JP7057954B2 - Protein-containing liquid formulation with improved storage stability - Google Patents

Description

The present invention relates to a liquid preparation containing a protein such as an antibody, which has improved storage stability, specifically, a liquid preparation in which aggregate formation is reduced or suppressed, and a method for producing the same.

One of the problems of biopharmacy such as antibody drug is protein aggregation. Accurate quantification and reduction is desired because aggregates can contribute to immunogenicity.

For the storage stability of liquid protein preparations, it has been known to add substances such as surfactants, serum albumin, polysaccharides, (poly) amino acids, organic acids, and salts as stabilizers ( Patent Document 1, Patent Document 2, Patent Document 3).

Furthermore, regarding the long-term storage stability of proteins such as antibodies, a correlation may be found between the colloidal stability and structural stability of the antibody solution, which is a physicochemical parameter, and the formation of aggregates of a specific size. Is known (Non-Patent Document 1).

However, since the method of acquiring physicochemical parameters for many prescription conditions has been limited in the past, agglomeration formation (for example, agglomerate amount) can be performed with high accuracy by using a plurality of physicochemical parameters related to the above correlation. ) Was not known. Therefore, it has not been realized to accurately estimate the relationship between the physicochemical parameters and the amount of aggregates with respect to the storage stability of the antibody.

Japanese Unexamined Patent Publication No. 2016-145192 US Publication No. 2011-0020328 Re-Table 2005/033143 Gazette

Susumu Uchiyama, Annual Meeting of the Japanese Society of Toxicology 40.1 (0), P.M. 2043, W4-3 (Prediction of cohesiveness of antibody drugs), 2013

The selection of pharmaceutical conditions for biopharmacy, for example protein drugs such as antibody drugs, has been lacking in rational methods, so short-term by heating or shaking based on experience or by setting limited conditions. Accelerated tests (that is, tests that estimate long-term storage stability in a short period of time) have been performed by methods such as quantifying the amount of aggregates generated by a limited method and selecting a specific composition. However, the pharmaceutical conditions selected by such a method are based on the result of an accelerated test, which is not selected after understanding the physicochemical properties of the protein solution, and therefore the optimality of the conditions is rational. Since it is difficult to give a specific explanation and therefore it is difficult to confirm that it is optimal, and since the tendency of changes in physical properties when the composition is changed is unknown, it is necessary to search for the composition semi-ad-hoc again. In addition to being inefficient, there was the problem of taking an uncertain approach.

In order to solve such problems, the present inventors have physics such as diffusion interaction parameters (sometimes referred to as "second virial coefficient"), aggregation start temperature, modification midpoint temperature, etc. in a liquid preparation of an antibody. Obtain a chemical parameter and use the value as an index of the storage stability of the antibody to create a general formula for determining the rate of monomer reduction (or "aggregate formation rate"). Based on this, it is possible to predict the optimization conditions for making a liquid preparation containing a protein such as an antibody stable in storage, and this time, we have completed an invention characterized by the following.

The present invention includes the following features.
[1] A method for producing a storage-stable liquid preparation containing a protein.
(1) A step of preparing one or more liquid formulation candidates, which comprises a protein having a predetermined concentration and the buffer having a pH, sodium chloride and at least one of different types and / or amounts of additives in a predetermined buffer.
(2) A step of measuring the aggregation start temperature (Tagg) and diffusion interaction parameter (kD) of the protein for each of the liquid formulation candidates.
(3) The values of the aggregation start temperature (Tagg) and the diffusion interaction parameter (kD) for the liquid product candidate determined in the second step are set to the following formula I:
[Equation I]
Monomer reduction rate = -AB x kD + C x Tagg + (D x kD) x [-E x (Tagg-F)]
(In the formula, Tagg is the aggregation initiation temperature, kD is the diffusion interaction parameter, and A, B, C, D, E and F are different coefficients depending on the type of protein.)
To determine the rate of monomer reduction of the protein by substituting into,
(4) A step of selecting a liquid formulation candidate having the minimum monomer reduction rate determined in the third step from the one or a plurality of liquid formulation candidates.
(5) A step of producing a liquid preparation containing the protein based on the protein concentration, the buffer, and the type and amount of the pH, salt and additive in the liquid preparation candidate selected in the fourth step.
How to include.
[2] A method of selecting a storage-stable liquid preparation from among liquid preparation candidates containing a protein.
(1) A step of preparing one or more liquid formulation candidates, which comprises a protein having a predetermined concentration and the buffer having a pH, sodium chloride and at least one of different types and / or amounts of additives in a predetermined buffer.
(2) A step of measuring the aggregation start temperature (Tagg) and diffusion interaction parameter (kD) of the protein for each of the liquid formulation candidates.
(3) The values of the aggregation start temperature (Tagg) and the diffusion interaction parameter (kD) for the liquid product candidate determined in the second step are set to the following formula I:
[Equation I]
Monomer reduction rate = -AB x kD + C x Tagg + (D x kD) x [-E x (Tagg-F)]
(In the formula, Tagg is the aggregation initiation temperature, kD is the diffusion interaction parameter, and A, B, C, D, E and F are different coefficients depending on the type of protein.)
To determine the rate of monomer reduction of the protein by substituting into,
(4) A step of selecting a liquid formulation candidate having the minimum monomer reduction rate determined in the third step from the one or a plurality of liquid formulation candidates.
How to include.
[3] The method according to [1] or [2], wherein the protein is an antibody.
[4] The method according to [3], wherein the antibody is an antibody against human CD20.
[5] The method according to [4], wherein the antibody is TKM-011 or BM-ca.
[6] The method according to any one of [1] to [5], wherein the additive is sugar or sugar alcohol.
[7] The method according to [6], wherein the sugar or sugar alcohol is sorbitol, sucrose or a mixture thereof.
[8] The method according to any one of [1] to [7], wherein the additive further contains a surfactant.
[9] The method according to [8], which comprises the above-mentioned surfactant in an amount of 0.02% by weight or less.
[10] The method according to [8] or [9], wherein the surfactant is a nonionic surfactant.
[11] The method according to [10], wherein the nonionic surfactant is polysorbate.
[12] A storage-stable liquid preparation containing the antibody TKM011 or BM-ca against human CD20, wherein the storage-stable liquid preparation contains 0.05 to 30% by weight of the antibody, and has a pH of 4.0 to 6. A liquid preparation comprising 0 to 15% by weight of sugar or sugar alcohol, containing an acetate buffer, and containing 0 to 100 mM of sodium chloride.
[13] The liquid preparation according to [12], further comprising a surfactant.
[14] The liquid preparation according to [13], which contains the above-mentioned surfactant in an amount of 0.02% by weight or less.
[15] The liquid preparation according to [13] or [14], wherein the surfactant is a nonionic surfactant.
[16] The liquid preparation according to [15], wherein the nonionic surfactant is polysorbate.
[17] The liquid preparation according to any one of [12] to [16], wherein the sugar or sugar alcohol is sorbitol or sucrose.
[18] A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, which contains 0.05 to 30% by weight of the antibody, has a pH of 4.5 to 5.5, and is sorbitol or. A liquid preparation comprising 1 to 15% by weight of sucrose, containing an acetate buffer, containing in an amount of 0.015% by weight or less of polysorbate, and containing 0 to 60 mM of sodium chloride.
[19] A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, containing 0.05 to 30% by weight of the antibody, pH 5.0, and 6 to 12 weight of chloride. A liquid formulation comprising%, containing 20 mM acetate buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride.
[20] A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, containing 0.05 to 30% by weight of the antibody, pH 5.0, and 6% by weight of sorbitol. , A liquid formulation comprising 20 mM acetate buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride.
[21] A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, containing 0.05 to 30% by weight of the antibody, pH 5.0, and 6 to 12 weight of chloride. A liquid formulation comprising%, containing 20 mM acetate buffer, containing 0.01% by weight of polysorbate, and containing 60 mM of sodium chloride.
[22] A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, containing 0.05 to 30% by weight of the antibody, having a pH of 5.0, containing no sugar, and 20 mM. A liquid preparation comprising an acetic acid buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride.
[23] A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, containing 0.05 to 30% by weight of the antibody, pH 5.0, and 3% by weight of sorbitol. , A liquid formulation comprising 20 mM acetate buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride.
[24] A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, containing 0.05 to 30% by weight of the antibody, pH 5.0, sugar-free, 20 mM. A liquid preparation comprising acetate buffer, 0.01% by weight of polysorbate, and 60 mM of sodium chloride.

