Pharmaceutical support device, method for operating pharmaceutical support device, and program for operating pharmaceutical support device

Abstract

A pharmaceutical support device is provided with a processor. The processor performs: a process for acquiring protein information relating to a protein included in a biomedical product to be stored and including site information relating to a plurality of sites constituting the protein; prescription information relating to a prescription of a candidate preservation solution, which is a candidate of a preservation solution of a biomedical product to be preserved, and measurement data indicating the preservation stability of the candidate preservation solution as measured by an actual test; deriving, on the basis of the protein information and the prescription information, a first feature quantity relating to the storage stability of the candidate storage solution for which the entire protein is targeted, and a second feature quantity relating to the storage stability of the candidate storage solution for which each of the plurality of sites is targeted; and inputting the measurement data, the first feature amount, and the second feature amount into a machine learning model, and outputting, from the machine learning model, a score indicating the storage stability of the candidate storage solution.

Description

Technical Field

[0001] The technology of the present invention relates to a pharmaceutical preparation support device, a working method of the pharmaceutical preparation support device, and a working procedure of the pharmaceutical preparation support device. Background Art

[0002] Recently, biopharmaceuticals have attracted attention due to their high efficacy and few side effects. Biopharmaceuticals, for example, use proteins such as interferon and antibodies as active ingredients. Biopharmaceuticals are stored in a storage solution. In order to stably maintain the quality of biopharmaceuticals, it is important to make the prescription of the storage solution (also called the preparation prescription) suitable for the biopharmaceuticals.

[0003] The storage solution contains a buffer, an additive, and a surfactant. The prescription of the storage solution includes, for example, the type and concentration of each of the buffer, additive, and surfactant, and the hydrogen ion index (pH (Potential Of Hydrogen) value) of the storage solution itself.

[0004] In the past, when determining the prescription of the storage solution, a variety of storage solutions were prepared by changing the combination of buffer, additives and surfactants, and the storage stability of each storage solution was confirmed by actual experiments. However, such operations are very laborious. Therefore, for example, the following technology is proposed: as in WO2021 / 041384 and <Theresa K. Cloutier, etc., Machine Learning Models of Antibody-Excipient Preferential Interactions for Use in Computational Formulation Design, Mol. Pharmaceutics, 17, 9, 3589-3599, 2020.> (hereinafter, marked as non-patent literature), based on the information of the protein of the biopharmaceutical or the information of the additive of the storage solution, the molecular dynamics (MD: Molecular Dynamics) method or the machine learning model is used to predict the storage stability of the storage solution without conducting experiments. Summary of the invention

[0005] Technical issues to be solved by the invention

[0006] In the technology described in WO2021 / 041384 and non-patent literature, the storage stability of the storage solution is not confirmed by actual experiments. In addition, in the technology described in WO2021 / 041384 and non-patent literature, the storage stability of each of the multiple parts constituting the protein is not analyzed in detail. Therefore, there is a risk of obtaining a predicted result that is far from the original storage stability.

[0007] One embodiment of the technology according to the present invention provides a pharmaceutical preparation support device, an operating method of the pharmaceutical preparation support device, and an operating program of the pharmaceutical preparation support device, which are capable of improving the prediction accuracy of the storage stability of a storage solution.

[0008] Means for solving technical problems

[0009] The pharmaceutical drug manufacturing support device of the present invention includes a processor, and the processor performs the following processing: obtaining protein information related to a protein contained in a biopharmaceutical to be stored and including site information related to a plurality of sites constituting the protein, prescription information related to a candidate storage solution, i.e., a prescription of a candidate storage solution, which is a candidate for a storage solution of the biopharmaceutical to be stored, and measurement data indicating the storage stability of the candidate storage solution measured by an actual test; deriving a first feature value related to the storage stability of the candidate storage solution with the protein as a whole as a target, and a second feature value related to the storage stability of the candidate storage solution with each of the plurality of sites as a target, based on the protein information and the prescription information; and inputting the measurement data, the first feature value, and the second feature value into a machine learning model, and outputting a score indicating the storage stability of the candidate storage solution from the machine learning model.

[0010] Preferably, the protein information includes at least one of the amino acid sequence of the protein and the three-dimensional structure of the protein.

[0011] Preferably, the prescription information includes the types and concentrations of each of the buffer, additives, and surfactants contained in the candidate storage solution and the hydrogen ion index of the candidate storage solution itself.

[0012] Preferably, the processor derives the first feature value and the second feature value using a molecular dynamics method.

[0013] Preferably, the first characteristic quantity includes at least one of the solvent accessible surface area, spatial aggregation tendency and space charge map of the entire protein, and the second characteristic quantity includes at least one of the solvent accessible surface area, spatial aggregation tendency and space charge map of the site constituting the protein.

[0014] Preferably, the first feature value includes an index indicating compatibility between the entire protein and an additive contained in the candidate storage solution, and the second feature value includes an index indicating compatibility between a site constituting the protein and the additive.

[0015] Preferably, the measurement data includes at least any one of subvisible particle aggregation analysis data of the protein in the candidate storage solution, dynamic light scattering-based analysis data of the protein in the candidate storage solution, size exclusion chromatography-based analysis data of the protein in the candidate storage solution, and differential scanning calorimetry-based analysis data of the protein in the candidate storage solution.

[0016] Preferably, the processor presents auxiliary information corresponding to the score and assisting in determining whether the candidate storage solution can be adopted.

[0017] Preferably, the processor performs the following processing: determining whether the score satisfies a preset selection condition; and presenting the determination result as auxiliary information.

[0018] Preferably, the processor performs the following processing: outputting a plurality of scores for each of the plurality of candidate storage solutions; determining a ranking of the candidate storage solutions based on the plurality of scores; and presenting the ranking determination result as auxiliary information.

[0019] Preferably, the processor inputs the prescription information into the machine learning model in addition to the measurement data, the first feature value, and the second feature value.

[0020] Preferably the protein is an antibody.

[0021] Preferred sites are any of the fragment antigen binding region, fragment crystallizable region, fragment variable region, various structural domains, and various complementarity determining regions.

[0022] The working method of the pharmaceutical preparation support device of the present invention comprises the following steps:

[0023] acquiring protein information related to a protein contained in a biopharmaceutical to be stored and including site information related to a plurality of sites constituting the protein, prescription information related to a candidate storage solution, i.e., a candidate storage solution, which is a candidate for the storage solution of the biopharmaceutical to be stored, and measurement data indicating the storage stability of the candidate storage solution measured by actual experiments;

[0024] deriving a first feature value related to the storage stability of a candidate storage solution for the entire protein and a second feature value related to the storage stability of a candidate storage solution for each of a plurality of parts based on the protein information and the prescription information; and

[0025] The measurement data, the first feature value, and the second feature value are input to the machine learning model, and a score indicating the storage stability of the candidate storage solution is output from the machine learning model.

[0026] The operating program of the pharmaceutical preparation support device of the present invention is used to cause a computer to execute a process including the following steps:

[0027] acquiring protein information related to a protein contained in a biopharmaceutical to be stored and including site information related to a plurality of sites constituting the protein, prescription information related to a candidate storage solution, i.e., a candidate storage solution, which is a candidate for the storage solution of the biopharmaceutical to be stored, and measurement data indicating the storage stability of the candidate storage solution measured by actual experiments;

[0028] deriving a first feature value related to the storage stability of a candidate storage solution for the entire protein and a second feature value related to the storage stability of a candidate storage solution for each of a plurality of parts based on the protein information and the prescription information; and

[0029] The measurement data, the first feature value, and the second feature value are input to the machine learning model, and a score indicating the storage stability of the candidate storage solution is output from the machine learning model.

[0030] Effects of the Invention

[0031] According to the technology of the present invention, it is possible to provide a pharmaceutical preparation support device, an operating method of the pharmaceutical preparation support device, and an operating program of the pharmaceutical preparation support device that can improve the prediction accuracy of the storage stability of the storage solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a diagram showing a pharmaceutical support server and an operator terminal.

[0033] Figure 2 It is a diagram showing antibody information.

[0034] Figure 3 It is a diagram showing a series of processes including preparation of a candidate storage solution, addition of an antibody, a stress test, and measurement of measurement data, as well as prescription information and measurement data.

[0035] Figure 4 It is a diagram showing the basic structure of an antibody.

