Method for producing low viscous and highly concentrated biopharmaceutical drug products in liquid formulation
11510871 · 2022-11-29
Assignee
Inventors
- Martin Scholz (Munich, DE)
- Kristina Kemter (Garching bei Munich, DE)
- Jens Altrichter (Kavelstorf, DE)
- Thomas Kriehuber (Garching bei Munich, DE)
Cpc classification
A61K9/0019
HUMAN NECESSITIES
International classification
A61K47/18
HUMAN NECESSITIES
A61K47/26
HUMAN NECESSITIES
A61K39/395
HUMAN NECESSITIES
Abstract
The present invention relates to a method of producing low viscous and highly concentrated biopharmaceutical drug products comprising a biomolecule of interest, the method comprising: (a) a first phase of preparing a drug substance of the biomolecule of interest, said first phase comprising at least one processing step selected from (a1) harvesting, (a2) purification, (a3) re-buffering, and (a4) enrichment, wherein said at least one processing step in this first phase is carried out in the presence of a composition comprising at least three amino acids, wherein the combination of said at least three amino acids provides at least one positively charged functional group, at least one anti-oxidative functional group, at least one osmolytic function, and at least one buffering function, and (b) a second phase of further processing the drug substance prepared in (a) to obtain a low viscous and highly concentrated biopharmaceutical drug product, said second phase comprising at least one processing step selected from (b1) re-buffering, (b2) freezing, (b3) thawing, and (b4) filling; wherein said at least one processing step in this second phase is carried out in the presence of a composition comprising (i) at least three amino acids, wherein the combination of said at least three amino acids provides at least one positively charged functional group, at least one anti-oxidative functional group, at least one osmolytic function, and at least one buffering function; and (ii) one or more sugar(s); in an amino acid:sugar ratio between 10:1 to 1:100 (w/w). The present invention further relates to a low viscous and highly concentrated biopharmaceutical drug product obtained or obtainable by the method of the invention.
Claims
1. A method of producing low viscous and highly concentrated biopharmaceutical drug products comprising an antibody, the method comprising: (a) a first phase of preparing a drug substance comprising an antibody, said first phase comprising at least one processing step of concentration, wherein said at least one processing step in this first phase is carried out in the presence of a composition comprising at least three amino acids, wherein the combination of said at least three amino acids provides at least one positively charged functional group, at least one anti-oxidative functional group, at least one osmolytic function, and at least one buffering function; and (b) a second phase of further processing the drug substance prepared in (a) to obtain a low viscous and highly concentrated biopharmaceutical drug product, said second phase comprising at least one processing step selected from (b1) re-buffering, and (b2) filling, wherein said at least one processing step in this second phase is carried out in the presence of a composition comprising (i) at least three amino acids, wherein the combination of said at least three amino acids provides at least one positively charged functional group, at least one anti-oxidative functional group, at least one osmolytic function, and at least one buffering function, and (ii) one or more sugar(s); in an amino acid:sugar ratio between 10:1 to 1:100 (w/w); and further comprising a step of storing the biopharmaceutical drug product at a concentration ranging from 100 to 500 mg/ml obtained in (b) in liquid formulation for at least 1.5 days.
2. The method of claim 1, wherein the low viscous and highly concentrated biopharmaceutical drug product obtained in (b) is further processed for administration as a liquid formulation.
3. The method of claim 2, wherein the liquid formulation comprises (i) at least three amino acids, wherein the combination of said at least three amino acids provides at least one positively charged functional group, at least one anti-oxidative functional group, at least one osmolytic function, and at least one buffering function, and (ii) one or more sugar(s); and wherein the ratio between the amino acids and the sugar is adjusted to be between 4:1 to 1:1 (w/w).
4. The method of claim 3, wherein the liquid formulation is further adjusted such that the ratio between the antibody and the at least three amino acids of (i) is between 3.5:1 to 1:2 (w/w).
5. The method of claim 1 wherein the composition in step (a) contains between 0.5 mg/ml and 10 mg/ml tryptophan and between 0.5 mg/ml and 30 mg/ml histidine.
6. The method of claim 1, wherein the antibody is a therapeutic antibody.
7. The method of claim 1, wherein the antibody is a diagnostic antibody.
8. The method of claim 1, wherein the antibody is an antibody for experimental purposes.
9. The method of claim 3, wherein the ratio between the amino acids and the sugar is adjusted to between 2.5:1 and 1:1.
10. The method of claim 5, wherein the ratio between the antibody and sum of excipients is adjusted to be between 1:1 and 1:500.
11. The method of claim 4, wherein the ratio between the amino acids and the sugar is adjusted to between 2.5:1 and 1:1.
12. The method of claim 5, wherein the ratio between the amino acids and the sugar is adjusted to between 4:1 and 1:1.
Description
(1) The figures show:
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21) The examples illustrate the invention.
EXAMPLE 1
(22) The in vitro study of the functional and structural integrity of freeze dried and subsequently stored highly concentrated adenoviral vectors, as a predictive model for liquid storage-associated loss of molecular integrity, showed that compositions based on amino acids and sugar, stabilize viral vectors.
(23) 1.1 Materials and Methods
(24) Composition 1 and 2 contained the 7 amino acids alanine, arginine, glycine, glutamic acid, lysine, histidine and tryptophan in a concentration corresponding to the sum of the amino acids of 40 g/l. But in composition 1, a 5 fold increase of the tryptophan concentration and a 1.667 fold increase of the histidine and glutamic acid concentration under reduction of the concentrations of the other amino acids arginine, glycine, lysine and the retention of the alanine concentration compared to composition 2 resulted in the same concentration according to the sum of amino acids of 40 g/l. Further, an additional surfactant polysorbate 80 in a concentration of 0.05 g/l was added to composition 1 in contrast to composition 2. Both compositions contained trehalose as the corresponding sugar in an amino acid to trehalose ratio of 1:2. The pH value was adjusted in all compositions to 7.
(25) An adenoviral stock solution stored at −80° C. with a concentration of 7.5*10.sup.10 IFU/ml in the original supplier formulation (Firma Sirion; Martinsried/Munich; Germany) was employed.
(26) 1.1.1 Sample Preparation and Freeze Drying
(27) The adenoviral vector stock solution was re-buffered by dilution of the stock solution to a concentration of 1*10.sup.8 IFU/ml with either composition 1 or composition 2. For comparison the stock solution was diluted with either the original supplier formulation or with PBS to the same concentrations.
(28) In order to prepare the samples for freeze drying, the different adenoviral formulations were aliquoted in volumes of 500 μl in 2R freeze drying vials (Schott AG; Mainz; Germany) and subsequently freeze-dried using the following drying parameters:
(29) TABLE-US-00001 Protocol Step Target T (° C.) Slope (h) Hold (h) Pressure (mbar) Introduction 20 0 0 1000 Freezing −50 2:00 2:00 1000 Sublimation −50 0:01 0:30 0.045 −35 3:00 30:00 0.045 Secondary 20 3:00 7:00 0.009 Drying
(30) After freeze drying, the samples were visually inspected and one part of the samples was stored for a short time at 2-8° C. until analysis of the initial infective titer at the time point t=0.
