CN107574243B - Molecular marker, reference gene and application thereof, detection kit and construction method of detection model - Google Patents
Molecular marker, reference gene and application thereof, detection kit and construction method of detection model Download PDFInfo
- Publication number
- CN107574243B CN107574243B CN201610509983.0A CN201610509983A CN107574243B CN 107574243 B CN107574243 B CN 107574243B CN 201610509983 A CN201610509983 A CN 201610509983A CN 107574243 B CN107574243 B CN 107574243B
- Authority
- CN
- China
- Prior art keywords
- breast cancer
- value
- prediction
- gene
- models
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- C—CHEMISTRY; METALLURGY
- C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
- C12N—MICROORGANISMS OR ENZYMES; COMPOSITIONS THEREOF; PROPAGATING, PRESERVING, OR MAINTAINING MICROORGANISMS; MUTATION OR GENETIC ENGINEERING; CULTURE MEDIA
- C12N15/00—Mutation or genetic engineering; DNA or RNA concerning genetic engineering, vectors, e.g. plasmids, or their isolation, preparation or purification; Use of hosts therefor
- C12N15/09—Recombinant DNA-technology
- C12N15/11—DNA or RNA fragments; Modified forms thereof; Non-coding nucleic acids having a biological activity
-
- C—CHEMISTRY; METALLURGY
- C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
- C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
- C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
- C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Organic Chemistry (AREA)
- Genetics & Genomics (AREA)
- Wood Science & Technology (AREA)
- Zoology (AREA)
- General Engineering & Computer Science (AREA)
- Biotechnology (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biochemistry (AREA)
- Physics & Mathematics (AREA)
- Proteomics, Peptides & Aminoacids (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Microbiology (AREA)
- Analytical Chemistry (AREA)
- Immunology (AREA)
- Plant Pathology (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention relates to the technical field of biology, in particular to a primer composition, application thereof, a detection kit and a construction method of a detection model. By adopting follow-up information as a comparison method, the kit provided by the invention has 70% of prediction accuracy for the 3-10 years of relapse or death risk of ER and PR positive breast cancer patients after operation, wherein the prediction accuracy of a low risk group and a high risk group is 81.1% and 54.4% respectively. The prediction accuracy of the corresponding FFPE pathological detection result is 71.9 percent and 56.8 percent respectively. The risk prediction model matched with the kit only needs the Ct value of the molecular marker, the age of a patient, the pT stage and the LN number, does not depend on other clinical pathological information, has better evaluation performance on the breast cancer prognosis than a simple pathological prediction result, can reduce improper treatment caused by pathological prediction errors to a certain extent, and further improves the technical method for breast cancer prognosis evaluation.
Description
Technical Field
The invention relates to the technical field of biology, in particular to a molecular marker, an internal reference gene, application thereof, a detection kit and a construction method of a detection model.
Background
Breast cancer is one of the main reasons threatening the life and health of women all over the world, and the American cancer society published American cancer statistics in 2013 shows that the incidence of breast cancer is the first of female cancers and the mortality is the second. The latest american national cancer center data showed 232,340 new breast cancers and 39,620 deaths in us women in 2013. In the united states, on average, one woman has breast cancer in every 8 women. Although China belongs to a country with low incidence of breast cancer, the annual incidence and mortality rate are obviously increased. Of the 130 new breast cancer patients diagnosed worldwide each year, about 15% are from china. Statistical data of the Chinese breast cancer network show that the new breast cancer reaches 3% -4% per year in China, which exceeds the world level by 1% -2%, and the incidence rate is the first of the female susceptibility to tumors. The development of techniques for prevention, diagnosis, prognosis and individual treatment of breast cancer is urgently needed.
Breast cancer is a highly heterogeneous group of tumors with numerous prognostic influencing factors, and breast cancer patients with the same clinical staging, histological grade and hormone receptor expression receive the same treatment regimen, and the prognosis may also vary. How to accurately judge the prognosis of breast cancer patients and formulate corresponding individualized treatment schemes, and avoid the harm and burden of patients caused by over-treatment and improper treatment is a problem which needs to be solved urgently in clinic.
With the rapid development of molecular biology technology, it is possible to discover and detect genes related to breast cancer prognosis by using molecular biology methods such as Polymerase Chain Reaction (PCR), probe hybridization, and gene chips. In 2002, Van't Veer et al screen 117 cases of breast cancer by a DNA chip technology and find 70 genes related to the prognosis of the breast cancer; in 2004, the American scientist verified 675 breast cancer samples by RT-PCR to obtain 21 genes related to prognosis, and Genomic Health developed the product Oncotype related to breast cancer prognosis based on the research
Oncotype
Is the only breast cancer prognosis detection product which is commonly recommended by NCCN guideline, ASCO clinical guideline and St Gallen clinical consensus 3 most authoritative clinical guidelines in the world. This is achieved byIn addition, Yasuto Naoi et al found 95 genes related to prognosis by studying cancer tissue samples of ER-positive and lymph node-negative breast cancer patients in Japanese population using DNA chip technology. Nielsen research group found that 50 gene combinations could provide more prognostic information for breast cancer than clinical factors and immunohistochemical staining, and that FFPE samples were used instead of fresh samples or quick frozen samples for detection, expanding the range of detectable samples. Oncotype as a product for detecting prognosis of breast cancer in 2002-2013 years
Mammaprint、ProsIgnaTM、MapQuant DxTMAnd obtaining FDA and CE certification in sequence. However, these products are developed based on the European and American population at present, and not only are the products expensive after entering China, but also whether the genes and the detection models thereof are suitable for the Chinese population is yet to be verified. Therefore, the development of economic and effective Chinese breast cancer prognosis detection technology has important significance.