[0036] Figure 5 This is a block diagram showing computers constituting the pharmaceutical manufacturing support server.

[0037] Figure 6 This is a block diagram showing a processing unit of a CPU of a pharmaceutical pharmaceutical support server.

[0038] Figure 7 It is a diagram showing the processing of the feature quantity derivation unit.

[0039] Figure 8 This is a diagram showing the processing of the prediction unit.

[0040] Fig. 9 This is a diagram showing the processing of the prediction unit and the distribution control unit.

[0041] Fig.10 This is a diagram showing an information input screen.

[0042] Fig.11 A diagram showing an auxiliary information display screen.

[0043] Fig.12This is a flowchart showing the processing flow of the pharmaceutical support server.

[0044] Fig.13 This is a diagram showing a conventional process of selecting a storage solution suitable for a biopharmaceutical to be stored from a plurality of candidate storage solutions.

[0045] Fig.14 This is a diagram showing the flow of this example of selecting a storage solution suitable for a biopharmaceutical to be stored from a plurality of candidate storage solutions.

[0046] Fig.15 This is a diagram showing another example of the second feature amount.

[0047] Fig.16 This is a diagram showing still another example of the second feature amount.

[0048] Fig.17 This is a diagram showing another example of the selection condition.

[0049] Fig.18 This is a diagram showing the processing of the prediction unit according to the second embodiment. DETAILED DESCRIPTION

[0050] [First embodiment]

[0051] As an example, Figure 1 As shown, the pharmaceutical support server 10 is connected to the operator terminal 11 via the network 12. The pharmaceutical support server 10 is an example of a "pharmaceutical support device" involved in the technology of the present invention. The operator terminal 11 is, for example, installed in a pharmaceutical facility and operated by an operator such as a pharmacist who is engaged in the pharmaceutical preparation of biopharmaceuticals in the pharmaceutical facility. The operator terminal 11 has a display 13 and an input device 14 such as a keyboard and a mouse. The network 12 is, for example, a WAN (Wide Area Network) such as the Internet or a public communication network. In addition, in Figure 1 In the example, only one operator terminal 11 is connected to the pharmaceutical manufacturing support server 10 , but in reality, a plurality of operator terminals 11 in a plurality of pharmaceutical facilities are connected to the pharmaceutical manufacturing support server 10 .

[0052] The operator terminal 11 sends a prediction request 15 to the pharmaceutical support server 10. The prediction request 15 is for causing the pharmaceutical support server 10 to select a candidate storage solution 35 (see Figure 3 ) is a request for prediction of storage stability of a biopharmaceutical. The prediction request 15 includes antibody information 16 and prescription information / measurement data set group 17. The antibody information 16 is related to an antibody 37 (reference Figure 3) The antibody information 16 is input by the operator operating the input device 14. The antibody 37 is an example of the "protein" involved in the technology of the present invention, and the antibody information 16 is an example of the "protein information" involved in the technology of the present invention.

[0053] The prescription information / measurement data set group 17 includes a plurality of prescription information 18 and measurement data 19 groups. A prescription information 18 and a measurement data 19 group are registered for each of a plurality of, for example, several to dozens of, candidate storage solutions 35. The prescription information 18 is information related to the prescription of the candidate storage solution 35. The measurement data 19 is data indicating the storage stability of the candidate storage solution 35 measured by actual tests. These prescription information 18 and measurement data 19 are also input by the operator operating the input device 14. In addition, although not shown in the figure, the prediction request 15 also includes a terminal ID (Identification Data) for uniquely identifying the operator terminal 11 that is the source of the prediction request 15, etc.

[0054] When receiving the prediction request 15, the pharmaceutical preparation support server 10 derives auxiliary information 20 for assisting in determining whether the candidate storage solution 35 can be adopted. The pharmaceutical preparation support server 10 distributes the auxiliary information 20 to the operator terminal 11 that is the source of the prediction request 15. When receiving the auxiliary information 20, the operator terminal 11 displays the auxiliary information 20 on the display 13 so that the operator can view the auxiliary information 20.

[0055] As an example, Figure 2 As shown, antibody information 16 includes an amino acid sequence 25 of antibody 37 and a three-dimensional structure 26 and site information 27 of antibody 37. Antibody information 16 is obtained by analyzing antibody 37 using well-known techniques such as mass spectrometry, X-ray crystal structure analysis, and electron microscopy. The amino acid sequence 25 is a sequence that describes the order of peptide bonds of amino acids constituting antibody 37 such as asparagine (ASn for short), glutamine (Glu for short), and arginine (Arg for short) from the amino terminal to the carboxyl terminal. The amino acid sequence 25 is also called a primary structure. The three-dimensional structure 26 represents the secondary structure, tertiary structure, and quaternary structure of the amino acids constituting antibody 37. In addition to the exemplified α helix, the secondary structure also includes β folding, β turn, etc. The tertiary structure includes the exemplified dimer coiled coil structure, etc. The quaternary structure is a dimer, a trimer, and a tetramer. According to antibody 37, there are also antibodies (i.e., monomers) that do not have a quaternary structure. In this case, as exemplified, the quaternary structure is not registered. The three-dimensional structure 26 can also be obtained by inputting the amino acid sequence 25 into "AlphaFold2", a three-dimensional structure prediction program for proteins developed by DeepMind.

[0056] The site information 27 is information about a plurality of sites constituting the antibody 37. Here, as the site information 27, the sites representing the fragment antigen-binding region FabR (Fragment antigen-binding Region), the fragment crystallizable region FcR (Fragment Crystallizable Region) and the fragment variable region FvR (Fragment variable Region) are exemplified (all refer to Figure 4 ) corresponds to which part of the amino acid sequence 25. The site information 27 can be obtained by analyzing the antibody 37 using a well-known technique such as X-ray crystal structure analysis, or can be obtained from paper information or a database such as the Protein Data Bank.

[0057] As an example, Figure 3 As shown, the prescription information 18 includes the type and concentration 30 of the buffer contained in the candidate storage solution 35, the type and concentration 31 of the additive, and the type and concentration 32 of the surfactant. The prescription information 18 is prepared by the operator based on the antibody information 16 and his own experience. As can be seen from the L-arginine hydrochloride and refined white sugar of the exemplified additives, a plurality of buffers, additives, and surfactants are sometimes used respectively. In this case, the type and concentration are registered for each of the plurality of types. In addition, the prescription information 18 also includes the hydrogen ion index 33 of the candidate storage solution 35 itself. In addition, the molecular formulas of the buffer, additives, and surfactants can be added to the prescription information 18.

[0058] The operator actually prepares the candidate storage solution 35 in the test tube 36 according to the prescription information 18. Then, the antibody 37 is added to the prepared candidate storage solution 35. Then, a stress test such as repeated freezing and thawing and stirring for 2 weeks is performed. During the 2-week stress test, the measurement data 19 is actually measured using the measuring device 38. Figure 3 , an example is shown in which measurement data 19_1 and 19_2 are measured in the first week and the second week of the pressure test, respectively. The measurement device 38 transmits the measurement data 19 to the operator terminal 11 .

[0059] As shown in the measurement data 19_1 of the first week, the measurement data 19 includes aggregation analysis data (hereinafter referred to as SVP aggregation analysis data) 40 of the sub-visible particles (SVP: Sub-Visible Particle) of the antibody 37 in the candidate storage solution 35 and analysis data (hereinafter referred to as DLS analysis data) 41 based on the dynamic light scattering method (DLS: Dynamic Light Scattering) of the antibody 37 in the candidate storage solution 35. The SVP aggregation analysis data 40 indicates the amount of SVP of the antibody 37 in the candidate storage solution 35 that causes the decrease in the efficacy of the biopharmaceutical. The DLS analysis data 41 also indicates the amount of nanoparticles of the antibody 37 in the candidate storage solution 35 that causes the decrease in the efficacy of the biopharmaceutical.

