EA043996B1 - ANALYSIS TO DETERMINE PROTEIN SELF-ASSOCIATION POTENTIAL USING CONCENTRATION-DEPENDENT NANOPARTICLE-BASED SELF-INTERACTION SPECTROSCOPY - Google Patents

Description

Field of technology

This invention generally relates to the selection of therapeutic proteins for development and commercialization, and more particularly to the selection of proteins that have a greater likelihood of easily achieving high protein concentrations with a concomitant lower likelihood of aggregate formation. This invention particularly relates to the evaluation of protein colloidal interactions and the selection of proteins for high concentration preparation.

State of the art

Large biological molecules - "biotherapeutic agents" - represent an important class of drugs. For example, monoclonal antibodies have exceptional therapeutic specificity, long biological half-lives, and high safety profiles.

Given their size and complexity, biotherapeutics are difficult to manufacture and develop. Most biotherapeutics are formulated as lyophilized solids or liquid formulations, and the stability of these molecules is an important issue. Stability affects not only the shelf life of the final product, but also the production process.

A large molecule, by definition, has a large molecular weight. Therefore, several moles of a drug contain a relatively large mass. For example, an antibody has a molecular weight of about 150,000 g/mol, whereas a small molecule drug such as atorvastatin has a molecular weight of about 560 g/mol. Consequently, a significant mass of protein preparation is required to achieve an effective dose. As more drugs are formulated for subcutaneous injection, this large mass must be contained in a small volume. To provide therapeutically effective doses, subcutaneous formulations of biotherapeutic drugs must achieve high (about 50 mg/ml or more) or very high (about 100-250 mg/ml) concentrations, all while maintaining viscosity at an acceptable level.

As large molecules fill a small space and the distance between protein molecules decreases, the frequency of colloidal interactions, i.e. the effect of the field of one molecule on another increases. Protein self-interactions (colloidal interactions) are generally energetically weak, usually reversible, and nonspecific. These interactions are protein dependent and are influenced by pH, salts and other additives. In some cases, interactions may be generally attractive, repulsive, or neutral. The nature and magnitude of the interaction may depend on the protein concentration.

Colloidal interactions cover a wide range of interactions (potential energies), which are quantified using virial coefficients. B ₂₂ represents pair interactions, B ₂₂₂ represents three body interactions, B ₂₂₂₂ represents four body interactions, etc. A positive virial coefficient represents repulsive interactions, a negative virial coefficient represents attractive interactions, and a zero virial coefficient represents the ideal state.

Colloidal interactions are typically present in protein solutions that exceed ~2 mg/mL and become significant in solutions that exceed ~40 mg/mL. At lower concentrations, steric and electrostatic forces predominate. The situation becomes more complex as protein concentration increases. When protein concentration increases during production and preparation, colloidal interactions become suboptimal.

Colloidal interactions influence various downstream processes during protein synthesis. These forces can be beneficial or harmful depending on the nature and extent of the interactions. Colloidal interactions affect chromatographic performance, ultrafiltration and diafiltration (UF/DF), dialysis, solution viscosity and handling, and protein stability in solution. Colloidal interactions also affect long-term storage stability, which can result in the formation of oligomers and multimers, as well as protein aggregates. Therefore, estimating the virial coefficient of a protein provides important information before committing resources to the development of a specific protein therapeutic.

Virial coefficient analysis is a well-developed and very active area of research. Current methods for estimating the virial coefficient ( _B22 or the closely related _A2 ) for specific proteins include membrane osmometry, sedimentation equilibrium analytical ultracentrifugation, self-interaction chromatography, static light scattering (SLS), diffusion or sedimentation interaction parameters, and nanoparticle-based self-interaction spectroscopy ( SINS).

For example, Cohen & Benedek (US 4080264 A, published March 3, 1978) describe a quasi-elastic light scattering spectroscopic method for measuring antigen or antibody aggregation. Publicover & Vincze (US 7514938 B2, published April 7, 2009) describe the use of dielectric relaxation spectroscopy (DRS) to study the interaction and aggregation of micron- and submicron-scale particles coated with protein, including antibodies. Holman et al. (US 20070291265 A1, published December 20, 2007) describes a bifurcated fiber optic system for measuring light scattering and concentration signals for measuring macromolecular aggregation. Obrezanova et al. (mAbs, 7(2):352-363,

- 1 043996

2015) describes the use of size exclusion high performance liquid chromatography (SEHPLC) and oligomer detection assay, which is an optical density microtiter plate antibody capture assay, to systematically measure the aggregation propensity of more than 500 antibodies. Geoghegan et al. (mAbs, 8(5):941-950, 2016) describes the use of hydrophobic interaction chromatography (HIC) retention time, SINS affinity capture, and dynamic light scattering to measure the self-interaction, viscosity, and stability of monoclonal antibodies. A review of current methods used to assess colloidal protein interactions is presented by Geng et al. (J. Pharm. Sci., 103(11):3356-3363, 2014).

Nanoparticle-based self-interaction spectroscopy (SINS) for assessing self-association of monoclonal antibodies is described by Sule et al. (Biophys. I, 101:1749-1757, 2011). Briefly, antibodies are adsorbed onto gold nanoparticles, which are then combined with buffer in 96-well microtiter plates. As the nanoparticles aggregate due to self-association of the clumped antibodies, the absorption spectrum of the nanoparticles changes. This change in surface plasmon resonance correlates with particle association. The assay provides a binary readout of self-association or lack of self-association depending on whether the absorption peak shifts or changes in amplitude. Sule et al. (Mol. Pharmaceutics, 10:1322-1331, 2013) describe an improved SINS technique called affinity capture nanoparticle-based self-interaction spectroscopy (AC-SINS). Here, human polyclonal antibodies are first attached to the nanoparticles, then the coated particles are contacted with low concentration/low purity antibody samples, which are then captured by the anti-human coating(s). AC-SINS enables rapid screening of cell supernatants without the need for extensive antibody purification. AC-SINS, like SINS, provides a binary assessment of positive self-association or lack of self-association of a subject's antibodies.

Proteins are complex molecules and often exhibit unpredictable behavior under different conditions. The prior art offers some means for observing protein self-association, but does not offer a quick and accurate means for determining the propensity of a protein to self-associate under many different conditions. Some proteins are stable under some conditions but unstable under other conditions. There remains a need in the field to analyze the determination of the dynamic range of a protein's ability to attract and repel under changing conditions.

The essence of the invention

The inventors have developed a method for determining the self-assembly potential of a protein (ie, the inherent repulsive or attractive interaction of a protein) over a dynamic range of conditions. Therefore, this method can be understood as a method for determining the stability of self-associating proteins or a method for determining the stability of aggregated proteins. As further described herein, the protein may be a single type of protein or a combination of any proteins of different origins. The technique, called concentration-dependent self-interaction nanoparticle spectroscopy (CD-SINS), measures the propensity of a protein to self-assemble under varying concentrations and under varying conditions of ionic strength and pH. The method allows protein evaluation under high or very high concentration conditions and informs the selection of proteins for commercial development. The method also provides information on how to prepare a pharmaceutical composition containing high concentrations of one or more biopharmaceutical drugs while maintaining a long shelf life without association of the biopharmaceutical drug or drugs. The method also provides information on how to formulate a composition that satisfies the required physical property specifications, such as, for example, suitable rheology/viscosity to provide a particular desired route of administration. The invention also allows for faster screening of biopharmaceuticals using less material to obtain a suitable composition or select suitable candidates for further drug development.

In a first aspect, the invention provides a bioanalytical mixture containing a plurality of nanoparticles, a protein and a buffered salt. In a mixture, the protein exists in at least two phases: the adhesive (sticking) phase and the soluble phase.

The adhesive phase includes proteins that adhere to (i.e. coat) the nanoparticles. The soluble phase includes proteins that dissolve in the buffered saline solution. In some embodiments, the protein exists in a third phase, where the proteins are self-associated as a dimer, trimers, tetramers, or higher order multimers, including aggregated coated nanoparticles.

In one embodiment, the nanoparticle is a gold nanoparticle. In some embodiments, the nanoparticle has a diameter of from about 20 nm to about 100 nm. In a specific embodiment, the nanoparticle is a gold nanoparticle having a diameter of about 20 nm.

In one embodiment, each nanoparticle is coated with a protein. In some embodiments, most or all of the surfaces of the nanoparticles are saturated with protein, meaning that the surface is completely occupied by protein and there is no free surface available for protein attachment. In some embodiments, most or all of the surfaces of the nanoparticles are saturated with protein before adding

- 2 043996 by adding other components to the mixture, such as a suitable salt or buffered salt.

In one embodiment, the mixture contains from about 6x1011 to about 7x1011 nanoparticles per milliliter of mixture. In a specific embodiment, the mixture contains about 6.3 x 1011 nanoparticles per ml. Without being bound by theory and assuming that the average protein can be modeled as a 10 nm sphere, the theoretical upper limit for the number of 10 nm spheres required to cover a 20 nm sphere is about 30. One can estimate that approximately 15 to 20 antibody molecules are bound by one 20 nm nanoparticle. Therefore, it is estimated that ~2.5 μg/mL of antibody is the minimum concentration required to completely cover 6.3 x 1011 nanoparticles (20 nm)/mL. In some embodiments, the mixture contains from about 2 μg/ml to about 512 μg/ml protein.

In some embodiments, the protein included in the mixture is a therapeutic antibody. In some embodiments, the protein comprises an immunoglobulin Fc domain. In some embodiments, the protein is an antigen binding protein. Antigen-binding proteins include antibodies, antibody fragments, antibody derivatives, Fc fusion proteins, and receptor Fc fusion proteins. In one embodiment, the protein is a monoclonal antibody. In a more specific embodiment, the monoclonal antibody is a human or humanized antibody. In some embodiments, the monoclonal antibody may be a monospecific antibody or a bispecific antibody.

In some embodiments, the buffered salt includes a buffer and a salt. In some embodiments, the buffer provides ionic strength. In some embodiments, the salt buffers the mixture. In principle, the salt can be any suitable salt known in the art, and can be, for example, any chloride, bromide, phosphate, sulfate or ammonium salt or any combination thereof. Non-limiting examples include, for example, sodium chloride, potassium chloride, potassium phosphate, ammonium chloride, etc. In one embodiment, the salt includes sodium chloride. The salt (eg, sodium chloride) may be present in a concentration of from about 2 mM to about 300 mM. In specific embodiments, the salt is present in the mixture at a concentration of about 2 mM, about 20 mM, or about 200 mM.

In one embodiment, the buffer contains 2-(I-morpholino)ethanesulfonic acid (MES). In a specific embodiment, MES is present in the mixture at a concentration of about 10 mM and a pH of about 6.

In a second aspect, the invention provides a method for determining the ability of a protein to self-associate. In one embodiment, the method includes the steps of combining a protein, a plurality of nanoparticles, and a buffered salt to form a sample;

exciting the sample with light;

measuring the light transmitted through the sample; and calculating a first absorption intensity factor of the sample.

The process is repeated at least one more time, each time changing one or more parameters and obtaining a second, third, etc. absorption coefficient. The set of absorption intensity coefficients obtained for a given protein can be plotted for analysis. These parameters include salt type (ie, neutral, chaotropic, kosmotropic), salt concentration, pH, protein concentration, inclusion or exclusion of additional components. When the absorption intensity ratio exceeds a threshold value, the protein is considered suitable for dosing at high concentrations. A protein that is considered suitable for high concentration dosing is expected to remain stable at high concentration and be less prone to aggregation.

In some embodiments, the protein is a therapeutic antibody. In some embodiments, the protein comprises an immunoglobulin Fc domain. In some embodiments, the protein is an antigen binding protein. Antigen-binding proteins include antibodies, antibody fragments, antibody derivatives, Fc fusion proteins, and receptor-Fc fusion proteins. In one embodiment, the protein is a monoclonal antibody. In a more specific embodiment, the monoclonal antibody is a human or humanized antibody. In some embodiments, the monoclonal antibody may be a monospecific antibody or a bispecific antibody.

In some embodiments, the protein is added to the sample to a final concentration of from about 2 μg/ml to about 512 μg/ml.

In some embodiments, the nanoparticle is a gold nanoparticle. In some embodiments, the gold nanoparticle has a diameter of from about 20 nm to about 100 nm. In one embodiment, the diameter of the gold nanoparticle is about 20 nm.

In some embodiments, nanoparticles are added to the sample to a final concentration of about 5x1011 nanoparticles per ml to about 8x1011 nanoparticles per ml, about 6x1011 nanoparticles per ml to about 6.5x1011 nanoparticles per ml, or about 6.3x1011 nanoparticles per ml.

The buffered salt may contain a buffer, a buffer that imparts ionic strength, a salt, a salt that has buffering capacity, or a salt and a buffer. In one embodiment, the salt includes chloride

- 3 043996 sodium. The salt (eg, sodium chloride) may be present in a concentration of from about 2 mM to about

300 mM. In some specific embodiments, the salt is present in the mixture at a concentration of about mM, about 5 mM, about 10 mM, about 20 mM, about 50 mM, about 75 mM, about 100 mM, about 110 mM, about 120 mM, about 150 mM, about 175 mM, about 200 mM, or about 300 mM.

