necessary for antibody production. Animals can be immunized with cells bearing the relevant antigen. Spleen cells can then be isolated from the immunized animals, and the spleen cells can be fused with a myeloma cell line to produce hybridomas (see, for example, Meyaard et al. (1997) Immunity 7:283-290; Wright et al. (2000) Immunity 13:233-242; Preston et al., supra; Kaithamana et al. (1999) J. Immunol. 163:5157-5164).

[0184] Antibodies can be conjugated to, for example, small drug molecules, enzymes, liposomes, polyethylene glycol (PEG). Antibodies are suitable for therapeutic, diagnostic, kit or other purposes, and include antibodies conjugated to, for example, dyes, radioisotopes, enzymes or metals (such as colloidal gold) (see, for example, Le Doussal et al. (1991) J. Immunol . 146:169-175; Gibellini et al. (1998) J. Immunol . 160:3891-3898; Hsing and Bishop (1999) J. Immunol . 162:2804-2811; Everts et al. (2002) J. Immunol . 168:883-889).

[0185] Methods are available for flow cytometry, which include fluorescence-activated cell sorting (FACS) (see, for example, Owens, et al. (1994) Flow Cytometry Principles for Clinical Laboratory Practice , John Wiley and Sons, Hoboken, NJ; Givan (2001) Flow Cytometry, 2nd Edition; Wiley-Liss, Hoboken, NJ; Shapiro (2003) Practical Flow Cytometry , John Wiley and Sons, Hoboken, NJ). Fluorescent reagents suitable for modifying nucleic acids are available, including nucleic acid primers and probes, polypeptides and antibodies used, for example, as diagnostic reagents (Molecular Probes (2003) Catalogue , Molecular Probes, Inc., Eugene, OR; Sigma-Aldrich (2003) Catalogue , St. Louis, MO).

[0186] Standard methods for describing the histology of the immune system (see, for example, Muller-Harmelink (ed.) (1986) Human Thymus: Histopathology and Pathology , Springer Verlag, New York, NY; Hiatt, et al. (2000) Color Atlas of Histology , Lippincott, Williams, and Wilkins, Phila, PA; Louis, et al. (2002) Basic Histology: Text and Atlas , McGraw-Hill, New York, NY).

[0187] Software packages and databases are available for determining, for example, antigen fragments, leader sequences, protein folding, functional domains, glycosylation sites, and sequence alignments (see, for example, GenBank, Vector NTI® Suite (Informax, Inc, Bethesda, MD); GCG Wisconsin Package (Accelrys, Inc., San Diego, CA); DeCypher® (TimeLogic Corp., Crystal Bay, Nevada); Menne, et al. (2000) Bioinformatics 16:741-742; Menne, et al. (2000) Bioinformatics Applications Note 16:741-742; Wren, et al. (2002) Comput. Methods Programs Biomed. 68:177-181; von Heijne (1983) Eur. J. Biochem. 133:17-21; von Heijne (1986) Nucleic Acids Res. 14:4683-4690). Examples

[0188] Example 1. Generation and Screening of Anti-PD-L1 Hybridomas

[0189] Balb / C mice were immunized with an adjuvant containing human PD-L1-Fc fusion protein (R&D Systems® catalog number 156-B7-100). This fusion protein contains the extracellular domain of PD-L1 (Phe19-Thr239) fused to a human IgG1 fragment (Pro100-Lys 300). After 12 immunizations, the lymph nodes of two mice with high titers against human PD-L1 were collected and electrofused to generate two batches of hybridomas, given the laboratory names MEB033 and MEB037.

[0190] The supernatants of the MEB033 and MEB037 hybridoma batches were screened to identify hybridomas producing antibodies against human PD-L1. The screening was performed using a protein-based ELISA directed against binding to the human hPD-L1-Fc protein; and a cell-based ELISA directed against binding to the human PD-L1-CHO human PD-L1-CHO stable transfectant and parental CHO cells (as a negative control). Supernatants from 88 MEB037 clones and 23 MEB033 hybridoma clones tested positive for the presence of anti-PD-L1 antibodies (data not shown), and their samples were tested for IHC reactivity against FFPE tissue sections from normal human tonsils (data not shown).

[0191] Among these 88 clones, based on comparison with the staining patterns obtained with the commercially available anti-PD-L1 antibodies listed in the table below, only 11 clones from the MEB037 batch produced staining patterns with sufficient intensity and apparent specificity to warrant further evaluation:

[0192] 。

[0193] When the inventors compared the various staining patterns obtained with the experimental antibodies to those obtained with these commercially available anti-human PD-L1 antibodies, they observed significant differences between the staining patterns, including the location of the staining and the type of cells stained. To attempt to explain these differences, the inventors performed many other experiments and found that none of these commercially available antibodies provided the combination of properties required for IHC of PD-L1 expression in FFPE sections: (1) sensitivity, i.e., the ability to detect normal physiological expression in positive control tissues (e.g., human tonsils) as well as expression in tumor tissues (e.g., human melanoma samples); (2) specificity, i.e., the staining pattern needed to be related to the known anatomical / cellular distribution of PD-L1 and needed to be neutralizable; and (3) robustness, i.e., little to no variation in the staining pattern when analyzing "duplicate" tissue sections.