(31) The other part of the samples was stored according to the guidelines of the International Council for Harmonization (ICH) for 21 or 42 days at 25° C. under environmental conditions of 60% residual humidity, or for 7 or 28 days at 40° C. under environmental conditions of 75% residual humidity.
(32) 1.1.2 Determination of the Infective Titers for Adenoviral Vectors in Cell Culture
(33) In order to analyze the infective titer of the adenoviral vector formulations, an antibody based virus titration experiment in HEK 293 cell culture using the detection of the adenoviral Hexon protein after successful amplification of the adenovirus in the infected cells was applied. 2.5*10.sup.5 HEK 293 (CCS) cells (Firma Sirion; Martinsried/Munich; Germany) were seeded per well of a 24-well micro titer plate in a volume of 500 μl. The adenoviral vector formulations were reconstituted either directly after freeze drying or at the indicated time points upon storage at 25° C. and at 40° C. As a positive control an aliquot of the adenoviral stock solution stored at −80° C. with a concentration of 7.5*10.sup.10 IFU/ml in the original supplier formulation (Firma Sirion; Martinsried/Munich; Germany) was used. Subsequently, serial dilutions of the adenoviral samples were prepared and 50 μl of the resulting dilutions per well were used for infection of the cells. The plates were incubated for 42 hours at 37° C. After infection, cells were fixed with methanol, incubated with the primary anti-Hexon protein antibody (Santa Cruz Biotechnology, Inc.; Dallas; Texas: USA), subsequently incubated with an horse radish peroxidase (HRP)-conjugated secondary anti-mouse antibody (Cell Signaling Technology; Danvers; Massachusetts; USA) specific for the primary antibody and an HRP enzymatic reaction with diaminobenzidine (Carl Roth GmbH and Co. KG; Grafrath; Germany) was carried out, wherein a brown colouring indicates infected cells. The number of infected cells was quantified by counting the brown coloured cells under the microscope, wherein each infected cell is counted as one infectious viral particle.
(34) 1.2 Results
(35) The in vitro infectivity assay revealed that a formulation of adenoviral vector preparations in the stabilizing compositions 1 and 2 early in the production process of a freeze dried biopharmaceutical product resulted in infective titers that correspond to those of the positive control depicted as dashed line in
(36) These differences were even more striking after storage of the freeze-dried preparations. A complete loss of function of the viral vectors freeze-dried in the original supplier formulation (
EXAMPLE 2
(37) The in vitro study of the functional and structural integrity of different adenoviral vector preparations after freeze and thaw stress as a predictive model for stress-associated loss of molecular integrity during processing showed that compositions based on amino acids and sugar, stabilize viral vectors even during freeze and thaw cycles
(38) 2.1 Materials and Methods
(39) 2.1.1 Sample Preparation and Further Processing
(40) High titers of adenoviral vector stocks of the adenoviral type 5 vectors containing the coding DNA for the eGFP protein 5*10.sup.8 HEK293 cells were transduced with adenoviral particles. 48 h after transduction, the cells were harvested and the release of viral particles was performed via Na-Deoxycholat and DNase I treatment. Viral particles were purified by CsCl gradient ultracentrifugation usually followed by buffer exchange in the original supplier formulation on PD10 columns and subsequent determination of the infective titer. The resulting high titer adenoviral stocks were subsequently aliquoted and stored at −80° C.
(41) Sample preparation—process step 1: Adenoviral vector formulations were prepared by re-buffering of the adenoviral vector preparations immediately after CsCl gradient ultracentrifugation. The obtained adenoviral vector band was harvested and dialysed at 2-8° C. in either composition 1 or 2 (as described in 1.1). The resulting formulations were aliquoted and stored at −80° C.
(42) Sample preparation—process step 2: Frozen (−80° C.) adenoviral stock solutions (7.5*10.sup.10 IFU/ml; Sirion, Martinsried/Munich, Germany) were thawed (room temperature; RT) in the original supplier buffer and subsequently dialysed at 2-8° C. in compositions 1 and 2.
(43) 2.1.2 Repeated Freeze and Thaw Cycles with Adenoviral Samples from Process Step 1 and Step 2 Preparations
(44) In order to analyze the stability of the adenoviral vector preparations during subsequent stress conditions, 50 μl of the adenoviral vectors, formulated in composition 1 or 2 were subjected to repeated freeze (−80° C.) and thaw (RT) cycles. The in vitro infectivity (described in 1.1.2) was determined at the initial time point t=0 and after 5 and 10 freeze thaw cycles by virus titration in HEK 293 cell cultures (described in 1.1.2). In parallel, the hydrodynamic radii of the adenoviral particles were measured by DLS.
(45) 2.1.3 Dynamic Light Scattering (DLS) Measurement
(46) DLS was carried out on samples taken before freeze drying directly after re-buffering compared to an untreated positive control corresponding to an aliquot of the adenoviral stock solution stored at −80° C. as well as on samples after reconstitution of the adenoviral vector formulations. In the latter case, DLS was carried out either immediately after freeze drying (t=0) or at the relevant time points upon storage at 25° C. (21 days, 42 days) and at 40° C. (7 days, 28 days).
(47) To this end, 5 μl of the samples were pipetted into a special DLS cuvette and analysed in a DynaPro Nanostar DLS instrument (Wyatt Technology Europe GmbH; Dernbach; Germany). For each experimental formulation, a blank measurement was performed under the same conditions. The DLS measurements were performed with acquisition times between 20 and 40 seconds in 10 or 20 cycles. The resulting correlation curves were analysed using the DynaPro DLS software.
(48) 2.2 Results
(49) The in vitro infectivity assay revealed that composition 1 fully retained the infective titers of both adenoviral vector preparations from process step 1 and step 2 (
(50) Upon additional freeze and thaw cycles (five and ten), composition 1 retained the full infective titer, regardless of the production process step and time point of re-buffering (FIGS. 4 A and B). In contrast, composition 2 resulted in remarkably different effects when prepared in the two different process steps 1 and 2. The infective titers of composition 2 samples obtained according to process step 2 significantly further decreased after five and even stronger after ten freeze and thaw cycles (
(51) In parallel to the determination of the infective titers before and after repeated freeze and thaw cycles, the hydrodynamic radii of the corresponding adenoviral particles were analyzed using Dynamic Light Scattering (DLS) (
(52) In summary and conclusion, composition 1 generally exhibited excellent stabilizing efficacy for the adenoviral vector particles during both applied early production steps. In contrast, although composition 2 showed stabilizing efficacy when used directly after ultracentrifugation, reduced stabilizing efficacy was observed when used later in the production process as compared to composition 1.
(53) The DLS data correlate with the in vitro infectivity data. This leads to the conclusion that the use of specifically tailored stabilizing compositions based on amino acids early in the production process of viral vector compositions is important for the stability during further processing steps in biopharmaceutical manufacturing.
