Method for filling vials containing liquid formulations

Abstract

The present invention relates to the pharmaceutical field, in particular to a method relating to the filling of containers containing liquid formulations in vials.

Description

[Technical field]

[0001] The present invention relates to the pharmaceutical field, in particular to a method relating to the filling of containers containing liquid formulations in vials. [Background technology]

[0002] When filling containers for liquid and lyophilized products, a slight excess volume - excess volume or overfill volume (V overfill ) - also called the target volume (label claim volume or label indicated volume (V label (also called ) can be extracted / withdrawn and dispensed (USP <1> , U.S.P. <1151> , U.S.P. <697> , Ph.Eur. monograph 2.9.17). Overfilling is necessary because losses occur at various levels, such as the container closure system and the withdrawal device, leaving a residual volume. Said overfill volume must be determined for each given drug product and for each presentation (formulation, vial size, etc.) very early in the development of said drug product. A test to estimate the extractable volume is proposed in the European Pharmacopoeia (Ph.Eur. monograph 2.9.17, published July 2019).

[0003] Excess volume should be minimized whenever possible to prevent unsafe handling (FDA Guidance for Industry, 2015) and limit drug product waste (Gotham et al., 2019; Hatswell et al., 2019). Excess volume includes hold-up volume (V) in vials, withdrawal syringes, and needles. HU), it is recommended to define it based on the variability of the filling line as well as the variability of the extractable volume test method (Manger 2019). In practice, the excess volume for a given formulation is calculated as a multiple of the standard deviation obtained from historical data (Sethuraman et al., 2010; Joglekar, 2010; Kruszynski, 2016; Levine, 2017) (Figure 1, Panel A). It is often defined experimentally, taking into account hold-up volume and filling process tolerances (Dixon and Gudinas 2018; Sethuraman et al., 2010), and does not include other sources of variation such as the variability of the extractable volume test method. V HU It has also been proposed to take into account the viscosity and / or specificity of the container to derive regression models that can be used to predict the viscosity (Mehta et al., 2020; Jameel et al., 2015; Akers et al., 2016), and the Health Authorities have proposed recommended excess volumes per type of container (FDA Guidance USP <1151> ; see Table 1).

[0004] However, a concern for regulators is that too much or too little formulation is filled into those vials. Such over- and under-dosing can result in medication errors and may lead to misuse of leftover medication or pooling of vials to obtain a single dose (FDA Guidance for Industry, 2015). Of further concern to regulators is that dose pooling or repeated use of a single vial may increase patient exposure to adverse events, most notably those caused by microbial contamination.

[0005] There remains a need to predict the overfill volume of a liquid formulation required and therefore optimize the filling method. Predictive models can provide an initial guess for the volume range to be tested to support the definition of the overfill volume. Summary of the Invention

[0006] definition "Excess Volume", "Overfill Volume" or "V overfill " refers to a target volume (labeled volume, label claim volume or V label It refers to a slight excess volume of a liquid formulation that is added to a container (such as a glass vial) that ensures that the excess volume (also called excess volume) can be extracted / drawn out. The excess volume is typically expressed in μl or ml.

[0007] "Hold-up volume", "residual volume" or V HU The term "hold-up volume" refers to the remaining volume in the vial and / or withdrawal syringe during an extractable volume test or upon extraction and / or administration of a liquid formulation, and depends, among other things, on the viscosity of the liquid formulation and the neck diameter of the vial. Hold-up volume is typically expressed in μl or ml.

[0008] The term "container" as used herein generally refers to the reservoir suitable for holding formulation in liquid form.The examples of the container that can be used in the present invention include ampule, glass vial, tube, bottle, syringe (such as pre-filled syringe), cartridge, or other such reservoir that is suitable for delivering liquid formulation to patient by injection.

[0009] "Sigma filling" (i.e. σ filling The term ) refers to the fill volume standard deviation / variability of a filling device that is used or to be used to fill containers with a liquid formulation. Fill variability is typically determined by weighing several containers that are then filled at a given target fill weight. This is typically expressed in μl or ml.

[0010] "Sigma analysis" (i.e. σ analysis The term variability / standard deviation of the extractable volume test is typically determined by performing the extractable volume test on several replicate vials. This is typically expressed in μl or ml.

[0011] "Sigma sum" (i.e. σ total The term ) corresponds to the total standard deviation (or total variability) combining the filling standard deviation and the extractable volume test standard deviation. It is typically expressed in μl or ml.

[0012] "RSD filling The term "Relative Filling Standard Deviation (Precision)" corresponds to RSD filling = σ filling / (V label +V HU ) is calculated as follows:

[0013] "RSD analysis The term "RSD" corresponds to the standard deviation / variability of the relative extractable volume test method. analysis = σ analysis / V HU It is calculated as follows:

[0014] The term "k" is a value corresponding to a tolerance factor, i.e. a safety margin set at the discretion of the person skilled in the art performing the calculations, for example a safety margin of 90%, 95%, 99%, etc. It is chosen from a normal distribution table (see http: / / www.z-table.com / ).

[0015] "B filling The term "fill bias" corresponds to the fill bias, i.e. the difference between the target and the actual average fill volume. The fill bias is typically determined by weighing several containers that are then filled at a given target fill weight. This is typically expressed in μl or ml.

