Oral cancer detection kit, reagent use, reagents, oral cancer detection device, information acquisition method and program

Abstract

An oral cancer detection kit includes a reagent for detecting the methylation of a promoter region in each of a plurality of genes in DNA contained in a sample collected from the oral cavity of a subject. The plurality of genes includes KLLN, CASP8, CHFR and GSTP1, and whether or not the subject has oral cancer can be determined on the basis of information on methylation occurring in a promoter region which is acquired using the reagent.

Description

[Technical Field]

[0001] This invention relates to oral cancer detection. for This relates to the use of kits and reagents, reagents, oral cancer detection devices, information acquisition methods, and programs. [Background technology]

[0002] Among oral precancerous lesions, oral leukoplakia, oral erythroplakia, oral lichen planus, and oral candidiasis have the potential to develop into oral cancer. It is said that most oral precancerous lesions follow a chronic course, with only a portion progressing to cancer. Precancerous lesions exhibit a variety of pathological conditions at the cellular level, ranging from mild abnormalities to early-stage cancer. A definitive diagnosis of whether an oral precancerous lesion is oral cancer requires an invasive tissue biopsy. Because tissue biopsies are invasive and require pathological diagnosis, they are limited to being performed at specialized facilities.

[0003] In addition to tissue biopsy, oral precancerous lesions are evaluated using the relatively less invasive method of cytological scraping. Cytological scraping has been reported to have relatively good sensitivity and specificity in diagnosing oral cancer. However, cytological scraping also requires specialized facilities, and its sensitivity for borderline lesions remains at around 60%. Furthermore, it is difficult to make a diagnosis using cytological scraping when it is difficult to distinguish between inflammatory cells and oral cancer cells.

[0004] Patent Document 1 discloses a method for determining oral cancer based on the expression level of microRNAs in serum obtained by blood sampling, which is less invasive than tissue biopsy. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2020-68673 [Overview of the project] [Problems that the invention aims to solve]

[0006] According to the method disclosed in Patent Document 1, although a sample is obtained by blood sampling, the burden on the subject is reduced compared to collecting tissue, but it is not non-invasive. Further, the method disclosed in Patent Document 1 is not optimized for discriminating between subjects with oral pre-cancerous lesions and subjects with oral cancer.

[0007] An inspection method for non-invasively detecting oral cancer has not yet been put into practical use. In the case of patients with multiple cancers, lesions may appear in multiple sites, so oral cancer may be examined multiple times. In order to reduce the burden on the subject due to repeated examinations, there is a need to detect oral cancer non-invasively and simply.

[0008] The present invention has been made in view of the above circumstances, and an object thereof is to provide an oral cancer detection kit, use of a reagent, a reagent, an oral cancer detection device, an information acquisition method, and a program that can detect oral cancer non-invasively and simply. for

Means for Solving the Problems

Means for Solving the Problems

[0009] The oral cancer detection kit according to the first aspect of the present invention for comprises a reagent for detecting methylation of each promoter region of a plurality of genes in DNA contained in a sample collected from the oral cavity of a subject, wherein the plurality of genes are KLLN, CASP8, CHFR, and GSTP1 be .

[0010] The plurality of genes may further include TP73, as well.

[0011] The plurality of genes may further include RARB, as well.

[0013] The sample may be the gargle of the subject, as well.

[0014] The use according to the second aspect of the present invention is Oral cancer detection for Use of a reagent for detecting methylation of each promoter region of a plurality of genes in DNA contained in a sample collected from the oral cavity of a subject in the manufacture of a kit, 、 before wherein the plurality of genes are KLLN, CASP8, CHFR and GSTP1.

[0015] The reagent according to the third aspect of the present invention is A reagent for detecting methylation of each promoter region of a plurality of genes in DNA contained in a sample collected from the oral cavity of a subject for detecting oral cancer, 、 before wherein the plurality of genes are KLLN, CASP8, CHFR and GSTP1.

[0016] The oral cancer detection device according to the fourth aspect of the present invention is Comprising a determination unit that determines whether or not the subject has oral cancer based on information regarding methylation of each promoter region of a plurality of genes in DNA contained in a sample collected from the oral cavity of the subject, The information regarding the methylation is, A value indicating the degree of methylation of the methylation site in the promoter region, wherein the plurality of genes are KLLN, CASP8, CHFR and GSTP1.

[0017] The information acquisition method according to the fifth aspect of the present invention is Including an acquisition step of acquiring information for determining whether or not the subject has oral cancer based on information regarding methylation of each promoter region of a plurality of genes in DNA contained in a sample collected from the oral cavity of the subject, The information regarding the methylation is, A value indicating the degree of methylation of the methylation site in the promoter region, The aforementioned multiple types of genes are, These are KLLN, CASP8, CHFR, and GSTP1.