EXAMPLE 3
(54) The analysis of the molecular integrity and viscosity of highly concentrated therapeutic antibody formulations during processing and liquid storage showed that specific amino acid and sugar compositions reduced the propensity for aggregation and particularly fragmentation of antibodies in a model for drug substance to drug product processing.
(55) 3.1 Materials and Methods
(56) Compositions 3 and 4 contained the 4 basic amino acids arginine, glycine, tryptophan and histidine in a concentration according to the sum of the amino acids to 50 g/l. In the case of composition 3 the amino acid composition was in combination with 80 g/l trehalose and in the case of composition 4 in combination with 32.2 g/l trehalose. As additional compounds the compositions 3 and 4 contained 1.5 g/l methionine and 0.4 g/l polysorbat 20. Composition 4 contained two additional compounds, a chelating agent EDTA and an antioxidant ascorbic acid. The resulting sum of excipients was 131.5 g/l in the case of composition 3 and 85 g/l in the case of composition 4. The ratio of the sum of basic amino acids to trehalose was in composition 3 1:1.55 (w/w) with trehalose in excess whereas in composition 4 the ratio of the basic amino acids to trehalose was 1.6:1 (w/w) with the amino acids in excess. The ratio of the antibody to the sum of excipients was 1:0.9 (w/w) in composition 3 and 1:1.4 (w/w) in composition 4. The pH value was adjusted to 5.5.
(57) As a model protein, the commercially available liquid therapeutic highly concentrated antibody Herceptin® (Roche; Basel; Switzerland) containing trastuzumab in a concentration of 120 mg/ml in the original supplier formulation (79.45 g/l trehalose; 3.13 g/l histidine buffer 20 mM; 1.49 g/l methionine; 0.4 g/l polysorbat 20; 0.024 g/l rHuPh20 (recombinant human hyaluronidase), pH 5.5) was used.
(58) 3.1.1 Sample Preparation
(59) The samples of the untreated antibody formulation in the original liquid supplier formulation were directly aliquoted in sterile HPLC vials from the original container for storage at 25° C., 30° C. and 40° C. Another part of the original liquid supplier formulation of trastuzumab at an antibody concentration of 120 mg/ml was re-buffered using dialysis at 2-8° C. into the compositions according to the invention. The resulting formulations were sterile filtrated, aliquoted in sterile HPLC vials and stored at 25° C., 30° C. and 40° C. The aggregation and fragmentation before storage, directly after sample preparation, and at indicated time points during storage were analyzed using SEC.
(60) 3.1.2 Size Exclusion Chromatography (SEC)
(61) 3.1.2 Size Exclusion Chromatography (SEC)
(62) Protein aggregation and fragmentation were quantified by SEC. Analytics were performed on an UHPLC system UltiMate3000 (Thermo Scientific; Darmstadt; Germany) equipped with a UV-280 nm detector and a TSK-gel G3000SW.sub.XL 7.8×300 mm column (Tosoh Bioscience, Tokyo, Japan) at 30° C. and with a flow rate of 0.5 ml/min. Prior to the SEC analysis, the samples containing 25 mg/ml antibody or higher concentrations according to the other examples were diluted to reach a concentration of 2.5 mg/ml IgG using the SEC running buffer PBS and aliquoted into special HPLC vials. The injection volume was 25 μl. The running buffer for SEC was Dulbecco's PBS pH 7.1 (PAA Laboratories, Pasching, Austria). Molecular weight standards (BSA, Thermo Scientific; Waltham, Mass., USA) and a placebo buffer were run in each sequence. Quantification of aggregation and fragmentation in % was determined by comparing the area under the curves of the monomer peaks, the sum of the high molecular weight species and the sum of the low molecular weight species using the Chromeleon 7 Chromatography Data Software (Thermo Scientific, Germany).
(63) 3.1.3 Viscosimetry
(64) After sample preparation according to paragraph 5.1.1 the viscosities of the highly concentrated antibody formulations based on amino acid compositions 3 and 4 compared to the viscosity of the untreated liquid original supplier formulation were determined using a falling ball viscosimeter (Anton Paar GmbH; Ostfildern-Schamhausen; Germany). After determination of the density of a highly concentrated protein sample (120 mg/ml) and the calibration of the capillary with water at 20° C. using the falling angle of 70°, the ball was introduced into the capillary and approximately 500 μl of the antibody formulations were carefully filled into the capillary. The filled capillary was inserted into the capillary block of the instrument and samples were measured as ten separate assays at 20° C. and a falling angle of 70°.
(65) 3.2 Results
(66) Sample Preparation
(67) The SEC profile of the untreated liquid trastuzumab formulation from the original container showed only a small aggregate peak with 0.19%, a monomer peak with 99.77% and 0.04% fragments. After re-buffering of this formulation in the amino acid based compositions 3 and 4 according to paragraph 3.1 comparable SEC profiles were analyzed, suggesting a stabilizing effect of the amino acid based formulations on the antibody during the process of re-buffering (
(68) Liquid Storage
(69) Already after 1.5 days storage at 40° C. in the original untreated liquid supplier formulation, an increase in aggregate formation was determined (0.22%) and a slight increase in fragmentation was found (0.09%). The monomer peak was slightly decreased to 99.69%. In contrast, the storage for 1.5 days at 40° C. of the antibody in two different amino acid based compositions 3 and 4 according to paragraph 3.1 revealed a decreased propensity of the antibody for aggregation and to a minor extent for fragmentation (
(70) These results suggest that the amino acid based formulations have a stronger stabilizing efficacy than the original formulation during storage at 40° C. and 30° C. against both the formation of aggregates and particularly in composition 4 the formation of fragments. The results further indicate that particularly composition 4, with amino acids in the excess over trehalose showed better stabilization against aggregation and fragmentation compared to composition 3. This observation was confirmed after storage of 3 months at 30° C. In the original formulation the aggregate content was 0.35% whereas in composition 3 the aggregate content was 0.26% and in composition 4 the aggregate content was 0.20%. The fragmentation in the original formulation was 0.40%, in composition 3, 0.46% and in composition 4 only 0.33%. Quantitative statistical analysis of the course of liquid storage of highly concentrated antibody formulations (120 mg/mL) at accelerated aging conditions for 3 months at 30° C. further substantiated the above findings. Accelerated aging (
(71) The stabilizing efficacy of composition 4 with a ratio of amino acids to trehalose of 1.6:1 (w/w) was confirmed by the following examples.
(72) In addition, analysis of the chemical degradation pattern upon the course of liquid storage of highly concentrated antibody formulations (120 mg/mL) at accelerated aging conditions for 3 months at 30° C. in composition 3 and 4 compared to the original formulation using CEX-HPLC further highlighted the advantageous effect of an adjustment of the base amino acid compositions e.g. by addition of sugar in an appropriate ratio and one or more antioxidant. As with higher concentrated (120 mg/mL) trastuzumab, composition 3 led to a reduction in basic species (
(73) This data further substantiate the claimed invention that the adjustment of the applied basic amino acid composition comprised of the at least three amino acids arginine, glycine, histidine and/or tryptophan used during the early drug substance processing steps according to the requirements of the specific biomolecule (e.g. final concentration and viscosity of the drug product) stabilizes the biomolecule during further processing such as filling, freeze drying, storage of the dried or the liquid product.
