Treatment plan formulation support device and treatment plan formulation support method

Abstract

To provide a treatment plan drafting assistance device and a treatment plan drafting assistance method capable of assisting the drafting of a treatment plan suitable to an individual patient.SOLUTION: Processing by a treatment plan drafting device includes: acquiring data related to a lesion of a patient obtained by analyzing a lesion site or a body fluid of the patient; storing pathway data being data of a pathway representing a connection between genes by a directed graph, and medicine data including medicines and genes to be a treatment object by the medicines; calculating a gene chain score representing the degree of association between gene abnormality of the patient in each gene and gene abnormality of the patient in an upstream gene that can reach the gene if following an edge in the pathway about the gene included in the pathway data; and calculating a medicine score representing the degree of association between a disease and the medicines on the basis of the medicine data.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a treatment plan planning support device and a treatment plan planning support method. [Background technology]

[0002] In contrast to conventional cancer treatment, which selects drugs based on the organ in which the cancer is located and the stage of the disease, cancer genomic medicine, which selects drugs that are expected to be effective against gene mutations that affect the disease, became covered by health insurance in Japan in 2019. In cancer genomic therapy, drugs are selected based on genetic information, and as needed, referencing pathways, which are networks of gene interactions. However, interpreting data such as gene mutations requires specialized knowledge, and handling the vast network of interconnected pathways is time-consuming.

[0003] Patent Document 1 discloses a device that analyzes pathways forming a broad and complex network and supports the formulation of treatment strategies. Specifically, it includes a process of quantifying abnormalities such as gene mutations and gene expression levels in each patient as an abnormality score, and a process of calculating indicator values ​​related to drug efficacy for multiple abnormal genes through pathway analysis. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Special table 2016-513303 publication [Overview of the project] [Problems that the invention aims to solve]

[0005] However, Patent Document 1 only considers the number of abnormalities and the type of intergenetic interaction, such as suppression or enhancement, when considering the transmission of abnormalities from other genes upstream and downstream. Therefore, the present invention provides a treatment strategy planning support device and a treatment strategy planning support method that can support the formulation of treatment strategies appropriate for individual patients. In particular, this device and the like are characterized by considering the influence of the type of abnormality, its effect on the living body, and whether or not it is related to disease on downstream genes in the pathway, and by quantifying the accumulation of abnormalities occurring in each gene.

[0006] [Means for solving the problem]

[0007] In one embodiment of the subject matter of the present invention, A data acquisition unit that acquires data related to the patient's lesions by analyzing the patient's lesion site or bodily fluids, A storage unit that stores pathway data, which is pathway data representing the sequence of genes as a directed graph, and drug data, which includes a drug and the genes that the drug targets for treatment. A gene chain score calculation unit extracts gene chains from the pathway data in which the upstream gene is a key gene that satisfies the conditions determined by the patient's lesion data, and calculates a gene chain score for each gene chain that represents the degree of association between the key gene and the downstream gene of the gene chain. A drug score calculation unit calculates a drug score by adding the gene chain score of gene chains whose downstream gene is the same gene to the drug targeting the downstream gene in the drug data. The present invention provides a treatment plan planning support device characterized by being equipped with the following features. Other means will be described within the descriptions of embodiments for carrying out the invention. [Effects of the Invention]

[0008] According to the present invention, it is possible to provide a treatment plan formulation support device and a treatment plan formulation support method that can assist in formulating treatment plans appropriate for individual patients. [Brief explanation of the drawing]

[0009] [Figure 1] This is a hardware configuration diagram of the treatment strategy planning support device for Example 1. [Figure 2] This is an example of a screen operated in Example 1. [Figure 3] This figure shows an example of the processing flow in Example 1. [Figure 4] It is a diagram showing an example of various data handled in Example 1. [Figure 5A] It is a diagram for explaining an example of a pathway. [Figure 5B] It is a diagram for explaining an example of a gene chain extracted from a pathway. [Figure 6] It is an example of the output screen of Example 1. [Figure 7] It is an example of a screen operated in Example 2. [Figure 8] It is a diagram showing an example of the processing flow of Example 2. [Figure 9] It is an example of the output screen of Example 2. [Figure 10] It is an example of the output screen of Example 2. [Figure 11] It is a hardware configuration diagram of the treatment policy formulation support device of Example 3. [Figure 12] It is a diagram showing an example of biomarker data handled in Example 3. [Figure 13] It is an example of a screen operated in Example 3. [Figure 14] It is a diagram showing an example of the processing flow of Example 3. [Figure 15] It is an example of the output screen of Example 3.