The present invention accurately determines the storage stability (particularly aggregate formation) in a liquid preparation containing a protein such as an antibody simply by determining the aggregation initiation temperature (Tagg) and the diffusion interaction parameter (kD) of the antibody in the preparation candidate sample. It provides the advantage of being able to predict well and in a short time.

The measurement results of the diffusion interaction parameters (kD) (A) and the aggregation start temperature (Tagg) (B) of the anti-humanized CD20 antibody (“TKM-011”) in each sample under the storage conditions shown in Table 5 below are shown. show. Measurement results (A) of turbidity (absorbance (ABS) (A350)) after storage of the sample under the storage conditions shown in Table 5 (described later) at 4 ° C for 1 month, 3 months and 6 months, and 40 ° C for 1 month. The measurement result (B) of the turbidity (ABS (A350)) after storage for 3 months and 6 months is shown. Measurement results (A) of the soluble component (ABS (A280)) after storage of the sample under the storage conditions shown in Table 5 (described later) at 4 ° C for 1 month, 3 months and 6 months, and 40 ° C for 1 month and 3 months. And the measurement result (B) of the soluble component (ABS (A280)) after storage for 6 months is shown. Measurement results (A) of SEC (monomer%) of soluble components after storage of the samples under the storage conditions shown in Table 5 (described later) at 4 ° C for 1 month, 3 months and 6 months, and 40 ° C for 1 month and 3 months. And the measurement result (B) of the SEC (monomer%) of the soluble component after storage for 6 months is shown. A comparison of chromatograms after storing sample 13 (acetate buffer, 6% sucrose) at 4 ° C and 40 ° C for 6 months is shown. The inset is a 10-fold magnified view of the chromatogram. Measurement results (A) of flow imaging (FI) analysis (particle size 1 μm or more) after storage of the samples under the storage conditions shown in Table 5 (described later) at 4 ° C for 1 month, 3 months and 6 months, and 40 ° C 1 The measurement result (B) of the flow imaging (FI) analysis (particle size 1 μm or more) after storage for months, 3 months and 6 months is shown. Radar diagrams of samples 11 and 12 under the storage conditions shown in Table 5 (described later) after storage at 4 ° C for 1 month, 3 months and 6 months (A350, recovery rate, monomer area, monomer%, FI (p / ml)). ) Is shown. Radar diagrams of samples 13 and 14 under the storage conditions shown in Table 5 (described later) after storage at 4 ° C for 1 month, 3 months and 6 months (A350, recovery rate, monomer area, monomer%, FI (p / ml)). ) Is shown. Radar diagrams of samples 15 and 16 under the storage conditions shown in Table 5 (described later) after storage at 4 ° C for 1 month, 3 months and 6 months (A350, recovery rate, monomer area, monomer%, FI (p / ml)). ) Is shown. Radar diagrams of samples 17 and 19 under the storage conditions shown in Table 5 (described later) after storage at 4 ° C for 1 month, 3 months and 6 months (A350, recovery rate, monomer area, monomer%, FI (p / ml)). ) Is shown. Radar diagrams of samples 11 and 12 under the storage conditions shown in Table 5 (described later) after storage at 40 ° C. for 1 month, 3 months and 6 months (A350, recovery rate, monomer area, monomer%, FI (p / ml)). ) Is shown. Radar diagrams of samples 13 and 14 under the storage conditions shown in Table 5 (described later) after storage at 40 ° C. for 1 month, 3 months and 6 months (A350, recovery rate, monomer area, monomer%, FI (p / ml)). ) Is shown. Radar diagrams of samples 15 and 16 under the storage conditions shown in Table 5 (described later) after storage at 40 ° C. for 1 month, 3 months and 6 months (A350, recovery rate, monomer area, monomer%, FI (p / ml)). ) Is shown. Radar diagrams of samples 17 and 19 under the storage conditions shown in Table 5 (described later) after storage at 40 ° C. for 1 month, 3 months and 6 months (A350, recovery rate, monomer area, monomer%, FI (p / ml)). ) Is shown.