[0016] "R.B. filling The term " corresponds to the average relative filling bias, RB filling =B filling / (V label +V HU ) is calculated as follows:

[0017] The term "therapeutic peptide," "therapeutic polypeptide" or "therapeutic protein" refers to a peptide, polypeptide, or a protein such as a cytokine, growth factor, hormone, antibody or fusion protein for therapeutic use. Preferably, the peptide, polypeptide or protein is recombinant, i.e., made by recombinant methods.

[0018] As used herein, the term "antibody" includes, but is not limited to, monoclonal antibodies, polyclonal antibodies and recombinant antibodies produced by recombinant techniques known in the art. "Antibody" includes antibodies of any species, particularly mammalian species; human antibodies of any isotype, including, for example, IgG1, IgG2a, IgG2b, IgG3, IgG4, IgE, IgD, and antibodies produced as dimers of this basic structure, including IgGA1, IgGA2, or pentamers, such as IgM, and modified variants thereof; non-human primate antibodies, such as from chimpanzees, baboons, rhesus monkeys or cynomolgus monkeys; rodent antibodies, such as from mice or rats; rabbit, goat or horse antibodies; camelid antibodies (e.g., from camels or llamas, such as Nanobodies™) and derivatives thereof; avian antibodies, such as chicken antibodies; or fish antibodies, such as shark antibodies. The term "antibody" also refers to a "chimeric" antibody in which at least a first portion of a heavy and / or light chain antibody sequence is derived from a first species and a second portion of a heavy and / or light chain antibody sequence is derived from a second species. Chimeric antibodies of interest herein include "primatized" antibodies that contain variable domain antigen-binding sequences derived from a non-human primate (e.g., Old World monkeys such as baboons, rhesus monkeys, or cynomolgus monkeys) and human constant region sequences. "Humanized" antibodies are chimeric antibodies that contain sequences derived from a non-human antibody. In most cases, humanized antibodies are human antibodies (recipient antibodies) in which residues from the recipient's hypervariable regions are replaced with residues from the hypervariable regions [or complementarity determining regions (CDRs)] of a non-human species (donor antibody) such as mouse, rat, rabbit, chicken, or non-human primate that have the desired specificity, affinity, and activity. In most cases, residues of the human (recipient) antibody are further replaced by corresponding non-human residues outside the CDRs, i.e., within the framework regions (FRs). Furthermore, humanized antibodies may contain residues that are not found in the recipient antibody or in the donor antibody. These modifications are made to further refine the antibody properties. Humanization reduces the immunogenicity of non-human antibodies in humans and thus facilitates the application of antibodies to treat human diseases.Humanized antibodies and several different techniques for producing them are well known in the art. The term "antibody" also refers to human antibodies that can be produced as an alternative to humanization. For example, it is possible to produce transgenic animals (e.g. mice) that, upon immunization, can produce a full repertoire of human antibodies in the absence of endogenous mouse antibody production. Other methods for obtaining human antibodies / antibody fragments in vitro are based on display technologies, such as phage display or ribosome display technologies, using recombinant DNA libraries that are at least partially made artificially or from a donor's immunoglobulin variable (V) domain gene repertoire. Phage and ribosome display technologies for producing human antibodies are well known in the art. Human antibodies can also be produced from isolated human B cells that are ex vivo immunized with an antigen of interest and then fused to generate hybridomas that can then be screened for optimal human antibodies. The term "antibody" refers to both glycosylated and non-glycosylated antibodies. Furthermore, the term "antibody" as used herein refers not only to full-length antibodies, but also to antibody fragments, more specifically antigen-binding fragments thereof. An antibody fragment contains at least one heavy or light chain immunoglobulin domain and binds to one or more antigens as known in the art. Examples of antibody fragments according to the invention include Fab, modified Fab, Fab', modified Fab', F(ab')2, Fv, Fab-Fv, Fab-dsFv, Fab-Fv-Fv, scFv and Bis-scFv fragments. The fragments may also be diabodies, tribodies, triabodies, tetrabodies, minibodies, single domain antibodies (dAbs), such as sdAbs, VL, VH, VHH or camelid antibodies (e.g. from camel or llama, such as Nanobody™) and VNAR fragments. An antigen-binding fragment according to the invention can also comprise a Fab linked to one or two scFvs or dsscFvs, each scFv or dsscFv binding to the same or different targets (e.g., one scFv or dsscFv that binds to a therapeutic target and one scFv or dsscFv that increases half-life, e.g., by binding to albumin).Examples of such antibody fragments are FabdsscFv (also called BYbe®), or Fab-(dsscFv)2 (also called TrYbe®, see for example WO 2015 / 197772). Antibody fragments as defined above are known in the art.

[0019] The term ρ corresponds to the density of a liquid formulation. It is expressed here in g / mL. However, alternatively, it can be expressed in e.g. g / L or kg / L (mainly for liquids) or kg / m 3 or g / cm 3 (mainly for solids) can be expressed as:

[0020] The term <<η>> corresponds to the dynamic viscosity of a liquid formulation, which is expressed interchangeably in cP or alternatively in mPa.s. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] In order to meet the label claim volume, containers containing vialed liquid formulations are typically filled with a slight excess volume (overfilled). In certain cases, if a person skilled in the art is willing to take a high risk of not meeting the label claim volume, when overfilling the container, V overfill =V HU However, according to current practice, a person skilled in the art may overfill =V HU +k×σ filling I think so.