In one embodiment, the buffer contains 2-(K-morpholino)ethanesulfonic acid (MES). In a specific embodiment, MES is present in the mixture at a concentration of about 10 mM and a pH of about 6.

In some embodiments, the excitation light is white light having wavelengths spanning the visible spectrum. In some embodiments, transmitted light is measured at several wavelengths ranging from about 450 nm to about 750 nm.

The absorption intensity factor is a measure of the relative intensity of light absorption relative to a standard or control. Control can be external or internal. In one embodiment, the absorption intensity factor is calculated by dividing the maximum intensity (optical density or absorbance) wavelength of the sample's peak absorption by the observed initial absorption intensity. In one embodiment, the initial observed absorption intensity is the absorption intensity observed at 450 nm. In this embodiment, the absorption intensity factor is the peak absorbance (A _peak )/absorbance at 450 nm (A4 ₅₀ or A _out ). In this embodiment, the threshold value of the absorption intensity factor is from about 1.5 to about 2. In a particular embodiment, the threshold value of the absorption intensity factor is about 1.7.

In one embodiment, a protein that is considered suitable for dosing at a high concentration is combined with an excipient to form a drug substance (DS) or drug product (DP) formulation. In one embodiment, the protein is formulated to a final concentration of from about 50 mg/ml to about 250 mg/ml.

In one embodiment, the excipient includes a tonicity modifier, a buffer, a surfactant, a stabilizer, or any combination of two or more of these. In one embodiment, the tonicity modifier is a salt. In a specific embodiment, the salt is sodium chloride. In one embodiment, the buffer maintains a pH of about 6 to about 7. In a particular embodiment, the buffer is histidine. In another specific embodiment, the buffer is a phosphate. In one embodiment, the surfactant is a polysorbate, such as polysorbate 20 or polysorbate 80. In one embodiment, the stabilizer is a sugar, such as sucrose or trehalose. In another embodiment, the stabilizer is an amino acid such as proline or arginine.

In a third aspect, the invention relates to a method for producing a therapeutic protein. In one embodiment, the method includes the steps of selecting a protein having high colloidal stability from a variety of different proteins with varying unknown colloidal stability;

producing the selected protein in the host cell;

protein purification; and combining the protein at high concentration with an excipient to form a drug substance or drug product where the protein is stable.

In one embodiment, less than 10% of the protein aggregates in the formulated drug substance or drug product.

In one embodiment, the step of selecting a protein having high colloidal stability includes the steps of combining the protein with a nanoparticle and a buffered salt to form a sample;

exciting the sample with light;

measuring light passing through a sample; and calculating the absorption intensity coefficient of the sample.

This selection stage is repeated one or more times, changing one or more parameters. Variable parameters include salt type, salt concentration, pH, protein concentration, and inclusion or exclusion of additional components. When the absorption intensity factor exceeds a threshold value, the protein is selected as having high colloidal stability.

In some embodiments, the excitation light is white light containing wavelengths spanning the visible spectrum. In some embodiments, transmitted light is measured at several wavelengths ranging from about 450 nm to about 750 nm.

The absorption intensity factor is a measure of the relative intensity of light absorption relative to a standard or control. Control can be external or internal. In some embodiments, the absorption intensity factor is calculated by dividing the maximum intensity (optical density or absorbance) of the absorption peak wavelength of a sample by (1) the original absorption intensity observed in that sample (intrinsic) or by (2) the original absorption intensity observed in sample of nanoparticles in the absence of protein

- 4 043996 (external).

In one embodiment, the initial observed absorption intensity is the absorption intensity observed at 450 nm. In this embodiment, the absorption intensity factor is the peak absorbance (A _peak )/absorbance at 450 nm (A ₄₅₀ or A _out ). In this embodiment, the threshold value of the absorption intensity factor is from about 1.5 to about 2. In a particular embodiment, the threshold value of the absorption intensity factor is about 1.7.

In a further embodiment, the invention relates to a composition containing a biotherapeutic drug (such as, for example, a protein or antibody) at a high concentration. Specifically, the composition may be such that no more than about 10% of the total protein species is present as an irreversible aggregate at the biotherapeutic drug concentration. Alternatively, the composition is such that the threshold value, as discussed herein, is in the range of, for example, about 1.5 to about 2.0 (A _peak /A ₄₅₀ ). Another alternative is where the threshold value is, for example, about 1.5, about 1.6, about 1.7, about 1.8, about 1.9, or about 2.0 (A _peak /A ₄₅₀ ). In some embodiments, the threshold value is in the range of about 0.7 to about 1.0 (A _peak /A _control ). A further alternative is where the threshold value is, for example, about 0.7, about 0.8, about 0.9, or about 1.0 (A _peak /A _control ). In yet another further embodiment, the composition is such that no more than about 10% of the total protein species is present as an irreversible aggregate at the biotherapeutic drug concentration, and such that the composition has a threshold value as discussed herein, e.g. range, for example, from about 1.5 to about 2.0 (A _peak /A ₄₅₀ ), or where the threshold value is, for example, about 1.5, about 1.6, about 1.7, about 1.8, about 1.9 or about 2.0 (A _peak /A4 ₅₀ ), and/or where the threshold value is in the range of about 0.7 to about 1.0 (A _peak /A _control ), or where the threshold value is, for example about 0.7, about 0.8, about 0.9 or about 1.0 (A _peak /A _control ). The concentration of the biopharmaceutical may be about 50 mg/ml or more. Alternatively, the concentration of the biopharmaceutical may be in the range, for example, from about 50 mg/ml to about 500 mg/ml, such as, for example, from about 50 mg/ml to about 250 mg/ml, such as, for example, from about 100 mg/ml to about 250 mg/ml.

In one embodiment, the invention provides a composition that can be prepared by a method for determining the self-assembly potential of a protein, as described herein.

In yet another aspect, the invention relates to a composition containing a biotherapeutic drug (such as, for example, a protein or antibody) at a high concentration for use in medicine. Depending on the drug, one skilled in the art will know which clinical conditions can be treated by administering the composition to a subject in need thereof. The composition may be such that no more than about 10% of the total protein species is present as an irreversible aggregate at that biotherapeutic drug concentration. Alternatively, the composition is such that the threshold value, as discussed herein, is in the range of, for example, about 1.5 to about 2.0 (A _peak /A ₄₅₀ ). Another alternative is where the threshold value is, for example, about 1.5, about 1.6, about 1.7, about 1.8, about 1.9, or about 2.0 (A _peak /A ₄₅₀ ). In some embodiments, the threshold value is in the range of about 0.7 to about 1.0 (A _peak /A _control ). A further alternative is where the threshold value is, for example, about 0.7, about 0.8, about 0.9, or about 1.0 (A _peak /A _control ). In yet another further embodiment, the composition is one in which no more than about 10% of the total protein species is present as an irreversible aggregate at that biotherapeutic drug concentration, and thus the composition has a threshold value as discussed herein, e.g. , in the range, for example, from about 1.5 to about 2.0 (A _peak /A ₄₅₀ ), or where the threshold value is, for example, about 1.5, about 1.6, about 1.7, about 1, 8, about 1.9 or about 2.0 (A _peak /A45 ₀ ), and/or where the threshold value is in the range of about 0.7 to about 1.0 (A _peak /A _control ), or where the threshold value is, for example, about 0.7, about 0.8, about 0.9, or about 1.0 (A _peak /A _control ). The concentration of the biopharmaceutical may be about 50 mg/ml or more. Alternatively, the concentration of the biopharmaceutical may be in the range, for example, from about 50 mg/ml to about 500 mg/ml, such as, for example, from about 50 mg/ml to about 250 mg/ml, such as from about 100 mg/ml to about 250 mg/ml. The composition may be administered in a single dose or multiple doses as deemed necessary to obtain the desired treatment result. In an alternative embodiment, the invention relates to a composition containing a biotherapeutic drug (such as, for example, a protein or antibody) at a high concentration for the production of a drug for the treatment of a disease or diseases that are treatable or curable by the biotherapeutic drug.

In one embodiment, the invention relates to a method of preparing a biopharmaceutical composition (such as, for example, a protein or antibody) such that the resulting composition has a suitable rheology. In particular, a method for determining the ability of a protein to self-assemble,

- 5 043996 as described herein, allows the determination of a procedure for providing a stable composition containing a high concentration of a biopharmaceutical drug. One of the important attributes of such a composition is viscosity. Therefore, the method of the invention will provide a composition with the desired physical properties so that the composition can be administered by the desired route or mode of administration. One example route of administration may be by injection. In such a case, it is important to have a viscosity of the prepared composition that allows injection, for example, by means of a syringe and catheter. Such a catheter may be a catheter of a size such as, for example, 6G, 8G, 9G, 10G, 11G, 12G, 13G, 14G, 16G, 19G, 20G, 21G, 22G, 23G, 24G, or 26G. The term viscosity refers to dynamic or absolute viscosity (at 20°C and normal pressure), which is a measure of the resistance of a fluid to deform under either shear or tensile stress. Thus, viscosity describes a fluid's internal resistance to mobility and can be thought of as a measure of fluid friction. Therefore, the lower the viscosity of something, the greater its ease of movement (fluidity). In this context, suitable viscosity ranges are from about 10-10,000 mPa-s, such as, for example, about 20-9000 mPa-s, such as, for example, about 30-8000 mPa-s, such as about 40 -7000 mPa-s, such as, for example, about 50-6000 mPa-s, such as, for example, about 70-5000 mPa-s, such as, for example, about 90-4000 mPa-s, such as, for example, about 100-3000 mPa-s or about 10 mPa-s, or about 20 mPa-s, or about 30 mPa-s, or about 40 mPa-s, or about 50 mPa-s, or alternatively from about 1 mPa-s to about 20 mPa-s, such as, for example, about 2 mPa-s, such as, for example, about 3 mPa-s, such as, for example, about 4 mPa-s, such as, for example, about 5 mPa-s s, such as, for example, about 6 mPa-s, such as, for example, about 7 mPa-s, such as, for example, about 10 mPa-s, about 13 mPa-s, about 15 mPa-s, such as, for example, about 20 mPa-s.

Other units of viscosity are well known, for example 1 cP is equivalent to 1 mPa-s. Without wishing to be bound by any particular theory, a typical low concentration protein (ie, less than or equal to 10 mg/ml) has a viscosity of about 10 cP. As such, a highly concentrated protein composition, such as an antibody composition, exhibiting a viscosity below about 10 cP (below about 10 mPa-s) is highly suitable for use as a biopharmaceutical composition. The protein composition (at high concentration) having a viscosity of about 10 cP to about 15 cP (about 10 mPa-s to about 15 mPa-s) is safe for use in the drug manufacturing and development process. The protein composition (at high concentration) having a viscosity of about 15 cP to about 20 cP (about 15 mPa-s to about 20 mPa-s) warns of the need for caution during drug manufacturing and development. A protein composition (at high concentration) having a viscosity greater than about 20 cP (greater than about 20 mPa-s) has been reported as problematic in drug manufacturing and development. Viscosity can be adjusted by various components and ingredients as described herein. Soluble and highly soluble also refer to a protein having a suitable viscosity for any of the applications described herein, for example, for the preparation and use of a biopharmaceutical composition.

In one embodiment, the protein is an antigen binding protein, such as an antibody, antibody fragment, or receptor-Fc fusion protein.

In one embodiment, the host cell in which the protein is produced is a Chinese hamster ovary (CHO) cell or a CHO cell derivative such as a CHO-Κι cell or an EESYR® cell (see Chen et al., US 771997 B2, publ. August 10, 2010).

In one embodiment, the protein purification step includes subjecting the protein to (1) one or more steps of affinity chromatography, ion exchange chromatography, hydrophobic interaction chromatography, mixed mode chromatography, and hydroxyapatite chromatography; and (2) one or both ultrafiltration and diafiltration steps.

In one embodiment, the protein is formulated at a final concentration that is at least 50 mg/ml. In one embodiment, the protein concentration is from about 50 mg/ml to about 250 mg/ml.

In one embodiment, the excipient with which the protein is combined to form a drug substance or drug product contains one or more tonicity modifiers, a buffer, a surfactant, and a stabilizer. In one embodiment, the tonicity modifier is a salt. In a specific embodiment, the salt is sodium chloride. In one embodiment, the buffer maintains a pH of about 6 to about 7. In a particular embodiment, the buffer is histidine. In another specific embodiment, the buffer is a phosphate. In one embodiment, the surfactant is a polysorbate, such as polysorbate 20 or polysorbate 80. In one embodiment, the stabilizer is a sugar, such as sucrose or trehalose. In another embodiment, the stabilizer is an amino acid such as proline or arginine.

- 6 043996

Brief description of graphic materials

In fig. Figure 1 shows the absorption profile of dispersed gold nanoparticles of 20 nm (line A) and agglomerated gold nanoparticles of 20 nm (line B). The Y-axis displays absorbance in arbitrary absorbance units. The X-axis displays the wavelength of transmitted light in nanometers (nm).

In fig. Figure 2 shows a scatter plot of absorption intensity coefficient data for various concentrations of monoclonal antibody 2 (mAb2) and 20 nm gold nanoparticles at 2 mM salt (circles), 20 mM salt (squares), and 200 mM salt (triangles). The Y-axis displays the ratio of the peak absorbance intensity of each sample condition to the initial absorbance intensity. The X-axis displays mAb2 concentration in micrograms per milliliter.