Disclosure of Invention
In view of the above, the invention provides a molecular marker and an application thereof, a detection kit and a construction method of a detection model. The kit is better than the clinical pathological evaluation result in the breast cancer prognosis evaluation detection performance, can reduce the over-treatment and the improper treatment caused by the pathological diagnosis error to a certain extent, meets the requirement of individualized accurate treatment of breast cancer patients, and further perfects the technical method in the aspect of domestic breast cancer prognosis prediction.
In order to achieve the above object, the present invention provides the following technical solutions:
the invention provides genomic compositions comprising the molecular markers MAPT and/or MS4a 1.
The invention provides a gene composition, which consists of molecular markers BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CD68, BAG1, MAPT and MS4A 1.
In some embodiments of the invention, the genomic composition further comprises the reference genes ACTB, GAPDH, GUSB, NUP214, VCAN.
The invention also provides application of the genome composition in preparing a detection device for predicting 3-10-year recurrence and/or death risk after breast cancer operation.
In some embodiments of the invention, the 3-10 year prognosis of breast cancer and/or the assessment of risk of mortality for use is in particular: obtaining total RNA of a sample to be detected, obtaining cDNA through reverse transcription, obtaining Ct values of the molecular marker and the reference gene by a fluorescence quantitative PCR method, averaging the Ct values of the reference gene to obtain an average Ct value (Ct ') of a reference gene combination, then subtracting the Ct values of the molecular marker and the Ct' value of the reference gene combination respectively for normalization to obtain a delta Ct, and analyzing the delta Ct value and the age, pT value and LN value of a detected person through a breast cancer postoperative 3-10-year recurrence or death risk prediction model constructed by a random forest algorithm to obtain a result. Wherein, pT value is pathological stage, LN value is lymph node metastasis number. And comparing the value obtained by the analysis with a threshold value to obtain a result, wherein the threshold value is 5. The numerical value obtained by analysis is more than or equal to 5, and the numerical value obtained by analysis is less than 5, and the numerical value is the prognosis difference.
In some embodiments of the present invention, the method for constructing the model for detecting 3-10 years of recurrence or death risk assessment after breast cancer prognosis in the application is as follows: the method comprises the steps of constructing a mathematical matrix by using a delta Ct value of a molecular marker of a sample to be detected, the age, the pT value and the LN value of a detected person, randomly selecting 1/2 as a training set, 1/2 as a verification set, establishing a prediction model comprising 10000 decision trees by using a random forest algorithm, randomly sampling for more than or equal to 1000 times, establishing more than or equal to 1000 prediction models, selecting more than or equal to 39 optimal models with the highest follow-up information consistency rate from the more than or equal to 1000 prediction models as sub models of a final model, and adopting the median of more than or equal to 39 sub models as a final prognosis risk prediction value.
The random forest is composed of a plurality of decision trees, and the decision trees are constructed by adopting an attribute and sample dual-random method, so that the random forest is also called as a random decision tree. In random forest, there is no correlation between the decision trees. When test data enters a random forest, each decision tree is used for classification, and finally the class with the highest classification result in all decision trees is taken as a final result, namely a decision tree 'voting' result. In the invention, optimization is carried out on the basis of the traditional random forest algorithm, samples are randomly sampled for 1000 times, 1000 models are established, 39 optimal models with higher accuracy and specificity values are selected from the 1000 models as the sub-models of the final model, and the median of the 39 sub-models is used as the final prediction result.
Age, PT stage, LN transfer number of untreated early, middle term ER or PR positive breast cancer patients at first diagnosis, and Ct values of 14 molecular markers BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CD68, BAG1, MAPT, MS4A1 and 5 housekeeping genes ACTB, GAPDH, GUSB, NUP214 and VCAN are input into 39 prediction models for analysis to obtain prediction analysis results, and a 3-10-year relapse or death risk value is obtained, and the prognosis is good or poor according to a risk threshold (the threshold is 5).
In some embodiments of the invention, the test sample is an FFPE sample from an untreated early or mid-ER or PR positive breast cancer patient at first visit.
The invention also provides a primer group for amplifying the gene composition, and the sequence is shown as SEQ ID No. 1-SEQ ID No. 28.
The invention also provides a probe set for amplifying the gene composition, and the sequence is shown as SEQ ID No. 29-SEQ ID No. 42.
The invention also provides a primer group for amplifying the reference gene of the gene composition, which is shown as SEQ ID No. 43-SEQ ID No. 47.
The invention also provides a probe set for amplifying the internal reference gene of the gene composition, which is shown as SEQ ID No. 48-SEQ ID No. 52.
The invention also provides a detection kit for predicting the recurrence and/or death risk of breast cancer 3-10 years after operation, which comprises the primer group and/or the probe group and reagents commonly used in the kit.
The invention also provides a construction method of a model for evaluating and detecting 3-10 years of breast cancer recurrence or death risk, a mathematical matrix is constructed by the delta Ct value of a sample molecular marker to be detected and the age, pT value and LN value of a detected person, 1/2 is randomly selected as a training set, 1/2 is selected as a verification set, a prediction model comprising 10000 decision trees is established through a random forest algorithm, random sampling is carried out for more than or equal to 1000 times, more than or equal to 1000 prediction models are established, more than or equal to 39 optimal models with the highest follow-up information consistency rate are selected from the more than or equal to 1000 prediction models as a sub model of a final model, and the median of more than or equal to 39 sub models is adopted as a final prognosis risk prediction value.