The present invention will be described in more detail below.

The present invention is a method for producing a storage-stable liquid preparation containing a protein according to the first aspect.
(1) Prepare one or more liquid formulation candidates containing a protein having a predetermined concentration and the buffer having a pH, sodium chloride (NaCl) and at least one of different types and / or amounts of additives in a predetermined buffer. Process,
(2) A step of measuring the aggregation start temperature (Tagg) and diffusion interaction parameter (kD) of the protein for each of the liquid formulation candidates.
(3) The values of the aggregation start temperature (Tagg) and the diffusion interaction parameter (kD) for the liquid product candidate determined in the second step are set to the following formula I:
[Equation I]
Monomer reduction rate = -AB x kD + C x Tagg + (D x kD) x [-E x (Tagg-F)]
(In the formula, Tagg is the aggregation initiation temperature, kD is the diffusion interaction parameter, and A, B, C, D, E and F are different coefficients depending on the type of protein.)
To determine the rate of monomer reduction of the protein by substituting into,
(4) A step of selecting a liquid formulation candidate having the minimum monomer reduction rate determined in the third step from the one or a plurality of liquid formulation candidates.
(5) A step of producing a liquid preparation containing the protein based on the protein concentration, the buffer, and the type and amount of the pH, salt and additive in the liquid preparation candidate selected in the fourth step.
Provide methods including.

The "predetermined concentration" of protein in the present specification means that the concentration of the protein is specified, for example, 10% by weight. Also, the "predetermined" buffer in the specification refers to a buffer being specified, for example, a 20 mM acetate buffer.

The present invention is a method for selecting a storage-stable liquid preparation from among liquid preparation candidates containing a protein according to the second aspect.
(1) Prepare one or more liquid formulation candidates containing a protein having a predetermined concentration and the buffer having a pH, sodium chloride (NaCl) and at least one of different types and / or amounts of additives in a predetermined buffer. Process,
(2) A step of measuring the aggregation start temperature (Tagg) and diffusion interaction parameter (kD) of the protein for each of the liquid formulation candidates.
(3) The values of the aggregation start temperature (Tagg) and the diffusion interaction parameter (kD) for the liquid product candidate determined in the second step are set to the following formula I:
[Equation I]
Monomer reduction rate = -AB x kD + C x Tagg + (D x kD) x [-E x (Tagg-F)]
(In the formula, Tagg is the aggregation initiation temperature, kD is the diffusion interaction parameter, and A, B, C, D, E and F are different coefficients depending on the type of protein.)
To determine the rate of monomer reduction of the protein by substituting into,
(4) A step of selecting a liquid formulation candidate having the minimum monomer reduction rate determined in the third step from the one or a plurality of liquid formulation candidates.
Provide methods including.
The above selection method corresponds to the steps (1) to (4) in the above manufacturing method.

1. 1. Protein As used herein, "protein" includes any therapeutic protein used in medical applications. Such proteins include proteins from mammals including humans, antibodies, proteins from plants, proteins from pathogenic microorganisms, proteins from pathogenic viruses, etc., which are naturally or artificially produced. Proteins (eg, recombinant proteins) are included. In particular, proteins that tend to form protein aggregates when stored in the form of liquid formulations are suitable for production by the above method of the invention. One example of such a protein is an antibody.

The antibody includes any of a complete antibody, an antibody fragment, a recombinant antibody, a single chain antibody, a human antibody, a humanized antibody, and a chimeric antibody.

The antibody drug contains an antibody that specifically binds to a protein-related disease-related molecule as an active ingredient, and has a function of suppressing the function of the molecule in a living body. Antibody drugs include, but are not limited to, for example, cancers (eg, B-cell non-Hodgkin lymphoma, Hodgkin lymphoma, acute lymphoblastic leukemia, breast cancer, acute myeloid leukemia, melanoma, gastric cancer, colon / rectal cancer, head and neck). Autoimmune diseases (eg rheumatoid arthritis, psoriasis vulgaris, multiple sclerosis, severe Crohn's disease, etc.), infectious diseases (eg RS virus infection, etc.), Alzheimer's disease, osteoporosis, acute post-renal transplantation Drugs comprising antibodies as active ingredients for the treatment of rejection, heart disease (eg, myocardial ischemia, etc.), allergic diseases (eg, asthma, etc.), eye diseases (eg, age-related luteal degeneration, etc.) are included. Target proteins are also diverse, such as CD3, CD11, CD20, CD25, CD33, CD52, CD80 / CD86, TNFα, IL-6R, IL-1β, EGFR, RSVF protein, IgE, CCR4, HER2, PD- 1, IL12, TLA4, VEGFR2, VEGF and the like are included, but the present invention is not limited thereto. In the examples described later, an antibody against CD20 (for example, TKM-011 or BM-ca) was exemplified, and the optimum formulation conditions were examined.

In recent years, from the viewpoint of convenience in the medical field, the form of the preparation has shifted to a liquid preparation (vial) having a low protein component concentration or a liquid preparation (syringe) having a high protein component concentration instead of a solid preparation such as a freeze-dried preparation. (S. Uchiyama, Biochim. Biophys. Acta. 2014; 1844: 2041-2052). As described above, even if the liquid protein preparation can be stored at 4 ° C for 2 to 3 years, if stress is applied during that period, the protein is decomposed and aggregated, and the generated aggregate causes allergic shock. Therefore, it is necessary to optimize the formulation conditions (for example, protein concentration, pH, salt concentration, type and amount of additives, etc.) to make the formulation highly stress resistant and highly stable. The stress on the pharmaceutical product can occur at any stage of manufacturing, transportation, storage and administration. Specifically, the pharmaceutical product is stressed by liquid feeding, interfacial contact, sweeping, vibration, stirring, handling, impact, pressurization and depressurization, temperature change, light, oxygen, change in solution composition, and the like.