[0022] It has been the inventors' finding that it was possible to better predict or determine the hold-up volume of the vialed liquid formulation in the container and the withdrawal syringe, as well as the excess volume of the vialed liquid formulation to be added to the container containing the liquid formulation, based on only a few parameters, including the remaining volume of the liquid formulation in the container of the type of interest, the variability of the container and the extractable volume test, and the tolerance factor. Further parameters such as the formulation viscosity (measured at 20° C. for example) and the container neck diameter can also be included. Thanks to these methods, it is possible to more accurately predict / determine the excess volume to be added for each liquid formulation and each type of container, resulting, among other advantages, in a reduction of experimental tests to support the definition of the overfill / excess volume. The methods described herein can be used for any type of container, making it possible to capture all sources of variation in order to limit the risk of not meeting the extractable volume specification. Furthermore, these models can help to define the maximum overfill volume value in order to avoid possible patient safety concerns. Overall, these methods result in better process control. The basic concept of the present invention is illustrated in FIG. 1.

[0023] The primary objective of the present invention is to provide a method for determining the excess volume (V) of a liquid formulation to be added to a container containing the liquid formulation in order to ensure that the label-claimed volume can be withdrawn from the container. overfill ), comprising the steps of: a) receiving a first set of values, said first set of values ​​comprising: i. The expected remaining volume of the liquid formulation in the container and the withdrawal syringe after dispensing the liquid formulation from the container using the withdrawal syringe (V HU ); ii. The total variability (σ) of the container filling and extractable volume test of the filling device used to fill the liquid formulation into the container total ); iii. A predetermined tolerance factor (k); A process comprising: b) The following equation (1): V overfill =V HU +k×σtotal +A As shown in Figure 1, at least V HU and k×σ total and estimating the excess volume by performing a summation of A and B, where A represents any one or more additional terms in the summation; c) outputting the predicted excess volume to be added; The method includes performing the steps of:

[0024] The excess volume is the label claim volume (V label The label claim volume (V) is the volume of the liquid formulation that is required in the container, in addition to the label claim volume, prior to dispensing of the liquid formulation, to ensure that the volume of the liquid formulation can be reliably withdrawn from the container and withdrawal / dispensing device (such as a withdrawal / dosing syringe) and thus dispensed, for example, to at least one subject. Thus, the label claim volume (V label The excess volume of liquid formulation to be added to a container containing a liquid formulation (V) is the volume of liquid formulation required in the container in addition to the label claim volume prior to dispensing of the liquid formulation to ensure that the liquid formulation can be withdrawn from the container. overfill ), comprising the steps of: a) receiving a first set of values, said first set of values ​​comprising: i. The expected remaining volume of the liquid formulation in the container and the withdrawal syringe after dispensing the liquid formulation from the container using the withdrawal syringe (V HU ); ii. The total variability (σ) of the container filling and extractable volume test of the filling device used to fill the liquid formulation into the container total ); iii. A predetermined tolerance factor (k); A process comprising: b) The following equation (1): V overfill =V HU +k×σ total +A As shown in Figure 1, at least V HU and k×σ total and estimating the excess volume by performing a summation of where A represents any one or more additional terms in the summation; and c) outputting the predicted excess volume to be added; Also described herein are methods that include performing:

[0025] Another object of the present invention is to provide a method for determining the excess volume (V) of a liquid formulation to be added to a container containing the liquid formulation. overfill ), comprising the steps of: a) collecting a first set of values, said first set of values ​​comprising: i. The expected remaining volume of the liquid formulation in the container and the withdrawal syringe after dispensing the liquid formulation from the container using the withdrawal syringe (V HU ); ii. The total variability (σ) of the container filling and extractable volume test of the filling device used to fill the liquid formulation into the container total ); iii. A predetermined tolerance factor (k); A process comprising: b) The following equation (1): V overfill =V HU +k×σ total +A As shown in Figure 1, at least V HU and k×σ total where A represents any one or more additional terms in the summation; The method includes:

[0026] The excess volume is the label claim volume (V label ) can be reliably withdrawn from the container and withdrawal / dispensing device (such as a withdrawal / administration syringe) and then dispensed, for example, to at least one subject. Thus, the label claim volume (V labelThe excess volume of liquid formulation to be added to a container containing a liquid formulation (V) is the volume of liquid formulation required in the container in addition to the label claim volume prior to dispensing of the liquid formulation to ensure that the liquid formulation can be withdrawn from the container. overfill ), comprising the steps of: a) collecting a first set of values, said first set of values ​​comprising: i. The expected remaining volume of the liquid formulation in the container and the withdrawal syringe after dispensing the liquid formulation from the container using the withdrawal syringe (V HU ); ii. The total variability (σ) of the container filling and extractable volume test of the filling device used to fill the liquid formulation into the container total ); iii. A predetermined tolerance factor (k); A process comprising: b) The following equation (1): V overfill =V HU +k×σ total +A As shown in Figure 1, at least V HU and k×σ total where A represents any one or more additional terms in the summation; Also described herein are methods comprising:

[0027] Overall, in the method according to the invention, k is preferably 1.64(z 0.95 )~2.78(z 0.9973 ) range. The specific value depends on the willingness to take the risk of not delivering on the label claim volume. This is an individual or company decision. It is based on the well-known z-table used in hypothesis testing.