Mode for Carrying Out the Invention

[0010] Hereinafter, preferred embodiments of the treatment policy formulation support device and the treatment policy formulation support method according to the present invention will be described with reference to the accompanying drawings. In the following description and the accompanying drawings, components having the same functional configuration are denoted by the same reference numerals, and redundant description is omitted.

[0011] 〈Example 1〉 In Example 1, calculating the degree of association between a disease and a drug will be described. That is, a doctor or the like who is an operator can select a drug suitable for an individual patient by confirming the degree of association between the disease and the drug.

[0012] The hardware configuration of the treatment plan planning support device 100 in Example 1 will be explained using Figure 1. The treatment plan planning support device 100 is a so-called computer, and is configured with a processing unit 101, memory 102, storage unit 103, and network adapter 104 connected via a bus 107 in a signal transmission / reception manner. The treatment plan planning support device 100 is also connected to a sample analyzer 109, a gene database (DB) 110, and a drug database (DB) 111 via the network adapter 104 and network 108 in a signal transmission / reception manner, as well as to an input unit 105 and a display unit 106 in a signal transmission / reception manner. Here, "signal transmission / reception manner" means a state in which signals can be transmitted to and from each other or from one to the other, electrically or optically, whether wired or wireless. The following describes each part.

[0013] The arithmetic unit 101 is a device that performs various calculations and controls the operation of each part, and is specifically a CPU (Central Processing Unit), etc. The arithmetic unit 101 loads programs stored in the storage unit 103 and data required by the programs into the memory 102 and executes them. The memory 102 stores the programs executed by the arithmetic unit 101 and the progress of the calculation processing. Examples of such programs include a data acquisition unit, a gene chain score calculation unit, and a drug score calculation unit (not shown).

[0014] The memory unit 103 is a device that stores programs executed by the arithmetic unit 101 and data necessary for program execution. Specifically, it is a device that reads and writes to recording devices such as HDDs (Hard Disk Drives) and SSDs (Solid State Drives), or recording media such as IC cards, SD cards, and DVDs. The memory unit 103 may also have various types of data pre-stored in it, such as patient lesion data 121, pathway data 122, and drug data 123. Examples of these types of data will be described later with reference to Figure 4. Various types of data, including data necessary for program execution, may be transmitted and received from a network 108 such as a LAN (Local Area Network).

[0015] The network adapter 104 is used to connect the treatment plan planning support device 100 to a network 108 such as a LAN, telephone line, or the internet.

[0016] The specimen analyzer 109 analyzes specimens containing patient cells, bodily fluids, etc., and adds the analysis results to, for example, patient lesion data 121. The gene database 110 is a database system that stores gene-related data, such as pathway data 122, described in known literature, etc. The drug database 111 is a database system that stores drug-related data, such as drug data 123, described in known literature, etc.

[0017] Furthermore, the patient's lesion data 121 may be the result of directly analyzing cells in the patient's lesion site, or it may be the result of indirectly analyzing bodily fluids such as blood, urine, saliva, etc., of the patient that are used to analyze the patient's lesion site (liquid biopsy). In other words, the data obtained by analyzing the patient's lesion site by any method is the "patient's lesion data 121".

[0018] The display unit 106 is a device that displays the execution results of a program performed by the calculation unit 101 (display processing unit), and is, for example, a liquid crystal display or a touch panel. The input unit 105 is an operating device that allows the operator to give operating instructions to the treatment plan planning support device 100, and is, for example, a keyboard, mouse, or touch panel. The mouse may be replaced with other pointing devices such as a trackpad or trackball. If the display unit 106 is a touch panel, the touch panel also functions as the input unit 105. The treatment plan planning support device 100 operates when the operator operates the screen displayed on the display unit 106.

[0019] Using FIG. 2, a setting screen 200, which is an example of a screen operated in Example 1, will be described. An operator sets input data, parameters, and output data using the setting screen 200. The setting screen 200 includes a patient data input section 201, a data specifying section 202 related to a patient's lesion, an attenuation rate setting section 203, a propagation distance setting section 204, a condition setting section 205 for important genes, and a setting button 206.