Therefore, in the following, a method of obtaining results as quickly as possible regarding the storage stability of a liquid preparation containing a protein and a method of selecting a preparation having high storage stability as efficiently as possible will be described.

2. 2. Selection and Production Method of Liquid Formulation Stable for Storage According to the present invention, efficient formulation conditions can be determined.
The physicochemical parameters involved in protein aggregation in protein solution include structural stability and colloidal stability in solution.

Colloidal stability can be evaluated by the ease of association of protein molecules, indicating that the dispersed state of the protein is good when the colloidal stability is high, while the dispersed state of the protein is good when the colloidal stability is low. Indicates bad. Colloidal stability is not limited to this, but can be evaluated by obtaining a second virial coefficient (B2) for a protein solution such as an antibody by an ultracentrifugal sedimentation equilibrium method. A positive B2 indicates that a repulsive force acts between the protein molecules, while a negative B2 indicates that an attractive force acts between the proteins. The diffusion interaction parameter (kD) has a high correlation with this second virial coefficient (B2), and the larger the kD, the better the dispersibility. Here, the relationship between kD and B2 can be expressed by the formula kD = B2-R (where R is a constant that changes depending on the type and composition of the antibody or solvent). B2 is an absolute value from which it can be determined whether the intramolecular force is repulsive or attractive, whereas kD is due to the contribution of the constant R, and from that value it is determined whether the intramolecular force is repulsive or attractive. It is not possible. As a relative comparison, the larger the value of kD, the better the dispersibility.

Structural stability indicates the degree to which a protein molecule can exist without denaturation, and the higher the structural stability, the less likely it is to denature. Since proteins usually aggregate when denatured, structural stability can be evaluated by the aggregation start temperature (Tagg) when the protein is heated. The aggregation start temperature can be determined by monitoring the temperature at which the size of the particles due to aggregation increases by a dynamic light scattering method. That is, when the temperature of the protein solution is gradually increased, the temperature at which the particle size begins to increase is the aggregation start temperature. In the dynamic light scattering method, for example, the fluctuation of the scattered light according to the diffusion coefficient can be detected, and the particle size can be measured by using, for example, the Stokes-Einstein equation. The higher the stability, the higher the agglomeration does not occur, so the structural stability can be evaluated by the agglomeration start temperature.

We have now found that both the aggregation initiation temperature (Tagg) and the diffusion interaction parameter (kD) are pH, types of salts (eg, sodium chloride, etc.) and additives (eg, sugars, surfactants, etc.) and /. Or we found that there is a strong correlation with quantity.

We further determined these physicochemical parameters and performed a correlation analysis for the rate of monomer reduction. First, the stepwise method was used to determine the physicochemical parameters that correlate with the rate of monomer reduction, that is, the explanatory variables in the multiple regression analysis. As a result, it was found that the aggregation start temperature, the diffusion interaction parameter, and their multiplication have a correlation with the monomer decrease rate. Then, using those explanatory variables, the following formula I is derived by performing multiple regression analysis on the rate of monomer decrease, and the storage stability test is performed by substituting the measured values of Tagg and kD into the formula I. We found that we could predict the result of.

[Equation I]
Monomer reduction rate = -AB x kD + C x Tagg + (D x kD) x [-E x (Tagg-F)]
(In the formula, Tagg is the aggregation start temperature, kD is the diffusion interaction parameter (indicating the concentration dependence of the diffusion coefficient), and A, B, C, D, E and F are different depending on the type of protein. It is a coefficient.)

The term "storage stability" as used herein is based on the dispersed state of protein components and the degree of aggregation due to, for example, denaturation, when the liquid preparation is stored for a long period of time, for example, 1 month or more or 6 months or more. "Good storage stability" means that the protein component has good dispersibility in the liquid, and the aggregation start temperature is high, or there is no aggregation or it is difficult to aggregate.

The specific procedure for determining the formula for determining the rate of monomer decrease is as follows.
In the first step, a plurality of liquid product candidates having different pH and salt concentration (for example, sodium chloride concentration) are prepared.

In the second step, the protein aggregation start temperature (Tagg) and the diffusion interaction parameter (kD) are measured for each of the liquid product candidates prepared in the first step. Thereby, the correlation between the physicochemical parameters Tagg and kD and the pH and / or the salt concentration is obtained.

Based on the assumption that higher values of kD and Tagg are more stable in the third step, the pH and / or salt concentration that raises kD and Tagg is calculated from the correlation determined in the second step. Add varying types and / or varying amounts (or different types and / or amounts) of additives (eg, sugars, sugar alcohols, surfactants, or mixtures thereof, etc.) and measure kD and Tagg again. , KD and Tagg are high to determine the conditions. This determines the solution conditions (pH, salt concentration and additives) that increase Tagg and kD.

In the fourth step, a stability test is carried out on the liquid pharmaceutical candidate candidates specified in the third step. As a result, numerical parameters related to aggregate formation such as the rate of monomer decrease are acquired.

In the fifth step, the relational expression between the result of the stability test obtained in the fourth step and kD and Tagg is calculated, and a stable region map is obtained.

Further, the coefficient can be calculated as follows.
The objective variable is the monomer reduction rate, the explanatory variables are set to parameters such as kD and Tagg, and the explanatory variables are determined by the stepwise method. Then, the model is fitted with kD and Tagg as explanatory variables by the least squares method, and the coefficients are determined.

For antibody TKM-011, formulas I A, B, C, D, E and F were determined, thus deriving Formula II below.

[Equation II]
Monomer reduction rate = -2623086 + 8670 x kD + 36966 x Tagg + (8.37 + kD) x [-5453 x (Tagg-45.3)]

In the method for producing a storage-stable liquid preparation containing a protein,
In the first step, one or more liquid formulation candidates are prepared, which comprises a protein having a predetermined concentration and the buffer having a pH, sodium chloride and at least one of different types and / or amounts of additives in a predetermined buffer.

In the second step, the protein aggregation start temperature (Tagg) and the diffusion interaction parameter (kD) are measured for each of the liquid product candidates.

That is, Tagg and kD, which are physicochemical parameters of liquid preparation candidates when the composition of pH, salt (ionic strength), additives, etc. is changed, are obtained, while "stress loading and after a lapse of time" "Aggregate amount and formation rate of each size" is evaluated.