[0060] Furthermore, the measurement data 19 includes analysis data 42 of the antibody 37 in the candidate storage solution 35 based on size exclusion chromatography (SEC: Size Exclusion Chromatography) (hereinafter referred to as SEC analysis data) and analysis data 43 of the antibody 37 in the candidate storage solution 35 based on differential scanning calorimetry (DSC: Differential Scanning Calorimetry) (hereinafter referred to as DSC analysis data). The SEC analysis data 42 indicates the molecular weight distribution of the antibody 37 in the candidate storage solution 35. The DSC analysis data 43 indicates values ​​related to thermophysical property values ​​such as the glass transition point and crystallization temperature of the antibody 37 in the candidate storage solution 35. Figure 3 In the figure, for the sake of convenience, only one measuring device 38 is depicted, but in reality, various measuring devices 38 are prepared to measure these analysis data 40 to 43.

[0061] The same candidate storage solution ID is assigned to the prescription information 18 and the measurement data 19 of one candidate storage solution 35. The prescription information 18 and the measurement data 19 are associated with each other by the candidate storage solution ID. Figure 3 , only the preparation of one candidate storage solution 35 , the addition of the antibody 37 , the pressure test, and the measurement of the measurement data 19 are illustrated, but in reality, the preparation, addition, pressure test, and the measurement of the measurement data 19 are performed on each of a plurality of candidate storage solutions 35 .

[0062] As an example, Figure 4 As shown, antibody 37 basically has four polypeptide chains, namely, two identical heavy chains HC (Heavy Chain) and two identical light chains LC (Light Chain). Antibody 37 has a structure in which the two heavy chains HC and the two light chains LC are bonded by disulfide bonds DB (Disulfide Bonds), and has a bilaterally symmetrical Y shape.

[0063] The heavy chain HC is composed of the variable domain VH (Variable domain, Heavy Chain) and the constant domain CH (Constant domain, Heavy Chain) 1, CH2 and CH3. The light chain LC is composed of the variable domain VL (Variable domain, Light Chain: Light chain) and the constant domain CL (Constant domain, Light Chain: Light chain). The constant domains CH1 and CH2 of the heavy chain HC are connected by the hinge region HR (Hinge Region). The variable domains VH and VL are collectively referred to as the variable region. The constant domains CH1 to CH3 and CL are collectively referred to as the constant region.

[0064] The variable domain VH includes three complementarity determining regions CDR-H (Complementarity Determining Region, Heavy Chain: heavy chain) 1, CDR-H2 and CDR-H3. In addition, the variable domain VL also includes three complementarity determining regions CDR-L (Complementarity Determining Region, Light Chain: light chain) 1, CDR-L2 and CDR-L3. These complementarity determining regions CDR-H1 to CDR-H3 and CDR-L1 to CDR-L3 are the bonding sites with the antigen and are also called hypervariable regions.

[0065] The region consisting of variable domains VH and VL and constant domains CH1 and CL is the fragment antigen binding region FabR. The region consisting of constant domains CH2 and CH3 and a portion of hinge region HR is the fragment crystallizable region FcR. The region consisting of variable domains VH and VL is the fragment variable region FvR. As is well known, fragment antigen binding region FabR and fragment crystallizable region FcR can be produced by papain digestion. Similarly, fragment variable region FvR can be produced by pepsin digestion.

[0066] As an example, Figure 5 As shown, the computer constituting the pharmaceutical pharmaceutical support server 10 includes a storage 50 , a memory 51 , a CPU (Central Processing Unit) 52 , a communication unit 53 , a display 54 , and an input device 55 . These are connected to each other via a bus 56 .

[0067] The memory 50 is a hard disk drive built into a computer constituting the pharmaceutical support server 10 or connected via a cable or a network. Alternatively, the memory 50 is a disk array formed by connecting a plurality of hard disk drives. The memory 50 stores control programs such as an operating system, various application programs, and various data attached to these programs. In addition, a solid-state drive may be used instead of a hard disk drive.

[0068] The memory 51 is a working memory for the CPU 52 to perform processing. The CPU 52 loads the program stored in the storage 50 into the memory 51 and performs processing according to the program. Thus, the CPU 52 centrally controls each part of the computer. The CPU 52 is an example of a "processor" involved in the technology of the present invention. In addition, the memory 51 can be built into the CPU 52.

[0069] The communication unit 53 controls the transmission of various information with external devices such as the operator terminal 11. The display 54 displays various screens. Various screens have operation functions based on GUI (Graphical User Interface). The computer constituting the pharmaceutical support server 10 receives input of operation instructions from the input device 55 through various screens. The input device 55 is a keyboard, a mouse, a touch panel, a microphone for voice input, etc.

[0070] As an example, Figure 6 As shown, a working program 60 is stored in the memory 50 of the pharmaceutical support server 10. The working program 60 is an application for causing the computer to function as the pharmaceutical support server 10. That is, the working program 60 is an example of a "working program of a pharmaceutical support device" involved in the technology of the present invention. The memory 50 also stores antibody information 16 and prescription information / measurement data set group 17. In addition, the memory 50 also stores a storage stability prediction model 61 and a selection condition 62. The storage stability prediction model 61 is an example of a "machine learning model" involved in the technology of the present invention.

[0071] When the operation program 60 is activated, the CPU 52 and the memory 51 etc. of the computer constituting the pharmaceutical support server 10 cooperate to function as a receiving unit 65, a read / write (hereinafter referred to as RW (Read Write)) control unit 66, a feature value derivation unit 67, a prediction unit 68 and a distribution control unit 69.

[0072] The receiving unit 65 receives the prediction request 15 from the operator terminal 11. As described above, the prediction request 15 includes the antibody information 16 and the prescription information / measurement data set group 17, and the prescription information / measurement data set group 17 includes the prescription information 18 and the measurement data 19. Therefore, the receiving unit 65 obtains the antibody information 16, the prescription information 18, and the measurement data 19 by receiving the prediction request 15. The receiving unit 65 outputs the antibody information 16 and the prescription information / measurement data set group 17 to the RW control unit 66. In addition, the receiving unit 65 outputs the terminal ID of the operator terminal 11 (not shown) to the distribution control unit 69.

[0073] The RW control unit 66 controls the storage of various data in the memory 50 and the reading of various data in the memory 50. For example, the RW control unit 66 stores the antibody information 16 and the prescription information / measurement data set group 17 from the receiving unit 65 in the memory 50. In addition, the RW control unit 66 reads the antibody information 16 and the prescription information / measurement data set group 17 from the memory 50, outputs the antibody information 16 and the prescription information 18 to the feature quantity derivation unit 67, and outputs the measurement data 19 to the prediction unit 68. In addition, the RW control unit 66 reads the storage stability prediction model 61 from the memory 50, and outputs the storage stability prediction model 61 to the prediction unit 68. In addition, the RW control unit 66 reads the selection condition 62 from the memory 50, and outputs the selection condition 62 to the distribution control unit 69.

[0074] The feature value deriving unit 67 derives a feature value group 75 which is a collection of a plurality of feature values ​​related to the storage stability of the candidate storage solution 35 based on the antibody information 16 and the prescription information 18. The feature value deriving unit 67 outputs the derived feature value group 75 to the predicting unit 68.

[0075] The prediction unit 68 derives a score 76 indicating the storage stability of the candidate storage solution 35 from the measurement data 19 and the feature value group 75 using the storage stability prediction model 61. The prediction unit 68 outputs the derived score 76 to the distribution control unit 69.

[0076] The distribution control unit 69 generates auxiliary information 20 based on the score 76. The distribution control unit 69 controls the distribution of auxiliary information 20 to the operator terminal 11 that sent the prediction request 15. At this time, the distribution control unit 69 specifies the operator terminal 11 that sent the prediction request 15 based on the terminal ID from the receiving unit 65.

[0077] As an example, Figure 7As shown, the feature quantity deriving unit 67 derives a feature quantity group 75 using the molecular dynamics method. The feature quantity group 75 is composed of a first feature quantity 751, second feature quantities 752A, 752B, and 752C. The first feature quantity 751 is a feature quantity related to the storage stability of the candidate storage solution 35 for the entire antibody 37. The second feature quantity 752A is a feature quantity related to the storage stability of the candidate storage solution 35 for the fragment antigen-binding region FabR. The second feature quantity 752B is a feature quantity related to the storage stability of the candidate storage solution 35 for the fragment crystallizable region FcR. The second feature quantity 752C is a feature quantity related to the storage stability of the candidate storage solution 35 for the fragment variable region FvR. The feature quantity deriving unit 67 refers to the site information 27 of the antibody information 16 to derive the second feature quantities 752A to 752C. The fragment antigen-binding region FabR, the fragment crystallizable region FcR and the fragment variable region FvR are examples of the "site" involved in the technology of the present invention.

