Process for lyophilized pharmaceutical formulation of therapeutic protein

Abstract

To provide a simple process where a product is formulated in a targeted concentration and then filled into containers at a set volume.SOLUTION: This invention concerns a process for making a lyophilized pharmaceutical formulation of a therapeutic protein, the process comprising (a) providing a formulation of a bulk amount of the therapeutic protein, (b) measuring the concentration of the therapeutic protein in the bulk formulation, (c) adjusting the fill weight of the protein in the bulk formulation to achieve a fixed dose of the protein, and (d) lyophilizing the protein fill weight-adjusted formulation to achieve a final formulation in a container, where the product concentration after reconstitution with a fixed volume is within a predetermined acceptance range. The process is particularly suitable for formulations with low protein concentrations (e.g., 0.05 mg / mL to 20 mg / mL).SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to biopharmaceuticals, in particular to therapeutic proteins, methods of use thereof, pharmaceutical formulations thereof, and processes for producing pharmaceutical formulations. In particular, the present invention relates to a process for producing a lyophilized pharmaceutical formulation. [Background technology]

[0002] Over the past decade, advances in technology have made it possible to generate a wide variety of active molecules for pharmaceutical use. Advances in the understanding of the nature of biological mechanisms of action have made it possible to engineer these molecules for specific attributes that have an effect in small quantities.

[0003] Formulation of such products presents special challenges since these molecules can be larger and / or more complex (i.e., have multiple functional groups in addition to complex three-dimensional structures) compared to traditional organic and inorganic chemicals. To maintain the biological activity of the product, the formulation must preserve the conformational integrity of at least the core sequence of amino acids of the protein intact while protecting the multiple functional groups of the protein from degradation. Protein degradation pathways can involve chemical instability (i.e., any process involving modification of a protein by bond formation or cleavage resulting in a new chemical entity) or physical instability (i.e., changes in the higher-order structure of the protein). Chemical instability can be caused by deamidation, racemization, hydrolysis, oxidation, beta-elimination, or disulfide exchange. Physical instability can be caused by, for example, denaturation, aggregation, precipitation, or adsorption. The three most common protein degradation pathways are protein aggregation, deamidation, and oxidation. Cleland et al. (1993), Critical Reviews in Therapeutic Drug Carrier Systems 10(4):307-377.

[0004] These designed molecules, due to their synthetic nature, are commonly lyophilized (freeze-dried) and have been shown to provide improved storage stability. Freeze-drying is a commonly used technique for preserving proteins and serves to remove water from protein preparations of interest. Freeze-drying, or lyophilization, is a process in which the material to be dried is first frozen, followed by removal of the ice or freezing solvent by sublimation under a vacuum environment. Excipients may be included in the pre-lyophilized formulation to enhance stability during the freeze-drying process and / or to improve the stability of the lyophilized product upon storage. Pikal, M. (1990), Biopharm. 3(9) 26-30 and Arakawa et al. (1991), Pharm. Res. 8(3): 285-291. Summary of the Invention [Problem to be solved by the invention]

[0005] Designing molecules with specific biological targets and the required dosage of the resulting products poses new challenges to the manufacturing process. Current technology involves a simple process of formulating the product to a target concentration and then filling containers at a defined volume. [Means for solving the problem]

[0006] The present invention relates to a process for producing a lyophilized formulation. Specifically, the present invention relates to compounding, filling, and ensuring the presence of a required amount of product after reconstitution of the lyophilized product with a fixed volume of diluent for use.

[0007] According to the present invention there is provided a process for producing a lyophilized pharmaceutical formulation of a therapeutic protein comprising the steps of: (a) providing a bulk amount of a therapeutic protein formulation; (b) determining the concentration of the therapeutic protein in said bulk formulation; (c) adjusting the loading weight of the protein in said bulk formulation to obtain a fixed dose of protein; (d) lyophilizing the adjusted protein fill weight formulation to obtain a final formulation in a container; A process is provided in which the concentration of the product after reconstitution with a fixed volume is within a predetermined acceptable range.

[0008] In the above process, the protein concentration in the final formulation is preferably about 20 or 25 mg / mL or less, with about 0.5 mg / mL, about 0.05 mg / mL, about 18 mg / mL, about 20 mg / mL, and about 21 mg / mL being most preferred. Preferred therapeutic proteins for the process of the present invention are romiplostim, blinatumomab, infliximab, trastuzumab, AMG701, and AMG330. AMG701 and AMG330 are bispecific single chain antibody constructs, and other bispecific single chain antibody constructs (e.g., bispecific T cell engagers) are also preferred therapeutic proteins in the process of the present invention. Preferred pharmaceutical excipients present in the formulation include sugars, with trehalose, sucrose, and hydrates of either being most preferred. Preferred pharmaceutical excipients further include buffers, with histidine, citric acid monohydrate, sodium phosphate, potassium phosphate, and glutamic acid being preferred. Preferred excipients further comprise a surfactant, most preferably polysorbate 20 and polysorbate 80. Preferred excipients and therapeutic proteins for use in accordance with the process of the present invention are set forth in Table 1, with the approximate preferred concentrations of each listed under each protein and excipient.

[0009] [Table 1]

[0010] Furthermore, in accordance with the present invention, the formulation may contain other excipients as described herein below. [Brief description of the drawings]

[0011] [Figure 1] Response surface map showing an exemplary design space for a low dose product where formulation and fill weight follow a typical control strategy. The grey space represents where method / reconstitution variability has a greater than 50% chance of having a non-compliant protein concentration result. The large square indicates the current operating range for formulation development. The small rectangle indicates the valid operating range. [Diagram 2] Response surface map showing an exemplary design space for a low dose product where osmolality is considered in addition to bulk concentration and fill volume. The light grey area indicates stepped-in formulation concentration specifications that have a greater than 50% chance of method / reconstitution variability resulting in non-compliant protein concentration. The dark grey area indicates osmolality specifications. The rectangles indicate the in-process control (IPC) / alert limit operating range (ALOR), which is the design space range of fill weight (x-axis) and formulation bulk concentration (y-axis). [Diagram 3] Response surface maps showing examples of feasible, but not fault-tolerant, fill weight and formulation control strategies. The curves and double-headed arrows indicate possible fill target errors. The light grey area represents the stepped formulation concentration specifications, and the light grey area indicates where method / reconstitution variability has a greater than 50% chance of producing non-compliant protein concentration results. The dark grey area defines the osmolality specification. The rectangle indicates the IPC / ALOR. [Figure 4] Response surface map showing an example of a fill weight and formulation control strategy with difficulty in controlling the range ends, with low expected formulation bulk and low fill weight. The curve and double arrow on the left indicates possible fill target deviation at low fill volumes. The curve and double arrow on the right indicates possible fill target deviation at high fill volumes. The light grey area represents the stepped formulation concentration specification, where the light grey area indicates where there is a greater than 50% chance of method / reconstitution variability resulting in non-compliant protein concentration. The dark grey area defines the osmolality specification. The rectangle indicates the IPC / ALOR. [Diagram 5]FIG. 1 is a response surface map showing an example of a combination strategy that uses the results of bulk compounding and subsequently adjusts the fill weight set point based on the total product dose in the vial target to result in a viable and fault-tolerant lyophilization drug process. The curves and double-headed arrows indicate possible fill target errors. The light grey area defines the graded formulation concentration specifications found in FIGS. 1-4. The dark grey area defines the osmolality specification. The diamonds indicate the IPC / ALOR and are clipped by the osmolality specification at the maximum fill volume within the IPC / ALOR. [Figure 6] Normalized fill weight versus product protein concentration after reconstitution, as determined according to Example 1 below, is shown. [Figure 7] Normalized fill weight versus osmolality is shown, as determined according to Example 1 below. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Definition of Terms In the description that follows, a number of terms are used extensively. The following definitions are provided to facilitate understanding of the invention.

[0013] Unless otherwise stated, "a," "an," "the," and "at least one" are used interchangeably and mean one or more than one. Further, unless the context otherwise requires, singular terms shall include the plural and plural terms shall include the singular.

[0014] As used herein, a "pharmaceutical formulation" or "formulation" is a sterile composition of (i) a pharma- ceutical active drug, such as a biologically active protein, suitable for parenteral administration (including, but not limited to, intravenous, intramuscular, subcutaneous, aerosol, intrapulmonary, intranasal, or intrathecal administration) to a patient in need thereof, and (ii) one or more pharma- ceutical acceptable excipients, diluents, and other additives deemed safe by the Federal Drug Administration or other foreign or domestic authorities. Pharmaceutical formulations include liquid (e.g., aqueous) solutions that can be administered directly, and lyophilized powders that can be reconstituted into a solution by adding a diluent prior to administration. However, the term "pharmaceutical formulation" specifically excludes compositions for topical administration to a patient, compositions for oral ingestion, and compositions for parenteral nutritional supplementation.

