Genetic test method and genetic test system

Abstract

The purpose is to detect at high sensitivity patients having a high risk of postoperative recurrence based on an expression profile of RNA extracted from platelets collected from cancer patients. To this end, a first expression profile, which is an expression profile of RNA extracted from platelets collected from the postoperative blood of a subject, is compared with cancer sufferer group expression profile, which is a set of expression profiles of RNA extracted from platelets collected from the blood of cancer sufferers belonging to the cancer sufferer group, and a non-cancer sufferer group expression profile, which is a set of expression profiles of RNA extracted from platelets collected from the blood of non-cancer sufferers belonging to the non-cancer sufferer group, and which of the expression profiles the first expression profile is similar to is determined.

Description

Genetic testing method and genetic testing system

[0001] The present disclosure relates to a genetic testing method and a genetic testing system, and more specifically to a genetic testing method and a genetic testing system using RNA extracted from platelets in blood.

[0002] In early-stage lung cancer, postoperative recurrence after surgical resection is a major cause of poor prognosis, and identifying predictive markers for postoperative recurrence could have a significant impact on improving prognosis. Predicting postoperative recurrence is facilitated by preoperative gene expression information and clinicopathological information from tumor tissue, as well as minimally invasive liquid biopsy testing using liquid samples such as blood and body fluids. Among liquid biopsies, circulating tumor DNA (ctDNA) can be easily extracted and purified from blood, and its application to cancer diagnosis has been explored. By detecting ctDNA in blood collected after surgery and measuring the ctDNA mutation copy number or mutation frequency, the presence or absence of postoperative minimal residual disease (MRD) can be determined. This method for predicting postoperative recurrence is highly specific because it predicts postoperative recurrence based on direct information, namely, the mutated genes contained in ctDNA released from the tumor. In colorectal cancer, the presence or absence of MRD is determined after surgery using ctDNA; if ctDNA is detected and it is determined that MRD is present, the patient's prognosis is poor due to recurrence, whereas if ctDNA is not detected and it is determined that MRD is absent, the patient's prognosis is good and there is no recurrence, and furthermore, it has been reported that the patient's prognosis remains good even without postoperative chemotherapy (Non-Patent Document 1).

[0003] Meanwhile, platelets, present in large quantities in the blood of cancer patients, have attracted attention as a new biomarker for liquid biopsy other than ctDNA. It has been revealed that platelets, when affected by the tumor environment, become tumor-educated platelets (TEPs), which exhibit gene expression states different from those of non-cancer patients. Platelets are produced from megakaryocytes in the bone marrow and are non-nucleated intracellular fragments containing megakaryocyte-derived RNA transcripts. It has been reported that both mRNA splicing due to external stimuli and uptake of circulating mRNA occur in platelets while circulating in the blood, and these mechanisms are presumed to result in RNA that reflects the pathological condition. Because platelets are abundant in the blood, they are expected to serve as a new biomarker source that can be used for research and clinical diagnosis without the need for excessive blood sampling (Non-Patent Document 2, Non-Patent Document 3).

[0004] Kotani D et al., "Molecular residual disease and efficacy of adjuvant chemotherapy in patients with colorectal cancer" Nature Medicine, Vol. 29, 2023, p.127-134Best MG et al., "RNA-Seq of Tumor-Educated Platelets Enables Blood-Based Pan-Cancer, Multiclass, and Molecular Pathway Cancer Diagnostics" Cancer Cell, Vol. 28, 2015, p.666-676In 't Veld SGJG and Wurdinger T "Tumor-educated platelets" Blood, Vol.133, 2019, p.2359-2364

[0005] Although detecting ctDNA in postoperative blood samples to determine the presence or absence of MRD and predict postoperative recurrence has shown some success, the amount of ctDNA released from minute residual tumors is extremely small, limiting its detection. While recent advances in ctDNA measurement devices such as next-generation sequencers (NGS) and digital PCR, as well as their assay methods, have improved sensitivity to the point where several copies of ctDNA can be detected, there are cases of recurrence even when ctDNA is not detected in postoperative blood samples, pointing to the lack of sensitivity of ctDNA-based postoperative recurrence prediction tests. Possible causes of this lack of sensitivity include low levels of ctDNA eluted into the blood, or the number of mutation copies or frequency of mutations in ctDNA below the detection limit of the measurement method. While attempts to improve sensitivity have been made to collect blood samples multiple times postoperatively to measure ctDNA, cases of recurrence despite no ctDNA detected in multiple measurements have been reported. Furthermore, in clinical studies of lung cancer, when MRD was assessed using ctDNA, the disease-free survival period was longer in patients in whom ctDNA was not detected compared to patients in whom ctDNA was detected. However, when comparing patients in whom ctDNA was not detected and postoperative adjuvant therapy (postoperative adjuvant chemotherapy) was administered to them, the disease-free survival period was worse in the patients who did not receive it, so there is a risk that patients who require postoperative adjuvant therapy will be missed when predicting recurrence based on the presence or absence of ctDNA.

[0006] The object of the present invention is to provide a genetic testing method capable of detecting MRD with high sensitivity, which compensates for the lack of sensitivity of the highly specific ctDNA-based MRD determination technology, and to stratify patients into a group that will relapse after surgery and a group that will not relapse, so as not to miss any patients who will relapse after surgery.

[0007] A genetic testing method that is one embodiment of the present disclosure is a genetic testing method for testing the risk of postoperative recurrence in a subject who is a cancer patient, and includes: a first expression analysis step of measuring a first expression profile, which is an expression profile of RNA extracted from platelets collected from the subject's blood after surgery; and an examination step of comparing the first expression profile with a cancer-bearer group expression profile, which is a collection of expression profiles of RNA extracted from platelets collected from the blood of cancer-bearers belonging to a cancer-bearer group, and a non-cancer-bearer group expression profile, which is a collection of expression profiles of RNA extracted from platelets collected from the blood of non-cancer-bearers belonging to a non-cancer-bearer group, to determine which expression profile the first expression profile resembles; and in the examination step, if the first expression profile resembles the cancer-bearer group expression profile, the risk of recurrence is assessed as high, and if the first expression profile resembles the non-cancer-bearer group expression profile, the risk of recurrence is assessed as low.