[0078] The first feature quantity 751 includes the solvent accessible surface area (hereinafter referred to as SASA (Solvent Accessible Surface Area)) 801 of the overall hydrophobicity of the antibody 37, the overall spatial aggregation tendency (hereinafter referred to as SAP (Spatial Aggregation Propensity)) 811 of the antibody 37, and the overall spatial charge map (hereinafter referred to as SCM (Spatial Charge Map)) 821 of the antibody 37. The larger the SASA801 and the larger the SAP811, the lower the stability of the antibody 37. SCM821 is a value indicating the degree of charge of the antibody 37. These SASA801, SAP811 and SCM821 are derived only from the antibody information 16 without referring to the prescription information 18. Therefore, SASA801, SAP811 and SCM821 are the same in each candidate storage solution 35.

[0079] Furthermore, the first feature quantity 751 also includes the preferential interaction coefficient (hereinafter referred to as PIC (Preferential Interaction Coefficient)) 831 described in the non-patent literature for the entire antibody 37. PIC831 is a value indicating the ease of binding between the antibody 37 and the additive, more specifically, the degree to which the additive covers the surface of the antibody 37, and is an index indicating the compatibility between the antibody 37 and the additive. According to PIC831, the possibility of causing aggregation of the antibody 37, which causes the decrease in the efficacy of the biopharmaceutical, can be known. PIC831 is different from the above-mentioned SASA801, SAP811, and SCM821, and is derived from the two information of the antibody information 16 and the prescription information 18. Therefore, PIC831 is different in each candidate storage solution 35. Therefore, with respect to the plurality of first feature quantities 751 of each candidate storage solution 35, SASA801, SAP811, and SCM821 are the same, but PIC831 is different. In addition, similarly to the prescription information 18 and the measurement data 19, the candidate storage solution ID is assigned to the first feature quantity 751. Similarly, the candidate storage solution ID is assigned to the second feature quantities 752A to 752C.

[0080] The second feature 752A includes SASA802A of the hydrophobicity of the fragment antigen binding region FabR, SAP812A of the fragment antigen binding region FabR, and SCM822A of the fragment antigen binding region FabR. The larger the SASA802A is, or the larger the SAP812A is, the lower the stability of the fragment antigen binding region FabR is. SCM822A is a value representing the degree of charge of the fragment antigen binding region FabR. As in the case of the first feature 751, these SASA802A, SAP812A, and SCM822A are also derived only from the antibody information 16, and therefore are common to each candidate storage solution 35.

[0081] In addition, the second feature 752A also includes PIC832A of the fragment antigen-binding region FabR. PIC832A indicates the ease of binding between the fragment antigen-binding region FabR and the additive, and more specifically, indicates the degree to which the additive covers the surface of the fragment antigen-binding region FabR, and is an index indicating the compatibility between the fragment antigen-binding region FabR and the additive. As in the case of the first feature 751, PIC832A is also derived based on both the antibody information 16 and the prescription information 18, and therefore is different in each candidate storage solution 35.

[0082] The second feature 752B includes SASA802B of the hydrophobicity of the fragment crystallizable region FcR, SAP812B of the fragment crystallizable region FcR, and SCM822B of the fragment crystallizable region FcR. The larger the SASA802B and the larger the SAP812B, the lower the stability of the fragment crystallizable region FcR. SCM822B is a value indicating the degree of charge of the fragment crystallizable region FcR. As in the case of the first feature 751, etc., these SASA802B, SAP812B, and SCM822B are also derived only from the antibody information 16, and therefore are common to each candidate storage solution 35.

[0083] Furthermore, the second feature quantity 752B also includes PIC832B of the fragment crystallizable region FcR. PIC832B indicates the ease of binding between the fragment crystallizable region FcR and the additive, and more specifically, indicates the degree to which the additive covers the surface of the fragment crystallizable region FcR, and is an index indicating the compatibility between the fragment crystallizable region FcR and the additive. As in the case of the first feature quantity 751, etc., PIC832B is also derived based on both the antibody information 16 and the prescription information 18, and therefore is different in each candidate storage solution 35.

[0084] The second feature 752C includes SASA802C of the hydrophobicity of the fragment variable region FvR, SAP812C of the fragment variable region FvR, and SCM822C of the fragment variable region FvR. The larger the SASA802C is, or the larger the SAP812C is, the lower the stability of the fragment variable region FvR is. SCM822C is a value indicating the degree of charge of the fragment variable region FvR. As in the case of the first feature 751, etc., these SASA802C, SAP812C, and SCM822C are also derived only from the antibody information 16, and therefore are common to each candidate storage solution 35.

[0085] Furthermore, the second feature quantity 752C also includes PIC832C of the fragment variable region FvR. PIC832C indicates the ease of binding between the fragment variable region FvR and the additive, and more specifically, indicates the degree to which the additive covers the surface of the fragment variable region FvR, and is an indicator of the compatibility between the fragment variable region FvR and the additive. As in the case of the first feature quantity 751, etc., PIC832C is also derived based on both the antibody information 16 and the prescription information 18, and therefore is different in each candidate storage solution 35. In addition, PIC831, 832A, 832B, and 832C can also be derived using a machine learning model such as a support vector machine (SVM) instead of a molecular dynamics method.

[0086] As an example, Figure 8As shown, the prediction unit 68 inputs the measurement data 19 and the feature value group 75 (the first feature value 751 and the second feature values ​​752A to 752C) into the storage stability prediction model 61. Then, the score 76 is output from the storage stability prediction model 61. The score 76 is represented by a continuous numerical value of 0 to 100, for example. The storage stability prediction model 61 is a machine learning model that predicts the score 76 of the continuous numerical value, for example, using a support vector regression (SVR: Support Vector Regression) algorithm. In addition, similarly to the first feature value 751, the score 76 is also assigned with a candidate storage solution ID.

[0087] The storage stability prediction model 61 may be learned in the pharmaceutical manufacturing support server 10 or in a device other than the pharmaceutical manufacturing support server 10. Furthermore, the storage stability prediction model 61 may be learned continuously after operation.

[0088] As an example, Fig. 9 As shown in Table 90, the prediction unit 68 outputs a plurality of scores 76 to the storage stability prediction model 61 for each of the plurality of candidate storage solutions 35. As shown in Table 90, the distribution control unit 69 determines the ranking of the plurality of candidate storage solutions 35 based on the plurality of scores 76. Furthermore, the distribution control unit 69 determines whether the score 76 satisfies the selection condition 62. The distribution control unit 69 distributes the score 76, the result of determining the ranking based on the score 76, and the result of determining whether the score 76 satisfies the selection condition 62 to the operator terminal 11 as the auxiliary information 20.

[0089] exist Fig. 9 , an example is shown in which the ranking of five candidate storage solutions 35, whose candidate storage solution IDs are PS0001 to PS0005, is determined, and whether the score 76 satisfies the selection condition 62 is determined. Regarding the score 76, PS0001 is 75 points, PS0002 is 62 points, PS0003 is 89 points, PS0004 is 47 points, and PS0005 is 82 points. Regarding the ranking, PS0003 with the highest score of 89 points is ranked first, PS0005 with the second highest score of 82 points is ranked second, PS0001 with 75 points is ranked third, PS0002 with 62 points is ranked fourth, and PS0004 with the lowest score of 47 points is ranked fifth. In addition, the selection condition 62 is "a score of 75 points or more". Therefore, the determination results of PS0003, PS0005, and PS0001, whose scores 76 are 75 or more, are "satisfied", and the determination results of PS0002 and PS0004, whose scores 76 are less than 75, are "unsatisfied".

[0090] Fig.10An example of the information input screen 100 displayed on the display 13 of the operator terminal 11 is shown. The information input screen 100 has an input area 101 for the antibody information 16 and an input area 102 for the information of the candidate storage solution 35. The input area 102 has an input area 103 for the prescription information 18, an input area 104_1 for the measurement data 19_1 of the first week, and an input area 104_2 for the measurement data 19_2 of the second week.

[0091] In the input area 101 of the antibody information 16 , an input box 110A for the amino acid sequence 25 , an input box 110B for the three-dimensional structure 26 , and an input box 110C for the site information 27 are provided.