In fig. Figure 3 shows a scatter plot of these absorption intensity coefficients of various concentrations of monoclonal antibody (mAb) 1 (open symbols) and mAb5 (closed symbols) in the presence of 20 nm gold nanoparticles at 2 mM salt (squares), 20 mM salt (triangles) and 200 mM salt ( mugs). The Y-axis displays the ratio of the peak absorbance intensity of each sample condition to the initial absorbance intensity. The x-axis represents mAb concentration in micrograms per milliliter. The boxed area on the left (A) represents conditions with higher colloidal repulsive conditions having a high selection value, the boxed area in the middle (B) represents conditions with mixed repulsive and attractive conditions having a moderate or mixed selection value, and the boxed area in the middle (B) represents conditions with mixed repulsive and attractive conditions having a moderate or mixed selection value, and The boxed area on the right (C) represents attraction conditions that have suboptimal or low selection values.

In fig. Figure 4 shows an overlay of several absorption profiles of dispersed 20 nm gold nanoparticles in combination with human serum albumin (HSA) of various concentrations (3.125, 6.25, 12.5, 25, 50, 100, 200 and 400 μg/ml in order from top to bottom at the rightmost [750 nm] part of the curve). The Y axis displays arbitrary absorbance in arbitrary units. The X-axis displays the wavelength of transmitted light in nanometers (nm).

In fig. 5 is a scatter plot of absorption intensity coefficient data for various concentrations of HSA obtained from the raw absorption spectra shown in FIG. 4, and 20 nm gold nanoparticles at 2 mM salt (circles), 20 mM salt (squares), and 200 mM salt (triangles). The Y-axis displays the ratio of the peak absorbance intensity of each sample condition to the initial absorbance intensity. The X-axis displays the HSA concentration in micrograms per milliliter.

In fig. Figure 6 shows a scatter plot of absorption intensity coefficient data for various concentrations of monoclonal antibody 1 (mAb1) and 20 nm gold nanoparticles at 2 mM salt (circles), 20 mM salt (squares), and 200 mM salt (triangles). The Y-axis displays the ratio of the peak absorbance intensity of each sample condition to the initial absorbance intensity. The x-axis displays mAb1 concentration in micrograms per milliliter.

In fig. Figure 7 shows a scatter diagram of the scattered light intensity of mAb1 solutions at various concentrations and concentrations of ionic salts: 2 mM NaCl (circles), 20 mM NaCl (squares), and 200 mM NaCl (triangles). The Y axis displays light scattering intensity in arbitrary absorbance units. The X-axis displays mAb1 concentration in grams per liter. The solid line represents an ideal solid sphere of comparable diameter.

In fig. Figure 8 shows a scatter plot of absorption intensity coefficient data for various concentrations of monoclonal antibody 6 (mAb6) and 20 nm gold nanoparticles at 2 mM salt (circles), 20 mM salt (squares), and 200 mM salt (triangles). The Y-axis displays the ratio of the peak absorbance intensity of each sample condition to the initial absorbance intensity. The X-axis displays mAb6 concentration in micrograms per milliliter.

In fig. Figure 9 shows a scatter diagram of the scattered light intensity of mAb6 solutions at various concentrations and concentrations of ionic salts: 2 mM NaCl (circles), 20 mM NaCl (squares), and 200 mM NaCl (triangles). The Y axis displays light scattering intensity in arbitrary absorbance units. The X-axis displays mAb6 concentration in grams per liter. The solid line represents an ideal solid sphere of comparable diameter.

In fig. 10 is a scatter plot of absorbance intensity coefficient data for various concentrations of monoclonal antibody 3 (mAb3) and 20 nm gold nanoparticles at 2 mM salt (circles), 20 mM salt (squares), and 200 mM salt (triangles). The Y-axis displays the ratio of the peak absorbance intensity of each sample condition to the initial absorbance intensity. The x-axis displays mAb3 concentration in micrograms per milliliter.

In fig. 11 is a scatter plot of absorption intensity coefficient data for various concentrations of monoclonal antibody 4 (mAb4) and 20 nm gold nanoparticles at 2 mM salt (circles), 20 mM salt (squares), and 200 mM salt (triangles). The Y-axis displays the ratio of the peak absorbance intensity of each sample condition to the initial absorbance intensity. The X-axis displays mAb4 concentration in micrograms per milliliter.

In fig. Figure 12 shows a scatter plot of these absorption intensity coefficients of various concentrations of monoclonal antibody 5 (mAb5) and 20 nm gold nanoparticles at 2 mM salt

- 7 043996 (circles), 20 mM salt (squares) and 200 mM salt (triangles). The Y-axis displays the ratio of the peak absorbance intensity of each sample condition to the initial absorbance intensity. The X-axis displays mAb5 concentration in micrograms per milliliter.

In fig. 13 is a scatter plot of CD-SINS data on absorption intensity ratios of various concentrations of monoclonal antibody 7 (mAb7) formulated in various buffers, pH, and ionic strengths. Each symbol represents a different buffer composition, pH, and ionic strength with different mAb7 concentrations (x-axis).

In fig. 14 is a CD-SINS scatter plot of absorption intensity ratios of various concentrations of monoclonal antibody 7 (mAb7) formulated in the presence of various sugars, concentrations of sugars and/or amino acids. Each symbol represents a different sugar/amino acid composition with different concentrations of mAb7 (x-axis).

In fig. 15 is a CD-SINS scatter plot of absorption intensity ratios of various concentrations of monoclonal antibody 7 (mAb7) (8, 32, 128 and 512 μg/ml mAb7) combined with various benzoate compounds. When mAb7 was combined with p-aminobenzoic acid [PABA(-o-), open circle], a favorable dynamic colloidal interaction profile emerged in which the absorption intensity ratio (y-axis) exceeded 1.6.

In fig. 16 is a CD-SINS scatter plot of absorption intensity ratios of various concentrations of monoclonal antibody 7 (mAb7) in the presence of various concentrations of p-aminobenzoic acid (PABA). The x-axis represents the log2 concentration of mAb7 in micrograms per milliliter. The Y axis displays the absorption intensity factor. Open circles (-o-) indicate no PABA, open squares (-□-) indicate 12 mM PABA, open triangles (-Δ-) indicate 18 mM PABA, filled circles (-·-) indicate 24 mM PABA, filled squares (--) represent 30 mM PABA, and filled triangles (-▲-) represent 36 mM PABA.

In fig. 17 shows a scatter plot of the constant shear viscosity (in mPa-s) of a mAb7 solution containing no PABA (filled circles [-·-]) or 20 mM PABA (filled triangles [-▲-]), as a function of mAb7 concentration in g/ l. Open symbols represent extrapolated viscosities of solutions expected to contain 100 g/L mAb7.

Detailed description of the invention

Before describing the present invention, it should be understood that the present invention is not limited to the specific methods and experimental conditions described, since such methods and conditions may vary. It should also be understood that the terminology used herein is intended to describe specific embodiments only and is not intended to be limiting, as the scope of the present invention will be limited only by the appended claims.

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 this invention relates. As used herein, the term about, when used in relation to a specific numerical value given, means that the value may differ from the given value by no more than 15%. For example, as used herein, the expression about 100 includes 85 and 115 and all integer and fractional values in between (eg, 86, 86.001, 87, 88, 88.3, 89, etc.).

Absolute amounts and relative amounts of fillers, components and other materials may be described by weight or moles. Units of mass can be expressed in grams, milligrams, micrograms, etc.). The term mass, such as w/v or m/v, means mass. Relative amounts may be expressed as a percentage by weight (i.e., percentage), where one (1) weight-to-volume (w/v) percent means 1 gram of material per 100 milliliters of volume. Also, for example, one (1) part of component A to one (1) part of component B by weight means, for example, that for every one (1) gram of component A there is one (1) gram of component B. For example, one percent (1 %) of the mass of component A means, for example, that for every 100 grams of the total mass of the particle there is one (1) gram of component A. Relative amounts of a component can also be expressed in terms of moles or the number of molecules per given volume, such as millimoles per liter ( millimoles (mM)) or to another component, for example X parts of component A to Y parts of component B in moles means that for every X moles of A there are Y moles of B

Although any methods and materials similar or equivalent to those described herein may be used in the practice or testing of this invention, exemplary methods and materials are described herein. All publications mentioned in this document are incorporated herein by reference in their entirety.

Dynamic colloidal stability assay: concentration-dependent nanoparticle-based self-interaction spectroscopy (CD-SINS).

A developed SINS-based assay is proposed to determine the dynamic range of virials for proteins. The protein exhibits a complex functional state that is very sensitive to the chemical and physical conditions of the protein's environment. Therefore, estimating the virial coefficient of a protein in a specific set of environmental conditions is not sufficient to determine the virial coefficient of that protein in other environments. Obtaining multiple virials for a given protein under numerous conditions is laborious and time-consuming. The present inventors describe a high-throughput assay for determining the dynamic colloidal stability of a protein that informs the selection of a protein that can be formulated or otherwise maintained at high concentration without, or at least minimizing, problems associated with aggregation potential aggregation.

This assay uses SINS in a novel way by assessing nanoparticle aggregation under varying protein concentrations, varying ionic strength, variable pH, and other conditions. As described above and in Sules (2011), gold nanoparticles are coated with protein and placed in a buffered solution. As nanoparticles aggregate due to protein self-association, the surface plasmon resonance of the particles changes: the maximum absorption shifts toward a longer wavelength (red shift); and absorption has a lower intensity, i.e. the absorption profile expands and moves to the left (Fig. 1). In this embodiment, uncoated gold nanoparticles measuring 20 nm have a peak absorbance at about 520-530 nm.

Comparison of traditional SINS with the well-known static light scattering (SLS) method for the determination of B ₂₂ or A ₂ shows general agreement for the majority of antibodies tested. In table Figure 1 compares the peak absorbance of beads coated with specific various monoclonal antibodies and human serum albumin (HSA) in the SINS assay with A ₂ virials for the same proteins determined by SLS. As shown in table. 1, SINS analysis easily distinguishes attracting (negative A ₂ ) from repulsive (positive A ₂ ) systems. However, traditional SINS and SLS analyzes do not show dynamic range for repulsive conditions. In addition, several discrepancies between conventional SINS and SLS indicate complex behavior of some proteins.

Table 1

Protein Peculiarity systems 2 mM NaCl 20 mM NaCl 200 mM NaCl SINS a ₂ SINS a ₂ SINS a ₂ HSA Repulsion 526 1.17E-4 526 9.68E-5 528 7.85E-5* mkAtI Repulsion 530 1.72E-4 531 6.70E-5 532 1.34E-5 mAb2 Mixed 583 -2.36E-5 536 -1.44E-5 534 1.47E-5 MKATZ Attraction 556 -3.39E-5 555 -4.68E-5 565 nd mAb4 Repulsion 531 -5.90E-5 533 -1.06E-5 535 1.38E-5 mAb5 Mixed 562 -1.09E-5 535 -2.14E-5 530 -7.67E-6

The inventors observed that some proteins are capable of being stable at high concentrations, but exhibit negative A2 or red-shifted maximum absorption spectra ( _Xmax ), for example, mAb5 can easily reach a concentration of about ~200 g/L and be stable, but its A ₂ is negative and its Z _max is red-shifted at lower ionic strength. Similarly, mAb2 can easily reach concentrations of ~175 g/L and remain stable, but its A ₂ is negative at lower salt concentrations and its X _max shifts red at lower ionic strength.

This document describes an improved SINS method. The method provides data on the dynamic colloidal stability of proteins to assess protein-specific state early in development, while bypassing the need for large quantities of protein. This method uses surface plasmon resonance of coated nanoparticles in the presence of varying amounts of protein in solution, exceeding the amount required to coat the nanoparticles. Absorption profiles are observed, the absorption intensity factor for each sample variant is calculated and plotted to determine dynamic colloidal stability.

In one embodiment, for each sample in a plurality of samples used to determine a plurality of individual absorbance intensity factor values for a given protein of interest, the nanoparticles are combined with varying amounts of protein greater than the minimum amount of protein required to completely coat the particles. The amount of minimal protein used depends on the molecular weight of the protein (i.e., its hydrodynamic radius), the size (surface area) of the nanoparticles, and the concentration of the nanoparticles. For example, when 20 nm gold nanoparticles are used at levels of about 6 to 6.5 x 1011 particles per ml, about 2.5 μg/ml of about 50 to 150 kDa protein is sufficient to completely coat the nanoparticles. Therefore, protein levels greater than 2.5 μg/mL are used for each sample of multiple samples under these conditions. Here, for example, the protein included in the sample is from about 2.6 μg/ml to about 512 μg/ml or more, about 3 ± 1 μg/ml, about 4 ± 1 μg/ml, about 5 ± 1 μg/ml , about 6±1 μg/ml, about 7±1 μg/ml, about 8±1 μg/ml,

- 9 043996 about 9±1 µg/ml, about 10±1 µg/ml, about 15±5 µg/ml, about 20±5 µg/ml, about 25±5 µg/ml, about 30±5 µg/ml , about 40±5 μg/ml, about 50±10 μg/ml, about 60±10 μg/ml, about 70±10 μg/ml, about 80±10 μg/ml, about 90±10 μg/ml, about 100 ± 10 μg/ml, about 125 ± 15 μg/ml, about 150 ± 25 μg/ml, 175 ± 25, 200 ± 25, 225 ± 25, 250 ± 25 300 ± 50, 350 ± 50, 400 ± 50, 450±50, 500±50 or 512±50 µg/ml. In some embodiments, the plurality of samples includes two samples, three samples, four samples, five samples, six samples, seven samples, eight samples, nine samples, ten samples, or more containing a different protein concentration.