[0194] For example, the inventors found that the Sigma / ProSco PRS4059 antibody showed multiple bands on tonsil lysates (confirming that none of them represented PD-L1), failed to stain the LOX melanoma cell line confirmed to be PD-L1 positive by flow cytometry, and failed to distinguish positive from negative cell lines by IHC.

[0195] Some of these data are shown in Figure 4, in which immunohistochemical staining of tonsil sections identified 22C3 and 20C3 as two antibodies with rare and useful immunohistochemical properties on FFPE tissue, compared to the PRS4059 antibody (which was identified by Gadiot et al. (supra) as the only candidate among 15 anti-human PD-L1 antibodies suitable for detecting PD-L1 expression in FFPE tissue sections). In experiments conducted by the inventors, the Prosci antibody (PRS4059, lot number 40590604, used primarily at 0.4 mg / ml, followed by a rabbit polymer detection system (DAKO Envision)) stained all hematopoietic lineages in the tonsil with equal intensity ( Figure 4A ), while the 22C3 antibody selectively stained tonsillar crypt epithelial cells and follicular CD68+ myeloid cells (morphologically consistent with macrophages) ( Figure 4B ). Substantially the same staining pattern was observed with the 20C3 antibody (data not shown). In addition, 22C3 and 20C3 showed a consistent difference in staining intensity between these two discrete cell populations, with crypt epithelial cells staining far more strongly than follicular macrophages. All three antibodies could be neutralized by pre-incubation with the PD-L1 antigen, indicating that the reactivity was mediated through the antigen-binding domain (CDR).

[0196] Example 2. Quality assessment of 20C3 and 22C3 anti-PD-L1 antibodies

[0197] This example describes additional experiments conducted to evaluate the utility of the 20C3 and 22C3 antibodies in an IHC assay for FFPE tissue sections.

[0198] One experiment evaluated the ability of these two antibodies to detect the expression of multiple human PD-L1 (hPD-L1) proteins in an IHC assay of normal human FFPE tonsil sections, and a representative image of 22C3 is shown in Figure 5A . Immunohistochemical staining with 22C3 strongly labeled tonsillar crypt epithelial cells, as well as showed weak to moderate staining of the CD68+ follicular myeloid population (putative macrophages). Both antibodies (20C3 data not shown) labeled cells in both of these cell types in a clearly defined membrane / cell surface pattern. The appropriateness of hPD-L1 expression in the tonsil was confirmed on adjacent FFPE tonsil tissue sections by an independent method (in situ hybridization [ISH] for hPD-L1 mRNA). Additionally, the differential expression of hPD-L1 protein (crypt epithelial cells >> follicular macrophages) as evaluated by IHC was consistent with the relative abundance of hPD-L1 mRNA observed by ISH.

[0199] Another experiment evaluated the binding specificity of 20C3 and 22C3 to hPD-L1-expressing cells. HT144 cells, which are known to be negative for hPD-L1 expression by mRNA analysis (qPCR), and LOX melanoma cells, which are known to highly express hPD-L1 mRNA (qPCR), were stained with 1 μg / ml of purified mouse IgG from the seven hybridomas generated in the experiment described in Example 1 above. An irrelevant isotype control mouse antibody at the same concentration was also used. Primary mouse antibodies were detected using a fluorescently labeled anti-mouse secondary antibody. After staining and repeated washing, the cells were analyzed by flow cytometry, and the median fluorescence intensity of the population (>10,000 collected events) was calculated. The results are shown in Figure 6.

[0200] 20C3 and 22C3, along with other hPD-L1 antibodies, were used as flow cytometry reagents to detect cell surface hPD-L1. The significant right shift of the 20C3 and 22C3 histograms ( Figure 6A ) compared to the isotype control antibody curve reflects the selective detection of hPD-L1 on the hPD-L1-positive LOX melanoma cell line. The median fluorescence intensities related to these histograms and other results from this analysis are shown in Figure 6B . The selectivity of 20C3 and 22C3 binding was further confirmed by the lack of significant binding (i.e., the MFIs of 22C3 and 20C3 comparable to the isotype) on the negative cell line HT144. In contrast, Figure 6B the data in

[0201] show that the MFIs of 20C3 and 22C3 are at least 10-fold increased compared to the isotype on the hPD-L1-positive LOX melanoma cell line. Thus, by flow cytometry evaluation, 20C3 and 22C3 (except for clones 5F9, 7C8, 13D2, and 31D3) show selective binding to hPD-L1-expressing cells.

[0201] Another experiment evaluated the ability of the 22C3 antibody to detect hPD-L1 expression on engineered and human cell lines. A Chinese hamster ovary (CHO) cell line that is negative for hPD-L1 was transfected with an expression vector encoding human PD-L1 to generate an engineered positive control cell line. As shown in Figure 7B , 22C3 immunohistochemical staining of formalin-fixed paraffin-embedded (FFPE) cell pellets of the parental CHO cell line (negative control) and the transfected CHO cell line (positive control) showed appropriate positive and negative staining.

[0202] Other FFPE human cell line pellets (A375, HS578T, and LOX melanoma) were stained with 22C3 and showed various staining patterns and intensities (as shown in Figure 7BAs shown in). 22C3 staining was strong and uniform on Lox melanoma cells, but only rare single-positive cells were shown in the A375 and HS578T cell pellets. Similar staining was observed with the 20C3 antibody (data not shown). As evaluated by qPCR using ubiquitin mRNA as a baseline, the hPD-L1 expression levels detected by 22C3 in these three cell lines in the IHC analysis were well correlated with the PD-L1 mRNA levels in these cell lines.