The invention also provides an evaluation and detection method for 3-10 years of recurrence or death risk of breast cancer, which comprises the steps of obtaining total RNA of a sample to be detected, obtaining cDNA through reverse transcription, obtaining Ct values of the molecular marker and the reference gene by adopting a fluorescence quantitative PCR method, averaging the Ct values of the reference gene to obtain an average Ct value (Ct ') of the reference gene combination, subtracting the Ct values of the molecular marker and the Ct' value of the reference gene combination respectively for normalization to obtain delta Ct, analyzing the delta Ct value and the age, pT value and LN value of a detected person through a 3-10 years of recurrence or death risk prediction model constructed by a random forest algorithm to obtain a result, namely obtaining the 3-10 years of recurrence or death risk value, and predicting the prognosis good or poor according to a risk threshold (the risk threshold is 5).
The sample to be detected in the invention is an FFPE sample of an untreated early-stage or middle-stage ER or PR positive breast cancer initial diagnosis patient.
The technical scheme for solving the problems comprises the following steps: (1) through research on documents and databases, 192 candidate genes (not limited to those related to breast cancer prognosis and containing reference genes) related to breast cancer are selected, and TLDA gene expression detection chips (Applied Biosystems) are customized; (2) collecting complete demographic data, clinical data and follow-up data (recurrence transfer time and survival time) by the system, selecting untreated FFPE samples of early and middle ER or PR positive breast cancer patients for initial diagnosis, adopting a customized TLDA chip to detect 192 genes, and screening related molecular markers of internal reference genes and breast cancer prognosis; (3) verifying the candidate molecular markers and the reference genes obtained by screening in independent samples, constructing a prediction model of the relapse or death risk of the patient in 3-10 years after the operation by adopting a random forest algorithm, and evaluating the consistency rate of the prediction model and the follow-up result; (4) further validation with independent clinical samples: FFPE samples of early and medium-term breast cancer patients with ER or PR positive in 19 cases of known clinical follow-up data evaluate the consistency rate of the detection result and the follow-up result.
The invention provides a prognostic evaluation gene detection system for relapse or death of untreated early and middle stage ER or PR positive breast cancer patients 3-10 years after operation. The expression Ct values of 14 molecular markers and 5 reference genes related to breast cancer prognosis are detected by extracting total RNA in a Formalin-Fixed paraffin-Embedded (FFPE) breast cancer tissue sample and adopting a PCR method after reverse transcription of common primers. And (3) introducing the Ct value, the age, the pT value and the LN number of the detected person into a prediction model of the postoperative 3-10 years of relapse or death risk of the ER or PR positive early-stage and middle-stage breast cancer patient constructed by a random forest algorithm to carry out good prognosis or poor prognosis judgment. Compared with follow-up information, the accuracy of the system reaches 70%, and other clinical and pathological information is not required to be relied on except the age, pT stage and LN number of patients.
The kit provided by the invention has the advantages that the prediction accuracy of the kit on the primary breast cancer patient with low recurrence or death risk value in 3-10 years is 81.1%, the pathology prediction detection accuracy is 71.9%, the prediction accuracy sensitivity of the kit on the primary breast cancer patient with high recurrence or death risk value in 3-10 years is 54.4%, and the prediction accuracy is close to the corresponding pathology prediction detection accuracy of 56.8%. Compared with clinical follow-up information, the kit has the consistency rate of 70 percent, and does not need to depend on other clinical pathological information except the age, pT stage and LN number of patients.
The detection system and the kit are superior to clinical pathological prediction results in breast cancer prognosis evaluation detection performance, can reduce over-treatment and improper treatment caused by pathological diagnosis errors to a certain extent, meet the requirement of individualized accurate treatment of breast cancer patients, and further improve the technical method in the aspect of domestic breast cancer prognosis prediction.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.
FIG. 1 shows the results of analysis of the correlation between reference genes and test genes;
FIG. 2 shows the establishment of a model for assessing the risk of recurrence or death of 3-10 years after breast cancer prognosis.
Detailed Description
The invention discloses a molecular marker, an internal reference gene and application thereof, a detection kit and a construction method of a detection model, and a person skilled in the art can realize the detection by properly improving process parameters by referring to the content. It is expressly intended that all such similar substitutes and modifications which would be obvious to one skilled in the art are deemed to be included in the invention. While the methods and applications of this invention have been described in terms of preferred embodiments, it will be apparent to those of ordinary skill in the art that variations and modifications in the methods and applications described herein, as well as other suitable variations and combinations, may be made to implement and use the techniques of this invention without departing from the spirit and scope of the invention.
The technical scheme for solving the problems comprises the following steps: (1) through research on documents and databases, 192 candidate genes (not limited to those related to breast cancer prognosis and containing reference genes) related to breast cancer are selected, and TLDA gene expression detection chips (Applied Biosystems) are customized; (2) collecting complete demographic data, clinical data and follow-up data (recurrence transfer time and survival time) by the system, selecting untreated FFPE samples of early and middle ER or PR positive breast cancer patients for initial diagnosis, adopting a customized TLDA chip to detect 192 genes, and screening related molecular markers of internal reference genes and breast cancer prognosis; (3) verifying the candidate molecular markers and the reference genes obtained by screening in independent samples, constructing a prediction model of the relapse or death risk of the patient in 3-10 years after the operation by adopting a random forest algorithm, and evaluating the consistency rate of the prediction model and the follow-up result; (4) further validation with independent clinical samples: FFPE samples of early and medium-term breast cancer patients with ER or PR positive in 19 cases of known clinical follow-up data evaluate the consistency rate of the detection result and the follow-up result.
1. Selection of study samples
(1) Patients with untreated early and middle breast cancer at first visit;
(2) untreated early, mid-ER or (and) PR positive breast cancer patients;
(3) LN with or (and) without transfer, and LN transfer number;
(4) accurate and detailed follow-up information is available;
a total of 339 standard samples were used for the study.