As "amount or formation rate of agglomerates of each size", after stress application (short-term acceleration) such as heating and shaking or long-term storage test, the particle size of the agglomerates is 100 nm or less, 100 nm to 1 μm, 1 μm to 100 μm. , And a quantitative value may be obtained, and the formation rate may be calculated if necessary.

Correlation function between all formulation candidates and "aggregate amount or formation rate of each size", for "aggregate amount or formation rate of each size" under each condition, of "physicochemical parameters" Using the values, the stepwise method in correlation analysis is used to select "physicochemical parameters" that correlate with "aggregate amount or formation rate of each size", determine explanatory variables in multiple regression analysis, and then weight. By performing regression analysis, the following formula III can be derived, and the coefficient can be determined for each protein to predict aggregation stability.

[Equation III]
Aggregate amount or formation rate of each size = intercept + coefficient A × kD + coefficient B × Tagg + coefficient C × kD × Tagg

In the third step, each value of the aggregation start temperature (Tagg) and the diffusion interaction parameter (kD) for the liquid formulation candidate determined in the second step is substituted into the above formula I to obtain the above protein. Determine the rate of monomer reduction.

In the fourth step, the liquid formulation candidate having the minimum monomer reduction rate determined in the third step is selected from the one or a plurality of liquid formulation candidates.

So far, it has been reported that there is a correlation between a specific "physicochemical parameter" and some "aggregate amount", but two or more types of "physicochemical parameters" that are directly related to aggregation stability. There was no example in which the prediction was realized by combining and the "aggregate amount" of each size. In that sense, the method of the present invention is epoch-making in that it makes it possible to predict the storage stability of a liquid pharmaceutical product of an antibody by a simple method.

The aggregation start temperature (Tagg) and the diffusion interaction parameter (kD) can be measured by, for example, the following methods.

Polymer (aggregate) formation can be measured by analysis of particle size by SEC (Size Exclusion Column Chromatography).

Absorbance (ABS) (A280 soluble component and A350 turbidity) analysis can be used to measure soluble monomer%, monomer (peak) area, recovery, and presence of aggregates.

In FI (flow imaging), the number of particles can be counted and the particle shape can be measured by flow site particle image analysis (exemplified device: FlowCam ^TM (Fluid Imaging Technologies)). With such an apparatus, for example, protein aggregates can be detected.

Dynamic light scattering monitors the temperature at which the size of the particles grows. At this time, the higher the stability, the higher the temperature does not aggregate.

In the fifth step, a liquid preparation containing the above protein is produced based on the protein concentration, the buffer, and the type and amount of the pH, salt and additive in the liquid preparation candidate selected in the fourth step.

The buffer is preferably a buffer having a pH of 4 to 6, and more preferably an acetic acid buffer. The concentration of the buffer is, for example, 10 mM to 100 mM, 15 mM to 50 mM, and the like.

Additives include, for example, sugars, surfactants, sugar alcohols, alcohols, vitamins, amino acids and the like.

An example of the formulation of a liquid formulation determined by the above method is, but is not limited to, when the protein is an antibody, it contains 0.05 to 30% by weight of the antibody and has a pH of 4.0 to 6.0. , A liquid preparation containing 0 to 15% by weight of sugar or sugar alcohol, containing acetic acid buffer, and containing no sodium chloride or 100 mM or less, for example 60 mM or less.

The sugar or sugar alcohol is, for example, sorbitol, sucrose, trehalose, mannitol or a mixture thereof.

The liquid preparation may further contain a surfactant in an amount of 0.02% by weight or less. The surfactant is, for example, a nonionic surfactant, but any surfactant may be used as long as it can be used as an additive for pharmaceutical products. Examples of nonionic surfactants are polysorbate and poloxamer.

Furthermore, according to the embodiment, when the antibody is a liquid preparation which is an antibody against human CD20 (for example, TKM-011 or BM-ca), the preparation having improved storage stability is such that the antibody is 0.05 to 30. It contains an acetate buffer, which contains% by weight (eg, 0.05 to 5% by weight), has a pH of 4.5 to 5.5, contains 0 to 15% by weight of sorbitol or chloride (eg, 4 to 6% by weight), and contains. It is characterized by containing polysolvate in an amount of 0.015% by weight or less (for example, 0.01% by weight), and containing no sodium chloride or 60 mM or less of sodium chloride (that is, sodium chloride 0 to 60 mM). do.

Further non-limiting examples of the pharmaceutical product are shown below.
(Example 1) A storage-stable liquid preparation containing an antibody TKM-011 or BM-ca against human CD20, which contains 0.05 to 30% by weight of the antibody, has a pH of 5.0, and contains 6 to 12 chlorides. A liquid formulation comprising weight%, containing 20 mM acetate buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride.

(Example 2) A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, which contains 0.05 to 30% by weight of the antibody, has a pH of 5.0, and contains 6% by weight of sorbitol. A liquid formulation comprising, containing 20 mM acetate buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride.

(Example 3) A storage-stable liquid preparation containing an antibody TKM-011 or BM-ca against human CD20, which contains 0.05 to 30% by weight of the antibody, has a pH of 5.0, and contains 6 to 12 chlorides. A liquid formulation comprising wt%, 20 mM acetate buffer, 0.01 wt% polysorbate, and 60 mM sodium chloride.

(Example 4) A storage-stable liquid preparation containing an antibody TKM-011 or BM-ca against human CD20, containing 0.05 to 30% by weight of the antibody, having a pH of 5.0, and containing no sugar. A liquid formulation comprising 20 mM acetate buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride.

(Example 5) A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, which contains 0.05 to 30% by weight of the antibody, has a pH of 5.0, and contains 3% by weight of sorbitol. A liquid formulation comprising, containing 20 mM acetate buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride.

(Example 6) A storage-stable liquid preparation containing an antibody TKM-011 or BM-ca against human CD20, containing 0.05 to 30% by weight of the antibody, having a pH of 5.0, and containing no sugar. A liquid formulation comprising 20 mM acetate buffer, containing 0.01% by weight of polysorbate, and containing 60 mM of sodium chloride.