[0028] In an alternative embodiment of any of the methods described herein, the first set of values ​​collected / received (step a) includes a mean fill bias (B filling In such a case, equation (1) can be rewritten as equation (1bis): V overfill =VHU +B filling +k×σ total In other words, the prediction of the excess volume is the following version (1bis) of equation (1): V overfill =V HU +B filling +k×σ total As shown in Figure 1, at least V HU , k×σ total , and B. filling This includes performing a summation over

[0029] Including this parameter further improves the accuracy of the method, but is not required since the first formula already provides a high level of accuracy. filling If used, this can be experimentally predetermined.

[0030] In the context of the present invention as a whole, the value V HU can be determined or predicted according to various methods. HU One method for determining the solubility of ethanol involves the following main steps: a) Weigh the empty container into which the liquid formulation will be filled to obtain W tare obtaining b) filling the container with a volume of the liquid formulation; c) Weigh the filled container and measure W full obtaining d) Weight of the formulation contained in the container (W prod ) to obtain W prod =W full -W tare The process is as follows: d) withdrawing the liquid formulation from the container using a syringe; e) expelling the entire contents of the filled syringe into a separate, tared, clean and dry container without emptying the needle; f) Weigh the extracted liquid in the filled container to obtain a value W extr obtaining g) The following equation (2): V HU =(W prod -Wextr ) / ρ to find V HU where ρ is the density of the liquid formulation; Includes.

[0031] In the context of this alternative method, a given volume of liquid formulation is the V HU (e.g., as determined by current methods, e.g., USP <1151> and Table 1) and up to the volume required to completely fill the container. For example, if the labeled volume (or labeled size) is 2 mL and the formulation is mobile, the given volume to be filled can be any volume that is at least slightly more than 0.15 mL. In another example, if the labeled volume (or labeled size) is 10 mL and the formulation is viscous, the given volume to be filled can be any volume that is at least slightly more than 0.70 mL.

[0032] V HU An alternative method for determining the following main steps: a) Add a given weight of liquid formulation (W) to an empty container. prod ) filling the molded article; b) withdrawing the liquid formulation from the container using a syringe; c) expelling the entire contents of the filled syringe into a separate, tared, clean and dry container without emptying the needle; d) Weigh the extracted liquid in the filled container to obtain a value W extr obtaining e) The following equation (2): V HU =(W prod -W extr ) / ρ to find V HU where ρ is the density of the liquid formulation; Includes.

[0033] This method is illustrated in Figure 2. In the context of this alternative method, a given fill weight of a liquid formulation is calculated based on the V HU(e.g. USP <1151> The fill weight may be any weight so long as it corresponds to a volume that is slightly greater than, for example, 10%, and less than or equal to the fill weight corresponding to the volume required to completely fill the container.

[0034] However, the inventors have surprisingly found that the value V HU Thus, a method was identified that can predict or determine the remaining volume (V HU ) according to the following equation (3):

number

number

number

number

number

[0035] In the context of the present invention as a whole, σ total is given by the following equation (E4):

number

[0036] σ filling If is the filling standard deviation of the equipment used to fill the container, then σ analysis is the standard deviation of the analytical test method (alternatively called the extractable volume test variability). σ filling can be determined, for example, by weighing several containers that are then filled with a given target fill weight. analysis can be determined, for example, by performing extractable volume tests on several replicate vials.

[0037] As described herein, when any of the methods of predicting excess volume are performed by a computer, the output of the predicted excess volume is a value that is optionally transmitted to a database, a data set, a computer readable memory, a computer readable medium, a computer processor, a computer network, a print output device, a visual display, or a wireless receiver to enable displaying the predicted excess volume as text or a graph via any means, such as on a computer monitor or any device screen, printing the predicted excess volume as text or a graph, or communicating the predicted excess volume as audio, for example by output via a computer system.

[0038] 1. A method for filling a container with a liquid formulation, comprising the steps of: a) providing a container to be filled; b) filling the container with a total volume of the liquid formulation, the total volume of the liquid formulation corresponding to the label claim volume of the container plus an excess volume, the excess volume being determined according to any of the methods described herein; and c) optionally closing the container with at least a stopper; Also disclosed herein is a method comprising:

[0039] In the context of the present invention as a whole, the liquid formulation is preferably drawn from the container using a drawing syringe.Such a syringe can then be used to administer said formulation to a subject in need thereof, in which case the drawing needle of the syringe is preferably replaced by an administration needle.

[0040] In the context of the present invention, the container may be an ampoule, a glass vial, a tube, a bottle, a syringe, a cartridge, or other such reservoir suitable for storage of a liquid formulation. For example, the container may be a glass bottle, which may have a neck size of 13 mm to 56 mm, and a nominal volume of about 5 mL to 500 mL. As another example, the container may be a glass vial conforming to an ISO standard (such as the ISO 8362 standard), which has a neck size of 13 mm or 20 mm, and a nominal volume of about 2 mL to 100 mL. These vials are also known as 2R, 4R, 6R, 8R, 10R, 15R, 20R, 30R, 50R and 100R. For example, according to the ISO8362 standard, glass vials 2R and 4R have a neck diameter of 13 mm and a capacity of 4 and 6 mL, respectively, and glass vials 6R to 100R have a neck diameter of 20 mm and a capacity of 10, 11.5, 13.5, 19, 26, 32.5, 37.5, 62 and 123 mL, respectively. The methods disclosed herein may be applicable to containers with necks greater than 56 mm and nominal volumes greater than 500 mL, but the containers do not need to be overfilled, since they are typically used to contain formulations that require less precision for administration.