[0092] In the input area 103 of the prescription information 18, there are provided a drop-down menu 112 for selecting the type of buffer solution, an input box 113 for the concentration of the buffer solution, a drop-down menu 114 for selecting the type of additive, an input box 115 for the concentration of the additive, a drop-down menu 116 for selecting the type of surfactant, and an input box 117 for the concentration of the surfactant.

[0093] A buffer addition button 118 is provided below the buffer pull-down menu 112 and input box 113. When the addition button 118 is selected, the buffer pull-down menu 112 and input box 113 are added. An additive addition button 119 is provided below the additive pull-down menu 114 and input box 115. When the addition button 119 is selected, the additive pull-down menu 114 and input box 115 are added. A surfactant addition button 120 is provided below the surfactant pull-down menu 116 and input box 117. When the addition button 120 is selected, the surfactant pull-down menu 116 and input box 117 are added. Furthermore, a hydrogen ion index input box 121 is provided below the input area 103.

[0094] In the input area 104 of the measurement data 19, a file selection button 125 for selecting a file of the SVP aggregation analysis data 40 and a file selection button 126 for selecting a file of the DLS analysis data 41 are provided. In addition, in the input area 104 of the measurement data 19, a file selection button 127 for selecting a file of the SEC analysis data 42 and a file selection button 128 for selecting a file of the DSC analysis data 43 are provided.

[0095] When a file of each of the analysis data 40 to 43 is selected, file icons 129, 130, 131, and 132 are displayed next to the file selection buttons 125 to 128. The file icons 129 to 132 are not displayed when no file is selected.

[0096] An add button 135 for the candidate storage solution 35 is provided below the input region 102 for the information of the candidate storage solution 35. When the add button 135 is selected, the input region 102 for the information of the candidate storage solution 35 is added.

[0097] A prediction button 136 is provided below the add button 135 of the candidate storage solution 35. When the prediction button 136 is selected, a prediction request 15 including the antibody information 16 and the prescription information / measurement data set group 17 is sent from the operator terminal 11 to the pharmaceutical support server 10. The antibody information 16 includes the contents input into the input boxes 110A to 110C. The prescription information / measurement data set group 17 is composed of the contents selected by the pull-down menus 112, 114, and 116, the contents input into the input boxes 113, 115, 117, and 121, and the analysis data 40 to 43 selected by the file selection buttons 125 to 128.

[0098] Fig.11 An example of an auxiliary information display screen 140 displayed on the display 13 of the operator terminal 11 when the auxiliary information 20 is received from the pharmaceutical preparation support server 10 is shown. The auxiliary information display screen 140 has a display area 141 for the antibody information 16 and a display area 142 for the auxiliary information 20. The amino acid sequence 25, the three-dimensional structure 26, and the site information 27 are displayed in the display area 141 for the antibody information 16. A table 143 showing the ranking of each candidate storage solution 35, the score 76, and the judgment result of whether the score 76 satisfies the selection condition 62 is displayed in the display area 142 for the auxiliary information 20. In the table 143, "OK" is displayed as the judgment result when the score 76 satisfies the selection condition 62, and on the contrary, "NG" is displayed as the judgment result when it does not satisfy the selection condition 62. A display button 144 for the prescription information 18 is provided in the table 143. When the display button 144 is selected, a display screen of the prescription information 18 of the candidate storage solution 35 is popped up and displayed on the auxiliary information display screen 140. The auxiliary information display screen 140 disappears when the confirmation button 145 is selected.

[0099] Next, refer to Fig.12 The flowchart of the above structure is used to explain the operation. First, if the operation program 60 is started in the pharmaceutical support server 10, Figure 6 As shown, the CPU 52 of the pharmaceutical manufacturing support server 10 functions as a receiving unit 65 , an RW control unit 66 , a feature value deriving unit 67 , a predicting unit 68 , and a distribution control unit 69 .

[0100] The display 13 of the operator terminal 11 displays Fig.10The information input screen 100 is shown. After the operator inputs desired antibody information 16, prescription information 18, and measurement data 19 in the information input screen 100, the operator selects the prediction button 136. As a result, the prediction request 15 is transmitted from the operator terminal 11 to the pharmaceutical manufacturing support server 10.

[0101] In the pharmaceutical preparation support server 10, the prediction request 15 is received by the receiving unit 65. As a result, the antibody information 16 and the prescription information / measurement data set group 17 (prescription information 18 and measurement data 19) are acquired (step ST100). The antibody information 16 and the prescription information / measurement data set group 17 are output from the receiving unit 65 to the RW control unit 66, and are stored in the memory 50 under the control of the RW control unit 66 (step ST110).

[0102] The RW control unit 66 reads the antibody information 16 and the prescription information / measurement data set 17 from the memory 50 (step ST120). The antibody information 16 and the prescription information 18 are output from the RW control unit 66 to the feature quantity derivation unit 67. The measurement data 19 are output from the RW control unit 66 to the prediction unit 68.

[0103] like Figure 7 As shown, in the feature quantity derivation unit 67, a feature quantity group 75 is derived using the molecular dynamics method based on the antibody information 16 and the prescription information 18 (step ST130). The feature quantity group 75 includes a first feature quantity 751 related to the storage stability of the candidate storage solution 35 for the entire antibody 37. In addition, the feature quantity group 75 includes a second feature quantity 752A related to the storage stability of the candidate storage solution 35 for the fragment antigen-binding region FabR, a second feature quantity 752B related to the storage stability of the candidate storage solution 35 for the fragment crystallizable region FcR, and a second feature quantity 752C related to the storage stability of the candidate storage solution 35 for the fragment variable region FvR. The feature quantity group 75 is output from the feature quantity derivation unit 67 to the prediction unit 68.

[0104] Then, if Figure 8 As shown, in the prediction unit 68, the measurement data 19 and the feature value group 75 (the first feature value 751 and the second feature values ​​752A to 752C) are input to the storage stability prediction model 61, and the score 76 is output from the storage stability prediction model 61 (step ST140). The score 76 is output from the prediction unit 68 to the distribution control unit 69. The processing of steps ST130 and ST140 is continuously repeated while all the scores 76 of the plurality of candidate storage solutions 35 are not output ("No" in step ST150).

[0105] When all scores 76 of the plurality of candidate storage solutions 35 are output ("Yes" in step ST150), as shown in FIG. Fig. 9 As shown, in the distribution control unit 69, the ranking of the candidate storage solutions 35 is determined based on the plurality of scores 76 of each of the plurality of candidate storage solutions 35. Then, it is determined whether the score 76 satisfies the selection condition 62 (step ST160). The result of the ranking determination and the result of the determination of whether the score 76 satisfies the selection condition 62 are distributed as auxiliary information 20 to the operator terminal 11 that is the transmission source of the prediction request 15 under the control of the distribution control unit 69 (step ST170).

[0106] The display 13 of the operator terminal 11 displays Fig.11 The auxiliary information display screen 140 is shown. The operator browses the auxiliary information display screen 140 and selects one candidate storage solution 35 to be used as the storage solution of the biopharmaceutical.

[0107] As described above, the CPU 52 of the pharmaceutical support server 10 includes a receiving unit 65, a feature value deriving unit 67, and a prediction unit 68. The receiving unit 65 receives the prediction request 15 to obtain the antibody information 16, the prescription information 18, and the measurement data 19. The antibody information 16 is information related to the antibody 37 contained in the biopharmaceutical to be stored. The antibody information 16 includes a plurality of parts constituting the antibody 37, including part information 27 related to the fragment antigen binding region FabR, the fragment crystallizable region FcR, and the fragment variable region FvR. The prescription information 18 is information related to the prescription of the candidate storage solution 35, which is a candidate for the storage solution of the biopharmaceutical to be stored. The measurement data 19 is data indicating the storage stability of the candidate storage solution 35 measured by an actual test. The feature value deriving unit 67 derives the first feature value 751 related to the storage stability of the candidate storage solution 35 with the entire antibody 37 as the object, based on the antibody information 16 and the prescription information 18. Furthermore, the feature value deriving unit 67 derives second feature values ​​752A to 752C related to the storage stability of the candidate storage solution 35 with respect to the plurality of sites (here, the fragment antigen-binding region FabR, the fragment crystallizable region FcR, and the fragment variable region FvR) constituting the antibody 37 as the target. The prediction unit 68 inputs the measurement data 19, the first feature value 751, and the second feature values ​​752A to 752C into the storage stability prediction model 61, and outputs a score 76 indicating the storage stability of the candidate storage solution 35 from the storage stability prediction model 61. The score 76 is output based on the measurement data 19 measured by the actual test and the second feature values ​​752A to 752C related to the storage stability of the candidate storage solution 35 with respect to each of the plurality of sites constituting the antibody 37 as the target. Therefore, compared with the case where the measurement data 19 is not used and the case where only the first feature value 751 with respect to the entire antibody 37 is used, the prediction accuracy of the storage stability of the candidate storage solution 35 can be improved.

