Method for quantifying individual antibodies in mixture

Abstract

In particular, the present disclosure relates to a method of quantifying the amount of antibody molecules of a mixture comprising two or more antibody molecules, the method comprising: separating each of the two or more antibody molecules of the mixture by hydrophobic interaction chromatography high performance liquid chromatography (HIC-HPLC); and quantifying the amount of each antibody molecule; wherein the molecular weight of each antibody molecule is within 15 kDa of any other antibody molecule in the mixture, and each antibody molecule and another antibody molecule in the mixture are in a Kyteamp; in some embodiments, the Doooltt hydrophobicity scale differs by about 0.25 units or more, or each of the antibody molecules elute at a slightly overlapping run time that differs from the other antibody molecules in the mixture when operated separately with HIC-HPLC, or both.

Description

[0001] This application is a divisional application of invention patent application 201780050583.0.

[0002] CROSS-REFERENCE TO RELATED APPLICATIONS

[0003] This patent application claims the benefit of U.S. Provisional Application No.

[0004] 62 / 375,887, which is incorporated herein by reference in its entirety. Technical Field

[0005] The present disclosure relates to the field of assays for co-formulation of therapeutic antibodies. Background of the Invention

[0007] Administration of multiple rather than a single monoclonal antibody (mAb) to a patient can improve its diagnostic or therapeutic profile and efficiency. These mAbs can be co-formulated in a single drug product (DP), and the DP administered to the patient.

[0008] Regulatory agencies require a method to quantify individual mAbs to be incorporated into a co-formulated drug substance (cFDS) of a DP or into the DP itself. Developing a method to separate two or more antibody molecules and measure the concentration of each mAb is challenging because antibody molecules can have similar molecular weight, protein structure, and charge properties. Summary of the invention

[0009] The present disclosure includes a method for quantifying the amount of an antibody in a mixture comprising multiple antibodies. In some aspects, the method may include the following steps: separating each of the multiple antibodies in the mixture using hydrophobic interaction chromatography high performance liquid chromatography (HIC-HPLC); quantifying the amount of each antibody in the mixture; wherein the molecular weight of each antibody in the mixture is within 15 kDa of the molecular weight of any other antibody in the mixture, and the surface hydrophobicity of each antibody in the mixture differs from the surface hydrophobicity of another antibody in the mixture by more than about 0.25 units on the Kyte & Doolittle hydrophobicity scale, or each antibody in the mixture elutes at a different run time than another antibody in the mixture when run separately with HIC-HPLC, or both.

[0010] In some embodiments, the surface hydrophobicity of each antibody in the mixture differs from the surface hydrophobicity of each other antibody in the mixture by about 0.5 to about 1.0 units on the Kyte & Doolittle hydrophobicity scale. In other embodiments, the surface hydrophobicity of each antibody in the mixture is determined by calculating the surface hydrophobicity based on the protein structure or structural model, rapid screening for solubility in ammonium sulfate or PEG 8000, or rapid screening for molecular effects by affinity self-capture nanoparticle spectroscopy (AC-SINS).

[0011] In additional embodiments, the first antibody in the mixture elutes at a first run time during the HIC-HPLC run, the second antibody in the mixture elutes at a second run time during the HIC-HPLC run, and the first run time and the second run time do not overlap. In further embodiments, the first antibody in the mixture and the second antibody in the mixture have protein sequences that are at least 90% homologous; the first antibody and the second antibody have protein structures that are at least 90% homologous, as determined by their protein sequences; or the first antibody and the second antibody have isoelectric points (pi) that are within about 0.6 of each other, as determined by their protein sequences.

[0012] In some embodiments, the plurality of antibodies comprises three antibodies. In other embodiments, one or more of the antibodies in the mixture are monoclonal antibodies. In other embodiments, one or more of the antibodies in the mixture are human monoclonal antibodies. In other embodiments, two or more of the antibodies in the mixture are of the same isotype. In some embodiments, two or more of the antibodies in the mixture are variants of each other. In other embodiments, two or more of the antibodies in the mixture bind to the same antigen.

[0013] In some embodiments, the mixture is a co-formulated composition. In additional embodiments, the co-formulated composition is configured to treat MERS in human patients. In further embodiments, the co-formulated composition is configured to treat Ebola hemorrhagic fever in human patients. In further embodiments, the co-formulated composition is configured to treat macular degeneration in human patients. In further embodiments, two or more antibodies in the co-formulated composition are configured to treat infectious diseases in human patients. In some embodiments, the co-formulated composition is contained in a pharmaceutical product.

[0014] In some embodiments, the HIC-HPLC is performed in a buffer of about pH 5.0 to about pH 7.0. In further embodiments, the method further comprises generating a chromatogram from the HIC-HPLC, wherein for the elution of each antibody in the mixture, the chromatogram shows a peak that does not overlap with other peaks in the chromatogram.

[0015] The present disclosure also includes a method for quantifying the amount of an antibody in a mixture comprising a plurality of antibodies, the method comprising: separating each of the plurality of antibodies in the mixture using hydrophobic interaction chromatography high performance liquid chromatography (HIC-HPLC), wherein the molecular weight of each antibody in the mixture is within 15 kDa of the molecular weight of each other antibody in the mixture; quantifying the amount of each antibody in the mixture; and generating a chromatogram from the HIC-HPLC, wherein for the elution of each antibody in the mixture, the chromatogram shows a peak that does not overlap with other peaks in the chromatogram. In some embodiments, one or more of the plurality of antibodies is a human monoclonal antibody. In further embodiments, the surface hydrophobicity of each antibody in the mixture differs from the surface hydrophobicity of another antibody in the mixture by more than about 0.25 units on the Kyte & Doolittle hydrophobicity scale, or each antibody in the mixture elutes at a different run time than another antibody in the mixture when run separately with HIC-HPLC, or both.

[0016] Many other aspects and embodiments are provided in accordance with these and other aspects of the disclosure.Other features and aspects of the disclosure will become more fully apparent from the following detailed description and appended claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated into and constitute a part of this specification, illustrate various examples and, together with the description, are used to explain the principles of the present disclosure. Any features (e.g., devices, methods, etc.) of the embodiments or examples described herein may be combined with any other embodiments or examples and are covered by the present disclosure.

[0019] Figure 1 An exemplary chromatogram of a HIC-HPLC run of a co-formulation containing an anti-Ebola mAb is shown.

[0020] Figures 2 to 8 An exemplary chromatogram of a HIC-HPLC run of a co-formulation comprising an anti-MERS mAb is shown.