For example, a plurality of samples in one embodiment comprises 6.3x1011 particles/ml of 20 nm gold nanoparticles and an antibody at a first concentration of about 3.125 μg/ml, a second concentration of 6.25 μg/ml, a third concentration of about 12.5 μg/ml, the fourth concentration is about 25 μg/ml, the fifth concentration is about 50 μg/ml, the sixth concentration is about 100 μg/ml, the seventh concentration is about 200 μg/ml and the eighth concentration is about 400 μg/ml.

In one embodiment, for each sample of a plurality of samples used to determine a plurality of individual absorbance intensity factor values for a given protein of interest, each nanoparticle/protein combination is combined with a variable salt concentration. For example, the sample may contain a neutral protein salt, such as sodium chloride, at a concentration of about 1 μM, about 10 μM, about 100 μM, about 1 mM, about 2 mM, about 4 mM, about 6 mM, about 8 mM, about 10 mM, about 20 mM, about 30 mM, about 40 mM, about 50 mM, about 60 mM, about 70 mM, about 80 mM, about 90 mM, about 100 mM, about 150 mM, about 175 mM, about 200 mM , about 250 mM or about 300 mM or more. In another embodiment, the salt may be a chaotropic salt such as guanidinium chloride, lithium perchlorate, lithium acetate, magnesium chloride, or the like. The sample may also contain, in addition to or instead of the chaotropic salt, another chaotropic agent such as butanol, ethanol, phenol, propanol, sodium dodecyl sulfate, thiourea or urea. In another embodiment, the sample may contain a kosmotropic salt, such as a carbonate, sulfate or phosphate salt, such as ammonium sulfate.

The plurality of samples may contain two, three, four, five, six, seven, eight, nine or ten or more subsets of samples having different salt concentrations. Thus, for example, a plurality of samples may contain ten different protein concentrations at three different salt concentrations for a total of thirty samples. In a specific embodiment, one subset of the plurality of samples contains about 2 mM sodium chloride, another subset of the plurality of samples contains about 20 mM sodium chloride, and another subset of the plurality of samples contains about 200 mM sodium chloride. A salt of 2 mM is considered low ionic strength, and 200 mM is considered high ionic strength. High salt tonicity up to and including 300 mM, although not relevant to most protein therapeutic formulations, is used in protein manufacturing during the purification process by hydrophobic interaction chromatography (HIC). Therefore, the invention is useful for determining whether a protein at high concentration under high salt conditions will be suitable for purification by HIC or the like.

White light is applied to each sample and its absorption spectrum is obtained. The absorption intensity factor is calculated for each sample from the set and plotted on a graph. In those samples in which nanoparticles do not aggregate or where aggregation is low, i.e. when the protein exhibits repulsion or low aggregation potential, the absorption peak wavelength and absorption intensity are similar to the uncoated control nanoparticles. In those samples in which nanoparticles aggregate, i.e. when a protein exhibits attraction or high aggregation potential, the absorption peak wavelength shifts toward red and the absorption profile levels off as peak intensity decreases compared to uncoated control nanoparticles. Each sample of a plurality of samples for a given protein may exhibit different absorption profiles (ie, different absorption intensity ratios), in some cases showing the ability to attract, and in other cases showing the ability to repel. These proteins have a dynamic range of colloidal stability and can be colloidally stable under certain conditions. For some proteins, the protein may exhibit repulsion across all parameters (patterns). Such a protein is considered stable or has stable dynamic colloidal stability. For other proteins, the protein may exhibit attraction under all conditions of the parameters tested. Such a protein is considered colloidally unstable.

Those proteins that exhibit repulsion in at least one test condition (at least one sample) can be selected for large-scale production and formulation as a stable drug substance. These proteins are less likely to form aggregates during production and storage. In addition, the conditions under which the protein exhibits repulsive properties can provide information regarding manufacturing steps and excipients in formulations for the protein as a drug. For example, the use of concentration-dependent SINS (CD-SINS), as described herein, can determine the feasibility of protein production, for example, in subsequent ultrafiltration and diafiltration steps. Protein selection

- 10 043996 low-viscosity solution for large-scale production can increase efficiency and reduce costs in the bioprocess for protein production (Shire S.J., 2009, Formulation and manufacturability of biologies, Curr. Opin. Biotechnol., 20(6):708-14). CD-SINS is also generally available for a wide range of excipients (both conventional and specific). CD-SINS can be used to obtain structure-activity relationships between different chemical series used to formulate a protein drug and can be used to select excipients that alter self-association (see Example 11 and Figures 13 and 14).

General criteria for selecting molecules and conditions that have acceptable colloidal stability dynamics are depicted in FIG. 3, which shows absorption intensity coefficient curves for two proteins, mAb1 and mAb5, under three ionic strength conditions. The three areas provide (a) a safe process that involves stable molecules that are expected to be stable at high concentrations;

(c) a process with possible adverse consequences that involves molecules exhibiting absorption intensity ratios above a threshold level (eg, A _peak /A4 ₅₀ >1.7) under some conditions; and (c) potentially suboptimal, which includes molecules that do not reach threshold absorption coefficients under all conditions.

For example, mAb1 (open symbols) is stable in terms of colloidal stability, falling into the safe process category at lower salt conditions (2 mM NaCl, open squares; 20 mM NaCl, open triangles) and the potentially adverse process category at higher salt conditions. high salt conditions (200 mM NaCl, open circles). Conversely, mAb5 (filled symbols) is less robust, falling into the potentially suboptimal category at lower salinity conditions and into the process domain with possible adverse effects at higher salinity conditions.

Concentration-dependent SINS (CD-SINS) demonstrates how protein/solvent systems evolve with increasing protein concentration. It captures a variety of colloidal interactions that are manifested by various proteins at high protein concentrations: from repulsive to ideal (neutral); attracting to the ideal; ideal to attractive; and insensitive. It captures many of the different aspects of colloidal interactions, such as charge-induced repulsion or attraction, undergoes electrostatic shielding, and qualitatively provides a fairly wide dynamic range. CD-SINS identifies possible hydrophobically mediated problems and provides insights into typically suboptimal molecules. The method provides an analytical tool for assessing the feasibility of developing a protein at high concentration with markedly minimal protein requirements that is suitable for automation.

Reduced viscosity of attractant or mixed colloidal protein.

Those proteins that have unfavorable dynamic colloidal interaction profiles, e.g. tend to self-associate, can usually have high viscosity at higher concentrations. Such high viscosity at high concentration may make the protein undesirable for parenteral administration. The CD-SINS assay described herein is useful for screening viscosity-lowering excipients. In one aspect, a method is provided for producing a protein composition with reduced viscosity. In one embodiment, a filler that potentially reduces viscosity.

In some embodiments, the viscosity reducing excipient reduces the viscosity of the protein solution by at least 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90 , 95, 100%, 2.5, 3, 3.5, 4, 4.5, 5 times or more than 5 times relative to the viscosity of the protein solution without a viscosity-reducing filler.

In some embodiments, the viscosity reducing filler is an amino acid or a salt of an amino acid. In some embodiments, the viscosity reducing filler is a hydrocarbon, an alkane, an alkene and an alkyne, a fatty acid, a fatty acid group, a benzene-containing structure, benzoic acid, a substituted hydrocarbon, a substituted alkane, a substituted alkene, a substituted alkyne, a substituted fatty acid, a substituted tail. fatty acid, substituted benzene-containing structure, substituted benzoic acid, sulfonic acid, aminobenzoic acid, alkylated benzoic acid, hydroxybenzoic acid or ammonium salt of benzoic acid. In one embodiment, the viscosity reducing filler is para-aminobenzoic acid (PABA). In one embodiment, PABA is included in the protein composition at a concentration of 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 or 25 mM. In one embodiment, the viscosity reducing excipient is PABA combined at a concentration of >12 mM or at a concentration of 20 mM in a protein solution.

Definitions.

As used herein, colloidal stability, which can be used interchangeably with self-association propensity or aggregation propensity, refers to the net effect of molecular forces in space, such as electrostatic, van der Waals, etc., that can lead to an attractive, neutral, or repulsive protein interaction potential.

As used herein, the term absorption intensity factor means the ratio of the maximum absorption intensity of a sample to the initial absorption intensity. The original intensity can be obtained from the sample's absorption at an arbitrary wavelength that is significantly shorter or longer than the expected absorption wavelength. In some embodiments, the expected absorption wavelength for 20 nm gold nanoparticles ranges from about 500 to 600 nm and peaks at about 530 nm. Therefore, the initial absorption intensity may be the absorption intensity of the sample at a wavelength less than 500 or greater than 600. For example, the initial absorption intensity may be the absorption intensity of the sample at 450 nm. In this case, the absorption intensity factor is the absorption intensity at the peak absorption wavelength (A _peak ) of the sample divided by the absorption intensity at 450 nm (A450 or A _out ).

In other embodiments, the initial absorption intensity may be the absorption peak intensity of the uncoated nanoparticles (control nanoparticles). In this case, nanoparticles without protein, but in the same buffered salt in which the experimentally coated nanoparticle samples were taken, are subjected to absorption analysis. The peak absorption intensity of the uncoated nanoparticle control (A _control ) serves as the initial absorption intensity. In this case, the absorption intensity coefficient is calculated as A _peak / A _control .

Absorbance is usually expressed in arbitrary units, which are abbreviated when calculating the coefficient. Absorbance is measured by passing a white light source through the sample. The sample absorbs light of a certain wavelength with a certain strength, according to the optical properties of the sample. The intensity of light that passes through the sample, i.e. transmitted light is measured at several wavelengths, and the intensity of absorbed light is calculated.

The absorption spectrum of gold nanoparticles is determined by the surface plasmon resonance of the particles. The electric field from the incident light interacts with free electrons on the surface of the gold particles and results in strong absorption in the visible region. This optical property depends on the size, shape and agglomeration state of the nanoparticles. For example, smaller particles (20 nm in diameter) absorb at a lower wavelength (~522 nm) and with a narrower and more intense peak (Αξ0.6) than larger particles (e.g., 250 nm particles absorb most strongly at 570-660 nm with an intensity of about 40% of the intensity of a 20 nm particle). The irregularly shaped particle shows a broader peak of lower intensity that is red-shifted relative to the spherical particle. Similarly, agglomerated particles show a broader peak with lower intensity, which is red-shifted relative to the discrete dispersed spherical particles (Fig. 1).

As used herein, the term threshold value refers to the absorption intensity factor above which a protein is considered suitable to achieve a stable high concentration in solution. A protein sample having an absorbance coefficient below the threshold value indicates that the protein, at least under the experimental conditions of pH, ionic strength and protein density, may not be suitable to achieve a stable high concentration in solution. This possible suitability of a protein to achieve high concentration in solution is inversely related to the attractive ability or self-association potential of the protein and positively related to the repulsive ability of the protein. In some embodiments, the threshold value is about 1.5, 1.6, 1.7, 1.8, 1.9, or 2 (A _peak /A ₄₅₀ ). In some embodiments, the threshold value is about 0.7, 0.8, 0.9, or 1 (Apik/Akiprole).

As used herein, the term dynamic range refers to the ability of a protein to self-associate under a range of conditions. Prior art self-association studies that use nanoparticle-based self-interaction spectroscopy (SINS) measure only the self-association potential of proteins under one set of conditions. A protein that is considered attractive in a single point assay may be excluded, whereas a protein that is considered repulsive in a single point assay may be accepted as suitable for high concentration formulation and production. In this case, a series of absorption intensity factors are obtained under different experimental conditions and plotted to obtain a dynamic range profile for a particular protein. Under some conditions, the protein may have an absorption intensity ratio below the threshold value, while under other conditions the protein may have an absorption intensity ratio below the threshold value. Such a protein will have an acceptable degree of repulsion under these conditions, which may influence the choice of protein production processes and excipients and drug concentrations. For example, monoclonal antibody 2 (mAb2) exhibits an acceptable degree of repulsion under high concentration/high ionic strength conditions (eg, 200 mM NaCl) but an unacceptable degree of attraction under low ionic strength conditions (Figure 2).

Experimental conditions used to determine the dynamic range of a protein include protein concentration; size, number of nanoparticle shapes; the material from which the nanoparticle is made; method of coating nanoparticles with protein; nature and number of components in the buffer

- 12 043996 or solvent; pH; and ionic strength of the solution. For example, the pH of a buffer can affect the overall charge of a protein depending on the protein's isoelectric point (pI), and the charge in turn can affect the protein's second-order virial coefficient (negative B22 is attractive and positive B22 is repulsive). Ionic strength (i.e., salt content) is known to influence protein self-attraction, with increasing ionic strength protecting proteins from charge-based repulsion and promoting attraction.