[0203] The selective binding and relative affinity of 22C3 for hPD-L1-expressing cells were evaluated in cell-based ELISA experiments. Several cell lines (hPDL1-CHOK1 cells, parental CHOK1 cells, hPDL2-CHOK1 cells, and LOX cells) were plated onto individual wells of a collagen-coated 96-well plate and grown to confluence. The medium was removed and replaced with fresh CHOK1 medium (DMEM / F12 containing 10% BCS) containing increasing concentrations of the primary antibody ranging from 1.4 ng / ml to 3,000 ng / ml. The following primary antibodies were used: each of the two different production batches of antibodies 20C3 and 22C3, in which mouse IgG1 isotype, anti-PD-L1 antibody (Biolegend), and anti-PD-L2 antibody served as controls. The primary antibody was incubated at 37 °C for 1 hour, washed three times with PBS / 0.01% Tween 20, and a secondary antibody (goat anti-human IgG, Fc-specific HRP (Southern Biotech, catalog number 1030-05) conjugate) diluted 1:2000 with CHOK1 medium was added. The secondary antibody was incubated at 37 °C for 1 hour and washed five times as above. ELISA was performed using TMB, stopped with 0.1 N phosphoric acid, and the absorbance at 450 nm (subtracting the background at 650 nm) was read. The results shown in Figure 8 demonstrated the selective binding of 22C3 and 20C3 to cells expressing hPD-L1, and for both the cells engineered with hPD-L1 CHOK1 and LOX cells, the binding affinity of 22C3 was greater than that of 20C3.

[0204] A similar ELISA experiment was performed to evaluate whether either the 20C3 or 22C3 antibody binds to mouse PD-L1. No significant binding of either antibody to mouse PD-L1 was observed (data not shown).

[0205] Perform an antibody binding competition assay ("cross-blocking") between different anti-hPD-L1 antibody pairs identified in the experiment described in Example 1. The assay uses the ForteBio® Octet® platform, which is based on biolayer interferometry. Briefly, this technique measures the binding of the initial antibody (mAb1) to the surface of the biosensor tip as a wavelength shift (Δλ) that causes an increase in the optical thickness (Y-axis) of the biosensor tip over time (X-axis). The tip consists of an anti-huIgG sensor that binds hPD-L1-Fc. The change in optical thickness upon binding of the anti-PD-L1 antibody is reflected by an upward-sloping curve starting from the first vertical red dotted line (see Figure 9A , Figure 9B and Figure 9C for the graphs), which represents the addition of a saturating concentration of mAb1 (10 μg / ml) to the solution. After reaching equilibrium (about 1000 seconds), a second antibody is injected into the assay solution (indicated by the second vertical red dotted line in Figure 9A , Figure 9B and Figure 9C ). Binding of the second mAb2, as indicated by an additional shift in the curve, indicates that the two antibodies bind to non-overlapping epitopes, while less to no shift in the curve indicates that the two antibodies bind to overlapping or identical epitopes.

[0206] Overall, the results showed that 22C3 binding competes with additional binding of all other anti-hPDL-1 clones (tested as mAb2) except 5H9 ( Figure 9A ). Similarly, 20C3 also failed to compete with 5H9 binding but showed moderate additional binding with 22C3 and 4B7 ( Figure 9B ). When considered together, these data indicate that 20C3 and 22C3 bind to overlapping epitopes.

[0207] The ability of the 22C3 antibody to detect multiple hPD-L1 expressions in different tumor types was evaluated by performing IHC analysis on FFPE sections prepared from the following tumors: bladder, esophagus, head and neck, kidney, HCC, breast, lung, ovary, and gastric tumors. A semi-quantitative "all-or-none" interpretation of the staining intensity was used for the initial screening of the reactivity of 22C3 with these tumor tissue sections. As Figure 10 described, 22C3 was able to detect a wide range of PD-L1 expressions from essentially no staining (score = 0) to significantly strong expression (score = 4), thus demonstrating the utility of 22C3 IHC analysis in guiding future tumor types that may respond to inhibition of the immunosuppressive PD-1 / PD-L1 interaction.

[0208] The utility of the 22C3 antibody for stratifying patients likely to respond to treatment that blocks the interaction of PD-1 with PD-L1 was evaluated in a study that utilized 22C3 immunohistochemical evaluation of archival samples obtained from 18 melanoma patients enrolled in a Phase 1 (P001) MK-3475 (an anti-PD1 therapeutic antibody developed by Merck and Co., Inc.) trial. Two pathologists independently evaluated the cases and classified them as "positive", "negative", or "indeterminate", and representative images depicting these three categories are shown in Figure 11A . The concordance between the pathologists for this sample set (n = 18) was 100%.

[0209] Clinical response was evaluated using immune-related response criteria (irRC) and correlated with the IHC results. For this analysis, "indeterminate" cases were considered negative, resulting in an analysis sensitivity of 72% and a specificity of 86%. Figure 11B The results shown in

[0210] demonstrate that 22C3 immunohistochemical staining of FFPE tissue has utility as a patient selection biomarker.