[0021] Fig. 9 An exemplary chromatogram of a HIC-HPLC run of a co-formulation containing an anti-Ebola mAb is shown.

[0022] Fig.10 Graph showing the linearity of HIC-HPLC, depicting the peak areas for the quantities of individual anti-Ebola mAbs run.

[0023] Fig.11 The graph is a range graph depicting the percent recovery of mAb for the amounts of individual anti-Ebola mAbs run.

[0024] Fig.12Graph showing the precision of HIC-HPLC with different mixture sample batches, depicting the percent recovery of mAb for the amount of individual anti-Ebola mAb run.

[0025] Fig.13 Graph showing the precision of HIC-HPLC with different ratios of individual anti-Ebola mAbs, depicting the percent recovery of mAb against the amount of individual anti-Ebola mAb run.

[0026] Fig.14 Graph of storage stability of mAbs in cFDS, depicting the amount of individual anti-Ebola mAbs against storage time in months.

[0027] Fig.15 An exemplary chromatogram of a cation exchange ultra-high pressure liquid chromatography (CEX-UPLC) run of a co-formulation containing an anti-Ebola mAb is shown.

[0028] Fig.16 A graph showing the linearity of CEX-UPLC, depicting the peak area versus the amount of individual anti-Ebola mAbs.

[0029] Fig.17 An exemplary chromatogram of a size exclusion ultra-high pressure liquid chromatography (SE-UPLC) run of a co-formulation comprising an anti-Ebola mAb is shown.

[0030] Fig.18 Exemplary imaged capillary isoelectric focusing (iCIEF) curves of co-formulations containing anti-Ebola mAb are shown.

[0031] Fig.19 An exemplary chromatogram of a reverse phase ultra-high pressure liquid chromatography (RP-UPLC) run of a co-formulation containing an anti-Ebola mAb is shown. DETAILED DESCRIPTION

[0032] The term "antibody" is sometimes used interchangeably with the term "immunoglobulin". In short, it can refer to a complete antibody comprising two light chain polypeptides and two heavy chain polypeptides. Complete antibodies include different antibody isotypes, including IgM antibodies, IgG antibodies, IgA antibodies, IgD antibodies and IgE antibodies. The term "antibody" can include, for example, polyclonal antibodies, monoclonal antibodies (mAbs), chimeric or chimeric antibodies, humanized antibodies, primatized antibodies, deimmunized antibodies and fully human antibodies. Antibodies can be made or derived from various five species, for example, mammals, such as humans, non-human primates (e.g., orangutans, baboons or chimpanzees), horses, cattle, pigs, sheep, goats, dogs, cats, rabbits, guinea pigs, gerbils, hamsters, rats and mice. Antibodies can be purified or recombinant antibodies. Antibodies can also be engineered proteins or antibody-like proteins comprising at least one immunoglobulin domain (e.g., fusion proteins). The engineered protein or antibody-like protein can also be a bispecific antibody or a trispecific antibody, or a dimer, trimer or multimer antibody, or a diabody, DVD-Ig, CODV-Ig, or

[0033] The terms "antibody variant", "antibody variant" and the like refer to an antibody that is altered from another antibody, wherein the variant antibody is a deletion variant, insertion variant and / or substitution variant of the other antibody.

[0034] As used herein, the term "human antibody" is intended to include antibodies with variable regions and constant regions derived from human germline immunoglobulin sequences. Human mAb may include amino acid residues such as CDR and particularly CDR3 that are not encoded by human germline immunoglobulin sequences (e.g., mutations caused by random or site-specific mutagenesis in vitro or by somatic mutations in vivo). However, as used herein, the term "human antibody" is not intended to include mAbs, wherein the CDR sequences derived from the germline of another mammalian species (e.g., mice) have been grafted onto human FR sequences. The term includes antibodies that are recombinantly prepared in non-human mammals or in cells of non-human mammals. The term is not intended to include antibodies isolated from human subjects or generated in human subjects.

[0035] As used herein, the terms "treat," "treating," or "treatment" refer to a reduction or alleviation of the severity of at least one symptom or indication of a disease or disorder as a result of the administration of a co-formulation of two or more antibodies to a subject in need thereof. The term includes inhibition of the progression of the disease. The term also includes a positive prognosis for the disease.

[0036] The terms "prevent," "preventing," or "prevention" refer to the inhibition of the manifestation of a disease or disorder, any symptom or indication of the disease or disorder following administration of a co-formulation of two or more antibodies.

[0037] For the terms "for example" and "such as" and their grammatical equivalents, the phrase "but not limited to" should be understood to include the following term unless otherwise expressly indicated. As used herein, the term "about" and the indicator "~" are meant to account for variations due to experimental error. All measured values ​​reported herein should be understood to be modified by the term "about", whether or not the term is explicitly used, unless otherwise expressly indicated. As used herein, the singular forms "a", "an", and "the" include plural referents unless the context clearly indicates otherwise.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the invention belongs. Methods and materials for use in the present invention are described herein; other suitable methods and materials known in the art may also be used. The materials, methods and examples are illustrative only and are not intended to be limiting. All publications, patent applications, patents, sequences, database entries and other references mentioned herein are incorporated by reference in their entirety. In the event of a conflict, the present specification (including definitions) shall prevail.

[0039] Antibody molecules, such as monoclonal antibody molecules, can be co-formulated to treat one or more diseases or disorders in a patient, including a human patient.The terms "patient" and "subject" are used interchangeably herein.

[0040] A co-formulated drug product (DP) may include a co-formulated drug substance (cFDS) (also referred to herein as a co-formulation) containing two or more (e.g., three) human monoclonal antibody (mAb) molecules. The cFDS is prepared by mixing purified mAbs in a predetermined ratio. A method is needed by regulatory agencies to quantify each of the individual mAbs in the cFDS.

[0041] The mAb molecules in the co-formulation may be similar to each other: they may be immunoglobulins (such as IgG1) of approximately the same molecular weight (eg, about 145 kDa), with similar protein structure and charge properties.

[0042] The methods disclosed herein for quantifying the molecular weight of similar mAbs in co-formulations are precise, accurate, reproducible, suitable for use in a quality control environment; do not use expensive equipment; and do not require tedious sample preparation.