[0015] As used herein, "shelf life" refers to a storage period during which there is minimal degradation (e.g., no more than about 5%-10% degradation) of the active ingredient (e.g., antibody) in a pharmaceutical formulation when the pharmaceutical formulation is stored under a particular storage condition (e.g., 2-8°C). Techniques for assessing degradation vary depending on the identity of the protein in the pharmaceutical formulation. Exemplary techniques include size exclusion chromatography (SEC)-HPLC, e.g., to detect aggregation, reversed phase (RP)-HPLC, e.g., to detect protein fragmentation, ion exchange-HPLC, e.g., to detect changes in protein charge, as well as mass spectrometry, fluorescence spectroscopy, circular dichroism (CD) spectroscopy, Fourier transform infrared spectroscopy (FT-IR), and Raman spectroscopy to detect conformational changes in proteins. All of these techniques can be used alone or in combination to assess the degradation of proteins in a pharmaceutical formulation and determine the shelf life of the formulation. The pharmaceutical formulations of the present invention preferably do not increase in degradation (e.g., fragmentation, aggregation, or unfolding) by more than about 5-10% over a period of two years when stored at 2-8°C.

[0016] As used herein, a "stable" formulation of a biologically active protein is one that exhibits either (i) at least a 20% reduction in aggregation and / or loss of biological activity when stored at 2-8° C. for at least 2 years compared to a control formulation sample, or (ii) reduced aggregation and / or reduced loss of biological activity under heat stress conditions (e.g., 1-12 weeks at 25° C., 1-12 weeks at 40° C., 7-8 days at 52° C., etc.). In one embodiment, a formulation is considered stable if the protein in the formulation retains its physical stability, chemical stability, and / or biological activity.

[0017] A protein can be said to "retain its physical stability" in a formulation if it shows no signs of aggregation, precipitation, and / or denaturation, for example, as measured by visual inspection for color and / or clarity, or by UV scattering or size exclusion chromatography (SEC) or electrophoresis for turbidity or aggregate formation, etc.

[0018] A protein can be said to "retain its chemical stability" in a formulation if, for example, the chemical stability at a given time point results in no new chemicals resulting from modification of the protein by bond formation or cleavage. In further embodiments, chemical stability can be assessed by detecting and quantifying chemically altered forms of the protein. Chemical alterations can involve, for example, size changes (e.g., clipping) that can be assessed using size exclusion chromatography, SDS-PAGE, and / or matrix-assisted laser desorption / ionization / time-of-flight mass spectrometry (MALDI / TOF MS). Other types of chemical alterations include charge changes (e.g., resulting from deamidation) that can be assessed by, for example, ion exchange chromatography. Oxidation is another commonly found chemical modification.

[0019] A protein can be said to "retain its biological activity" in a pharmaceutical formulation if, for example, the percentage of biological activity of the formulated protein (e.g., an antibody) is about 50% to about 200%, about 60% to about 170%, about 70% to about 150%, about 80% to about 125%, or about 90% to about 110% as compared to the unmodified protein, as measured in an assay (e.g., an antigen binding assay) compared to a control solution. In further embodiments, a protein can be said to "retain its biological activity" in a pharmaceutical formulation if, for example, but not limited to, the biological activity of the protein at a given time is at least 1%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 100%.

[0020] As used herein, the terms "comprising" and "comprises" are intended to mean that the formulations and methods include the recited elements, but do not exclude other unrecited elements. The terms "consisting essentially of" and "consists essentially of," when used to define the formulations and methods as including the recited elements, exclude unrecited elements that would alter the basic nature of the formulation and / or method, but do not exclude other unrecited elements. Thus, a formulation consisting essentially of the elements defined herein does not exclude contaminants from any separation and purification methods, or trace amounts of other elements, such as pharma- ceutically acceptable carriers (e.g., phosphate buffered saline), preservatives, and the like, but does exclude, for example, additional unspecified amino acids, and the like. The terms "consisting of" and "consists of," when used to define the formulations and methods, exclude more than trace amounts of other components, and substantial method steps for administering the compositions described herein. Embodiments defined by each of these transitional phrases are within the scope of this disclosure and the invention(s) embodied herein.

[0021] As used herein, the term "isolated" refers to a protein (e.g., an antibody) that has been identified and separated and / or recovered from a component of its natural environment. Contaminating components of its natural environment are substances that would interfere with diagnostic or therapeutic uses of the protein, and may include enzymes, hormones, and other proteinaceous or non-proteinaceous solutes. In preferred embodiments, the protein is purified (1) to greater than 95% by weight of the antibody as measured by the Lowry method, most preferably greater than 99% by weight, (2) to the extent sufficient to obtain at least 15 residues of N-terminal or internal amino acid sequence by use of a spinning cup sequencer, or (3) to homogeneity by SDS-PAGE under reducing or non-reducing conditions using Coomassie blue, or preferably silver staining. Isolated proteins include proteins in situ, within recombinant cells, since at least one component of the protein's natural environment is not present. Ordinarily, however, isolated proteins are prepared by at least one purification step.

[0022] The present invention relates to a process for the pharmaceutical formulation of therapeutic proteins such as antibodies. "Antibody" (Ab) and the synonym "immunoglobulin" (Ig) are glycoproteins with the same structural characteristics. While antibodies exhibit binding specificity to a specific antigen, immunoglobulins include both antibodies and other antibody-like molecules that lack antigen specificity. The latter type of polypeptide is produced at low levels, for example, by the lymphatic system and at higher levels by myelomas. Thus, as used herein, the term "antibody" or "antibody peptide" refers to intact antibodies, antibody derivatives, antibody analogs, genetically engineered antibodies, antibodies with a detectable label, antibodies that compete for specific binding with an antibody disclosed herein, or antigen-binding fragments thereof (e.g., Fab, Fab', F(ab')2, Fv, single domain antibodies) that compete for specific binding with the intact antibody, including chimeric antibodies, humanized antibodies, fully human antibodies, and bispecific antibodies. In certain embodiments, the antigen-binding fragments are produced, for example, by recombinant DNA techniques. In further embodiments, antigen-binding fragments are produced by enzymatic or chemical cleavage of intact antibodies. Antigen-binding fragments include Fab, Fab', F(ab) 2 , F(ab') 2 , Fv, and single chain antibodies.

[0023] As used herein, the term "intact antibody" refers to an antibody that comprises two heavy chains and two light chains. Thus, this term includes, but is not limited to, fully human antibodies, genetically engineered antibodies, bispecific antibodies, and antibody derivatives, provided that such antibodies contain two heavy chains and two light chains.

[0024] The term "monoclonal antibody," as used herein, is not limited to antibodies produced through hybridoma technology. The term "monoclonal antibody" refers to an antibody that is derived from a single clone, including any eukaryotic, prokaryotic, or phage clone, and not the method by which it is produced.

[0025] The monoclonal antibodies and antibody constructs formulated according to the invention expressly include "chimeric" antibodies (immunoglobulins) in which a portion of the heavy and / or light chain is identical or homologous to corresponding sequences in antibodies derived from a particular species or belonging to a particular antibody class or subclass, while the remainder of the chain is identical or homologous to corresponding sequences in antibodies derived from another species or belonging to another antibody class or subclass, as well as fragments of such antibodies, so long as they exhibit the desired biological activity (U.S. Pat. No. 4,816,567; Morrison et al. (1984), Proc. Natl. Acad. Sci. USA, 81:6851-6855). Chimeric antibodies of interest herein include "primitized" antibodies that contain variable domain antigen-binding sequences derived from a non-human primate (e.g., Old World monkeys, apes, etc.) and human constant region sequences. Various methods for producing chimeric antibodies have been described. See, e.g., Morrison et al. (1985), Proc. Natl. Acad. Sci. USA 81:6851; Takeda et al. (1985), Nature 314:452; U.S. Pat. No. 4,816,567 to Cabilly et al.; U.S. Pat. No. 4,816,397 to Boss et al.; EP Pat. Nos. 0171496, 0173494, and GB Pat. No. 2177096 to Tanaguchi et al.

[0026] The monoclonal antibodies and antibody constructs formulated according to the present invention expressly include antibodies that are designated "human" or "fully human." The terms "human antibody" and "fully human antibody" refer to antibodies that have the amino acid sequence of a human immunoglobulin and that do not express endogenous immunoglobulins, including antibodies isolated from a human immunoglobulin library or from an animal into which one or more human immunoglobulin genes have been introduced, respectively, such as Xenomouse® antibodies and those described by Kucherlapati et al. in U.S. Patent No. 5,939,598.

[0027] The term "genetically engineered antibody" refers to an antibody whose amino acid sequence has been altered from that of a natural antibody. Due to the relevance of recombinant DNA technology in the generation of antibodies, it is not necessary to be limited to the sequence of amino acids found in natural antibodies, and antibodies can be redesigned to obtain desired properties. There are many possible variations, ranging from changes to just one or a few amino acids to complete redesign of, for example, the variable and / or constant regions. Changes in the constant region are generally made to improve or alter properties such as complement fixation, interaction with membranes, and other effector functions, as well as manufacturability and viscosity. Changes in the variable region are made to improve antigen-binding properties.

[0028] A "Fab fragment" is a fragment of one light chain and one heavy chain. H1 and a variable region. The heavy chain of a Fab molecule cannot form disulfide bonds with another heavy chain molecule.

[0029] A "Fab' fragment" is a fragment consisting of one light chain and two heavy chains separated by a C-terminal fragment such that interchain disulfide bonds can be formed between them to form an F(ab')2 molecule. H1 Domain and C H2 and one heavy chain that contains more of the constant region between the domains.

[0030] "F(ab')2 fragment" refers to a fragment of two light chains and two heavy chains that have been split into two fragments such that interchain disulfide bonds are formed between them. H1 Domain and C H2 and two heavy chains containing a portion of the constant region between the domains.