(74) Viscosity Measurements
(75) The measured dynamic viscosities in the highly concentrated antibody formulations based on amino acids were found to be remarkably reduced compared to the corresponding viscosity of the untreated liquid original supplier formulation. The dynamic viscosity in composition 3 was 4 mPa*s and in composition 4 was 3.5 mPa*s. In contrast, the dynamic viscosity in the highly concentrated liquid original supplier formulation was 4.8 mPa*s (
(76) Composition 4 and the original supplier formulation contained the antibody in an approximately comparable antibody to excipient ratio of 1.4:1. (w/w) But, in composition 4 the adjustment of the amino acid to trehalose ratio and concomitant of the corresponding antibody to excipient ratio resulted in an impact on both the stabilizing efficacy during liquid storage at elevated temperatures and in a remarkable decrease of the viscosity of the formulation compared to the original formulation. Already the adjustment of the amino acid to trehalose ratio in composition 3 and the resulting antibody to excipient ratio resulted in an increased stabilizing efficacy and in a decrease in the formulation viscosity compared to the original formulation but to a minor extent in comparison to the further adjustments resulted in the effects of composition 4.
(77) Thus, these data further substantiate the finding that the combination of amino acids with trehalose in a balanced ratio and the simultaneous adjustment of the ratio antibody to the sum of excipients have a strong impact on the stabilizing efficacy of the formulation concerning aggregation and particularly of fragmentation of the antibody during liquid storage at elevated temperatures. Moreover, beside the above mentioned stabilizing efficacy the adjustment of the compositions in this manner results in a significant decreased formulation viscosity of highly concentrated therapeutic antibody formulations.
EXAMPLE 4
(78) The analysis of the molecular integrity of highly concentrated therapeutic antibody formulations during processing and liquid storage showed that specific amino acid and sugar compositions as well as amino acid to sugar ratios (w/w) reduced the propensity for aggregation and particularly fragmentation of antibodies in a model for drug product stability.
(79) 4.1 Materials and Methods
(80) Composition 3 and 4_1 are similar formulations applied in Example 5 according to paragraph 5.1. But in the case of composition 4_1 the pH adjustment to pH 5.5 was performed using HCl instead of citric acid. Both compositions contained the similar ratios of the sum of amino acids to trehalose according to paragraph 5.1 in Example 5. Composition 4_2 was also a variant of composition 4 according to paragraph 5.1 of Example 5. Composition 4_2 contained the 4 basic amino acids arginine, glycine, tryptophan and histidine under addition of an additional amino acid alanine. In composition 4_2 the sugar fraction was a mixture of trehalose and saccharose in a ratio of 3:1 (w/w). The amount of methionine was slightly increased to 3.5 g/l and addition excipients, e.g. a chelating agent EDTA and ascorbic acid were further supplied.
(81) The ratio of amino acids to sugar was slightly reduced in composition 4_2 to 1:1 (w/w). In the original formulation, the antibody to excipient ratio was 1.6:1 (w/w), in composition 3, 1:1.1 (w/w); in composition 4_1, 1.76:1 (w/w) and in composition 4_2, 1.12:1 (w/w) The pH was adjusted to 5.5.
(82) As a model protein, the commercially available freeze-dried Herceptin® (Roche; Basel; Switzerland) was used. Preparation of the samples was performed by reconstitution of the commercially available freeze-dried Herceptin® in a desired volume of water.
(83) 4.1.1 Sample Preparation
(84) The resulting formulation was dialysed at 2-8° C. against the composition of the original liquid formulation (79.45 g/l trehalose; 3.13 g/l histidine buffer (20 mM); 1.49 g/l methionine; 0.4 g/l polysorbat 20; pH 5.5) and against the amino acid based compositions according to paragraph 6.1. Subsequent concentration of the resulting IgG formulations were done to obtain 135 mg/ml antibody in the original formulation, 145 mg/ml antibody in composition 3, 150 mg/ml antibody in composition 4_1 and 151 mg/ml antibody in composition 4_2. Subsequently, the formulations were sterile filtrated, aliquoted in sterile HPLC vials and stored at 5° C., 25° C., 30° C. and 40° C. The aggregation and fragmentation before storage, directly after sample preparation, and at indicated time points during storage were analyzed using SEC.
(85) 4.1.2 Size Exclusion Chromatography
(86) SEC was performed according to paragraph 3.1.2.
(87) 4.2 Results
(88) Liquid Storage
(89) Already after the initial storage time of 8 days at 40° C. in the original formulation, the formation of aggregates and fragments was remarkably increased (0.77% aggregates and 0.75% fragments versus 0.35% aggregates and no fragments before liquid storage, respectively). In contrast, storage for 8 days at 40° C. in all amino acid based formulations tested clearly limited aggregation and fragmentation. In composition 3, aggregation was 0.39% and fragmentation was 0.51%. In composition 4_1, aggregation was 0.38% and interestingly, fragmentation was further reduced to 0.31%. In composition 4_2, a slightly further reduction of aggregation and fragmentation was detected (0.36% aggregates and 0.28% fragments) as depicted in
(90) This data confirmed the results of the previous experiment concerning the efficacy of composition 4 (composition 4_1 in this Example 4) to further reduce particularly the formation of fragments during liquid storage.
(91) Comparable results were found after storage for 1 month at 30° C. Storage in the original formulation led to 0.53% aggregates and 0.50% fragments. The corresponding SEC analysis for composition 3 revealed 0.4% aggregates and 0.51% fragments. Liquid Storage for 1 month at 30° C. in composition 4_1 resulted in remarkably decreased aggregate formation (0.35%) and fragment formation (0.30%).
(92) Comparable results were shown in composition 4_2 with an aggregate formation of 0.32% and fragment formation of 0.31% (
(93) After liquid storage for 6 months at 25° C., aggregation of the antibody in the original formulation was increased to 0.98% and fragmentation reached 1.00%. In composition 3, a smaller increase in aggregation to 0.71% and in fragmentation to 0.9% was shown. In both compositions, 4_1 and 4_2, only nearly the half of aggregation and fragmentation compared to the original formulation was found; composition 4_1: 0.58% aggregates and 0.53% fragments and composition 4_2: 0.58% aggregates and 0.56% fragments (
(94) Comparable results were found after liquid storage for 6 months at 2-8° C. with smaller changes in aggregation and fragmentation compared to the storage at 25° C. (
(95) Quantitative statistical analysis of the whole course of storage of highly concentrated antibody formulations (150 mg/mL) for 6 months at 25° C. in composition 4_1 and 4_2 compared to the original formulation revealed significantly reduced aggregates and fragments (p<0.01) and retained a stable monomer peak during storage for 6 months at 25° C. (
(96) The additional analysis of the chemical degradation profile of the highly concentrated antibody formulations during storage for six months at 25° C. in composition 4_1 and 4_2 compared to the original formulations using CEX HPLC underlined the previous SEC results and the results of the previous examples. After six months storage of 150 mg/mL trastuzumab at 25° C., lower amounts of basic species were observed for composition 4_1 and 4_2 (p<0.05, p<0.01) compared to the original formulation (
(97) In summary, these results confirm that the balanced mixture of amino acids and sugar is necessary for the prevention of antibody aggregation and fragmentation during liquid storage and that the adjustment of the ratio of amino acids to sugar at least to amino acids in excess and more preferred amino acids and sugar in the ratio approx. 1:1 (w/w) resulted in a further increased stabilizing efficacy.