[0211] Example 3. Epitope Mapping of Human PD-L1 against the 22C3 Anti-PD-L1 Antibody

[0212] HDX-MS epitope mapping was performed using the antibody 22C3 and the PD-L1-His protein (which contains the extracellular domain of mature human PD-L1 fused to an eleven-mer histidine tag (SEQ ID NO: 38)). Regions 156 to 178 and 196 to 206 on the extracellular domain of human PD-L1 (SEQ ID NO: 38) showed strong protection when bound to the antibody 22C3 (average deuteration level difference > 10%). Additionally, regions 3 to 9, 10 to 13, 88 to 93, and 135 to 147 showed minor but significant protection (average deuteration level difference of 5% to 10%).

[0213] All references cited herein are incorporated by reference to the extent as if each individual disclosure, database entry (e.g., Genbank sequence or GeneID entry), patent application, or patent were specifically and individually incorporated by reference. In accordance with 37 C.F.R.§1.57(b)(1), for this statement regarding incorporation by reference of references, the applicant means each individual disclosure, database entry (e.g., Genbank sequence or GeneID entry), patent application, or patent, each document being clearly identified in accordance with 37 C.F.R.§1.57(b)(2), even if such citation is not immediately adjacent to the specific statement of incorporation by reference. The inclusion of a specific statement of incorporation by reference in the specification does not in any way weaken this general statement regarding incorporation by reference. The citation of references herein is not intended to admit that the references are relevant prior art, nor does it constitute any admission as to the content or date of these disclosures or documents.

[0214] The invention is not limited to the scope of the specific embodiments described herein. In fact, those skilled in the art will recognize from the foregoing description and the accompanying drawings various modifications of the invention other than those described herein. Such modifications are intended to fall within the scope of the appended claims.