[0031] The terms "Fv fragment" and "single chain antibody" refer to a polypeptide that contains both heavy and light chain antibody variable regions but lacks the constant regions. Like a whole antibody, it is capable of selectively binding to a specific antigen. With a molecular weight of only about 25 kDa, the Fv fragment is much smaller than a conventional antibody (150-160 kDa), which is composed of two heavy protein chains and two light chains, and even smaller than a Fab fragment (about 50 kDa, one light chain and half a heavy chain).

[0032] A "single domain antibody" refers to a single domain Fv unit, e.g., V H Or V L Single domain antibodies are antibody fragments consisting of a heavy chain and a light chain. Like whole antibodies, they can selectively bind to specific antigens. With a molecular weight of only 12-15 kDa, single domain antibodies are much smaller than conventional antibodies (150-160 kDa) composed of two heavy protein chains and two light chains, and even smaller than Fab fragments (about 50 kDa, one light chain and half a heavy chain) and single chain variable fragments (about 25 kDa, two variable domains, one from a light chain and one from a heavy chain). The first single domain antibodies were genetically engineered from heavy chain antibodies found in camelids. Most of the research on single domain antibodies is currently based on heavy chain variable domains, but nanobodies derived from light chains and light chains have also been shown to specifically bind target epitopes.

[0033] As used herein, the term "bispecific" refers to an antibody construct that is "at least bispecific", i.e., it comprises at least a first binding domain and a second binding domain, where the first binding domain binds to one antigen or target (e.g. CD3) and the second binding domain binds to another antigen or target (e.g. BCMA; e.g. CD33). Thus, an antibody construct according to the invention comprises specificity for at least two different antigens or targets. The term "bispecific antibody construct" of the present invention also encompasses multispecific antibody constructs, such as trispecific antibody constructs comprising three binding domains or constructs with four or more (e.g. four, five...) specificities.

[0034] When the antibody constructs according to the invention are (at least) bispecific, they do not occur in nature and they differ significantly from naturally occurring products. Thus, a "bispecific" antibody construct or immunoglobulin is an artificial hybrid antibody or immunoglobulin having at least two different binding sites with different specificities. Bispecific antibody constructs can be produced by a variety of methods including fusion of hybridomas or linking of Fab' fragments. See, for example, Songsivilai & Lachmann, Clin. Exp. Immunol. 79:315-321 (1990).

[0035] The at least two binding domains and the variable domain of the antibody construct of the present invention may or may not comprise a peptide linker (spacer peptide). The term "peptide linker" according to the present invention comprises an amino acid sequence that interconnects the amino acid sequences of one (variable and / or binding) domain and the other (variable and / or binding) domain of the antibody construct of the present invention. An essential technical feature of such a peptide linker is that it does not contain any polymerization activity. Suitable peptide linkers include those described in US Pat. Nos. 4,751,180 and 4,935,233 or WO 88 / 09344. Peptide linkers may also be used to attach other domains or modules or regions (such as half-life extending domains) to the antibody construct of the present invention.

[0036] If a linker is used, it is preferred that this linker is of sufficient length and sequence to ensure that each of the first and second domains retains its different binding specificity independently of each other. In the case of peptide linkers connecting at least two binding domains (or two variable domains) in the antibody construct of the invention, these peptide linkers preferably contain only a few amino acid residues, for example 12 amino acid residues or less. Thus, peptide linkers of 12, 11, 10, 9, 8, 7, 6 or 5 amino acid residues are preferred. Conceivable peptide linkers with less than 5 amino acids contain 4, 3, 2 or 1 amino acid, with Gly-rich linkers being preferred. A particularly preferred "single" amino acid for the above "peptide linker" is Gly. Thus, the above peptide linker may consist of a single amino acid Gly. Another preferred embodiment of the peptide linker is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e. Gly4Ser (SEQ ID NO: 1) or a polymer thereof, i.e. (Gly4Ser)x, where x is 1 or a larger integer (e.g. 2 or 3). The above peptide linker features, including the lack of promotion of secondary structures, are known in the art and are described, for example, in Dall'Acqua et al. (Biochem. (1998) 37, 9266-9273), Cheadle et al. (Mol. Immunol. (1992) 29, 21-30), and Raag and Whitlow (FASEB (1995) 9(1), 73-80). Moreover, peptide linkers that do not promote any secondary structures are preferred. The interconnection of the above domains can be provided, for example, by genetic engineering as described in the Examples. Methods for preparing fused, operably linked bispecific single chain constructs and expressing them in mammalian cells or bacteria are well known in the art (e.g., WO 99 / 54440 or Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2001).

[0037] According to a particularly preferred embodiment and as described in the accompanying examples, each of the AMG701 and AMG330 antibody constructs of the invention is a "bispecific single chain antibody construct", more preferably a bispecific "single chain Fv" (scFv). Although the two domains of the Fv fragment, VL and VH, are encoded by separate genes, they can be linked by a synthetic linker as described herein above, which allows them to be produced using recombinant methods as a single protein chain in which the VL and VH regions pair to form a monovalent molecule (see, for example, Huston et al. (1988) Proc. Natl. Acad. Sci USA 85:5879-5883). These antibody fragments are obtained using conventional techniques known to those skilled in the art, and the fragments are evaluated for function in the same manner as complete or full-length antibodies. Thus, a single chain variable fragment (scFv) is a fusion protein of the variable region of the heavy chain (VH) and the variable region of the light chain (VL) of an immunoglobulin, connected by a short linker peptide, usually about 10 to about 25 amino acids, preferably about 15 to 20 amino acids. The linker is usually rich in glycine for flexibility and serine or threonine for solubility, and can either link the N-terminus of VH to the C-terminus of VL or vice versa. The protein retains the specificity of the original immunoglobulin despite the removal of the constant region and the introduction of the linker.

[0038] Bispecific single chain molecules are known in the art and are described in WO 99 / 54440; Mack, J. Immunol. (1997), 158, 3965-3970; Mack, PNAS, (1995), 92, 7021-7025; Kufer, Cancer Immunol. Immunother., (1997), 45, 193-197; Loeffler, Blood, (2000), 95, 6, 2098-2103; Bruehl, Immunol., (2001), 166, 2420-2426; Kipriyanov, J. Mol. Biol., (1999), 293, 41-56. Techniques described for the production of single chain antibodies (see, inter alia, U.S. Pat. No. 4,946,778) can be adapted to produce single chain antibody constructs which specifically recognize a target of choice.

[0039] Bivalent (also called divalent) or bispecific single chain variable fragments (bi-scFv or di-scFv with the format (scFv)2) can be created by linking two scFv molecules (e.g., using a linker as described herein above). If these two scFv molecules have the same binding specificity, the resulting (scFv)2 molecule is preferably called bivalent (i.e., it has a valency of two for the same target epitope). If these two scFv molecules have different binding specificities, the resulting (scFv)2 molecule is preferably called bispecific. Linking can be done by creating a single peptide chain with two VH and two VL regions to generate tandem scFvs (see, for example, Kufer P. et al., (2004) Trends in Biotechnology 22(5):238-244). Another possibility is to create scFv molecules with a linker peptide that is too short (e.g., about 5 amino acids) to allow the two variable regions to fold together, allowing the scFvs to dimerize, a form known as a diabody (see, e.g., Hollinger, Philipp et al., (July 1993) Proceedings of the National Academy of Sciences of the United States of America 90(14):6444-8).

[0040] As described herein above, the present invention provides preferred embodiments in which the antibody construct is in a format selected from the group consisting of (scFv)2, scFv-single domain mAb, diabodies and oligomers of any of these formats.

[0041] According to another preferred embodiment of the antibody construct of the invention, the heavy (VH) and light (VL) chains of the binding domains binding to the target antigens CD3 and either CD33 or BCMA are not directly linked via the above mentioned peptide linker, but rather the binding domains are formed due to the formation of a bispecific molecule as described for diabodies. Thus, the VH chain of the CD3 binding domain may be fused to the VL of the CD33 or BCMA binding domain via such a peptide linker and at the same time the VH chain of the CD3 binding domain is fused to the VL of the CD33 or BCMA binding domain via such a peptide linker.

[0042] As used herein, the terms "amino-terminus" and "carboxyl-terminus" and their shortened forms "N-terminus" and "C-terminus" are used to denote positions within a polypeptide. Where the context permits, these terms are used with respect to a particular sequence or portion of a polypeptide to indicate proximity or relative position. For example, a particular sequence located carboxyl-terminal to a reference sequence within a polypeptide is located close to the carboxyl terminus of the reference sequence, but is not necessarily located at the carboxyl terminus of the complete polypeptide.

[0043] As used herein, the term "amino acid" refers to any natural and / or unnatural or synthetic amino acids, including both glycine and the D or L optical isomers, including but not limited to N-acetyl analogs of the D or L optical isomers (e.g., N-acetylarginine), as well as amino acid analogs and peptidomimetics. In some embodiments, the term "amino acid" refers to a monomeric amino acid.