[0018] A program according to the sixth aspect of the present invention is: Computers, Based on information regarding the methylation of each promoter region of multiple genes in the DNA contained in a sample taken from the subject's oral cavity, this unit functions as a determination unit to determine whether or not the subject has oral cancer. It is a program for that purpose. , The information regarding the methylation is, A value indicating the degree of methylation of the methylation site in the promoter region, The aforementioned multiple types of genes are, These are KLLN, CASP8, CHFR, and GSTP1. [Effects of the Invention]

[0019] According to the present invention, oral cancer can be detected non-invasively and easily. [Brief explanation of the drawing]

[0020] [Figure 1] (A) is a block diagram showing the hardware configuration of an oral cancer detection device according to an embodiment of the present invention. (B) is a block diagram showing the functions of an oral cancer detection device. [Figure 2] Figure 1 shows a flowchart of the determination process by the oral cancer detection device. [Figure 3] This figure shows the patient operating characteristic (ROC) curve for oral cancer diagnosis in the example. [Figure 4] (A) is a figure showing the ROC curve based on the abnormal methylation score for the training set according to Example 2. (B) is a figure showing the distribution of the abnormal methylation score for the training set according to Example 2. [Figure 5] (A) is a figure showing the ROC curve based on the abnormal methylation score for the test set according to Example 2. (B) is a figure showing the distribution of abnormal methylation scores for the test set according to Example 2. [Figure 6] (A) is a figure showing the ROC curve based on abnormal methylation scores for patients who underwent cytology according to Example 2. (B) is a figure showing the distribution of abnormal methylation scores for patients who underwent cytology according to Example 2. [Modes for carrying out the invention]

[0021] Embodiments of the present invention will be described with reference to the drawings. However, the present invention is not limited to the embodiments and drawings described below. In the embodiments described below, expressions such as “having,” “including,” or “containing” also include the meaning of “consisting of” or “composed of.”

[0022] The oral cancer detection device 100 according to this embodiment will be described with reference to Figure 1. The oral cancer detection device 100 is a device for determining whether or not a subject has oral cancer (whether or not they are suffering from oral cancer) by analyzing data obtained from a subject's biological sample. As shown in Figure 1(A), the oral cancer detection device 100 has a configuration in which a storage unit 10, a RAM (Random Access Memory) 20, an input device 30, a display device 40, and a CPU (Central Processing Unit) 50 are communicated via a bus 60.

[0023] The storage unit 10 includes non-volatile storage media such as ROM (Read Only Memory), HDD (Hard Disk Drive), and flash memory. The storage unit 10 stores various data and software programs, as well as the oral cancer diagnosis program 11.

[0024] RAM20 functions as the main memory of CPU50, and when CPU50 executes the oral cancer diagnosis program 11, the oral cancer diagnosis program 11 is loaded into RAM20. Data input from input device 30 is temporarily stored in RAM20.

[0025] The input device 30 is hardware for the user to input data into the oral cancer detection device 100. The input device 30 inputs data (information) regarding the methylation of gene promoter regions in DNA contained in a sample taken from the subject's oral cavity to the CPU 50. The CPU 50 stores the input methylation data in the storage unit 10.

[0026] While the subjects are not particularly limited, the oral cancer detection device 100 according to this embodiment can accurately distinguish between subjects with oral cancer and subjects with oral precancerous lesions, even if the subjects to be analyzed include subjects with oral precancerous lesions rather than oral cancer, as shown in the example below. Therefore, the subjects preferably have oral precancerous lesions.

[0027] The sample is not particularly limited as long as it contains cells from the oral cavity, and may include, for example, the subject's saliva, oral swab, sputum, and mouthwash. Mouthwash, also called gargle solution, is the liquid that is expelled from the mouth after the subject rinses their mouth with a certain amount of liquid. Gargling may be gargling to wash the throat, but preferably, gargling is gargling to wash the mouth. Mouthwash contains saliva secreted by the subject and shed oral cells. Mouthwash can be collected non-invasively. Furthermore, using mouthwash as a sample has the advantage of making it easier to collect samples from patients with oral precancerous lesions who also have xerostomia (dry mouth). The following explanation will assume that mouthwash is used as the sample.

[0028] Methylation data are values ​​that indicate the degree of methylation at methylation sites in the promoter region of a gene in DNA (genomic DNA). DNA methylation mainly refers to modification by methyl groups occurring on cytosine. Methylation of specific bases in the promoter region suppresses the expression of genes whose expression is regulated by that promoter region. Methylation of tumor suppressor genes may be involved in carcinogenesis. Here, "methylation site" refers to a site (base) in the gene sequence that can be methylated. In the following, data on methylation in the promoter region of genes contained in the mouthwash of subjects will also be referred to as "methylation data".

