Application technology for accurate early screening and diagnosis of cervical cancer and cervical lesion triage

Abstract

The invention discloses an application technology for accurate early screening and diagnosis of cervical cancer and cervical lesion triage, and relates to the field of application of biotechnology and medical technology. A large model is utilized to solve the problem that most kits can only analyze the sensitivity and specificity of results of two purposes, the analysis of more than three purposes of results (cervical normal / cervicitis, LSIL, HSIL (or CIN2, CIN3) and cervical cancer patients) is effectively solved, samples such as cervical exfoliated cells / cervical tissues / cervical paraffin tissues / cervical paraffin sections and the like are collected, and the samples are subjected to quantitative analysis. By detecting the DNA expression quantity / expression level / content after HPV typing and analyzing the DNA methylation size of a marker gene, the application of early screening and diagnosis of cervical cancer and triage of cervical lesions is successfully realized by utilizing the idea of a large model; the system can also be used for treating cervical lesion patients through non-invasive measurement, evaluating the prognosis treatment effect of individuals, excavating the change trend of the illness condition of the individual patients, and reducing the pain caused by repeated pathologic biopsy examination for multiple times of the illness patients.

Description

Technical Field

[0001] The present invention relates to the field of early screening, early diagnosis and early triage prevention of cervical cancer using big data and molecular biology analysis, and in particular to a mathematical model and method for integrating DNA expression levels of high-risk human papillomavirus (HPV) typing and the risk of related gene methylation. Background Art

[0002] Cervical cancer is a major malignant tumor that women are prone to worldwide. Currently, cervical cancer screening includes cervical cytology and human papillomavirus (HPV) testing. Cytology testing includes Pap smears and TCT liquid-based cell compression films, and its diagnostic level is greatly affected by the subjective factors of physicians. Compared with cytology testing, HPV-DNA typing testing can type high-risk and low-risk HPV. Cervical cancer is usually associated with persistent high-risk HPV infection, such as HPV16 or HPV18 or HPV45. In addition to population screening and HPV vaccination, effective treatments for cervical cancer are still limited. Research on the risk level of methylation of cervical cancer-related genes has made some progress. However, with the rapid development of big data, artificial intelligence, and the digitalization of the Internet of Things, many test kits are still in the qualitative stage and lack quantification; in particular, a large number of test kits can only discuss two target results after specificity and sensitivity studies, and lose the results of evaluating specificity and sensitivity for more than three target results. In order to overcome the above shortcomings, a mathematical model and method were established by linking the DNA expression of high-risk HPV typing associated with high-risk HPV infection with the risk of methylation of cervical cancer-related genes through the evolution and progression of cervical lesions in cervical cancer pathology. This new research has not been reported in domestic and foreign research. The present invention aims to extract the DNA of cervical cells to be tested through a sample of human cervical exfoliated cells or cervical lesion tissue, detect the data of HPV typing DNA expression and cervical cancer-related gene methylation risk research at one time, and analyze and predict the lesion risk of a normal cervix or lesion cervix of an individual woman, so as to obtain an application for accurate early screening and early diagnosis of cervical cancer and cervical lesion triage. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide an application for detecting human cervical exfoliated cells or cervical lesion tissue samples, accurately predicting early screening and early diagnosis of cervical cancer and cervical lesion triage. It can also be measured non-destructively to assist the existing prognosis diagnosis and treatment effect evaluation to improve the five-year survival rate of cervical cancer patients.

[0004] The present invention relates to a cervical cancer having an obvious marker gene, collecting cervical lesion tissue as a sample, and detecting the DNA methylation level of the gene marker using methylation-sensitive restriction endonuclease-PCR / Southern method, pyrosequencing, combined bisulfite restriction endonuclease method (Combined Bisulfite Restriction Analysis, COBRA) and other measurement methods. The risk of the progression of pathological cervical lesions is analyzed by DNA methylation level, characterized in that for patients with different degrees of cervical lesions (including normal cervical / cervical inflammation patients, CIN 1, CIN2, CIN 2, cervical cancer, pathology / histology / clinical staging of cervical cancer, LSIL and HSIL patients), a functional equation is used to represent the patients or risk levels of the above-mentioned different degrees of cervical lesions, such as the function y=A exp(Bx); y=A×x B (A, B, C are model parameters), etc.; or the patients with different degrees of cervical lesions described above (or similar expressions of the evolution process from normal cervix to cervical lesions to the occurrence and development of cervical cancer) are directly represented by numbers to represent categorical variables, and any two or more categorical variables are selected as variables for classification labels; categorical variables can also be expressed in any n-dimensional vector form (n>=2).