[0044] Generally, the nomenclatures and techniques used in connection with cell and tissue culture, molecular biology, immunology, microbiology, genetics, and protein and nucleic acid chemistry, and hybridization described herein are well known and commonly used in the art. Unless otherwise stated, the methods and techniques of the present invention are generally performed according to conventional methods well known in the art and as described in the general and more specific references cited and discussed throughout this specification. See, for example, Sambrook et al. (2001), Molecular Cloning: A Laboratory Manual, 3rd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, and Ausubel et al. (1992), Current Protocols in Molecular Biology, Greene Publishing Associates, and Harlow and Lane (1990), Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Any enzymatic reactions and purification techniques are performed according to manufacturer's instructions, as commonly accomplished in the art, or as described herein. The nomenclature used in connection with, and the laboratory procedures and techniques of, analytical chemistry, synthetic organic chemistry, and medicinal and pharmaceutical chemistry described herein are those well known and commonly used in the art. Standard techniques may be used for chemical syntheses, chemical analyses, pharmaceutical preparation, formulation, and delivery, and treatment of patients.

[0045] Use of intermediate blend concentrations The goal of protein formulation is to convert a high purity recombinant protein solution (the drug substance) into a stable and effective biopharmaceutical dosage form. Kamerzell et al. (2011), 63(13):1118-59 (incorporated by reference). The first step, sometimes referred to as preformulation characterization, involves determining the physicochemical properties and instability pathways of the protein, which allows the design of formulations containing various excipients to ensure protein stability under the given storage conditions. At the same time, analytical methods must be developed to monitor the physicochemical integrity and biological activity of the protein under formulation conditions (e.g., in the presence of excipients), along with specifications that define acceptable limits for any changes in these parameters. Subsequently, specific formulations at various protein concentrations with target levels of various pharmaceutical excipients are experimentally tested to ensure stability, solubility, and isotonicity over the shelf life. In addition, a primary container is selected (e.g., vials, cartridges, or prefilled syringes) to store the protein-excipient mixture and facilitate parenteral administration by the patient or medical professional. The entire biopharmaceutical drug or vaccine dosage form (protein, excipients, primary container, and delivery device) must be designed both for scalability to allow commercial production under sterile conditions and to comply with all of the regulatory guidance for the production and testing of biopharmaceutical dosage forms for human use.

[0046] Formulation development typically begins with identifying the correct pH and buffer system through a biophysical screening strategy. Buffers are added to the protein solution to stabilize the pH, which in turn stabilizes the protein, since protein stability is typically linked to its specific narrow pH range properties. See Garidel and Bassarab (2009), "Impact of formulation design on stability and quality," in Quality for Biologics: Critical Quality Attributes, Process and Change Control, Production Variation, Characterisation, Impurities and Concerns, Biopharm Knowledge Publishing, London, UK, pp. 94-113, which is incorporated by reference. Other formulation excipients, such as stabilizers (e.g., sugars), bulking agents (e.g., mannitol), and surfactants (e.g., polysorbate 20), are then added to the buffered protein solution. In the case of lyophilized formulations, these liquid formulation mixtures are then freeze-dried.

[0047] In lyophilization in other formulation processes, the key control is the loading weight of the therapeutic protein that delivers the required dose to the patient. For high concentration products, a simple strategy of controlling the product concentration at the compounding stage, followed by loading weight control as part of the loading process, is adequate. However, when delivering very low doses of products in the microgram or even milligram range, this simple strategy to ensure dose delivery begins to create problems in ensuring that the patient can receive the required dose. A related concern is that the product label may not reflect the product in the container, and that it may be difficult to remove the material for dosing. Thus, there is a need for a process that provides better control over the loading weight of the therapeutic protein in lyophilized pharmaceutical formulations.

[0048] As mentioned above, for high concentration therapeutic proteins, a simple strategy of controlling product concentration at the compounding stage, followed by filling weight control as part of the filling process, is adequate, but can cause problems at low concentrations. The design space for filling weight control of low dose products when the filling weight and compounding targets are controlled, shows high variability individually, with a large probability that the product will not meet the release specification. Due to manufacturing variability, precise control of compounding bulk cannot be achieved due to the scale of the process and the inherent variability of the manufacturing equipment. Another problem is that high variability is inherent in lyophilized products, because the product needs to be reconstituted, which is also a variable process.

[0049] A typical control strategy (shown in Figure 1) has a greater than 50% chance of having an incompatible protein concentration result. In Figure 1, the observed overall variability of the lyophilization homogeneity data is 0.01 mg / mL. To guarantee minimum capability, the average operating range needs to be constrained to correspond to a specification range "stepped" by 3 x 0.01 = 0.03 mg / mL. Processes that target within the white space have a quality of less than 0.1% out-of-specification (OOS) vials per batch. In contrast, processes that target within the shaded area have a higher OOS rate. As can be seen from Figure 1, the current operating range (large square) includes the edges of failure of the desired formulation. However, the effective operating range (smaller rectangle in Figure 1) requires tighter fill tolerances than can be achieved as part of both formulation variability and filler weight control.

[0050] Thus, processes that also need to consider other product quality attributes (e.g., osmolality, pH) and protein concentration may not be producible or viable in a standard manner. Figure 2 shows an exemplary design space where osmolality specifications are considered. Grey coloring indicates a wide range of non-compliant conditions that fall within the IPC / alert limit operating range.

[0051] The standard method may also result in a fill weight and formulation control strategy that is viable but not fault tolerant (see Figure 3). As in Figure 1, the IPC / ALOR includes areas where the formulation concentration specifications are not met. Within the areas where the formulation concentration specifications are met (the white areas in the IPC / ALOR), there is little room for error in the fill target. Thus, as shown in Figure 3, the standard method may result in a process that is viable but not sufficiently fault tolerant and has difficulty controlling the range edges.

[0052] Increasing the fill volume may not have enough process variation to produce a viable formulation with adequate fault tolerance. Figure 4 shows a response surface map with increased fill volume. The fill target range may be within the IPC / ALOR and product specifications, but the fault tolerance may remain imperfect. A fill weight range outside the allowable range of the formulation bulk concentration has insufficient fault tolerance.

[0053] When the formulation bulk results are used to subsequently adjust the loading weight of the therapeutic protein, the operating range may be altered sufficiently to achieve a viable and sufficiently fault-tolerant formulation. As shown in Figure 5, the modified operating range (IPC / ALOR) allows the expected loading target range to be well within the concentration and osmolality specifications. In this way, loading target ranges can be provided within the concentration and other specifications with sufficient fault tolerance.

[0054] To date, this technique has been used for low dose stock keeping units (SKUs) for romiplostim (Nplate®), blinatumomab (Blincyto®), infliximab, and trastuzumab. Further details regarding the use of this technique are provided in the Examples below.

[0055] Excipients in general One problem of formulation is to stabilize the product against the stresses of manufacturing, transportation and storage, which can be achieved by certain formulation excipients.In general, excipients can be classified based on their mechanism of stabilizing proteins against various chemical and physical stresses.Some excipients mitigate the effects of specific stresses or regulate the specific sensitivity of specific proteins.Other excipients have a more general effect on the physical and covalent stability of proteins.

[0056] Common excipients for liquid and lyophilized protein formulations are listed in Table 2 (see Kamerzell et al. (2011), Advanced Drug Delivery Rev. 63(13):1118-59).

[0057] [Table 2]

[0058] [Table 3]

[0059] [Table 4]

[0060] Other excipients are known in the art and can be found in Powell et al. (1998), "Compendium of Excipients for Parenteral Formulations," PDA J. Pharm. Sci. Tech., 52:238-311, incorporated herein by reference.

[0061] Given the teachings and instructions provided herein, one of ordinary skill in the art will understand what amounts or ranges of excipients may be included in any particular formulation to obtain a biopharmaceutical formulation of the present invention. For example, the amount and type of salt to be included in a biopharmaceutical formulation of the present invention may be selected based on the desired osmolality (i.e., isotonic, hypotonic, or hypertonic) of the final solution, as well as the amount and osmolality of other components to be included in the formulation. Similarly, by way of illustration with reference to the type of polyol or sugar to be included in the formulation, the amount of such excipient will depend on its osmolality.

[0062] One skilled in the art can determine what amount or range of excipients may be included in any particular formulation to obtain a biopharmaceutical formulation of the invention that promotes retention of the stability of the biopharmaceutical. For example, the amount and type of salt to be included in the biopharmaceutical formulation of the invention can be selected based on the desired osmolality (i.e., isotonic, hypotonic, or hypertonic) of the final solution, as well as the amount and osmolality of other components to be included in the formulation. Similarly, by way of illustration with reference to the type of polyol or sugar to be included in the formulation, the amount of such excipients will depend on its osmolality.

[0063] For example, about 5% (w / v) sorbitol can achieve isotonicity, while about 9% (w / v) of sucrose excipient is required to achieve isotonicity. Selection of the range of amounts or concentrations of one or more excipients that can be included in the biopharmaceutical formulations of the present invention is exemplified above with reference to salts, polyols, and sugars. However, one of ordinary skill in the art will understand that the considerations described herein and further exemplified with reference to specific excipients are equally applicable to any type and combination of excipients, such as, for example, salts, amino acids, other isotonicity agents, surfactants, stabilizers, bulking agents, cryoprotectants, lyoprotectants, antioxidants, metal ions, chelating agents, and / or preservatives.

[0064] Additionally, when a particular excipient in a formulation is reported, for example, as a percentage (%) weight / volume, one of skill in the art will recognize that the equivalent molar concentration of that excipient is also contemplated.

