A method, system, device and medium for determining cancer gene markers

Abstract

The present invention discloses a method, system, device and medium for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis evaluation, and belongs to the technical field of cancer gene markers. The method comprises the following steps: obtaining RNA samples of peripheral blood mononuclear cells of two groups of subjects; obtaining the expression level of each gene based on the obtained RNA samples, and obtaining a gene expression profile; obtaining genes with different expression levels in the two groups of subjects as candidate gene markers based on the gene expression profile, wherein the two groups of subjects can be any of cancer and non-cancer, this type of cancer and non-this type of cancer, cancer prognosis recurrence and cancer prognosis non-recurrence, and cancer prognosis good and cancer prognosis poor. Using the method of the present invention, gene markers for cancer diagnosis, recurrence monitoring or prognosis evaluation can be determined, and the corresponding risk assessment model can be constructed using the determined gene markers, with excellent performance.

Description

Technical Field

[0001] The present invention belongs to the technical field of cancer gene markers, and in particular, relates to a method, system, device and medium for determining cancer gene markers. Background Art

[0002] Cancer is a serious disease that poses a huge threat to human health and life. In vitro diagnostic technology plays a vital role in the early detection, diagnosis and recurrence diagnosis of cancer. In vitro diagnostic technology mainly includes methods such as tumor marker detection, histological examination, and genetic testing. These technologies can help doctors determine whether cancer is present and determine the type and stage of cancer by detecting abnormal biomarkers in body fluids, tissues or cells. However, current in vitro diagnostic technology still has certain limitations in sensitivity and specificity, so it needs to be continuously improved and developed.

[0003] Brain cancer is a malignant tumor that occurs in the skull, of which glioma is the most common. Gliomas originate from brain glial cells and are highly malignant and invasive. At present, the diagnosis of brain cancer mainly relies on imaging and histological examinations. Imaging examinations such as magnetic resonance imaging (MRI) can show the location, size, and morphology of brain tumors, but there are still certain limitations in judging whether the tumor is benign or malignant. Histological examination is to obtain tumor tissue through surgery or biopsy and perform pathological examination to determine the type and grade of the tumor. However, histological examination requires surgery or puncture, which causes certain trauma and risks to patients. Therefore, the development of non-invasive, highly sensitive and highly specific in vitro diagnostic methods is of great significance for the early diagnosis and recurrence monitoring of brain cancer.

[0004] In addition to brain cancer, other cancers such as breast cancer, lung cancer, and colorectal cancer also require the support of in vitro diagnostic technology. In vitro diagnostic methods for breast cancer include imaging examinations such as breast ultrasound, mammography, and magnetic resonance imaging, as well as breast puncture biopsy and histological examination. In vitro diagnostic methods for lung cancer include chest X-ray, CT scan, and fiberoptic bronchoscopy. In vitro diagnostic methods for colorectal cancer include colonoscopy, fecal occult blood test, and genetic testing. These methods can help doctors determine the presence, type, and stage of cancer and develop individualized treatment plans for patients. However, these methods still have certain limitations, such as the risk of invasive operations and insufficient sensitivity. Therefore, developing new in vitro diagnostic technologies to improve the early diagnosis rate and treatment effect of cancer is an important direction of current research. Summary of the invention

[0005] In order to solve at least one of the above technical problems, the technical solution adopted by the present invention is as follows:

[0006] The first aspect of the present invention provides a method for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment, comprising the following steps:

[0007] S1, obtaining RNA samples of peripheral blood mononuclear cells from two groups of subjects, wherein the two groups of subjects are selected from one of the following situations:

[0008] (1) One group includes multiple cancer patients and the other group includes multiple non-cancer subjects,

[0009] (2) One group included patients with multiple cancer prognoses of recurrence, and the other group included patients with multiple cancer prognoses of non-recurrence,

[0010] (3) one group includes multiple patients with good cancer prognosis, and the other group includes multiple patients with poor cancer prognosis;

[0011] S2, obtaining the expression level of each gene based on the RNA sample obtained in step S1, and obtaining a gene expression profile, wherein the gene expression profile includes the expression level of each gene in each subject;

[0012] S3, based on the gene expression profile obtained in step S2, obtaining genes with different expression levels between the two groups of subjects as candidate gene markers,

[0013] Wherein, the cancer is selected from one of brain cancer, breast cancer, cervical cancer, colorectal cancer and ovarian cancer.

[0014] In some embodiments of the present invention, in the first situation, the cancer patients refer to patients suffering from brain cancer, breast cancer, cervical cancer, colorectal cancer and / or ovarian cancer, that is, the multiple cancer patients may suffer from different cancers; the non-cancer subjects refer to subjects who do not suffer from brain cancer, breast cancer, cervical cancer, colorectal cancer or ovarian cancer. In this case, the non-cancer subjects are healthy people, but it does not exclude that the non-cancer subjects suffer from cancers other than brain cancer, breast cancer, cervical cancer, colorectal cancer and ovarian cancer, such as gastric cancer.

[0015] In some other embodiments of the present invention, in the first situation, the multiple cancer patients refer to patients suffering from the same type of cancer. In this case, the non-cancer subjects may include other patients with different types of cancer from the cancer patients, and of course, all of them may be healthy people.

[0016] In the present invention, by selecting two groups of subjects in different situations, gene markers for cancer diagnosis, recurrence monitoring and / or prognosis evaluation can be obtained respectively. It should be noted that the technical solution of the present invention is intended to screen and determine gene markers based on peripheral blood mononuclear cells. The inventor unexpectedly found that this solution is not applicable to all cancer types.

[0017] In the present invention, brain cancer refers to a malignant tumor that occurs in the brain or central nervous system of the human body. These tumors can originate from various cell types in the brain tissue, including neurons, glial cells, meningeal cells, etc. Brain cancer is usually called a primary brain tumor, which initially forms in the brain, although sometimes tumors may also be formed by cancer cells from other parts of the body metastasized to the brain. Brain cancer can be divided into many types, the most common of which include:

[0018] Gliomas: including glioblastoma, astrocytoma, oligodendroglioma, etc., which originate from glial cells in the brain.

[0019] Meningiomas: They arise from the meninges and are usually benign.

[0020] Ependymomas arise from the endocardial cells surrounding the brain ventricles and usually occur in children and adolescents.

[0021] Craniopharyngiomas arise from residual tissue of the craniopharyngioma, usually in the upper part of the craniopharyngioma.

[0022] Metastatic Brain Cancer: Cancer that develops from cancer cells elsewhere in the body and spreads to the brain.

[0023] In some embodiments of the present invention, the brain cancer includes glioma and other intracranial tumors other than glioma. Among them, the other intracranial tumors include but are not limited to the following:

[0024] Pituitary Adenomas: These tumors originate from the pituitary gland cells and may cause abnormal secretion of pituitary hormones or compress surrounding structures, causing symptoms.

[0025] Neurofibromatosis: also known as von Recklinghausen's disease, is an autosomal dominant genetic disease caused by disorders in the development of ectoderm and mesoderm tissues. It is characterized by the involvement of multiple systems and organs, with the central nervous system being the most obvious.

[0026] Schwannomas: These tumors arise from the nerve sheath cells and usually develop in the acoustic nerve or other peripheral nerves.

[0027] Ependymomas: These tumors arise from the endocardial cells that surround the brain ventricles and usually occur in children and adolescents.

[0028] Intracranial diffuse large B-cell lymphoma, activated B-cell subtype: Intracranial diffuse large B-cell lymphoma (DLBCL) is an aggressive tumor derived from mature B cells. It is the most common type of non-Hodgkin's lymphoma, accounting for about 25% to 50% of all non-Hodgkin's lymphomas. The activated B-cell subtype is a non-specific type.

[0029] Craniopharyngiomas: These tumors arise from remnant craniopharyngiomas, usually in the upper portion of the craniopharyngioma.

[0030] Meningiomas: These tumors originate in the meninges and are usually benign. They can occur anywhere in the meninges, including inside and outside the skull and in the spine.

[0031] Arachnoid Cyst: A cystic lesion that forms beneath the arachnoid membrane of the brain or spinal cord. The arachnoid membrane is a thin membrane that covers the brain and spinal cord and is adjacent to the pia mater. It helps protect and support the central nervous system.

[0032] Metastatic poorly differentiated neuroendocrine carcinoma: a malignant tumor that originates from neuroendocrine cells and has highly malignant characteristics.

[0033] In the present invention, the recurrence or non-recurrence refers to the patient's recurrence or non-recurrence within 3 months, 6 months, 1 year, 2 years, 3 years, 5 years, 10 years or the entire survival period of the prognosis. Recurrence is usually worse than the primary condition because recurrent cancer is often more difficult to treat and more likely to continue to progress. Cancer recurrence monitoring can help doctors evaluate the patient's disease evolution, understand the growth rate and location of recurrent tumors, and develop individualized treatment plans based on the monitoring results. Different types of recurrence may require different treatment strategies, so timely monitoring of recurrence helps guide treatment decisions.

[0034] In the present invention, the good prognosis refers to good recovery after disease treatment, no recurrence of the disease, and significantly improved life expectancy; the poor prognosis refers to poor recovery after disease treatment, recurrence of the disease, shortened life expectancy, etc.

[0035] In some embodiments of the present invention, in step S2, the expression level of each gene is obtained using at least one method selected from reverse transcription PCR, fluorescent quantitative PCR, in situ hybridization, Northern blotting, RNA sequencing, microarray chip and proteomics, wherein:

[0036] Reverse transcription polymerase chain reaction (RT-PCR): is a commonly used method that transcribes RNA into cDNA through reverse transcription, then amplifies the target gene or genomic fragment by PCR, and finally quantifies the RNA level by gel electrophoresis or quantitative PCR.

[0037] Real-time fluorescence quantitative PCR (qPCR): It is a highly sensitive method for quantitative detection of RNA levels. It can monitor the increase in the amount of DNA during the PCR process in real time, thereby quantifying the RNA content.

[0038] In situ hybridization (ISH): is a method used to detect the location and distribution of specific RNA sequences in tissue sections or cells. It usually uses affinity-labeled probes to bind to RNA sequences and then observes them under a microscope.

[0039] Northern blotting: A method of isolating RNA from a sample and then probing the RNA with a specific probe. Although less used in modern laboratories, it is still an effective method for quantifying RNA levels.

[0040] RNA sequencing (RNA-Seq): It is a high-throughput method that can obtain comprehensive information about RNA, including the expression levels of different genes, splicing variations, etc., by transcribing RNA into cDNA and then sequencing the cDNA.

[0041] Microarray: It is a commonly used high-throughput technology that detects RNA in samples through probes fixed on the surface of the chip to analyze the expression level of RNA.

[0042] Proteomics: It is the detection of protein content at the omics level. Protein spectrum identification, quantitative proteomics and other technologies can realize the detection of protein expression levels. Proteins are effectors of biological functions, and their levels largely depend on the corresponding mRNA levels, so they can characterize gene expression levels.

[0043] In some embodiments of the present invention, in step S3, a statistical method or an artificial intelligence method is used to obtain genes with different expression levels between the two groups of subjects.

[0044] Artificial Intelligence (AI) is an interdisciplinary subject based on computer science, integrating mathematics, informatics, control science and other subjects.

[0045] Furthermore, the artificial intelligence method is implemented by means of machine learning. Machine learning includes many different algorithms, deep learning is one of them, and other methods include logistic regression, decision tree, random forest, support vector machine, naive Bayes, K nearest neighbor, neural network, etc.

[0046] In some specific embodiments of the present invention, RNA sequencing is used to obtain the expression level of each gene to obtain a gene expression profile, and a differential analysis algorithm is used to obtain genes with different expression levels in the two groups of subjects.

[0047] Furthermore, the difference analysis algorithm includes but is not limited to one of limma, edgeR and DESeq2, wherein:

[0048] limma (Linear Models for Microarray Data): is a statistical method for analyzing microarray data that identifies differentially expressed genes based on linear models. Although originally designed for microarray data, it can also be applied to RNA sequencing data analysis.

[0049] edgeR: is a statistical method for RNA sequencing data analysis that uses the negative binomial distribution model to identify differentially expressed genes. edgeR focuses on analyzing small sample RNA sequencing data, especially providing reliable differential expression analysis in small sample conditions.

[0050] DESeq2: It is also a statistical method for RNA sequencing data analysis, which uses negative binomial distribution model and mixed effect model to identify differentially expressed genes. DESeq2 performs well in large sample RNA sequencing data analysis, especially in dealing with technical repetitions and batch effects.

[0051] A second aspect of the present invention provides a system for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment, comprising the following modules:

[0052] An information input module, used to obtain a gene expression profile based on RNA samples of peripheral blood mononuclear cells of two groups of subjects, wherein the gene expression profile includes the expression level of each gene in each subject;

[0053] an identification module, connected to the information input module, for obtaining genes with different expression levels between the two groups of subjects as candidate gene markers based on the gene expression profile,

[0054] Wherein, the two groups of subjects are selected from one of the following situations:

[0055] (1) One group includes multiple cancer patients and the other group includes multiple non-cancer subjects,

[0056] (2) One group included patients with multiple cancer prognoses of recurrence, and the other group included patients with multiple cancer prognoses of non-recurrence,

[0057] (3) One group includes multiple patients with good cancer prognosis, and the other group includes multiple patients with poor cancer prognosis.

[0058] Wherein, the cancer is selected from one of brain cancer, breast cancer, cervical cancer, colorectal cancer and ovarian cancer.

[0059] In some embodiments of the present invention, the expression level of each gene is obtained by at least one method selected from reverse transcription PCR, fluorescent quantitative PCR, in situ hybridization, Northern blotting, RNA sequencing, microarray chip and proteomics.

[0060] In some embodiments of the present invention, the identification module uses statistical methods or artificial intelligence methods to obtain genes with different expression levels between two groups of subjects.

[0061] The third aspect of the present invention provides a computer device, comprising: a memory for storing a computer program; a processor for implementing the steps of a method for determining genetic markers for cancer diagnosis, recurrence monitoring or prognosis assessment as described in any one of the first aspects of the present invention when executing the computer program.

[0062] The fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of a method for determining genetic markers for cancer diagnosis, recurrence monitoring or prognosis assessment as described in any one of the first aspects of the present invention are implemented.

[0063] Based on the same principle and process of the present invention, the method of the present invention can also be used to determine:

[0064] (1) Candidate genes that distinguish gliomas from other tumors and healthy people;

[0065] (2) candidate genes to distinguish other intracranial tumors other than gliomas from healthy subjects;

[0066] (3) candidate genes for distinguishing a cancer type selected from the group consisting of breast cancer, cervical cancer, colorectal cancer and ovarian cancer from other tumors other than the cancer type and healthy persons;

[0067] (4) A candidate gene that can be used for recurrence monitoring or prognosis assessment of a cancer selected from the group consisting of breast cancer, cervical cancer, colorectal cancer and ovarian cancer.

[0068] In some embodiments, the candidate genes that can distinguish glioma from other tumors and healthy people determined by the method of the present invention include ALOX15B, CEACAM4, GUCY2D, IL1R2, LINC00211, LPL, MFGE8, MYBPC1, OLAH, RSPO2, SH2D4A, SLC1A3, SLCO4A1-AS2, SRGAP1, ZBTB16, DLC1, FMN1, MPV17L, TMOD1, TSHR, DAAM2, LHFPL6, SLCO4A1, BLM, DNAH11, GLDN, LINC01736, THRB, B3GNT8, FKBP5, HEIH, METTL7A, TFC P2L1, CPM, MACIR, INSR, LINC01614, RPS24P19, MYOM2, TEX2, CD163, FOXD4, LINC01960, CD180, RPL36P9, FAM117B, LSP1P5, MMP9, RHOBTB2, ARHGEF18-AS1, EBI3, IL18R1, TMSB4XP4, MGAM, SESN1, SUGT1-DT, LINC02940, PALLD, SPON2, TMEM276, SYCP3, CITED4, GPRC5C, TMEM236, TMIGD3, BSND, HELZ-AS1, SCRG1 and SLC7A8. Further, a glioma binary risk assessment model was constructed using regression (in a specific implementation scheme, a generalized linear regression model GLMnet), which can distinguish glioma patients from other tumor patients and healthy people.

[0069] In some embodiments, the candidate genes that can distinguish brain cancer (including glioma and other intracranial tumors) from other tumors and healthy people determined by the method of the present invention include SPRY1, DDIT4, DEFA3, MMP8, ZBTB16, LTF, LCN2, ADAMTS2, MFGE8, SLCO4A1, OLFM4, GLDN, SRGAP1, SLC1A3, IL1R2, HBB, ALOX15B, THBS1, THRB, BCAS4, TPST1, AKAP13-A S1, CELF2-AS1, AREG, BSND, PDLIM4, MTND1P11, MT1G, FN1, GAPDHP60, NBPF19, CD68, CAVIN1, MRS2P2, LHFPL6, CHN2-AS1, AICDA, RAI14, NUDT19P5, GNAO1, CADM1, H1-4, LPL, FOXP1-IT1, KCNK5, LRRN1, GPER1, PTPRJ-AS1, SHISA8 and TSHR. Further, the risk of brain cancer can be calculated using a statistical method (ssGSEA in a specific embodiment), which can distinguish brain cancer (including glioma and other intracranial tumors) from other tumor patients and healthy people.

[0070] In some embodiments of the present invention, the candidate genes that can distinguish other intracranial tumors other than gliomas from healthy people are determined by the method of the present invention, including: SPRY1, DDIT4, AREG, PDLIM4, COL6A3, CELF2-AS1, AKAP13-AS1, DPYD-AS2, AFAP1, MTND1P11, GAPDHP60, H1-4, ZBTB16, RAI14, TRIL, MAP1B, ERVFRD-1, BRD7P3, CHN2-AS1, FLJ13224, HNRNPA3 P15, BSND, MTND5P32, FOXP1-IT1, RPS6KA2-AS1, PDLIM4, MTND1P11, MT1G, FN1, GAPDHP60, NBPF19, CD68, CAVIN1, MRS2P2, LHFPL6, CHN2-AS1, AICDA, RAI14, NUDT19P5, GNAO1, CADM1, H1-4, LPL, FOXP1-IT1, KCNK5, LRRN1, GPER1, PTPRJ-AS1, SHISA8 and TSHR. Further, a statistical method (ssGSEA in a specific embodiment) is used to calculate the risk score of other intracranial tumors, which can distinguish patients with other intracranial tumors from healthy people.

[0071] In some embodiments of the present invention, candidate genes that can distinguish breast cancer from other tumors and healthy people include: FAM3B, FHIP1A, MTND1P11, MTCO1P11, ITGA10, NBPF19, ZBTB47-AS1 and LYST. Further, a statistical method (ssGSEA in a specific embodiment) is used to calculate the breast cancer risk score, which can distinguish breast cancer patients from tumor patients and healthy people.

[0072] In some embodiments of the present invention, candidate genes that can distinguish cervical cancer from other tumors and healthy people are determined to include: ZNF683, MSLN, ID1, CCL3, DAPK1-IT1, CD8A, CCL2, TNF, KIF19, LINC00987, SLC1A7, CD8B, CCL4, LAG3, PZP, CTLA4, THEMIS, RPL12P42, CHRM4 and TENT5C. Further, a statistical method (ssGSEA in a specific embodiment) is used to calculate the cervical cancer risk score, which can distinguish cervical cancer patients from tumor patients and healthy people.

[0073] In some embodiments of the present invention, candidate genes that can distinguish colorectal cancer from other tumors and healthy people include: TMEM45A, BCAS4, ADAMTS2, FN1, KRT8, MAP1B, AKAP12, CD177, FOLR3, MID1, TMSB4XP4, KRT18, PTPRF, RAB26, CSNK2A3, VPS13C-DT, RPS15AP6, FLT3, TPST1, LPIN3, VSIG4, APOL4, IGFBP4, EIF4A1P2, SRGAP1, EIF4A1P10, PRR16, RN7SL832P, ISL2, CLEC10A, SHMT1 and TSSK6. Further, a colorectal cancer risk score is calculated using a statistical method (ssGSEA in a specific embodiment), which can distinguish colorectal cancer patients from tumor patients and healthy people.

[0074] In some embodiments of the present invention, candidate genes that can be determined to distinguish ovarian cancer from other tumors and healthy people include: HBB, ALAS2, NR4A3, CXCL3, CXCL8, EREG, CXCL2, MARK2P8, GOS2, NIPAL4, NRIP3, MAFF, F3, PHLDA1, MIR222HG, EHD4-AS1, OLR1, MAILR, NEU4, HBEGF, PDCD1, FOSL1, IL1B, L INC01093, RGCC, CCL3L1, SPRY2, LMOD2, SNAI1, DUSP8, TAMALIN, PDYN-AS1, CCL4L2, TP53INP2, NR4A1, PTX3, NKX3-1, FFAR3, DUSP2, BTG3-AS1, CCL4, TNFAIP3, DNAAF1, OSM, RRAD, ATF3, CD83, SELENBP1, JUN and ADORA2A. Further, a statistical method (ssGSEA in a specific embodiment) is used to calculate an ovarian cancer risk score, which can distinguish ovarian cancer patients from tumor patients and healthy people.

[0075] In an embodiment of the present invention, the candidate genes for distinguishing a certain cancer type from other tumors and healthy persons other than the cancer type can also be used for recurrence monitoring and / or prognosis assessment of the cancer.

[0076] Another aspect of the present invention provides a gene marker combination capable of distinguishing glioma from other tumors and healthy people, the gene marker combination comprising ALOX15B, CEACAM4, GUCY2D, IL1R2, LINC00211, LPL, MFGE8, MYBPC1, OLAH, RSPO2, SH2D4A, SLC1A3, SLCO4A1-AS2, SRGAP1, ZBTB16, DLC1, FMN1, MPV17L, TMOD1, TSHR, DAAM2, LHFPL6, SLCO4A1, BLM, DNAH11, GLDN, LINC01736, THRB, B3GNT8, FKBP5, HEIH, METTL7A, TFCP At least one of 2L1, CPM, MACIR, INSR, LINC01614, RPS24P19, MYOM2, TEX2, CD163, FOXD4, LINC01960, CD180, RPL36P9, FAM117B, LSP1P5, MMP9, RHOBTB2, ARHGEF18-AS1, EBI3, IL18R1, TMSB4XP4, MGAM, SESN1, SUGT1-DT, LINC02940, PALLD, SPON2, TMEM276, SYCP3, CITED4, GPRC5C, TMEM236, TMIGD3, BSND, HELZ-AS1, SCRG1, and SLC7A8.

[0077] Another aspect of the present invention provides a gene marker combination that can distinguish brain cancer (including glioma and other intracranial tumors) from other tumors and healthy people, and the gene marker combination includes SPRY1, DDIT4, DEFA3, MMP8, ZBTB16, LTF, LCN2, ADAMTS2, MFGE8, SLCO4A1, OLFM4, GLDN, SRGAP1, SLC1A3, IL1R2, HBB, ALOX15B, THBS1, THRB, BCAS4, TPST1, AKAP13-A At least one of S1, CELF2-AS1, AREG, BSND, PDLIM4, MTND1P11, MT1G, FN1, GAPDHP60, NBPF19, CD68, CAVIN1, MRS2P2, LHFPL6, CHN2-AS1, AICDA, RAI14, NUDT19P5, GNAO1, CADM1, H1-4, LPL, FOXP1-IT1, KCNK5, LRRN1, GPER1, PTPRJ-AS1, SHISA8 and TSHR.

[0078] Another aspect of the present invention provides a gene marker combination capable of distinguishing other intracranial tumors other than glioma from healthy people, the gene marker combination comprising SPRY1, DDIT4, AREG, PDLIM4, COL6A3, CELF2-AS1, AKAP13-AS1, DPYD-AS2, AFAP1, MTND1P11, GAPDHP60, H1-4, ZBTB16, RAI14, TRIL, MAP1B, ERVFRD-1, BRD7P3, CHN2-AS1, FLJ13224, HNRNPA3P15 , BSND, MTND5P32, FOXP1-IT1, RPS6KA2-AS1, PDLIM4, MTND1P11, MT1G, FN1, GAPDHP60, NBPF19, CD68, CAVIN1, MRS2P2, LHFPL6, CHN2-AS1, AICDA, RAI14, NUDT19P5, GNAO1, CADM1, H1-4, LPL, FOXP1-IT1, KCNK5, LRRN1, GPER1, PTPRJ-AS1, SHISA8 and TSHR.

[0079] Another aspect of the present invention provides a gene marker combination capable of distinguishing breast cancer from other tumors and healthy people, wherein the gene marker combination includes at least one of FAM3B, FHIP1A, MTND1P11, MTCO1P11, ITGA10, NBPF19, ZBTB47-AS1 and LYST.

[0080] Another aspect of the present invention provides a gene marker combination capable of distinguishing cervical cancer from other tumors and healthy people, wherein the gene marker combination includes at least one of ZNF683, MSLN, ID1, CCL3, DAPK1-IT1, CD8A, CCL2, TNF, KIF19, LINC00987, SLC1A7, CD8B, CCL4, LAG3, PZP, CTLA4, THEMIS, RPL12P42, CHRM4 and TENT5C.

[0081] Another aspect of the present invention provides a gene marker combination that can distinguish colorectal cancer from other tumors and healthy people, and the gene marker combination includes at least one of TMEM45A, BCAS4, ADAMTS2, FN1, KRT8, MAP1B, AKAP12, CD177, FOLR3, MID1, TMSB4XP4, KRT18, PTPRF, RAB26, CSNK2A3, VPS13C-DT, RPS15AP6, FLT3, TPST1, LPIN3, VSIG4, APOL4, IGFBP4, EIF4A1P2, SRGAP1, EIF4A1P10, PRR16, RN7SL832P, ISL2, CLEC10A, SHMT1 and TSSK6.

[0082] Another aspect of the present invention provides a gene marker combination that can distinguish ovarian cancer from other tumors and healthy people, the gene marker combination includes HBB, ALAS2, NR4A3, CXCL3, CXCL8, EREG, CXCL2, MARK2P8, G0S2, NIPAL4, NRIP3, MAFF, F3, PHLDA1, MIR222HG, EHD4-AS1, OLR1, MAILR, NEU4, HBEGF, PDCD1, FOSL1, IL1B , LINC01093, RGCC, CCL3L1, SPRY2, LMOD2, SNAI1, DUSP8, TAMALIN, PDYN-AS1, CCL4L2, TP53INP2, NR4A1, PTX3, NKX3-1, FFAR3, DUSP2, BTG3-AS1, CCL4, TNFAIP3, DNAAF1, OSM, RRAD, ATF3, CD83, SELENBP1, JUN and ADORA2A.

[0083] Another aspect of the present invention provides the use of an expression level detection reagent of any of the above gene marker combinations in a reagent kit for distinguishing the corresponding cancer from other tumors and healthy people, thereby completing the diagnosis, recurrence monitoring and prognosis evaluation of the corresponding cancer.

[0084] Another aspect of the present invention also provides a system for cancer diagnosis, recurrence monitoring or prognosis assessment, comprising:

[0085] A data input module is used to obtain the expression level of each gene in the gene marker combination;

[0086] The prediction module is connected to the data input module and is used to perform cancer diagnosis, recurrence monitoring or prognosis evaluation based on the expression level of each gene using statistical methods or artificial intelligence methods.

[0087] Optionally, the cancer is glioma, and the gene marker combination includes ALOX15B, CEACAM4, GUCY2D, IL1R2, LINC00211, LPL, MFGE8, MYBPC1, OLAH, RSPO2, SH2D4A, SLC1A3, SLCO4A1-AS2, SRGAP1, ZBTB16, DLC1, FMN1, MPV17L, TMOD1, TSHR, DAAM2, LHFPL6, SLCO4A1, BLM, DNAH11, GLDN, LINC01736, THRB, B3GNT8, FKBP5, HEIH, METTL7A, TFCP2L1, CPM, MACI At least one of R, INSR, LINC01614, RPS24P19, MYOM2, TEX2, CD163, FOXD4, LINC01960, CD180, RPL36P9, FAM117B, LSP1P5, MMP9, RHOBTB2, ARHGEF18-AS1, EBI3, IL18R1, TMSB4XP4, MGAM, SESN1, SUGT1-DT, LINC02940, PALLD, SPON2, TMEM276, SYCP3, CITED4, GPRC5C, TMEM236, TMIGD3, BSND, HELZ-AS1, SCRG1, and SLC7A8.

[0088] Optionally, the cancer is brain cancer (including glioma and other intracranial tumors), and the gene marker combination includes SPRY1, DDIT4, DEFA3, MMP8, ZBTB16, LTF, LCN2, ADAMTS2, MFGE8, SLCO4A1, OLFM4, GLDN, SRGAP1, SLC1A3, IL1R2, HBB, ALOX15B, THBS1, THRB, BCAS4, TPST1, AKAP13-AS1, CELF2-AS1 , AREG, BSND, PDLIM4, MTND1P11, MT1G, FN1, GAPDHP60, NBPF19, CD68, CAVIN1, MRS2P2, LHFPL6, CHN2-AS1, AICDA, RAI14, NUDT19P5, GNAO1, CADM1, H1-4, LPL, FOXP1-IT1, KCNK5, LRRN1, GPER1, PTPRJ-AS1, SHISA8 and TSHR.

[0089] Optionally, the cancer is an intracranial tumor other than glioma, and the gene marker combination includes SPRY1, DDIT4, AREG, PDLIM4, COL6A3, CELF2-AS1, AKAP13-AS1, DPYD-AS2, AFAP1, MTND1P11, GAPDHP60, H1-4, ZBTB16, RAI14, TRIL, MAP1B, ERVFRD-1, BRD7P3, CHN2-AS1, FLJ13224, HNRNPA3P15, BSND, MTND At least one of 5P32, FOXP1-IT1, RPS6KA2-AS1, PDLIM4, MTND1P11, MT1G, FN1, GAPDHP60, NBPF19, CD68, CAVIN1, MRS2P2, LHFPL6, CHN2-AS1, AICDA, RAI14, NUDT19P5, GNAO1, CADM1, H1-4, LPL, FOXP1-IT1, KCNK5, LRRN1, GPER1, PTPRJ-AS1, SHISA8 and TSHR.

[0090] Optionally, the cancer is breast cancer, and the gene marker combination includes at least one of FAM3B, FHIP1A, MTND1P11, MTCO1P11, ITGA10, NBPF19, ZBTB47-AS1 and LYST.

[0091] Optionally, the cancer is cervical cancer, and the gene marker combination includes at least one of ZNF683, MSLN, ID1, CCL3, DAPK1-IT1, CD8A, CCL2, TNF, KIF19, LINC00987, SLC1A7, CD8B, CCL4, LAG3, PZP, CTLA4, THEMIS, RPL12P42, CHRM4 and TENT5C.

[0092] Optionally, the cancer is colorectal cancer, and the gene marker combination includes at least one of TMEM45A, BCAS4, ADAMTS2, FN1, KRT8, MAP1B, AKAP12, CD177, FOLR3, MID1, TMSB4XP4, KRT18, PTPRF, RAB26, CSNK2A3, VPS13C-DT, RPS15AP6, FLT3, TPST1, LPIN3, VSIG4, APOL4, IGFBP4, EIF4A1P2, SRGAP1, EIF4A1P10, PRR16, RN7SL832P, ISL2, CLEC10A, SHMT1 and TSSK6.

[0093] Optionally, the cancer is ovarian cancer, and the gene marker combination includes HBB, ALAS2, NR4A3, CXCL3, CXCL8, EREG, CXCL2, MARK2P8, G0S2, NIPAL4, NRIP3, MAFF, F3, PHLDA1, MIR222HG, EHD4-AS1, OLR1, MAILR, NEU4, HBEGF, PDCD1, FOSL1, IL1B, LINC01093, R At least one of GCC, CCL3L1, SPRY2, LMOD2, SNAI1, DUSP8, TAMALIN, PDYN-AS1, CCL4L2, TP53INP2, NR4A1, PTX3, NKX3-1, FFAR3, DUSP2, BTG3-AS1, CCL4, TNFAIP3, DNAAF1, OSM, RRAD, ATF3, CD83, SELENBP1, JUN and ADORA2A.

[0094] Artificial Intelligence (AI) is an interdisciplinary subject based on computer science, integrating mathematics, informatics, control science and other subjects.

[0095] Furthermore, the artificial intelligence method is implemented by means of machine learning. Machine learning includes many different algorithms, deep learning is one of them, and other methods include logistic regression, decision tree, random forest, support vector machine, naive Bayes, K nearest neighbor, neural network, etc.

[0096] Beneficial effects of the present invention

[0097] Compared with the prior art, the present invention has the following technical effects:

[0098] The method of the present invention can be used to determine gene markers for diagnosis, recurrence monitoring or prognosis evaluation of multiple cancer types, and to construct corresponding risk assessment models using the determined gene markers, with excellent performance.

[0099] The method of the present invention can also be used to determine candidate genes for distinguishing multiple cancer types from other tumors and healthy people.

[0100] The method of the present invention can also be used to determine candidate genes for recurrence monitoring or prognosis assessment of multiple cancer types.

[0101] The algorithm model constructed using the candidate genes determined by the method of the present invention has excellent performance in disease identification, recurrence monitoring or prognosis evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] Figure 1The performance of the risk assessment model constructed using glioma candidate genes in Example 2 of the present invention in identifying gliomas in the training set is shown.

[0103] Figure 2 The performance of the risk assessment model constructed using glioma candidate genes in Example 2 of the present invention in identifying gliomas in the validation set is shown.

[0104] Figure 3 The performance of the risk assessment model constructed by using candidate genes for distinguishing brain cancer (including glioma and intracranial tumor) from other tumors and healthy people in Example 3 of the present invention in identifying brain cancer is shown.

[0105] Figure 4 The performance of the risk assessment model constructed using candidate genes for distinguishing other intracranial tumors (non-glioma) from healthy persons in Example 4 of the present invention in identifying other intracranial tumors (non-glioma) is shown.

[0106] Figure 5 The performance of the risk assessment model constructed using candidate genes for distinguishing breast cancer from healthy people in Example 5 of the present invention in identifying breast cancer is shown.

[0107] Figure 6 The performance of the risk assessment model constructed using candidate genes for distinguishing cervical cancer from healthy people in Example 6 of the present invention in identifying cervical cancer is shown.

[0108] Figure 7 The performance of the risk assessment model constructed using candidate genes for distinguishing colorectal cancer from healthy people in Example 7 of the present invention in identifying colorectal cancer is shown.

[0109] Figure 8 The performance of the risk assessment model constructed using candidate genes for distinguishing gastric cancer from healthy subjects in Example 8 of the present invention in identifying gastric cancer is shown.

[0110] Fig. 9 The performance of the risk assessment model constructed using candidate genes for distinguishing ovarian cancer from healthy subjects in Example 9 of the present invention in identifying ovarian cancer is shown.

[0111] Fig.10 The performance of the risk assessment model constructed using candidate genes for distinguishing glioma recurrence and non-recurrence in Example 10 of the present invention in identifying glioma recurrence is shown.

[0112] Fig.11 The performance of the risk assessment model constructed using candidate genes that distinguish between pre- and post-operative gliomas in Example 11 of the present invention for identifying post-operative gliomas is shown.

[0113] Fig.12 The results of the postoperative survival analysis of glioma in Example 11 of the present invention are shown.

[0114] Fig.13 The risk assessment model constructed by using candidate genes for distinguishing pre-operative and post-operative glioma in Example 11 of the present invention predicts disease progression results.

[0115] Fig.14 The performance of the risk assessment model constructed using candidate genes for distinguishing gliomas from other intracranial tumors in Example 12 of the present invention in identifying gliomas in an independent validation set is shown. DETAILED DESCRIPTION

[0116] Unless otherwise indicated, implied from the context, or customary in the prior art, all parts and percentages in this application are based on weight, and the tests and characterization methods used are all synchronized with the filing date of this application. Where applicable, the contents of any patent, patent application or disclosure involved in this application are fully incorporated herein by reference, and their equivalent patent families are also introduced as references, especially the definitions of relevant terms in the art disclosed in these documents. If the definition of a specific term disclosed in the prior art is inconsistent with any definition provided in this application, the definition of the term provided in this application shall prevail.

[0117] The following examples are used to demonstrate preferred embodiments of the present invention. It will be appreciated by those skilled in the art that the techniques disclosed in the following examples represent techniques discovered by the inventors that can be used to implement the present invention and therefore can be considered as preferred embodiments of the present invention. However, it will be appreciated by those skilled in the art based on this specification that many modifications may be made to the specific embodiments disclosed herein and still achieve the same or similar results without departing from the spirit or scope of the present invention.

[0118] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0119] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many technical equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the claims.

[0120] The experimental methods in the following examples are conventional methods unless otherwise specified. The instruments and equipment used in the following examples are conventional laboratory instruments and equipment unless otherwise specified; the experimental materials used in the following examples are purchased from conventional biochemical reagent stores unless otherwise specified.

[0121] Example 1 Peripheral blood mononuclear cell RNA sequencing

[0122] 1. Peripheral Blood Collection

[0123] The inventors collected peripheral blood samples from 59 glioma patients, 16 patients with other intracranial tumors (non-glioma), 6 breast cancer patients, 4 cervical cancer patients, 10 colorectal cancer patients, 9 gastric cancer patients, 3 ovarian cancer patients, 1 liver cancer patient and 11 healthy people.

[0124] 2. Monocyte Isolation

[0125] Use 10mL DPBS to dilute whole blood, with a blood sample and lymphocyte separation solution volume ratio of 2:1, first add the layering solution and then the diluted blood, so that the layering solution is at the bottom and the diluted blood is stacked on the layering solution; use a horizontal centrifuge, centrifuge horizontally for 20 minutes, extract mononuclear cells after centrifugation, transfer to another centrifuge tube, add DPBS to wash again, centrifuge horizontally for 10 minutes, and discard the supernatant; after discarding the supernatant, add 80μL of Buffer solution to the centrifuge tube, resuspend the precipitate in the tube, and mix slowly; then add 20μL of CD14 magnetic beads, mix gently, and incubate in the refrigerator for 15 minutes, keeping mixing every 5 minutes; after incubation, add 5mL of Buffer solution to the tube and centrifuge for 5 minutes; discard the supernatant again and resuspend the cell precipitate with 500μL Buffer; finally, put the sorting column into the magnetic rack, first use 500μL of Buffer to rinse the sorting column, and then slowly drip the cells. The collected cells at the bottom are CD14 - Mononuclear cells, use the piston to push the liquid out of the sorting column, which is CD14 + Monocytes.

[0126] 3. Mononuclear cell RNA extraction and sequencing

[0127] RNA was extracted from each sample to obtain transcriptome RNA; samples with RNA fragments that were too large were broken into 200-350 base pairs by ultrasonic fragmentation; the fragmented RNA molecules were subjected to end repair, purine addition, library adapter connection and other operations; the obtained RNA fragments were hybridized with single-stranded biotin-labeled DNA probe molecules with a library length of 120 bases, and the captured RNA library molecules were separated with streptavidin-coated magnetic beads; high-throughput sequencing was performed on a second-generation sequencer.

[0128] 4. Sequencing Data Processing

[0129] Each sample sequence obtained by sequencing is filtered to remove unqualified sequences and adapter sequences; wherein the unqualified sequence can be at least one of the following:

[0130] Reads with an uncertain base N ratio greater than 10% and reads with bases with a base quality value Q less than 10 accounting for more than 50% of the entire reads.

[0131] Each filtered sample sequence is aligned to the reference genome sequence, each sample sequence after alignment is screened to obtain a unique aligned sample sequence, the position information of each unique aligned sample sequence relative to the reference genome sequence is determined, and the position information is sorted. Specifically:

[0132] First, each sample sequence obtained by filtering is aligned to the reference genome sequence through the RNA alignment mapping program to obtain the position of each sample sequence on the reference genome.

[0133] Then, the alignment results were quantified, the number of aligned sequences on the gene / transcript was counted, and the gene expression profile of monocytes was obtained.

[0134] Example 2 Identification of specific highly expressed genes in peripheral blood mononuclear cells of glioma

[0135] The above 119 samples are divided into a discovery data set and a validation data set in a ratio of 7:3, as shown in Table 1.

[0136] Table 1 Sample grouping

[0137]

[0138] In the discovery dataset, the following steps were performed to identify genes that were specifically highly expressed in peripheral blood mononuclear cells of glioma:

[0139] Step a, divide the discovery dataset into training set and test set in a ratio of 3:1;

[0140] Step b: In the training set, the edgeR algorithm was used to calculate the differentially expressed genes between glioma and other tumors and healthy people. Genes with FC≥1.5 and FDR<0.1 were selected as genes that were highly expressed in peripheral monocytes of glioma, and after sorting by FDR, the top 100 highly expressed genes were selected as the candidate gene set;

[0141] Step c, repeating steps a and b 50 times to obtain 50 candidate gene sets;

[0142] Step d, count the frequency of each highly expressed gene, and select candidate genes with a frequency ≥ 30 as stable specific highly expressed genes in glioma peripheral blood mononuclear cells, as shown in Table 2, a total of 69 genes.

[0143] Table 2 Candidate genes for distinguishing glioma from other tumors and healthy people

[0144]

[0145] The performance of the 69 genes in identifying gliomas in the discovery dataset and validation dataset is shown in Table 3 , and the results show that 51 genes with AUC>0.5 for identifying gliomas in the validation set.

[0146] Table 3 Performance of glioma candidate genes in identifying glioma in the discovery dataset and validation dataset

[0147]

[0148] Furthermore, the inventors used glioma as case samples and other tumors and healthy samples as control samples in the discovery data set, and based on the above 69 candidate genes, used the generalized linear regression model GLMnet to construct a glioma binary classification risk assessment model.

[0149] Performance of the risk assessment model in identifying gliomas in the discovery dataset Figure 1 The performance of identifying gliomas in the validation dataset is shown in Figure 2 shown.

[0150] Example 3 Identification of Specific Highly Expressed Genes in Peripheral Blood Mononuclear Cells of Brain Cancer

[0151] Using the method of Example 2, candidate genes for distinguishing brain cancer (including glioma and intracranial tumor) from other tumors and healthy people were screened, as shown in Table 4, a total of 50 genes.

[0152] Table 4 Candidate genes for distinguishing brain cancer (including glioma and intracranial tumor) from other tumors and healthy people

[0153]

[0154] The above candidate genes were used to construct a brain cancer risk score. The specific method was to use the above genes to construct a gene set. Based on this gene set, the single sample gene set enrichment analysis (ssGSEA) algorithm was used to calculate the gene set activation level score of the sample to be tested, which was defined as the brain cancer risk score. The results are as follows Figure 3 As shown, the brain cancer risk score in peripheral monocytes of brain cancer patients was significantly higher than that in other tumor patients and healthy people.

[0155] Example 4 Identification of specific highly expressed genes in peripheral blood mononuclear cells of intracranial tumors

[0156] Using the method of Example 2, candidate genes for distinguishing other intracranial tumors (non-glioma) from healthy people were screened, as shown in Table 5, a total of 50 genes.

[0157] Table 5 Candidate genes for distinguishing other intracranial tumors (non-glioma) from healthy subjects

[0158]

[0159] The above candidate genes were used to construct other intracranial tumor risk scores. The specific method was to use the above genes to construct a gene set. Based on this gene set, the single sample gene set enrichment analysis (ssGSEA) algorithm was used to calculate the gene set activation level score of the sample to be tested, which was defined as the other intracranial tumor risk score. Figure 4 As shown, the risk scores of other intracranial tumors in peripheral mononuclear cells of patients with other intracranial tumors were significantly higher than those in healthy people.

[0160] Example 5 Identification of Specific Highly Expressed Genes in Peripheral Blood Mononuclear Cells of Breast Cancer

[0161] Using the method of Example 2, candidate genes for distinguishing breast cancer from other tumors and healthy people were screened, as shown in Table 6, a total of 8 genes.

[0162] Table 6 Candidate genes for distinguishing breast cancer from other tumors and healthy people

[0163]

[0164] The above candidate genes were used to construct a breast cancer risk score. The specific method was to use the above genes to construct a gene set. Based on this gene set, the single sample gene set enrichment analysis (ssGSEA) algorithm was used to calculate the gene set activation level score of the sample to be tested, which was defined as the breast cancer risk score. The results are as follows Figure 5 As shown, the breast cancer risk score in peripheral mononuclear cells of breast cancer patients was significantly higher than that of other tumor patients and healthy people.

[0165] Example 6 Identification of Specific Highly Expressed Genes in Peripheral Blood Mononuclear Cells of Cervical Cancer

[0166] Using the method of Example 2, candidate genes for distinguishing cervical cancer from other tumors and healthy people were screened, as shown in Table 7, a total of 20 genes.

[0167] Table 7 Candidate genes for distinguishing cervical cancer from other tumors and healthy people

[0168]

[0169] The above candidate genes were used to construct a cervical cancer risk score. The specific method was to use the above genes to construct a gene set. Based on this gene set, the single sample gene set enrichment analysis (ssGSEA) algorithm was used to calculate the gene set activation level score of the sample to be tested, which was defined as the cervical cancer risk score. The results are as follows Figure 6 As shown, the cervical cancer risk score in peripheral mononuclear cells of cervical cancer patients was significantly higher than that of other tumor patients and healthy people.

[0170] Example 7 Identification of genes specifically highly expressed in peripheral blood mononuclear cells of colorectal cancer

[0171] Using the method of Example 2, candidate genes for distinguishing colorectal cancer from other tumors and healthy people were screened, as shown in Table 8, totaling 32 genes.

[0172] Table 8 Candidate genes for distinguishing colorectal cancer from other tumors and healthy people

[0173]

[0174] The above candidate genes were used to construct a colorectal cancer risk score. The specific method was to use the above genes to construct a gene set. Based on this gene set, the single sample gene set enrichment analysis (ssGSEA) algorithm was used to calculate the gene set activation level score of the sample to be tested, which was defined as the colorectal cancer risk score. Figure 7 As shown, the colorectal cancer risk score in peripheral mononuclear cells of colorectal cancer patients was significantly higher than that in other tumors / healthy people.

[0175] Example 8 Identification of Specific Highly Expressed Genes in Gastric Cancer Peripheral Blood Mononuclear Cells

[0176] Using the method of Example 2, candidate genes for distinguishing gastric cancer from other tumors and healthy people were screened, as shown in Table 9, a total of 3 genes.

[0177] Table 9 Candidate genes for distinguishing gastric cancer from other tumors and healthy people

[0178]

[0179] The above candidate genes were used to construct a gastric cancer risk score. The specific method was to use the above genes to construct a gene set. Based on this gene set, the single sample gene set enrichment analysis (ssGSEA) algorithm was used to calculate the gene set activation level score of the sample to be tested, which was defined as the gastric cancer risk score. Figure 8 As shown, there was no significant difference in gastric cancer risk score between peripheral mononuclear cells of gastric cancer patients and other tumor / healthy people.

[0180] Example 9 Identification of Specific Highly Expressed Genes in Peripheral Blood Mononuclear Cells of Ovarian Cancer

[0181] Using the method of Example 2, candidate genes for distinguishing ovarian cancer from other tumors and healthy people were screened, as shown in Table 10, totaling 50 genes.

[0182] Table 10 Candidate genes for distinguishing ovarian cancer from other tumors and healthy people

[0183]

[0184] The above candidate genes were used to construct an ovarian cancer risk score. The specific method was to use the above genes to construct a gene set. Based on this gene set, the single sample gene set enrichment analysis (ssGSEA) algorithm was used to calculate the gene set activation level score of the sample to be tested, which was defined as the ovarian cancer risk score. The results are as follows Fig. 9 As shown, the ovarian cancer risk score in peripheral mononuclear cells of ovarian cancer patients was significantly higher than that of other tumor patients and healthy people.

[0185] Example 10 Brain Cancer Recurrence Risk Assessment Model - Independent Validation Set

[0186] 28 peripheral blood mononuclear cell samples of glioma without recurrence within 2 years after surgery and 25 peripheral blood mononuclear cell samples of glioma recurred 1-2 years after surgery were collected, and the gene expression profiles of the mononuclear cells of each sample were obtained using the same method as in Example 1.

[0187] The glioma risk score of the sample was directly calculated using the glioma binary risk assessment model constructed in Example 2. Fig.10 As shown, the glioma risk score in peripheral mononuclear cells of glioma recurrence samples was significantly higher than that in glioma non-recurrence samples.

[0188] Example 11 Prognostic Progress Assessment of Brain Cancer - Independent Validation Set

[0189] Paired peripheral blood mononuclear cell samples were collected from 29 glioma patients before and one week after surgery, and the gene expression profiles of the mononuclear cells in each sample were obtained using the same method as in Example 1.

[0190] The glioma risk score of the sample was directly calculated using the glioma binary risk assessment model constructed in Example 2. Fig.11 As shown, the tumor score of peripheral mononuclear cells decreased significantly one week after surgery.

[0191] Further survival analysis was performed on the tumor scores of peripheral mononuclear cells after surgery. Fig.12 As shown in the results, in patients with glioma, especially low-grade glioma, patients with high tumor scores of peripheral mononuclear cells after surgery had a shorter time to disease progression. Furthermore, combined with the EOR analysis of the surgical resection area, it was found that in low-grade glioma LGG, the tumor score of peripheral mononuclear cells can predict disease progression independently of EOR, such as Fig.13 shown.

[0192] Example 12 Independent Validation Set of Example 2

[0193] In order to verify the stability of Example 2, the inventors collected 10 patients with glioma, 17 patients with other intracranial tumors (including 7 cases of meningioma, 4 cases of meningioma, 3 cases of craniopharyngioma, 2 cases of chordoma and 1 case of neuroepithelial tumor), and 11 cases of other cancers (including 4 cases of breast cancer, 5 cases of intestinal cancer and 2 cases of gastric cancer), and obtained the gene expression profile of monocytes of each sample using the same method as Example 1.

[0194] The glioma risk score of the sample was directly calculated using the glioma binary risk assessment model constructed in Example 2. Fig.14 As shown, the glioma risk score in peripheral mononuclear cells of glioma samples was significantly higher than that of other intracranial tumors and other cancers.

[0195] All documents mentioned in the present invention are cited as references in this application, just as each document is cited as reference individually. In addition, it should be understood that after reading the above teachings of the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the claims attached to this application.

Claims

1. A method for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment, characterized in that, The following steps are involved: S1, obtaining RNA samples of peripheral blood mononuclear cells from two groups of subjects, wherein the two groups of subjects are selected from one of the following situations: (1) One group includes multiple cancer patients and the other group includes multiple non-cancer subjects, (2) One group included patients with multiple cancer prognoses of recurrence, and the other group included patients with multiple cancer prognoses of non-recurrence, (3) one group includes multiple patients with good cancer prognosis, and the other group includes multiple patients with poor cancer prognosis; S2, obtaining the expression level of each gene based on the RNA sample obtained in step S1, and obtaining a gene expression profile, wherein the gene expression profile includes the expression level of each gene in each subject; S3, based on the gene expression profile obtained in step S2, obtaining genes with different expression levels between the two groups of subjects as candidate gene markers, Wherein, the cancer is selected from one of brain cancer, breast cancer, cervical cancer, colorectal cancer and ovarian cancer.

2. A method for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment according to claim 1, characterized in that: The brain cancer includes glioma and other intracranial tumors except glioma.

3. A method for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment according to claim 1, characterized in that: The recurrence or non-recurrence refers to the patient's recurrence or non-recurrence within 3 months, 6 months, 1 year, 2 years, 3 years, 5 years, 10 years or the entire survival period of the prognosis.

4. The method for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment according to claim 1, characterized in that: In step S2, the expression level of each gene is obtained by using at least one method selected from reverse transcription PCR, fluorescent quantitative PCR, in situ hybridization, Northern blotting, RNA sequencing, microarray chip, and proteomics.

5. The method for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment according to claim 1, characterized in that: In step S3, genes with different expression levels between the two groups of subjects are obtained using statistical methods or artificial intelligence.

6. A system for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment, characterized in that: Includes the following modules: An information input module, used to obtain a gene expression profile based on RNA samples of peripheral blood mononuclear cells of two groups of subjects, wherein the gene expression profile includes the expression level of each gene in each subject; an identification module, connected to the information input module, for obtaining genes with different expression levels between the two groups of subjects as candidate gene markers based on the gene expression profile, Wherein, the two groups of subjects are selected from one of the following situations: (1) One group includes multiple cancer patients and the other group includes multiple non-cancer subjects, (2) One group included patients with multiple cancer prognoses of recurrence, and the other group included patients with multiple cancer prognoses of non-recurrence, (3) One group includes multiple patients with good cancer prognosis, and the other group includes multiple patients with poor cancer prognosis. Wherein, the cancer is selected from one of brain cancer, breast cancer, cervical cancer, colorectal cancer and ovarian cancer.

7. A system for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment according to claim 6, characterized in that: The expression level of each gene is obtained by using at least one method selected from reverse transcription PCR, fluorescent quantitative PCR, in situ hybridization, Northern blotting, RNA sequencing, microarray chip and proteomics.

8. The system for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment according to claim 6, characterized in that: The identification module uses statistical methods or artificial intelligence methods to obtain genes with different expression levels between two groups of subjects.

9. A computer device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of a method for determining gene markers for cancer diagnosis, recurrence monitoring or prognosis assessment as described in any one of claims 1 to 5 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of a method for determining genetic markers for cancer diagnosis, recurrence monitoring or prognosis assessment as described in any one of claims 1 to 5 are implemented.

Images