[0020] The patient data input section 201 is a box for inputting data related to a patient. For example, a patient ID for identifying the patient is input, and data 121 related to the patient's lesion pre-linked to the patient ID may be read out in the subsequent processing flow.

[0021] The data specifying section 202 related to a patient's lesion is a box for specifying the path to a file storing data 121 related to the patient's lesion. When the operator inputs a file path to the data specifying section 202 related to a patient's lesion and presses a reference button, data 121 related to the patient's lesion is read out based on the specified path. The file storing data 121 related to the patient's lesion may be configured to allow multiple selections.

[0022] The attenuation rate setting section 203 is a box for setting an attenuation rate D. The attenuation rate D is a parameter indicating the rate at which the influence of gene mutations attenuates between genes of a gene chain included in a pathway, and a real number of 0 < D < 1 is set. The propagation distance setting section 204 is a box for setting a propagation distance L. The propagation distance L is a parameter indicating the upper limit with respect to the length of a gene chain extracted from a pathway, and a natural number is set.

[0023] The condition setting section 205 for important genes is a box for setting conditions for important genes. The conditions for important genes are used in the process of extracting important genes based on data 121 related to the patient's lesion, which will be described later. A plurality of parameters and thresholds used for determining important genes may be displayed in the condition setting section for important genes.

[0024] The settings button 206 is pressed when all settings on the settings screen 200 have been completed. When the settings button 206 is pressed, the process described below begins.

[0025] An example of the processing flow of Embodiment 1 will be explained step by step using Figure 3. Hereafter, the case where each processing step is executed by the arithmetic unit 101 will be described. Note that each processing step may also be executed by dedicated hardware such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field-Programmable Gate Array). For convenience of explanation, Figures 4, 5A, and 5B will be referenced as appropriate during the explanation of Figure 3.

[0026] (S301) The calculation unit 101 (data acquisition unit) reads the input data and parameters. The input data is read according to the path set in the settings screen 200 in Figure 2, for example. In other words, the data 121 related to the patient's lesions is read by the calculation unit 101.

[0027] Using Figure 4, we will explain an example of patient lesion data 121, pathway data 122, and drug data 123. An example of patient lesion data 121 may include items such as gene, chromosome, location, base of the human standard genome sequence, patient's base, and type of mutation. That is, the chromosome and location are shown as the site of the gene mutation, along with the bases before and after the gene mutation and the type of mutation. For example, it is shown that the gene Gene1 underwent a missense mutation where base A was changed to base G at location 4574 of chromosome chr1.

[0028] Furthermore, the patient's lesion data 121 may include molecular measurements directly related to each gene. Examples of molecular measurements include, but are not limited to, gene expression levels and methylation levels.

[0029] Pathway data 122 includes edge data and node data. The edge data includes an ID that identifies the edge, the name of the node connected to the starting point of the edge, and the name of the node connected to the ending point of the edge. The node data includes the node name and the type of node. The node types include genes and molecules present in living organisms. Based on the edge data and node data, a pathway is constructed that represents the connections between genes as a directed graph.

[0030] Figure 5A shows an example of a pathway. Pathway 500, illustrated in Figure 5A, includes seven nodes indicated by circled numbers and eight edges indicated by arrows connecting the nodes. In other words, pathway 500 is constructed by connecting the nodes indicated by circled numbers with the edges indicated by arrows. The circled numbers are node names used to identify genes, and the direction of the arrows indicates the direction of gene expression regulation between genes. Actual pathways are vast networks, often numbering in the tens of thousands, of both edges and nodes.

[0031] Returning to the explanation of Figure 4, drug data 123 includes the drug and the gene it targets.

[0032] (S302) Returning to Figure 3, the calculation unit 101 extracts all genes that satisfy predetermined conditions based on the input data and parameters read in S301 and stores them in a list as important genes. When the patient's lesion data 121 set in the patient data input unit 201 or the patient's lesion data specification unit 202 on the setting screen 200 is read, genes that satisfy the conditions specified in the important gene condition setting unit 205 are extracted as important genes.

[0033] (S303) The calculation unit 101 reads the first important gene from the list containing the genes extracted in S302. If no important genes are stored in the list, that is, if nothing was extracted in S302, the process ends.