[0008] The expression profile of RNA extracted from platelets collected from cancer patients makes it possible to detect patients at high risk of postoperative recurrence with high sensitivity. Other issues and novel features will become apparent from the description of this specification and the accompanying drawings.

[0009] 1 is a flowchart for examining the risk of recurrence from the expression profile of platelet-derived RNA from a subject before and after surgery.

[0023] FIG. 1 is a schematic diagram showing an example of the results of comparing the pre- and post-operative expression profile of a subject with the expression profiles of a group of healthy subjects and a group of cancer patients prepared in advance.

[0024] FIG. 1 is an example of an explanatory diagram provided in a test result report.

[0025] FIG. 1 is a flowchart for examining the risk of recurrence from the expression profile of platelet-derived RNA from a subject after surgery.

[0026] FIG. 1 is a schematic diagram showing an example of the results and interpretation when ctDNA testing and TEP testing are performed simultaneously.

[0027] FIG. 1 is a schematic diagram showing an example of the results and interpretation when ctDNA testing and TEP testing are performed sequentially.

[0028] FIG. 1 is a schematic diagram showing an example of the configuration of a genetic testing system.

[0029] FIG. 1 is a functional block diagram of a testing device.

[0030] FIG. 1 is an example of the data structure of an expression profile of a cancer-bearing group.

[0031] FIG. 1 is a schematic diagram showing an example of a method for separating a platelet fraction by centrifugation.

[0032] FIG. 1 is a schematic diagram showing an example of a method for separating a platelet fraction by ultrasonic separation.

[0033] FIG. 1 is a heat map summarizing the results of whole transcriptome analysis of platelet-derived RNA, PBMCs, and tumor tissue. 1 is a diagram showing the results of clustering ultrasonically separated platelet gene expression data from cancer patients and healthy subjects in Example 1. In Example 1, it is a diagram showing genes whose expression was enhanced in the clusters into which cancer patients were classified by gene ontology (GO) analysis. In Example 1, it is a diagram showing genes whose expression was suppressed in the clusters into which cancer patients were classified by gene ontology (GO) analysis. In Example 2, it is a diagram showing the results of clustering to determine whether the expression profile of a non-small cell lung cancer patient after surgery is similar to the expression profile of a healthy subject or the expression profile of a cancer patient before surgery. In Example 3, it is a graph in which the gene sets characteristic of cancer patients used in Example 2 are arranged on the horizontal axis in descending order of expression level ratio in cluster α and cluster β, and the median expression levels of samples classified into each of cluster α and cluster β are plotted on the vertical axis.In Example 3, with regard to the expression profile of platelet-derived RNA in postoperative non-small cell lung cancer patients, the gene sets characteristic of cancer patients used in Example 2 were arranged on the horizontal axis in descending order of expression level ratio in cluster α and cluster β, and the expression levels of platelet-derived RNA in postoperative non-small cell lung cancer patients were plotted.

[0010] The objectives, features, advantages, and ideas related thereto of the present disclosure will be apparent to those skilled in the art from the description of this specification. Those skilled in the art can easily reproduce the contents of the present disclosure from the description of this specification. The embodiments and specific examples of the invention described below show preferred embodiments of the present disclosure and are presented for illustration or explanation purposes, and are not intended to limit the present disclosure thereto. It will be apparent to those skilled in the art that various changes and modifications can be made based on the description of this specification without departing from the intent and scope of the present disclosure disclosed herein.

[0011] As will be explained in detail in Example 1 below, the inventors have discovered that the gene expression profile of RNA extracted from platelets makes it possible to distinguish between cancer-bearing and non-cancer-bearing individuals (those in a cancer-bearing state), and that when gene expression profiles of RNA extracted from platelets and RNA extracted from tumors are compared, the results of gene expression analyses of platelets and tumors are completely inconsistent, demonstrating that TEP (platelets) is a method for indirectly obtaining information about tumors. Furthermore, when genes with differential gene expression between platelets from cancer-bearing and non-cancer-bearing individuals were extracted, activation of signaling pathways leading to activation of platelet functions such as aggregation and angiogenesis associated with tumor formation and growth was observed, rather than genes specifically activated or suppressed in tumors. These results suggest that TEP is sensitized to tumor signals by platelets and cells affected by tumors around the tumor, and it is expected that TEP will be able to detect MRD with higher sensitivity than detecting minute amounts of ctDNA leaked from tumors. Therefore, as will be described in detail in Examples 2 and 3 below, the present inventors found that by comparing the expression profiles of RNA contained in the platelets of cancer-bearing and non-cancer-bearing individuals to extract a set of genes with differences in gene expression, and using this gene set, it is possible to distinguish between patients with postoperative recurrence and patients without recurrence, leading to the completion of the present invention.

[0012] The present disclosure is useful in fields such as basic research, testing, and drug discovery that involve molecular diagnosis of cancer, such as early cancer screening, evaluation of treatment efficacy, prediction of recurrence, and monitoring.

[0013] (1) Genetic Testing Method The genetic testing method (TEP test) of the present disclosure will be specifically described with reference to the drawings.

[0014] FIG. 1 is a flowchart (first embodiment) for assessing the risk of recurrence from the expression profiles of platelet-derived RNA before and after surgery in a subject who is a cancer patient, and examining whether postoperative adjuvant therapy can be omitted.

[0015] First, blood samples are collected from a subject before and after surgery, and a platelet fraction is extracted (S101). There are no particular restrictions on the subject, as long as the subject is a cancer patient who can be operated on. The cancer type can be any type, including lung cancer, breast cancer, colon cancer, pancreatic cancer, and malignant lymphoma. The stage is not limited to stage I or II, as long as the cancer is operable. The timing of blood collection after surgery may be one week, two weeks, one month, two months, or six months after surgery, or any timing within six months after surgery. Furthermore, blood may be collected once after surgery, or multiple times at different times after surgery. If blood is collected multiple times, this flow chart is repeated each time blood is collected.