[0108] And, as an example, Fig.13 and Fig.14 As shown, it is also possible to reduce the cumbersome work of preparing a plurality of candidate storage solutions 35 and confirming the storage stability of each candidate storage solution 35 through actual testing.

[0109] Fig.13 and Fig.14 An example is shown in which one candidate storage solution 35 is selected as a storage solution for a biopharmaceutical from 6×5=30 candidate storage solutions 35 in which the hydrogen ion index is divided into six stages at 0.5 scales, such as 5.0, 5.5, 6.0, 6.5, 7.0, and 7.5, and five types of additives, namely, sorbitol, sucrose, trehalose, proline, and L-arginine hydrochloride, are used.

[0110] exist Fig.13 1 shows a conventional process as a comparative example. Conventionally, a 4-week pressure test was performed on 30 candidate storage solutions 35, and measurement data 19_1 of the first week, measurement data 19_2 of the second week, measurement data 19_3 of the third week, and measurement data 19_4 of the fourth week were actually measured. The operator analyzed the storage stability of each candidate storage solution 35 based on these measurement data 19, and selected one candidate storage solution 35 to be used as the storage solution of the biopharmaceutical from the 30 candidate storage solutions 35.

[0111] In contrast, in the process of this example, if Fig.14 As shown, the operator completes the stress test in two weeks, which is half of the four weeks. Based on the measurement data 19_1 of the first week and the measurement data 19_2 of the second week measured in the two-week stress test and the first feature quantity 751 and the second feature quantity 752A to 752C derived from the antibody information 16 and the prescription information 18, a score 76 indicating the storage stability of the candidate storage solution 35 is derived. Then, the auxiliary information 20 corresponding to the score 76 is presented to the operator. The measurement data 19_1 of the first week and the measurement data 19_2 of the second week are the results of the midway of the four-week stress test. The operator analyzes the storage stability of each candidate storage solution 35 based on the auxiliary information 20, and selects one candidate storage solution 35 to be adopted as the storage solution of the biopharmaceutical from the 30 candidate storage solutions 35.

[0112] Thus, in the conventional process, a 4-week pressure test is required, but in the process of this example, a 2-week pressure test can be completed. Therefore, it is possible to reduce the cumbersome work of preparing a plurality of candidate storage solutions 35 and confirming the storage stability of each candidate storage solution 35 through actual testing.

[0113] In addition, as a conventional process, there is a method of determining the prescription of the storage solution in stages, such as narrowing the range of the types and concentrations of the buffer solution in the first stage, narrowing the range of the types and concentrations of the additives in the second stage, ..., narrowing the numerical range of the hydrogen ion index in the Nth stage. According to the process of this example, the work in each stage can be reduced. In addition, the stage can be omitted according to the situation, such as omitting the first stage and starting from the second stage.

[0114] like Figure 2 As shown, the antibody information 16 includes the amino acid sequence 25 of the antibody 37 and the three-dimensional structure 26 of the antibody 37. Therefore, it is possible to derive a feature value group 75 that more accurately represents the storage stability of the candidate storage solution 35. In addition, the antibody information 16 only needs to include at least one of the amino acid sequence 25 of the antibody 37 and the three-dimensional structure 26 of the antibody 37.

[0115] like Figure 3 As shown, prescription information 18 includes the types and concentrations 30 to 32 of the buffer, additives and surfactants contained in candidate storage solution 35 and hydrogen ion index 33 of candidate storage solution 35 itself. Therefore, feature quantity group 75 that more accurately represents the storage stability of candidate storage solution 35 can be derived.

[0116] like Figure 7 As shown, the feature value deriving unit 67 uses the molecular dynamics method to derive the feature value group 75. The molecular dynamics method is a computer simulation method that is popular in the biological field and can be introduced without any problem. Therefore, the feature value group 75 can be easily derived without much effort.

[0117] like Figure 7As shown, the first feature quantity 751 includes SASA801, SAP811, and SCM821 of the entire antibody 37. In addition, the first feature quantity 751 includes PIC831, which is an index indicating the compatibility of the entire antibody 37 with the additives contained in the candidate storage solution 35. Similarly, the second feature quantities 752A to 752C include SASA802A to 802C, SAP812A to 812C, SCM822A to 822C, and PIC832A to 832C of the fragment antigen-binding region FabR, the fragment crystallizable region FcR, and the fragment variable region FvR. Therefore, the score 76 that more accurately indicates the storage stability of the candidate storage solution 35 (the stability of each of the antibody 37, the fragment antigen-binding region FabR, the fragment crystallizable region FcR, and the fragment variable region FvR) can be output from the storage stability prediction model 61. The first feature 751 may include at least one of SASA801, SAP811, and SCM821 of the entire antibody 37. The second feature 752A may include at least one of SASA802A, SAP812A, and SCM822A. The same applies to the second feature 752B and 752C.

[0118] like Figure 3 As shown, the measurement data 19 includes SVP aggregation analysis data 40, DLS analysis data 41, SEC analysis data 42, and DSC analysis data 43 of the antibody 37 in the candidate storage solution 35. Therefore, it is possible to output a score 76 that more accurately represents the storage stability of the candidate storage solution 35 from the storage stability prediction model 61. In addition, the measurement data 19 only needs to include at least any one of the SVP aggregation analysis data 40, the DLS analysis data 41, the SEC analysis data 42, and the DSC analysis data 43.

[0119] like Fig. 9 As shown, the prediction unit 68 outputs a plurality of scores 76 for each of the plurality of candidate storage solutions 35. The distribution control unit 69 presents the auxiliary information 20 corresponding to the score 76, that is, the auxiliary information 20 for assisting in determining whether the candidate storage solution 35 can be adopted, by distributing it to the operator terminal 11. More specifically, the distribution control unit 69 determines the ranking of the candidate storage solutions 35 based on the plurality of scores 76, and presents the ranking determination result as the auxiliary information 20. Furthermore, the distribution control unit 69 presents the determination result of whether the score 76 satisfies the selection condition 62 as the auxiliary information 20. Therefore, the operator smoothly selects the candidate storage solution 35 to be adopted as the storage solution of the biopharmaceutical.

[0120] Biopharmaceuticals containing antibody 37 as a protein are called antibody pharmaceuticals, and are widely used in the treatment of chronic diseases such as cancer, diabetes, and rheumatoid arthritis, as well as in the treatment of rare diseases such as hemophilia and Crohn's disease. Therefore, according to this example in which the protein is antibody 37, the development of antibody pharmaceuticals widely used in the treatment of various diseases can be further promoted.

[0121] The sites are the fragment antigen-binding region FabR, the fragment crystallizable region FcR, and the fragment variable region FvR. Therefore, the score 76 that more accurately represents the storage stability of the candidate storage solution 35 can be output from the storage stability prediction model 61.

[0122] In addition, the site is not limited to the exemplified fragment antigen binding region FabR, fragment crystallizable region FcR and fragment variable region FvR. Fig.15 As shown, the site may be various structural domains, that is, variable domains VH and VL, constant domains CH1 to CH3, and CL. In this case, the feature quantity deriving unit 150 derives a second feature quantity 752D related to the storage stability of the candidate storage solution 35 for the variable domain VH and a second feature quantity 752E related to the storage stability of the candidate storage solution 35 for the variable domain VL. In addition, the feature quantity deriving unit 150 derives second feature quantities 752F, 752G, and 752H related to the storage stability of the candidate storage solution 35 for each of the constant domains CH1 to CH3, and a second feature quantity 752I related to the storage stability of the candidate storage solution 35 for the constant domain CL. The feature quantity group 75 is composed of the first feature quantity 751 and the second feature quantities 752D to 752I.