[0034] (S304) The calculation unit 101 (gene strand score calculation unit) extracts gene strands from the pathway in which important genes read from the list are located at the uppermost level and whose length is less than or equal to the propagation distance. A gene strand is a series of nodes and edges arranged in one direction according to the order in which they appear when traversing a directed graph, with the starting node and the endpoint node reachable by traversing the directed graph from the starting node. The length of a gene strand is represented by the number of nodes it contains, and a gene strand with a length of 1 has the same starting and ending nodes and contains no edges. The propagation distance is the value set in the propagation distance setting unit 204 of the setting screen 200, as exemplified in Figure 2.

[0035] Figure 5B illustrates an example of gene strands extracted from a pathway. Figure 5B lists all gene strands from pathway 500, as exemplified in Figure 5A, where the gene with the circle number 1 is the upstream gene and the strand is 5 or less in length. In other words, there are 12 gene strands extracted from pathway 500, and some of them have the downstream gene k as one of the circle numbers 1 to 7.

[0036] (S305) Returning to Figure 3, the calculation unit 101 determines whether or not there is a drug in the drug data 123 that targets the downstream gene k. If a drug is found, the process proceeds to S306; otherwise, the process proceeds to S308.

[0037] (S306) The calculation unit 101 (gene chain score calculation unit) calculates the gene chain score c(k) for each of the gene chains extracted in S304 at the downstream gene k. The gene chain score c(k) represents the degree of association between the important gene at the upstream end and the downstream gene k. When the length of the gene chain is 1, the gene chain score c(k) is calculated, for example, by the following formula.

[0038]

number

[0039] Here, a(k) is the mutation importance in the downstream gene k, and b(k) may be a molecular metric of the downstream gene k or a numerical value calculated by inputting it into a specific function. The mutation importance a(k) can be calculated, for example, by the following formula.

[0040]

number

[0041] Here, v j This is the j-th mutation of the important gene i, which is at the very top of the gene chain, f(v j ) is a score according to the type of mutation, e(v j This score is related to the likelihood of the mutation.

[0042] In the human body, a process called translation generates 20 different amino acids corresponding to gene sequences, and these amino acids are then converted into proteins that perform various functions. Mutations can be classified into silent mutations, which do not alter the amino acids corresponding to the sequence and have no effect on the body, and nonsense mutations, which affect the translation function of specific sequence regions and stop amino acid production. The impact of mutations on translation varies.

[0043] Therefore, the score f(v) for each gene mutation. j The value (e(v)) is set to reflect the magnitude of the amino acid change caused by the mutation. The types of mutations include missense mutations, silent mutations, and nonsense mutations, as exemplified in patient lesion data 121 in Figure 4, and are assigned to each mutation. A score e(v) related to the likelihood of the mutation is also assigned. j ) is a gene mutation v j It is set by various numerical values ​​in or by inputting them into any function to calculate a value. A score e(v) related to the likelihood of mutation. j For example, the calculation of mutation v j Mapping quality and depth may also be used.

[0044] Note that the calculation of the mutation importance a(k) is not limited to Equation 2, and the formula may be modified. Also, when the input data in S301 does not include gene mutations, the mutation importance a(k) may be set to 0 for all genes. Similarly, when the input data does not include molecular measurement quantities, the molecular measurement quantity b(k) may be set to 0 for all genes. The gene chain score c(k) of a gene chain with an arbitrary length is calculated, for example, by the following formula.

[0045]

Equation

[0046] Here, n is an arbitrary gene included in the gene chain, and L kn is the distance between gene n and the most downstream gene k, and A(L kn , D) is a function of the distance L kn and the attenuation rate D.

[0047] In Equation 3, when A(L kn , D) = D ^ (L kn ), since 0 < D < 1, the contribution of the mutation importance a(n) to the gene chain score c(k) decreases as the distance L kn becomes longer, and increases as the distance L kn becomes shorter. Note that the calculation of the gene chain score c(k) is not limited to Equation 3, and the formula may be modified.

[0048] (S307) The arithmetic unit 101 (drug score calculation unit) calculates the drug score s(m) of drug m by adding the gene chain score c(k) calculated in S306 for each drug m targeting the most downstream gene k. When there are multiple drugs targeting the most downstream gene k, the gene chain score c(k) calculated in S306 is added to the drug score s(m) of each drug.