[0029] Methylation data can be obtained by analyzing the above-mentioned tumor suppressor genes using a known method that can distinguish between methylated and unmethylated base sequences at methylation sites. For example, by treating a sample containing the gene with bisulfite, unmethylated cytosine is converted to uracil, resulting in a base sequence different from that containing methylated cytosine. Using this, the presence or absence of amplification and the base sequence of the amplified DNA can be determined by performing PCR (Polymerase Chain Reaction) using two sets of primers capable of amplifying both the methylated and unmethylated cytosine base sequences. The length of the primers is not particularly limited, but for example, a length of 20 to 45 bases is sufficient to obtain an amplification product of 80 to 500 bases in length that includes the methylation site.

[0030] Alternatively, the MS-MLPA (Methylation Specific-Multiplex Ligation-dependent Probe Amplification) method can be used, in which a probe capable of hybridizing to a target region containing a methylation site is set, and a PCR amplification product is obtained only when the probe hybridizes. Each probe used in the MS-MLPA method has a cleavage site for methylation-sensitive restriction enzymes. If the methylation site in the target region hybridized by the probe is methylated, the restriction enzyme does not act, and amplification occurs by PCR. On the other hand, if the methylation site in the target region hybridized by the probe is not methylated, the restriction enzyme acts, and amplification does not occur by PCR. As a result, an amplification product is obtained and detected as a peak value only when methylation occurs. According to the MS-MLPA method, the methylation rate can be determined by calculating the relative peak value of restriction enzyme-treated samples compared to the peak value of untreated samples in the same sample.

[0031] Various enzymes are known as methylation-sensitive restriction enzymes, such as AccII, HhaI, HapII, and HaeIII. Methylation data can be obtained using commercially available kits that utilize the MS-MLPA method. Probes are designed according to the target nucleotide sequence. The length of the probe is not particularly limited; for example, a probe with a length of 80 to 500 nucleotides that can hybridize to the region containing the methylation site can be used.

[0032] The display device 40 is a display for outputting the results of the oral cancer diagnosis made by the CPU 50. The CPU 50 reads the oral cancer diagnosis program 11 stored in the memory unit 10 into the RAM 20 and executes the oral cancer diagnosis program 11, thereby realizing the functions described below.

[0033] Figure 1(B) is a block diagram showing the functions implemented by the CPU 50. The oral cancer diagnosis program 11 causes the CPU 50 to function as a diagnosis unit 1 and an output unit 2.

[0034] The determination unit 1 determines whether or not a subject has oral cancer based on methylation data of the promoter region of genes in DNA contained in the mouthwash collected from the subject's oral cavity. In this embodiment, the genes to be analyzed are at least one of 25 types of tumor suppressor genes, which are shown in Table 1: TP73, CASP8, VHL, RARB, MLH1, RASSF1, FHIT, APC, ESR1, CDKN2A, CDKN2B, DAPK1, KLLN, PTEN, CD44, GSTP1, ATM, CADM1, CDKN1B, CHFR, BRCA2, CDH13, HIC1, BRCA1, and TIMP3.

[0035] [Table 1]

[0036] The methylation data, for example, for a single gene KLLN, is the percentage of KLLN in the mouthwash where a specific methylation site is methylated (methylation rate). The determination unit 1 determines whether or not the subject has oral cancer based on the methylation rate calculated for each of the 25 types of genes. Preferably, the determination unit 1 determines whether or not the subject has oral cancer based on the methylation rate calculated for at least one selected from the group consisting of KLLN, CASP8, CHFR, GSTP1, and CDKN1B. The determination unit 1 compares the methylation rate of the gene with a cutoff value C1 set in advance for each gene to determine whether or not the subject has oral cancer. For example, if the methylation rate of a predetermined methylation site in KLLN is r KLLN Therefore, if the cutoff value C1 set for KLLN is 3%, the determination unit 1 determines r KLLN A subject is determined to have oral cancer if the percentage is 3% or higher, and not to have oral cancer if the percentage is less than 3%.

[0037] The determination unit 1 may determine whether the subject has oral cancer by combining methylation data on a plurality of types of genes. That is, the determination unit 1 determines whether the subject has oral cancer based on the methylation data of each promoter region of a plurality of types of genes in the DNA contained in the gargle collected from the oral cavity of the subject. The methylation data on a plurality of types of genes is, for example, the total S of the methylation ratios of the methylation sites calculated for each of the above 25 types of genes r is. The determination unit 1 determines whether the subject has oral cancer based on S r . For example, the determination unit 1 compares the cutoff value C2 regarding the total methylation ratio set in advance with S r to determine whether the subject has oral cancer. If C2 is 20%, the determination unit 1 determines that the subject has oral cancer when S r is 20% or more, and determines that the subject does not have oral cancer when S r is less than 20%.