In some embodiments, the data points (ie, absorption rates) are taken at various salt concentrations or without salt at all. For example, salt, such as sodium chloride, can be included in any concentration ranging from trace amounts to supersaturation. In some embodiments, sodium chloride is included in the buffered salt at a concentration of about 0.1 to about 1 mM, 1 to 500 mM, or 2 to 300 mM. In some embodiments, data is taken from two, three, four, five, six, seven, eight, nine, ten, or more different salt concentration conditions. In one embodiment, at least three salt concentration conditions are used, for example, 2, 20 and 200 mM sodium chloride. In another embodiment, at least five salt concentration conditions are used. Non-limiting examples of the five salt concentration conditions may include 1, 2, 20 mM, 200 nM and 300 mM sodium chloride or 1, 5, 20 mM, 100 nM and 300 mM sodium chloride. One skilled in the art can easily tailor the assay using different salts and/or salt conditions depending on the nature of the experimental or therapeutic use of the protein of interest.

As used herein, the term buffered salt refers to an aqueous solution containing a buffer and a salt. The salt can be in any concentration, and the buffer can have buffering capacity over any pH range. In some embodiments, the salt and buffer may be the same, such as calcium carbonate. Salts are known to influence protein interactions and are often used to precipitate proteins to influence protein folding to stabilize pharmaceutical formulations to ensure tonicity and to regulate protein behavior during chromatography. Salts can be kosmotropic (positive hydrogen bond free energy) or chaotropic (negative hydrogen bond free energy) agents. Cosmotropes facilitate water-water interactions and are commonly used for salting out proteins. Examples of kosmotropic ions include carbonate, sulfate, phosphate, magnesium, lithium, zinc and aluminum. Chaotropes disrupt hydrogen bonds and weaken hydrophobic effects, thereby promoting protein denaturation. Examples of chaotropic salts include guanidinium chloride, lithium perchlorate, lithium acetate and magnesium chloride. Salts such as sodium chloride are towards the middle of the Hofmeister series, which means they are not effective in either salting in or salting out. Depending on the purpose of the protein self-association assay, buffered salts may be selected for their ability to salt out, salt out, or remain effectively neutral at physiological concentrations.

Buffers are included to control pH, which in turn influences the charge properties of proteins and subsequent structure and function. Depending on the purpose of the protein self-association assay, buffers may be selected for, for example, their ability to enhance protein purification, provide long-term protein stability, or provide patient comfort during administration of the therapeutic protein. Suitable buffers are well known in the art and include MES, TRIS, PIPES, MOPS, phosphate, citrate-phosphate, citrate, acetate, carbonate-bicarbonate, histidine, imidazole, and the like. In one particular embodiment, the buffer used in the protein self-association assay is MES (2-N-morpholino)ethanesulfonic acid), which has a convenient buffer range of about pH 5.5-6.7. In a specific embodiment, MES is included in the buffered salt at about 10 mM, pH 6.

The buffer is used in the buffered salt component of the analytical system and reagents. The buffer is also used to control the pH of solutions that contain protein after analysis.

Buffers are used in cell culture production media used for protein production. Buffers are also used in solutions used during protein purification and during various routine operations. Buffers are used in formulated drug substances and in the final formulation of the drug product.

Buffers are widely used in protein formulations and are well known in the art and include histidine, succinate, citrate, acetate, phosphate and the like. The buffer can be included in the composition of a medicinal substance (SLS) or a medicinal product (LP) in a concentration from 1 mM to 100 mM. In some specific embodiments, the buffer is included in an amount of about 10 mM. In some embodiments, the buffer is present at a concentration of 5±0.75 to 15±2.25 mM; from 6±0.9 to 14±2.1 mM; from 7±1.05 to 13±1.95 mM; from 8±1.2 to 12±1.8 mM; from 9±1.35 to 11±1.65 mM; 10±1.5 mM; or about 10 mM. In some specific embodiments, the SLV or LP buffer system contains histidine, phosphate and/or acetate at 10 ± 1.5 mM.

In some embodiments, the buffer is selected from a chemical capable of buffering anywhere in the pH range of about 3 to about 9 or in the pH range of about 3.7 to about 8.0.

- 13 043996

For example, the pre-lyophilized solution may have a pH of about 3.4, about 3.6, about 3.8, about 4.0, about 4.2, about 4.4, about 4.6, about 4.8, about 5 ,0, about 5.2, about 5.4, about 5.6, about 5.8, about 6.0, about 6.2, about 6.4, about 6.6, about 6.8, about 7 .0, about 7.2, about 7.4, about 7.6, about 7.8 or about 8.0.

The buffer may be a combination of individual buffers, such as for example a combination of histidine and acetate (hist-acetate buffer). In one embodiment, the buffer has a buffering range of about 3.5 to about 6, or about 3.7 to about 5.6, such as an acetate-buffered range. In one embodiment, the buffer has a buffering range of from about 5.5 to about 8.5, or from about 5.8 to about 8.0, such as a phosphate buffered range. In one embodiment, the buffer has a buffering range of from about 5.0 to about 8.0, or from about 5.5 to about 7.4, such as a histidine-buffered range.

As used herein, the term nanoparticle refers to a spherical, sub-spherical or spheroidal particle having a diameter ranging from 10 ^-9 M (0.001 microns) to 10 ^-6 M (1 micron). Nanoparticles can contain any material, including organic polymers, metals, semiconductor materials, magnetic materials, and combinations of materials. Metal nanoparticles are particularly suitable for surface plasmon resonance assays because they have free electrons on their surface that can interact with the electric field from incident light, resulting in a strong absorption spectrum. Gold nanoparticles and silver nanoparticles are examples of metal nanoparticles useful for measuring changes in plasmon resonance. The size, shape and material of nanoparticles influence the intensity and wavelength of maximum absorption. Gold nanoparticles with a diameter of 20 to 400 nm are useful. In some embodiments, the nanoparticle used in the self-association dynamic range assay is a gold nanoparticle (AuNP) with a diameter of about 20, 30, 40, 50, 60, 80, or 100 nm.

As used herein, the term light refers to electromagnetic radiation (EMR). Light can have a single wavelength, a narrow range of wavelengths, a wide range of wavelengths such as white light, or a combination of wavelengths. The light can be in the form of a laser beam or scattered light. A self-aggregation assay is a spectrophotometric assay that measures the effect a sample has on light. Thus, light is applied to the sample (i.e. incident light), the sample interacts with the light such as the generation of plasmons and absorption of certain wavelengths of light (i.e. absorbed light), and a spectrum of light is passed through the sample and onto the detector ( i.e., transmitted light). In some embodiments, the incident light is white light containing EMR having wavelengths between 400 and 800 nm. In some embodiments, transmitted light is detected and measured at wavelengths between 400 and 800 nm, from which the absorption intensity is determined.

As used herein, the term self-association refers to the nonspecific binding of a particular protein to another protein of the same species. By non-specific is meant an association with weak forces that are considered non-biological. To distinguish nonspecific from specific interactions, the association of two identical halves of an antibody to form an intact antibody is considered a specific interaction. Conversely, associations of two or more identical intact antibodies through van der Waals or hydrophobic interactions to form dimers, trimers or higher order multimers that are reversible or irreversible are nonspecific.

As used herein, the term stable or stability refers to the retention of an acceptable degree of physical structure (thermodynamic stability), chemical structure (kinetic stability), or biological function (functional stability) of a protein or other biological macromolecule over time. A protein can be stable even if it does not retain 100% of its physical structure, chemical structure, or biological function after storage for a certain period of time. Under certain circumstances, maintaining about 90%, about 95%, about 96%, about 97%, about 98%, or about 99% of protein structure or function after storage for a certain period of time can be considered stable.

Stability can be measured by determining the percentage of native protein remaining in the sample. The percentage of protein that retains its native form can be determined using size exclusion chromatography, which separates high molecular weight protein aggregates from the lower molecular weight native protein. A stable protein will retain 90% or more of its native structure over time. A stable protein contains no more than 10% of all types of protein in the form of an irreversible aggregate.

A stable protein has a low rate of aggregate formation. A stable protein is subject to an increase in the formation of high molecular weight species, i.e. aggregation that is less than 15%, less than 14%, less than 13%, less than 12%, less than 11%, less than 10%, less than 9%, less than 8%, less than 7%, less less than 6%, less than 5%, less than 4%, less than 3%, less than 2%, less than 1% or less than 0.5% when stored at a temperature of about 5°C to about 25°C. for up to 7 months - 14 043996, up to 8 months, up to 9 months, up to 10 months, up to 11 months, up to 12 months, up to 13 months, up to 14 months, up to 15 months, up to 16 months, up to 17 months, up to 18 months, up to 19 months, up to 20 months, up to 21 months, up to 22 months, up to 23 months or up to 24 months.

Other methods can be used to assess protein stability, such as, for example, differential scanning calorimetry (DSC) to determine thermal stability, controlled stirring to determine mechanical stability, and absorbance at 350 nm or about 405 nm to determine solution turbidity. In one embodiment, a protein may be considered stable if, after storage for 6 months or more at a temperature of from about 5°C to about 25°C, the change in OD405 of the composition is less than about 0.05 (e.g., 0.04, 0. 03, 0.02, 0.01 or less) from the OD405 protein at time zero.

As used herein, the term protein means any amino acid polymer having more than about 50 amino acids covalently linked through amide bonds. Proteins contain one or more amino acid polymer chains, known in the art as polypeptides. A protein may contain one or more polypeptides to form one functioning biomolecule. Polypeptides typically contain more than 50 amino acids, while peptides typically contain 50 amino acids or less. Proteins may contain one or more covalent and non-covalent modifications. Disulfide bridges (i.e. between cysteine residues to form cystine) may be present in some proteins. These covalent bonds can be within a single polypeptide chain or between two separate polypeptide chains. For example, disulfide bonds are essential for the proper structure and function of insulin, immunoglobulins, protamine and the like. For information regarding disulfide bond formation see the recent review by Oka & Bulleid, Forming disulfides in the endoplasmic reticulum, 1833 (11), Biochim. Biophys. Acta., 2425-9 (2013).

In addition to disulfide bond formation, proteins can undergo other post-translational modifications. These modifications include lipidation (e.g., myristoylation, palmitoylation, phasesoylation, geranylgeranylation, and formation of glycosylphosphatidylinositol (GPI) anchors), alkylation (e.g., methylation), acylation, amidation, glycosylation (e.g., adding glycosyl groups to arginine, asparagine, cysteine, hydroxylysine, serine, threonine, and/or tryptophan) and phosphorylation (i.e., adding a phosphate group to serine, threonine, tyrosine, and/or histidine). Regarding post-translational modification of proteins produced in eukaryotes, see the recent review by Mowen & David, Unconventional post-translational modifications in immunological signaling, 15(6), Nat. Immunol. 512-20 (2014); and Blixt & Westerlind, Arraying the post-translational glycoproteome (PTG), 18, Curr. Opin. Chem. Biol., 62-9 (2014).

Examples of proteins include therapeutic proteins, recombinant proteins used in research or therapy, decoy proteins and other receptor-Fc fusion proteins, chimeric proteins, antibodies, monoclonal antibodies, human antibodies, bispecific antibodies, antibody fragments, nanobodies, recombinant antibody chimeras, cytokines , chemokines, peptide hormones and the like. Proteins can be produced using recombinant cell production systems such as the insect baculovirus system, yeast systems (eg, Pichia sp.), mammalian systems (eg, CHO cells and CHO derivatives such as CHO-Κι cells). For a recent review regarding the discussion of therapeutic proteins and their production, see Ghaderi et al., Production platforms for biotherapeutic glycoproteins. Occurrence, impact, and challenges of non-human sialylation, 28, Biotechnol. Genet. Eng. Rev., 147-75 (2012).

As used herein, the term antigen binding protein refers to any protein that binds another molecular entity. The molecular entity may be a peptide, polypeptide, protein, epitope, hapten, antigen, or biological molecule. For example, an antigen binding protein includes a receptor molecule that binds a ligand, where the ligand is an antigen. Antigen-binding proteins include antibodies, antibody fragments (eg, Fabs), single-chain antibodies, ScFv molecules, recombinant proteins containing receptors and portions of receptors, ligand molecules, recombinant proteins containing ligands or portions of ligands, recombinant molecules containing multiple receptors or receptor fragments ( e.g. receptor-Fc fusion proteins) etc.

An antibody or immunoglobulin molecule is a type or subtype of antigen-binding protein. A canonical immunoglobulin protein (eg, IgG) contains four polypeptide chains, two heavy (H) chains and two light (L) chains, linked by disulfide bonds. Each heavy chain contains a heavy chain variable region (abbreviated herein as HCVR or VH) and a heavy chain constant region. The heavy chain constant region contains three domains, CH1, CH2 and CH3. Each light chain contains a light chain variable region (abbreviated herein as LCVR or VL) and a light chain constant region. The light chain constant region contains one domain, CL. The VH and VL regions can be further subdivided into hypervariable regions, called complementarity determining regions (CDRs), alternating with regions that are more conserved, called framework regions (FR). Each VH and VL consists of three CDRs and four FRs, arranged from the amino terminus to the carboxy terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4 (heavy chain CDRs can be abbreviated as HCDR1, HCDR2 and HCDR3; light chain CDRs chains may be abbreviated as LCDR1, LCDR2 and LCDR3. As used herein, the term monoclonal antibody simply means an antibody with the same molecular identity, usually produced by clones of the parent cell that produces a single antibody. Monoclonal antibodies differ from polyclonal antibodies in that polyclonal antibodies are a collection of antibodies made by different cells.A monoclonal antibody can have monovalent affinity, which means that both halves of the antibody are identical and bind to the same epitope, or bivalent affinity, as in a bispecific antibody, which means that one half of the antibody binds to the other epitope compared to the other half of the antibody.