[0049] For example, when drug A targets the gene marked with circle number 2, the drug score for drug A is calculated by adding the gene chain scores of the 2nd and 9th gene chains out of the 12 gene chains listed in Figure 5B. Similarly, when drug B targets the genes marked with circle number 2 and 6, the drug score for drug B is calculated by adding the gene chain scores of the 5th, 6th, and 12th gene chains in addition to the 2nd and 9th. Here, the drug score s(m) for drug m is initialized to 0 when it is first called.

[0050] (S308) The calculation unit 101 (display processing unit) determines in S302 whether all of the important genes stored in the list have been read. If all important genes have been read from the list, the process proceeds to S309; ​​if there are any important genes that have not been read, the process returns to S304 via S310.

[0051] (S309) The calculation unit 101 creates a correspondence table between drug m and drug score s(m) and displays it on the display unit 106. It is desirable that the correspondence table be sorted in descending order of drug score s(m) value.

[0052] (S310) The calculation unit 101 reads the next important gene from the list and proceeds to S304.

[0053] An example of the output screen 600 displayed in S309 will be explained using Figure 6. The output screen 600 includes a list of recommended drugs 601, which lists drugs in order of their degree of relevance to the disease. In other words, the list of recommended drugs 601 is a correspondence table created in S309, in which drugs m are arranged in descending order of their drug score s(m). The list of recommended drugs 601 may also include a column displaying the target gene, which is the gene that drug m treats.

[0054] Based on the processing flow described above, drugs are listed in descending order of their relevance to the patient's disease, based on at least a portion of the patient's disease-related data 121 obtained by analyzing the patient's lesion site or cells in body fluids. The physician, acting as the operator, can then select the appropriate drug for each individual patient by reviewing the drugs listed in descending order of their relevance to the disease.

[0055] <Example 2> Example 1 describes how to calculate gene chain scores and drug scores using pathway data 122 as is. Pathway data contains a mixture of highly reliable interactions and interactions with insufficient evidence and therefore low reliability. Furthermore, the literature on gene regulation is increasing daily, and new interactions are frequently added and existing interactions are modified. In such circumstances, the gene chain scores and drug scores calculated by the treatment strategy planning support device 100 are required to be as robust as possible to local changes in the pathway data.

[0056] Therefore, in Example 2, we will describe how to evaluate the robustness of the drug score to local changes in the pathway data by repeatedly calculating the drug score while randomly invalidating a portion of the pathway data 122. Note that the hardware configuration of Example 2 is the same as in Example 1, so we will omit the explanation.

[0057] Using Figure 7, we will explain a setting screen 700, which is an example of a screen operated in Embodiment 2. The operator uses the setting screen 700 to set input data, parameters, and output data. Compared to the setting screen 200 of Embodiment 1, the setting screen 700 has an invalidation ratio setting unit 701 and a repeat count setting unit 702 added to it.

[0058] The invalidation rate setting unit 701 is a box in which the invalidation rate C is set. The invalidation rate C is a parameter indicating the ratio of the number of edges to be invalidated to the total number of edges included in the pathway, and a value of 0% < C < 100% is set. The repetition count setting unit 702 is a box in which the repetition count x is set. The repetition count x is the number of times to repeat calculating the drug score while randomly invalidating a part of the pathway. That is, the drug score for the number of times set by the repetition count setting unit 702 is calculated.

[0059] Using FIG. 8, an example of the processing flow of Example 2 will be described. Since S301 to S308 and S310 are the same processes as those in Example 1, the description thereof will be omitted, and S801 and S802 added after S302, S803 added after S308, and S804 replaced with S309 will be described.

[0060] (S801) The arithmetic unit 101 repeats the loop from S801 to S803 x times.

[0061] (S802) The arithmetic unit 101 (gene chain score calculation unit) invalidates the edges included in the pathway data at a ratio C. That is, when the number of edges included in the pathway data is N, C·N / 100 edges are randomly selected and invalidated. The invalidation of an edge is a process of removing an edge randomly selected from all the edges included in the pathway data. For example, in the pathway 500 of FIG. 5A, when the edge between the circled numbers 4 and 6 is invalidated, the 5th and 12th gene chains among the gene chains listed in FIG. 5B are not extracted in S304. As a result, the drug score of the drug targeting the gene of the circled number 6 decreases.