EXAMPLE 5
(98) The analysis of the molecular integrity and viscosity of highly concentrated therapeutic antibody formulations during processing and liquid storage showed that specific amino acid and sugar compositions as well as amino acid to sugar ratios (w/w) as well as the adjustment of tryptophan and histidine concentrations and ratios (w/w) reduced the propensity for aggregation and particularly fragmentation of antibodies in a model for drug product stability.
(99) 5.1 Materials and Methods
(100) Compositions 4_3, 4_4 and 4_5 were obtained by concentrating composition 4_2 from the previous example. Composition 4_3 is similar to composition 4_2 according to paragraph 4.1 in Example 4, but the sum of excipients was reduced from 135 g/l to 90 g/l in composition 4_3. The amino acid to sugar mixture ratio was preserved to 1:1 (w/w) and the antibody to excipient ratio was 2.22:1 (w/w). In composition 4_4 the similar mixture of amino acids and sugar mixture was used compared to composition 4_2 in paragraph 4.1 in Example 4 and to composition 4_3 in this Example 5, but the amino acids to sugar ratio was increased to 3.4:1 (w/w) and the antibody to excipient ratio was increased to 3.33:1 (w/w). In composition 4_5 also the same mixture of excipients was used, but the amino acid concentration of histidine was increased and the concentration of tryptophan was decreased. The ratio trehalose to saccharose was reduced to 2:1 (w/w) compared to 3:1 (w/w) in the previous experiments, and the amino acid to sugar ratio was reduced to 1.5:1 (w/w). The antibody to excipient ratio was comparable to composition 4_3 adjusted to 2.22:1 (w/w). The antibody to excipient ratio was 2.4:1 (w/w) in the case of the original liquid suppler formulation. The pH value was adjusted to 5.5.
(101) As a model protein, the commercially available liquid therapeutic highly concentrated antibody Herceptin® (Roche; Basel; Switzerland) according to paragraph 5.1 in Example 5 was used.
(102) 5.1.1 Sample Preparation
(103) In order to get higher concentrated trastuzumab preparations compared to the Examples 3 and 4, the antibody in the original liquid supplier formulation was concentrated to obtain 200 mg/ml and the antibody in compositions 4_3, 4_4 and 4_5 were concentrated after an additional dialysis step at 2-8° C. Highly concentrated formulations were prepared for the subsequent storage experiment by sterile filtration and subsequent aliquoting in sterile HPLC vials. The highly concentrated samples were stored at 5° C., 25° C., 30° C. and 40° C. The aggregation and fragmentation were analyzed before storage, directly after sample preparation, and at indicated time points during storage using SEC.
(104) 5.1.2 Size Exclusion Chromatography
(105) SEC was performed according to paragraph 3.1.2.
(106) 5.1.3 Measurements of Viscosities of the Highly Concentrated Antibody Formulations
(107) The viscosities of the highly concentrated antibody formulations according to this example were measured using a falling ball viscosimeter according to paragraph 3.1.3 in example 3.
(108) 5.2 Results
(109) Sample Preparation
(110) The SEC profile of the untreated liquid trastuzumab formulation from the original container showed only a small aggregate peak with 0.19%, a monomer peak with 99.77% and 0.04% fragments (example 3;
(111) Liquid Storage
(112) Interestingly, during the whole course of liquid storage of these particular highly concentrated formulations at different temperatures the fragmentation of the antibody was only a minor event in the original formulation as well as in the amino acids based compositions according to the invention. This effect might be a result of the increased antibody to excipient ratios and was already observed in a previous experiment with low concentrated trastuzumab formulations.
(113) Liquid storage for 3 days at 40° C. resulted in increased aggregation in the original formulation to 2.11% aggregates and 0.1% fragments. In composition 4_3 this storage period resulted in an aggregation of about 0.22% and fragmentation of about 0.07%. Storage for 3 days at 40° C. in composition 4_4 resulted in a formation of aggregates to about 0.26% and fragment formation of about 0.06%. In composition 4_5, the aggregate content was only 0.18% and the fragmentation was analyzed to about 0.08% (
(114) Further liquid storage for 14 days at 40° C. resulted in an aggregate content of about 2.28% in the original formulation and 0.33% fragmentation. In the composition 4_3 the aggregation was only 0.43% and the fragmentation was similar to 0.33%.
(115) Comparable results were obtained in composition 4_4 with aggregation of about 0.40% and fragmentation about 0.33%. Also in composition 4_5, comparable results were found for aggregate formation (0.27%) and fragment formation at 0.34% (
(116) After long term liquid storage for 1½ months at 30° C. in original formulation the aggregate formation was 1.84% and fragment formation 0.25%. In composition 4_3, the aggregation was only 0.28% and fragmentation 0.23% comparable to the original formulation. In composition 4_4, the aggregate formation was slightly increased to 0.38% and the fragment formation was about 0.22%. In composition 4_5, the aggregation after storage for 1½ months at 30° C. was only 0.22% and the fragmentation reached 0.25% (
(117) Real time liquid storage for 3 months at 25° C. resulted in an aggregate formation of about 1.89% in the original formulation and 0.25% fragmentation. In composition 4_3, the long term storage at 25° C. resulted in 0.35% aggregates and 0.22% fragments. The aggregation was slightly increased in composition 4_4 after 3 months storage at 25° C. to 0.40% and the fragment formation was retained at 0.22%. In composition 4_5, the lowest aggregation was found with 0.27% and a comparable fragmentation with 0.25% was achieved (
(118) At all analytic time points during liquid storage of the antibody at different temperatures composition 4_5 showed the best stabilizing efficacy against aggregation. In composition 4_3 and composition 4_4, the aggregation propensity of the antibody during liquid storage at different temperatures was more or less comparable whereas in composition 4_5 the antibody showed the lowest aggregate formation at the indicated analytic time points. Between composition 4_3 and composition 4_4 the former showed slightly superior stabilizing efficacy.
(119) Evaluation of the propensity of the antibody for aggregation and fragmentation and the associated loss of monomer peak during liquid storage of highly concentrated antibody formulations (200 mg/mL) in compositions 4_3, 4_4 and 4_5 compared to the original formulation for 3 months at 25° C. further confirmed the above detailed results.