[0215] It is believed that the foregoing written description is sufficient to enable those skilled in the art to practice the invention. Those skilled in the art will recognize from the foregoing description various modifications of the invention other than those shown and described herein, and such modifications fall within the scope of the appended claims. Sequence Listing <110> Merck Sharp & Dohme Corp. R. H. Pierce L. Liang P. Bohne M.E. Buehler <120> Antibodies that Bind to Human Programmed Death Ligand 1 (PD-L1) <130> 23411 <150> US 61 / 745,386 <151> 2012-12-21 <160> 36 <170> PatentIn version 3.5 <210> 1 <211> 17 <212> PRT <213> Artificial Sequence <220> <223> Consensus Sequence <220> <221> MISC_FEATURE <222> (8)..(8) <223> The Xaa at position 8 can be His or Asn <220> <221> MISC_FEATURE <222> (9)..(9) <223> The Xaa at position 9 can be Ser or Thr <220> <221> MISC_FEATURE <222> (10)..(10) <223> The Xaa at position 10 can be Arg or Ser <400> 1 Lys Ser Ser Gln Ser Leu Leu Xaa Xaa Xaa Thr Arg Lys Asn Tyr Leu 1 5 10 15 Ala <210> 2 <211> 7 <212> PRT <213> Mus musculus <400> 2 Trp Ala Ser Thr Arg Glu Ser 1 5 <210> 3 <211> 8 <212> PRT <213> Artificial sequence <220> <223> Consensus sequence <220> <221> MISC_FEATURE <222> (1)..(1) <223> The Xaa at position 1 can be Gln or Lys <400> 3 Xaa Gln Ser Tyr Asp Val Val Thr 1 5 <210> 4 <211> 112 <212> PRT <213> Artificial Sequence <220> <223> Consensus Sequence <220> <221> MISC_FEATURE <222> (22)..(22) <223> Xaa at position 22 can be Ser or Thr <220> <221> MISC_FEATURE <222> (31)..(31) <223> Xaa at position 31 can be His or Asn <220> <221> MISC_FEATURE <222> (32)..(32) <223> Xaa at position 32 can be Ser or Thr <220> <221> MISC_FEATURE <222> (33)..(33) <223> Xaa at position 33 can be Arg or Ser <220> <221> MISC_FEATURE <222> (95)..(95) <223> Xaa at position 95 can be Gln or Lys <400> 4 Asp Ile Val Met Ser Gln Ser Pro Ser Ser Leu Ala Val Ser Ala Gly 1 5 10 15 Glu Lys Val Thr Met Xaa Cys Lys Ser Ser Gln Ser Leu Leu Xaa Xaa 20 25 30 Xaa Thr Arg Lys Asn Tyr Leu Ala Trp Tyr Gln Gln Lys Pro Gly Gln 35 40 45 Ser Pro Lys Leu Leu Ile Tyr Trp Ala Ser Thr Arg Glu Ser Gly Val 50 55 60 Pro Asp Arg Phe Thr Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu Thr 65 70 75 80 Ile Ser Ser Val Gln Ala Glu Asp Leu Ala Val Tyr Tyr Cys Xaa Gln 85 90 95 Ser Tyr Asp Val Val Thr Phe Gly Ala Gly Thr Lys Leu Glu Leu Lys 100 105 110 <210> 5 <211> 5 <212> PRT <213> Artificial Sequence <220> <223> Consensus Sequence <220> <221> MISC_FEATURE <222> (4)..(4) <223> Xaa at position 4 can be Ile or Met <400> 5 Ser Tyr Trp Xaa His 1 5 <210> 6 <211> 17 <212> PRT <213> Artificial Sequence <220> <223> Consensus Sequence <220> <221> MISC_FEATURE <222> (7)..(7) <223> Xaa at position 7 can be Asp or Gly <220> <221> MISC_FEATURE <222> (9)..(9) <223> Xaa at position 9 can be His or Asn <220> <221> MISC_FEATURE <222> (12)..(12) <223> Xaa at position 12 can be Ser or Asn <220> <221> MISC_FEATURE <222> (13)..(13) <223> Xaa at position 13 can be Glu or Gln <220> <221> MISC_FEATURE <222> (16)..(16) <223> Xaa at position 16 can be Ile or Met <400> 6 Tyr Ile Asn Pro Ser Ser Xaa Tyr Xaa Glu Tyr Xaa Xaa Lys Phe Xaa 1 5 10 15 Asp <210> 7 <211> 13 <212> PRT <213> Artificial Sequence <220> <223> Consensus Sequence <220> <221> MISC_FEATURE <222> (5)..(5) <223> Xaa at position 5 can be Ile or Val <220> <221> MISC_FEATURE <222> (13)..(13) <223> Xaa at position 13 can be Phe or Tyr <400> 7 Ser Gly Trp Leu Xaa His Gly Asp Tyr Tyr Phe Asp Xaa 1 5 10 <210> 8 <211> 122 <212> PRT <213> Artificial Sequence <220> <223> Consensus Sequence <220> <221> MISC_FEATURE <222> (3)..(3) <223> Xaa at position 3 can be His or Gln <220> <221> MISC_FEATURE <222> (4)..(4) <223> Xaa at position 4 can be Leu or Val <220> <221> MISC_FEATURE <222> (13)..(13) <223> Xaa at position 13 can be Glu or Lys <220> <221> MISC_FEATURE <222> (34)..(34) <223> Xaa at position 34 can be Ile or Met <220> <221> MISC_FEATURE <222> (37)..(37) <223> Xaa at position 37 can be Ile or Leu <220> <221> MISC_FEATURE <222> (56)..(56) <223> Xaa at position 56 can be Asp or Gly <220> <221> MISC_FEATURE <222> (58)..(58) <223> Xaa at position 58 can be His or Asn <220> <221> MISC_FEATURE <222> (61)..(61) <223> Xaa at position 61 can be Asn or Ser <220> <221> MISC_FEATURE <222> (62)..(62) <223> The Xaa at position 62 can be Glu or Gln <220> <221> MISC_FEATURE <222> (65)..(65) <223> The Xaa at position 65 can be Ile or Met <220> <221> MISC_FEATURE <222> (74)..(74) <223> The Xaa at position 74 can be Lys or Arg <220> <221> MISC_FEATURE <222> (75)..(75) <223> The Xaa at position 75 can be Ala or Ser <220> <221> MISC_FEATURE <222> (77)..(77) <223> The Xaa at position 77 can be Ser or Thr <220> <221> MISC_FEATURE <222> (82)..(82) <223> The Xaa at position 82 can be His or Gln <220> <221> MISC_FEATURE <222> (84)..(84) <223> The Xaa at position 84 can be Ile or Thr <220> <221> MISC_FEATURE <222> (103)..(103) <223> The Xaa at position 103 can be Ile or Val <220> <221> MISC_FEATURE <222> (111)..