[0065] Those skilled in the art will recognize that the concentrations of the above excipients share interdependencies within a particular formulation. As an example, if the protein / peptide concentration is high, or if the concentration of stabilizer is high, the concentration of bulking agent may be reduced. Furthermore, those skilled in the art will recognize that to maintain the isotonicity of a particular formulation without bulking agent, the concentration of stabilizer may be adjusted accordingly (i.e., a "tonicifying" amount of stabilizer is used).

[0066] Buffer The pH of a solution affects the chemical integrity of the amino acid residues of a protein (e.g., deamidation of Asn and oxidation of Met) and the maintenance of its higher-order structure. Therefore, those skilled in the art use buffers to control the pH of the solution and optimize the stability of the protein. The maximum stability of a protein drug is usually within a narrow pH range. Several approaches are useful for this purpose, such as accelerated stability studies and calorimetric screening studies (Remmele et al. (1999), Biochemistry, 38(16):5241-7). Once the formulation is complete, the preparation must be manufactured and maintained within a given specification throughout its shelf life. Therefore, buffers are almost always used to control the pH in the formulation.

[0067] Organic acids, phosphates and Tris have been commonly used as buffers in protein formulations (see Table 3). The buffering capacity of a buffering species is greatest when the pH is equal to the pKa and decreases as the pH increases or decreases from this value. Ninety percent of the buffering capacity is within one pH unit of its pKa. Buffering capacity also increases proportionally as the buffering agent concentration increases.

[0068] [Table 5]

[0069] In addition to the above, some therapeutic proteins can be self-buffering at pharma- ceutically relevant concentrations.The formulation of such proteins may not contain any traditional buffering agents.See US Patent Application No. 2012 / 0028877, which is incorporated herein by reference.

[0070] Sugars and carbohydrates Sugars are frequently used to stabilize proteins in both liquid and lyophilized formulations. Disaccharides such as sucrose and trehalose are believed to stabilize proteins by selective hydration at high concentrations in the liquid state and by specific interactions with proteins and formation of a viscous glassy matrix in the solid state. Sugar molecules can increase the viscosity of monoclonal antibody solutions, presumably through a selective hydration mechanism. Sugar alcohols such as sorbitol can stabilize proteins in solution and in the lyophilized state. Mannitol is often used as a bulking agent in lyophilized formulations. Lactose is used as a carrier molecule in aspirated formulations of proteins. Cyclodextrin derivatives can stabilize proteins in liquid formulations of antibodies, vaccine antigens, smaller proteins such as growth factors, interleukin-2, and insulin.

[0071] Stabilizers and fillers Bulking agents are typically used in lyophilization formulations to improve product elegance and prevent blowout. Conditions in the formulation are generally designed so that the bulking agent crystallizes from the frozen amorphous phase (either during freezing or annealing above Tg') to impart cake structure and bulk. Mannitol and glycine are examples of commonly used bulking agents.

[0072] Stabilizers include classes of compounds that can act as cryoprotectants, lyoprotectants, and glass formers. Cryoprotectants act to stabilize proteins in the frozen state during freezing or at low temperatures (P. Cameron, ed., Good Pharmaceutical Freeze-Drying Practice, Interpharm Press, Inc., Buffalo Grove, IL, (1997)). Lyoprotectants stabilize proteins in freeze-dried solid dosage forms by preserving the native-like conformational properties of proteins during the dehydration step of freeze-drying. Glass state properties have been classified as "robust" or "fragile" depending on their relaxation properties as a function of temperature. To provide stability, it is important that cryoprotectants, lyoprotectants, and glass formers remain in the same phase as the protein. Sugars, polymers, and polyols fall into this category and can in some cases fulfill all three roles.

[0073] Polyols encompass a class of excipients that includes sugars (e.g., mannitol, sucrose, sorbitol) and other polyhydric alcohols (e.g., glycerol and propylene glycol). Polyethylene glycol (PEG) polymers are included in this category. Polyols are commonly used as stabilizing excipients and / or tonicity agents in both liquid and lyophilized parenteral protein formulations. In the case of the Hofmeister series, the polyols are kosmotropic and selectively excluded from the protein surface. Polyols can protect proteins from both physical and chemical degradation pathways. The energetically most favorable protein conformations are those with the smallest surface areas, since the selectively excluded cosolvents increase the effective surface tension of the solvent at the protein interface.

[0074] Mannitol is a common bulking agent in lyophilized formulations because it crystallizes from the amorphous protein phase during freeze-drying to provide structural stability to the cake (e.g., Leukine®, Enbrel®-Lyo, Betaseron®). It is generally used in combination with a cryoprotectant and / or lyoprotectant such as sucrose. Because mannitol has a tendency to crystallize under freezing conditions, sorbitol and sucrose are the preferred tonicity / stabilizers in liquid formulations to protect the product from the freeze-thaw stresses encountered during shipping or when freezing the bulk prior to manufacture. Sorbitol and sucrose are much more resistant to crystallization and therefore less likely to phase separate from the protein. It is interesting to note that while mannitol is used in isotonic amounts in several commercial liquid formulations such as Actimmune®, Forteo®, and Rebif®, these drugs carry a "do not freeze" warning on their product labels. Reducing sugars (containing free aldehyde or ketone groups), such as glucose and lactose, should be avoided because they can react with and glycate surface lysine and arginine residues of proteins via the Maillard reaction of aldehydes with primary amines (Chevalier F, et al., Nahrung, 46(2):58-63(2002); Humeny A, et al., J Agric Food Chem. 50(7):2153-60(2002)). Sucrose can be hydrolyzed under acidic conditions to fructose and glucose (Kautz CF and Robinson AL, JACS, 50(4)1022-30(1928)), resulting in glycation.

[0075] In certain embodiments of the compositions of the present invention, a stabilizer (or a combination of stabilizers) is added to the lyophilized formulation to prevent or reduce lyophilization-induced or storage-induced aggregation and chemical degradation. A cloudy or opaque solution upon reconstitution indicates that the protein has precipitated. The term "stabilizer" refers to an excipient that can prevent aggregation or other physical and chemical degradation (e.g., autolysis, deamidation, oxidation, etc.) in aqueous and solid states. Stabilizers conventionally used in pharmaceutical compositions include, but are not limited to, sucrose, trehalose, mannose, maltose, lactose, glucose, raffinose, cellobiose, gentiobiose, isomaltose, arabinose, glucosamine, fructose, mannitol, sorbitol, glycine, arginine HCL, dextran, starch, hydroxyethyl starch, cyclodextrins, N-methylpyrollidene, polyhydroxy compounds including polysaccharides such as cellulose and hyaluronic acid, sodium chloride (Carpenter et al. (1991), Develop. Biol. Standard 74:225).

[0076] Osmolyte Osmolytes commonly used as excipients in protein formulations are listed in Table 2. Other osmolytes commonly found in nature that may be useful as excipients include taurine, betaine, trimethylamine, N-oxide (TMAO), choline-O-sulfate, and sarcosine.

[0077] Proteins and Polymers Protein-based excipients add complexity to formulations, especially in the development of analytical methods to monitor the stability of protein-based drugs or vaccines in the presence of protein-based excipients. Polymers have been evaluated as excipients (e.g., bulking agents) in lyophilized protein formulations. Controlled release formulations of protein drugs and vaccines have been investigated in which proteins are formulated with polymers such as polylactic-co-glycolic acid (PLGA) and polyethylene glycol (PEG). Many additional water-soluble polymers (e.g., hydroxyethylcellulose (HEC), carboxymethylcellulose (CMC)) have been used for topical formulation of protein drugs.

[0078] PEG can stabilize proteins by two different temperature-dependent mechanisms. It has been shown that at low temperatures it is selectively excluded from protein surfaces, whereas at high temperatures it interacts with the unfolded forms of proteins due to its amphiphilic nature (Lee and Lee (1987), Biochemistry, 26(24):7813-9). At low temperatures it can protect proteins by selective exclusion, whereas at high temperatures it can protect proteins, presumably by reducing the number of productive collisions between unfolded molecules. PEG is also a cryoprotectant and has been used in Recombinate®, a lyophilized formulation of recombinant antihemophilic factor.

[0079] Antioxidants Many different causes can cause protein residues to become oxidized. Oxidative protein damage can be minimized by careful control of manufacturing processes and product storage, including factors such as atmospheric oxygen, temperature, light exposure, and chemical contamination. When such controls are inadequate, antioxidant excipients can be included in the formulation.

[0080] The most commonly used pharmaceutical antioxidant excipients are reducing agents, oxygen / free radical scavengers, or chelating agents. Antioxidants in therapeutic protein formulations must be water soluble and must remain active throughout the shelf life of the product. Reducing agents and oxygen / free radical scavengers function by scavenging reactive oxygen species in solution. Chelating agents (e.g., EDTA) may exert their effect by binding trace metal impurities that promote free radical formation. For example, in liquid formulations of acidic fibroblast growth factor, EDTA inhibits the oxidation of cysteine ​​residues catalyzed by metal ions. EDTA is used in commercial products such as Kineret® and Ontak®.

[0081] Metal ions Generally, transition metal ions are undesirable in protein formulations because they can catalyze physical and chemical decomposition reactions in proteins. However, if a particular metal ion functions as a cofactor for the protein, it is included in the formulation. If the metal ion forms a coordination complex, it can also be used in suspension formulations of proteins (e.g., zinc suspension of insulin). The use of magnesium ions (10-120 mM) to inhibit the isomerization of aspartic acid to isoaspartic acid has been proposed (WO 2004 / 039337).