The present invention will be described in more detail with reference to the following examples. However, the scope of the invention is not limited by these examples.

[Example 1]
Examination of storage stabilization conditions for TKM-011 liquid preparation (Part 1)
1. 1. Test procedure Apply antibody TKM-011 vial (5 mg / mL, 10 mL) (manufactured by Special Immunology Laboratory) to the cation exchange column HiTrap SP HP (manufactured by GE Healthcare) and eluent (start buffer: 20 mM sodium acetate, pH 5). .0, Elution buffer: 100 mM phosphoric acid, 400 mM NaCl, pH 8.0) was flowed to remove polysorbate (PS) 80, and then dialyzed in distilled water to recover TKM-011.

The amino acid sequences of the H-chain variable region and the L-chain variable region of the antibody TKM-011 (humanized antibody; also referred to as BM-ca) are described in US Pat. Nos. 8,101,179B2 for SEQ ID NO: 19 and SEQ ID, respectively. NO: 25 amino acid sequence.

Test samples were prepared to meet the following test conditions.

7 antibodies were prepared per condition (antibody 1 mg / mL, 1 mL, N = 1).
Test conditions:
Samples were prepared for various combinations of pH, solvent (buffer), salt, sugar, and polysorbate 80.
(1) Diffusion interaction parameter kD (dynamic light scattering method DLS); 67 conditions
pH; 5.0,5.5,6.0,6.5,7.0
Solvent; 20 mM acetic acid, 20 mM citric acid, 20 mM histidine, 20 mM phosphate salt; 0,20 mM NaCl, 150 mM NaCl, 200 mM NaCl
Sugar; 0,5% sucrose, 10% sucrose, 5% sorbitol, 10% sorbitol Polysorbate 80; 0,0.01%, 0.07%
(2) Aggregation start temperature Tagg (dynamic light scattering method DLS); 27 conditions
pH; 5.0,6.0,7.0
Solvent; 20 mM acetic acid, 20 mM citric acid, 20 mM histidine, 20 mM phosphate salt; 0,20 mM NaCl, 200 mM NaCl
Sugar; 0,5% sucrose, 10% sucrose Polysorbate 80; 0,0.01%, 0.07%
(3) Aggregation start temperature Tagg (trace protein characteristic analysis system OPTIM ^TM (Avacta)); 128 conditions
pH; 5.0,5.5,6.0,6.5,7.0
Solvent; 20 mM acetic acid, 20 mM citric acid, 20 mM histidine, 20 mM phosphate salt; 0,20 mM NaCl, 150 mM NaCl, 200 mM NaCl
Sugar; 0,5% sucrose, 10% sucrose, 5% sorbitol, 10% sorbitol Polysorbate 80; 0,0.01%, 0.07%

There were two types of storage conditions, 4 ° C and 40 ° C.
In addition, the storage period was set to three types of 1, 3, and 6 months.

Measurements and analyzes were performed at 1, 3 and 6 months after storage. The measurement was performed using SEC, ABS (A280 soluble component, A350 turbidity), and FI.

2. 2. Measurement results It was found that there is a strong correlation between Tm (CH2) of antibody TKM-011 and pH, NaCl, and sugar (sucrose or sorbitol) (Table 1). Here, Tm indicates the denaturation midpoint temperature of the differential scanning calorimetry (DSC) curve of the heat denaturation process and is an index of thermal stability, and Tm (CH2) is the Tm of the constant region CH2 of the antibody TKM-011.

Furthermore, in the table below, the estimated value is the coefficient of each explanatory variable calculated by multiple regression analysis, and the larger the absolute value, the greater the influence on the objective variable, and the positive or negative indicates the positive or negative of the correlation. The standard error refers to the standard deviation of the estimated value, and is an index showing the magnitude and estimation accuracy of the variation of the estimated value itself obtained from the sample. The p-value represents the significance probability of the coefficients of each explanatory variable in multiple regression analysis. The t-value represents the magnitude of the influence of each explanatory variable on the objective variable in the multiple regression analysis, and the larger the absolute value, the stronger the influence.

Specifically, from the results in Table 1, there is a negative correlation for pH and NaCl. That is, this indicates that the Tm value is higher at low pH and low concentration NaCl, in other words, the stability is high.

Further, since both sugars have a positive correlation, it is shown that the high-concentration sugar has a high Tm value, in other words, has high stability.

The strength of the correlation is lower in the order of NaCl>Sorbitol>pH> Sucrose. Correlation analysis was performed using JMP ^(R) Pro12.2.0 (SAS Institute INC.) Software.

Furthermore, it was found that there is a correlation between Tagg (OPTIM) of antibody TKM-011 and pH, NaCl, sugar (sucrose, sorbitol) and polysorbate 80 (PS80) (Table 2). Correlation analysis was performed using JMP ^(R) Pro12.2.0 (SAS Institute INC.) Software.

The specific results are as follows.
Correlated: pH, NaCl, sugar (sucrose, sorbitol), PS80
Positive correlation: Sugar Negative correlation: pH, NaCl, PS80
Strength of correlation: NaCl>Sorbitol>Sucrose>pH> PS80

Furthermore, it was found that there is a correlation between Tagg (DLS) of antibody TKM-011 and pH, NaCl and sucrose (Table 3). Correlation analysis was performed using JMP ^(R) Pro12.2.0 (SAS Institute INC.) Software.

The specific results are as follows.
Correlated: pH, NaCl, sucrose
Positive correlation: Sugar Negative correlation: pH, NaCl, PS80
Strength of correlation: NaCl>Sucrose> pH

Furthermore, it was found that there is a correlation between the kD of antibody TKM-011 and pH, NaCl and sugar (sucrose, sorbitol) (Table 4). Correlation analysis was performed using JMP ^(R) Pro12.2.0 (SAS Institute INC.) Software.

The specific results are as follows.
Correlated: pH, NaCl, sugar (sucrose, sorbitol)
Negative correlation: pH, NaCl, strength of sugar correlation: Sucrose>pH>NaCl> Sorbitol

Next, the influence of the solvent type was investigated. Acetic acid buffer, citric acid buffer, histidine buffer and phosphate buffer were used as solvent types. Tm (CH2), Tagg (OPTIM) and kD were measured in the same manner as above.