(111) <223> Xaa at position 111 can be Phe or Tyr <400> 8 Gln Val Xaa Xaa Gln Gln Ser Gly Ala Glu Leu Ala Xaa Pro Gly Ala 1 5 10 15 Ser Val Lys Met Ser Cys Lys Ala Ser Gly Tyr Ile Phe Thr Ser Tyr 20 25 30 Trp Xaa His Trp Xaa Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile 35 40 45 Gly Tyr Ile Asn Pro Ser Ser Xaa Tyr Xaa Glu Tyr Xaa Xaa Lys Phe 50 55 60 Xaa Asp Lys Ala Thr Leu Thr Ala Asp Xaa Xaa Ser Xaa Thr Ala Tyr 65 70 75 80 Met Xaa Leu Xaa Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Ser Gly Trp Leu Xaa His Gly Asp Tyr Tyr Phe Asp Xaa Trp 100 105 110 Gly Gln Gly Thr Thr Leu Thr Val Ser Ser 115 120 <210> 9 <211> 17 <212> PRT <213> Mus musculus <400> 9 Lys Ser Ser Gln Ser Leu Leu Asn Ser Arg Thr Arg Lys Asn Tyr Leu 1 5 10 15 Ala <210> 10 <211> 8 <212> PRT <213> Mus musculus <400> 10 Gln Gln Ser Tyr Asp Val Val Thr 1 5 <210> 11 <211> 20 <212> PRT <213> Mus musculus <400> 11 Met Asp Ser Gln Ala Gln Val Leu Ile Leu Leu Leu Leu Trp Val Ser 1 5 10 15 Gly Thr Phe Gly 20 <210> 12 <211> 132 <212> PRT <213> Mus musculus <400> 12 Met Asp Ser Gln Ala Gln Val Leu Ile Leu Leu Leu Leu Trp Val Ser 1 5 10 15 Gly Thr Phe Gly Asp Ile Val Met Ser Gln Ser Pro Ser Ser Leu Ala 20 25 30 Val Ser Ala Gly Glu Lys Val Thr Met Ser Cys Lys Ser Ser Gln Ser 35 40 45 Leu Leu Asn Ser Arg Thr Arg Lys Asn Tyr Leu Ala Trp Tyr Gln Gln 50 55 60 Lys Pro Gly Gln Ser Pro Lys Leu Leu Ile Tyr Trp Ala Ser Thr Arg 65 70 75 80 Glu Ser Gly Val Pro Asp Arg Phe Thr Gly Ser Gly Ser Gly Thr Asp 85 90 95 Phe Thr Leu Thr Ile Ser Ser Val Gln Ala Glu Asp Leu Ala Val Tyr 100 105 110 Tyr Cys Gln Gln Ser Tyr Asp Val Val Thr Phe Gly Ala Gly Thr Lys 115 120 125 Leu Glu Leu Lys 130 <210> 13 <211> 112 <212> PRT <213> Mus musculus <400> 13 Asp Ile Val Met Ser Gln Ser Pro Ser Ser Leu Ala Val Ser Ala Gly 1 5 10 15 Glu Lys Val Thr Met Ser Cys Lys Ser Ser Gln Ser Leu Leu Asn Ser 20 25 30 Arg Thr Arg Lys Asn Tyr Leu Ala Trp Tyr Gln Gln Lys Pro Gly Gln 35 40 45 Ser Pro Lys Leu Leu Ile Tyr Trp Ala Ser Thr Arg Glu Ser Gly Val 50 55 60 Pro Asp Arg Phe Thr Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu Thr 65 70 75 80 Ile Ser Ser Val Gln Ala Glu Asp Leu Ala Val Tyr Tyr Cys Gln Gln 85 90 95 Ser Tyr Asp Val Val Thr Phe Gly Ala Gly Thr Lys Leu Glu Leu Lys 100 105 110 <210> 14 <211> 5 <212> PRT <213> Mus musculus <400> 14 Ser Tyr Trp Met His 1 5 <210> 15 <211> 10 <212> PRT <213> Mus musculus <400> 15 Gly Tyr Ile Phe Thr Ser Tyr Trp Met His 1 5 10 <210> 16 <211> 17 <212> PRT <213> Mus musculus <400> 16 Tyr Ile Asn Pro Ser Ser Asp Tyr Asn Glu Tyr Ser Glu Lys Phe Met 1 5 10 15 Asp <210> 17 <211> 13 <212> PRT <213> Mus musculus <400> 17 Ser Gly Trp Leu Val His Gly Asp Tyr Tyr Phe Asp Tyr 1 5 10 <210> 18 <211> 19 <212> PRT <213> Mus musculus <400> 18 Met Glu Arg His Trp Ile Phe Leu Phe Leu Phe Ser Val Thr Ala Gly 1 5 10 15 Val His Ser <210> 19 <211> 141 <212> PRT <213> Mus musculus <400> 19 Met Glu Arg His Trp Ile Phe Leu Phe Leu Phe Ser Val Thr Ala Gly 1 5 10 15 Val His Ser Gln Val Gln Val Gln Gln Ser Gly Ala Glu Leu Ala Glu 20 25 30 Pro Gly Ala Ser Val Lys Met Ser Cys Lys Ala Ser Gly Tyr Ile Phe 35 40 45 Thr Ser Tyr Trp Met His Trp Leu Lys Gln Arg Pro Gly Gln Gly Leu 50 55 60 Glu Trp Ile Gly Tyr Ile Asn Pro Ser Ser Asp Tyr Asn Glu Tyr Ser 65 70 75 80 Glu Lys Phe Met Asp Lys Ala Thr Leu Thr Ala Asp Lys Ala Ser Thr 85 90 95 Thr Ala Tyr Met Gln Leu Ile Ser Leu Thr Ser Glu Asp Ser Ala Val 100 105 110 Tyr Tyr Cys Ala Arg Ser Gly Trp Leu Val His Gly Asp Tyr Tyr Phe 115 120 125 Asp Tyr Trp Gly Gln Gly Thr Thr Leu Thr Val Ser Ser 130 135 140 <210> 20 <211> 122 <212> PRT <213> Mus musculus <400> 20 Gln Val Gln Val Gln Gln Ser Gly Ala Glu Leu Ala Glu Pro Gly Ala 1 5 10 15 Ser Val Lys Met Ser Cys Lys Ala Ser Gly Tyr Ile Phe Thr Ser Tyr 20 25 30 Trp Met His Trp Leu Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile 35 40 45 Gly Tyr Ile Asn Pro Ser Ser Asp Tyr Asn Glu Tyr Ser Glu Lys Phe 50 55 60 Met Asp Lys Ala Thr Leu Thr Ala Asp Lys Ala Ser Thr Thr Ala Tyr 65 70 75 80 Met Gln Leu Ile Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Ser Gly Trp Leu Val His Gly Asp Tyr Tyr Phe Asp Tyr Trp 100 105 110 Gly Gln Gly Thr Thr Leu Thr Val Ser Ser 115 120 <210> 21 <211> 17 <212> PRT <213> Mus musculus <400> 21 Lys Ser Ser Gln Ser Leu Leu His Thr Ser Thr Arg Lys Asn Tyr Leu 1 5 10 15 Ala <210> 22 <211> 8 <212> PRT <213> Mus musculus <400> 22 Lys Gln Ser Tyr Asp Val Val Thr 1 5 <210> 23 <211> 20 <212> PRT <213> Mus musculus <400> 23 Met Asp Ser Gln Ala Gln Val Leu Ile Leu Leu Leu Leu Trp Val Ser 1 5 10 15 Gly Thr Cys Gly 20 <210> 24 <211> 132 <212> PRT <213> Mus musculus <400> 24 Met Asp Ser Gln Ala Gln Val Leu Ile Leu Leu Leu Leu Trp Val Ser 1 5 10 15 Gly Thr Cys Gly Asp Ile Val Met Ser Gln Ser Pro Ser Ser Leu Ala 20 25 30 Val Ser Ala Gly Glu Lys Val Thr Met Thr Cys Lys Ser Ser Gln Ser 35 40 45 Leu Leu His Thr Ser Thr Arg Lys Asn Tyr Leu Ala Trp Tyr Gln Gln 50 55 60 Lys Pro Gly Gln Ser Pro Lys Leu Leu Ile Tyr Trp Ala Ser Thr Arg 65 70 75 80 