[0041] In the context of the present invention as a whole, liquid formulation is either liquid formulation or liquid formulation obtained after reconstitution of freeze-dried or spray-dried formulation.Formulation can be any formulation that contains either chemical compound (alternatively referred to herein as small molecule drug or SMD) or biological compound as active ingredient.When formulation contains biological compound, said biological compound can be any therapeutic peptide, polypeptide or protein, such as cytokine, growth factor, hormone, antibody or fusion protein.

[0042] In the context of the present invention as a whole, the viscosity of the liquid formulation is preferably comprised between 1 and 100, more preferably between 1 and 50, expressed in cP or alternatively in mPa.s. Thus, the viscosity of the liquid formulation is preferably between 1 and 100 cP (alternatively expressed as 1 and 100 mPa.s), more preferably between 1 and 50 cP (alternatively expressed as 1 and 50 mPa.s). In the context of the present invention as a whole, the viscosity η is preferably measured at room temperature, for example between 15 and 25°C, for example at 18 or 20°C. [Brief description of the drawings]

[0043] [Figure 1A] Current state of the art approach. [Figure 1B] Our approach to define the excess volume, Eq. (1). [Figure 1C] The approach according to the invention for defining the excess volume, equation (1bis). [Diagram 2] One method for vial filling and extractable volume testing. [Diagram 3] Change in viscosity with concentration of sorbitol in water. [Figure 4A] Hold-up volumes in vials and withdrawal syringes (VHU) of formulations of one monoclonal antibody with IgG1 configuration (hexagons), one antibody with fAb configuration (crosses), various monoclonal antibodies with IgG4 configuration (circles), two bispecific antibodies with IgG4 configuration (triangles), one single domain antibody (diamonds), one trispecific antibody (stars) and various small molecule drugs (SMDs) (squares) in 13 mm (2R) or 20 mm (6R, 10R or 20R) vials. Sorbitol model of vials with vial neck diameters of 13 mm (dashed red) and 20 mm (flat blue). Error bars reflect standard deviation from replicate batches. [Figure 4B] Data from FIG. 4A plotted against vial neck diameter. 95% confidence intervals for the sorbitol model are shown (dotted lines). [Diagram 5]Excess volume prediction model for vials with vial neck diameters of 13 mm (dashed line) and 20 mm (flat line) within the viscosity range of 1–40 mPa.s.

[0044] example Basis of the Invention Liquid formulations (also known as liquid formulations in vials) are typically overfilled in their containers to meet the label claim volume, taking into account losses in the container, closure system, and dispensing equipment. Guidelines (e.g., USP <1151> Not only do deviations from the maximum overfill value defined by ) need to be justified based on experimental data, but any overfill volume setting also requires justification. The objective of this study was to predict the overfill volume required for vialed liquid drug products using a total variability approach that includes the variability of the fill and extractable volume studies.

[0045] Briefly, glass vials of sizes ranging from 2R to 20R volumes were filled with sorbitol-based aqueous solutions with viscosities ranging from 1 to 40 mPa.s at 20 °C. Viscosity and vial neck diameter were shown to be the main contributors to the hold-up volume of sorbitol-based aqueous solutions in the vial and withdrawal syringe. Models constructed from the sorbitol-based aqueous solution data were used to estimate the hold-up volume (alternatively V) of various molecules of therapeutic interest. HU (called the sigma-based sigma-based method) was successfully estimated.

[0046] Therefore, a total variability approach is proposed to predict excess volume of vialed liquid formulations, taking into account the variability of formulation viscosity, vial neck diameter, filling variability and extractable volume testing. The use of this predictive model may allow for reduced testing to support excess volume definition, especially in the early stages of development when the availability of drug substance may be limited. It may replace the standard method currently used.

[0047] material The formulations used in the Examples section are defined in Table 5.

[0048] method Vial filling: The determination of the hold-up volume was focused on three factors: viscosity (seven levels ranging from 1 to 40 mPa.s), vial format (four levels: 2R, 6R, 10R and 20R) as well as fill volume (five levels per vial format: 2R-1.10, 1.20, 1.30, 1.40 and 1.50 mL; 6R-3.20, 3.30, 3.40, 3.50 and 3.60 mL; 10R-5.20, 5.35, 5.50, 5.65 and 5.80 mL; 20R-10.20, 10.40, 10.60, 10.80 and 11.00 mL). A full factorial design approach (all combinations of factors and levels) was followed, resulting in 140 experimental conditions. Tare weight (W tare -vial, stopper and overseal) were measured (see Figure 2). A minimum of two replicates per condition were prepared. Vials were weight filled taking into account the density of the sorbitol solution (Table 2). Net fill weight was recorded (W fill ). All vials were stoppered and crimped with aluminum overseals.