[0082] Metal ions have been found to confer stability and / or increased activity in formulations of human deoxyribonuclease (rhDNase, Pulmozyme®). +2 ions (up to 100 mM) increased the stability of the enzyme through specific binding sites (Chen et al. (1999), J Pharm Sci. 88(4):477-82). Indeed, removal of calcium ions from solution by EGTA led to increased deamidation and aggregation. However, this effect was not observed in the presence of Ca +2 ions, and other divalent cations (Mg +2 , Mn+2 , and Zn +2 ) was observed to destabilize rhDNase.

[0083] A similar effect was observed in the formulation of factor VIII. +2 and Sr +2 ions stabilized the protein, while Mg +2 , Mn +2 , and Zn +2 , Cu +2 , and Fe +2 Others, such as Fatouros, et al. (1997), Int. J. Pharm., 155, 121-131, destabilized it. In another study with factor VIII, Al +3 A significant increase in the rate of aggregation was observed in the presence of ions (Derrick et al. (2004), J. Pharm. Sci., 93(10):2549-57). The authors noted that other excipients, such as buffer salts, were not affected by Al. +3 It has been noted that ions are often contaminating, illustrating the need for using excipients of appropriate quality in formulated products. Vaccines containing live or dead attenuated picornaviruses, such as Hepatitis A and polio, are conformationally stabilized by magnesium. Metal ions such as calcium, magnesium, and zinc improve the stability of oxytocin in aqueous solutions. Insulin can bind to zinc, resulting in the formation of crystalline forms of dimers and hexamers, which allows the preparation of different formulations with different in vivo release profiles. The chemical and thermal stability of hexameric insulin formulations changes in the presence of different concentrations of zinc and phenol.

[0084] Specific Ligand One approach to improve the conformational stability of protein therapeutics is to exploit the protein's inherent ligand binding sites. For example, Pulmozyme® not only requires divalent metal ions for its enzymatic activity, but it has improved conformational stability in the presence of calcium ions. Both acidic and basic fibroblast growth factors (aFGF and bFGF) have been clinically evaluated for their ability to promote wound healing, but both proteins naturally bind to strongly negatively charged proteoglycans on the cell surface. A variety of other strongly negatively charged compounds also bind to aFGF, dramatically stabilizing it through interactions with the protein's polyanion binding sites.

[0085] Surfactants Protein molecules have a high tendency to interact with surfaces and are therefore prone to adsorption and denaturation at air-liquid, vial-liquid, and liquid-liquid (silicone oil) interfaces. This degradation pathway is inversely dependent on protein concentration and results in soluble and insoluble protein aggregates, or loss of protein from solution due to adsorption to surfaces. In addition to container surface adsorption, surface-induced degradation is exacerbated by physical agitation, such as that experienced during transport and handling.

[0086] Surfactants are commonly used in protein formulations to prevent surface-induced degradation. Surfactants are amphiphilic molecules that have the ability to outcompete proteins for interfacial locations. The hydrophobic portion of the surfactant molecule occupies the interfacial location (e.g., air / liquid), while the hydrophilic portion of the molecule remains oriented toward the bulk solvent. At sufficient concentrations (typically near the detergent's critical micelle concentration), a surface layer of surfactant molecules serves to prevent protein molecules from adsorbing at the interface. This minimizes surface-induced degradation.

[0087] The most commonly used surfactants are the non-ionic fatty acid esters of sorbitan polyethoxylates, namely polysorbate 20 and polysorbate 80 (e.g., in the formulations Avonex®, Neupogen®, Neulasta®). The two differ only in the length of the aliphatic chain, which confers hydrophobic properties to the molecules C-12 and C-18, respectively. Polysorbate 80 is more surface active than polysorbate 20 and has a lower critical micelle concentration. Both polysorbate 20 and polysorbate 80 have been shown to protect against agitation-induced aggregation. Polysorbate 20 and polysorbate 80 also protect against stress induced by freezing, lyophilization, and reconstitution. Both polysorbates 20 and 80 can contain peroxides that can oxidize proteins and themselves can decompose by either oxidation or hydrolysis, with varying effects on protein stability. Controlling the concentration of polysorbate 20 or 80 in a formulation can also be difficult due to its complex behavior during membrane filtration (especially at concentrations where polysorbates form micelles in solution). The surfactant Poloxamer 188 has also been used in several commercial liquid products such as Gonal-F®, Norditropin® and Ovidrel®. In general, non-ionic surfactants are believed to stabilize proteins by outcompeting protein molecules primarily for hydrophobic surfaces (e.g., air-water interfaces), thereby preventing the protein from unfolding at these hydrophobic interfaces. Non-ionic surfactants can also prevent protein molecules from adsorbing to other hydrophobic surfaces present during processing. In addition, non-ionic surfactants can directly interact with hydrophobic regions in protein molecules. Monoclonal antibodies can affect the critical micelle concentration of polysorbate 20 compared to buffers alone.

[0088] Detergents can also affect the thermodynamic conformational stability of proteins. Again, the effect of a given excipient is protein specific. For example, polysorbates have been shown to decrease the stability of some proteins and increase the stability of others. Detergent protein destabilization makes sense in terms of the hydrophobic tails of the detergent molecules that can be involved in specific binding to partially or totally unfolded protein states. These types of interactions can cause a shift in the conformational equilibrium towards more extended protein states (i.e., increasing the exposure of hydrophobic parts of the protein molecule, complementing the binding polysorbates). Alternatively, if the native state of a protein exhibits several hydrophobic surfaces, detergents binding to the native state can stabilize that conformation.

[0089] Another aspect of polysorbates is that they are inherently prone to oxidative degradation. They often contain sufficient amounts of peroxide as raw material to cause oxidation of protein residue side chains, especially methionine. Due to the possibility of oxidative damage resulting from the addition of stabilizers, it is emphasized that the lowest effective concentration of excipients should be used in the formulation. In the case of surfactants, the effective concentration for a given protein is determined by the stabilization mechanism. If the stabilization mechanism of the surfactant is related to the prevention of surface denaturation, it is assumed that the effective concentration will be near the critical micelle concentration of the detergent. Conversely, if the stabilization mechanism is related to specific protein-detergent interactions, the effective surfactant concentration is related to the protein concentration and the stoichiometry of the interaction (Randolph et al. (2002), Pharm Biotechnol., 13:159-75).

[0090] Surfactants may also be added in appropriate amounts to prevent surface-related aggregation phenomena during freezing and drying (Chang (1996), J. Pharm. Sci. 85:1325). Exemplary surfactants include anionic, cationic, nonionic, zwitterionic, and amphoteric surfactants, including those derived from natural amino acids. Anionic surfactants include, but are not limited to, sodium lauryl sulfate, dioctyl sodium sulfosuccinate and dioctyl sodium sulfonate, chenodeoxycholic acid, N-lauroyl sarcosine sodium salt, lithium dodecyl sulfate, 1-octanesulfonic acid sodium salt, sodium cholate hydrate, sodium deoxycholate, and glycodeoxycholic acid sodium salt. Cationic surfactants include, but are not limited to, benzalkonium chloride or benzethonium chloride, cetylpyridinium chloride monohydrate, and hexadecyltrimethylammonium bromide. Zwitterionic surfactants include, but are not limited to, CHAPS, CHAPSO, SB3-10, and SB3-12. Nonionic surfactants include, but are not limited to, digitonin, Triton X-100, Triton X-114, TWEEN-20, and TWEEN-80. In another embodiment, surfactants include lauromacrogol 400, polyoxyl 40 stearate, polyoxyethylene hydrogenated castor oil 10, 40, 50, and 60, glycerol monostearate, polysorbate 40, 60, 65, and 80, soy lecithin and other phospholipids such as DOPC, DMPG, DMPC, and DOPG, sucrose fatty acid esters, methylcellulose, and carboxymethylcellulose.

[0091] salt Salts are often added to increase the ionic strength of the formulation, which can be important for protein solubility, physical stability, and isotonicity. Salts can affect the physical stability of proteins in a variety of ways. Ions can stabilize the native state of proteins by binding to charged residues on the protein surface. Alternatively, they can stabilize the denatured state by binding to peptide groups (-CONH-) along the protein backbone. Salts can also stabilize the native conformation of proteins by shielding repulsive electrostatic interactions between residues within the protein molecule. Electrolytes in protein formulations can also shield attractive electrostatic interactions between protein molecules that can lead to protein aggregation and insolubility.

[0092] The effect of salts on protein stability and solubility varies greatly depending on the properties of the ionic species. The Hofmeister series was devised in the 1880s as a way to rank electrolytes based on their ability to precipitate proteins (Cacace et al. (1997), Quarterly Reviews of Biophysics, 30(3):241-277). In this report, the Hofmeister series is used to represent the magnitude of protein stabilizing effects of ionic and nonionic cosolutes. In Table C, the cosolutes are listed in order from stabilizing (cosmotropic) to destabilizing (chaotropic) in relation to their general effect on proteins in solution. In general, the difference in effect between anions is much greater than that observed for cations, and for both species, the effect is most evident at concentrations higher than can be tolerated in parenteral formulations. High concentrations of kosmotropes (e.g., >1 molar ammonium sulfate) are commonly used to precipitate proteins from solution by a process called "salting out," which selectively excludes the kosmotrope from the protein surface and reduces the solubility of the protein in its native (folded) conformation. Removal or dilution of the salt restores the protein to solution. The term "salting in" refers to the use of destabilizing ions (e.g., guanidine and chloride) that improve protein solubility by solvating the peptide bonds of the protein backbone. Increasing concentrations of chaotropes favor the denatured (unfolded) state conformation of the protein as the solubility of the peptide chains increases. The relative effects of ions on "salting in" and "salting out" define the ion's position in the Hofmeister series.