[0005] The present invention does not include women with other autoimmune diseases or malignant tumors; intact hymen; combined uncontrolled internal and external diseases; and patients with a history of cervical surgery. By collecting body fluids (blood, urine, etc.) related to normal cervix, cervical lesions and cervical cancer tissues, cervical scrapings, exfoliated cells, paraffin tissue or paraffin section genes and biological samples of fresh tissues, for example, HPV-DNA typing test is performed on women with exfoliated cells of cervical tissue. A special cell brush is inserted through the vagina to 1.5 cm into the endocervical canal, and rotated in a fixed direction for 8 times with appropriate force at the junction of squamous-columnar epithelium. After collection, the sample is immediately placed in a preservation solution and stored at 4°C for inspection. After lysis, DNA is extracted, and high-risk HPV DNA detection is performed using a fluorescent quantitative polymerase chain reaction (PCR) instrument. The PCR amplification conditions and reaction system are all referred to the instructions of the kit, and the DNA expression amount of high-risk HPV DNA typing is analyzed according to the system's built-in software. Similarly, all molecular tests such as quantitative PCR, fluorescence in situ hybridization (FISH), real-time fluorescence quantitative PCR (qPCR), fluorescence quantitative PCR (qPCR), PCR-fluorescent probe, microarray technology, RNA-Seq and protein sequencing, biochip detection, etc. can obtain the expression level / expression amount / content of high-risk HPV-16, 18, 31, 33, 35, 45, 51, 52, 56, and 58 typing DNA, and use HPV-DAN16, HPV-DAN18, HPV-DAN31, HPV-DAN33, HPV-DAN35, HPV-DAN45, HPV-DAN51, HPV-DAN52, HPV-DAN56, and HPV-DAN58 as markers to represent the meaning, and any one or two or more markers from the above markers can be used as a variable.

[0006] The present invention relates to the second and third paragraphs of the above invention content, by adopting univariate or multivariate linear functions (including all linear models such as linear regression equations); or by adopting univariate or multivariate nonlinear functions (including all nonlinear models such as nonlinear regression equations); or by adopting Beyes, decision trees, support vector machine classification algorithms, machine learning classification, deep learning classification, artificial intelligence classification recognition, and classification algorithm recognition methods of applied mathematical classification algorithms, respectively achieving the application purposes of reference, screening, prediction, auxiliary diagnosis and diagnosis of cervical cancer and cervical lesion triage; or for the purpose of prognosis and treatment of cervical cancer.

[0007] The present invention relates to the second, third and fourth paragraphs of the above invention content, the concepts, ideas and methods described above, and the combination of any one or more of HPV-DAN16, HPV-DAN18, HPV-DAN31, HPV-DAN33, HPV-DAN35, HPV-DAN45, HPV-DAN51, HPV-DAN52, HPV-DAN56 and HPV-DAN58 with other mRNA expression, protein expression, DNA methylation level and other markers to achieve the above 3 application purposes. Or achieve the purpose of prognosis and treatment of cervical cancer.

[0008] The present invention relates to the second, third and fourth paragraphs or the second, third, fourth and fifth paragraphs of the above invention content, the concepts, ideas and methods described, construct two or more function equations / mathematical models / discriminant functions / similar recognition functions / similar recognition equations / similar recognition methods, and integrate their different recognition methods for comprehensive evaluation, so as to improve the accuracy of screening, prediction, auxiliary diagnosis and diagnosis of cervical cancer and its cervical lesion triage. DETAILED DESCRIPTION

[0009] Embodiment: By selecting cervical cancer-related marker genes, collecting samples of diseases at different stages of cervical tissue evolution, detecting the corresponding DNA methylation size, and using big data model calculations, the method of the second paragraph of the present invention is used to obtain the evaluation interval of cervical cancer [11.41, 291.79); for a 55-year-old woman who perceives abnormal cervical lesions, the DNA of her cervical tissue exfoliated cells is extracted, and HPV-DNA typing is performed using modern detection methods. The method of the third paragraph of the present invention is used to obtain the HPV-DNA18 of this woman, which is equal to 357.60, HPV-DNA 16, which is equal to 329.70, and HPV-DNA 45, which is equal to 231.70. At the same time, the combination of HPV-DNA 16, HPV-DNA 18 and HPV-DNA 45 markers is used as a variable (three). The present invention is related to the big data calculation method of the fourth paragraph of the invention content, and a specific function equation (mathematical model) is obtained: y = 0.223 × HPV-DNA18-0.118 × HPV-DNA16-0.087 × HPV-DNA45-0.805, where y is the predicted value. Substituting the expression value of the high-risk HPV typing DNA into the above specific function equation, a specific value of 19.8773 is obtained. This specific value is within the evaluation interval of cervical cancer [11.41, 291.70), and it is predicted that the woman is a cervical cancer patient. This prediction and inference result was proven. The squamous-columnar epithelium, columnar epithelium, and squamous epithelium of the woman's cervix were stained in reverse order using acetic acid and iodine. Samples were taken from suspicious sites with abnormalities. Single samples were taken from four weeks of the cervix of patients without significant abnormalities and sent for testing. Experienced testing technicians analyzed the prepared and stained samples. The results of a cervical tissue pathology biopsy under an electronic colposcope proved that the woman had cervical cancer.