(120) Aggregate peaks (elution time≈14 minutes) corresponding to antibody dimers were significantly reduced in composition 4_5 (p<0.01; p<0.0001), and to a minor extent in composition 4_3 and composition 4_4 (p<0.01, p<0.001) compared to the original formulation (
(121) These results suggest that both changes in the composition, the concentration changes of histidine and tryptophan and the change of the ratio of trehalose to saccharose had a positive impact on the stabilizing efficacy particularly on the aggregation. Fragmentation was slightly more reduced in the amino acid based compositions containing higher concentrations of tryptophan and histidine in the buffer concentration (see previous examples and composition 4_3 and 4_4 compared to composition 4_5). The ratio amino acids:trehalose/saccharose mixture was slightly increased in composition 4_4 (1.5:1 (w/w)) compared to 1:1 (w/w) in composition 4_3 derived from composition 4_2 of the previous experiment.
(122) Thus, the adjustment of the concentrations of tryptophan and histidine in line with the adjustment of the amino acid to sugar ratio is important for preventing aggregation and fragmentation in conjunction during liquid storage of an antibody. Furthermore, the strong increase of the antibody to excipient ratio to 3.3:1 (w/w) in composition 4_4 led to a slight increase in the aggregation propensity of the antibody during liquid storage. The parallel adjustment of the antibody to excipient ratio was shown to have an impact on the stabilizing efficacy of the amino acid based compositions according to the invention.
(123) Similar observations were made by analyzing the chemical degradation of the antibody during liquid storage at elevated temperature. To this end, the samples were stored for 3 months at 25° C. and analyzed using CEX. Composition 4_4 and 4_5 resulted in a not significant increased formation of acidic charge variants (lowest degree in composition 4_5) but a reduced formation of basic charge variants particularly in the case of composition 4_3 and 4_4 (p<0.05) compared to the original formulation. The loss of the main peak area was partly prevented by composition 4_3 and 4_4 (p>0.05) and was comparable to the original formulation in composition 4_5 (
(124) Viscosity Measurements
(125) For the analysis of the dynamic viscosities of higher concentrated antibody formulations compared to example 5 the concentrations of the formulations were adjusted to 200 mg/ml and 220 mg/ml according to the sample preparation method in paragraph 7.1.1 of this example (
EXAMPLE 6
(126) The analysis of both the molecular integrity as well as the chemical stability during processing and during subsequent liquid storage of highly concentrated antibody formulations showed that specific amino acid and sugar compositions that do not comprise proline were able to reduce the propensity for aggregation during processing as well as during subsequent liquid storage. In particular chemical changes were remarkably reduced compared to the original trastuzumab formulation of the freeze dried product comprising glycine and proline.
(127) 6.1 Materials and Methods
(128) Composition 4_1 corresponds to the formulation applied in Example 3 and compositions 4_3 and 4_5 correspond to the formulations applied in Example 5. The stabilizing effect of these formulations was compared to the original liquid supplier formulation (79.45 g/l trehalose; 3.13 g/l histidine buffer 20 mM; 1.49 g/l methionine; 0.4 g/l polysorbat 20; pH 5.5). The pH was adjusted in these formulations to 5.5. In addition, the stabilizing effect of the inventive compositions was compared to the original supplier formulation of the freeze-dried product (20 g/l trehalose; 0.9 mg/ml histidine buffer approx. 5 mM; 0.1 g/l polysorbat 20; pH 6) with addition of the amino acids glycine and proline in accordance with the teaching of U.S. Pat. No. 9,364,542 B2 (Example 16;
(129) As a model drug, commercially available freeze-dried Herceptin® (Roche; Basel; Switzerland), a therapeutic humanized IgG1 monoclonal antibody (trastuzumab), was used. Reconstitution of the freeze-dried drug in the desired volume of water resulted in an antibody concentration of 21 mg/mL in the original supplier formulation (20 g/l trehalose; 0.9 mg/ml histidine buffer approximately 5 mM; 0.1 g/l polysorbat 20; pH 6).
(130) 6.1.1 Sample Preparation
(131) The resulting formulation was dialysed at 2-8° C. against the composition of the original liquid formulation (79.45 g/l trehalose; 3.13 g/l histidine buffer, approx. 20 mM; 1.49 g/l methionine; 0.4 g/l polysorbat 20; pH 5.5) and the original supplier formulation of the freeze-dried product (20 g/l trehalose; 0.9 mg/ml histidine buffer, approx. 5 mM; 0.1 g/l polysorbat 20; pH 6), wherein this formulation additionally contained the amino acids glycine and proline in the concentrations described in U.S. Pat. No. 9,364,542 B2 (Example 16).
(132) In parallel, dialysis was performed against the amino acid based compositions of the present invention as detailed in paragraph 8.1 Subsequent concentration of the resulting IgG formulations was performed in order to obtain 200 mg/ml antibody. Subsequently, the formulations were sterile filtrated, aliquoted in sterile HPLC vials and either stored at 25° C. or 40° C., or, for short term storage, at 55° C., as described in U.S. Pat. No. 9,364,542 B2 (Example 16;
(133) 6.1.2 Size Exclusion Chromatography
(134) SEC was performed as described in section 3.1.2 above.
(135) 6.1.3 Cation Exchange Chromatography
(136) CEX-HPLC (UV-280 nm detector; UHPLC UltiMate3000 Thermo Scientific, Germany) and a cation exchange column TSK-gel CM-STAT 4.5×100 nm (Tosoh Bioscience, Tokyo, Japan) was used at 45° C. and with a flow rate of 0.8 ml/min (injection volume 25 μl). Prior to the CEX-HPLC analysis, samples were diluted to 2.5 mg/mL IgG in running buffer A (10 mM sodium phosphate buffer pH 7.5). The immobilized trastuzumab molecules were eluted in a sodium chloride gradient using 0% to 30% buffer B (10 mM sodium phosphate buffer pH 7.5; 100 mM sodium chloride). Relative areas under the curves (% AUC) were determined with the Chromeleon 7 Chromatography Data Software (Thermo Scientific).
(137) 6.2 Results
(138) Sample Preparation
(139) During the course of sample preparation using dialysis and subsequent concentration in order to obtain an antibody concentration of 200 mg/mL, the previously obtained results of Examples 3 and 5 were confirmed. Specifically, when the preparation process was carried out in one of the inventive compositions (composition_4_1, composition_4_3 or composition_4_5), a retention of the amount of aggregates and monomers was obtained that is comparable to the trastuzumab standard (0.65-0.70% aggregates; 99.30-99.35% monomers). These results suggest that the retention of the structural integrity of the antibody prepared in the inventive compositions is comparable to the levels of structural intact antibodies before preparation (
(140) The process of preparing the highly concentrated antibody in the original liquid supplier formulation, on the other hand, led to an increase in the percent area of the peak corresponding to the antibody dimers (i.e. at a retention time of 14 min) to about 0.84% and a corresponding reduction of the monomer peak to about 99.16%. Moreover, the corresponding preparation process of the highly concentrated trastuzumab in the original supplier formulation of the freeze-dried product in combination with the additional amino acids glycine and proline also resulted in a strong increase in the formation of aggregates to about 0.79% and, consequently, in the reduction of the monomer peak to 99.21%.