[0093] [Table 6]

[0094] To maintain isotonicity in parenteral formulations, salt concentrations are generally limited to less than 150 mM for combinations of monovalent ions. In this concentration range, the mechanism of salt stabilization is likely due to screening of electrostatic repulsive intramolecular forces or attractive intermolecular forces (Debye-Huckel screening). Interestingly, chaotropic salts have been shown to be more effective at stabilizing protein structure by this mechanism than kosmotropes of similar concentrations. Chaotropic anions are believed to bind more strongly than kosmotropic ions. In the case of covalent protein degradation, the Debye-Huckel theory predicts that the effect of ionic strength on this mechanism will be different. Thus, published reports of protein stabilization by sodium chloride are accompanied by reports that sodium chloride accelerated covalent degradation. The mechanism by which salts affect protein stability is protein specific and can vary significantly as a function of solution pH. An example of an excipient that may be useful in enabling the delivery of protein drugs is the excipient of some high concentration antibody formulations. Over the past few years, salts have been shown to be effective in reducing the viscosity of such formulations (Liu et al. (2005, 2006), J. Pharm Sci., 94(9):1928-40, erratum in J Pharm Sci., 95(1):234-5).

[0095] Preservatives Preservatives are necessary when developing multi-use parenteral formulations that involve more than one withdrawal from the same container. Their primary function is to inhibit microbial growth and ensure sterility of the product over the shelf life or use period of the formulation. Commonly used preservatives include benzyl alcohol, phenol, and m-cresol. Although preservatives have a long history of use, developing protein formulations containing preservatives can be difficult. Preservatives almost always have a destabilizing effect on proteins (aggregation), which is a major factor limiting their use in multi-dose protein formulations (Roy et al. (2005), J. Pharm. Sci., 94(2):382-96). Benzyl alcohol has also been shown to affect protein structure and stability in a concentration-, temperature-, and time-dependent manner. Due to these destabilizing effects, many lyophilized protein formulations are reconstituted with diluents that contain benzyl alcohol to minimize the time of contact with the protein before administration.

[0096] Most protein drugs are formulated for single use only. However, if a multi-dose formulation is possible, it offers the added benefit of patient convenience and increased marketability. Human growth hormone (hGH) is a good example, where the development of a preservative formulation has led to the commercialization of a more convenient multi-use injection pen. At least four such pen devices containing preservative formulations of hGH are currently available. Norditropin® (liquid), Nutropin AQ® (liquid), and Genotropin (lyophilized-dual chamber cartridge) contain phenol, while Somatrope® contains m-cresol.

[0097] During formulation development of a preservative dosage form, several aspects must be considered. The effective preservative concentration in the formulation must be optimized. This requires testing a given preservative in the dosage form at a range of concentrations that provide antimicrobial effect without compromising protein stability. For example, three preservatives were successfully screened using differential scanning calorimetry (DSC) in the development of a liquid formulation for interleukin-1 receptor (type I). The preservatives were ranked based on their effect on stability at concentrations commonly used in commercial products (Remmele et al. (1998), Pharm. Res., 15(2):200-8).

[0098] As might be expected, the development of liquid formulations containing preservatives is more challenging than lyophilized formulations. Freeze-dried products can be lyophilized without preservatives and reconstituted with a preservative containing diluent at the time of use. This reduces the time the preservative is in contact with the protein and greatly minimizes the associated stability risks. For liquid formulations, the efficacy and stability of the preservative must be maintained over the entire shelf life of the product, which is usually about 18-24 months. It is important to note that the efficacy of the preservative needs to be demonstrated in the final formulation containing the active drug and all excipient components.

[0099] Some preservatives may cause injection site reactions, which is another factor that needs to be considered when selecting a preservative. In a clinical trial focused on the evaluation of preservatives and buffers in Norditropin®, it was observed that the perception of pain was lower in formulations containing phenol and benzyl alcohol compared to those containing m-cresol (Kappelgaard (2004), Horm. Res. 62 Suppl 3:98-103). Interestingly, among the commonly used preservatives, benzyl alcohol has anesthetic properties (Minogue and Sun (2005), Anesth. Analg. 100(3):683-6). EXAMPLES

[0100] All publications, patents and patent applications discussed and cited herein are incorporated herein by reference in their entirety. It is understood that the disclosed invention is not limited to the specific methodology, protocols and materials described, which may vary. It is also understood that the terms used herein are for the purpose of describing specific embodiments only, and are not intended to limit the scope of the appended claims.

[0101] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein which equivalents are intended to be encompassed by the following claims.

[0102] Example 1: This experiment demonstrates that by adjusting the loading weight target, the protein concentration can be precisely targeted for each reconstituted formulation.

[0103] material A buffer containing mannitol, sucrose, L-histidine, and polysorbate 20 at pH 5 · Container vial, 3cc, I-type glass, untreated, 13mm finish with stopper, 13mm, 4432 / 50 V-50, Romiplostim filtered bulk

[0104] method 1. Dilute the drug to the target product formulation (0.5 mg / mL) with the required amount of dilution buffer. 2. Filter the blended solution using a 0.22 μm polyvinylidene fluoride (PVDF) filter. 3. Vials and stoppers: Ensure that 3 cc vials are cleaned and depyrogenated. 4. Fill a sufficient number of vials to the respective fill weight targets; 0.307, 0.322, 0.342, 0.357, 0.373 g. 5. Partially stopper the vials and place in the lyophilizer. 6. Perform the required lyophilization cycle with appropriate freezing, vacuum, primary and secondary drying, followed by stoppering and removal of the lyophilized product in sealed vials. 7. Reconstitute the product with the stated reconstitution volume of 0.32 mL of water for injection. 8. Measure the concentration of protein obtained in the vial using ultraviolet (UV) absorption. Absorbance is defined as the amount of light of a particular wavelength that is absorbed as it passes through the sample. An absorbance unit is a function of absorbance specific to the molecule, its concentration, and the path length of the sample. The aromatic amino acids phenylalanine, tyrosine, and tryptophan in protein molecules absorb light in the UV range of 260-290 nm. UV absorbance in this range is commonly used to measure the presence of protein in a solution. 9. Determine the osmolality of the product using the freezing point depression method. 10. Analyses are performed to determine the effect of adjusted fill weight targets on both protein and osmolality.

[0105] [Table 7]

[0106] The R of 0.9964 shown in Figure 6 2 A strong linear relationship was observed between reconstituted protein concentration and loading capacity, as shown by: This linear relationship can be rationalized based on the protein mass balance shown below. C 再構成 * V 再構成 =(C 配合 -C 喪失 ) * V 充填 [Equation 1]

number

[0107] C再構成 If V is the protein concentration after reconstitution, 再構成 is the product volume after reconstitution, and C 配合 is the concentration of the formulated bulk protein, C 喪失 is the loss of protein due to binding to the filter and vessel, and V 充填 is the fill volume in each vial. Because the formulation bulk concentration, protein binding losses, and reconstitution volume are constant for a given run, the protein concentration after reconstitution is proportional to the fill volume as shown in Equation 2.

[0108] In pilot-scale experiments, 10 replicates for each loading condition were used to measure the variability in protein concentration of the final reconstituted formulation (DP), as shown in Table 6. The variability observed therefore represents both process and analytical variability, including the variability associated with reconstituting the product.

[0109] [Table 8]

[0110] Effect of filling weight on osmolality (reconstituted DP) The osmolality of the final formulation after reconstitution was tested relative to the osmolality of vials filled at five different fill weight targets. The results of the osmolality before filling and after reconstitution are summarized in Table 7. Based on the results at the target fill weight of 0.341 grams, the osmolality did not change after filling and reconstitution, indicating that no measurable amount of excipients was lost. When the reconstitution volume is kept constant, the osmolality increases slightly at higher fill volumes and decreases slightly at lower fill volumes.

[0111] [Table 9]

[0112] A linear relationship was found when fill weight / volume (normalized based on a target of 0.341 g) was plotted against product osmolality (normalized based on a target osmolality of 313 mOsm / kg at the target fill weight) as shown in Figure 7. This linear relationship can be explained by the definition of osmolality and its relationship to buffer component concentrations.

[0113] Osmolality is a measure of solute concentration and is defined as the number of osmoles (Osm) per kilogram of solvent (osmoles / kg or Osm / kg). Unlike molarity (moles / L), osmolality measures moles of solute particles (such as dissociated ions) rather than moles of solute. The osmolality of a solution can be calculated according to the following formula: Osmolality (osm / kg) = density * Σφ i n i C i [Equation 3] where φ is the osmotic coefficient, n is the number of particles (e.g., ions) into which the molecule dissociates, and C is the molar concentration of the solute (moles / L).