[0016] The method for fractionating the platelet fraction is not limited to a specific method. For example, a blood collection tube containing blood may be centrifuged using a swing rotor centrifuge, and the platelets contained in the fraction above the buffy coat may be separated using a pipette tip. Alternatively, platelets may be separated by ultrasonic separation. In this method, blood is added from the side of a channel through which a buffer, whose density is lower than that of white blood cells and red blood cells but higher than that of platelets, flows. After passing through a region where a standing wave is generated so that the center of the channel is the antinode of the standing wave, platelets from which white blood cells and red blood cells have been removed are separated from the side of the channel.

[0017] Next, RNA contained in the platelet fraction of the subject is extracted, and its expression profile is compared with the expression profiles of a group of healthy subjects and a group of cancer patients prepared in advance (S102). In this flow, the group of healthy subjects corresponds to the non-cancer group, and the group of cancer patients corresponds to the cancer group.

[0018] The genes for which expression profiles are compared are selected by collecting blood samples from multiple healthy individuals and cancer patients in advance, analyzing the expression of RNA extracted from platelets in the blood using a next-generation sequencer (NGS), microarray, real-time PCR, or digital PCR, and selecting characteristic genes whose gene expression levels differ between healthy individuals and cancer patients. Genes selected for expression profile comparison are sometimes called characteristic genes. The analyzed expression profiles are stored as a database.

[0019] In this case, the expression profiles of the cancer patient group prepared in advance may be a group collected by cancer type, such as lung cancer, breast cancer, colon cancer, pancreatic cancer, or malignant lymphoma; a group collected by stage or presence or absence of metastasis; or a group collected regardless of cancer type or stage. Furthermore, the selected feature genes are preferably a combination of genes that can accurately distinguish between a group of healthy subjects and a group of cancer patients by clustering. It is also possible to select the optimal combination and number of genes as feature genes by dividing the group by cancer type or stage, or it is also possible to select the optimal combination and number of genes as feature genes regardless of cancer type or stage. The number of feature genes selected may be 5, 10, 100, 200, or 400.

[0020] To extract characteristic genes, a machine learning discrimination model for discriminating between a group of healthy subjects and a group of cancer patients may be constructed using expression profiles of multiple healthy subjects and cancer patients as a training set. Expression profiles of multiple healthy subjects and cancer patients are obtained, gene expression information is normalized, and low-expression genes are filtered. Genes specifically expressed in the cancer patient group are then extracted, and a discrimination model is constructed using a machine learning classification method, for example, K-nearest neighbor classification or logistic regression classification. The accuracy of the constructed discrimination model may be evaluated by using as a validation set preoperative blood samples from cancer patients whose recurrence or non-recurrence is unknown, collected independently and prospectively, and blood samples from healthy subjects. Furthermore, accuracy may be evaluated by dividing preoperative and postoperative blood samples from cancer patients into a recurrence group and a non-recurrence group based on the presence or absence of recurrence several years after surgery.

[0021] The method for quantifying the expression profile is not limited to a specific method as long as it can identify the type of gene and quantify its expression level. For example, NGS, microarray, real-time PCR, digital PCR, etc. can be used. The method used to obtain the expression profiles of healthy individuals and cancer patients stored as a database may be the same as the method used to quantify the expression profile of the subject. NGS is preferably used when the number of feature genes used to compare the expression profiles is 10 or more, more preferably 100 or more. Microarray is preferably used when the number of feature genes is 10 or more, and preferably 100 or less. Real-time PCR is preferably used when the number of feature genes is 10 or less. Digital PCR is preferably used when the number of feature genes is 20 or less, and more preferably when any of the feature genes is expressed at a low expression level of 1,000 copies or less per measurement.

[0022] The expression profile of a subject is compared with the expression profiles of multiple healthy individuals and cancer patients prepared in advance as follows. For example, the expression profiles of the subject, including the preoperative and postoperative ones, are clustered for preselected feature genes, and the similarity is evaluated based on whether the subject's expression profile is included in the cluster of expression profiles of the healthy individuals or the cancer patients. The comparison method may use a discriminant model constructed when extracting feature genes. In this case, weighting is performed based on the discriminant model for each preselected feature gene, and the similarity is determined by scoring.

[0023] As a result, if the subject's preoperative expression profile is more similar to that of the cancer patient group than to that of the healthy control group, and if the subject's postoperative expression profile is more similar to that of the healthy control group than to that of the cancer patient group (Yes in S103), the risk of recurrence is evaluated as low, and, for example, omission of postoperative adjuvant therapy is proposed as a treatment policy (S105).On the other hand, if the subject's preoperative expression profile is similar to that of the healthy control group or if the subject's postoperative expression profile is similar to that of the cancer patient group (No in S103), the risk of recurrence is evaluated as high, and, for example, implementation of postoperative adjuvant therapy is proposed (S104).

[0024] Furthermore, if the postoperative expression profile of even one of the multiple blood samples taken from a subject after surgery is judged as No in step S103, the risk of recurrence is assessed as high regardless of previous evaluations, and postoperative adjuvant therapy is suggested.

[0025] Figure 2A is a schematic diagram showing an example of the results of comparing a subject's pre- and post-operative expression profile with the expression profiles of a group of healthy subjects and a group of cancer patients prepared in advance. For the characteristic genes that characterize the pre-selected healthy subjects and cancer patients, the subject's pre- and post-operative expression profile is clustered together with the expression profile data of the healthy subjects and cancer patients in the database. The clustering results are displayed as a heat map in which the expression level of each gene is colored, and groups with similar balances of expression levels are displayed in separate clusters.