[0062] (S803) The arithmetic unit 101 returns the process until the loop from S801 to S803 is repeated x times. Since the drug score s(m) for the drug m is calculated every time the loop is repeated, the drug score calculated at the execution count y is sy It is denoted as (m).

[0063] (S804) The calculation unit 101 creates a correspondence table between drug m and the average drug score and displays it on the display unit 106. As a prerequisite, the calculation unit 101 (drug score calculation unit) calculates the average drug score μ, which is the average value of all drug scores s(m) up to the time of execution count y, at the time of execution count y. y (m) is calculated using the following formula.

[0064]

number

[0065] The calculated average drug score is used in the correspondence table between drug m and average drug score. y By identifying drugs with a relatively high (m) value, the physician performing the procedure can select a drug that is appropriate for each individual patient.

[0066] The calculation unit 101 (drug score calculation unit) further calculates the average drug score s y Standard deviation σ at the number of executions y of (m) y (m) may be calculated using the following formula, and the calculation result may be displayed on the display unit 106 for each drug m.

[0067]

number

[0068] Drug score s y By identifying drugs m with a relatively low standard deviation of (m), the operator (physician, etc.) can select drugs that yield a robust score. y Robustness may also be evaluated by dividing the mean of (m) by its standard deviation.

[0069] An example of the output screen 900 displayed in S804 will be explained using Figures 9 and 10. The output screen 900 includes a recommended drug list 901 and a graph 902, which list drugs in order of their relevance to the disease. The recommended drug list 901 is a correspondence table created in S804, where the drug score s y The drugs m are listed in descending order of their average value of (m). The recommended drug list 901 may also include display buttons to show the target genes and detailed reports for drug m.

[0070] When the display button included in the list of recommended drugs 901 is pressed, graph 902 is displayed as a detailed report on drug m. In graph 902, the vertical axis is the drug score s. y (m) The horizontal axis is the number of executions y, and the drug score s is calculated each time the loop from S801 to S803 is repeated. y This shows the trend of (m).

[0071] Figure 9 shows an example where the display button for drug m1 is pressed, so the vertical axis of graph 902 represents the drug score s y (m1). Note that the rankings of the drugs and the drug scores may also be displayed in graph 902. Figure 11 shows an example where the rankings of the data points in graph 902 are displayed above and the drug scores are displayed below. Furthermore, the mean and standard deviation calculated by equations 4 and 5 may be displayed in the margin of the output screen 900.

[0072] Figure 10 shows an example of a screen where the list of invalid edge IDs 1001 is displayed when the cursor is moved near a data point in graph 902. The list of invalid edge IDs 1001 lists the IDs of edges that were invalidated when the execution count was y. By checking the list of invalid edge IDs 1001 and understanding which edges were invalidated when the execution count was y, the operator can investigate the cause if a drug of interest is ranked low, for example.

[0073] Following the processing flow described above, similar to Example 1, drugs are listed in order of their degree of association with the patient's disease, based on data related to the patient's lesions. The physician, acting as the operator, can select a drug appropriate for each individual patient by reviewing the drugs listed in order of their degree of association with the disease. Furthermore, the drug score calculation is repeated, along with randomly invalidating a portion of the pathway data, and the statistical values ​​of the calculation results are displayed, allowing for evaluation of the robustness of the drug score to local changes in the pathway data.

[0074] <Example 3> Example 1 presents a method for recommending potentially target genes and effective drugs based on pathway analysis, using information on the genes targeted by the drug. Clinical trials have extensively explored biomarkers related to drug efficacy and treatment prognosis. Example 3 presents a method for more evidence-based drug recommendations using information on drugs and their biomarkers.

[0075] The hardware configuration of Example 4 will be explained using Figure 11. Example 4 differs from Example 3 in that it newly includes biomarker data 124 in the storage unit 103.

[0076] Figure 12 illustrates the biomarker data 124. The biomarker data 124 includes information on the drug, the genes and genetic abnormalities that are biomarkers related to the drug's therapeutic effect, information on the therapeutic effect, and the level of evidence.