[0123] And, as an example, Fig.16 As shown, the site may be various complementarity determining regions, namely, complementarity determining regions CDR-H1 to CDR-H3 and CDR-L1 to CDR-L3. In this case, the feature quantity deriving unit 152 derives second feature quantities 752J, 752K, and 752L related to the storage stability of the candidate storage solution 35 for each of the complementarity determining regions CDR-H1 to CDR-H3 and second feature quantities 752M, 752N, and 752O related to the storage stability of the candidate storage solution 35 for each of the complementarity determining regions CDR-L1 to CDR-L3. The feature quantity group 75 is composed of the first feature quantity 751 and the second feature quantities 752J to 752O.

[0124] The site may be a fragment antigen binding region FabR, a fragment crystallizable region FcR, a fragment variable region FvR, and various domains. Furthermore, the site may also be a fragment antigen binding region FabR, a fragment crystallizable region FcR, a fragment variable region FvR, and various complementarity determining regions. Furthermore, the site may also be various domains and various complementarity determining regions. In short, the site may be any one of a fragment antigen binding region FabR, a fragment crystallizable region FcR, a fragment variable region FvR, various domains, and various complementarity determining regions.

[0125] The selection criteria are not limited to Fig. 9 The content of setting a threshold value for the score 76, such as "a score of 75 points or more" in the selection condition 62 illustrated in FIG. Fig.17 As in the selection condition 160 shown, the selection condition 160 may be related to the ranking of the score 76.

[0126] The selection condition 160 is "1st, 2nd". Therefore, the judgment results of PS0003 ranked 1st and PS0005 ranked 2nd are "satisfied", and the judgment results of PS0001, PS0002 and PS0004 ranked 3rd to 5th (not 1st and 2nd) are "unsatisfied". In addition, as a selection condition related to the ranking, for example, the content such as "top 20%" can also be used.

[0127] [Second embodiment]

[0128] As an example, in Fig.18 In the second embodiment shown, in addition to the measurement data 19 and the feature value group 75, the prescription information 18 is also input into the storage stability prediction model 165. The types of buffer, additives and surfactants are input with numerical values ​​preset for each type. Since the prescription information 18 is added to the input data of the storage stability prediction model 165, the prediction accuracy of the score 76 can be further improved. In addition, the amino acid sequence 25 and the three-dimensional structure 26 in the antibody information 16 can also be input. In this case, the preset numerical values ​​are also input for the amino acid sequence 25 and the three-dimensional structure 26.

[0129] In the above-mentioned embodiments, the description is made on the premise that there are plural candidate storage solutions 35 , but the present invention is not limited thereto, and the candidate storage solution 35 may be only one type.

[0130] Proteins are not limited to the exemplified antibody 37. They may also be cytokines (interferons, interleukins, etc.), hormones (insulin, glucagon, follicle-stimulating hormone, erythropoietin, etc.), growth factors (IGF (Insulin-Like Growth Factor: insulin-like growth factor)-1, bFGF (Basic Fibroblast Growth Factor: basic fibroblast growth factor)), coagulation factors (factor 7, factor 8, factor 9, etc.), enzymes (lysosomal enzymes, DNA (Deoxyribonucleic Acid: deoxyribonucleic acid) degradation enzymes, etc.), Fc (Fragment Crystallizable: crystallizable fragment) fusion proteins, receptors, albumin, protein vaccines. In addition, antibodies also include bispecific antibodies, antibody-drug conjugates, low molecular weight antibodies, sugar chain modified antibodies, etc.

[0131] You can also Fig.11 The screen data of the auxiliary information display screen 140 shown is distributed from the pharmaceutical manufacturing support server 10 to the operator terminal 11 , instead of distributing the auxiliary information 20 from the pharmaceutical manufacturing support server 10 to the operator terminal 11 .

[0132] The method of providing the auxiliary information 20 for the operator to view is not limited to the auxiliary information display screen 140. A printed material of the auxiliary information 20 may be provided to the operator, or an e-mail with the auxiliary information 20 attached may be sent to the operator's mobile terminal.

[0133] The pharmaceutical manufacturing support server 10 may be installed in each pharmaceutical manufacturing facility or in a data center independent of the pharmaceutical manufacturing facility. Furthermore, the operator terminal 11 may assume part or all of the functions of the processing units 65 to 69 of the pharmaceutical manufacturing support server 10 .

[0134] In each of the above-mentioned embodiments, for example, as a hardware structure of a processing unit (Processing Unit) that performs various processes such as a receiving unit 65, an RW control unit 66, a feature quantity derivation unit 67, 150 and 152, a prediction unit 68 and a distribution control unit 69, various processors (Processor) shown below can be used. As described above, various processors include a general-purpose processor, i.e., a CPU 52, that executes software (a work program 60) to function as various processing units, and processors such as FPGA (Field Programmable Gate Array) whose circuit structure can be changed after manufacturing, i.e., programmable logic devices (PLD), ASIC (Application Specific Integrated Circuit), etc., which have a circuit structure specially designed for performing specific processing, i.e., a dedicated circuit, etc.

[0135] A processing unit may be composed of one of these various processors, or a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs and / or a combination of a CPU and an FPGA). Furthermore, multiple processing units may be composed of one processor.

[0136] As an example of a processor constituting multiple processing units, there is the following method: as represented by computers such as client terminals and servers, a processor is constituted by a combination of one or more CPUs and software, and the processor functions as multiple processing units. The second method is as represented by a system on chip (SoC), etc., a processor that uses an IC (Integrated Circuit) chip to implement the functions of the entire system including multiple processing units. In this way, various processing units are constituted using one or more of the above-mentioned various processors as a hardware structure.

[0137] Furthermore, as the hardware configuration of these various processors, more specifically, a circuit (circuitry) formed by combining circuit elements such as semiconductor elements can be used.

[0138] From the above contents, the techniques described in the following supplementary notes can be understood.

[0139] [Supplementary Item 1]

[0140] A pharmaceutical preparation support device comprising a processor,

[0141] The processor performs the following processing:

[0142] Acquiring protein information related to a protein contained in a biopharmaceutical to be stored and including site information related to a plurality of sites constituting the protein, prescription information related to a candidate storage solution, i.e., a candidate storage solution, for the biopharmaceutical to be stored, and measurement data indicating storage stability of the candidate storage solution measured by actual tests;

[0143] deriving, based on the protein information and the prescription information, a first feature value related to the storage stability of the candidate storage solution for the entire protein and a second feature value related to the storage stability of the candidate storage solution for each of the plurality of parts; and

[0144] The measurement data, the first feature value, and the second feature value are input into a machine learning model, and a score indicating the storage stability of the candidate storage solution is output from the machine learning model.

[0145] [Supplementary Item 2]

[0146] The pharmaceutical preparation support device according to Supplementary Item 1, wherein:

[0147] The protein information includes at least one of the amino acid sequence of the protein and the three-dimensional structure of the protein.

[0148] [Supplementary Item 3]

[0149] The pharmaceutical preparation support device according to Supplementary Item 1 or Supplementary Item 2, wherein:

[0150] The prescription information includes the types and concentrations of the buffer, additives and surfactants contained in the candidate storage solution and the hydrogen ion index of the candidate storage solution itself.

[0151] [Supplementary Item 4]

[0152] The pharmaceutical preparation support device according to any one of Supplementary Items 1 to 3, wherein

[0153] The processor derives the first feature value and the second feature value using a molecular dynamics method.

[0154] [Supplementary Item 5]

[0155] The pharmaceutical preparation support device according to Supplementary item 4, wherein:

[0156] The first characteristic quantity includes at least one of the solvent accessible surface area, spatial aggregation tendency and spatial charge map of the protein as a whole,

[0157] The second characteristic quantity includes at least one of a solvent accessible surface area, a spatial aggregation tendency, and a spatial charge map of a site constituting the protein.

[0158] [Supplementary Item 6]

[0159] The pharmaceutical preparation support device according to any one of Supplementary Items 1 to 5, wherein

[0160] The first feature value includes an index indicating compatibility of the protein as a whole with an additive contained in the candidate storage solution.

[0161] The second feature value includes an index indicating compatibility between a site constituting the protein and the additive.