[0026] In Figure 2A, as a result of clustering the expression profile of the subject, the subject's preoperative expression profile 203 is classified into the same cluster as the expression profile 201 of a group of cancer patients prepared in advance, and the subject's postoperative expression profile 204 is classified into the same cluster as the profile 202 of a group of healthy subjects prepared in advance. From this result, it is reported that the subject's risk of recurrence can be evaluated as low, and that postoperative adjuvant therapy may be omitted. Together with the report, a diagram showing the clustering results of the subject's expression profile, such as that of Figure 2A, may also be provided to the user.

[0027] Figure 2B is an example of an explanatory diagram provided with a report indicating whether or not postoperative adjuvant therapy should be omitted. In Figure 2B, if the subject's preoperative and postoperative expression profile resembles that of a cancer patient group, it is considered positive, and if it resembles that of a healthy control group, it is considered negative, and the corresponding test results are marked. Users can compare reports such as those shown in Figures 2A and 2B with clinical information such as information on the tumor removed during surgery to consider whether or not to implement the proposed postoperative adjuvant therapy.

[0028] 3 is a flowchart (second embodiment) for assessing the risk of recurrence from the expression profile of platelet-derived RNA after surgery in a subject and determining whether postoperative adjuvant therapy can be omitted. Note that duplicated explanations of content similar to that in the flowchart shown in FIG. 1 will be omitted.

[0029] First, blood is collected from the subject after surgery, and a platelet fraction is separated (S301). Next, RNA contained in the subject's platelet fraction is extracted, and its expression profile is compared with the expression profiles of a pre-prepared non-recurrence patient group and a pre-prepared recurrence patient group (S302). Non-recurrence patients are cancer patients who did not recur after surgery, while recurrence patients are cancer patients who recur after surgery. Although blood from non-recurrence and recurrence patients is collected after surgery, the recurrence / non-recurrence status may not be known at this stage, and the recurrence / non-recurrence classification can be performed ex post. In this flow, the non-recurrence patient group corresponds to the non-cancer group, and the recurrence patient group corresponds to the cancer group.

[0030] In this case, the expression profiles of the recurrent patient group and the non-recurrent patient group prepared in advance may be groups collected by cancer type (e.g., lung cancer, breast cancer, colon cancer, pancreatic cancer, malignant lymphoma), groups collected by stage or presence or absence of metastasis, or groups collected regardless of cancer type or stage. Furthermore, the selected feature genes are preferably a combination of genes that can accurately distinguish the non-recurrent patient group and the recurrent patient group into two groups by clustering. The optimal combination and number of genes can be selected as feature genes by dividing the patients into groups by cancer type or stage, or the optimal combination and number of genes can be selected as feature genes regardless of cancer type or stage. The number of feature genes selected may be 5, 10, 100, 200, or 400.

[0031] The expression profile of a subject is compared with the expression profiles of multiple pre-prepared non-recurrent and recurrent patients as follows. For example, pre-selected feature genes are clustered together with the subject's postoperative expression profile, and the similarity is evaluated based on whether the subject's expression profile is included in the cluster of expression profiles of the non-recurrent or recurrent patient group. The comparison method may use a discriminant model constructed when extracting feature genes.

[0032] As a result, if the postoperative expression profile of the subject is similar to that of the non-recurrence patient group (Yes in S303), the risk of recurrence is evaluated as low, and for example, omission of postoperative adjuvant therapy is proposed as a treatment policy (S305).On the other hand, if the postoperative expression profile of the subject is similar to that of the recurrence patient group (No in S303), the risk of recurrence is evaluated as high, and for example, implementation of postoperative adjuvant therapy is proposed (S304).

[0033] 4A and 4B are schematic diagrams showing an example of test results and interpretations of test results when ctDNA testing is performed in addition to TEP testing to consider whether to perform or omit postoperative adjuvant therapy. As described above, ctDNA testing is a highly specific testing method, and by using it in combination with the highly sensitive TEP testing of this embodiment, it is expected that the risk of recurrence can be evaluated comprehensively with high accuracy. Here, the TEP testing may be either the first embodiment shown in FIG. 1 or the second embodiment shown in FIG. 3.

[0034] Figure 4A is a schematic diagram showing an example of the results and interpretation when ctDNA and TEP tests are performed simultaneously. If either the ctDNA or TEP test is positive, postoperative adjuvant therapy is recommended; if both the ctDNA and TEP tests are negative, postoperative adjuvant therapy is suggested to be omitted. The positive or negative result of the ctDNA test is determined based on the mutation information obtained from the tumor gene mutation analysis results for each patient, by quantifying the mutation frequency or mutation copy number of multiple or a representative mutation using a device such as NGS, real-time PCR, or digital PCR. For example, a positive result is determined when the mutation frequency or mutation copy number exceeds a threshold. The threshold may be determined in advance from a population of healthy individuals, or the detection sensitivity of the device may be used as the threshold.

[0035] Figure 4B is a schematic diagram showing an example of the results and interpretation when ctDNA and TEP tests are performed sequentially. TEP testing is performed on patients who are negative in ctDNA testing. Patients who are negative in both ctDNA and TEP testing are assessed as having a low risk of recurrence, and we suggest omitting postoperative adjuvant therapy.

[0036] In the example of Figure 4A, a treatment plan for postoperative patients can be determined quickly, but the cost of two tests is required for all postoperative patients. In the example of Figure 4B, TEP tests are not performed for patients whose ctDNA test is positive, so the cost of testing can be reduced, but it takes time to determine a treatment plan because the two tests are performed sequentially.

[0037] (2) Genetic Testing System A genetic testing system for carrying out the above-described genetic testing method will now be described. Fig. 5A shows a schematic diagram of an example of the configuration of the genetic testing system.

[0038] The RNA extraction unit 502 separates a platelet fraction from the collected blood 501 and extracts RNA from the fractionated platelet fraction. Here, an example is shown in which the RNA extraction unit 502 includes a centrifuge 503 that can centrifuge a centrifuge tube, a fraction collection device 504 that collects the platelet fraction from the centrifuge tube after centrifugation with a pipette, and an RNA extraction device 505 that extracts RNA from the platelet fraction. The RNA extraction unit 502 may be configured as an integrated device, or each device may be configured as an independent device. Furthermore, instead of using the fraction collection device 504, the platelet fraction from the centrifuge tube after centrifugation may be manually collected by the user.