[0077] Using Figure 13, we will explain a setting screen 1300, which is an example of a screen operated in Example 3. The operator uses the setting screen 1300 to set input data, parameters, and output data. Compared to the setting screen 200 of Example 1, the setting screen 1300 has an additional condition 1301 for the gene abnormality to be extracted, instead of the condition 205 for the important gene.

[0078] An example of the processing flow in Example 3 will be explained using Figure 14. Note that S301, S305-S310 are the same processes as in Example 1, so their explanation will be omitted. After S301, S1401 and S1402, which are replaced by S302-S303, S1403 and S1404, which are added thereafter, and S1405, which is replaced by S304, will be explained.

[0079] (S1401) The calculation unit 101 extracts all gene abnormalities that satisfy predetermined conditions based on the input data and parameters read in S301 and stores them in a list. When the patient lesion data 121 set in the patient data input unit 201 or 202 of the setting screen 200 is read, gene abnormalities that satisfy the conditions specified in 1401 are extracted.

[0080] (S1402) The calculation unit 101 reads out one of the gene abnormalities i extracted in S1401.

[0081] (S1403) Next, the calculation unit 101 (drug score calculation unit) refers to the biomarker data 124 and searches for a drug d associated with the gene abnormality i. If drug d exists, the process proceeds to S1404; otherwise, it proceeds to S1405.

[0082] (S1404) The calculation unit 101 (drug score calculation unit) calculates the marker score m(i) of the gene abnormality i based on the information of the drug d searched in S1403. When the length of the gene strand is 1, the marker score m(i) is calculated by, for example, the following formula.

[0083]

number

[0084] Here, the calculation of the marker score m(i) for gene abnormality i is not limited to formula 6, and the formula may be modified. In other words, the marker score can be anything that quantifies the magnitude of drug efficacy based on the evidence level or the content of the evidence. The calculation unit 101 (drug score calculation unit) further adds the marker score m(i) to the drug score s(d) of drug d to obtain a "second drug score".

[0085] Here, h j (j=1,2,3…) represents the level of evidence in the literature for drug d that uses gene abnormality i as a biomarker, and g(h j ) is evidence level h j This is a numerical value that quantifies the gene abnormality i. If there are multiple studies on drug d that use gene abnormality i as a biomarker, the calculation unit 101 (drug score calculation unit) may adopt the index value of the study with the highest level of evidence. Note that the drug score s(d) of drug d is assumed to be initialized to 0 when it is first called.

[0086] (S1405) Next, the calculation unit 101 extracts a gene strand from the pathway in which the gene with the gene abnormality i read from the list is located at the very top and whose length is less than or equal to the propagation distance. The definitions of gene strand, very top, and propagation distance are the same as in S304 of Examples 1 and 2.

[0087] Using Figure 15, an example of the output screen 1500 displayed in S309 of Figure 14 will be explained. The output screen 1500 includes a list of recommended drugs 1501, which lists drugs in order of their degree of relevance to the disease. The list of recommended drugs 1501 is a correspondence table created in S309, in which drugs d are arranged in descending order of drug score s(d). The list of recommended drugs 1501 may also include information on target genes and biomarkers for drug d.

[0088] Multiple embodiments of the present invention have been described above. The present invention is not limited to the above embodiments, and the components can be modified and implemented without departing from the spirit of the invention. Furthermore, the multiple components disclosed in the above embodiments may be combined as appropriate. In addition, some components may be deleted from all the components shown in the above embodiments.

[0089] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are described in detail for the purpose of explaining the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace parts of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add configurations from other embodiments to the configuration of one embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations.

[0090] Furthermore, each of the aforementioned configurations, functions, processing units, and processing means may be implemented in hardware, either partially or entirely, by designing them as integrated circuits, for example. Alternatively, each of the aforementioned configurations and functions may be implemented in software by having the processor interpret and execute programs that realize each function. Information such as programs, tables, and files that realize each function can be stored in memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD. In addition, each processing unit, such as the data acquisition unit, and the storage unit 103 in the arithmetic unit 101 may be distributed over the internet, and processing may be executed in that state. Furthermore, the control lines and information lines shown are those deemed necessary for explanatory purposes, and not all control lines and information lines are necessarily shown in the actual product. In practice, it is reasonable to assume that almost all components are interconnected. [Explanation of symbols]