A bispecific antibody includes any antibody capable of selectively binding two or more epitopes. Bispecific antibodies typically contain two different heavy chains, with each heavy chain specifically binding to different epitopes, either on two different molecules (eg, antigens) or on the same molecule (eg, the same antigen). If a bispecific antibody is capable of selectively binding two different epitopes (a first epitope and a second epitope), the affinity of the first heavy chain for the first epitope will typically be at least one to two, three or four orders of magnitude lower than the affinity of the first heavy chain for the second epitope. and vice versa. The epitopes recognized by a bispecific antibody may be on one or a different target (eg, the same or a different protein). Bispecific antibodies can be produced, for example, by combining heavy chains that recognize different epitopes of the same antigen. For example, nucleic acid sequences encoding heavy chain variable sequences that recognize different epitopes of the same antigen can be fused to nucleic acid sequences encoding different heavy chain constant regions, and such sequences can be expressed in a cell that expresses the light chain immunoglobulin. A typical bispecific antibody has two heavy chains, each having three heavy chain CDRs, followed (from N-terminus to C-terminus) by a CH1 domain, a hinge domain, a CH2 domain, and a CH3 domain, and an immunoglobulin light chain, which to some extent may not exhibit antigen-binding specificity, but which can associate with each heavy chain or, which can associate with each heavy chain and which can bind one or more epitopes bound by the antigen-binding regions of the heavy chain, or which can bind with each heavy chain and allow binding or one or both heavy chains with one or both epitopes.

The expression heavy chain or immunoglobulin heavy chain includes the sequence of the immunoglobulin heavy chain constant region from any organism, and, unless otherwise indicated, includes the heavy chain variable domain. Heavy chain variable domains include three heavy chain CDRs and four FR regions unless otherwise noted. Heavy chain fragments include CDR, CDR and FR and combinations thereof. A typical heavy chain has after the variable domain (from N-terminus to C-terminus) a CH1 domain, a hinge, a CH2 domain, and a CH3 domain. A functional heavy chain fragment includes a fragment that is capable of specifically recognizing an antigen (eg, recognizing an antigen with a KD in the micromolar, nanomolar or picomolar range), which is capable of being expressed and secreted from a cell, and which contains at least one CDR.

The expression light chain includes the immunoglobulin light chain constant region sequence from any organism and, unless otherwise specified, includes in humans the kappa and lambda light chains. Light chain variable (VL) domains typically comprise three light chain CDRs and four framework (FR) regions, unless otherwise noted. Typically, the full length of the light chain includes, from the amino terminus to the carboxyl terminus, a VL domain, which includes FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4, and a light chain constant domain. Light chains that can be used in this invention include those, for example, that selectively bind neither the first nor the second antigen and selectively bind to an antigen binding protein. Suitable light chains include those that can be identified by screening the most commonly used light chains in existing antibody libraries (wet libraries or in silico), where the light chains do not substantially affect the affinity and/or selectivity of the antigen binding domains of the antigen binding proteins. Suitable light chains include those that can bind one or both epitopes that are bound by the antigen binding regions of the antigen binding protein.

The expression variable domain includes the amino acid sequence of an immunoglobulin light or heavy chain (modified as desired), which includes the following amino acid regions in sequence from N-terminus to C-terminus (unless otherwise indicated): FR1, CDR1, FR2, CDR2, FR3, CDR3 , FR4. The variable domain includes an amino acid sequence capable of folding into a canonical domain (VH or VL) having a double beta-sheet structure, the beta-sheet being linked by a disulfide bond between a residue of the first beta-sheet and the second beta-sheet.

- 16 043996

The expression complementarity determining region or the term CDR includes the amino acid sequence encoded by the nucleic acid sequence of the immunoglobulin genes of an organism, which typically (i.e., in a wild-type animal) appears between two framework regions in the light or heavy chain variable region of an immunoglobulin molecule (e.g., an antibody or T cell receptor). The CDR may be encoded by, for example, a germline sequence or a rearranged or disordered sequence and, for example, a naïve or mature B cell or T cell. In some circumstances (eg, for CDR3), CDRs may be encoded by two or more sequences (eg, germline sequences) that are not contiguous (eg, in a disordered nucleic acid sequence) but are contiguous in a B cell nucleic acid sequence, e.g. , as a result of splicing or joining of sequences (for example, VDJ recombination to form the heavy chain CDR3).

Fc-containing proteins include antibodies, bispecific antibodies, immunoadhesins, Fc receptor fusion proteins and other binding proteins that contain at least a functional portion of the CH2 and CH3 region of an immunoglobulin. The functional part refers to the region CH2 and CH3, which can bind the Fc receptor (eg FcyR or FcRn, ie neonatal Fc receptor) and/or which can participate in complement activation. If the CH2 and CH3 region contains deletions, substitutions and/or insertions or other modifications that render it unable to bind any Fc receptor and also unable to activate complement, the CH2 and CH3 region is not functional.

Fc-containing proteins may contain modifications in immunoglobulin domains, including in cases where the modifications affect one or more effector functions of the binding protein (for example, modifications that affect FcyR binding, FcRn binding, and thus the period half-life and/or CDC activity.Such modifications include, but are not limited to, the following modifications and combinations thereof with reference to the EU numbering of the immunoglobulin constant region: 238, 239, 248, 249, 250, 252, 254, 255 , 256, 258, 265, 267, 268, 269, 270, 272, 276, 278, 280, 283, 285, 286, 289, 290, 292, 293, 294, 295, 296, 297, 298, 301, 303 , 305, 307, 308, 309, 311, 340, 342, 344, 356, 358, 359, 360, 361, 362, 373, 375, 376, 378, 380, 382, 383, 384, 386, 388, 389 , 398, 414, 416, 419, 428, 430, 433, 434, 435, 437, 438 and 439.

By way of example, and not by way of limitation, the binding protein is an Fc-containing protein and exhibits an increased serum half-life (compared to the same Fc-containing protein without said modification(s)) and has a modification at position 250 (e.g., E or Q); 250 and 428 (for example, L or F); 252 (for example, L/Y/F/W or T), 254 (for example, S or T) and 256 (for example, S/R/Q/E/D or T); or modification to 428 and/or 433 (eg, L/R/SI/P/Q or K) and/or 434 (eg, H/F or Y); or modification to 250 and/or 428; or a modification to 307 or 308 (eg, 308F, V308F) and 434. In another example, the modification may comprise a modification to 428L (eg, M428L) and 434S (eg, N434S); modification 428L, 2591 (for example, V259I) and 308F (for example, V308F); modification 433K (for example, Н433К) and 434 (for example, 434Y); modification 252, 254 and 256 (for example, 252Y, 254T and 256E); modification 250Q and 428L (for example, T250Q and M428L); modification 307 and/or 308 (for example, 308F or 308P).

Some recombinant Fc-containing proteins contain receptors or receptor fragments, ligands or ligand fragments that have related binding partners in biological systems. Receptor-Fc fusion proteins refer to recombinant molecules that contain a soluble receptor fused to the Fc domain of an immunoglobulin. Some receptor Fc fusion proteins may contain the ligand binding domains of multiple different receptors. These receptor-Fc fusion proteins are known as decoys or decoy molecules. Rilonacept and aflibercept are examples of market decoys that antagonize IL1R (see US Pat. No. 7,927,583) and VEGF (see US Pat. No. 7,087,411), respectively. Other recombinant Fc-containing proteins include those recombinant proteins that contain a peptide fused to an Fc domain, such as Centocor's MIMETIBODYTM technology. Recombinant Fc-containing proteins are described in S. Huang, Receptor-Fc fusion therapeutics, traps and MIMETIBODY technology, 20 (6), Curr. Opin. Biotechnol., 692-9 (2009).

As used herein, the term tonicity modifier or tonicity adjuster is a substance or combination of substances that provides tonicity or osmolality of a formulation or solution. The formulation may require a physiological osmolarity, which is approximately 0.29 osmol of solute per kilogram of solvent (290 mOsm). A composition having a physiological osmolality is generally referred to as physiologically isotonic. The formulation may have an osmolality below 290 mOsm (physiologically hypotonic) or above 290 mOsm (physiologically hypertonic). A tonicity modifier is added to the formulation to bring the formulation to the appropriate tonicity. The term tonicity can be used interchangeably with osmolality or osmolarity.

Tonicity modifiers include salts, which are added to adjust ionic strength or conductivity, and non-salt tonicity modifiers. Commonly used salts include sodium chloride, potassium chloride, magnesium chloride and calcium chloride. Non-salt tonicity modifiers include sugars, sugar alcohols, monosaccharides and disaccharides, examples of which include sorbitol, mannitol, sucrose, trehalose, glycerol, maltose and lactose.

As used herein, the term surfactant refers to an additive or filler that reduces the surface tension at an interface. Some surfactants have a lipophilic part and a hydrophilic part. Surfactants are believed to provide additional stability by reducing hydrophobic protein-protein interactions and resulting in the formation of high molecular weight particles (ie aggregates). One or more surfactants may be included in protein-containing solutions, including SLV, LP, protein processing and production solutions, and protein self-association assay solutions. Surfactants can be ionic or non-ionic. Non-ionic surfactants include, for example, alkyl poly(ethylene oxide), alkyl polyglucosides (eg octyl glucoside and decyl maltoside), fatty alcohols such as cetyl alcohol and oleyl alcohol, cocamide MEA, cocamide DEA and cocamide TEA. Specific nonionic surfactants include, for example, polyoxyethylene sorbitan esters (eg, polysorbates) such as polysorbate 20, polysorbate 28, polysorbate 40, polysorbate 60, polysorbate 65, polysorbate 80, polysorbate 81, and polysorbate 85; poloxamers such as poloxamer 188, poloxamer 407; polyethylene polypropylene glycol; or polyethylene glycol (PEG). Polysorbate 20 is also known as TWEEN 20, sorbitan monolaurate and polyoxyethylene sorbitan monolaurate. Polysorbate 80 is also known as TWEEN 80, sorbitan monooleate and polyoxyethylene sorbitan monooleate.

The amount of surfactant contained in the protein-containing solution may vary depending on the specific properties and purposes required for the solution. In some embodiments, the solution may contain from about 0.001% (w/v) to about 0.5% (w/v) surfactant (eg, polysorbate 20 or polysorbate 80). For example, the solution may contain about 0.001%; about 0.0015%; about 0.002%; about 0.0025%; about 0.003%; about 0.0035%; about 0.004%; about 0.0045%; about 0.005%; about 0.0055%; about 0.006%; about 0.0065%; about 0.007%; about 0.0075%; about 0.008%; about 0.0085%; about 0.009%; about 0.0095%; about 0.01%; about 0.015%; about 0.016%; about 0.017%; about 0.018%; about 0.019%; about 0.02%; about 0.021%; about 0.022%; about 0.023%; about 0.024%; about 0.025%; about 0.026%; about 0.027%; about 0.028%; about 0.029%; about 0.03%; about 0.031%; about 0.032%; about 0.033%; about 0.034%; about 0.035%; about 0.036%; about 0.037%; about 0.038%; about 0.039%; about 0.04%; about 0.041%; about 0.042%; about 0.043%; about 0.044%; about 0.045%; about 0.046%; about 0.047%; about 0.048%; about 0.049%; about 0.05%; about 0.051%; about 0.052%; about 0.053%; about 0.054%; about 0.055%; about 0.056%; about 0.057%; about 0.058%; about 0.059%; about 0.06%; about 0.061%; about 0.062%; about 0.063%; about 0.064%; about 0.065%; about 0.066%; about 0.067%; about 0.068%; about 0.069%; about 0.07%; about 0.071%; about 0.072%; about 0.073%; about 0.074%; about 0.075%; about 0.076%; about 0.077%; about 0.078%; about 0.079%; about 0.08%; about 0.081%; about 0.082%; about 0.083%; about 0.084%; about 0.085%; about 0.086%; about 0.087%; about 0.088%; about 0.089%; about 0.09%; about 0.091%; about 0.092%; about 0.093%; about 0.094%; about 0.095%; about 0.096%; about 0.097%; about 0.098%; about 0.099%; about 0.10%; about 0.15%; about 0.20%; about 0.25%; about 0.30%; about 0.35%; about 0.4 0%; about 0.45%; or about 0.50% surfactant (eg, polysorbate 20 or polysorbate 80).

As used herein, the term stabilizer refers to a molecule or compound or combination of chemical elements (ie, more than one chemical element) that serves to stabilize the native conformation of a protein. Stabilizers stabilize proteins in solution through one or more of the following mechanisms:

(1) an increase in the surface tension of water;

(2) by eliminating the filler protein, which forms a layer of water around the protein;

(3) negative peptide bond interaction; and (4) repulsive interaction with the protein surface.