Total RNA extraction of FFPE sample
The total RNA of the FFPE sample is extracted by using a High Pure FFPET RNA Isolation Kit (Roche) at a concentration of 25ng-400 ng/. mu.L, an OD260/280 of 1.8-2.0 and an OD260/230 of 1.5-2.0.
TLDA (Applied Biosystems) chip assay.
The following experiments were performed on good and poor prognosis samples using 26 of known clinical follow-up information.
(1) Carrying out reverse transcription reaction on the total RNA to obtain a cDNA sample;
(2) carrying out TLDA chip detection on the cDNA product;
(3) and analyzing and processing data to obtain candidate molecular markers and reference genes.
4. Real-time quantitative RT-PCR (qRT-PCR) method
Verification of candidate molecular markers was performed using 289 samples of known clinical follow-up information.
(1) Carrying out reverse transcription reaction on the total RNA to obtain a cDNA sample;
(2) carrying out RT-PCR detection on the cDNA product;
(3) and (5) analyzing and processing data.
5. Method for preparing diagnostic kit
Determining internal reference genes and gene expression differences of 26 breast cancer prognosis good samples and 26 breast cancer prognosis poor samples by customizing a TLDA chip detection method, and screening the internal reference genes and the differential expression genes. And (3) verifying the candidate molecular marker in a large sample size by reverse transcription fluorescence quantitative PCR. The last 14 genes and 5 internal reference genes selected to be related to the prognosis of breast cancer constitute a diagnostic kit (BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CD68, BAG1, MAPT, MS4A 1; ACTB, GAPDH, GUSB, NUP214, VCAN). The diagnostic kit comprises primers and probes of the genes, and qRT-PCR other conventional reagents. The kit also comprises a prediction model, wherein the expression levels of BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CD68, BAG1, MAPT and MS4A1 are detected by taking the mean value of ACTB, GAPDH, GUSB, NUP214 and VCAN as a reference gene, and the prognosis recurrence or death risk of the breast cancer patient in 3-10 years after surgery is comprehensively evaluated by combining clinical information such as age, PT, LN and the like, so that the good prognosis and the poor prognosis are judged.
6. Establishment of risk assessment detection model
(1) Establishing a model for predicting the recurrence or death risk of the breast cancer after 3-10 years of operation.
And (3) evaluating the postoperative recurrence or death risk value of the detection sample by adopting a random forest algorithm in a machine learning method, and establishing a breast cancer prognosis evaluation gene detection model. The random forest is composed of a plurality of decision trees, and the decision trees are constructed by adopting an attribute and sample dual-random method, so that the random forest is also called as a random decision tree. In random forest, there is no correlation between the decision trees. When test data enters a random forest, each decision tree is used for classification, and finally the class with the highest classification result in all decision trees is taken as a final result, namely a decision tree 'voting' result. In the invention, optimization is carried out on the basis of the traditional random forest algorithm, samples are randomly sampled for 1000 times, 1000 models are built, 39 optimal models with the highest accuracy are selected from the 1000 models to serve as sub models of a final model, and the median of the 39 sub models is used as a final prediction result.
The following is a further description of the invention:
in the first stage of research, 192 candidate genes related to the breast cancer are detected by adopting a TLDA detection technology, the gene expression difference of 26 cases of good samples of the breast cancer prognosis and 26 cases of poor samples of the breast cancer prognosis is detected, and differential expression genes are screened out. Different expression levels of the genes at 2-ΔCtIn the expression, "Δ Ct" represents a Ct sample-Ct reference, and the relative expression level was calculated by normalizing the selected reference gene as a reference. Wherein the screening process of the reference gene comprises the following steps: screening candidate reference genes by using four genes, i.e., genorm, bestkeeper, norm finder and delta Ct, based on a stability algorithm and considering the biological function of a gene with small fluctuation and the relation between the gene and a tumor; calculating the correlation between the Ct mean of all candidate reference gene combinations and the Ct mean of 192 genes, wherein the combination with the highest correlation is the reference gene and comprises the following steps: ACTB, GAPDH, GUSB, NUP214, VCAN. Candidate gene screening criteria: (1) the fold difference of two groups of good prognosis and poor prognosis reaches 2 times or less than 0.5, and the proportion of cases with Ct less than 35 reaches 50%; (2) the fold difference of two groups with good prognosis and poor prognosis of the lymph node-free metastasis group is more than 2 times, and the statistical difference is<0.05; (3) the two groups of fold differences are not obvious in overall analysis, but reported in documents related to breast cancer prognosis, and the proportion of cases with Ct < 35 reaches 90%. Genes that meet the above screening criteria include: BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CD68, BAG1, MAPT, MS4A1
Firstly, the breast cancer prognosis evaluation gene with Chinese female characteristics provided by the invention: at present, foreign similar products are developed based on European and American people, people of different ethnic groups have different gene expressions, 19 genes are screened out in the invention, wherein MAPT and MS4A1 are screened out based on female breast cancer patients in China and are related to relapse or death evaluation after 3-10 years of operation, and although reports of the genes are related to breast cancer, direct reports related to breast cancer prognosis are not found. Secondly, the invention establishes a new reference gene combination which is different from other inventions and products, and the gene combination is slightly influenced by the RNA quality in the FFPE sample, so that the detection result of the molecular marker is more reliable. Thirdly, comprehensive analysis is carried out by adopting a prediction model of a random forest algorithm, and the model carries out the risk prediction of relapse or death of the early-stage and middle-stage ER + or PR + breast cancer patients after the operation for 3-10 years.
In conclusion, the present invention provides a prognostic evaluation gene test system for ER or PR positive stage I and II breast cancer patients who relapse or die 3-10 years after surgery without treatment. Compared with follow-up information, the accuracy of the system reaches 70%, and other clinical and pathological information is not required to be relied on except the age, pT stage and LN number of patients.