[0038] Also, the methylation data may be the number N of types of genes in which the methylation sites are methylated among 25 types of genes. For example, r KLLN and the methylation ratio r CASP8 of a predetermined methylation site in CASP8 exceed 0, and the methylation ratios of the other genes are all 0, then the number N is 2. Note that as a criterion for determining that a methylation site is methylated, the cutoff value C1 set for each of the above genes may be used. For example, the cutoff value C1 set for KLLN is 3%, and when r KLLN is 3% or more, KLLN is counted as a gene in which the methylation site is methylated.

[0039] The determination unit 1 determines whether or not a subject has oral cancer based on the number N. For example, the determination unit 1 compares a preset cutoff value C3 with N to determine whether or not a subject has oral cancer. If the cutoff value C3 is 9, the determination unit 1 determines that the subject has oral cancer if N is 9 or greater, and determines that the subject does not have oral cancer if N is 8 or less.

[0040] Furthermore, the determination unit 1 may determine that the subject has oral cancer if the number of genes whose methylation rate exceeds the cutoff value C1 (first reference value), calculated for each of the multiple genes selected from the genes to be analyzed above, exceeds the number-based cutoff value C4 (second reference value). For example, the multiple genes are KLLN, CASP8, CHFR, and GSTP1. Specifically, if C1 is set for each of KLLN, CASP8, CHFR, and GSTP1, and C4 is 2, the determination unit 1 determines that the subject has oral cancer if there are two or more genes among KLLN, CASP8, CHFR, and GSTP1 whose methylation rate exceeds their respective C1 values, and determines that the subject does not have oral cancer if there is one or fewer such genes.

[0041] Furthermore, in addition to KLLN, CASP8, CHFR, and GSTP1, TP73 or RARB may be further selected as genes to be analyzed as multiple genes. Preferably, the multiple genes are KLLN, CASP8, CHFR, GSTP1, TP73, and RARB. The determination unit 1 may determine that the subject has oral cancer if the methylation rate of at least one of these multiple genes is equal to or greater than the cutoff value set for the methylation rate of each gene, and determine that the subject does not have oral cancer if it is less than C1.

[0042] The cutoff values ​​C1 to C4 used by the determination unit 1 for determination can be set by, for example, comparing the methylation data of subjects with oral cancer with the methylation data of subjects without oral cancer, using a known method. Alternatively, the cutoff values ​​C1 to C4 can be obtained from a model constructed using a known data mining technique. This model is constructed using supervised learning.

[0043] Supervised learning is a machine learning technique that uses a set of combinations of explanatory variables and their associated target variables as training data, and learns by fitting the training data to the model. Fitting is performed by extracting features of the explanatory variables included in the training data and selecting features for each target variable, extracting features of data belonging to that target variable, or generating criteria for identifying the target variable. Through fitting, a model is constructed that outputs the target variable that should correspond to the input explanatory variables. Depending on the model, it is also possible to output a target variable that corresponds to explanatory variables not included in the training data.

[0044] When determining whether or not a person has oral cancer, the explanatory variables in the training data are one or more methylation data points, and the dependent variable is information indicating whether or not the subject corresponding to that methylation data has oral cancer. For example, the sum of the methylation percentages S could be used as the methylation data. r When using the number of types of genes in which the methylation site is methylated (N), the methylation data of subjects who do not have oral cancer is S r1 And if N1 is the explanatory variable, then S r1 The dependent variable "0", which indicates that the patient does not have oral cancer, is associated with "N1". On the other hand, the methylation data of oral cancer patients is S r2 And if N2, then the explanatory variable "S r2 The target variable "1", which is information indicating the presence of oral cancer, is associated with "N2". Preferably, the training data is a set of combinations of methylation data for multiple subjects and information indicating whether or not the subject corresponding to the methylation data has oral cancer.

[0045] Any known method can be used for supervised learning. Examples of supervised learning methods include discriminant analysis, canonical discriminant analysis, linear classification, multiple regression analysis, logistic regression analysis, support vector machines, decision trees, neural networks, convolutional neural networks, perceptrons, and k-nearest neighbors.

[0046] In logistic regression analysis, the following equation is used as the model, where p is the dependent variable and x is the independent variable. Note that in equation 1, this a i is x i These are the partial regression coefficients for . The partial regression coefficients can be obtained using training data by known methods, such as the least squares method or the maximum likelihood method. p=1 / {1+exp(-(a1x1+a2x2+···+a n x n +b))} (Formula 1)

[0047] The constructed model and each cutoff value C1 to C4 are stored in the memory unit 10. The determination unit 1 inputs the subject's methylation data into the model stored in the memory unit 10 to obtain information indicating whether or not the subject has oral cancer as output. For example, the information indicating whether or not the subject has oral cancer is either information indicating that the subject has oral cancer or information indicating that the subject does not have oral cancer.