Claims

1. An application technology for accurate early screening and diagnosis of cervical cancer and cervical lesion triage, characterized in that: For patients with different degrees of cervical lesions (including normal cervical / cervical inflammation patients, CIN 1, CIN 2, CIN 3, cervical cancer, pathological / histological / clinical staging of cervical cancer, LSIL and HSIL patients), a functional equation is used to represent the patients or risk levels of the above different degrees of cervical lesions, such as the function y=A exp(Bx); y=A×x B (A, B, C are model parameters), etc.; or the patients with different degrees of cervical lesions described above (or similar expressions of the evolution process from normal cervix to cervical lesions to the occurrence and development of cervical cancer) are directly represented by numbers to represent categorical variables, and any two or more categorical variables are selected as variables for classification labels; categorical variables can also be expressed in any n-dimensional vector form (n>=2).

2. An application technology for accurate early screening and diagnosis of cervical cancer and cervical lesion triage, characterized in that: By collecting biological samples of body fluids (blood, urine, etc.) related to normal cervix, cervical lesions and cervical cancer tissues, cervical scrapings, exfoliated cells, paraffin tissue or paraffin section genes and fresh tissues, and then going through the HPV typing process, all molecular tests such as quantitative PCR, fluorescence in situ hybridization (FISH), real-time fluorescence quantitative PCR (qPCR), fluorescence quantitative PCR (qPCR), PCR-fluorescent probe method, microarray technology, RNA-Seq and protein sequencing, biochip detection, etc. can be used to obtain high-risk HPV-1 6, 18, 31, 33, 35, 45, 51, 52, 56, 58 typing DNA expression level / expression amount / content, and HPV-DAN16, HPV-DAN18, HPV-DAN31, HPV-DAN33, HPV-DAN35, HPV-DAN45, HPV-DAN51, HPV-DAN52, HPV-DAN56, HPV-DAN58 are used as markers to represent the meaning, and any one or two or more markers selected from the above markers are used as variables.

3. An application technology for accurate early screening and diagnosis of cervical cancer and cervical lesion triage, characterized in that: By using the above 1 and 2, by adopting univariate or multivariate linear functions (including all linear models such as linear regression equations); or by adopting univariate or multivariate nonlinear functions (including all nonlinear models such as nonlinear regression equations); or by adopting the Beyes, decision tree, support vector machine classification algorithm, machine learning classification, deep learning classification, artificial intelligence classification recognition, and classification algorithm recognition methods of applied mathematical classification algorithms, the application purposes of reference, screening, prediction, auxiliary diagnosis and diagnosis of cervical cancer and cervical lesion triage can be achieved respectively; or for the purpose of prognosis and treatment of cervical cancer.

4. An application technology for accurate early screening and diagnosis of cervical cancer and cervical lesion triage, characterized in that: Using the concepts, ideas and methods described in 1, 2 and 3 above, any one or a combination of HPV-DAN16, HPV-DAN18, HPV-DAN31, HPV-DAN33, HPV-DAN35, HPV-DAN45, HPV-DAN51, HPV-DAN52, HPV-DAN56 and HPV-DAN58 can be studied together with other mRNA expression, protein expression, DNA methylation level and other markers to achieve the application purpose of 3 above. Or achieve the purpose of prognosis and treatment of cervical cancer.

5. An application technology for accurate early screening and diagnosis of cervical cancer and cervical lesion triage, characterized in that: Among normal cervical patients / cervical inflammation patients, CIN 1, CIN 2, CIN 2, cervical cancer, pathological / histological / clinical staging of cervical cancer, LSIL and HSIL patients, any combination of two or more was used to compare with HPV-DAN16, HPV-DAN18, HPV - DAN31, HPV-DAN33, HPV-DAN35, HPV-DAN45, HPV-DAN51, HPV-DAN52, HPV - The data range of DNA expression level / expression amount / content of any one or more combinations of DAN56 and HPV-DAN58 is compared in a table; other similar comparison methods are also included. To achieve the application purpose of reference, screening, prediction, auxiliary diagnosis and diagnosis of cervical cancer and cervical lesion triage. For the purpose of cervical cancer prognosis and treatment.

6. An application technology for accurate early screening and diagnosis of cervical cancer and cervical lesion triage, characterized in that: Utilize the concepts, ideas and methods described in 1, 2, 3 or 1, 2, 3, 4 above to construct two or more function equations / mathematical models / discriminant functions / similar recognition functions / similar recognition equations / similar recognition methods, and combine their different recognition methods for comprehensive evaluation, so as to improve the accuracy of screening, prediction, auxiliary diagnosis and diagnosis of cervical cancer and triage of cervical lesions.

Images