[0091] 100 Treatment policy planning support device 101 Calculation Unit (Data Acquisition Unit, Gene Chain Score Calculation Unit, Drug Score Calculation Unit, Display Processing Unit) 102 memory 103 Storage section 105 Input section 106 Display section 108 Network 109 Sample analyzer 110 Gene Databases 111 Drug Database 121 Data related to patient lesions 122 Pathway Data 123 Drug Data 124 Biomarker Data 200 Settings screen 500 Pathways 501 gene strands 600 Output screen 700 Settings screen 900 Output screen 1001 List of Disabled Edge IDs 1300 Settings screen 1500 Output screen

Claims

1. A treatment plan formulation support device that assists in formulating treatment plans, A data acquisition unit that acquires data related to the patient's lesions by analyzing the patient's lesion site or bodily fluids, A storage unit that stores pathway data, which is pathway data representing the sequence of genes as a directed graph, and drug data, which includes a drug and the genes that the drug targets for treatment. A gene chain score calculation unit extracts gene chains from the pathway data in which the upstream gene is a key gene that satisfies the conditions determined by the patient's lesion data, and calculates a gene chain score for each gene chain that represents the degree of association between the key gene and the downstream gene of the gene chain. A drug score calculation unit calculates a drug score by adding the gene chain score of gene chains whose downstream gene is the same gene to the drug targeting the downstream gene in the drug data. A treatment plan formulation support device characterized by being equipped with the following features.

2. A treatment plan planning support device according to claim 1, The system further includes a display processing unit that lists the drugs in descending order of their drug scores and displays them on the display unit. A treatment plan planning support device characterized by the following features.

3. A treatment plan planning support device according to claim 1, The drug score calculation unit, The calculation of the drug score is repeated while randomly disabling some of the edges included in the pathway. A treatment plan planning support device characterized by the following features.

4. A treatment plan planning support device according to claim 3, The drug score calculation unit, The average value of the repeatedly calculated drug score is calculated, The system further includes a display processing unit that lists the genes in descending order of the average drug score and displays them on the display unit. A treatment plan planning support device characterized by the following features.

5. A treatment plan planning support device according to claim 3, The drug score calculation unit, To calculate the standard deviation of the drug score that is repeatedly calculated, A treatment plan planning support device characterized by the following features.

6. A treatment plan planning support device according to claim 3, A display processing unit displays a graph showing the trend of the repeatedly calculated drug score on the display unit. To further prepare, A treatment plan planning support device characterized by the following features.

7. A treatment plan planning support device according to claim 6, The display processing unit, Displaying disabled edges in the aforementioned graph, A treatment plan planning support device characterized by the following features.

8. A treatment plan planning support device according to claim 1, The aforementioned storage unit includes: The data includes biomarkers such as drugs and genes and genetic abnormalities that are biomarkers related to the therapeutic effect of the drugs, and is further stored. A treatment plan planning support device characterized by the following features.

9. A treatment plan planning support device according to claim 8, The drug score calculation unit, For genetic abnormalities that meet the conditions defined by the patient's genetic abnormality data, the drug data is searched, and if information on a drug matching the genetic abnormality is found, a marker score is calculated that quantifies the magnitude of the drug's efficacy based on the evidence level or the content of the evidence. A treatment plan planning support device characterized by the following features.

10. A treatment plan planning support device according to claim 9, The drug score calculation unit, The value obtained by adding the marker score to the drug score is defined as the second drug score, and the second drug score is defined as the drug score. A treatment plan planning support device characterized by the following features.

11. A treatment plan planning support device according to claim 9, A treatment plan planning support device further comprising a display processing unit that lists the drugs in descending order of their drug scores and displays them on a display unit.

12. A treatment plan planning support device that assists in formulating treatment plans, By analyzing the patient's lesion site or bodily fluids, we obtain data related to the patient's lesions. The system references a memory unit that stores pathway data, which is pathway data representing the sequence of genes as a directed graph, and drug data, which includes drugs and the genes targeted by those drugs. The gene chain whose upstream is a key gene that satisfies the conditions determined by the patient's lesion data is extracted from the pathway data, and a gene chain score representing the degree of association between the key gene and the downstream gene of the gene chain is calculated for each gene chain. The drug score is calculated by adding the gene chain score of gene chains whose downstream gene is the same gene to the drug data for each drug that targets the downstream gene. A treatment plan formulation support method characterized by the following.

Images