[0048] The determination unit 1 inputs information indicating whether or not the subject has oral cancer to the output unit 2. The output unit 2 displays the information indicating whether or not the subject has oral cancer, input by the determination unit 1, on the display device 40.

[0049] Next, the determination process by the oral cancer detection device 100 will be explained with reference to the flowchart shown in Figure 2. In this flowchart, the determination unit 1 determines that a subject has oral cancer if the number of genes among KLLN, CASP8, CHFR, and GSTP1 with a methylation rate of C1 or higher is C4 or higher, and determines that the subject does not have oral cancer if the number of such genes is less than C4. In addition to C4, C1 for each of KLLN, CASP8, CHFR, and GSTP1 is pre-stored in the storage unit 10. The input methylation data includes the methylation rate r for each of KLLN, CASP8, CHFR, and GSTP1. KLLN , r CASP8 , r CHFR and r GSTP1 This includes the methylation rates for the 25 genes listed above.

[0050] The determination unit 1 waits for the subject's methylation data to be entered by the user via the input device 30 (Step S1; No). Once the subject's methylation data is entered (Step S1; Yes), the determination unit 1 refers to the memory unit 10 and determines the C1 and r for KLLN, CASP8, CHFR, and GSTP1 respectively. KLLN , r CASP8 , r CHFR and r GSTP1 The determination unit 1 compares the number of genes with a methylation rate of C1 or higher and counts the number of genes with a methylation rate of C1 or higher for each (step S2). Next, the determination unit 1 compares the number of genes with a methylation rate of C1 or higher with C4 and determines whether or not the subject has oral cancer (step S3). If the number of genes is C4 or higher (step S3; Yes), the output unit 2 displays information indicating that the subject has oral cancer via the display device 40 (step S4). On the other hand, if the number of genes is less than C4 (step S3; No), the output unit 2 displays information indicating that the subject does not have oral cancer via the display device 40 (step S5). Then, the determination unit 1 ends the determination process.

[0051] As described in detail above, the oral cancer detection device 100 according to this embodiment determines whether or not a subject has oral cancer based on methylation data of a sample such as mouthwash collected from the subject's oral cavity. By using a sample that can be easily collected from the oral cavity, oral cancer can be detected non-invasively and simply. In particular, mouthwash can be collected inexpensively and repeatedly. Furthermore, another advantage is that mouthwash can be easily collected even by non-experts. According to the determination using methylation data according to this embodiment, subjects with oral cancer and subjects with oral precancerous lesions can be distinguished with high accuracy, as shown in the example below.

[0052] Furthermore, since mouthwash can be easily obtained, oral cancer can be detected rapidly. Because mouthwash collection is non-invasive, gene methylation can be evaluated from the subject's mouthwash at any desired time.

[0053] The determination unit 1 may determine the result by adding, subtracting, multiplying, or dividing arbitrary values ​​to the methylation data values, such as the methylation rate, the total methylation rate, and the number of methylated genes. Alternatively, the determination may be made using values ​​obtained by transforming the methylation data using known transformation methods, such as exponential transformation, logarithmic transformation, angular transformation, square root transformation, probit transformation, reciprocal transformation, Box-Cox transformation, or power transformation. The determination unit 1 may also determine the result using values ​​obtained by transforming the methylation data with weights according to the subject's sex or age.

[0054] The CPU 50 may also function as a model building unit that constructs the above model. The model building unit constructs a model by supervised learning using training data stored in the memory unit 10. More specifically, the model building unit performs supervised learning using training data in which one or more methylation data and information indicating whether or not the subject corresponding to the methylation data has oral cancer are used as explanatory variables and dependent variables, respectively. The model building unit stores the constructed model in the memory unit 10. As a result, the determination unit 1 inputs the subject's methylation data into the model stored in the memory unit 10 to determine whether or not the subject has oral cancer.

[0055] The oral cancer detection device 100 may be equipped with a communication interface and connected to a network. The determination unit 1 may receive methylation data transmitted via communication means from an external device connected to the network and determine whether or not the subject has oral cancer. Furthermore, the output unit 2 may transmit information indicating whether or not the subject has oral cancer to an external device via the communication interface.

[0056] The oral cancer detection program 11 and various software programs used in the oral cancer detection device 100 can be distributed by storing them on computer-readable recording media such as CD-ROMs (Compact Disc Read Only Memory), DVDs (Digital Versatile Discs), magneto-optical discs, USB (Universal Serial Bus) memory, memory cards, and HDDs. By installing the oral cancer detection program 11 and various software programs on a specific or general-purpose computer, that computer can function as the oral cancer detection device 100. Alternatively, the oral cancer detection program 11 and various software programs may be stored on a storage device owned by another server on the Internet, and the oral cancer detection program 11 and various software programs can be downloaded from that server.