As a result, it was found that the acetic acid buffer was improved for all of Tm, kD and Tagg under the conditions of low pH and low salt concentration.

The above results are summarized below.

(1) Statistical processing revealed that NaCl was not added to all parameters, and that the pH on the acidic side further enhanced the stability of the antibody in solution.

(2) Regarding the addition of sugar, it was clarified that it is preferable to add it from the viewpoint of structural stability, but it is preferable not to add it from the viewpoint of colloidal stability. It was revealed that the addition of PS80 did not have a significant effect.

(3) Examples of pharmaceutical conditions that are expected to show good stability are as follows.
20 mM acetate buffer pH 5.0, NaCl concentration 0 mM, sorbitol 5%, PS80 0.01%

(4) Examples of pharmaceutical conditions that are expected to exhibit moderate stability are as follows.
20 mM phosphate buffer pH 5.0, NaCl concentration 200 mM

(5) Examples of pharmaceutical conditions that are expected to show poor stability are as follows.
20 mM phosphate buffer pH 7.0, NaCl concentration 200 mM

[Example 2]
Examination of storage stabilization conditions for TKM-011 liquid preparation (Part 2)
1. 1. Test procedure Using the same test procedure as in Example 1 above, the test samples in Table 5 below were evaluated for colloidal stability and structural stability. The samples were selected from the results of accelerated tests (shaking and heating), and were selected under 8 conditions (13,15,19,14,11,17,16,12) with good total evaluation, and 2 conventional conditions (41,42). It consists of a total of 10 conditions.

The test conditions were 40 ± 2 ° C and 4 ± 2 ° C, 75% RH ± 5%, and 1, 3, 6 months, and SEC, ABS (A280 soluble component and A350 turbidity), FI immediately after storage. Was measured and stability was evaluated. Tagg and kD were measured before the start of the storage test.

2. 2. Measurement results The results of measuring Tagg, kD, SEC, ABS (A280 soluble component and A350 turbidity) and FI are shown in FIG. 1A with kD, FIG. 1B with Tagg, FIG. 2 with A350 turbidity, and FIG. 3 with A280 soluble component. FIG. 4 shows SEC (monomer%), and FIG. 6 shows FI total particle%.

After the 40 ° C. 1-month storage test, after the 40 ° C. 3-month storage test, and after the 40 ° C. 6-month storage test, white turbidity cannot be visually confirmed in sample 13 (acetic acid buffer, 6% sucrose), but sample 41 (1st pumpin). ), In sample 42 (2nd buffer), it became clearly cloudy after the storage test at 40 ° C. for 1 month, and a precipitate was confirmed.

In FIG. 1A, it is shown that the larger the kD, the better the dispersibility. From this point of view, the samples showing a large kD are 11, 13 and 19, and the samples showing a negative but relatively small kD are 12, 15 and 17.

In FIG. 1B, the higher the temperature of Tagg, the higher the temperature at which aggregate formation is started. Looking at the samples whose Tagg is 45 ° C. or higher, they are 11 to 17 and 19.

From FIG. 2A, the value of A350 was almost the same as the value at 0 hours in the acetic acid buffer sample (11 to 19) when stored at 4 ° C. for 6 months, but the value of A350 was slightly similar to the value in the conventional condition (41). Has risen. From FIG. 2B, it was confirmed that the value of A350 at 40 ° C. for 6 months had the same tendency as the result of storage at 40 ° C. for 3 months. In the acetic acid buffer sample, the value of A350 increased in samples 12, 14 and 16 as compared with the value at 0 hour, and in the citric acid buffer sample and the conventional conditions, the value of A350 increased significantly.

From FIG. 3A, the value of A280 was slightly decreased in the samples 14 to 19 of the acetic acid buffer samples (11 to 19) after storage at 4 ° C. for 6 months as compared with the value at 0 hour. Under the conventional condition (41), the value of A280 was slightly larger than the value at 0 hours, and it is considered that the scattering of aggregates of a size that cannot be removed even after centrifugation affected the value of A280.

From FIG. 3B, when stored at 40 ° C. for 6 months, the value of A280 was clearly lower than the value of 0 hours in the samples 12 and 14 of the acetate buffer samples (11 to 19), but the value of A280 was 0 hours in the other samples. It was slightly larger than the value of A280, and it is considered that the scattering of aggregates of a size that cannot be removed even after centrifugation affected the value of A280. Under the conventional conditions (sample 41), the value was clearly lower than 0 hours.

Note that FIG. 3 shows the recovery rate of the A280 value of the supernatant after centrifugation at 13,200 rpm for 30 minutes, with 0 hour as 100% (the same applies hereinafter).

From FIG. 4A, when stored at 4 ° C. for 6 months, the value of the soluble component SEC monomer (%) was slightly lower than the value at 0 hour. Further, from FIG. 4B, the value of the SEC monomer (%) of the soluble component was clearly lower than the value at 0 hours when stored at 40 ° C. for 6 months.

In addition, from FIG. 5, a comparison of chromatograms of sample 13 (acetic acid buffer, 6% sucrose) after storage at 4 ° C and 40 ° C for 6 months showed that the main peak was significantly reduced in the sample stored at 40 ° C for 3 months. An increase in the peaks considered to be oligomers and multiple peaks considered to be derived from decomposition products was confirmed. Other samples had similar chromatographic patterns.

From FIG. 6A, when stored at 4 ° C. for 6 months, the number of particles increased by 1 μm or more from 0 hours.
From FIG. 6B, in the acetic acid buffer sample (11 to 19), the particles of 1 μm or more increased about 10 times from 0 hours in the samples 11 to 15 when stored at 40 ° C. for 6 months. Under the conventional conditions (samples 41 and 42), the number of particles having a size of 1 μm or more increased 100 times or more from 0 hours.

Furthermore, stable storage conditions based on the measurement results of SEC (monomer%), FI, A280 (recovery) and A280 (soluble) after storage at 4 ° C for 6 months and 40 ° C for 6 months are selected in good order. The results are shown in Tables 6 and 7.