(141) Liquid Storage—SE-HPLC
(142) Short term liquid storage of the antibody under extreme (i.e. physiologically and pharmaceutically irrelevant high) temperature conditions (24 h at 55° C.) as carried out in U.S. Pat. No. 9,364,542 B2 revealed an efficient stabilizing effect of the inventive compositions against aggregation and fragmentation compared to the original liquid supplier formulation as well as to the original supplier formulation of the freeze-dried product with glycine and proline as additives. The percent areas of the peaks corresponding to the formation of aggregates were nearly completely retained in the inventive compositions, particularly in composition_4_3 and composition 4_5, comparable to the trastuzumab standard stored at −80° C. (0.65-0.70% aggregates; 99.30-99.35% monomers).
(143) In contrast, short term storage for 24 h hat 55° C. in the original supplier formulation with the additives glycine and proline and, more pronounced, in the original liquid supplier formulation, resulted in a remarkably increased aggregation with a more pronounced decrease in the percent area of the monomer peak (
(144) Liquid storage for 7 days, 14 days, 28 and 42 days, respectively, at 40° C./75% RH revealed an increased propensity of the antibody for both aggregation as well as fragmentation in all formulations but to different extents. Most strikingly, formulation of the antibody in the inventive compositions (composition_4_1, composition_4_3 and composition_4_5) resulted in a remarkably reduced aggregation and fragmentation as compared to the original liquid supplier formulation as well as to the original supplier formulation in combination with the amino acids glycine and proline (
(145) Moreover, liquid storage for 14 days and 28 days at 25° C. further substantiated the observation that the inventive solutions are able to prevent aggregation and fragmentation upon the course of storage at ambient and elevated temperatures and even at extreme temperature conditions such as 55° C. (Table 3, 4).
(146) Liquid Storage—CEX-HPLC
(147) In the U.S. Pat. No. 9,364,542 B2 only the formation of macroscopic insoluble aggregates of the antibody using turbidity measurements and, in some examples, the analysis of the formation of soluble aggregates using SE-HPLC during short term storage under physiologically and pharmaceutically irrelevant high temperature conditions such as 55° C. was analyzed. Here, the extent of chemical changes in the antibody molecule during short term storage at 55° C., as well as during long term storage at 40° C. and 25° C., was additionally analyzed. Already short term storage of trastuzumab for 24 h at 55° C. triggered chemical changes in all formulations to varying degrees.
(148) In the original supplier formulation of the freeze-dried product with addition of the amino acids glycine and proline (Original+G/P), a significantly increased percentage of acidic charge variants of the antibody (>30%) was observed during short term storage at 55° C. This observation was associated, in consequence, with a remarkable decrease of the percent area of the main peak (53.23%) as shown in Table 5.
(149) In contrast, short term storage of the antibody in the compositions according to the invention resulted in a remarkably reduced formation of acidic charge variants (e.g. <23% in composition_4_5). The increase of acidic charge variants provides evidence for protein deamidation or glycation and is an important criterion for negative selection of test formulations in industrial manufacturing standards. Therefore, different compositions according to the invention were tested in comparison with Original+G/P to study the modifications of acidic charge variants during three day storage at 55° C., up to 21 days at 40° C., and up to two months at 25° C. (
EXAMPLE 7
(150) The analysis of the molecular integrity after dialysis, concentration, and subsequent storage of high concentrated antibody formulations (200 mg/mL) by means of SE-HPLC shows the relevance of the at least three amino acid combinations alone or in combination with trehalose to limit the increase of aggregates in manufacturing relevant processing steps and subsequent storage.
(151) 7.1 Materials and Methods
(152) Composition 5 contained the two base amino acids histidine and methionine and is similar to composition 8, which corresponds to the original supplier formulation (histidine, methionine, trehalose), except that it does not contain any sugar. Composition 6 comprises the three amino acids histidine, methionine, and glycine without sugar, whereas composition 9 is similar, but contains trehalose. Composition 7 comprises the five amino acids histidine, methionine, alanine, arginine and tryptophan, whereas composition 10 is similar, but contains trehalose.
(153) As a model protein, the commercially available liquid therapeutic highly concentrated antibody Herceptin® (Roche; Basel; Switzerland) containing trastuzumab in a concentration of 120 mg/ml in the original supplier formulation (79.45 g/l trehalose; 3.13 g/l histidine buffer 20 mM; 1.49 g/l methionine; 0.4 g/l polysorbat 20; 0.024 g/l rHuPh20 (recombinant human hyaluronidase), pH 5.5) was used.
(154) 9.1.1 Sample Preparation
(155) The liquid therapeutic highly concentrated antibody Herceptin® was dialysed at 2-8° C. against a 5 mM histidine buffer pH 5.5. After determination of the protein concentration and subsequent adjustment of the protein concentration to 100 mg/ml the antibody was formulated into the compositions according to paragraph 9.1 as well as the original liquid supplier formulation by 1 per 5 dilution of the dialyzed high concentrated antibody using 1.25 fold concentrated formulations to antibody concentrations of 20 mg/ml. For experiments with high concentrated antibody formulations selective compositions and the antibody formulated in the original liquid supplier formulation were concentrated up to 200 mg/ml. Subsequently, the formulations were sterile filtrated and aliquoted in sterile HPLC vials and stored at 45° C. for up to 14 days. The aggregation and fragmentation before storage, directly after dialysis, directly after the concentration step, and after storage at seven days and 14 days during liquid storage were analyzed using SE-HPLC.
(156) 7.1.2 Size Exclusion Chromatography
(157) SEC was performed according to paragraph 3.1.2.
(158) 7.2 Results
(159) Liquid Storage—SE-HPLC
(160) As shown in Table 6, increased aggregation was observed after seven days storage at 45° C. subsequent to previous dialysis and concentration steps. It was surprisingly found that formulations with two amino acids and without sugar in general exhibited the highest values (e.g. 0.65% after seven days; 1.29% after 14 days) compared with compositions comprising three amino acids (0.53% after seven days; 1.11 after 14 days), and compared with compositions comprising five amino acids (0.4% after seven days; 0.88% after 14 days). The main peaks were stabilized accordingly (Table 6). This observation was confirmed when the same amino acid combinations were supplemented with trehalose. Specifically, formulations with two amino acids with trehalose (corresponding to the original formulation) in general exhibited the highest values (e.g. 0.57% after seven days; 1.07% after 14 days) compared with compositions comprising three amino acids (0.43% after seven days; 0.93 after 14 days), and compared with compositions comprising five amino acids (0.34% after seven days; 0.8% after 14 days). The main peaks were stabilized accordingly (Table 6).