[0057] In another embodiment, a method is provided that includes an evaluation step of evaluating the methylation of the above gene in the mouthwash of a subject in order to obtain data for determining whether or not the subject has oral cancer.

[0058] Furthermore, other embodiments provide a method for obtaining information useful for detecting oral cancer. The information acquisition method includes an acquisition step of obtaining information for determining whether or not a subject has oral cancer based on information regarding methylation in the promoter region of the above-mentioned gene. Another embodiment provides an oral cancer treatment method comprising the acquisition step and an administration step of administering an oral cancer treatment drug to a subject based on the information obtained in the acquisition step. Yet another embodiment provides an oral cancer diagnostic method that includes a diagnostic step of diagnosing whether or not a subject has oral cancer based on information regarding methylation in the promoter region of the above-mentioned gene.

[0059] Furthermore, other embodiments provide a method for detecting oral cancer. The oral cancer detection method includes an auxiliary step that assists in determining whether or not a subject has oral cancer based on information regarding methylation in the promoter region of the above-mentioned gene.

[0060] In another embodiment, an oral cancer screening kit is provided. The oral cancer screening kit comprises a reagent for detecting methylation in the promoter region of at least one of the above-mentioned genes. Preferably, the reagent is the reagent, primer, probe, and methylation-sensitive restriction enzyme required for the above-mentioned bisulfite treatment. The primer is not particularly limited as long as it is designed to include the methylation site in the base sequence of the PCR product.

[0061] Examples of primer and probe base sequences include the base sequence or partial sequence of the target region containing the methylation site, or a base sequence complementary to the entire base sequence or partial sequence.

[0062] The probe hybridizes with the PCR product. The hybridization conditions are stringent, for example, that the probe hybridizes with nucleic acids with complementary base sequences but not with nucleic acids with non-complementary base sequences. Stringent conditions can be appropriately determined based on, for example, the Molecular Cloning: A Laboratory Manual, 3rd Edition (2001), and include, for example, 0.2×SSC, 0.1% SDS, and incubation at 65°C. Primers and probes can be chemically synthesized, for example, using a commercially available automated nucleic acid synthesizer. When obtaining methylation data using the MS-MLPA method, the base sequences of the probes for each of the above-mentioned genes are shown, for example, in SEQ ID NOs: 1 to 28.

[0063] The oral cancer screening kit may include a reagent that detects methylation of the promoter regions of multiple genes in DNA contained in a sample taken from the subject's oral cavity. In this case, the multiple genes are, for example, KLLN, CASP8, CHFR, and GSTP1. Based on the methylation data of the promoter regions of each gene obtained by the reagent, it is determined whether or not the subject has oral cancer. The multiple genes may be KLLN, CASP8, CHFR, GSTP1, and TP73, or KLLN, CASP8, CHFR, GSTP1, and RARB, or KLLN, CASP8, CHFR, GSTP1, TP73, and RARB.

[0064] In another embodiment, a reagent is provided for detecting the methylation of each promoter region of multiple genes in DNA contained in a sample taken from the oral cavity of a subject in the manufacture of an oral cancer detection kit, or a reagent is provided for detecting oral cancer by detecting the methylation of each promoter region of multiple genes in DNA contained in a sample taken from the oral cavity of a subject. Based on the methylation data obtained by the reagent, it is determined whether or not the subject has oral cancer.

[0065] The present invention will be described in more detail by the following examples, but the present invention is not limited to these examples. [Examples]

[0066] [Example 1] (subject) This study included 22 patients with a clinical diagnosis of oral precancerous lesions who visited the Oral and Maxillofacial Center at Kagoshima University Hospital between 2019 and 2020. Tissue biopsies were performed on all 22 subjects after specimen collection, followed by pathological examination. Pathological examination revealed malignant lesions in 2 of the 22 patients with precancerous lesions (squamous cell cancer (SCC) intraepithelial) and in 3 cases (SCC stage 1).

[0067] (Sample collection) Specimen collection was performed when subjects were in good general health, without fever or cold symptoms. Prior to tissue biopsy, subjects gargled with 20 mL of sterile purified water for 30 seconds, and the resulting gargle solution was collected. The collected gargle solution was immediately placed in a refrigerator and stored at 4°C until DNA extraction.

[0068] (DNA extraction) DNA was extracted from 200 μL of rinse solution using the DNeasy Blood and Tissue Kit (QIAGEN). After extraction, the DNA was quantified using NanoDrop (Thermo-Fisher), and the DNA concentration was adjusted to 10 ng / μL for analysis as follows.