[0039] The gene expression analysis unit 506 measures the expression profile of the subject. The gene expression analyzer 508 for measuring gene expression profiles includes an NGS, real-time PCR, or digital PCR. The type of device used as the gene expression analyzer 508 can be determined based on the number of genes used to compare expression profiles and the throughput. The user mixes a reagent for measuring an expression profile appropriate for the type of gene expression analyzer 508 with RNA extracted from the platelet fraction, reacts the mixture, and then sets the mixture in the gene expression analyzer 508. For this reason, the gene expression analyzer 508 may be equipped with an automatic reagent mixer 507 for mixing a reagent for measuring an expression profile appropriate for the type of gene expression analyzer 508 with RNA extracted from the platelet fraction.

[0040] The testing device 509, which tests a subject's risk of postoperative recurrence, evaluates the risk of postoperative recurrence based on the similarity determined by comparing the subject's expression profile measured by the gene expression analysis unit 506 with an existing expression profile. The testing device 509 is realized by an information processing device (computer) 510, which primarily includes a processor (CPU) 511, memory 512, storage device 513, input interface (I / F) 514, output I / F 515, communication I / F 516, and bus 517, as shown in FIG. 5B . The processor 511 functions as a functional unit that provides predetermined functions by executing programs loaded into the memory 512. The storage device 513 stores data and programs used by the functional unit. The input I / F 514 is connected to input devices such as a keyboard, pointing device, and operation panel, and the output I / F 515 is connected to a display device. The communication I / F 516 enables communication with other information processing devices via a network. These are communicatively connected to each other via the bus 517.

[0041] In the following description, when describing processing by a program, the program or functional units may be described as the main components, but the main hardware components are a processor or an information processing device configured to include the processor. The information processing device executes processing according to a program read into memory using resources such as memory and a communication interface as appropriate through the processor. While FIG. 5B shows an example of a CPU as the processor, a GPU or the like may also be used. Furthermore, processing to realize a function is not limited to software program processing, but can also be implemented using a dedicated circuit. The dedicated circuit may be an FPGA, an ASIC, or the like.

[0042] FIG. 5C shows a functional block diagram of the testing device 509. The input unit 521 receives the expression profile of the subject measured by the gene expression analysis unit 506. As described in FIG. 1 or 3, the comparison unit 522 compares the expression profile of the subject with a cancer-bearer group expression profile 526, which is a collection of RNA expression profiles extracted from platelets collected from the blood of cancer-bearing individuals belonging to a cancer-bearing group, and a non-cancer-bearer group expression profile 527, which is a collection of RNA expression profiles extracted from platelets collected from the blood of non-cancer-bearing individuals belonging to a non-cancer-bearing group, both stored in the memory unit 525. The suggestion unit 524 presents the comparison results to the user. The presented content may include a report on the implementation or omission of postoperative adjuvant therapy. Furthermore, if ctDNA test results for the subject are obtained from the ctDNA test result input unit 523, a report integrating the two test results is created. The report created by the suggestion unit 524 may be displayed on a monitor or printed on a printer.

[0043] The cancer-bearing group expression profile 526 and the cancer-non-bearing group expression profile 527 have a similar data structure and are stored in storage unit 525. The cancer-bearing group expression profile 526 and the non-cancer-bearing group expression profile 527 are obtained by clustering the cancer-bearing group expression profile and the non-cancer-bearing group expression profile together, and a predetermined number (here, N) of gene sets with a large expression level ratio between a cluster mainly composed of the cancer-bearing group expression profile and a cluster mainly composed of the non-cancer-bearing group expression profile are selected as feature genes, and the expression levels of the feature genes for each sample are stored.

[0044] FIG. 5D shows an example of the data structure of the expression profile 526 of the cancer-bearing group, and an example of the data structure of the expression profile 527 of the cancer-free group is similar.

[0045] As described in the Examples below, it has been confirmed that genes whose gene expression levels differ between cancer-bearing and non-cancer-bearing individuals in platelet RNA are not genes whose expression levels are increased or decreased in tumor tissue, but rather that many of them are involved in the activation of signaling pathways leading to the activation of platelet functions such as aggregation ability and angiogenesis. Therefore, the characteristic genes stored as the expression profile 526 of the cancer-bearing individuals and the expression profile 527 of the non-cancer-bearing individuals preferably include genes involved in the activation of signaling pathways leading to the activation of platelet function or angiogenesis. More specifically, it is preferable that the signaling pathways related to platelet function and angiogenesis include genes involved in the pathway of platelet response to wounds and the pathway of platelet activation, and further include genes involved in the signaling pathways of additional functions, such as the VEGFA-VEGFR2 pathway, the RHO GTPase pathway, and the cytokine production pathway.

[0046] Example 1 First, in Example 1, blood samples were collected from healthy individuals and preoperative non-small cell lung cancer (NSCLC) patients, and the results of platelet-derived RNA expression analysis were described. For comparison, human peripheral blood mononuclear cells (PBMCs) and tumor tissue were also collected as samples other than platelets. Furthermore, to evaluate the influence of the platelet collection method, samples were prepared by two methods: centrifugation and ultrasonic separation.

[0047] Blood samples from 10 healthy volunteers and 10 preoperative non-small cell lung cancer patients were collected in citrate blood collection tubes, and platelet fractions were isolated using two separation methods: centrifugation and ultrasonic separation. Figure 6A shows a schematic diagram of the platelet fraction isolation by centrifugation. In this example, 3 mL of blood was centrifuged at a centrifugal force of 150 g for 20 minutes using a swing-out rotor centrifuge, and the portion up to 5 mm above the buffy coat 602 was isolated as platelet fraction 601. Note that a red blood cell fraction 603 was present below the buffy coat 602.