[0049] Extractable Volume Test: Vials were filled with sorbitol solution. Syringes were sized appropriately for the volume to be extracted: 2R vial - 3mL syringe, 6R vial - 5mL syringe, 10R vial - 10mL syringe, 20R vial - 20mL syringe. A recommended 21G needle (e.g. USP 10001) with a length of at least 1 inch was used. <1> ;USP <697> A 19Gx1 1 / 2 inch needle wider than the standard (see JP General Test 6.05; FDA Guidance for Industry, 2015 or ICH Q4B Annex 2(R1)) was used to limit the withdrawal force required for viscous solutions during the extractable volume test. The syringe and needle were connected. The Flip-off® disk was removed and the needle was used to puncture the vial stopper. No air was expelled from the syringe into the vial to aid in the extraction. The entire contents of the inverted vial were extracted into the syringe as much as possible. The needle was removed from the vial stopper. With the needle pointing up, the syringe was tapped lightly to collapse any air bubbles. Air was carefully expelled from the syringe and needle until the first signs of liquid appeared from the tip of the needle. The syringe contents were expelled (without emptying the needle) into a tared glass beaker. The net weight (W extr The total vial weight after solution withdrawal (W including the Flip-off® disk weight) was recorded. resid ) was recorded (Figure 2).

[0050] The hold-up volume in the vial and the withdrawal syringe (V HU ) is expressed by the following formula: V HU =(W fill -W extr ) / ρ(Equation 1) was used to calculate.

[0051] The hold-up volume in the vial (V HUv ) is expressed by the following formula: V HUv =(W resid -W tare ) / ρ(Equation 2) was used to calculate.

[0052] The hold-up volume in the withdrawal syringe (V HUs ) is the difference between: V HUs =V HU -V HUv (Formula 3) was obtained indirectly by

[0053] A mixture model (α = 0.05) was used to calculate the hold-up volume results (V HU The effects of fill volume, solution viscosity, vial type (fixed effects) and analyst (random effects) on V HU and viscosity values. Variability of the relative extractable volume test method obtained by regression analysis (log-log model) (RSD analysis = σ analysis / V HU ) was defined as a combination of vial-to-vial variability (root mean square error of the model) and analyst variability.

[0054] Determination of filling process variability: The filling process variability was estimated using in-process fill weight values ​​from 82 batches involving 17 vialed drug product formats filled using a peristaltic pump. Viscosity, concentration and fill volume ranged from 1 to 20 mPa.s, 1 to 160 mg / mL and 1.0 to 16.8 mL, respectively. Target fill weights (V fill The dependence of actual fill weight by batch (random effect) and by time (fixed effect) was assessed using mixed models (α = 0.05). A logarithmic transformation was applied to actual and target fill weights. Two sources of filling variability were estimated (Figure 1): the mean relative fill bias (RB filling =B filling / (V label +V HU ), the root mean square of the best linear unbiased predictor of batch effect) and the relative filling precision (RSD filling = σ filling / (V label +V HU ) Root mean square error of the model, vial-to-vial variability).

[0055] Determination of Total Variability: Total variability was calculated as the root sum of the squares of the variability of the Fill Precision and Extractable Volume test methods (Equation 1):

number

[0056] Data analysis: Statistical analysis was performed using JMP 11.0.0 (SAS Institute). Graphs were generated using Prism 8.1.1 (GraphPad Software).

[0057] result Literature and preliminary test data show an exponential increase in viscosity (at 20 °C) with sorbitol concentration (Figure 3; Jiang et al., 2013), PEG concentration (Metha et al., 2020) or antibody concentration (Shieu et al., 2014). Seven sorbitol-based aqueous solutions were prepared to cover a viscosity range of 1-40 mPa.s typical of a number of formulations (Table 2).

[0058] V HUv and V HU The values ​​were surprisingly found to be independent of fill volume, but dependent on solution viscosity and vial format (Table 3). V HUv Value and V HU Post-hoc multiple comparisons of the values ​​showed that V HUv Data and V HU Two distinct groups were identified within the data: 13 mm (2R) and 20 mm (6R, 10R and 20R) neck diameter vials.

[0059] V HUs No significant effect of fill volume and solution viscosity on the V value was observed. HUsMean and standard deviation values ​​were obtained (n>=87): 88±35 μL (3 mL syringe), 91±38 μL (5 mL syringe), 107±44 μL (10 mL syringe) and 128±51 μL (20 mL syringe).

[0060] The presented model expresses V as a function of sorbitol viscosity (η in mPa.s at 20 °C; Eq. 3). HU Built to predict:

number

[0061] If the vessel neck diameter is 13 mm, Equation 3 becomes V HU =174×η 1 / 7 If the vessel neck diameter is 20 mm, equation 3 becomes V HU =262×η 1 / 7 (Actually, the value

number

[0062] Variability of Extractable Volume Test Method (RSD analysis ) was calculated to be 23.56% (Table 4). The vial-by-vial contributions to analyst and analytical variability were 32% and 68%, respectively.

[0063] The predictive potential of the sorbitol model was evaluated for 20 formulations of various proteins / antibodies and SMD molecules of therapeutic interest (Table 6). Overall, the sorbitol model has good predictive properties, but exhibits poor predictive properties for V in the range of 100-200 μL. HU values ​​slightly overestimated (Figure 4).

[0064] No significant effect of viscosity on fill variability was observed. Because greater variation was observed at low fill volumes, in-process fill weight data was analyzed by vial neck diameter. The batch-to-batch and vial-to-vial contributions to fill variability were 23% and 77% (for 13 mm vials), and 47% and 53% (for 20 mm vials), respectively. Fill precision (RSD filling ) values ​​were 1.23% and 0.47% for the 13 mm and 20 mm neck diameter vials, respectively (complete set of data not shown).

[0065] Average filling bias (B filling ) were 0.68% and 0.44% for the 13 mm and 20 mm vials, respectively (Table 4).

[0066] The proposed strategy to predict the excess volume of liquid drug product vials is shown in Figure 1 (panels B / C) and in Equation 5 (derived from Equation 1) and alternative Equation 5bis (derived from Equation 1bis). The input parameters of the model are V label (expressed in mL), formulation viscosity at 20° C. (η, in mPa.s) and vial neck diameter (13 mm or 20 mm). label The tolerance factor k to ensure extraction of a volume equal to or greater than the 99.73% quantile of the standard normal distribution (z 0.9973 =2.78).

[0067] Equation 5:

number

number

[0068] This model is shown in Figure 5. The presented extractable volumes cover the range of 20%-80% of the nominal capacity of 13 mm (2R-4R) and 20 mm (6R-20R) vials.

[0069] Consideration Early stage development formulations are typically filled into small volume vials below the nominal volume to limit formulation waste from partially used units. This approach is based on the use of a vial and withdrawal device (V HU This does not take into account waste resulting from excess volume added to compensate for the hold-up volume in the vial. The excess volume to ensure extraction of the labeled volume from small volume vials may be greater than that recommended by FDA guidelines and should therefore be justified based on experimental data to meet regulatory requirements.

[0070] Availability of formulation material is often limited, especially in the early stages of development, and there has been interest in predictive modeling to limit extractable volume testing. Based on extractable volume data from aqueous polyethylene glycol (PEG) 400 solutions in the viscosity range of 1-30 mPa.s, V in vials (2R, 6R, and 10R) and syringes (1-10 mL) was HU A predictive model was previously proposed. This model predicts the different V HU However, our approach predicts identical V values ​​for 6R and 10R vials since they share the same stopper internal geometry. HUv / V HU Suggests value.

[0071] Experimental V HUv and V HUs The values ​​(Table 3) are in the same order of magnitude as those previously reported, 150-200 µL and 100 µL, respectively. V for PEG 400 and several molecules of therapeutic interest HU The values ​​(Figure 4) were overestimated by the sorbitol model (Equation 3), possibly due to differences in surface tension, adsorption to the container, glass surface area, or vial shoulder shape. Predictive models can help reduce testing, but this is due to the fact that the predicted V for a given formulation or presentation is overestimated. HU We emphasize the need to experimentally verify the values.

[0072] The excess volume is traditionally called the hold-up volume (V HU ) and the tolerances of the filling process are defined. The total filling process variation values ​​are within the range reported in the literature (0.25%-1.0%). The proposed overfill prediction models (Equation 1 and Equation 1bis, or Equation 5 and Equation 5bis) are based on a total variability approach that includes both process (filling precision and bias) and analytical (extractable volume test) variability. This methodology makes it possible to capture all sources of variation in order to limit the risk of out-of-specification test results.

[0073] conclusion We propose a total variability approach to excess (or overfill) volume prediction for vialed liquid drug products, taking into account formulation viscosity, vial neck diameter, filling variability and extractable volume test variability.

[0074] The use of this predictive model may allow for reduced testing to support the definition of overfill volumes, especially in the early stages of development where material availability may be limited. This will also prove useful in later stages of development as well as once the formulation is commercialized. These methods also result in better process control. [Table 1] [Table 2] [Table 3] [Table 4] [Table 5] [Table 6] References JPEG2024520340000018.jpg227165 JPEG2024520340000019.jpg99165

Claims

1. Label claim volume (V ｌａｂｅｌ an excess volume of the liquid formulation to be added to the container containing the liquid formulation (V), which is the volume of liquid formulation required in the container in addition to the label claim volume to ensure that the liquid formulation can be withdrawn from the container; ｏｖｅｒｆｉｌｌ ) comprising the steps of: a) receiving a first set of values, said first set of values ​​comprising: i. the expected remaining volume of the liquid formulation in the container and withdrawal syringe after dispensing the liquid formulation from the container using a withdrawal syringe (V ＨＵ ). ii. The total variability (σ) of the container fill and extractable volume test of the filling device used to fill the container with the liquid formulation. ｔｏｔａｌ ). iii. A predetermined tolerance factor (k); A process comprising: b) The following equation (1): V ｏｖｅｒｆｉｌｌ =V ＨＵ +k×σ ｔｏｔａｌ ++A As shown in FIG. ＨＵ and k×σ ｔｏｔａｌ predicting said excess volume by performing a summation of A and B, where A represents any one or more additional terms in the summation; c) outputting the predicted excess volume to be added; The method of claim 1, further comprising:

2. Label claim volume (V ｌａｂｅｌ an excess volume of the liquid formulation to be added to the container containing the liquid formulation (V), which is the volume of liquid formulation required in the container in addition to the label claim volume to ensure that the liquid formulation can be withdrawn from the container; ｏｖｅｒｆｉｌｌ ) comprising the steps of: a) collecting a first set of values, said first set of values ​​comprising: i. the expected remaining volume of the liquid formulation in the container and withdrawal syringe after dispensing the liquid formulation from the container using a withdrawal syringe (V ＨＵ ). ii. The total variability (σ) of the container fill and extractable volume test of the filling device used to fill the container with the liquid formulation. ｔｏｔａｌ ). iii. A predetermined tolerance factor (k); A process comprising: b) The following equation (1): V ｏｖｅｒｆｉｌｌ =V ＨＵ +k×σ ｔｏｔａｌ ++A As shown in FIG. ＨＵ and k×σ ｔｏｔａｌ predicting said excess volume by performing a summation of A method comprising:

3. The method of claim 1 or claim 2, wherein k has a value in the range of 1.64 to 2.

78.

4. The first set of values ​​may optionally be a mean fill bias (B ｆｉｌｌｉｎｇ ) wherein predicting the excess volume comprises the following version (1bis) of equation (1): V ｏｖｅｒｆｉｌｌ =V ＨＵ +B ｆｉｌｌｉｎｇ +k×σ ｔｏｔａｌ As shown in FIG. ＨＵ , k×σ ｔｏｔａｌ , and B ｆｉｌｌｉｎｇ 3. The method of claim 1 or claim 2, comprising performing a summation over

5. The following steps: a) Weighing an empty container into which the liquid formulation will be filled, W ｔａｒｅ obtaining b) filling said container with a given volume of said liquid formulation; c) Weighing the filled container to obtain W ｆｕｌｌ obtaining d) the weight (W) of the formulation contained in the container ｐｒｏｄ ) to obtain W ｐｒｏｄ =W ｆｕｌｌ -W ｔａｒｅ The process is as follows: d) withdrawing the liquid formulation from the container using a syringe; e) expelling the entire contents of the filled syringe into a separate, tared, clean and dry container without emptying the needle; f) weighing the extracted liquid in the filled container to obtain a value W ｅｘｔｒ obtaining g) The following equation (2): V ＨＵ = (W ｐｒｏｄ -W ｅｘｔｒ ) / ρ to V ＨＵ where ρ is the density of the liquid formulation; By carrying out V ＨＵ The method of claim 1 or claim 2, further comprising the step of determining:

6. The following steps: a) Add a given weight of liquid formulation (W ｐｒｏｄ ) filling the mixture; b) withdrawing the liquid formulation from the container using a syringe; c) expelling the entire contents of the filled syringe into a separate, tared, clean and dry container without emptying the needle; d) weighing the extracted liquid in the filled container to obtain a value W ｅｘｔｒ obtaining e) The following equation (2): V ＨＵ = (W ｐｒｏｄ -W ｅｘｔｒ ) / ρ to V ＨＵ where ρ is the density of the liquid formulation; By carrying out V ＨＵ The method of claim 1 or claim 2, further comprising the step of determining:

7. The first set of values ​​to be collected further includes a viscosity η corresponding to a viscosity of the liquid formulation and a neck diameter of the container, and is calculated based on the following equation (3): [0010] According to V ＨＵ wherein: [0025] The method of claim 1 or claim 2, wherein: is a value that depends on the neck diameter.

8. B ｆｉｌｌｉｎｇ The method of claim 4 , wherein is experimentally predetermined.

9. σ ｔｏｔａｌ But, the following equation (4): [0030] where σ ｆｉｌｌｉｎｇ is the filling standard deviation, and σ ａｎａｌｙｓｉｓ 3. The method of claim 1 or claim 2, wherein x is the standard deviation of the analytical test method.

10. 3. The method of claim 1 or claim 2, wherein the liquid formulation is dispensed from the container using the withdrawal syringe.

11. 2. The method of claim 1, wherein the output predicted excess volume is a value transmitted to a database, a data set, a computer readable memory, a computer readable medium, a computer processor, a computer network, a print output device, a visual display, or a wireless receiver, optionally to enable displaying the predicted excess volume as text or a graph via any means, such as on a computer monitor or any device screen, printing the predicted excess volume as text or a graph, or communicating the predicted excess volume as audio, for example by output via a computer system.

12. 1. A method for filling a container with a liquid formulation, comprising the steps of: a) providing a container to be filled with the liquid formulation; b) filling the container with a total volume of the liquid formulation, the total volume of the liquid formulation corresponding to the label claim volume of the container plus an excess volume, the excess volume being determined according to claim 1 or claim 2; and c) optionally closing said container with at least a stopper; The method includes:

13. 13. The method of claim 1, claim 2, or claim 12, wherein the container is an ampoule, a glass vial, a tube, a bottle, a syringe, a cartridge, or other such reservoir suitable for storing a liquid formulation.

14. The total remaining volume of the liquid formulation in the container and the withdrawal syringe (V ＨＵ ) , comprising the steps of: ＨＵ But, the following equation (3): [0045] where η is the viscosity of the liquid formulation; [0050] 【number】 is a value that depends on the neck diameter, method.

15. i) if the container neck diameter is 13 mm, the value [006] is 174, expressed in μl, and ii) if the container neck diameter is 20 mm, the value [0070] is 262 and is expressed in μl The method of claim 7.

16. i) When the container neck diameter is 13 mm, the value [0080] is 174, expressed in μl, and ii) if the container neck diameter is 20 mm, the value [0090] is 262 and is expressed in μl The method of claim 14.

Images