Good conditions when selected from accelerated test results The top 5 types of samples 11, 13, 14, 15, 19 are stored at 4 ° C and 40 ° C for 6 months, and when the number of particles of 1 μm or more in FI is selected in good order, they are ranked high. The result is located in.

Of the above samples, the sample showing the radar diagram (A350, recovery rate, monomer area, monomer%, FI (p / ml)) that provides good conditions (storage at 4 ° C for 1 month, 3 months, 6 months) is a sample. 11 and 12 (FIG. 7A), samples 13 and 14 (FIG. 7B), samples 15 and 16 (FIG. 7C), and samples 17 and 19 (FIG. 7D). In addition, a radar diagram (A350, recovery rate, monomer area, monomer%, FI (p / ml)) when these samples were stored at 40 ° C. for 1 month, 3 months, and 6 months is shown in FIGS. 8A for samples 11 and 12. Samples 13 and 14 are shown in FIG. 8B, samples 15 and 16 are shown in FIG. 8C, and samples 17 and 19 are shown in FIG. 8D, respectively. It can be seen that even at 40 ° C., when stored for 1 month, all show good storage stability.

Table 8 shows the results of ranking the samples of the pharmaceutical conditions shown in Table 5 above in order of good order of the pharmaceutical conditions, judging from the radar diagrams (stored at 4 ° C and 40 ° C).

From Table 8, it was found that the pharmaceutical conditions of the samples 17, 19, 11, 13, 15, 16, 14 and 12 impart excellent storage stability.

[Example 3]
Preparation of Monomer Decrease Rate Formula for Antibody TKM-011 The monomer decrease rate was calculated from the amount of change over time in the area value of the monomer obtained by SEC analysis. After that, for the rate of monomer decrease, the values of the physicochemical parameters were used, the explanatory variables were determined by the stepwise method, and then multiple regression analysis was performed to calculate the equations and the coefficients. The stepwise method and multiple regression analysis were performed using JMP ^(R) Pro12.2.0 (SAS Institute INC.) Software.

The specific procedure is as follows.

In the first step, a plurality of liquid product candidates having different pH and sodium chloride concentration are prepared.
In the second step, the protein aggregation start temperature (Tagg) and the diffusion interaction parameter (kD) are measured for each of the liquid formulation candidates prepared in the first step. As a result, the correlation between the physicochemical parameters Tagg and kD and the pH and salt concentration is obtained.

Based on the premise that the higher the kD and Tagg values are, the more stable it is in the third step, the pH and salt concentration that increase the kD and Tagg are calculated from the correlation determined in the second step, and the additive is added. (Sugar, polysorbate 80, etc.) is added, and kD and Tagg are measured again to determine the conditions under which kD and Tagg are high. Thereby, the solution conditions (pH, salt concentration, additives) for increasing Tagg and kD are determined.

In the fourth step, a stability test is carried out on the liquid pharmaceutical candidate candidates specified in the third step. As a result, numerical parameters related to aggregate formation such as the rate of monomer decrease are acquired.

In the fifth step, the relational expression between the result of the stability test obtained in the fourth step and kD and Tagg is calculated, and a stable region map is obtained.

By the above procedure, the following formula was obtained with respect to the monomer reduction rate of the pharmaceutical product of antibody TKM-011.

<Formula>
Monomer reduction rate = -2623086 + 8670 x kD + 36966 x Tagg + (8.37 + kD) x [-5453 x (Tagg-45.3)]

The coefficient was calculated as follows.
The objective variable was set to the monomer decrease rate, the explanatory variables were set to parameters such as kD and Tagg, and the explanatory variables were determined by the stepwise method. As a result, it was confirmed that kD and Tagg have an influence as explanatory variables when the rate of monomer decrease is used as the objective variable from the p-value. Then, the model was fitted with kD and Tagg as explanatory variables by the least squares method, and the coefficients were determined.

By measuring the aggregation start temperature (Tagg) and diffusion interaction parameter (kD) under various storage conditions and substituting each of these values into the above equation, the rate of monomer decrease (ie, coagulation) in each condition. Aggregation formation rate) can be determined.

For the samples under the pharmaceutical conditions shown in Table 5 above, the aggregation start temperature (Tagg) and the diffusion interaction parameter (kD) were obtained (Table 9), and the sample ranking based on the results of calculating the monomer reduction rate was determined. The order of decreasing monomer reduction rate was as follows.
(Low) 15, 17, 19, 12, 13, 16, 11, 14, 41, 42 (High)

Compared with the results of the overall ranking in Table 8 above, a relatively similar tendency was observed. If necessary, by considering the results of the monomer reduction rate determined from the above equation and the results of, for example, A350 (4 ° C. and 40 ° C.) shown in the radar diagrams (FIGS. 7 and 8). , It is also possible to modify the overall ranking.

INDUSTRIAL APPLICABILITY According to the present invention, the storage stability (particularly aggregate formation) of a liquid preparation containing a protein such as an antibody drug is determined, and the aggregation initiation temperature (Tagg) and the diffusion interaction parameter (kD) of the protein in the preparation candidate sample are determined. It is possible to make an accurate and short-time prediction with just one.

Claims (4)

A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, which contains 0.05 to 30% by weight of the antibody, has a pH of 5.0 , and contains 3 to 6 % by weight of sorbitol or sucrose . A liquid preparation comprising %, containing 15 to 50 mM acetate buffer, containing 0.01 to 0.015% by weight of polysorbate, and containing no sodium chloride. A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, comprising 0.05 to 30% by weight of the antibody, pH 5.0, and 6% by weight of chloride. A liquid formulation comprising 20 mM acetate buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride. A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, containing 0.05 to 30% by weight of the antibody, pH 5.0, containing 6% by weight of sorbitol, 20 mM. A liquid preparation comprising an acetate buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride. A storage-stable liquid preparation containing the antibody TKM-011 or BM-ca against human CD20, containing 0.05 to 30% by weight of the antibody, pH 5.0, containing 3% by weight of sorbitol, 20 mM. A liquid preparation comprising an acetate buffer, containing 0.01% by weight of polysorbate, and containing no sodium chloride.

Images