[0069] (Methylation analysis) Methylation analysis was performed using the MS-MLPA method with ME001-D1 Tumour suppressor mix 1 (MRC-Holland), which contains probes for 25 tumor suppressor genes as shown in Table 1. The base sequences of the probes for each gene are shown in SEQ ID NOs. 1 to 28 (see Table 1). In the MS-MLPA method, each probe has a cleavage site by the HhaI restriction enzyme. If the region corresponding to the probe is methylated, the restriction enzyme does not act, and the region is amplified by PCR and detected as a peak value. If the region corresponding to the probe is not methylated, the restriction enzyme acts, and amplification does not occur. Furthermore, the MS-MLPA method compares HhaI-treated and HhaI-untreated samples of the same sample to obtain the peak value of the probe in the methylated region. This peak value reflects the relative amount of methylation (methylation rate, hereinafter also referred to as "methylation %") of each tumor suppressor gene.

[0070] Probes were added to the extracted DNA, and hybridization and ligation were performed to prepare HhaI-untreated and HhaI-treated samples, which were then subjected to PCR. The amplified fragments were analyzed using an ABI PRISM 3130XL Genetic Analyzer (ABI), and peak values ​​were calculated using Gene Mapper (ABI). Furthermore, after normalizing the fragment analysis data using the analysis software Coffalyser (MRC-Holland), the methylation percentage of each tumor suppressor gene was calculated.

[0071] (statistical analysis) Based on pathological diagnoses, 22 patients were divided into a malignant group (carcinoma in situ and early invasive carcinoma) and a non-malignant group. The detection of the malignant group was used as the endpoint. Binary logistic regression analysis was performed based on the methylation percentage of each tumor suppressor gene, the total number of methylated tumor suppressor genes, and the sum of the methylation percentages of each tumor suppressor gene (total methylation percentage), and the AUC (Area Under the Curve) was calculated from the ROC curve. Furthermore, useful cutoff values ​​for determination were set based on the sensitivity and specificity obtained from the ROC curve.

[0072] Using the top four genes with high AUC (KLLN, CASP8, CHFR, and GSTP1), an abnormal methylation score (0 to 4 genes) was created, which represents the number of tumor suppressor genes whose methylation percentage exceeds the respective cutoff value, and the malignancy group was determined. These statistical analyses were performed using SPSS software version 26. The significance level for all tests was set at p=0.05.

[0073] (result) Table 2 shows the characteristics of the subjects, the number of methylated genes, and the methylation percentage. [Table 2]

[0074] Binary logistic regression analysis of methylation percentages for each tumor suppressor gene revealed that KLLN, CASP8, CHFR, GSTP1, and CDKN1B were statistically significant. The number of methylated genes and total methylation percentage were also statistically significant. Table 3 shows the cutoff values ​​obtained from the AUC and ROC curves for the statistically significant variables.

[0075] [Table 3]

[0076] Table 4 shows the accuracy of the determination when using the methylation percentages of KLLN, CASP8, CHFR, GSTP1, and CDKN1B, as well as the cutoff values ​​for the number of methylated genes and total methylation percentage. PPV and NPV represent the positive predictive value and negative predictive value, respectively.

[0077] [Table 4]

[0078] Figure 3 shows the ROC curve based on the abnormal methylation score. The AUC was 0.935, and there were two cutoff values. Table 5 shows the accuracy of the determination when these cutoff values ​​were used.

[0079] [Table 5]

[0080] The abnormal methylation score analysis demonstrated high accuracy, with both sensitivity and specificity exceeding 80% and an AUC of 0.9 or higher.

[0081] [Example 2] (subject) The 22 subjects from Example 1, plus 34 patients with a clinical diagnosis of oral precancerous lesions, comprised 56 subjects. These subjects were divided into a training set of 40 (9 malignant, 31 non-malignant) for learning data and a test set of 16 (6 malignant, 10 non-malignant) for test data. Methylation analysis and statistical analysis were performed on both sets of subjects in the same manner as in Example 1. Additionally, tissue biopsies were performed on the additional subjects after sample collection, and pathological examinations were conducted.

[0082] Tables 6 and 7 show the clinicopathological characteristics of subjects in the training set and the test set, respectively. In both the training set and the test set, no statistically significant differences were observed between the malignant and non-malignant groups in terms of smoking and alcohol consumption.

[0083] [Table 6]

[0084] [Table 7]

[0085] (result) Table 8 shows the diagnostic performance of 14 genes that showed an AUC of 0.6 or higher in the training set. These 14 genes are considered candidate genes for the detection of early-stage cancer.

[0086] [Table 8]

[0087] Of the 14 genes shown in Table 8, six genes (TP73, CASP8, RARB, KLLN, GSTP1, and CHFR) were used to create abnormal methylation scores (0 to 6), and the training set was used to determine whether the cells were malignant or non-malignant. As shown in Table 9, statistically significant differences were observed between the malignant and non-malignant groups using non-parametric and exact tests.

[0088] [Table 9]

[0089] For the training set, the ROC curve based on the abnormal methylation score and the distribution of the abnormal methylation score are shown in Figure 4(A) and Figure 4(B), respectively.

[0090] Using the abnormal methylation scores based on the six genes mentioned above, malignant and non-malignant groups were determined for each test set. As shown in Table 10, the AUC was 0.833, indicating high accuracy, and good results were obtained in terms of specificity and negative predictive value.

[0091] [Table 10]

[0092] For the test set, the ROC curve based on the abnormal methylation score and the distribution of the abnormal methylation score are shown in Figure 5(A) and Figure 5(B), respectively.

[0093] Of the 44 patients who underwent commonly used cytological examinations, only one case of SCC could be correctly identified by cytological examination. The results of the cytological examination and tissue biopsy are shown in Table 11.

[0094] [Table 11]

[0095] Of the 44 patients, excluding 10 patients who were difficult to differentiate (IFN), the ROC curves and distribution of abnormal methylation scores based on the six genes mentioned above are shown in Figures 6(A) and 6(B), respectively. The abnormal methylation score still showed a high AUC even in the group of patients who underwent cytology, demonstrating good diagnostic capability. As shown in Table 12, the abnormal methylation score showed useful diagnostic accuracy in terms of sensitivity, positive predictive value, and negative predictive value when determining malignant and non-malignant groups. Therefore, the abnormal methylation score is useful for detecting malignant lesions in patients with oral potential malignancies, even in situations where cytology is difficult.

[0096] [Table 12]

[0097] This invention allows for various embodiments and modifications without departing from the broad spirit and scope of the invention. Furthermore, the embodiments described above are for illustrative purposes only and do not limit the scope of the invention. In other words, the scope of the invention is indicated not by the embodiments, but by the claims. Various modifications made within the scope of the claims and the equivalent scope of the meaning of the invention are considered to be within the scope of the invention.

[0098] This application is based on Japanese Patent Application No. 2021-29431, filed on 26 February 2021. The entire specification, claims, and drawings of Japanese Patent Application No. 2021-29431 are incorporated herein by reference. [Industrial applicability]

[0099] This invention is suitable for the examination, diagnosis, and treatment of oral cancer. [Explanation of Symbols]

[0100] 1 Judgment unit, 2 Output unit, 10 Memory unit, 11 Oral cancer diagnosis program, 20 RAM, 30 Input device, 40 Display device, 50 CPU, 60 Bus, 100 Oral cancer detection device

Claims

1. The system includes reagents for detecting methylation of the promoter regions of multiple genes in DNA contained in a sample taken from the oral cavity of a subject. The aforementioned multiple types of genes are, These are KLLN, CASP8, CHFR, and GSTP1. Oral cancer detection kit.

2. The aforementioned multiple types of genes are, Further including TP73, The oral cancer detection kit according to claim 1.

3. The aforementioned multiple types of genes are, Further including RARB, The oral cancer detection kit according to claim 1 or 2.

4. The aforementioned sample is The mouthwash used by the aforementioned subject, A kit for detecting oral cancer according to any one of claims 1 to 3.

5. In the manufacture of an oral cancer detection kit, the use of a reagent to detect methylation of promoter regions of multiple genes in DNA contained in a sample taken from the oral cavity of a subject, The aforementioned multiple types of genes are, These are KLLN, CASP8, CHFR, and GSTP1. use.

6. A reagent for detecting oral cancer, which detects methylation of the promoter regions of multiple genes in DNA contained in a sample taken from the oral cavity of a subject, The aforementioned multiple types of genes are, These are KLLN, CASP8, CHFR, and GSTP1. reagent.

7. The system includes a determination unit that determines whether or not a subject has oral cancer based on information regarding the methylation of each promoter region of multiple genes in the DNA contained in a sample taken from the subject's oral cavity. The information regarding the methylation is, A value indicating the degree of methylation of the methylation site in the promoter region, The aforementioned multiple types of genes are, These are KLLN, CASP8, CHFR, and GSTP1. Oral cancer detection device.

8. The process includes an acquisition step of obtaining information to determine whether or not a subject has oral cancer, based on information regarding the methylation of each promoter region of multiple genes in the DNA contained in a sample taken from the subject's oral cavity. The information regarding the methylation is, A value indicating the degree of methylation of the methylation site in the promoter region, The aforementioned multiple types of genes are, These are KLLN, CASP8, CHFR, and GSTP1. How to obtain information.

9. Computers, A program for functioning as a determination unit that determines whether or not a subject has oral cancer, based on information regarding the methylation of each promoter region of multiple genes in DNA contained in a sample taken from the subject's oral cavity, The information regarding the methylation is, A value indicating the degree of methylation of the methylation site in the promoter region, The aforementioned multiple types of genes are, These are KLLN, CASP8, CHFR, and GSTP1. program.

Images