Regardless of the specific mechanism, stabilizers are preferentially excluded from the protein surface, thereby enriching the water on the protein surface. In addition, unfavorable excipient protein interactions make protein unwinding thermodynamically unfavorable as the surface area of the protein increases during denaturation. Stabilizers include, for example, polyols, sugars, amino acids, salting out salts, or any combination thereof. Examples of useful stabilizers include polyethylene glycol, sorbitol, glycerin, mannitol, trehalose, sucrose, arginine, alanine, proline, glycine, sodium chloride, or any combination thereof. Sucrose and trehalose are the most commonly used sugars.

The present invention is further described by the following non-limiting clauses.

Clause 1. Method for determining the ability of a protein to self-associate, including

a) combining protein, nanoparticles and buffered salt to form a sample;

b) excitation of the sample with light;

c) measuring the light transmitted through the sample;

- 18 043996

d) calculating the absorption intensity factor of the sample, characterized in that the protein is stable at high concentrations when the absorption intensity factor exceeds a threshold value.

Clause 2. The method according to claim 1, characterized in that the protein is an antigen-binding protein.

Claim 3. The method of claim 2, wherein the antigen binding protein is an antibody, an antibody fragment, or a receptor-Fc fusion protein.

Clause 4. The method according to claim 3, characterized in that the antigen-binding protein is a human monoclonal antibody.

Claim 5. A method according to any of the preceding claims, characterized in that the protein is present in the sample at a concentration of from about 2 μg/ml to about 512 μg/ml.

Clause 6. The method according to any of the preceding claims, characterized in that the nanoparticle is a gold nanoparticle.

Clause 7. The method according to claim 6, characterized in that the gold nanoparticle has a diameter of from about 20 nm to about 100 nm.

Item 8. The method according to item 7, characterized in that the diameter of the gold nanoparticle is about 20 nm.

Claim 9. A method according to any of the preceding claims, wherein the sample contains from about 5x1011 nanoparticles per ml to about 8x1011 nanoparticles per ml.

Claim 10. Method according to any of the preceding claims, characterized in that the sample contains about 6-6.5 x 1011 nanoparticles per ml.

Claim 11. A method according to any of the preceding claims, wherein the salt is present in the sample at a concentration of from about 2 mM to about 250 mM.

Clause 12. Method according to any of the preceding claims, characterized in that the salt is sodium chloride.

Claim 13. Method according to any of the previous claims, characterized in that the salt concentration is about 2 mM, about 20 mM or about 200 mM.

Claim 14. A method according to any of the previous claims, characterized in that the transmitted light is measured at several wavelengths in the range from about 450 nm to about 750 nm.

Clause 15. The method according to any of the preceding clauses, characterized in that the absorption intensity coefficient is the ratio of the maximum absorption to the initial absorption.

Clause 16. The method according to any of the previous clauses, characterized in that the threshold value of the absorption intensity coefficient is about 1.7.

Clause 17. The method of any of the preceding claims, further comprising repeating steps a) to d) using a different protein concentration in the sample.

Clause 18. The method of any of the preceding clauses, further comprising repeating steps a) to d) using a different salt concentration in the sample.

Clause 19. The method of any of the preceding clauses, further comprising repeating steps a) to d) using a different pH value of the sample.

Clause 20. The method according to any of the preceding clauses, further comprising

f) combining a highly soluble protein at a high concentration with an excipient to form a drug substance.

Claim 21. A method according to any of the preceding claims, characterized in that the protein concentration is from about 50 mg/ml to about 500 mg/ml.

Clause 22. The method according to any of clauses 20 or 21, characterized in that the excipient is selected from the group consisting of a tonicity modifier, a buffer, a surfactant, a stabilizer, and a combination thereof.

Clause 23. The method according to claim 22, characterized in that the tonicity modifier is a salt.

Clause 24. Method according to any of the preceding claims, characterized in that the salt is NaCl.

Clause 25. A composition containing a biotherapeutic drug obtained by the method according to any of the previous clauses.

Clause 26. The composition according to claim 25, characterized in that no more than about 10% of the total number of types of biotherapeutic drugs is present in the form of an irreversible aggregate in the concentration of the biotherapeutic drug.

Item 27. The composition according to any one of claims 25, 26, characterized in that the composition is such that the threshold value is in the range, for example, from about 1.5 to about 2.0 (A _peak /A ₄₅₀ ).

Item 28. The composition according to any one of claims 25-27, characterized in that the threshold value is in the range from about 0.7 to about 1.0 (A _peak /A _control ).

Clause 29. The composition according to any one of claims 25-28, characterized in that the concentration of the biopharmaceutical drug is in the range, for example, from about 50 mg/ml to about

- 19 043996

500 mg/ml, about 50 mg/ml to about 250 mg/ml, or about 100 mg/ml to about 250 mg/ml.

Clause 30. Composition that can be obtained according to any of claims 25-29 for use in medicine.

Clause 31. A method according to any one of claims 1 to 24 for preparing a composition comprising a biopharmaceutical drug, characterized in that the composition has a viscosity of about 10-10000 mPa, about 20-9000 mPa, about 30-8000 mPa, about 40-7000 mPa, about 50-6000 MPa, about 70-5000 mPa, about 90-4000 mPa, about 100-3000 mPa, or about 10 mPa, or about 20 mPa, or about 30 mPa, or about 40 mPa, or about 50 mPa, or alternatively from about 1 mPa to about 20 mPa, about 2 mPa, about 3 mPa, about 4 mPa, about 5 mPa, about 6 mPa, about 7 mPa, about 10 mPa, 13 mPa, 15 mPa or about 20 mPa.

Clause 32. The method of claim 31, wherein the biopharmaceutical drug is one or more proteins or one or more antibodies or any mixtures thereof.

Clause 33. The method according to any one of claims 31, 32, characterized in that the antibody is a monoclonal antibody or a polyclonal antibody or a combination thereof.

Item 34. Bioanalytical mixture containing

a) at least two nanoparticles;

b) protein in at least two phases; And

c) salt or buffer.

Clause 35. The bioanalytical mixture according to claim 34, characterized in that the first of at least two protein phases is a soluble phase.

Item 36. The bioanalytical mixture according to item 34 or 35, characterized in that the second of at least two protein phases is an adhesive phase, where the protein adheres to the surface of each of the at least two nanoparticles.

Clause 37. Bioanalytical mixture according to any of the previous claims 34-36, characterized in that the third of at least two protein phases is an aggregated phase, where the protein self-associates to form an aggregate.

Clause 38. Bioanalytical mixture according to any of the previous claims 34-37, characterized in that one or more aggregated proteins are also attached to the surface of the nanoparticle.

Clause 39. Bioanalytical mixture according to any of the previous claims 34-38, characterized in that each of at least two nanoparticles contains gold.

Clause 40. Bioanalytical mixture according to any of the previous claims 34-39, characterized in that each of the at least two nanoparticles has a diameter of from about 20 nm to about 100 nm.

Clause 41. Bioanalytical mixture according to any of the previous claims 34-40, characterized in that the diameter is about 20 nm.

Clause 42. Bioanalytical mixture according to any of the previous claims 34-41, characterized in that each of the at least two nanoparticles is saturated with protein.

Clause 43. Bioanalytical mixture according to any of the previous claims 34-42, characterized in that the nanoparticles are present at a density of from about 6x1011 nanoparticles per ml to about 7x1011 nanoparticles per ml of the mixture.

Clause 44. Bioanalytical mixture according to any of the previous claims 34-43, characterized in that the protein is present at a concentration of from about 2 μg/ml to about 512 μg/ml.

Clause 45. Bioanalytical mixture according to any of the previous claims 34-44, characterized in that the protein is an antigen-binding protein.

Clause 46. The bioanalytical mixture according to any one of the preceding claims 34-45, characterized in that the antigen binding protein is selected from the group consisting of an antibody, an antibody fragment, an aptamer and a receptor-Fc fusion protein.

Clause 47. Bioanalytical mixture according to any of the previous claims 34-46, characterized in that the antigen-binding protein is an antibody.

Clause 48. Bioanalytical mixture according to any of the previous claims 34-47, characterized in that the antibody is a human monoclonal antibody.

Clause 49. Bioanalytical mixture according to any of the previous claims 34-48, characterized in that the salt is present in a concentration of from about 2 mM to about 300 mM.

Clause 50. Bioanalytical mixture according to any of the previous claims 34-49, characterized in that the salt is present at a concentration of about 2 mM, about 20 mM, or about 200 mM.

Clause 51. Bioanalytical mixture according to any of the previous claims 34-50, characterized in that the salt includes NaCl.

Examples

The following examples are presented to further illustrate the methods of this invention. These examples are illustrative only and are not intended to limit the scope of the invention in any way.

- 20 043996

Example 1. Materials.

Monoclonal antibodies (mAbs) were produced by EESYR® cells and purified by protein A chromatography, with final purification by anion exchange or hydrophobic interaction chromatography. Buffer components were obtained from Sigma-Aldrich, VWR, or JT-Baker and were of the highest purity available. Illustra NAP columns (cat. no. 17-0853-02) and XK26/100 Superdex 200 pg column (cat. no. 90-1002-73) were purchased from GE Healthcare Life Sciences. Amicon centrifugal filters (cat. no. UFC905024) were purchased from EMD Millipore. Slide-A-LyzerTM G2 Dialysis Cassettes (cat. no. 87732) were purchased from ThermoFisher Scientific. 20 nm gold nanoparticles (cat. no. HD.GC20) were purchased from BBI solutions. Thermo ScientificTM NuncTM MicrowellTM 96-well microplates (cat. no. 12-565-66) were used as the reaction container to obtain absorption spectra.

Example 2: Highly concentrated static light scattering (HC-SLS).

Concentrated (100 g/L) monoclonal antibodies (mkAtI, mAb2, mAb3, mAb4, mAb5 and mAb6) were each purified on Akta avant (GE Healthcare Life Sciences) through an XK26/100 Superdex 200 pg column at 10 mM. MES pH 6.0, 50 mM sodium chloride. The monomeric mAb fraction was collected and concentrated using an Easy-Load MAsterFlex L/S pump (Cole Parmer) in tandem with a VivaFlow 200 30000 MWCO HY membrane. 150 ml of each concentrated antibody was divided into 3 fractions and loaded into a 10,000 MWCO dialysis cassette and replaced with 2 L of 10 mM MES pH 6.0, 250 mM sodium chloride; 10 mM MES pH 6.0, 50 mM sodium chloride; and 10 mM MES pH 6.0, 10 mM sodium chloride. The solutions were concentrated using a 50,000 MWCO centrifugal filter unit to a final volume of approximately 15 mL in their respective buffers. Concentrations were measured using SoloVPE (C Technologies, Inc.). Antibodies could not be concentrated beyond a nominal 60 g/L in 10 mM MES pH 6.0, 10 mM sodium chloride. Two other salt concentrations (250 and 50 mM sodium chloride) were adjusted to nominally 80 g/L. An aliquot from the three conditions was taken and diluted to 10 g/L. The 80.10 g/L sample and buffer were attached to a CG-MALS device (Wyatt Technology), and the light scattering signal was measured as a function of mAb concentration.

Example 3: Standard Nanoparticle Self-Interaction Spectroscopy (SINS).

A subaliquot of each antibody (mAb1, mAb2, mAb3, mAb4, mAb5 and mAb6) under different salt conditions was added to a separate 15 ml Falcon tube. To this solution, 5 ml of a solution with an optical density (1 od) with 20 nm gold nanoparticles was added to achieve a final protein concentration of 50 μg/ml. Absorbance and λ _max spectra were recorded on a SPECTRAmax 340PC (Molecular Devices) after waiting 30 min.

Concentration-dependent self-interaction spectroscopy based on nanoparticles (CD-SINS). 100 mM Buffer solutions were prepared at the appropriate pH. Illustra NAP columns were conditioned with 2.4 mL of 100 mM buffer solution (e.g., MES or sodium phosphate depending on target pH). Concentrated (50-75 g/L) mAb stock solutions were subjected to buffer exchange using a conditioned demineralized column. The resulting concentration was measured using SoloVPE. Each antibody was subsequently diluted to 5.12 mg/ml in 100 mM buffer. The mAb was then added to column 1 of the microplate and serially diluted in 100 mM buffer to 0.04 mg/mL. 80 Gold nanoparticles were added to columns 2, 3 and 4 of a 96-well microplate. 40 Serially diluted mAbs were added to each column using a multichannel pipette containing gold nanoparticles maintaining the serial dilution from column 1. Using a multichannel pipette, 280 ml of sodium chloride stock solution 357, 71 and 14 were then added to columns 2, 3 and 4 respectively mm. The resulting solution was left to equilibrate for 30 min before absorption spectra were recorded on a SPECTRAmax 340PC. The maximum absorption intensity was normalized to the absorbance at 450 nm. This ratio (coefficient) was graphically expressed as a function of the final antibody concentration.

Example 4. Dynamic colloidal stability of human serum albumin (HSA).