Glu Ser Gly Val Pro Asp Arg Phe Thr Gly Ser Gly Ser Gly Thr Asp 85 90 95 Phe Thr Leu Thr Ile Ser Ser Val Gln Ala Glu Asp Leu Ala Val Tyr 100 105 110 Tyr Cys Lys Gln Ser Tyr Asp Val Val Thr Phe Gly Ala Gly Thr Lys 115 120 125 Leu Glu Leu Lys 130 <210> 25 <211> 112 <212> PRT <213> Mus musculus <400> 25 Asp Ile Val Met Ser Gln Ser Pro Ser Ser Leu Ala Val Ser Ala Gly 1 5 10 15 Glu Lys Val Thr Met Thr Cys Lys Ser Ser Gln Ser Leu Leu His Thr 20 25 30 Ser Thr Arg Lys Asn Tyr Leu Ala Trp Tyr Gln Gln Lys Pro Gly Gln 35 40 45 Ser Pro Lys Leu Leu Ile Tyr Trp Ala Ser Thr Arg Glu Ser Gly Val 50 55 60 Pro Asp Arg Phe Thr Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu Thr 65 70 75 80 Ile Ser Ser Val Gln Ala Glu Asp Leu Ala Val Tyr Tyr Cys Lys Gln 85 90 95 Ser Tyr Asp Val Val Thr Phe Gly Ala Gly Thr Lys Leu Glu Leu Lys 100 105 110 <210> 26 <211> 5 <212> PRT <213> Mus musculus <400> 26 Ser Tyr Trp Ile His 1 5 <210> 27 <211> 10 <212> PRT <213> Mus musculus <400> 27 Gly Thr Thr Phe Thr Ser Tyr Trp Ile His 1 5 10 <210> 28 <211> 17 <212> PRT <213> Mus musculus <400> 28 Tyr Ile Asn Pro Ser Ser Gly Tyr His Glu Tyr Asn Gln Lys Phe Ile 1 5 10 15 Asp <210> 29 <211> 13 <212> PRT <213> Mus musculus <400> 29 Ser Gly Trp Leu Ile His Gly Asp Tyr Tyr Phe Asp Phe 1 5 10 <210> 30 <211> 19 <212> PRT <213> Mus musculus <400> 30 Met Glu Arg His Trp Ile Phe Leu Phe Leu Phe Ser Val Thr Ala Gly 1 5 10 15 Val His Ser <210> 31 <211> 141 <212> PRT <213> Mus musculus <400> 31 Met Glu Arg His Trp Ile Phe Leu Phe Leu Phe Ser Val Thr Ala Gly 1 5 10 15 Val His Ser Gln Val His Leu Gln Gln Ser Gly Ala Glu Leu Ala Lys 20 25 30 Pro Gly Ala Ser Val Lys Met Ser Cys Lys Ala Ser Gly Tyr Thr Phe 35 40 45 Thr Ser Tyr Trp Ile His Trp Ile Lys Gln Arg Pro Gly Gln Gly Leu 50 55 60 Glu Trp Ile Gly Tyr Ile Asn Pro Ser Ser Gly Tyr His Glu Tyr Asn 65 70 75 80 Gln Lys Phe Ile Asp Lys Ala Thr Leu Thr Ala Asp Arg Ser Ser Ser 85 90 95 Thr Ala Tyr Met His Leu Thr Ser Leu Thr Ser Glu Asp Ser Ala Val 100 105 110 Tyr Tyr Cys Ala Arg Ser Gly Trp Leu Ile His Gly Asp Tyr Tyr Phe 115 120 125 Asp Phe Trp Gly Gln Gly Thr Thr Leu Thr Val Ser Ser 130 135 140 <210> 32 <211> 122 <212> PRT <213> Mus musculus <220> <221> MISC_FEATURE <222> (1)..(1) <223> Xaa at position 1 can be Gln or pyroglutamate <400> 32 Xaa Val His Leu Gln Gln Ser Gly Ala Glu Leu Ala Lys Pro Gly Ala 1 5 10 15 Ser Val Lys Met Ser Cys Lys Ala Ser Gly Tyr Thr Phe Thr Ser Tyr 20 25 30 Trp Ile His Trp Ile Lys Gln Arg Pro Gly Gln Gly Leu Glu Trp Ile 35 40 45 Gly Tyr Ile Asn Pro Ser Ser Gly Tyr His Glu Tyr Asn Gln Lys Phe 50 55 60 Ile Asp Lys Ala Thr Leu Thr Ala Asp Arg Ser Ser Ser Thr Ala Tyr 65 70 75 80 Met His Leu Thr Ser Leu Thr Ser Glu Asp Ser Ala Val Tyr Tyr Cys 85 90 95 Ala Arg Ser Gly Trp Leu Ile His Gly Asp Tyr Tyr Phe Asp Phe Trp 100 105 110 Gly Gln Gly Thr Thr Leu Thr Val Ser Ser 115 120 <210> 33 <211> 396 <212> DNA <213> Mus musculus <400> 33 atggattcac aggcccaggt tcttatattg ctgctgctat gggtatctgg tacctttggg 60 gacattgtga tgtcacaatc tccatcctcc ctggctgtgt cagcaggaga gaaggtcact 120 atgagctgca aatccagtca gagtctgctc aacagtagaa cccgaaagaa ctacttggct 180 tggtaccagc agaaaccagg gcagtctcct aaactgctga tctactgggc atccactagg 240 gaatctgggg tccctgatcg cttcacaggc agtggatctg ggacagattt cactctcacc 300 atcagcagtg tgcaggctga agacctggca gtttattact gccagcaatc ttatgatgtg 360 gtcacgttcg gtgctgggac caagctggag ctgaaa 396 <210> 34 <211> 423 <212> DNA <213> Mus musculus <400> 34 atggaaaggc actggatctt tctcttcctg ttttcagtaa ctgcaggtgt ccactcccag 60 gtccaggttc agcagtctgg ggctgaactg gcagaacctg gggcctcagt gaagatgtcc 120 tgcaaggcct ctggctacat ctttactagc tactggatgc actggctaaa gcagaggcct 180 ggacagggtc tggaatggat tggatacatt aatcccagca gtgattataa tgaatacagt 240 gagaaattca tggacaaggc cacattgact gcagacaaag cctccaccac agcctacatg 300 caactgatca gcctgacatc tgaggactct gcagtctatt actgtgcaag atcgggatgg 360 ttagtacatg gagactatta ttttgactac tggggccaag gcaccactct cacagtctcc 420 tca 423 <210> 35 <211> 396 <212> DNA <213> Mus musculus <400> 35 atggattcac aggcccaggt tcttatattg ctgctgctat gggtatctgg tacctgtggg 60 gacattgtga tgtcacagtc tccctcctcc ctggctgtgt cagcaggaga gaaggtcact 120 atgacctgca aatccagtca gagtctgctc cacactagca cccgaaagaa ctacttggct 180 tggtaccagc agaaaccagg gcagtctcct aaactgctga tctattgggc atccactagg 240 gaatctgggg tccctgatcg cttcacaggc agtggatctg ggacagattt cactctcacc 300 atcagcagtg tgcaggctga agacctggca gtttattact gcaaacaatc ttatgatgtg 360 gtcacgttcg gtgctgggac caagctggag ctgaaa 396 <210> 36 <211> 423 <212> DNA <213> Mus musculus <400> 36 atggaaaggc actggatctt tctcttcctg ttttcagtaa ctgcaggtgt ccactcccag 60 gtccaccttc agcagtctgg ggctgaactg gcaaaacctg gggcctcagt gaagatgtcc 120 tgcaaggctt ctggctacac gtttactagt tactggatac actggataaa gcagaggcct 180 ggacagggtc tggaatggat tggatacatt aatccttcct ctggttatca tgaatacaat 240 cagaaattca ttgacaaggc cacattgact gctgacagat cctccagcac agcctacatg 300 cacctgacca gcctgacgtc tgaagactct gcagtctatt actgtgcaag atcgggatgg 360 ttaatacatg gagactacta ctttgacttc tggggccaag gcaccactct cacagtctcc 420 tca 423

Claims

1. An isolated antibody or antigen-binding fragment thereof, comprising a light chain variable region and a heavy chain variable region, wherein: (a) the light chain variable region is selected from SEQ ID NO:25 and SEQ ID NO:24; and (b) the heavy chain variable region is selected from SEQ ID NO:32 and SEQ ID NO:

31.

2. An isolated nucleic acid encoding both an antibody light chain variable region and an antibody heavy chain variable region, wherein (a) the antibody light chain variable region is as shown in SEQ ID NO:24; and (b) the antibody heavy chain variable region is as shown in SEQ ID NO:

31.

3. The isolated nucleic acid of claim 2, comprising SEQ ID NO:35 and SEQ ID NO:

36.

4. An expression vector comprising the isolated nucleic acid of any one of claims 2 or 3.

5. A host cell comprising the expression vector of claim 4.

6. Use of a PD-L1 binding reagent for the preparation of a kit for performing a method of analyzing PD-L1 expression in a tissue sample taken from a human, the method comprising: (a) contacting the tissue sample with the PD-L1 binding reagent under conditions that permit specific binding of the PD-L1 binding reagent to human PD-L1, wherein the binding reagent comprises the antibody or antigen-binding fragment of claim 1, (b) removing unbound PD-L1 binding reagent, and (c) detecting the presence or absence of the bound PD-L1 binding agent.

7. The use of claim 6, further comprising quantifying the amount of the bound binding reagent.

8. A kit comprising the isolated antibody or antigen-binding fragment thereof of claim 1 and a set of reagents for detecting a complex of the antibody or antigen-binding fragment bound to human PD-L1.

9. The kit of claim 8, wherein the antibody or antigen-binding fragment comprises an antibody light chain variable region of SEQ ID NO:24 and an antibody heavy chain variable region of SEQ ID NO:31.

Images