Tables
(161) TABLE-US-00002 TABLE 1 SE-HPLC analysis of highly concentrated trastuzumab directly after sample preparation (t = 0) and at the indicated time points during liquid storage at 55° C. and 40° C. formulated in the original liquid supplier formulation and in the original supplier formulation of the freeze dried product with addition of the amino acids glycine and proline - quantification of aggregates, monomers and fragments expressed in percent areas under the corresponding peaks in the SE-HPLC chromatograms. original supplier formulation of the freeze-dried product original liquid supplier formulation with amino acid additives glycine and proline aggregates monomers fragments aggregates monomers fragments t = 0 0.82 99.18 0.78 99.23 24 h 55° C. 1.21 98.68 0.11 1.21 98.68 0.11 7 d 40° C. 1.28 98.58 0.14 1.26 98.57 0.17 14 d 40° C. 1.49 98.29 0.22 1.62 98.07 0.31 28 d 40° C. 1.82 97.70 0.49 2.25 97.23 0.54 42 d 40° C. 2.09 97.27 0.64 3.17 95.91 0.93
(162) TABLE-US-00003 TABLE 2 SE-HPLC analysis of highly concentrated trastuzumab directly after sample preparation (t = 0) and at the indicated time points during liquid storage at 55° C. and 40° C. formulated in the compositions according to the invention - quantification of aggregates. monomers and fragments expressed in percent areas under the corresponding peaks in the SE-HPLC chromatograms. composition_4_1 composition_4_3 composition_4_5 aggregates monomers fragments aggregates monomers fragments aggregates monomers fragments t = 0 0.69 99.31 0.70 99.31 0.65 99.36 24 h 55° C. 0.81 99.13 0.06 0.81 99.14 0.06 0.77 99.16 0.07 7 d 40° C. 0.90 99.00 0.12 0.85 99.04 0.12 0.85 98.97 0.18 14 d 40° C. 1.04 98.63 0.33 0.98 98.71 0.32 0.99 98.71 0.31 28 d 40° C. 1.32 98.06 0.62 1.24 98.38 0.39 1.24 98.13 0.63 42 d 40° C. 1.77 97.27 0.97 1.36 97.83 0.78 1.53 97.58 0.89
(163) TABLE-US-00004 TABLE 3 SE-HPLC analysis of highly concentrated trastuzumab directly after sample preparation (t = 0) and at the indicated time points during liquid storage at 25° C. formulated in the original liquid supplier formulation and in the original supplier formulation of the freeze dried product with addition of the amino acids glycine and proline - quantification of aggregates, monomers and fragments expressed in percent areas under the corresponding peaks in the SE-HPLC chromatograms. original supplier formulation of the freeze-dried original liquid supplier formulation product with amino acid additives glycine and proline aggregates monomers fragments aggregates monomers fragments t = 0 0.82 99.18 0.78 99.23 14 d 25° C. 1.23 98.77 0.05 1.15 98.79 0.07 28 d 25° C. 1.19 98.73 0.08 1.18 98.72 0.11 42 d 25° C. 1.27 98.63 0.11 1.28 98.62 0.12 2 months 25° C. 1.28 98.57 0.16 1.44 98.28 0.30
(164) TABLE-US-00005 TABLE 4 SE-HPLC analysis of highly concentrated trastuzumab directly after sample preparation (t = 0) and at the indicated time points during liquid storage at 25° C. formulated in the compositions according to the invention - quantification of aggregates, monomers and fragments expressed in percent areas under the corresponding peaks in the SE-HPLC chromatograms. composition_4_1 composition_4_3 composition_4_5 aggregates monomers fragments aggregates monomers fragments aggregates monomers fragments t = 0 0.69 99.31 0.695 99.305 0.65 99.36 14 d 25° C. 0.79 99.22 0.765 99.235 0.01 0.75 99.21 0.08 28 d 25° C. 0.87 99.01 0.13 0.835 99.025 0.14 0.79 99.08 0.14 42 d 25° C. 0.96 99.04 0.87 99.125 0.85 99.20 2 months 25° C. 0.98 98.87 0.16 0.905 98.905 0.19 0.81 98.99 0.21
(165) TABLE-US-00006 TABLE 5 Main peaks of CEX-HPLC chromatograms of highly concentrated trastuzumab directly after sample preparation (t = 0) and at the indicated time points during liquid storage at 55° C., 40° C., and 25° C. formulated in the original supplier formulation of the freeze dried product supplemented with the amino acids glycine and proline (original + G/P) and in compositions 4_1, 4_3 and 4_5 according to the present invention. Main peaks are expressed as percent areas under the corresponding chromatogram peaks. original + G/P composition_4_1 composition_4_3 composition_4_5 t = 0 67.10 68.52 67.99 68.14 24 h 55° C. 53.23 58.26 57.11 56.87 3 d 55° C. 32.22 41.72 41.05 40.22 1.5 d 40° C. 58.97 63.36 63.10 62.96 3 d 40° C. 56.88 63.44 62.97 62.95 7 d 40° C. 46.32 55.62 55.61 53.99 14 d 40° C. 30.53 43.10 43.14 41.87 21 d 40° C. 21.35 35.05 34.01 33.68 7 d 25° C. 62.58 66.90 66.47 66.38 14 d 25° C. 57.02 64.31 64.04 64.01 28 d 25° C. 48.98 60.93 60.89 60.06 42 d 25° C. 41.67 55.89 55.79 54.75 2 months 25° C. 35.65 50.63 50.44 49.59
(166) TABLE-US-00007 TABLE 6 SE-HPLC analysis of highly concentrated trastuzumab directly after sample preparation (t = 0) and at the indicated time points during liquid storage at 45° C. formulated in the compositions according to the invention - quantification of aggregates, monomers and fragments expressed in percent areas under the corresponding peaks in the SE-HPLC chromatograms. t = 0 Liquid storage after Liquid storage after (after dialysis) After concentration step concentration step 7 d 45° C. concentration step 14 d 45° C. aggre- mono- frag- aggre- mono- frag- aggre- mono- frag- aggre- mono- frag- t = 0 gates mers ments gates mers ments gates mers ments gates mers ments 2 amino acids 0.19 99.78 0.03 0.21 99.76 0.03 0.65 98.89 0.46 1.29 98.12 0.59 w/o sugar 3 amino acids 0.24 99.73 0.04 0.19 99.79 0.03 0.53 99.02 0.46 1.11 98.29 0.61 w/o sugar 5 amino acids 0.19 99.79 0.03 0.18 99.80 0.03 0.40 99.09 0.52 0.88 98.44 0.68 w/o sugar 2 amino acids 0.19 99.78 0.04 0.20 99.77 0.03 0.57 99.00 0.44 1.07 98.37 0.55 with trehalose (original formulation) 3 amino acids 0.19 99.78 0.03 0.18 99.80 0.03 0.43 99.06 0.52 0.93 98.44 0.63 with trehalose 5 amino acids 0.18 99.79 0.03 0.18 99.80 0.03 0.34 99.17 0.49 0.80 98.60 0.60 with trehalose