[0114] If the fill weight (or volume) is 10% higher than the target volume, the concentration of each excipient type in the formulation will increase by 10% when reconstituted to a constant volume. As shown in Equation 3, φ i and n i Since is constant, a 10% increase in the concentration of each species results in a 10% increase in osmolality. Similarly, a 10% decrease in loading volume results in a 10% decrease in excipient concentration and, consequently, a 10% decrease in osmolality.

[0115] Example 2 The protein blinatumomab has 55 μg / mL formulated in 200 mM L-lysine-HCl, 25 mM citric acid, 15% (w / v) trehalose dihydrate, 0.1% (w / v) polysorbate 80 at pH 7.0. Measure the acceptable range of formulated protein at the filtration bulk stage. Use bulk concentrations ranging from 48.0 μg / mL to 65.0 μg / mL. Based on the measured protein concentration, calculate the target fill weight to target a reconstituted formulation of 12.5 mcg / mL when reconstituted with 3 mL of water. C 再構成 * V 再構成 =(C 配合 -C 喪失 ) * V 充填 During the ceremony, C 再構成 * V 再構成 = target protein content 再構成 It is.

[0116] The target protein content may then be multiplied by the product density and divided by the measured drug concentration (adjusted for binding losses if necessary) to determine the target fill weight.

[0117] The target fill weight is calculated according to the following formula using the corresponding fill weight range (0.634-0.858gm):

number

number

[0118] Example 3 The infliximab formulation has 20±1.5 mg / mL infliximab, formulated with 10 mM sodium phosphate, 10% (w / v) sucrose, 0.01% (w / v) polysorbate 80, and has a pH of 7.2 after reconstitution with 10 mL of water for injection.

[0119] The target fill weight is calculated according to the following formula using the corresponding fill weight range (4.85-5.63 gm):

[0120] Target fill weight calculation

number

[0121] Example 4 Trastuzumab has 21 mg / mL, pH 6.1, formulated with 0.303 mg / mL L-histidine, 0.470 mg / mL L-histidine hydrochloride monohydrate, 19.1 mg / mL α,α-trehalose dihydrate, 0.0840 mg / mL polysorbate 20. The target fill weight is calculated according to the following formula with various fill weight ranges based on product delivery requirements. The fill weight targets range from 3.1 to 21.2 gm (varies depending on the respective presentation). This product has multiple compositions with a pooled drug substance concentration of 21 mg / mL.

[0122] The target fill weight for each formulation lot in this Example 4 and subsequent examples is calculated using the following information:

number

[0123] As an example,

number

[0124] As another example,

number

[0125] As a further example,

number

[0126] Example 5 AMG701 is a single chain variable domain bispecific T cell engager (BiTE®) anti-BCMA / anti-CD3 antibody construct (see NCI Drug Dictionary and other references). AMG701 with a protein concentration of 1 mg / mL was formulated with 10 mM L-glutamic acid, 9.0% (w / v) sucrose, 0.010% (w / v) polysorbate 80 at pH 4.2. The target fill weight is calculated according to the following formula with various fill weight ranges based on product delivery requirements. AMG701 has three representations, the fill weight target range for the first representation is 0.47-0.57 gm, for the second representation is 1.60-1.96 gm, and for the third representation is 3.28-4.01 gm.

[0127] For the first submission:

number

[0128] For the second submission:

number

[0129] For the third submission:

number

[0130] Example 6 AMG330 is an anti-CD33 / anti-CD3 single chain variable domain bispecific T cell engager (BiTE®) (see NCI Drug Dictionary and other references). AMG300 with a protein concentration of 0.5 mg / mL was formulated with 10 mM potassium phosphate, 8.0% (w / v) sucrose, 1.0% (w / v) sulfobutylether-β-cyclodextrin (SBE-CD), 0.010% (w / v) polysorbate 80 at pH 6.1. The target fill weight is calculated according to the following formula with various fill weight ranges based on product delivery requirements. The fill weight target is in the range of 1.2-1.5 gm. The target fill weight is calculated as follows:

number

[0131] In general, the methodology can be used when a target amount of product is desired in a container to ensure that the product after reconstitution is of a required concentration. This result is achieved by measuring the amount of reconstituted product volume: for example, a 1 mL volume containing a desired concentration of therapeutic protein of 1 mg / mL will yield a total amount of 1 mg of required protein content. Target protein content [mg] = (protein concentration [mg / mL] × reconstitution volume [mL])

[0132] The target protein content is then used to calculate the target fill weight based on the following formula: In-process, or measured, drug concentration is typically measured as part of the formulation process to account for any process variability. In some cases, adjustments to the target protein content are made to account for product loss due to conjugation. Reliability of the measured drug concentration is necessary to ensure accurate targeting of the fill weight.

number

[0133] This is a particular example because the therapeutic agent for both the in-process concentrated and reconstituted products can range from micrograms (mcg or μg) per milliliter (mL) to milligrams (mg) per milliliter (mL).

[0134] Validation of osmolality as mentioned above is required (based on experimental findings). In order to achieve the required amount of active agent filled into the container and ensure that the reconstituted product meets both the product (active agent or protein) concentration as well as osmolality specification limits, limits of the allowed measured drug concentration are established in combination with the fill weight target.

[0135] All publications, patents and patent applications discussed and cited herein are incorporated herein by reference in their entirety. It is understood that the disclosed invention is not limited to the specific methodology, protocols and materials described, which may vary. It is also understood that the terms used herein are for the purpose of describing specific embodiments only, and are not intended to limit the scope of the appended claims.

[0136] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein which equivalents are intended to be encompassed by the following claims.

Claims

1. 1. A process for producing a lyophilized pharmaceutical formulation of a therapeutic protein, comprising: a. providing a bulk amount of said therapeutic protein formulation; b. determining the concentration of said therapeutic protein in a bulk formulation; c. adjusting the loading weight of the protein in the bulk formulation to obtain a fixed dose of the protein; d. lyophilizing the adjusted protein fill weight formulation to obtain a final formulation in a container. Including, A process wherein the concentration of the product after reconstitution with a fixed volume is within a predetermined acceptable range.

2. The adjusted loading weight of the protein is calculated according to the formula: [0010] The process of claim 1 , wherein the calculated value is calculated according to:

3. 10. The process of claim 1, wherein the concentration of the protein in the final formulation is about 20 mg / mL or less.

4. 2. The process of claim 1, wherein the therapeutic protein is selected from romiplostim, blinatumomab, infliximab, trastuzumab, AMG 701, and AMG 330.

5. 2. The process of claim 1, wherein the therapeutic protein is romiplostim and the concentration of the therapeutic protein in the final formulation is about 0.5 mg / mL.

6. 10. The process of claim 1, wherein the final formulation comprises about 0.5 mg / mL romiplostim in about 10 mM histidine, about 4% w / v mannitol, about 2% w / v sucrose, and about 0.004% polysorbate 20 at about pH 5.

0.

7. 2. The process of claim 1, wherein the therapeutic protein is blinatumomab and the concentration of the therapeutic protein in the final formulation is about 55 μg / mL.

8. 2. The process of claim 1, wherein the formulation comprises about 55 μg / mL of blinatumomab in about 25 mM citric acid monohydrate, about 15% (w / v) trehalose, about 200 mM L-lysine hydrochloride, and about 0.1% (w / v) polysorbate 80 at about pH 7.

0.

9. 2. The process of claim 1, wherein the therapeutic protein is infliximab and the concentration of the therapeutic protein in the final formulation is about 20±1.5 mg / mL.

10. 2. The process of claim 1, wherein the final formulation comprises about 20±1.5 mg / mL infliximab, about 10 mM sodium phosphate, about 10% (w / v) sucrose, and about 0.01% (w / v) polysorbate 80 at about pH 7.

2.

11. 2. The process of claim 1, wherein the therapeutic protein is trastuzumab and the concentration of the therapeutic protein in the final formulation is about 21 mg / mL.

12. 2. The process of claim 1, wherein the final formulation comprises about 21 mg / mL trastuzumab, about 0.303 mg / mL L-histidine, about 0.470 mg / mL L-histidine hydrochloride monohydrate, about 19.1 mg / mL α,α-trehalose dihydrate, and 0.0840 mg / mL polysorbate 20 at about pH 6.

1.

13. 2. The process of claim 1, wherein the therapeutic protein is AMG701 and the concentration of the therapeutic protein in the final formulation is about 1 mg / mL.

14. 2. The process of claim 1, wherein the final formulation comprises about 1 mg / mL AMG 701, about 10 mM L-glutamic acid, about 9.0% (w / v) sucrose, and about 0.010% (w / v) polysorbate 80 at about pH 4.

2.

15. 2. The process of claim 1, wherein the therapeutic protein is AMG330 and the concentration of the therapeutic protein in the final formulation is about 0.5 mg / mL.

16. 2. The process of claim 1, wherein the final formulation comprises about 0.5 mg / mL AMG 330, about 10 mM potassium phosphate, about 8.0% (w / v) sucrose, about 1.0% (w / v) sulfobutylether-β-cyclodextrin (SBE-CD), and about 0.010% (w / v) polysorbate 80 at about pH 6.

1.

17. 10. The process of claim 1, wherein the concentration of the protein in the final formulation is about 25 mg / mL or less.

18. The process of claim 1 , wherein the therapeutic protein is a bispecific single chain antibody construct.

19. 2. The process of claim 1, wherein the therapeutic protein is selected from AMG701 and AMG330.

Images