[0048] Platelet fractionation by ultrasonic separation was performed using the AcouWash system (AcouSort). As shown in Figure 6B, the AcouWash system has two inlets 612 and 613 and two outlets 614 and 615 at both ends of a microchannel equipped with a piezoelectric transducer 611. Two milliliters of blood diluted 2.5 times with PBS (phosphate-buffered saline) was added to the microchannel through the side inlet 612, and 70% Ficoll-Paque PREMIUM 1.084 (GE Healthcare) was added through the center inlet 613. Blood cells in the added blood were ultrasonically separated by applying a 2 MHz standing wave to the microfluidic channel 616 using the piezoelectric transducer 611. White blood cells (WBCs) and red blood cells (RBCs) were collected through the center outlet 615, and platelets (Platelets) were collected through the side outlet 614.

[0049] To obtain PBMCs, 0.4 mL of blood was diluted 2-fold with PBS and layered on 0.6 mL of Ficoll-Paque PREMIUM 1.084. The mixture was centrifuged at 400 g for 30 minutes using a swing-out rotor centrifuge to remove the plasma fraction, and the buffy coat on the Ficoll layer was then separated. The buffy coat contained PBMCs.

[0050] Total RNA was extracted from the above platelet samples and PBMC samples using Isogen-LS (Nippon Gene).

[0051] In addition, RNA was isolated from formalin-fixed, paraffin-embedded (FFPE) specimens of tumor tissues excised during surgery using AllPrep DNA / RNA FFPE Kit (Qiagen).

[0052] Whole-transcriptome analysis was performed on these four samples using the AmpliSeq Transcriptome Human Gene Expression Kit (Thermo Fisher Scientific). For library preparation, cDNA was generated from 10 ng of total RNA using the SuperScript VILO cDNA Synthesis kit (Thermo Fisher Scientific). PCR Master Mix and the AmpliSeq human transcriptome gene expression primer pool (Thermo Fisher Scientific) were then added, and the cDNA was amplified for 12 cycles. After multiplex PCR, Ion Xpress Barcode Adapters (Thermo Fisher Scientific) were ligated to the PCR products and purified using Agencourt AMPure XP beads (Beckman Coulter). The purified libraries were pooled and sequenced using an Ion Torrent S5 instrument with the Ion 550 Chip Kit (Thermo Fisher Scientific) and Ion Torrent Suite v5.12 software (Thermo Fisher Scientific) for base calling, alignment to the human reference genome (hg19), quality control, and sequencing. The raw reads were then automatically analyzed using the AmpliSeqRNA plugin to generate gene-level expression values ​​for all 20,802 RefSeq human genes.

[0053] The expression values ​​of all 20,802 RefSeq human genes were normalized using Transcriptome Analysis Console ver. 4.0.2 (Thermo Fisher Scientific) to filter out low-expressing genes. Of the total 15,000 genes, 200 were selected based on the top-ranking expression ratio between cancer patients and healthy controls in platelet-derived RNA signatures obtained from ultrasonic separation. These genes were subjected to hierarchical clustering using the average linkage method, with the Pearson correlation coefficient minus 1 used as the distance measure. Functional and pathway enrichment analysis was performed using Metascape.

[0054] Figure 7 shows a heat map summarizing the results of whole transcriptome analysis of platelet-derived RNA (using two different separation methods), PBMCs, and tumor tissues. The heat map is colored according to the expression level of each gene, with higher expression levels indicated by darker colors. As shown in Figure 7, the whole transcriptome analysis clearly distinguished four major clusters (clusters A to D): platelets from cancer patients (NSCLC), platelets from healthy volunteers, PBMCs, and tumor tissues from cancer patients. Specifically, clusters A and B are clusters of platelet samples, cluster C is a cluster of PBMC samples, and cluster D is a cluster of tumor tissue samples. All 13 samples classified in cluster A were platelets from cancer patients, and of the 23 samples classified in cluster B, 19 samples, excluding four samples, were platelets from healthy volunteers.

[0055] On the other hand, platelets obtained by ultrasonic separation and centrifugation were contained in the same clusters in almost equal amounts, and were not clearly distinguishable. Specifically, cluster A contained 8 samples of platelets obtained by ultrasonic separation and 5 samples of platelets obtained by centrifugation, while cluster B contained 11 samples of platelets obtained by ultrasonic separation and 12 samples of platelets obtained by centrifugation.

[0056] Thus, the expression profiles of platelets were distinct from those of PBMCs and tumor tissues.Furthermore, the gene expression profiles of platelets from cancer patients were also distinct from those of platelets from healthy individuals.

[0057] To investigate signaling pathways associated with platelets from cancer patients, we applied gene ontology (GO) analysis to ultrasound-separated platelet gene expression data from cancer patients (non-small cell lung cancer patients) and healthy volunteers. The results are illustrated using Figures 8A-C. Figure 8A shows the results of unsupervised hierarchical clustering analysis of the top 200 genes by expression level ratio. Cancer patients and healthy volunteers were classified into different clusters. Most platelets from cancer patients (NSCLC) were classified into cluster X, whereas all platelets from healthy volunteers were classified into cluster Y. This suggests that cancer patients have unique platelet gene expression profiles and that platelet RNA from cancer patients may be affected by the presence of tumors in the body.

[0058] GO analysis was performed for each cluster gene set. Figure 8B shows the group of genes whose expression was increased in cluster X, and Figure 8C shows the group of genes whose expression was suppressed in cluster X. In addition to the pathway 801 for platelet activation, signaling, and aggregation, the set of highly expressed genes enriched in cluster X also predicted additional functional pathways, such as VEGFA-VEGFR2 signaling 802, ROCKs activation by RHO GTPase 803, and cytokine production 804.