[0162] [Supplementary Item 7]

[0163] The pharmaceutical preparation support device according to any one of Supplementary Items 1 to 6, wherein

[0164] The measurement data includes at least any one of aggregation analysis data of subvisible particles of the protein in the candidate storage solution, analysis data of the protein in the candidate storage solution based on dynamic light scattering, analysis data of the protein in the candidate storage solution based on size exclusion chromatography, and analysis data of the protein in the candidate storage solution based on differential scanning calorimetry.

[0165] [Supplementary Item 8]

[0166] The pharmaceutical preparation support device according to any one of Supplementary Items 1 to 7, wherein

[0167] The processor prompts auxiliary information corresponding to the score and assisting in determining whether the candidate preservation solution can be adopted.

[0168] [Supplementary Item 9]

[0169] The pharmaceutical preparation support device according to Supplementary item 8, wherein:

[0170] The processor performs the following processing:

[0171] Determining whether the score satisfies a predetermined selected condition; and

[0172] The judgment result is provided as the auxiliary information.

[0173] [Supplementary Item 10]

[0174] The pharmaceutical preparation support device according to Supplementary item 8 or Supplementary item 9, wherein

[0175] The processor performs the following processing:

[0176] outputting a plurality of said scores for each of a plurality of said candidate preservation solutions;

[0177] determining a ranking of the candidate preservation solutions based on a plurality of the scores; and

[0178] The decision result of the sorting is presented as the auxiliary information.

[0179] [Supplementary Note 11]

[0180] The pharmaceutical preparation support device according to any one of Supplementary Items 1 to 10, wherein

[0181] The processor inputs the prescription information into the machine learning model in addition to the measurement data, the first feature value, and the second feature value.

[0182] [Supplementary Item 12]

[0183] The pharmaceutical preparation support device according to any one of Supplementary Items 1 to 11, wherein

[0184] The protein is an antibody.

[0185] [Supplementary Item 13]

[0186] The pharmaceutical preparation support device according to Supplementary item 12, wherein:

[0187] The site is any one of a fragment antigen binding region, a fragment crystallizable region, a fragment variable region, various structural domains and various complementarity determining regions.

[0188] The technology of the present invention can also appropriately combine the above-mentioned various embodiments and / or various variations. Moreover, it is not limited to the above-mentioned embodiments, and various structures can of course be adopted as long as it does not deviate from the main purpose. In addition, the technology of the present invention also relates to a storage medium that stores the program non-temporarily in addition to the program.

[0189] The recorded contents and illustrated contents shown above are detailed descriptions of the parts involved in the technology of the present invention, which are only an example of the technology of the present invention. For example, the descriptions related to the above-mentioned structure, function, action and effect are descriptions related to an example of the structure, function, action and effect of the parts involved in the technology of the present invention. Therefore, without departing from the scope of the technical purpose of the present invention, it is of course possible to delete unnecessary parts of the recorded contents and illustrated contents shown above, or to add or replace new elements. In addition, in order to avoid complicated situations and facilitate understanding of the parts involved in the technology of the present invention, descriptions related to technical common sense that does not particularly need to be explained in terms of enabling the implementation of the technology of the present invention are omitted in the recorded contents and illustrated contents shown above.

[0190] In this specification, "A and / or B" has the same meaning as "at least one of A and B". That is, "A and / or B" may mean only A, only B, or a combination of A and B. Furthermore, in this specification, the same idea as "A and / or B" is also applicable to the case where three or more items are expressed by connecting them with "and / or".

[0191] All documents, patent applications, and technical standards described in this specification are incorporated herein by reference to the same extent as if each document, patent application, or technical standard was specifically and individually described as being incorporated by reference.

Claims

1. A pharmaceutical preparation support device comprising a processor, The processor performs the following processing: Acquiring protein information related to a protein contained in a biopharmaceutical to be stored and including site information related to a plurality of sites constituting the protein, prescription information related to a candidate storage solution, i.e., a candidate storage solution, for the biopharmaceutical to be stored, and measurement data indicating storage stability of the candidate storage solution measured by actual tests; deriving, based on the protein information and the prescription information, a first feature value related to the storage stability of the candidate storage solution for the entire protein and a second feature value related to the storage stability of the candidate storage solution for each of the plurality of parts; and The measurement data, the first feature value, and the second feature value are input into a machine learning model, and a score indicating the storage stability of the candidate storage solution is output from the machine learning model.

2. The pharmaceutical preparation support device according to claim 1, wherein: The protein information includes at least one of the amino acid sequence of the protein and the three-dimensional structure of the protein.

3. The pharmaceutical preparation support device according to claim 1, wherein: The prescription information includes the types and concentrations of the buffer, additives and surfactants contained in the candidate storage solution and the hydrogen ion index of the candidate storage solution itself.

4. The pharmaceutical preparation support device according to claim 1, wherein: The processor derives the first feature value and the second feature value using a molecular dynamics method.

5. The pharmaceutical preparation support device according to claim 4, wherein: The first characteristic quantity includes at least one of the solvent accessible surface area, spatial aggregation tendency and spatial charge map of the protein as a whole, The second characteristic quantity includes at least one of a solvent accessible surface area, a spatial aggregation tendency, and a spatial charge map of a site constituting the protein.

6. The pharmaceutical preparation support device according to claim 1, wherein: The first feature value includes an index indicating compatibility of the protein as a whole with an additive contained in the candidate storage solution. The second feature value includes an index indicating compatibility between a site constituting the protein and the additive.

7. The pharmaceutical preparation support device according to claim 1, wherein: The measurement data includes at least any one of aggregation analysis data of subvisible particles of the protein in the candidate storage solution, analysis data of the protein in the candidate storage solution based on dynamic light scattering, analysis data of the protein in the candidate storage solution based on size exclusion chromatography, and analysis data of the protein in the candidate storage solution based on differential scanning calorimetry.

8. The pharmaceutical preparation support device according to claim 1, wherein: The processor prompts auxiliary information corresponding to the score and assisting in determining whether the candidate preservation solution can be adopted.

9. The pharmaceutical preparation support device according to claim 8, wherein: The processor performs the following processing: Determining whether the score satisfies a predetermined selected condition; and The judgment result is provided as the auxiliary information.

10. The pharmaceutical preparation support device according to claim 8, wherein: The processor performs the following processing: outputting a plurality of said scores for each of a plurality of said candidate preservation solutions; determining a ranking of the candidate preservation solutions based on a plurality of the scores; and The decision result of the sorting is presented as the auxiliary information.

11. The pharmaceutical preparation support device according to claim 1, wherein: The processor inputs the prescription information into the machine learning model in addition to the measurement data, the first feature value, and the second feature value.

12. The pharmaceutical preparation support device according to claim 1, wherein: The protein is an antibody.

13. The pharmaceutical preparation support device according to claim 12, wherein: The site is any one of a fragment antigen binding region, a fragment crystallizable region, a fragment variable region, various structural domains and various complementarity determining regions.

14. A method for operating a pharmaceutical support device, comprising the following steps: Acquiring protein information related to a protein contained in a biopharmaceutical to be stored and including site information related to a plurality of sites constituting the protein, prescription information related to a candidate storage solution, i.e., a candidate storage solution, for the biopharmaceutical to be stored, and measurement data indicating storage stability of the candidate storage solution measured by actual tests; deriving, based on the protein information and the prescription information, a first feature value related to the storage stability of the candidate storage solution for the entire protein and a second feature value related to the storage stability of the candidate storage solution for each of the plurality of parts; and The measurement data, the first feature value, and the second feature value are input into a machine learning model, and a score indicating the storage stability of the candidate storage solution is output from the machine learning model.

15. An operating program of a pharmaceutical preparation support device, which is used to cause a computer to execute a process comprising the following steps: Acquiring protein information related to a protein contained in a biopharmaceutical to be stored and including site information related to a plurality of sites constituting the protein, prescription information related to a candidate storage solution, i.e., a candidate storage solution, for the biopharmaceutical to be stored, and measurement data indicating storage stability of the candidate storage solution measured by actual tests; deriving, based on the protein information and the prescription information, a first feature value related to the storage stability of the candidate storage solution for the entire protein and a second feature value related to the storage stability of the candidate storage solution for each of the plurality of parts; and The measurement data, the first feature value, and the second feature value are input into a machine learning model, and a score indicating the storage stability of the candidate storage solution is output from the machine learning model.

Images