20 nm gold nanoparticles at a concentration of about 6.3 x 1011 particles per milliliter were combined with various concentrations of human serum albumin (HSA) in 10 mM MES buffer at 2 mM NaCl, 20 mM NaCl, or 200 mM NaCl, pH 6. Individual samples containing 3.125 , 6.25, 12.5, 25, 50, 100, 200 and 400 μg/ml, were subjected to absorption spectroscopy. These spectrograms were expressed graphically (Fig. 4 depicts spectra for 200 mM NaCl samples), and the ratio of the maximum absorbance intensity (λ _max ) to the initial absorbance at 450 nm was calculated for each protein concentration and plotted (Fig. 5).

HSA exhibits an absorption intensity factor greater than 1.7 (A _peak /A4 ₅₀ ) for all ionic strengths and at higher protein concentrations, indicating a favorable dynamic colloidal interaction profile. Static light scattering experiments show positive virials ( _A2 ) in full agreement with the CD-SINS profile (Table 2).

- 21 043996

table 2

NaCl mM SLS (A ₂ ) 2 1.17E-4 20 9.68E-5 120 7.85E-5

Example 5. Monoclonal Antibody No. 1 (mAb1).

20 nm gold nanoparticles at a concentration of about 6.3 x 1011 particles per milliliter were combined with various concentrations of human mAb1 in 10 mM MES buffer at 2 mM NaCl, 20 mM NaCl, or 200 mM NaCl, pH 6. Individual samples containing 3.125, 6.25 , 12.5, 25, 50, 100, 200 and 400 μg/ml, were subjected to absorption spectroscopy. These spectrograms were expressed graphically and the ratio of the absorbance intensity of the maximum absorbance (X _max ) to the initial absorbance at 450 nm was calculated for each protein concentration and plotted (Fig. 6).

mAb1 exhibits an absorption intensity factor greater than 1.7 (A _peak /A ₄₅₀ ) for all ionic strengths and at higher protein concentrations, indicating a favorable dynamic colloidal interaction profile. Experiments on highly concentrated static light scattering show positive virials (A ₂ ) in full agreement with the CD-SINS profile (Table 3, Fig. 7).

Table 3

NaCl mM SLS (A ₂ ) 2 1.72E-4 20 6.70E-5 200 1.34E-5

Example 6. Monoclonal antibody No. 6 (mAb6).

20 nm gold nanoparticles at a concentration of about 6.3 x 1011 particles per milliliter were combined with various concentrations of human mAb6 in 10 mM MES buffer at 2 mM NaCl, 20 mM NaCl, or 200 mM NaCl, pH 6. Individual samples containing 3.125, 6.25 , 12.5, 25, 50, 100, 200 and 400 μg/ml, were subjected to absorption spectroscopy. These spectrograms were expressed graphically and the ratio of the absorbance intensity of the maximum absorbance (X _max ) to the initial absorbance at 450 nm was calculated for each protein concentration and plotted (Fig. 8).

mAb6 exhibits an absorption intensity factor greater than 1.7 (A _peak /A4 ₅ o) at lower ionic strengths and at higher protein concentrations, indicating a favorable dynamic colloidal interaction profile under these conditions. mAb6 Shows an increase in attractive interactions with increasing ionic strength. Highly concentrated static light scattering experiments show different virials (A2) in full agreement with the CD-SINS profile (Fig. 9).

Example 7. Monoclonal antibody No. 2 (mAb2).

20 nm gold nanoparticles at a concentration of about 6.3 x 1011 particles per milliliter were combined with various concentrations of human mAb2 in 10 mM MES buffer at 2 mM NaCl, 20 mM NaCl, or 200 mM NaCl, pH 6. Individual samples containing 3.125, 6.25 , 12.5, 25, 50, 100, 200 and 400 μg/ml, were subjected to absorption spectroscopy. These spectrograms were expressed graphically and the ratio of the absorbance intensity of the maximum absorbance (X _max ) to the initial absorbance at 450 nm was calculated for each protein concentration and plotted (Fig. 2).

mAb2 Shows absolute changes in overall protein interactions under different conditions. At low ionic strength, mAb2 exhibits an increase in attractive interactions. The molecule becomes generally repulsive as the ionic strength increases. At 200 mM NaCl, the antibody has favorable colloidal stability. High-concentration static light scattering experiments show variable virials (A ₂ ) in full agreement with the usual SINS wavelengths of maximum absorption (Table 1).

Example 8. Monoclonal antibody No. 3 (mAb3).

20 nm gold nanoparticles at a concentration of about 6.3 x 1011 particles per milliliter were combined with various concentrations of human mAb3 in 10 mM MES buffer at 2 mM NaCl, 20 mM NaCl, or 200 mM NaCl, pH 6. Individual samples containing 3.125, 6.25 , 12.5, 25, 50, 100, 200 and 400 μg/ml, were subjected to absorption spectroscopy. These spectrograms were expressed graphically and the ratio of the absorbance intensity of the maximum absorbance (X _max ) to the initial absorbance at 450 nm was calculated for each protein concentration and plotted (Fig. 10).

mAb3 is an example of a colloidally unstable protein that results in suboptimal composition and production at high concentrations. At all ionic strengths and protein concentrations, mAb3 exhibits attractive interactions. Experiments on highly concentrated steel

Claims (23)

tic light scattering shows negative virials (A2) in full agreement with the usual SINS wavelengths of maximum absorption (Table 1). Example 9. Monoclonal antibody No. 4 (mAb4). 20 nm gold nanoparticles at a concentration of about 6.3 x 1011 particles per milliliter were combined with various concentrations of human mAb4 in 10 mM MES buffer at 2 mM NaCl, 20 mM NaCl, or 200 mM NaCl, pH 6. Individual samples containing 3.125, 6.25 , 12.5, 25, 50, 100, 200 and 400 μg/ml, were subjected to absorption spectroscopy. These spectrograms were expressed graphically and the ratio of the absorbance intensity of the maximum absorbance (λ _max ) to the initial absorbance at 450 nm was calculated for each protein concentration and plotted (Fig. 11). mAb4 is an example of a colloidally stable protein with lower ionic strength, but exhibits less solubility at higher protein concentrations at low ionic strength. Static light scattering experiments, which image mixed virials (negative at lower ionic strength, positive at high ionic strength), show discrepancies with data generated by CD-SINS, which demonstrate a repulsive profile for mAb4 (Table 1). Example 10. Monoclonal antibody No. 5 (mAb6). 20 nm gold nanoparticles at a concentration of about 6.3 x 1011 particles per milliliter were combined with various concentrations of human mAb5 in 10 mM MES buffer at 2 mM NaCl, 20 mM NaCl, or 200 mM NaCl, pH 6. Individual samples containing 3.125, 6.25 , 12.5, 25, 50, 100, 200 and 400 μg/ml, were subjected to absorption spectroscopy. These spectrograms were expressed graphically and the ratio of the absorbance intensity of the maximum absorbance (λ _max ) to the initial absorbance at 450 nm was calculated for each protein concentration and plotted (Fig. 12). mAb5 is an example of a colloidal mixed protein that is attractive at low ionic strength and exhibits improved functionality at high protein concentrations with increasing ionic strength. Static light scattering experiments, which image only negative virials, show inconsistencies with the data generated by CD-SINS, which shows a mixed profile for mAb4 (Table 1). Example 11: Screening and selection of excipients. A highly viscosity formulated antibody that has an unfavorable dynamic colloidal interaction profile under some formulation conditions (mAb7) was used to screen additional excipients for their ability to improve the CG-SINS profile of the antibody. The CG-SINS profile of monoclonal antibody 7 (mAb7), prepared in various solutions containing different buffers and ionic strengths (Figure 13), and sugar and/or amino acid excipients (Figure 14), remained below the absorption _max /absorbance ₀ threshold , suitable for dosing at high concentrations (eg below 1.7). The viscosity of mAb7 prepared at a concentration of ~100 mg/mL was observed at ~100 cP. The CD-SINS assay was used to screen for the colloidal stabilizing effect of various substituted benzoate compounds on mAb7. Substituted benzoic acids include m-hydroxybenzoic acid, p-hydroxybenzoic acid, m-methylbenzoic acid, p-methylbenzoic acid, m-ethylbenzoic acid, p-ethylbenzoic acid, m-aminobenzoic acid, p-aminobenzoic acid, sulfonic acid and ammonium benzoate. In fig. 15 is a scatterplot of the CD-SINS absorbance ratio for mAb7 prepared with each of the substituted benzoic acids. When p-aminobenzoic acid (PABA, filled circles in Fig. 15) was combined with formulated mAb7, mAb7 exhibited favorable colloidal stability as evidenced by absorption ratios compared to a threshold value favorable for dispersion at high concentration, i.e. >1.6 or 1.7 as shown in FIG. 15. PABA was combined in a dose-dependent manner with mAb7 and subjected to CD-SINS assay. Incorporation of >12 mM PABA (eg, 18, 24, 30, and 36 mM PABA) with mAb7 produced a favorable (i.e., repulsive) colloidal interaction profile (Figure 16). The effect of PABA on the colloidal interaction profile of mAb7 appears to be in a saturation direction at approximately 20 mM PABA. 20 mM PABA was combined with 5, 10, 30, 50, 70 and 80 g/L mAb7 and subjected to viscometric analysis using the cone and plate wetting method using a rotational rheometer (Pathak et al., Biophys. J., 104(4) :913-923, 2013 Feb 19). The results are presented in Fig. 17. Extrapolation of these results to 100 g/L mAb7 using the Krieger-Doherty algorithm (see, for example, Lukham & Ukeje, J. Colloid Interface Sci., 220(2):347-356, 1999 Dec 15) showed approximately threefold decrease in viscosity under constant shear in those mAb7 formulations that contain 20 mM PABA. CLAIM 1. A method for determining the potential of a protein for self-association at its high concentration, including a) combining low concentration protein, nanoparticles and a buffered salt to form a sample where the nanoparticles are gold nanoparticles and where the low protein concentration is co- - 23 043996 ranges from approximately 2 μg/ml to approximately 10 mg/ml; b) excitation of the sample with light; c) measuring the light transmitted through the sample at many wavelengths in the range from 450 nm to approximately 750 nm; d) calculating the absorption intensity factor of the sample, where the absorption intensity factor is the ratio of the absorption intensity at the absorption maximum (λ _mαx ) to the initial absorbance at 450 nm; And e) determining that the protein retains 90% or more of its native structure at a high concentration where the absorption intensity factor is greater than 1.7, where the high protein concentration is from about 50 mg/ml to about 500 mg/ml. 2. The method according to claim 1, characterized in that the protein is an antigen-binding protein. 3. The method of claim 2, wherein the antigen binding protein is an antibody, an antibody fragment, or a receptor-Fc fusion protein. 4. The method according to claim 3, characterized in that the antigen-binding protein is a human monoclonal antibody. 5. The method of claim 1, wherein the low protein concentration is from about 2 μg/ml to about 512 μg/ml. 6. The method of claim 1, wherein the nanoparticle has a diameter of from about 20 nm to about 100 nm. 7. The method according to claim 6, characterized in that the diameter of the nanoparticle is approximately 20 nm. 8. The method according to claim 1, characterized in that the sample contains approximately 6-6.5x1011 nanoparticles per ml. 9. The method of claim 1, wherein the salt is present in the sample at a concentration of from about 2 mM to about 250 mM. 10. Method according to claim 9, characterized in that the salt is sodium chloride. 11. The method of claim 10, wherein the salt concentration is about 2 mM, about 20 mM, or about 200 mM. 12. The method according to any one of claims 1 to 11, further comprising repeating steps a) to e) using a different protein concentration in the sample. 13. The method according to any one of claims 1 to 12, further comprising repeating steps a) to e) using a different salt concentration in the sample. 14. The method according to any one of claims 1 to 13, further comprising repeating steps a) to e) using a different pH in the sample. 15. The method according to any one of claims 1-14, further comprising f) combining a highly soluble protein in high concentration with an excipient to form a formulated drug substance or drug product. 16. The method according to claim 15, characterized in that the excipient is selected from the group consisting of a tonicity modifier, a buffer, a surfactant, a stabilizer, and a combination thereof. 17. The method according to claim 16, characterized in that the tonicity modifier is a salt. 18. Method according to claim 17, characterized in that the salt is NaCl. 19. The method of claim 1, wherein the high protein concentration is from about 50 mg/ml to about 250 mg/ml. 20. The method according to claim 1, characterized in that (i) the protein is in the sample at a low concentration of from about 2 μg/ml to about 512 μg/ml; and (ii) the protein is in high concentration when it is present in the composition at a concentration of from about 50 mg/ml to about 500 mg/ml. 21. The method according to claim 1, characterized in that (i) the protein is an antigen binding protein, a receptor-Fc fusion protein, an antibody or an antibody fragment; (ii) the protein is present in the sample at a low concentration of from about 2 μg/ml to about 512 μg/ml; (iii) the nanoparticle is a gold nanoparticle with a diameter of from about 20 nm to about 100 nm; (iv) the salt is present in the sample at a concentration of from about 2 mM to about 250 mM; and (v) the protein is in high concentration when it is present in the composition at a concentration of from about 50 mg/ml to about 500 mg/ml. 22. The method according to claim 1, wherein the low protein concentration is less than or equal to 10 mg/ml. 23. The method of claim 1, wherein the high protein concentration is from about 100 mg/ml to about 250 mg/ml. -