The molecular marker, the reference gene and the application thereof, the detection kit and the raw materials and reagents used in the construction method of the detection model provided by the invention are all available in the market.
The invention is further illustrated by the following examples:
EXAMPLE 1 Collection of samples, working up of sample data
The FFPE sample of an untreated primary breast cancer patient is adopted, complete clinical follow-up data is collected systematically, and through the arrangement of the sample data, 341 samples meeting the following standards are selected by the inventor as TLDA (Taqman Low sensitivity Array, TLDA) chip detection and a series of subsequent qRT-PCR verification experimental samples:
(1) patients with untreated early and middle breast cancer at first visit;
(2) untreated early, mid-ER or (and) PR positive breast cancer patients;
(3) LN with or (and) without transfer, and LN transfer number;
(4) accurate and detailed follow-up information is provided.
Example 2TLDA chip screening for molecular markers and reference genes
TLDA chip detection is carried out on 26 breast cancer prognosis good samples and 26 breast cancer prognosis poor samples which meet the conditions, and relevant results are obtained. The method comprises the following specific steps:
(1) extraction of RNA from FFPE samples: taking 4 slices of 20 mu m slices or 8 slices of 10 mu m slices of each sample, extracting RNA according to the instruction of High Pure FFPET RNA Isolation Kit (Roche), and performing downstream reverse transcription experiment after quantitative quality control of the extracted RNA by NanoDrop-2000.
(2) The total RNA is subjected to reverse transcription reaction to obtain a cDNA sample: taking 1 μ g of total RNA according to
VILOTMMaster Mix kit (Invitrogen) instructions for reverse transcription.
(3) TLDA detection of cDNA samples: the above cDNA product and
after the Universal PCR Master Mix was mixed well, the detection experiment was performed on an ABI 7900 fluorescent quantitative PCR instrument according to the TLDA standard procedure.
(4) Data analysis and processing:
in the first stage of research, 192 candidate genes related to breast cancer are detected by adopting TLDA detection technology, the gene expression difference of 26 samples with good breast cancer prognosis and 26 samples with poor breast cancer prognosis is detected, and differential expression genes are screened out. Different expression levels of the genes at 2-ΔCtIn the expression, "Δ Ct" represents a Ct sample-Ct reference, and the relative expression level was calculated by normalizing the selected reference gene as a reference. Wherein the screening process of the reference gene comprises the following steps: screening candidate reference genes by using four genes, i.e., genorm, bestkeeper, norm finder and delta Ct, based on a stability algorithm and considering the biological function of a gene with small fluctuation and the relation between the gene and a tumor; calculating the correlation between the Ct mean of all candidate reference gene combinations and the Ct mean of 192 genes, wherein the combination with the highest correlation is the reference gene and comprises the following steps: ACTB, GAPDH, GUSB, NUP214, VCAN. Candidate gene screening criteria: (1) the fold difference of two groups of good prognosis and poor prognosis reaches 2 times or less than 0.5, and the proportion of cases with Ct less than 35 reaches 50%; (2) the fold difference of two groups with good prognosis and poor prognosis of the lymph node-free metastasis group is more than 2 times, and the statistical difference is<0.05; (3) the two groups of fold differences are not obvious in overall analysis, but reported in documents related to breast cancer prognosis, and the proportion of cases with Ct < 35 reaches 90%. Genes that meet the above criteria include: BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CThe functions of 14 genes above D68, BAG1, MAPT and MS4A1 are shown in Table 1 below.
TABLE 1 analysis of Gene function
| Serial number | Name of Gene | Functional relevance |
| 1. | BCL2 | Estrogen related compounds |
| 2. | PGR | Estrogen related compounds |
| 3. | SCUBE2 | Estrogen related compounds |
| 4. | ESR1 | Estrogen related compounds |
| 5. | MKi67 | Proliferation related |
| 6. | CCNB1 | Proliferation related |
| 7. | MYBL2 | Proliferation related |
| 8. | GRB7 | Her-2 correlation |
| 9. | ERBB2 | Her-2 correlation |
| 10. | MMP11 | Invasion correlation |
| 11. | CD68 | Differentiation group 68 |
| 12. | BAG1 | BCL2 binding to anti-apoptotic Gene 1 |
| 13. | MAPT | Microtubule-associated protein tau |
| 14. | MS4A1 | Transmembrane 4-domain subfamily a member 1 |
Example 3 Large sample size qRT-PCR validation of molecular markers
14 molecular markers and 5 reference genes screened by TLDA: BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CD68, BAG1, MAPT, MS4a 1; ACTB, GAPDH, GUSB, NUP214, VCAN. Single-tube qRT-PCR verification is carried out by adopting 289 FFPE samples which meet the sample collection requirements and have complete clinical follow-up information.
(1)289 FFPE sample RNA extraction: taking 4 slices of 20 mu m slices or 8 slices of 10 mu m slices of each sample, extracting RNA according to the instruction of High Pure FFPET RNA Isolation Kit (Roche), and performing downstream reverse transcription experiment after quantitative quality control of the extracted RNA by NanoDrop-2000.
(2)289 FFPE samples RNA were reverse transcribed into cDNA: taking 1 μ g of total RNA according to
VILOTMMaster Mix kit (Invitrogen) instructions for reverse transcription.
(3)289 FFPE sample cDNA products were qPCR tested: cDNA product, probe and primer of each sample,
After being mixed uniformly, the Universal Master Mix II is subjected to a detection experiment on an ABI 7900 fluorescent quantitative PCR instrument. The qPCR primer and probe sequences are shown in tables 2 to 5.