[0059] The results of GO analysis revealed differences in gene expression between platelet-derived RNA from healthy individuals and cancer patients, and revealed that gene groups related to the pathways of platelet response to wounds and platelet activation, which are signaling pathways related to platelet function, as well as gene groups related to the VEGFA-VEGFR2 pathway, RHO GTPase pathway, and cytokine production pathway, which are signaling pathways for additional functions, were found to have increased expression in platelet-derived RNA from cancer patients.

[0060] Example 2: Example 1 shows that there are differences in gene expression in platelet-derived RNA between healthy individuals (non-cancer individuals) and cancer patients (cancer individuals), and it is predicted that the pathological condition of a patient can be estimated from the gene expression profile of platelet-derived RNA. Thus, Example 2 shows an example of determining the presence or absence of postoperative recurrence in a cancer patient using the gene set that shows differences in expression levels between a group of healthy individuals and a group of cancer patients extracted in Example 1.

[0061] In Example 2, blood was collected from non-small cell lung cancer patients who had had blood collected before surgery, and blood was collected after surgery. RNA was extracted from the platelets obtained using the same procedure as in Example 1, and the expression profile was analyzed.

[0062] 9 shows the results of clustering to determine whether the postoperative expression profiles of non-small cell lung cancer patients resembled the expression profiles of healthy subjects or preoperative cancer patients, using the top 100 genes of the gene set extracted in Example 1. As shown in Figure 9, all three postoperative recurrence cases (type: T2, recurrence: Yes) were classified into cluster α, which was mainly classified with the expression profiles of the preoperative cancer patient group (type: T), i.e., the cancer-bearing group, and two of the four postoperative non-recurrence cases (type: T2, recurrence: No) were classified into cluster β, which was mainly classified with the expression profiles of the healthy subject group (type: N), i.e., the non-cancer-bearing group.

[0063] In Example 2, it was shown that by comparing the expression profiles of RNA contained in the platelets of cancer-bearing and non-cancer-bearing individuals to extract gene sets with differences in gene expression, and then using these gene sets for clustering, it is possible to distinguish between patients with postoperative recurrence and those without recurrence.

[0064] Example 3 In Example 3, a method for assessing the risk of postoperative recurrence in cancer patients, which is different from that in Example 2, will be described.

[0065] 10A shows a graph in which the expression level ratios between cluster α (which was frequently classified into the expression profiles of the preoperative cancer patient group) and cluster β (which was frequently classified into the expression profiles of the healthy subject group) for the gene set characteristic of cancer patients used in Example 2 are arranged on the horizontal axis in descending order, and the median expression levels of specimens classified into cluster α and cluster β are plotted on the vertical axis. Graph 1001 for cluster α has a negative slope, and graph 1002 for cluster β has a positive slope.

[0066] Figure 10B shows the results of plotting the expression levels of platelet-derived RNA from postoperative non-small cell lung cancer patients, with gene sets characteristic of cancer patients arranged on the horizontal axis in descending order of the expression level ratio between cluster α and cluster β, following the example of Figure 10A, for the expression profiles of platelet-derived RNA from postoperative non-small cell lung cancer patients collected in Example 2. Two of the three postoperative recurrence patients showed the same negative slope as cluster α (where many expression profiles from the preoperative cancer patient group were classified), and one showed no slope, while all four postoperative non-recurrence patients showed the same positive slope as cluster β (where many expression profiles from the healthy subjects were classified).

[0067] In Example 3, the expression profiles of RNA contained in platelets from cancer-bearing and non-cancer-bearing individuals were compared to extract gene sets with differences in gene expression, and expression level plots of these gene sets were created. It was shown that patients with postoperative recurrence and those without recurrence could be distinguished based on the similarity of the plot shapes.

[0068] 201...Expression profile of a cancer patient group, 202...Expression profile of a healthy subject group, 203...Preoperative expression profile of a subject, 204...Postoperative expression profile of a subject, 501...Blood, 502...RNA extraction unit, 503...Centrifuge, 504...Fraction collection device, 505...RNA extraction device, 506...Gene expression analysis unit, 507...Automatic reagent mixing device, 508...Gene expression analysis device, 509...Testing device, 510...Information processing device (computer), 511...Processor (CPU), 512...Memory, 513...Storage device, 514...Input interface, 515... Output interface, 516...communication interface, 517...bus, 521...input section, 522...comparison section, 523...ctDNA test result input section, 524...proposal section, 525...memory section, 526...expression profile of cancer-bearing group, 527...expression profile of cancer-free group, 601...platelet fraction, 602...buffy coat, 603...red blood cell fraction, 611...piezo transducer, 612...side inlet, 613...center inlet, 614...side outlet, 615...center outlet, 616...microfluidic channel, 1001, 1002...graph.

Claims

1. A genetic testing method for testing the risk of postoperative recurrence in a subject who is a cancer patient, comprising: a first expression analysis step of measuring a first expression profile, which is an expression profile of RNA extracted from platelets collected from the subject's blood after surgery; and a testing step of comparing the first expression profile with a cancer-bearer group expression profile, which is a collection of expression profiles of RNA extracted from platelets collected from the blood of cancer-bearers belonging to a cancer-bearer group, and a non-cancer-bearer group expression profile, which is a collection of expression profiles of RNA extracted from platelets collected from the blood of non-cancer-bearers belonging to a non-cancer-bearer group, to determine which expression profile the first expression profile resembles; wherein in the testing step, if the first expression profile resembles the cancer-bearer group expression profile, the risk of recurrence is assessed as high, and if the first expression profile resembles the non-cancer-bearer group expression profile, the risk of recurrence is assessed as low.

2. A genetic testing method as claimed in claim 1, wherein in the testing step, the first expression profile, the expression profile of the cancer-bearing group, and the expression profile of the non-cancer-bearing group are clustered together, and if the first expression profile belongs to a cluster mainly composed of the expression profile of the cancer-bearing group, the first expression profile is determined to be similar to the expression profile of the cancer-bearing group, and if the first expression profile belongs to a cluster mainly composed of the expression profile of the non-cancer-bearing group, the first expression profile is determined to be similar to the expression profile of the non-cancer-bearing group.