[0043] method

[0044] The present disclosure provides a method for quantifying the amount of antibody molecules of a mixture comprising two or more antibody molecules. The method may include separating each of two or more antibody molecules of a mixture by hydrophobic interaction chromatography high performance liquid chromatography (HIC-HPLC), and quantifying the amount of each antibody molecule; wherein the molecular weight of each antibody molecule is within 15kDa of any other antibody molecule in the mixture, and each antibody molecule differs from another antibody molecule in the mixture on the Kyte & Doolittle hydrophobicity scale (see, e.g., Kyte and Doolittle, Journal of Molecular Biology (J.Mol.Biol.), No. 157, pp. 105 to 132 (1982)) by more than about 0.25 units, or each of the antibody molecules is eluted with a slightly overlapping running time different from other antibody molecules in the mixture when running separately with HIC-HPLC, or both. In certain embodiments, for example, the relative hydrophobicity of each antibody molecule differs from each other antibody molecule in the mixture on the Kyte & Doolittle hydrophobicity scale by about 0.5 to about 1.0 units. In certain other embodiments, each of the antibody molecules, when run separately with HIC-HPLC, elutes with a slightly overlapping run time different from that of the other antibody molecules in the mixture. In certain embodiments, the mixture comprises a plurality of antibody molecules (e.g., two, three, four or five antibody molecules).

[0045] If two antibody molecules of a mixture have significant overlap in a chromatogram (the chromatogram monitors the elution curve of the antibody molecule by absorbance relative to the elution time) when running with HIC-HPLC, then this indicates that the two antibody molecules are eluted at a close time, resulting in poor separation in the column so that the antibody molecule is not purified (e.g., less than 50% pure). In such cases, individual antibodies are not fully quantifiable due to this significant overlap. In certain embodiments, two antibody molecules of a mixture have slightly overlap or no overlap in a chromatogram (the chromatogram monitors the elution curve of the antibody molecule by absorbance relative to the elution time) when running with HIC-HPLC, wherein the antibody molecule is eluted at a time that is not close, thereby resulting in separation so that the antibody molecule is significantly purified, and each antibody molecule is at least 90% pure, 91% pure, 92% pure, 93% pure, 94% pure, 95% pure, 96% pure, 97% pure, 98% pure, 99% pure or 100% pure.

[0046] In some embodiments, the methods disclosed herein are used to quantify individual populations of antibodies (also referred to as "antibody molecules") of an antibody mixture, the mixture comprising different populations of two or more antibodies, wherein the different populations have at least one amino acid difference therebetween. In some embodiments, the population of antibodies may have an amino acid sequence identical to the amino acid sequence of another population in the antibody mixture, but they may differ in post-translational modifications. In certain embodiments, the mixture is a co-formulated composition, such as or including a co-formulated drug substance (cFDS). In certain embodiments, the mixture comprises an excipient. In certain embodiments, the mixture comprises sucrose, and may be a sucrose drug substance comprising mAb molecules. Each antibody in the antibody mixture will have a similar molecular weight within 1 kilodalton (kDa), 2kDa, 3kDa, 4kDa, 5kDa, 10kDa or 15kDa of each other. In another aspect, the average molecular weight of all antibodies will be about 150kDa. This method can also be used with two or more trap molecules with similar molecular weights. In certain additional embodiments, the co-formulated composition comprises a VEGF trap (see, e.g., U.S. Patent No. 9,265,827 and U.S. Patent Application No. 14 / 943,490, both of which are incorporated herein by reference in their entirety). This method can also be used to monitor the concentration of each of the mAbs in a mixture (e.g., a co-formulation) to monitor storage stability. In certain embodiments, this method is used to separate antibody molecules that cannot be separated by another chromatographic method, such as by reverse phase high pressure liquid chromatography (HPLC) or ultra high pressure liquid chromatography (UPLC).

[0047] Hydrophobic interaction chromatography (HIC) separates antibodies based on the difference in their surface hydrophobicity with a reduced salt gradient. Separation using HIC is based on the reversible interaction between the antibody and the hydrophobic ligand that is bound to the chromatographic matrix. Although the hydrophobic amino acids of proteins and peptides are usually located away from the molecular surface, biomolecules have some hydrophobic groups that are exposed to allow interaction with the hydrophobic ligand on the medium. Hydrophobic interactions are enhanced by buffers with high ionic strength.

[0048] Any suitable HIC-HPLC column can be used for the methods disclosed herein, including but not limited to Dionex ProPacHIC-10, mAbPac HIC-10, mAbPac HIC-20, mAbPac HIC-Butyl (Dionex, Thermo Fisher Scientific, Sunnyville, CA). The buffer can be any suitable buffer. In certain embodiments, HIC-HPLC is performed in a buffer having a pH value between about pH 5.0 and about pH 7.0, or between about pH 6.0 and about pH 7.0, or between about pH 6.5 and about pH 7.5. In certain embodiments, the column is Dionex MabPac HIC-10 (100 x 4.6 mm, 5 μm, pore size 1000 Å°) and ProPac HIC-10 (100 x 4.6 mm, 5 μm, pore size 300 Å°).

[0049] A mixture comprising two or more antibodies is run with HIC-HPLC. The antibodies are eluted at independent times (run times). The antibodies are monitored by measuring their absorbance at, for example, 280 nm on the UV spectrum. A chromatogram can be used to monitor the run and record the run. In certain embodiments, the buffer used for HIC-HPLC is different gradients of 1 M ammonium sulfate (from 100% to 0%) in 100 mM phosphate buffer.

[0050] In certain embodiments, antibody molecules are quantified by comparing their absorbance at 280nm to a standard curve. The standard curve can be constructed by determining the absorbance of known antibodies of known concentrations derived from HIC-HPLC operation at, for example, 280nm. The greater the absorbance (or "optical density"), the higher the protein concentration. These data on the known concentrations of the antibody are used to prepare a standard curve, so that the concentration is depicted on the X axis and the absorbance obtained from the chromatogram of the elution curve of the antibody derived from the HIC-HPLC operation is depicted on the Y axis. Samples containing two or more antibodies of known concentrations are run, including antibodies derived from the standard curve. To analyze the data, the antibody is separated by HIC-HPLC. The elution curve of the antibody is shown on the chromatogram, and the antibody concentration is represented by the absorbance monitored over time by a spectrophotometer. The absorbance of each antibody (as identified by its elution time (run time)) is used to locate the corresponding absorbance value on the standard curve. The corresponding X axis value for this point on the standard curve is the concentration of the antibody in the sample.

[0051] In some aspects of the present disclosure, methods for quantifying individual antibodies in a mixture disclosed herein may be useful if the relative surface hydrophobicity of each antibody in the mixture differs from the relative surface hydrophobicity of each other antibody in the mixture by greater than about 0.25 units on the Kyte & Doolittle hydrophobicity scale (such as by about 0.5 units to about 1.0 units). The relative surface hydrophobicity of antibodies can be determined / estimated by a variety of methods.

[0052] For example, HIC-HPLC can be used to determine the relative surface hydrophobicity of different antibody molecules. Each of the antibody molecules is run individually with HIC-HPLC, and the run time (elution time) of each antibody is determined.

[0053] In certain embodiments, the difference in surface hydrophobicity between antibodies in a mixture is determined by one or more of the following methods: calculation of surface hydrophobicity based on protein structure or structural model, rapid screening for solubility in ammonium sulfate or PEG 8000, and rapid screening for molecular interactions by affinity self-capture interaction nanoparticle spectroscopy (AC-SINS).

[0054] Relative surface hydrophobicity can be determined based on the known structure of the antibody (based on, for example, a crystal structure or an NMR structure) or a structural model. Such calculation methods are known in the art.

[0055] Relative surface hydrophobicity can be calculated / estimated by determining the solubility of the antibody molecule, such as in ammonium sulfate or PEG 8000 (e.g., by a rapid screening method for solubility under conditions of increased hydrophobicity). The more soluble the protein, the lower the hydrophobicity (see, e.g., Kramer et al., Biophysical Journal, Vol. 102, April, pp. 1907-1915 (2012)).

[0056] Relative surface hydrophobicity can be calculated / estimated by affinity self-capture interaction nanoparticle spectroscopy (AC-SINS) for rapid screening of molecular interactions (see, e.g., Estep et al., mAbs, 7:3, pp. 553-561 (2015)). AC-SINS is a method that coats gold nanoparticles with polyclonal anti-human antibodies, uses these conjugates to immobilize human mAbs, and assesses mAb self-interactions by measuring the plasmon wavelength of the antibody conjugates as a function of ammonium sulfate concentration.

[0057] In certain other embodiments, antibodies in a mixture that can be separated using the described methods will have similar size or overall charge as determined by one or more of the following methods: protein structure, or structural modeling based on sequence and other known protein structures, calculated overall protein charge based on protein sequence, calculated hydrophobicity based on protein sequence, and protein sequencing.

[0058] In certain embodiments, the antibodies in the mixture have at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% homology protein structure or structural model based on sequence and other known protein structures. In certain embodiments, the antibodies in the mixture have calculated protein charge properties within 3, 4 or 5, 10 or 15 basic units of each other, wherein the protein charge properties are calculated based on the protein sequence. In certain embodiments, two or more antibodies in the mixture have at least 80%, at least 85%, at least 90%, at least 95%, at least 98% or at least 99% identity between their protein sequences.

[0059] In one aspect, if there are large differences between antibody molecules (as calculated or predicted by protein structure, sequence-based structural models, protein sequence, and other known aspects of protein structure), then methods using HIC-HPLC as disclosed herein to quantify antibodies in a mixture may not be applicable. If there are differences in the hydrophobicity profiles of the antibodies, then methods using HIC-HPLC may still facilitate separation of antibody species.

[0060] On the other hand, if there are large differences between antibody molecules (calculated / estimated by calculating the overall protein charge properties of each antibody molecule based on the protein sequence of each antibody molecule), then the method of quantifying antibodies in a mixture using HIC-HPLC as disclosed herein may not be applicable. If there are differences in the hydrophobicity profiles of the antibodies, then the method using HIC-HPLC may still facilitate the separation of antibody species.

[0061] On the other hand, if there are large differences between antibody molecules (calculated / estimated by detecting protein charge or size), then the method of quantifying antibodies in a mixture using HIC-HPLC as disclosed herein may not be applicable. If there are differences in the hydrophobicity profiles of the antibodies, the method using HIC-HPLC may still be beneficial for the separation of antibody species.

[0062] In certain embodiments, one or more of the antibodies in the antibody mixture is a monoclonal antibody. In certain embodiments, one or more of the monoclonal antibodies are human monoclonal antibodies. In certain embodiments, two or more of the antibody molecules are of the same isotype. In certain embodiments, two or more of the antibody molecules are variants of each other. In certain other embodiments, two or more of the antibody molecules bind to the same antigen. mAbs may be complete antibody molecules.

[0063] In certain embodiments, the mixture is a co-formulated composition. In certain other embodiments, the co-formulated composition comprises two or more mAbs effective for treating Middle East Respiratory Syndrome (MERS) in human patients. Antibodies for MERS coronavirus (MERS-CoV) are disclosed, for example, in U.S. Patent Application No. 14 / 717,760 and International Application Publication No. WO2015 / 179535A1. In certain other embodiments, the co-formulated composition comprises mAbs effective in treating Ebola hemorrhagic fever in human patients. Antibodies for Ebola virus are disclosed, for example, in U.S. Patent Application No. 15 / 005,334 filed on January 25, 2016. In certain other embodiments, the co-formulated composition comprises mAbs effective in treating macular degeneration in human patients. In certain other embodiments, the co-formulated composition comprises U.S. Provisional Application No.

[0064] 62 / 302,907 disclosed mAb alirocumab and evinacumab. In certain embodiments, the co-formulated composition comprises PD1 antibodies and other immuno-oncology antibody products, such as bispecific antibodies (see, e.g., U.S. Provisional Application No. 62 / 270,749, U.S. Patent Application No. 15 / 386,443 and U.S. Patent Application No. 15 / 386,453 PD-1 and CD3xCD20 and U.S. Provisional Application 62 / 365,006 and U.S. Patent Application No. 15 / 289,032 PD-1 and Lag3 disclosure). In certain embodiments, the co-formulated composition comprises anti-Zika virus antibodies (see, e.g., U.S. Provisional Application No. 62 / 363,546). In certain other embodiments, the co-formulated composition comprises trevogrumab antibodies and Activin A antibodies (see, e.g., U.S. Patent No. 8,871,209). In certain embodiments, two or more antibodies in the co-formulated composition can treat infectious diseases in human patients. The disclosures of each of the above-cited patent documents in this section are hereby incorporated by reference.

[0065] The anti-MERS mAb in the co-formulation can bind to, for example, the spike protein of MERS-CoV (e.g., the spike protein of MERS-CoV isolate EMC / 2012). The epitope of the spike protein can be within the receptor binding domain of the spike protein (e.g., amino acids selected from amino acids 367 to 606 of GenBank Accession No. AFS88936.1). The anti-MERS mAb in the co-formulated composition can be: a fully human monoclonal antibody that binds to the MERS-CoV spike protein; one that interacts with one or more amino acid residues in the receptor binding domain of the MERS-CoV spike protein selected from amino acid residues 367 to 606 of GenBank Accession No. AFS88936.1; one that binds to the MERS-CoV spike protein with a dissociation constant (Ko) of less than 18.5 nM (as measured in surface plasmon resonance analysis); or one that blocks more than 90% of the binding of the MERS-CoV spike protein to dipeptidyl peptidase 4 (DPP4).

[0066] A co-formulated composition can be any composition comprising two or more antibodies directed to the same or different targets, and effective for treating the same or different diseases or conditions in a patient, including a human patient.

[0067] Pharmaceutical compositions and preparations

[0068] Compositions comprising two or more antibody molecules can be formulated as pharmaceutical compositions (e.g., DP) for administration to a subject. Pharmaceutical compositions can comprise, for example, three antibody molecules. Any suitable pharmaceutical composition and formulation, as well as suitable methods for formulation and suitable routes and suitable sites for administration are within the scope of the present invention. In addition, unless otherwise indicated, any suitable dosage and frequency of administration are contemplated.

[0069] The mAbs in the co-formulation are purified by methods known in the art prior to co-administration.The co-formulation of two or more antibodies may be any suitable co-formulation.

[0070] The pharmaceutical composition / co-formulation may include a pharmaceutically acceptable carrier (i.e., excipient). "Pharmaceutically acceptable carrier" refers to and includes any and all solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents, diluents, glidants, and the like. The composition may include a pharmaceutically acceptable salt, e.g., an acid addition salt or a base addition salt (see, e.g., Berge et al., J Pharm Sci 66: 1-19 (1977)). The composition may include sucrose or may be coated with sucrose where appropriate.

[0071] In certain embodiments, protein compositions can be stabilized and can be formulated as solutions, microemulsions, dispersions, liposomes, lyophilized cakes, solids, etc. Aseptic injection solutions can be prepared as needed by incorporating two or more mAbs of the required amount into a suitable solvent with one or a combination of the components listed above, followed by filtration sterilization. Generally speaking, dispersions are prepared by incorporating two or more mAb molecules into a sterile vehicle comprising an alkaline dispersion medium and other required components derived from those listed above. The appropriate fluidity of the solution can be maintained, for example, by using a coating (such as lecithin), by maintaining the required particle size in the case of a dispersion and by utilizing a surfactant. The long-acting absorption of the injection composition can be achieved by including an agent (e.g., monostearate and gelatin) that delays absorption in the composition.

[0072] Examples

[0073] In order to better understand the present invention, the following examples are explained. These examples are for illustrative purposes only and are not to be considered as limiting the scope of the present invention in any way.

[0074] The anti-Ebola mAbs in the co-formulations used in Examples 1 and Examples 3 to 7 were whole fully human IgG1 monoclonal antibodies. The three mAbs (mAb A, mAb B, and mAb C) had similar molecular weights (e.g., about 145 kDa), protein structures, and charge properties (e.g., pI differences of about 0.6 or less, as determined by protein sequence). The isoelectric point (pI) of mAb A was determined to be 9.0, the pI of mAb B was determined to be 8.5, and the pI of mAb C was determined to be 9.1.

[0075] The anti-MERS mAb in the co-formulation used in Example 1 is an integral fully human anti-Ebola mAb. The two mAbs have similar molecular weights (e.g., within about 15 kDa of each other) and charge properties (see U.S. Patent Publication Nos. US2015 / 0337029, WO2015 / 179535A1, and U.S. Patent Publication No.

[0076] 14 / 717,760, the disclosures of each of which are hereby incorporated by reference herein).

[0077] Example 1

[0078] The HIC-HPLC method was used to quantify three anti-Ebola monoclonal antibodies (mAb1, mAb 2 and mAb 3) of similar molecular weight, protein structure and charge properties from the co-formulation by first isolating the three mAbs from the co-formulation and then quantifying each of them.

[0079] A Dionex ProPac HIC-10 column, Cat #063655 (Dionex, Thermo Fisher Scientific, Sunnyville, CA), 4.6 x 100 mm was used.

[0080] Preparation of mobile phase

[0081] The mobile phase includes mobile phase A and mobile phase B. Mobile phase A includes 1M ammonium phosphate and 100mM phosphate at pH 7.0. The preparation of mobile phase A includes: dissolving 13.8g of sodium dihydrogen phosphate monohydrate (NaH2PO4.H2O) and 132.1g of ammonium phosphate in 800mL of Milli Q; adjusting the pH to 7.0 with 50% NaOH; increasing the volume to 1000mL; and filtering the solution through a 0.22μM filter. Mobile phase B includes 100mM phosphate at pH 7.0. The preparation of mobile phase B includes: dissolving 13.8g of sodium dihydrogen phosphate monohydrate (NaH2PO4.H2O) in 900mL of Milli Q; adjusting the pH to 7.0 with 50% NaOH; increasing the volume to 1000mL; and filtering the solution through a 0.22μM filter.

[0082] HIC-HPLC method

[0083] HIC-HPLC was run through the aforementioned column at a flow rate of 0.5 mL / min. The column temperature was maintained at 30°C, and the co-formulation sample temperature was 5°C. The stop time of the column was 40 minutes. The 280 nm ultraviolet absorbance of the effluent was monitored using a UV detector. The following table 1 shows the mixing of mobile phase A and mobile phase B as the percentage of the mobile phase composition gradient to be introduced along with the column running time.

[0084] Table 1: Mobile phase gradient

[0085] Time (minutes) % Mobile Phase A % Mobile Phase B 0 60 40 4 60 40 19 0 100 32 0 100 33 60 40 40 60 40

[0086] Calibration of three individual mAbs

[0087] To calibrate the results for the three individual mAbs, a known amount of each mAb in the FDS was run. An FDS with a known concentration of each mAb was prepared and the concentrations were determined by reverse phase (RP) chromatography or Solo The FDS of each mAb is as follows:

[0088] Table 2: If you use RP or Solo The concentration of each mAb FDS measured

[0089]

[0090] A control sample was also prepared for calibration of the sequence. The control sequence had a concentration of 50.31 mg / mL, as determined by Solo Spectroscopy (eg, C Technologies, Inc.) was used. One injection of the control sample was 12.0 μL (603.72 μg).

[0091] Table 2 below lists the calibration samples used for each of the mAb FDS.

[0092] Table 2: Calibration samples

[0093]

[0094] Run the calibration sequence. During the calibration sequence, one injection of the control sample was introduced at the beginning of each sample collection (e.g., collection of mAb 1 FDS injection, collection of mAb 2 FDS injection, or collection of mAb 3 FDS injection), after every 20 to 24 injections of mAb FDS, and at the end of each injection sequence.

[0095] During one calibration sequence, one injection of each known amount of mAb was performed. During another calibration sequence, four injections of each known amount of mAb were performed to determine the repeatability of the injections.

[0096] A standard calibration curve was constructed for each mAb. The calibration curve for each mAb had an R2 ≥ 0.999, and the variability of repeated injections of the same amount of each mAb was ≤ 1%.

[0097] Comparison Instructions

[0098] A control HIC-HPLC run of each known mAb FDS was performed. The percent recovery of each known mAb FDS was calculated by measuring the concentration by HIC-HPLC and then dividing the concentration measured by HIC-HPLC by the known mAb concentration in Table 2. The percent recovery was calculated to be 90% to 110% of the known concentrations of all three known mAb FDSs.

[0099] Unknown sample

[0100] A sample co-formulation with a total of 600 μg of mAb (e.g., 12 μL of a 50 mg / mL mAb sample) was run by HIC-HPLC. For example, a 12 μL co-formulation including a total of 600 μg of mAb may include the equivalent of three injections of 200 μg / individual mAb injection. The individual mAbs in the co-formulation are mAb 1, mAb 2, and mAb 3.

[0101] Duplicate runs of samples were performed and the average concentration (mg / mL) of each individual mAb in the co-formulation and the ratio of each mAb to the total amount of mAb are reported.

[0102] Figure 1 A chromatogram showing a HIC-HPLC run of this co-formulation. Each mAb is shown by an independent line, and the cFDS as a whole is also shown by a line. The three antibodies are well separated by HIC-HPLC, with peaks slightly overlapping.

[0103] Example 2

[0104] Multiple HIC-HPLC methods were used to quantify two anti-MERS monoclonal antibodies of similar molecular weight and charge properties from co-formulations by first isolating the two mAbs from the co-formulation and then quantifying each of them.

[0105] mAb drug substance

[0106] Two mAb (mAb 1 and mAb 2) drug substances were combined. mAb 1 drug substance contained 52.3 mg / ml of mAb 1 and 10 mM His at pH 5.5 (Regeneron Pharmaceuticals, Tarrytown, NY). mAb 2 drug substance contained 40.4 mg / ml of mAb 2 and 10 mM His at pH 6.0 and contained 5% (w / v) sucrose (Regeneron Pharmaceuticals, Tarrytown, NY).

[0107] HIC-HPLC method

[0108] Seven HIC-HPLC methods were run. In each method, 50 μg of sample (mAb 1 and mAb 2) was loaded. The effluent was monitored at 280 nm using UV detection. Two solvents were used for the mobile phase in different ratios. Solvent A was 1 M (NH4)SO4 at pH 5.5, 20 mM acetic acid, and solvent B was 20 mM acetic acid at pH 5.5.

[0109] 1. Method 1

[0110] Table 3.1: Method 1 gradient and flow rate information

[0111] Time (minutes) Flow rate (mL / min) %A of mobile phase %B of mobile phase start 0.500 100 0 4 0.500 100 0 29 0.500 0 100 40 0.500 0 100 41 0.500 100 0 50 0.500 100 0

[0112] Figure 2 The chromatogram for this run is shown. The two antibodies are somewhat separated but with significant overlap.

[0113] 2. Method 2 - 0.5% per minute

[0114] Table 3.2: Method 2 gradient and flow rate information

[0115] Time (minutes) Flow rate (mL / min) %A of mobile phase %B of mobile phase start 0.500 40 60 4 0.500 40 60 40 0.500 20 80 41 0.500 0 100 51 0.500 0 100 52 0.500 40 60 56 0.500 40 60

[0116] Figure 3 The chromatogram for this run is shown. The two antibodies are somewhat separated but with significant overlap.

[0117] 3. Method 3 - 0.2% per minute

[0118] Table 3.3: Method 3 gradient and flow rate information

[0119] Time (minutes) Flow rate (mL / min) %A of mobile phase %B of mobile phase start 0.500 39 61 4 0.500 39 61 49 0.500 39 70 50 0.500 0 100 60 0.500 0 100 61 0.500 39 61 65 0.500 39 61

[0120] Figure 4 The chromatogram for this run is shown. The two antibodies are somewhat separated but with significant overlap.

[0121] 4. Method 4 – 2% per minute

[0122] Table 3.4: Method 4 gradient and flow rate information

[0123] Time (minutes) Flow rate (mL / min) %A of mobile phase %B of mobile phase start 0.500 100 0 4 0.500 100 0 34 0.500 40 60 35 0.500 0 100 45 0.500 0 100 46 0.500 100 0 50 0.500 100 0

[0124] Figure 5 The chromatogram for this run is shown. The two antibodies are somewhat separated, but there is still some overlap.

[0125] 5. Method 5 – 1% per minute

[0126] Table 3.5: Method 5 gradient and flow rate information

[0127] Time (minutes) Flow rate (mL / min) %A of mobile phase %B of mobile phase start 0.500 80 20 4 0.500 80 20 34 0.500 50 50 35 0.500 0 100 45 0.500 0 100 46 0.500 80 20 50 0.500 80 20

[0128] Figure 6 The chromatogram for this run is shown. The two antibodies are somewhat separated but with some overlap.

[0129] 6. Method 6 – 1% per minute

[0130] Table 3.6: Method 6 gradient and flow rate information

[0131] Time (minutes) Flow rate (mL / min) %A of mobile phase %B of mobile phase start 0.500 90 10 4 0.500 90 10 44 0.500 50 50 45 0.500 0 100 55 0.500 0 100 56 0.500 90 10 60 0.500 90 10

[0132] Figure 7The chromatogram for this run is shown. The two antibodies are somewhat separated but with some overlap.

[0133] 7. Method 7 – 0.5% per minute

[0134] Table 3.7: Method 7 gradient and flow rate information

[0135] Time (minutes) Flow rate (mL / min) %A of mobile phase %B of mobile phase start 0.500 87 130 4 0.500 87 13 74 0.500 52 48 75 0.500 0 100 85 0.500 0 100 86 0.500 87 13 90 0.500 87 13

[0136] Figure 8 The chromatogram for this run is shown. The two antibodies are somewhat separated but with some overlap.

[0137] Example 3

[0138] A HIC separation of a co-formulation of three anti-Ebola mAbs of similar molecular weight, protein structure, and charge properties was developed and evaluated on an Agilent 1100 HPLC system.

[0139] The HIC-HPLC method utilized a Dionex ProPac HIC-10 column with a binary mobile phase including mobile phase A (100 mM phosphate, 1 M ammonium sulfate, pH 7.0) and mobile phase B (100 mM phosphate, pH 7.0) at a flow rate of 0.5 mL / min. The HIC method was quantified using a five-point calibration curve for each mAb. Accuracy (% recovery) was determined by comparing the measured concentration to the theoretical concentration of each mAb. A gradient of 600 to 0 mM ammonium sulfate was used to run the column.

[0140] Fig. 9 The chromatogram of the HIC-HPLC run is shown. As can be seen by the three distinct peaks, the three antibodies are well separated and their peaks have little to no overlap.

[0141] Reproducibility and precision of HIC-HPLC

[0142] Reproducibility was assessed by analyzing triplicate cFDS samples at 70%, 100%, 125%, 150% of the target amount (600 μg / cFDS injection). The relative standard deviation (RSD) of the reproducibility was determined to be ≤ 0.2%.

[0143] Table 4: Repeatability data

[0144]

[0145] The reproducibility study was repeated on different days using different batches of columns and different batches of mobile phase. The RSD of the intermediate precision was determined to be less than or equal to 1.7%.

[0146] Table 5: Intermediate accuracy

[0147]

[0148] Linearity of HIC-HPLC

[0149] Linearity was determined by analyzing triplicate cFDS samples at 9 levels (75 to 1200 μg of cFDS; 25 to 400 μg of individual mAbs). Plots of peak areas of individual mAbs versus injected amount ( Fig.10 ) to obtain a linear curve: from 75 to 300 μg, R 2 ≥0.999; at 25 to 400 μg, R 2 ≥0.98.

[0150] Scope of HIC-HPLC

[0151] The range was such that the method demonstrated acceptable precision (≤1.7%), accuracy (93-106%), and linearity (R 2 The interval between the lowest and highest concentrations was ≥ 0.999). The range was determined to be 300 to 900 μg for cFDS samples (100 to 300 μg for each individual mAb). Fig.11 Graph showing % recovery of mAb versus amount of individual mAb.

[0152] Accuracy of HIC-HPLC of different batches of cFDS

[0153] Data obtained from intra-day and inter-day precision (cFDS Batch #1) and the study range (cFDS Batch #2) were used to assess method accuracy. Fig.12 Figure 2 is a graph of the % recovery of mAbs versus the amount of individual mAbs. The % recovery was within 93% to 106% at 75%, 100%, 125%, 150% of the target amount (600 μg of total protein, 200 μg of each mAb). The accuracy (% recovery) was determined to be 93% to 106%.

[0154] Accuracy of HIC-HPLC of individual mAbs at various ratios in cFDS

[0155] CFDS were formulated by mixing individual mAbs at 1:1:1, 2:1:1, 1:2:1, 1:1:2. Each cFDS was analyzed at 100% of the target amount (600 μg / injection, n=3). Fig.13 % recovery of mAb at different cFDS ratios relative to each individual mAb. % recovery ranged from 94% to 106% with a reproducibility of ≤ 0.3%.

[0156] Table 6: HIC-HPLC data summary

[0157]

[0158] Application of HIC-HPLC in co-formulation development

[0159] The use of HIC-HPLC to quantify each of the mAbs in a co-formulation or DP can have various applications in co-formulation development. For example, the concentration of each mAb in a co-formulation or DP can be monitored in storage stability studies.

[0160] Fig.14 A graph is shown monitoring the storage stability of three mAbs in DP as a function of time (months). HIC-HPLC is used to quantify the amount and thus concentration of each of the three mAbs at six time points over one year. Example data collection points from one time point in this study include measurements of the concentration of mAb A at 16.6 mg / mL, mAb B at 17.5 mg / mL, and mAb C at 17.1 mg / mL. In this case, the storage stability shows that there is no measurable change (≤4%) in the concentration of each of the mAbs over the monitoring period.

[0161] The HIC-HPLC method had acceptable reproducibility (≤0.2%), intermediate precision (≤1.7%), accuracy (93 to 106% recovery), and linearity (R 2 ≥0.999) and can be used to determine the concentration of each of the three mAbs in cFDS and DP.

[0162] The HIC-HPLC method can be used as a release method for quantifying each of the three mAbs in co-formulated DS and DP. This method can also be used to support formulation development.

[0163] Example 4

[0164] A cation exchange ultra-high pressure liquid chromatography (CEX-UPLC) separation of a co-formulation of three anti-Ebola mAbs of similar molecular weight, protein structure, and charge properties was developed and evaluated.

[0165] The CEX-UPLC method utilized a YMC-BioPro SP-F column with mobility containing 200 mM MES buffer and a 10 to 120 mM NaCl gradient with a pH of 6.5. Six point standard calibration curves were prepared for each of the three mAb molecules (mAb A, mAb B, mAb C). The linearity of each calibration curve was determined with R 2≥0.99. Three mAb molecules were separated from a co-formulation (ie, a mixture) using CEX-UPLC and a UPLC chromatogram was generated. The chromatogram shows the separation of each mAb with some overlap between mAb A and mAb C.

[0166] The reproducibility of the method was evaluated by analyzing triplicate samples of co-formulations run at three different amounts (150 μg, 225 μg, and 300 μg). Accuracy (% recovery) was determined by comparing the measured concentration of each sample to the theoretical value. The linear range of quantification for each mAb was determined to be 50 to 100 μg for each antibody (except mAb A), with R 2 ≥0.973 and an accuracy of 75% to 91%.

[0167] Fig.15 Shown is a chromatogram of a CEX-UPLC run separating 3 mAbs from a co-formulation. The three antibodies were separated with a small overlap between mAb A and mAb C.

[0168] Reproducibility and accuracy of CEX-UPLC

[0169] Reproducibility was evaluated by analyzing triplicate cFDS samples at five different percentages (25% to 100%) of the target amount (600 μg / cFDS injection). The RSD of the reproducibility was determined to be between 3.7% and 21.4%.

[0170] Table 7: Repeatability and accuracy data

[0171]

[0172]

[0173] *These data points are outside the scope of analysis for this method.

[0174] Linearity of CEX-UPLC

[0175] Data from the reproducibility study were used to evaluate the range and linearity of the CEX-UPLC method. Fig.16 ) generally obtained linear curves for each mAb. However, because the RSDs for the mAbs were above 10%, the data points for the mAbs at 100 μg were not statistically significant enough to conclude that there was a linear curve extending to 100 μg for mAb A.

[0176] Scope of CEX-UPLC

[0177] Some reproducibility (RSD ≤ 10%), accuracy (> 80%), and linearity (R 2 ≥0.97). However, the range of CEX-UPLC is not as clearly shown as that of HIC-HPLC.

[0178] Table 8: HIC-HPLC data summary

[0179]

[0180] The CEX-UPLC method separated the three mAbs based on their charge properties, with some overlap. The overlap meant that manual integration of the methods was required. The precision, accuracy and range of CEX-UPLC are not as optimal as those of HIC-HPLC.

[0181] Example 5

[0182] SE-UPLC of co-formulations containing three anti-Ebola mAbs of similar molecular weight was evaluated. SE-UPLC separated mAbs by size. A Waters Acquity H UPLC system was used. Two Waters BEH200 SEC columns were linked in series. The mobile phase consisted of 10 mM phosphate buffer at pH 6.0, and 1 M perchlorate. Fig.17 The chromatogram of SE-UPLC is shown. As shown, there is a significant overlap between the elution times of the three mAbs. Therefore, SE-UPLC is a suboptimal method for separating the three anti-MERS mAbs of similar molecular weight.

[0183] Example 6

[0184] iCIEF was evaluated for co-formulations of three anti-Ebola mAbs of similar molecular weight, protein structure, and charge properties. iCIEF separates proteins by their isoelectric point (pI). A ProteinSimple iCE3 Charge Variant Analyzer was used. 4% of pH 3 to 10 was used as ampholyte, and 2 M urea was used as buffer. Fig.18 The iCIEF curves generated using these instructions are depicted. As shown, there is a significant overlap between the elution times of mAb A and mAb C. Therefore, iCIEF is a suboptimal method for separating the three anti-MERS mABs.

[0185] Example 7

[0186] RP-UPLC of co-formulations of three anti-Ebola mAbs of similar molecular weight, protein structure and charge properties was evaluated. RP-UPLC separated mAbs by hydrophobicity. A Waters Acquity UPLC system was used. A ZORBAX 300SB-C8 column was used and the column was run at 80°C. The mobile phase contained 60 to 90% acetonitrile in 0.1% TFA. Fig.19 The chromatogram of RP-UPLC is shown. As shown, there is a significant overlap between the elution times of mAb A and mAb B. Therefore, RP-UPLC is a suboptimal method for separating three anti-MERS mAbs of similar molecular weight.

[0187] Example

[0188] The foregoing description discloses only exemplary embodiments of the present invention.

[0189] It should be understood that although the invention has been described in conjunction with its detailed description, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages and modifications are within the scope of the appended claims. Therefore, although only certain features of the invention have been shown and described, many modifications and variations will occur to those skilled in the art. Therefore, it should be understood that the appended claims are intended to cover all such modifications and variations that fall within the true spirit of the invention.

Claims

1. A method for quantifying a first antibody and a second antibody in a mixture, the method comprising: Separating the first antibody from the second antibody using hydrophobic interaction chromatography-high performance liquid chromatography (HIC-HPLC); quantifying the amount of the first antibody; as well as quantifying the amount of the second antibody; wherein each of the first antibody and the second antibody: When run alone by HIC-HPLC, elutes at a different run time than the other antibodies in the mixture; and It is a bispecific antibody. 2 . The method according to claim 1 , wherein the first antibody binds to two or more proteins selected from the following group: PD-1, CD3, CD20 and LAG3.

3. The method of claim 2, wherein the second antibody binds to two or more target proteins selected from the following group: PD-1, CD3, CD20 and LAG3.

4. The method of claim 1, wherein the first antibody, the second antibody, or both comprise a first antigen binding arm that specifically binds CD20 and a second antigen binding arm that specifically binds CD3.

5. The method of claim 1, wherein during a HIC-HPLC run, the first antibody is eluted at a first run time, the second antibody is eluted at a second run time, and the first and second run times do not overlap.

6. The method of claim 1, wherein the first antibody and the second antibody are human monoclonal antibodies.

7. The method of claim 1, wherein the first antibody and the second antibody have protein sequences that are at least 90% homologous; or The first antibody and the second antibody have a protein structure that is at least 90% homologous as determined by their protein sequences.

8. The method of claim 1, wherein the mixture is a co-formulated pharmaceutical product.

9. The method of claim 8, wherein the co-formulated drug product further comprises a third antibody that binds to one or more molecules selected from the group consisting of PD-1, CD3, CD20, and LAG3.

10. A method for quantifying an antibody in a co-formulated drug product comprising a plurality of antibodies, the method comprising: generating a standard curve for each of the multiple antibodies; Separating each of the plurality of antibodies using hydrophobic interaction chromatography-high performance liquid chromatography (HIC-HPLC), wherein each of the plurality of antibodies binds to two or more proteins selected from the group consisting of: PD-1, CD3, CD20, and LAG3; and After separation of each antibody, the amount of each antibody was quantified using a standard curve.

11. The method of claim 10, wherein each antibody in the plurality of antibodies is a bispecific antibody.

12. The method of claim 11, wherein each bispecific antibody comprises a first antigen binding arm that specifically binds CD20 and a second antigen binding arm that specifically binds CD3.

13. The method of claim 10, wherein each antibody elutes at a different run time than other antibodies in the plurality of antibodies when run separately by HIC-HPLC.

14. The method of claim 10, wherein the plurality of antibodies comprises three human monoclonal antibodies.

15. The method of claim 10, wherein the weight of each antibody in the plurality of mixtures is within 15 kDa of the weight of each other antibody in the plurality of mixtures.

16. The method of claim 10, wherein a first antibody in the plurality of antibodies binds to a first epitope of the antigen; and a second antibody in the plurality of antibodies binds to a second epitope of the antigen.

17. The method of claim 16, wherein the first antibody and the second antibody are human monoclonal antibodies; The first antibody has a first weight, the second antibody has a second weight, and the first weight differs from the second weight by less than 15 kDa; and The first antibody and the second antibody have protein sequences that are at least 90% homologous.

18. The method of claim 10, further comprising determining the surface hydrophobicity of each antibody in the plurality of antibodies prior to generating the chromatogram, wherein the surface hydrophobicity is determined by one or more methods selected from the group consisting of: calculation of surface hydrophobicity based on protein structure or structural model, rapid screening for solubility in ammonium sulfate or PEG 8000, or rapid screening for molecular interactions by affinity self-capture nanoparticle spectroscopy (AC-SINS).

19. The method of claim 18, wherein the surface hydrophobicity is a surface hydrophobicity calculated based on a protein structure or a structural model, and the surface hydrophobicity of each antibody in the plurality of antibodies differs from the surface hydrophobicity of another antibody in the co-formulated drug product by more than about 1.0 unit on the Kyte & Doolittle hydrophobicity scale.

Images