TABLE 2qRT-PCR primer sequences
TABLE 3qRT-PCR Probe sequences
TABLE 4 Housekeeping Gene qRT-PCR primer sequences
TABLE 5 Housekeeping Gene qRT-PCR Probe sequences
Example 4 prediction of 3-10 years after Breast cancer prognosis Risk prediction model establishment
And (3) evaluating the postoperative recurrence or death risk value of the detection sample by adopting a random forest algorithm in a machine learning method, and establishing a breast cancer prognosis evaluation gene detection model. The random forest is composed of a plurality of decision trees, and the decision trees are constructed by adopting an attribute and sample dual-random method, so that the random forest is also called as a random decision tree. In random forest, there is no correlation between the decision trees. When test data enters a random forest, each decision tree is used for classification, and finally the class with the highest classification result in all decision trees is taken as a final result, namely a decision tree 'voting' result. In the invention, optimization is carried out on the basis of the traditional random forest algorithm, samples are randomly sampled for 1000 times, 1000 models are built, 39 optimal models with the highest accuracy are selected from the 1000 models to serve as sub models of a final model, and the median of the 39 sub models is used as a final prediction result.
Example 5 further validation of independent clinical samples
FFPE samples from 19 ER or PR positive early and mid breast cancer patients with known clinical follow-up data: the PT stages are stages 1 and 2, with patients operating between 2004 and 2008, observed with follow-up visits of 2011 to 2015, with follow-up visits of 3-10 years or more.
Extracting the total RNA of the 19 FFPE samples by using a High Pure FFPET RNA Isolation Kit (Roche), carrying out reverse transcription reaction on the RNA after quality control is qualified to obtain a cDNA sample, carrying out qRT-PCR reaction on the cDNA product, and detecting internal reference genes ACTB, GAPDH, GUSB, NUP214 and VCAN, BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CD68, BAG1, MAPT and MS4A1 genes. The Ct value of the gene is introduced into a model for evaluating the recurrence or death risk of the breast cancer after 3-10 years of operation, which is constructed by a random forest method, to obtain a recurrence or death risk value of 3-10 years, and the prognosis is good or poor according to the prediction of a risk threshold. The consistency rate of the prediction analysis result and the known follow-up information is 73.6%, and the specific results are shown in table 6:
TABLE 619 cases of breast cancer patients with prognostic evaluation results
TABLE 7
Example 6
Currently, the decision of clinical treatment and treatment scheme of breast cancer ultimately depends on the result of pathological examination, and the result of pathological examination is the most important objective basis for judging prognosis. The detection system adopts FFPE samples of 289 cases of breast cancer patients with known clinical follow-up data collected by tumor hospitals of Tianjin medical university and tumor hospitals in Henan province to respectively detect 5 reference genes and 14 molecular markers.
The results are shown in Table 8.
TABLE 8
The kit provided by the invention has the accuracy of 81.1% for the primary breast cancer patient with low recurrence or death risk value in 3-10 years, the accuracy of pathological detection is 71.8%, the accuracy sensitivity of the kit to the primary breast cancer patient with high recurrence or death risk value in 3-10 years is 54.4%, and the accuracy is close to the accuracy of corresponding pathological detection by 56.8%. Compared with clinical follow-up information, the kit has the consistency rate of 70 percent, and does not need to depend on other clinical pathological information except the age, pT stage and LN number of patients.
The detection system and the kit are better than clinical pathological diagnosis results in breast cancer prognosis evaluation detection performance, can reduce over-treatment and improper treatment caused by pathological diagnosis errors to a certain extent, meet the requirement of individualized and accurate treatment of breast cancer patients, and further improve the technical method in the aspect of domestic breast cancer prognosis prediction.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (6)
1. The application of the genome composition in preparing a detection device for predicting the recurrence and/or death risk of 3-10 years after breast cancer operation;
the genome composition comprises 14 molecular markers and 5 reference genes; the molecular markers are BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CD68, BAG1, MAPT and MS4A 1; the reference genes are ACTB, GAPDH, GUSB, NUP214 and VCAN;
the 3-10 years after breast cancer operation relapse and/or death risk prediction specifically comprises the following steps: obtaining total RNA of a sample to be detected, obtaining cDNA through reverse transcription, obtaining Ct values of the molecular marker and the reference gene by adopting a fluorescence quantitative PCR method, averaging the Ct values of the reference gene to obtain an average Ct value (Ct ') of a reference gene combination, then subtracting the Ct values of the molecular marker and the Ct' value of the reference gene combination respectively for normalization to obtain a delta Ct, and analyzing the delta Ct value and the age, pT value and LN value of a detected person through a breast cancer postoperative 3-10-year recurrence or death risk prediction model constructed by a random forest algorithm to obtain a result;
the construction method of the model for predicting the 3-10-year relapse or death risk of the breast cancer after the operation comprises the following steps: the method comprises the steps of constructing a mathematical matrix by using a delta Ct value of a molecular marker of a sample to be detected, the age, the pT value and the LN value of a detected person, randomly selecting 1/2 as a training set, 1/2 as a verification set, establishing a prediction model comprising 10000 decision trees by using a random forest algorithm, randomly sampling for more than or equal to 1000 times, establishing more than or equal to 1000 prediction models, selecting more than or equal to 39 optimal models with the highest follow-up information consistency rate from the more than or equal to 1000 prediction models as sub models of a final model, and adopting the median of the more than or equal to 39 sub models as a final prognosis risk prediction value.
2. The use of claim 1, wherein the test sample is an untreated FFPE sample from an early or mid ER or PR positive breast cancer patient at first visit.