3. A genetic testing method as claimed in claim 1, wherein in the testing step, the cancer-bearing expression group profile and the cancer-free expression profile are clustered to obtain a first cluster mainly composed of the cancer-bearing expression profile and a second cluster mainly composed of the non-cancer-bearing expression profile, and the genetic testing method determines whether the first expression profile is similar to the cancer-bearing expression profile or the non-cancer-bearing expression profile based on the shape of an expression level plot obtained by plotting the expression levels of the genes in the first expression profile against the horizontal axis of genes arranged in descending order of expression level ratio between the first cluster and the second cluster.

4. A genetic testing method as claimed in claim 3, which determines similar expression profiles based on whether the slope of the expression level plot of the first expression profile matches the slope of the expression level plot of the expression profile of the cancer-bearing group or the slope of the expression level plot of the expression profile of the non-cancer-bearing group.

5. A genetic testing method as claimed in claim 1, wherein the blood of the cancer-bearing individual is postoperative blood from a cancer patient who has experienced a postoperative recurrence of cancer, and the blood of the non-cancer-bearing individual is postoperative blood from a cancer patient who has not experienced a postoperative recurrence of cancer.

6. A genetic testing method according to claim 1, further comprising a second expression analysis step of measuring a second expression profile, which is an expression profile of RNA extracted from platelets collected from the subject's preoperative blood, wherein in the testing step, the second expression profile is compared with the expression profile of the cancer-bearing group and the expression profile of the cancer-free group to determine which expression profile the second expression profile resembles, and wherein in the testing step, if the first expression profile resembles the expression profile of the cancer-bearing group or if the second expression profile resembles the expression profile of the non-cancer-bearing group, the risk of recurrence is assessed as high, and if the first expression profile resembles the expression profile of the non-cancer-bearing group and the second expression profile resembles the expression profile of the cancer-bearing group, the risk of recurrence is assessed as low.

7. A genetic testing method according to claim 6, wherein the blood of the cancer-bearing individual is the blood of a cancer patient, and the blood of the cancer-free individual is the blood of a healthy individual.

8. A genetic testing method as claimed in claim 1, wherein, in the testing step, even if a test for postoperative recurrence risk based on ctDNA in plasma collected from the subject's postoperative blood is negative, the risk of recurrence is highly evaluated if the first expression profile is similar to the expression profile of the cancer-bearing group.

9. A genetic testing method as claimed in claim 1, wherein in the testing step, the expression profiles of the cancer-bearing group and the non-cancer-bearing group are obtained by clustering the expression profiles of the cancer-bearing group and the non-cancer-bearing group together, and similarity to the first expression profile is determined for expression profiles of a predetermined number of genes having a large expression ratio between a first cluster mainly composed of the expression profiles of the cancer-bearing group and a second cluster mainly composed of the expression profiles of the non-cancer-bearing group.

10. The genetic testing method according to claim 9, wherein the predetermined number of genes includes genes related to any one of the pathways of platelet response to wounds, the pathway of platelet activation, the VEGFA-VEGFR2 pathway, the RHO GTPase pathway, and the cytokine production pathway.

11. A genetic testing method according to claim 1, wherein the platelets of the subject are collected as a platelet fraction separated from the subject's blood by centrifugation or ultrasonic separation.

12. A genetic testing method according to claim 1, further comprising a proposing step of proposing to the subject, as a treatment plan, omitting postoperative adjuvant therapy, if the risk of recurrence is assessed as low in the testing step.

13. A genetic testing system for testing the risk of postoperative recurrence in a cancer patient subject, comprising: an RNA extraction device that extracts RNA from platelets collected from the subject's blood; a gene expression analysis device that measures the expression profile of the extracted RNA; and a testing device that compares a first expression profile, which is an expression profile of RNA extracted from platelets collected from the subject's postoperative blood, with a cancer-bearer group expression profile, which is a collection of expression profiles of RNA extracted from platelets collected from the blood of cancer-bearing individuals belonging to a cancer-bearing group, and a non-cancer-bearer group expression profile, which is a collection of expression profiles of RNA extracted from platelets collected from the blood of non-cancer-bearing individuals belonging to a non-cancer-bearing group, and determines which expression profile the first expression profile resembles; wherein the testing device evaluates the risk of recurrence as high if the first expression profile is similar to the cancer-bearer group expression profile, and evaluates the risk of recurrence as low if the first expression profile is similar to the non-cancer-bearing group expression profile.

14. A genetic testing system as claimed in claim 13, wherein the testing device compares a second expression profile, which is an expression profile of RNA extracted from platelets collected from the subject's preoperative blood, with the expression profile of the cancer-bearing group and the expression profile of the non-cancer-bearing group, and determines which expression profile the second expression profile resembles; and the testing device evaluates the risk of recurrence as high if the first expression profile is similar to the expression profile of the cancer-bearing group or if the second expression profile is similar to the expression profile of the non-cancer-bearing group, and evaluates the risk of recurrence as low if the first expression profile is similar to the expression profile of the non-cancer-bearing group and the second expression profile is similar to the expression profile of the cancer-bearing group.

15. A genetic testing system as claimed in claim 13, wherein the testing device stores expression profiles of a predetermined number of genes that have a large expression ratio between a first cluster mainly composed of the expression profiles of the cancer-bearing group and a second cluster mainly composed of the expression profiles of the non-cancer-bearing group, the expression profiles being obtained when the expression profiles of the cancer-bearing group and the expression profiles of the non-cancer-bearing group are clustered together.

16. A genetic testing system according to claim 15, wherein the predetermined number of genes includes genes related to any one of the pathways of platelet response to wounds, the pathway of platelet activation, the VEGFA-VEGFR2 pathway, the RHO GTPase pathway, and the cytokine production pathway.

17. A genetic testing system as claimed in claim 13, wherein the testing device proposes omission of postoperative adjuvant therapy to the subject as a treatment plan when the risk of recurrence is assessed as low.

Images