3. A primer set and a probe set, characterized in that,
the primer group for amplifying the gene composition has a sequence shown as SEQ ID No. 1-SEQ ID No. 28;
the probe group for amplifying the gene composition has a sequence shown as SEQ ID No. 29-SEQ ID No. 42;
the genomic compositions include BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CD68, BAG1, MAPT, MS4a 1;
the primer group for amplifying the reference gene of the gene composition has a sequence shown as SEQ ID No. 43-SEQ ID No. 47;
the probe group of the reference gene for amplifying the gene composition has a sequence shown as SEQ ID No. 48-SEQ ID No. 52;
the reference genes comprise ACTB, GAPDH, GUSB, NUP214 and VCAN.
4. A test kit for the prediction of the risk of recurrence and/or death of 3-10 years after breast cancer surgery, comprising the primer set and probe set of claim 3 and reagents commonly used in the kit.
5. A method for constructing a model for predicting 3-10-year relapse or death risk after breast cancer surgery is characterized in that a mathematical matrix is constructed by a delta Ct value of a sample molecular marker to be tested and an internal reference gene, the age, the pT value and the LN value of a tested person, 1/2 is randomly selected as a training set, 1/2 is selected as a verification set, a prediction model comprising 10000 decision trees is established through a random forest algorithm, random sampling is carried out for more than or equal to 1000 times, more than or equal to 1000 prediction models are established, more than or equal to 39 optimal models with the highest follow-up information consistency rate are selected from the more than or equal to 1000 prediction models as sub models of a final model, and the median of more than or equal to 39 sub models is adopted as a final prognosis risk prediction value;
the molecular markers are BCL2, PGR, SCUBE2, ESR1, MKi67, CCNB1, MYBL2, GRB7, ERBB2, MMP11, CD68, BAG1, MAPT and MS4A 1;
the reference genes are ACTB, GAPDH, GUSB, NUP214 and VCAN.
6. The method of claim 5, wherein the test sample is an untreated FFPE sample from an early or middle ER or PR positive breast cancer patient at first visit.
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610509983.0A CN107574243B (en) | 2016-06-30 | 2016-06-30 | Molecular marker, reference gene and application thereof, detection kit and construction method of detection model |
| PCT/CN2017/090740 WO2018001295A1 (en) | 2016-06-30 | 2017-06-29 | Molecular marker, reference gene, and application and test kit thereof, and method for constructing testing model |
| JP2018568674A JP2019527544A (en) | 2016-06-30 | 2017-06-29 | Molecular marker, reference gene, and application thereof, detection kit, and detection model construction method |
| SG11201811263WA SG11201811263WA (en) | 2016-06-30 | 2017-06-29 | Molecular marker, reference gene, and application and test kit thereof, and method for constructing testing model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610509983.0A CN107574243B (en) | 2016-06-30 | 2016-06-30 | Molecular marker, reference gene and application thereof, detection kit and construction method of detection model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107574243A CN107574243A (en) | 2018-01-12 |
| CN107574243B true CN107574243B (en) | 2021-06-29 |
Family
ID=60785942
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610509983.0A Active CN107574243B (en) | 2016-06-30 | 2016-06-30 | Molecular marker, reference gene and application thereof, detection kit and construction method of detection model |
Country Status (4)
| Country | Link |
|---|---|
| JP (1) | JP2019527544A (en) |
| CN (1) | CN107574243B (en) |
| SG (1) | SG11201811263WA (en) |
| WO (1) | WO2018001295A1 (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108841962B (en) * | 2018-08-01 | 2021-11-19 | 博奥生物集团有限公司 | Non-small cell lung cancer detection kit and application thereof |
| CN109801680B (en) * | 2018-12-03 | 2023-02-28 | 广州中医药大学(广州中医药研究院) | Tumor metastasis and recurrence prediction method and system based on TCGA database |
| CN110493235A (en) * | 2019-08-23 | 2019-11-22 | 四川长虹电器股份有限公司 | A kind of mobile terminal from malicious software synchronization detection method based on network flow characteristic |
| CN110923317A (en) * | 2019-11-27 | 2020-03-27 | 福建省立医院 | Method for breast cancer prognosis prediction and primer group thereof |
| CN110942808A (en) * | 2019-12-10 | 2020-03-31 | 山东大学 | Prognosis prediction method and prediction system based on gene big data |
| CN111440869A (en) * | 2020-03-16 | 2020-07-24 | 武汉百药联科科技有限公司 | DNA methylation marker for predicting primary breast cancer occurrence risk and screening method and application thereof |
| KR102565378B1 (en) * | 2020-03-23 | 2023-08-10 | 단국대학교 산학협력단 | Biomarker for predicting the status of breast cancer hormone receptors |
| CN111500724B (en) * | 2020-04-28 | 2023-11-21 | 启程医学科技(山东)有限公司 | Primer set and probe combination for simultaneous detection of breast cancer six genes |
| CN112725444A (en) * | 2020-12-30 | 2021-04-30 | 杭州联川基因诊断技术有限公司 | Primer, probe, kit and detection method for detecting PGR gene expression |
| CN112646864A (en) * | 2020-12-30 | 2021-04-13 | 杭州联川基因诊断技术有限公司 | Primer, probe, kit and detection method for detecting ESR1 gene expression |
| CN112927795B (en) * | 2021-02-23 | 2022-09-23 | 山东大学 | Breast cancer prediction system based on bagging algorithm |
| CN113846149B (en) * | 2021-09-28 | 2024-06-11 | 领航基因科技（杭州）有限公司 | Digital PCR real-time analysis method of micropore array chip |
| CN114373511B (en) * | 2022-03-15 | 2022-08-30 | 南方医科大学南方医院 | Intestinal cancer model based on 5hmC molecular marker detection and intestinal cancer model construction method |
| CN117153392A (en) * | 2023-08-25 | 2023-12-01 | 云基智能生物科技(广州)有限公司 | Marker for prognosis prediction of gastric cancer, assessment model and construction method thereof |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101173313A (en) * | 2006-09-19 | 2008-05-07 | 天津医科大学附属肿瘤医院 | Mammary cancer diversion and prognosis molecule parting gene group, gene chip producing and using method |
| CN101195825A (en) * | 2007-12-10 | 2008-06-11 | 上海华冠生物芯片有限公司 | Gene for prognosis of breast cancer and uses thereof |
| WO2008079269A2 (en) * | 2006-12-19 | 2008-07-03 | Genego, Inc. | Novel methods for functional analysis of high-throughput experimental data and gene groups identified therfrom |
| CN101921858A (en) * | 2010-08-23 | 2010-12-22 | 广州益善生物技术有限公司 | Liquid phase chip for detecting breast cancer prognosis-related gene mRNA expression level |
| CN101965190A (en) * | 2005-04-04 | 2011-02-02 | 维里德克斯有限责任公司 | Laser microdissection and microarray analysis of breast tumors reveal estrogen receptor related genes and pathways |
| CN104263815A (en) * | 2014-08-25 | 2015-01-07 | 复旦大学附属肿瘤医院 | A group of genes used for prognosis of hormone receptor-positive breast cancer and applications thereof |
| WO2015135035A2 (en) * | 2014-03-11 | 2015-09-17 | The Council Of The Queensland Institute Of Medical Research | Determining cancer agressiveness, prognosis and responsiveness to treatment |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| NZ548254A (en) * | 2003-12-23 | 2008-09-26 | Santaris Pharma As | Oligomeric compounds for the modulation of BCL-2 |
| EP2304631A1 (en) * | 2008-06-16 | 2011-04-06 | Sividon Diagnostics GmbH | Algorithms for outcome prediction in patients with node-positive chemotherapy-treated breast cancer |
| JP2016537010A (en) * | 2013-09-09 | 2016-12-01 | ブリティッシュ コロンビア キャンサー エージェンシー ブランチ | Method and kit for predicting prognosis, and method and kit for treating breast cancer using radiation therapy |
| CN104004844A (en) * | 2014-05-28 | 2014-08-27 | 杭州美中疾病基因研究院有限公司 | Kit for jointly detecting breast cancer 21 genes and preparation method of kit |
-
2016
- 2016-06-30 CN CN201610509983.0A patent/CN107574243B/en active Active
-
2017
- 2017-06-29 JP JP2018568674A patent/JP2019527544A/en active Pending
- 2017-06-29 WO PCT/CN2017/090740 patent/WO2018001295A1/en active Application Filing
- 2017-06-29 SG SG11201811263WA patent/SG11201811263WA/en unknown
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101965190A (en) * | 2005-04-04 | 2011-02-02 | 维里德克斯有限责任公司 | Laser microdissection and microarray analysis of breast tumors reveal estrogen receptor related genes and pathways |
| CN101173313A (en) * | 2006-09-19 | 2008-05-07 | 天津医科大学附属肿瘤医院 | Mammary cancer diversion and prognosis molecule parting gene group, gene chip producing and using method |
| WO2008079269A2 (en) * | 2006-12-19 | 2008-07-03 | Genego, Inc. | Novel methods for functional analysis of high-throughput experimental data and gene groups identified therfrom |
| CN101195825A (en) * | 2007-12-10 | 2008-06-11 | 上海华冠生物芯片有限公司 | Gene for prognosis of breast cancer and uses thereof |
| CN101921858A (en) * | 2010-08-23 | 2010-12-22 | 广州益善生物技术有限公司 | Liquid phase chip for detecting breast cancer prognosis-related gene mRNA expression level |
| WO2015135035A2 (en) * | 2014-03-11 | 2015-09-17 | The Council Of The Queensland Institute Of Medical Research | Determining cancer agressiveness, prognosis and responsiveness to treatment |
| CN104263815A (en) * | 2014-08-25 | 2015-01-07 | 复旦大学附属肿瘤医院 | A group of genes used for prognosis of hormone receptor-positive breast cancer and applications thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2019527544A (en) | 2019-10-03 |
| CN107574243A (en) | 2018-01-12 |
| WO2018001295A1 (en) | 2018-01-04 |
| SG11201811263WA (en) | 2019-01-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107574243B (en) | Molecular marker, reference gene and application thereof, detection kit and construction method of detection model | |
| US11220716B2 (en) | Methods for predicting the prognosis of breast cancer patient | |
| KR101672531B1 (en) | Genetic markers for prognosing or predicting early stage breast cancer and uses thereof | |
| US10718030B2 (en) | Methods for predicting effectiveness of chemotherapy for a breast cancer patient | |
| CN110423816B (en) | Breast cancer prognosis quantitative evaluation system and application | |
| WO2017223216A1 (en) | Compositions and methods for diagnosing lung cancers using gene expression profiles | |
| CN111172285A (en) | miRNA group for early diagnosis and/or prognosis monitoring of pancreatic cancer and application thereof | |
| US20190010558A1 (en) | Method for determining the risk of recurrence of an estrogen receptor-positive and her2-negative primary mammary carcinoma under an endocrine therapy | |
| KR102602133B1 (en) | Kit for diagnosing metastasis of cervical cancer | |
| CN113444803B (en) | Cervical cancer prognosis marker microorganism and application thereof in preparation of cervical cancer prognosis prediction diagnosis product | |
| CN114214420B (en) | Application of exosome miR-106B-3P, IL B and the like in lung cancer diagnosis | |
| CN106520957B (en) | The susceptible SNP site detection reagent of DHRS7 and its kit of preparation | |
| CN115472294A (en) | Model for predicting transformation speed of small cell transformation lung adenocarcinoma patient and construction method thereof | |
| WO2022225447A1 (en) | Risk assessment method of breast cancer recurrence or metastasis and kit thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1248775 Country of ref document: HK |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |