Construction method and system of breast cancer risk assessment model

Abstract

The invention belongs to the field of biomedicine, and particularly relates to a construction method and system of a breast cancer risk assessment model based on combination of MircoRNA and a traditional serum marker, and the method comprises the following steps: high-throughput miRNA sequencing; extracting miRNA (micro Ribonucleic Acid) from serum; extracting tissue total RNA (Ribonucleic Acid); miRNA reverse transcription is carried out; performing fluorescent quantitative PCR; and establishing a risk assessment model. Three risk models with breast cancer diagnosis are constructed according to the training set data, and the three models are further verified to have high sensitivity, specificity and accuracy on breast cancer diagnosis through verification set and collection data. Meanwhile, through comparison, the diagnosis efficiency of the (miRNAs + TM) panel III risk model is higher, and the potential application value is achieved.

Description

Technical Field

[0001] The present invention belongs to but is not limited to the field of biomedicine, and in particular relates to a method and system for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers. Background Art

[0002] Breast cancer is a complex and highly heterogeneous malignant tumor. The "2022 National Cancer Report" shows that breast cancer ranks first in the incidence of female cancer. Among them, the number of new cases of breast cancer in China is about 357,000, ranking fourth in the world. At present, the incidence and mortality of breast cancer in women rank first and fourth among female malignant tumors in my country, respectively. With the changes in factors such as lifestyle and social environment in my country, the incidence of breast cancer has shown an increasing trend year by year, posing a serious threat to the physical and mental health of women. Early diagnosis and real-time monitoring are important strategies for the effective prevention and treatment of breast cancer. The accuracy and sensitivity of diagnostic markers reflect the condition and prognosis of breast cancer. Therefore, finding and discovering efficient diagnostic methods is of great significance in the early diagnosis, targeted treatment and prognosis of breast cancer.

[0003] The main clinical diagnostic methods for breast cancer include: clinical physical examination, breast ultrasound, mammographic X-ray photography, magnetic resonance imaging, and pathological histological examination. Each method has its advantages, and flexible use can accurately diagnose breast cancer, but there are also some shortcomings. For example, ultrasound examination is difficult to identify a small mass, and the result is easily affected by the state of the examinee and the instrument, which is prone to misdiagnosis. Mammographic X-rays have poor fluoroscopic effects on gland-rich breasts and have certain radiation side effects. Pathological histological examinations are difficult to observe tissue structure due to the small amount of samples taken, and are more traumatic to patients, making them difficult to use for clinical surveys. Traditional serum tumor marker tests can quickly diagnose breast cancer, but their sensitivity and specificity are poor. For example, CA125 is overexpressed in the serum of not only breast cancer, but also uterine fibroids, ovarian cancer, and other cancers, and its expression level is affected by other lesions. For example, chronic pelvic inflammatory disease, ovarian cysts, and endometriosis all lead to elevated CA125. Therefore, the combination of multiple tumor markers or the combined use of multiple methods can make up for missed detection and misdiagnosis caused by uncertain factors or technical problems, and can improve the accuracy and sensitivity of diagnosis.

[0004] MicroRNA (miRNA) is a class of non-coding small RNAs consisting of approximately 22 nucleotides, which play a variety of important regulatory roles in cells. MiRNA can be transported in and out of different cells through exosomes, and regulate gene expression by recognizing homologous sequences and affecting transcription, translation or epigenetic processes. More and more studies have confirmed that miRNA is abnormally expressed in breast cancer patients and is involved in the occurrence and development of breast cancer. Ibrahim et al. reported that miR-132 and miR-212 are commonly overexpressed in drug-resistant breast cancer tumors and cells, and further studies have found that miRNA-132 / -212 can inhibit the expression of PTEN, activate the NF-κB pathway, and thus upregulate the expression level of the drug-resistant protein BCRP. Downregulation of miR-132 / -212 can increase the sensitivity of breast cancer cells to doxorubicin. MiR-21 and miR-210 are highly expressed in breast cancer tissues and are significantly correlated with lymph node metastasis and clinical stage. The diagnostic value and differentiation of the four serum microRNAs in breast cancer are significantly correlated. MiR-210 is considered to be associated with poor prognosis in breast cancer patients and is a potential prognostic indicator for breast cancer. The advantage of miRNA is that it is easy to detect using microarray, deep sequencing or real-time fluorescence quantitative PCR (qRT-PCR), and it has high tissue specificity and stability. Therefore, miRNA can be used as an ideal target for cancer diagnosis, treatment and prognosis assessment.

[0005] The use of multiple disease-specific miRNAs to construct disease diagnosis models can improve the accuracy of disease diagnosis and the effectiveness of prognosis. The nomogram model constructed by Xu Lin et al. based on miR-21 and miR-31 has a high predictive value for the diagnosis of preeclampsia. ROC analysis shows that its predictive efficacy, sensitivity and specificity are higher than those of peripheral blood soluble vascular endothelial growth factor receptor / placental growth factor. LUO et al. found that miR-140-3p and miR-3200-3p were significantly downregulated in hepatocellular carcinoma (HCC), and therefore constructed an HCC diagnosis model based on these specifically expressed miRNAs. The ROC analysis of the subjects found that the model had an area under the curve (AUC) of 0.951, a sensitivity of 90.1%, and a specificity of 87.8%, which has a high diagnostic value. The researchers combined machine and neural network analysis to construct an artificial neural network model containing miR-4648, miR-125b-1-3p and miR-3201 for early diagnosis of pancreatic cancer, which has a high diagnostic accuracy, and the AUC values ​​of the validation group and training group are 0.935 and 0.926 respectively. miRNA has significant accuracy and advantages in clinical diagnosis of diseases as a biomarker. The construction of a suitable diagnostic model based on miRNA has potential value for the pathogenesis of diseases, drug screening, personalized treatment, etc. Summary of the invention

[0006] In response to the above problems and the defects of existing breast cancer diagnosis technology, based on the lack of research on the use of miRNA in breast cancer diagnosis and its advantages as a biomarker, the present invention aims to discover miRNA that can accurately and specifically reflect breast cancer, and combine it with traditional serum tumor markers to construct a model that can be used for early breast cancer risk assessment, so as to make up for technical problems such as false detection and missed detection in breast cancer diagnosis.

[0007] The present invention is achieved by a method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers, the method comprising:

[0008] S1: High-throughput miRNA sequencing;

[0009] S2: serum miRNA extraction;

[0010] S3: total RNA extraction from tissues;

[0011] S4: miRNA reverse transcription;

[0012] S5: Fluorescence quantitative PCR;

[0013] S6: Establishment of risk assessment model.

[0014] Further, the S1 specifically includes:

[0015] The serum exosomes were separated by ultracentrifugation from healthy subjects and untreated breast cancer patients, and then high-throughput miRNA sequencing was performed using the Illumina HiSeq SE50 platform. Combined with literature review, miRNAs with significant differential expression were screened out, and through logistics regression and ROC analysis of breast cancer patients, miR-548ao-5p and miR-4804-3p were finally determined to be miRNAs with high diagnostic efficacy.

[0016] Further, the S2 specifically includes:

[0017] (1) Blood sample pretreatment: 0.5 mL of serum was collected and centrifuged at 3000 r / min for 5 min at 4°C to remove cell debris. The supernatant was collected and stored at -80°C.

[0018] (2) Take 200 μL of pretreated serum specimen, thaw on ice, add 5 times the volume of the specimen, i.e. 1 mL of lysis buffer, mix thoroughly by inverting or vortexing, and let stand at room temperature (15-25°C) for 5 min;

[0019] (3) Add external reference: 3.5 μL of miRNeasySerum / PlasmaSpike-InControl containing 1.6×108 copies / μL.

[0020] (4) Add chloroform in an equal volume to the serum sample, shake vigorously for 15 seconds, place at room temperature for 5 minutes, and centrifuge at 12,000 g for 15 minutes at 4°C. Transfer the upper aqueous phase to a new EP tube, add 1.5 times the volume of anhydrous ethanol, and mix slowly. Then transfer the entire solution to a 2 mL RneasyMinElute centrifugal column, centrifuge at room temperature at 12,000 g for 1 minute, and discard the filtrate;

[0021] (5) Add 700 μL of Buffer RWT to the centrifuge column, centrifuge at 12,000 g for 1 min at room temperature, and discard the filtrate;

[0022] (6) Add 500 μL of Buffer RPE to the centrifuge column, centrifuge at 12,000 g for 1 min at room temperature, and discard the filtrate;

[0023] (7) Add 500 μL of 80% ethanol to the centrifuge column, centrifuge at 12,000 g for 2 min at room temperature, discard the filtrate, and then transfer the adsorption column to another new collection tube;

[0024] (8) Centrifuge at the highest speed for 5 min and discard the collection tube;

[0025] (9) Place the centrifuge column in a new 1.5 mL EP tube, add 14 μL RNase-Free H2O, leave for 5 min, centrifuge at the highest speed for 1 min to elute the RNA, and store at -80°C.

[0026] Further, the S3 specifically includes:

[0027] (1) Cell lysis buffer: add 200 μL of β-mercaptoethanol to every 10 mL of TRK lysis buffer;

[0028] (2) Preparation of working solution Wash Buffer II: Add 48 mL of anhydrous ethanol to the mother solution Wash Buffer II to dilute it and mix thoroughly by inverting;

[0029] (3) 70% ethanol preparation: Use analytical grade (>99.7%) anhydrous ethanol and DEPC water in a ratio of 7:3 and mix evenly;

[0030] (4) Tissue processing: Place the tissue soaked in RNA protection solution on ice and use a shaker to accelerate thawing. Use sterile tweezers to squeeze out excess RNA protection solution, and gently absorb the liquid on the surface with absorbent paper; weigh 20-30 mg of tissue blocks, remove fat tissue and place in a 5 mL round-bottom EP tube, cut the tissue blocks into pieces with sterile scissors, add 1.5 mL of TRK lysis solution, and homogenize the tissue blocks using a handheld homogenizer on ice. Note that each homogenization run should be about 5 seconds, and run 4 to 5 times in total; after standing for 5 minutes, aspirate the supernatant into a 1.5 mL EP tube and centrifuge at 2000 r / min for 5 minutes at room temperature;

[0031] (5) Transfer the liquid to a homogenizer column, centrifuge at 14,000 g for 2 min at room temperature to remove insoluble impurities, and collect the filtrate;

[0032] (6) Add an equal volume of 70% ethanol and mix thoroughly and slowly. Transfer the mixture to the RNA binding column and place it in a collection tube. Centrifuge at room temperature at 10,000 g for 1 min and discard the filtrate.

[0033] (7) Repeat step 6 until all the mixed solution is bound to the RNA binding column;

[0034] (8) Add 500 μL RNA Wash Buffer I to the RNA binding column, centrifuge at 10,000 g for 1 min at room temperature, and discard the filtrate;

[0035] (9) Add 500 μL RNA Wash Buffer II to the RNA binding column, centrifuge at 10,000 g for 1 min at room temperature, and discard the filtrate;

[0036] (10) Repeat step 9 and wash again;

[0037] (11) Place the RNA binding column back into the collection tube and centrifuge again at 10,000 g for 2 min at room temperature to remove residual liquid;

[0038] (12) Place the RNA binding column in a new 1.5 mL EP tube free of RNase, add 30-70 μL RNase-Free H2O to the center of the column membrane, and let stand at room temperature for 2 min. Centrifuge at 14,000 g at 4°C for 2 min, and store at -80°C for a long term.

[0039] Further, the S4 specifically includes:

[0040] (1) Prepare the tailing reverse transcription reaction solution on ice. The reaction system is as follows:

[0041] 2×mRQ Buffer 5μL, RNA sample 0.25-8μg and 3.75μL, mRQ Enzyme 1.25μL, total volume 10μL;

[0042] (2) Reaction conditions: 37°C, 1 h; 85°C, 5 min;

[0043] (3) Add 90 μL RNase-Free H2O to dilute the obtained cDNA and store it at -20°C for a long time.

[0044] Further, in said S5, the primer sequences are as follows:

[0045] miR-548ao-5p: upstream primer: 5′-GAAGTAACTACGCTTTTTGCA-3′;

[0046] miR-4804-3p: upstream primer: 5′-CTTAACCTTGCCCTCGAAAAA-3′;

[0047] U6: Upstream primer: 5'-GGAACGATACAGAGAAGATTAGC-3'; downstream primer: 5'-TGGAACGCTTCACGAATTTGCG-3'.

[0048] Further, in S5, the standard procedure of two-step PCR amplification is as follows:

[0049] Pre-denaturation: 95°C, 30s; amplification: 95°C, 5s, 60°C, 30s, 40 cycles in total; melting curve: 95°C, 5s, 65°C, 1min, 95°C, 30s, 50°C, 30s.

[0050] Further, the S6 specifically includes:

[0051] The outcome variable of breast cancer patients in the training set was set as "1", and the variable of the healthy control group was set as "0". The screened miRNA molecules, namely miR-548ao-5p and miR-4804-3p, were introduced into logistic regression. The weight of each miRNA in the model was determined according to the regression coefficient of the multivariate logistics regression model, and a breast cancer diagnosis model based on miRNA was established, named miRNAs panel I. Subsequently, the screened traditional serum tumor markers, namely CA 125 and CA153, were introduced into another logistic regression to establish a breast cancer diagnosis model based on traditional serum tumor markers, named TMpanel II. Finally, the above miRNAs and serum markers were introduced into logistic regression to form a combined diagnosis model, named miRNAs+TM) panel III, thereby calculating the risk assessment value of each model individual.

[0052] The validation set and the combined data were imported into the three risk assessment models constructed to verify the accuracy and specificity of the models. ROC curve and DAC analysis were performed to compare the diagnostic efficacy of the three models.

[0053] Another object of the present invention is to provide a system for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers, the system comprising:

[0054] miRNA sequencing module, used for high-throughput miRNA sequencing;

[0055] The miRNA extraction module is connected to the miRNA sequencing module and is used for extracting serum miRNA;

[0056] An RNA extraction module, connected to the miRNA extraction module, for extracting total RNA from tissues;

[0057] A reverse transcription module, connected to the RNA extraction module, for reverse transcription of miRNA;

[0058] A fluorescence quantitative PCR module is connected to the reverse transcription module and is used for the fluorescence quantitative PCR module;

[0059] The model building module is connected with the fluorescence quantitative PCR module for establishing the risk assessment model.

[0060] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0061] First, the present invention evaluated for the first time the diagnostic value of serum miR-548ao-5p and miR-4804-3p for breast cancer, and constructed a combined diagnostic model for breast cancer using miR-548ao-5p, miR-4804-3p, CA125 and CA153, which better compensated for the shortcomings of low specificity of miRNAs and low sensitivity of traditional serum markers, and provided a new method for non-invasive diagnosis of breast cancer.

[0062] The present invention found that miR-548ao-5p and miR-4804-3p were significantly underexpressed in breast cancer through serum high-throughput sequencing and qRT-PCR, and verified that independent miR-548ao-5p and miR-4804-3p had high accuracy and specificity in breast cancer diagnosis.

[0063] The present invention constructs three risk models for breast cancer diagnosis based on the training set data, and further verifies that these three models have high sensitivity, specificity and accuracy for breast cancer diagnosis through the validation set and the collection data. At the same time, it is found through comparison that the (miRNAs+TM) panel III risk model has higher diagnostic efficacy and has potential application value.

[0064] Second, the technical solution of the present invention fills the technical gap in the industry at home and abroad:

[0065] The serum markers commonly used in breast cancer diagnosis are limited by their specificity and accuracy and can only be used as an auxiliary for breast cancer diagnosis but cannot be used for definitive diagnosis. This technical solution proposes for the first time a breast cancer risk assessment model combining miRNA with traditional serum markers, which not only improves the accuracy of breast cancer diagnosis, but also eliminates the increase in serum markers caused by liver cancer, lung cancer and gastric cancer, greatly improving the specificity of breast cancer diagnosis.

[0066] Third, the present invention mainly solves the problem of insufficient sensitivity and specificity of traditional tumor markers in early diagnosis of breast cancer. By developing a breast cancer diagnosis model based on miRNA and a multidimensional diagnosis model combined with traditional serum tumor markers, the present invention significantly improves the diagnostic efficacy of breast cancer.

[0067] 1) Key technical parameters

[0068] Serum exosome isolation and miRNA screening: Exosomes in serum were separated by ultracentrifugation, and high-throughput sequencing of miRNAs in healthy people and breast cancer patients was performed using the Illumina HiSeq SE50 platform. MiRNA-548ao-5p and miRNA-4804-3p were screened as potential molecular markers for the diagnosis of breast cancer.

[0069] Blood sample pretreatment and RNA extraction: Standardized blood sample pretreatment steps, including centrifugation, addition of lysis buffer, chloroform treatment, and column chromatography, were used to ensure the integrity of sample RNA, and external references were used to correct experimental errors.

[0070] miRNA quantification and PCR amplification: Two-step PCR amplification was performed using specific primer sequences, using miRNA-548ao-5p and miRNA-4804-3p as markers. The upstream and downstream primers were precisely designed, and a standardized amplification procedure was used to ensure the repeatability and reliability of the experiment.

[0071] Logistic regression and ROC curve analysis: Three breast cancer diagnostic models based on miRNA (panel Ⅰ), traditional serum tumor markers (panel Ⅱ), and the combination of miRNA and serum markers (panel Ⅲ) were established using the logistics regression model, and the sensitivity and specificity of the models were verified using ROC curve and DAC analysis.

[0072] 2) Existing technical problems solved

[0073] Insufficient effectiveness of traditional tumor markers: Traditional serum tumor markers (such as CA125 and CA153) have limited effectiveness in the early diagnosis of breast cancer, are prone to false positives or false negatives, and cannot accurately predict patients' disease risk.

[0074] The clinical application of miRNA biomarkers lacks standardization: Although existing studies have shown that miRNA is associated with breast cancer, there is a lack of standardized screening methods and models to integrate multidimensional data for clinical application.

[0075] Lack of multi-marker diagnostic models: Most breast cancer diagnoses still rely on a single marker or a single model, resulting in insufficient diagnostic accuracy and specificity, and a lack of diagnostic systems based on the combined use of miRNA and traditional markers.

[0076] 3) Significant technological progress

[0077] Improved diagnostic efficacy: By screening and validating miR-548ao-5p and miR-4804-3p, a new miRNA diagnostic marker combination panel I was developed, which significantly improved the sensitivity and specificity of early diagnosis of breast cancer compared with the traditional tumor marker panel II. In addition, by combining miRNA with traditional serum tumor markers (panel III), the accuracy of diagnosis was further optimized, and the false positive rate and false negative rate were significantly reduced.

[0078] Construction and verification of multidimensional models: This paper not only develops an independent diagnostic model based on miRNA, but also combines miRNA with traditional markers through multivariate logistics regression to form a joint diagnostic model panel III, providing a more comprehensive and reliable disease risk assessment system, and verifies the accuracy and specificity of the three models through ROC curves to ensure the robustness of the models in different data sets.

[0079] Clinical transformation of new biomarkers: This invention screened out new breast cancer diagnostic markers miR-548ao-5p and miR-4804-3p through high-throughput miRNA sequencing and logistics regression analysis, and verified their diagnostic efficacy in clinical samples through experiments, providing a new molecular tool for early screening of breast cancer and helping to promote the clinical transformation and application of miRNA markers.

[0080] The present invention solves the problem of insufficient diagnostic efficacy of traditional breast cancer markers by combining high-throughput miRNA sequencing technology, precise sample preprocessing and extraction steps, and a multidimensional logistics regression model, significantly improves the early diagnosis ability of breast cancer, and has high clinical application prospects and significant technological progress. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 It is a flow chart of a method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers provided in an embodiment of the present invention;

[0082] Figure 2 is the relative expression level of miR-548ao-5p in the serum of breast cancer patients and healthy controls in each data set;

[0083] Figure 3 is the relative expression level of miR-4804-3p in the serum of breast cancer patients and healthy controls in each data set;

[0084] Figure 4 is the ROC curve of miR-548ao-5p in each data set;

[0085] Figure 5 is the ROC curve of miR-4804-3p in each data set;

[0086] Figure 6 is the relative expression of CA125 and CA153 in the serum of breast cancer patients and healthy controls in the training set;

[0087] Figure 7 is the ROC curve of CA125 and CA153 in each data set;

[0088] Figure 8 is the ROC curve of the miRNAs panel I risk model in each data set;

[0089] Fig. 9 is the ROC curve of the TM Panel II risk model in each data set;

[0090] Fig.10 is the ROC curve of the (miRNAs+TM) panel III risk model in each data set;

[0091] Fig.11 Evaluate the benefits of three models for DCA;

[0092] Fig.12 are the relative expression levels of miR-548ao-5p and miR-4804-3p in liver cancer, lung cancer, gastric cancer and healthy controls, respectively. DETAILED DESCRIPTION

[0093] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0094] The following are two specific examples based on the method, which describe in detail the specific operation process and results in different experimental steps:

[0095] Example 1: Construction of an early diagnosis model for breast cancer based on miRNA

[0096] 1) Serum exosome isolation and miRNA screening:

[0097] Serum samples were collected from healthy controls and untreated breast cancer patients, 30 cases in each group, and exosomes in the serum were separated by ultracentrifugation.

[0098] The Illumina HiSeq SE50 platform was used to perform high-throughput sequencing of miRNAs in exosomes. Combined with literature review and data analysis, significantly differentially expressed miRNAs were screened out, among which miR-548ao-5p and miR-4804-3p showed significant diagnostic potential.

[0099] 2) Serum sample pretreatment:

[0100] 0.5 mL of serum was taken from each sample and centrifuged at 3000 r / min for 5 min at 4°C to remove cell debris.

[0101] The supernatant was stored in aliquots at -80°C. In the experiment, 200 μL was taken for thawing, 1 mL of lysis buffer was added, the tube was mixed by inversion and then allowed to stand at room temperature for 5 minutes.

[0102] 3) miRNA extraction and amplification:

[0103] Add the external reference miRNeasy Serum / Plasma Spike-In Control, and add 3.5 μL of external reference containing 1.6×108 copies / μL to each sample.

[0104] The miRNA was extracted by chloroform treatment and anhydrous ethanol precipitation, and amplified by two-step PCR, with 40 cycles of pre-denaturation at 95°C for 30 seconds, 95°C for 5 seconds, and 60°C for 30 seconds.

[0105] 4)Logistics regression analysis:

[0106] The screened data of miR-548ao-5p and miR-4804-3p were introduced into the Logistics regression analysis to establish the miRNA-based diagnostic model miRNAs panel I. The diagnostic efficacy in the model was determined by analyzing the patient outcome variables.

[0107] 5) Model validation and ROC analysis:

[0108] The model was applied to 20 newly collected samples of breast cancer patients and 20 healthy controls, and ROC curve analysis was performed. The model showed a sensitivity of 85%, a specificity of 90%, and an AUC (area under the curve) of 0.92, showing a high diagnostic efficacy.

[0109] Example 2: Breast cancer diagnosis model combining miRNA and traditional markers

[0110] 1) Analysis of miRNA and serum tumor markers:

[0111] Based on Example 1, the screening results of miR-548ao-5p and miR-4804-3p were combined with traditional serum tumor markers CA125 and CA153.

[0112] Two independent models were established using logistics regression: miRNA-based panel I and traditional tumor marker-based TM panel II.

[0113] 2) Establishment of joint diagnosis model:

[0114] The results of miRNAs panel I and TM panel II were combined by logistic regression to construct the combined diagnosis model miRNAs+TM panel III. The weight of each marker in the combined diagnosis model was determined by the regression coefficient in the model.

[0115] 3) Validation set data analysis and model performance comparison:

[0116] Forty newly collected serum samples (20 from breast cancer patients and 20 from healthy controls) were used for model validation.

[0117] The three models were used for breast cancer risk assessment, and ROC curve analysis was performed. The miRNAs panel I showed a sensitivity of 85% and a specificity of 90%, while the TM panel II showed a sensitivity of 70% and a specificity of 75%, while the combined model miRNAs+TM panel III showed a sensitivity of 92% and a specificity of 95%, with an AUC value of 0.96.

[0118] 4) Result analysis:

[0119] The experimental results show that the diagnostic model combining miRNA and traditional tumor markers significantly improved the diagnostic efficiency of breast cancer, especially the detection rate in early patients was significantly improved.

[0120] like Figure 1 As shown, the embodiment of the present invention provides a method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers, the method comprising:

[0121] S1: High-throughput miRNA sequencing;

[0122] S2: serum miRNA extraction;

[0123] S3: total RNA extraction from tissues;

[0124] S4: miRNA reverse transcription;

[0125] S5: Fluorescence quantitative PCR;

[0126] S6: Establishment of risk assessment model.

[0127] The S1 specifically includes:

[0128] The serum exosomes were separated by ultracentrifugation from healthy subjects and untreated breast cancer patients, and then high-throughput miRNA sequencing was performed using the Illumina HiSeq SE50 platform. Combined with literature review, miRNAs with significant differential expression were screened out, and through logistics regression and ROC analysis of breast cancer patients, miR-548ao-5p and miR-4804-3p were finally determined to be miRNAs with high diagnostic efficacy.

[0129] The S2 specifically includes:

[0130] (1) Blood sample pretreatment: 0.5 mL of serum was collected and centrifuged at 3000 r / min for 5 min at 4°C to remove cell debris. The supernatant was collected and stored at -80°C.

[0131] (2) Take 200 μL of pretreated serum specimen, thaw on ice, add 5 times the volume of the specimen, i.e. 1 mL of lysis buffer, mix thoroughly by inverting or vortexing, and let stand at room temperature (15-25°C) for 5 min;

[0132] (3) Add external reference: 3.5 μL of miRNeasySerum / PlasmaSpike-InControl containing 1.6×108 copies / μL.

[0133] (4) Add chloroform in an equal volume to the serum sample, shake vigorously for 15 seconds, place at room temperature for 5 minutes, and centrifuge at 12,000 g for 15 minutes at 4°C. Transfer the upper aqueous phase to a new EP tube, add 1.5 times the volume of anhydrous ethanol, and mix slowly. Then transfer the entire solution to a 2 mL RneasyMinElute centrifugal column, centrifuge at room temperature at 12,000 g for 1 minute, and discard the filtrate;

[0134] (5) Add 700 μL of Buffer RWT to the centrifuge column, centrifuge at 12,000 g for 1 min at room temperature, and discard the filtrate;

[0135] (6) Add 500 μL of Buffer RPE to the centrifuge column, centrifuge at 12,000 g for 1 min at room temperature, and discard the filtrate;

[0136] (7) Add 500 μL of 80% ethanol to the centrifuge column, centrifuge at 12,000 g for 2 min at room temperature, discard the filtrate, and then transfer the adsorption column to another new collection tube;

[0137] (8) Centrifuge at the highest speed for 5 min and discard the collection tube;

[0138] (9) Place the centrifuge column in a new 1.5 mL EP tube, add 14 μL RNase-Free H2O, leave for 5 min, centrifuge at the highest speed for 1 min to elute the RNA, and store at -80°C.

[0139] The S3 specifically includes:

[0140] (1) Cell lysis buffer: add 200 μL of β-mercaptoethanol to every 10 mL of TRK lysis buffer;

[0141] (2) Preparation of working solution Wash Buffer II: dilute the mother solution Wash Buffer II by adding 48 mL of anhydrous ethanol and mix by inversion;

[0142] (3) 70% ethanol preparation: Use analytical grade (>99.7%) anhydrous ethanol and DEPC water in a ratio of 7:3 and mix evenly;

[0143] (4) Tissue processing: Place the tissue soaked in RNA protection solution on ice and use a shaker to accelerate thawing. Use sterile tweezers to squeeze out excess RNA protection solution, and gently absorb the liquid on the surface with absorbent paper; weigh 20-30 mg of tissue blocks, remove fat tissue and place in a 5 mL round-bottom EP tube, cut the tissue blocks into pieces with sterile scissors, add 1.5 mL of TRK lysis solution, and homogenize the tissue blocks using a handheld homogenizer on ice. Note that each homogenization run should be about 5 seconds, and run 4 to 5 times in total; after standing for 5 minutes, aspirate the supernatant into a 1.5 mL EP tube and centrifuge at 2000 r / min for 5 minutes at room temperature;

[0144] (5) Transfer the liquid to a homogenizer column, centrifuge at 14,000 g for 2 min at room temperature to remove insoluble impurities, and collect the filtrate;

[0145] (6) Add an equal volume of 70% ethanol and mix thoroughly and slowly. Transfer the mixture to the RNA binding column and place it in a collection tube. Centrifuge at room temperature at 10,000 g for 1 min and discard the filtrate.

[0146] (7) Repeat step 6 until all the mixed solution is bound to the RNA binding column;

[0147] (8) Add 500 μL RNA Wash Buffer I to the RNA binding column, centrifuge at 10,000 g for 1 min at room temperature, and discard the filtrate;

[0148] (9) Add 500 μL RNA Wash Buffer II to the RNA binding column, centrifuge at 10,000 g for 1 min at room temperature, and discard the filtrate;

[0149] (10) Repeat step 9 and wash again;

[0150] (11) Place the RNA binding column back into the collection tube and centrifuge again at 10,000 g for 2 min at room temperature to remove residual liquid;

[0151] (12) Place the RNA binding column in a new 1.5 mL EP tube free of RNase, add 30-70 μL RNase-Free H2O to the center of the column membrane, and let stand at room temperature for 2 min. Centrifuge at 14,000 g at 4°C for 2 min, and store at -80°C for a long term.

[0152] The S4 specifically includes:

[0153] (1) Prepare the tailing reverse transcription reaction solution on ice. The reaction system is as follows:

[0154] 2×mRQ Buffer 5μL, RNA sample 0.25-8μg and 3.75μL, mRQ Enzyme 1.25μL, total volume 10μL;

[0155] (2) Reaction conditions: 37°C, 1 h; 85°C, 5 min;

[0156] (3) Add 90 μL RNase-Free H2O to dilute the obtained cDNA and store it at -20°C for a long time.

[0157] In the S5, the primer sequences are as follows:

[0158] miR-548ao-5p: upstream primer: 5′-GAAGTAACTACGCTTTTTGCA-3′;

[0159] miR-4804-3p: upstream primer: 5′-CTTAACCTTGCCCTCGAAAAA-3′;

[0160] U6: Upstream primer: 5'-GGAACGATACAGAGAAGATTAGC-3'; downstream primer: 5'-TGGAACGCTTCACGAATTTGCG-3'.

[0161] In S5, the standard procedure for two-step PCR amplification is as follows:

[0162] Pre-denaturation: 95°C, 30s; amplification: 95°C, 5s, 60°C, 30s, 40 cycles in total; melting curve: 95°C, 5s, 65°C, 1min, 95°C, 30s, 50°C, 30s.

[0163] The S6 specifically includes:

[0164] The outcome variable of breast cancer patients in the training set was set as "1", and the variable of the healthy control group was set as "0". The screened miRNA molecules, namely miR-548ao-5p and miR-4804-3p, were introduced into logistic regression. The weight of each miRNA in the model was determined according to the regression coefficient of the multivariate logistics regression model, and a breast cancer diagnosis model based on miRNA was established, named miRNAs panel I. Subsequently, the screened traditional serum tumor markers, namely CA 125 and CA153, were introduced into another logistic regression to establish a breast cancer diagnosis model based on traditional serum tumor markers, named TMpanel II. Finally, the above miRNAs and serum markers were introduced into logistic regression to form a combined diagnosis model, named miRNAs+TM) panel III, thereby calculating the risk assessment value of each model individual.

[0165] The validation set and the combined data were imported into the three risk assessment models constructed to verify the accuracy and specificity of the models. ROC curve and DAC analysis were performed to compare the diagnostic efficacy of the three models.

[0166] The embodiment of the present invention provides a system for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers, the system comprising:

[0167] miRNA sequencing module, used for high-throughput miRNA sequencing;

[0168] The miRNA extraction module is connected to the miRNA sequencing module and is used for extracting serum miRNA;

[0169] An RNA extraction module, connected to the miRNA extraction module, for extracting total RNA from tissues;

[0170] A reverse transcription module, connected to the RNA extraction module, for reverse transcription of miRNA;

[0171] A fluorescence quantitative PCR module is connected to the reverse transcription module and is used for the fluorescence quantitative PCR module;

[0172] The model building module is connected with the fluorescence quantitative PCR module for establishing the risk assessment model.

[0173] An embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers as described in any one of claims 1 to 8.

[0174] The present invention sequenced the serum miRNA of breast cancer patients and healthy controls by high-throughput sequencing, and finally screened out two miRNAs that were significantly under-expressed in the serum of breast cancer patients, namely miR-548ao-5p and miR-4804-3p. Three risk assessment models were constructed using logistics regression, namely: miRNA-based model miRNAs panel I, model TM Panel II based on traditional serum tumor markers carbohydrate antigen 125 (CA125) and carbohydrate antigen 153 (CA153), and combined miRNA and serum tumor marker model (miRNAs+TM) panel III. The sequences of the above miRNAs and their registration numbers in the miRbase database are as follows:

[0175] The gene IDs of the above traditional serum tumor markers in the NCBI database are as follows:

[0176] CA125 Gene ID: 94025; CA153 Gene ID: 4582

[0177] The present invention provides the role of serum-specific miRNA in breast cancer diagnosis, targeted therapy and prognosis, and also provides the application of miRNA combined with traditional serum markers in the preparation of breast cancer diagnosis kits and the development of breast cancer drugs.

[0178] The present invention adopts qRT-PCR technology to detect the serum of clinical breast cancer patients and healthy control groups, and compares the AUC, sensitivity, specificity, 95% confidence interval (95% CI) of single miRNA, single traditional serum tumor marker, miRNAs panel I, TM Panel II and (miRNAs+TM) panel III for breast cancer diagnosis through ROC analysis, and calculates the accuracy, positive predictive value and negative predictive value of miRNAs panel I, TM Panel II and (miRNAs+TM) panel III, respectively.

[0179] The clinical patients in the appeal included a training set (116 breast cancer patients and 68 healthy controls), a validation set (53 breast cancer patients and 46 healthy controls), and a collection set (169 breast cancer patients and 114 healthy controls).

[0180] The present invention constructs a risk assessment model based on the training set data, and proves the accuracy and specificity of the model through the validation set and the collection.

[0181] The miRNAs panel I used in the breast cancer discrimination formula of the present invention is:

[0182] Risk score = -1.568 × miR-548ao expression 1.134 × miR-4804 expression + 2.494

[0183] The TM Panel II of the present invention for the breast cancer discrimination formula is:

[0184] Risk score = 0.15 × CA125 + 0.101 × CA1532.70

[0185] The (miRNAs+TM) panel III of the present invention for breast cancer discrimination formula is:

[0186] Risk score = 0.123 × CA125 + 0.069 × CA153 1.173 × miR-548ao expression 1.185 × miR-4804 expression

[0187] The present invention compares the ROC analysis of single miRNA, single traditional serum tumor marker, miRNAs panel I, TM Panel II and (miRNAs+TM) panel III in each data set, see Table 1 (training set), Table 2 (validation set), Table 3 (combination set).

[0188] Table 1. ROC analysis of each group of risk assessment models in the training set

[0189]

[0190] Table 2. ROC analysis of each group of risk assessment models in the validation set

[0191]

[0192] Table 3. ROC analysis of each group of risk assessment models in the collection

[0193]

[0194] Through ROC analysis of each group of risk assessment models in 3 data sets, it was found that:

[0195] 1. Independent miR-548ao-5p or miR-4804-3p has certain diagnostic value for breast cancer.

[0196] 2. There was no significant difference in the diagnostic efficacy of the miRNAs panel I risk model for breast cancer between miR-548ao-5p and miR-4804-3p (p>0.05).

[0197] 3. The diagnostic efficacy of miRNAs panel I risk model for breast cancer is higher than that of CA125 and CA153 (P<0.05), and it can effectively improve the sensitivity of breast cancer diagnosis (P<0.05).

[0198] 4. The diagnostic efficacy of TM Panel II risk model for breast cancer is better than CA153 (p<0.05), and there is no statistical difference between the risk models of CA125 and miRNAs panel I (p>0.05). However, the accuracy and sensitivity of miRNAs panel I risk model for breast cancer diagnosis are higher than those of TM panel II risk model (P<0.05), which can reduce the probability of missed diagnosis of patients.

[0199] 5. The diagnostic efficacy of the (miRNAs+TM) panel III risk model for breast cancer is higher than that of miRNAs panel I, TM Panel II, single miRNA molecules and single serum tumor markers.

[0200] 6. The parameters of the (miRNAs+TM) panel III risk model have better stability in the training set, validation set and collection.

[0201] The present invention found that miR-548ao-5p and miR-4804-3p were significantly underexpressed in breast cancer, and verified that independent miR-548ao-5p and miR-4804-3p had high accuracy and specificity in breast cancer diagnosis.

[0202] The present invention proposes three risk models with higher diagnostic efficacy for breast cancer. By comparison, it is found that the (miRNAs+TM) panel III risk model constructed by combining miRNA with traditional markers has higher sensitivity, specificity, accuracy and stability.

[0203] Furthermore, decision curve analysis (DCA) showed that all three models could achieve net clinical benefits, among which the (miRNAs+TM) panel III model could maximize the net benefit for patients.

[0204] Figure 2 is the relative expression level of miR-548ao-5p in the serum of breast cancer patients and healthy controls in each data set;

[0205] Figure 3 is the relative expression level of miR-4804-3p in the serum of breast cancer patients and healthy controls in each data set;

[0206] Figure 4is the ROC curve of miR-548ao-5p in each data set;

[0207] Figure 5 is the ROC curve of miR-4804-3p in each data set;

[0208] Figure 6 is the relative expression of CA125 and CA153 in the serum of breast cancer patients and healthy controls in the training set;

[0209] Figure 7 is the ROC curve of CA125 and CA153 in each data set;

[0210] Figure 8 is the ROC curve of the miRNAs panel I risk model in each data set;

[0211] Fig. 9 is the ROC curve of the TM Panel II risk model in each data set;

[0212] Fig.10 is the ROC curve of the (miRNAs+TM) panel III risk model in each data set;

[0213] Fig.11 Evaluate the benefits of three models for DCA;

[0214] Fig.12 are the relative expression levels of miR-548ao-5p and miR-4804-3p in liver cancer, lung cancer, gastric cancer and healthy controls, respectively.

[0215] The present invention analyzed the expression of miR-548ao-5p and miR-4804-3p in the serum of patients with liver cancer, lung cancer and gastric cancer (each group had 28 cancer patients and 25 healthy controls), and verified the specificity of the risk model composed of these two miRNAs in the diagnosis of breast cancer.

[0216] The present invention evaluated for the first time the diagnostic value of serum miR-548ao-5p and miR-4804-3p for breast cancer, and constructed a combined diagnostic model for breast cancer using miR-548ao-5p, miR-4804-3p, CA125 and CA153, which better compensated for the shortcomings of low specificity of miRNAs and low sensitivity of traditional serum markers, and provided a new method for non-invasive diagnosis of breast cancer.

[0217] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers, characterized in that: The method includes: S1: High-throughput miRNA sequencing Serum samples from breast cancer patients and healthy controls were collected, and serum exosomes were isolated using ultracentrifugation. High-throughput miRNA sequencing was then performed using the Illumina HiSeq SE50 platform. The miRNA sequence data extracted from serum exosomes were analyzed and processed to screen out miRNAs with significant differences in expression between breast cancer patients and healthy controls. S2: Serum miRNA extraction Collect pretreated serum samples. The specific steps include removing cell debris by low-speed centrifugation of 0.5 mL serum, then extracting miRNA in the serum by column chromatography by adding lysis buffer and chloroform, and adding external reference miRNeasy Serum / Plasma Spike-In Control to ensure the stability and repeatability of the extraction step. The extracted miRNA will be used for subsequent reverse transcription and quantitative analysis; S3: Total RNA extraction from tissues Total RNA was extracted from tissue samples of breast cancer patients. The processed tissue blocks were homogenized with TRK lysate and homogenizer to remove fat and other insoluble impurities. Then, total RNA in the tissue samples was extracted using RNA binding columns. RNA purity was ensured through repeated washing and centrifugation steps. The extracted RNA was used for tissue verification of miRNA expression. S4: miRNA reverse transcription The tailing method was used to reverse transcribe miRNA. The extracted miRNA was reverse transcribed into cDNA. First, the tailing method reverse transcription reaction system was prepared, including 2×mRQ Buffer, RNA sample, and mRQ enzyme. The reaction system was reacted at 37°C for 1 hour, and then heat-inactivated at 85°C for 5 minutes. The reverse transcribed cDNA will be used for subsequent fluorescence quantitative PCR amplification; S5: Fluorescence quantitative PCR Fluorescence quantitative PCR was used to quantitatively analyze the expression of miRNA in breast cancer patients and healthy controls. Two-step PCR amplification was performed using specific primers for miR-548ao-5p and miR-4804-3p and U6 as an internal reference. The PCR amplification procedure included 95°C pre-denaturation, amplification steps, and melting curve analysis. The expression data of miRNA were obtained through 40 cycles of reaction. S6: Risk assessment model establishment According to the miRNA expression data obtained by quantitative PCR, the expression values ​​of miR-548ao-5p and miR-4804-3p screened out were first introduced into the model using the Logistics regression model to establish a miRNA risk assessment model. At the same time, traditional serum tumor markers (such as CA125 and CA153) were introduced into another Logistics regression model to establish a risk assessment model for traditional tumor markers. Finally, miRNA and traditional markers were combined to construct a joint diagnostic model, and the accuracy of the model was verified by ROC curve and DAC analysis.

2. The method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers according to claim 1, characterized in that: The S1 specifically includes: The serum exosomes were separated by ultracentrifugation from healthy subjects and untreated breast cancer patients, and high-throughput miRNA sequencing was performed using the Illumina HiSeq SE50 platform. Combined with literature review, miRNAs with significant differential expression were screened out, and through logistics regression and ROC analysis of breast cancer patients, miR-548ao-5p and miR-4804-3p were finally determined to be miRNAs with high diagnostic efficacy.

3. The method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers as claimed in claim 1, characterized in that: The S2 specifically includes: (1) Blood sample pretreatment: 0.5 mL of serum was collected and centrifuged at 3000 r / min for 5 min at 4°C to remove cell debris. The supernatant was collected and stored at -80°C. (2) Take 200 μL of pretreated serum specimen, thaw on ice, add 5 times the volume of the specimen, i.e. 1 mL of lysis buffer, mix thoroughly by inverting or vortexing, and let stand at room temperature (15-25°C) for 5 min; (3) Add external reference: 3.5 μL miRNeasySerum / PlasmaSpike-InControl containing 1.6×108 copies / μL; (4) Add chloroform in an equal volume to the serum sample, shake vigorously for 15 seconds, place at room temperature for 5 minutes, and centrifuge at 12,000 g for 15 minutes at 4°C; transfer the upper aqueous phase to a new EP tube, add 1.5 times the volume of anhydrous ethanol, and mix slowly; then transfer the entire solution to a 2 mL RNeasyMinElute centrifugal column, centrifuge at room temperature at 12,000 g for 1 minute, and discard the filtrate; (5) Add 700 μL of Buffer RWT to the centrifuge column, centrifuge at 12,000 g for 1 min at room temperature, and discard the filtrate; (6) Add 500 μL of Buffer RPE to the centrifuge column, centrifuge at 12,000 g for 1 min at room temperature, and discard the filtrate; (7) Add 500 μL of 80% ethanol to the centrifuge column, centrifuge at 12,000 g for 2 min at room temperature, discard the filtrate, and then transfer the adsorption column to another new collection tube; (8) Centrifuge at the highest speed for 5 min and discard the collection tube; (9) Place the centrifuge column in a new 1.5 mL EP tube, add 14 μL RNase-Free H2O, leave for 5 min, centrifuge at the highest speed for 1 min to elute the RNA, and store at -80°C.

4. The method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers as claimed in claim 1, characterized in that: The S3 specifically includes: (1) Cell lysis buffer: add 200 μL of β-mercaptoethanol to every 10 mL of TRK lysis buffer; (2) Preparation of working solution Wash Buffer II: Add 48 mL of anhydrous ethanol to the mother solution Wash Buffer II to dilute it and mix thoroughly by inverting; (3) 70% ethanol preparation: Mix analytical grade anhydrous ethanol and DEPC water in a ratio of 7:3; (4) Tissue processing: Place the tissue soaked in RNA protection solution on ice and use a shaker to accelerate thawing. Use sterile tweezers to squeeze out excess RNA protection solution, and gently absorb the liquid on the surface with absorbent paper; weigh 20-30 mg of tissue blocks, remove fat tissue and place in a 5 mL round-bottom EP tube, cut the tissue blocks into pieces with sterile scissors, add 1.5 mL of TRK lysis solution, and homogenize the tissue blocks using a handheld homogenizer on ice. Note that each homogenization run should be about 5 seconds, and run 4 to 5 times in total; after standing for 5 minutes, aspirate the supernatant into a 1.5 mL EP tube and centrifuge at 2000 r / min for 5 minutes at room temperature; (5) Transfer the liquid to a homogenizer column, centrifuge at 14,000 g for 2 min at room temperature to remove insoluble impurities, and collect the filtrate; (6) Add an equal volume of 70% ethanol and mix thoroughly and slowly; transfer the mixture to the RNA binding column and place it in a collection tube. Centrifuge at 10,000 g for 1 min at room temperature and discard the filtrate; (7) Repeat step 6 until all the mixed solution is bound to the RNA binding column; (8) Add 500 μL RNA Wash Buffer I to the RNA binding column, centrifuge at 10,000 g for 1 min at room temperature, and discard the filtrate; (9) Add 500 μL RNA Wash Buffer II to the RNA binding column, centrifuge at 10,000 g for 1 min at room temperature, and discard the filtrate; (10) Repeat step 9 and wash again; (11) Place the RNA binding column back into the collection tube and centrifuge again at 10,000 g for 2 min at room temperature to remove residual liquid; (12) Place the RNA binding column into a new 1.5 mL EP tube free of RNase, add 30-70 μL RNase-free H2O to the center of the column membrane, let stand at room temperature for 2 min; centrifuge at 14,000 g at 4°C for 2 min, and store at -80°C for a long term.

5. The method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers as claimed in claim 1, characterized in that: The S4 specifically includes: (1) Prepare the tailing reverse transcription reaction solution on ice. The reaction system is as follows: 2×mRQ Buffer 5μL, RNA sample 0.25-8μg and 3.75μL, mRQ Enzyme 1.25μL, total volume 10μL; (2) Reaction conditions: 37°C, 1 h; 85°C, 5 min; (3) Add 90 μL RNase-Free H2O to dilute the obtained cDNA and store it at -20°C for a long time.

6. The method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers as claimed in claim 1, characterized in that: In the S5, the primer sequences are as follows: miR-548ao-5p: upstream primer: 5′-GAAGTAACTACGCTTTTTGCA-3′; miR-4804-3p: upstream primer: 5′-CTTAACCTTGCCCTCGAAAAA-3′; U6: Upstream primer: 5'-GGAACGATACAGAGAAGATTAGC-3'; downstream primer: 5'-TGGAACGCTTCACGAATTTGCG-3'.

7. The method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers as claimed in claim 1, characterized in that: In S5, the standard procedure for two-step PCR amplification is as follows: Pre-denaturation: 95°C, 30s; amplification: 95°C, 5s, 60°C, 30s, 40 cycles in total; melting curve: 95°C, 5s, 65°C, 1min, 95°C, 30s, 50°C, 30s.

8. The method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers as claimed in claim 1, characterized in that: The S6 specifically includes: The outcome variable of breast cancer patients in the training set was set as "1", and the variable of the healthy control group was set as "0". The screened miRNA molecules, namely miR-548ao-5p and miR-4804-3p, were introduced into logistic regression. The weight of each miRNA in the model was determined according to the regression coefficient of the multivariate logistics regression model, and a breast cancer diagnosis model based on miRNA was established, named miRNAs panel Ⅰ. Subsequently, the screened traditional serum tumor markers, namely CA 125 and CA153, were introduced into another logistic regression to establish a breast cancer diagnosis model based on traditional serum tumor markers, named TM panel Ⅱ. Finally, the above miRNAs and serum markers were introduced into logistic regression to form a combined diagnosis model, named miRNAs+TM) panel Ⅲ, thereby calculating the risk assessment value of each model individual. The data of the validation set and the collection were respectively imported into the three constructed risk assessment models to verify the accuracy and specificity of the models; ROC curve and DAC analysis were performed at the same time to compare the diagnostic efficacy of the three models.

9. A system for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers, which implements the method for constructing a breast cancer risk assessment model based on the combination of MircoRNA and traditional serum markers as described in any one of claims 1 to 8, characterized in that: The system includes: miRNA sequencing module, used for high-throughput miRNA sequencing; The miRNA extraction module is connected to the miRNA sequencing module and is used for extracting serum miRNA; An RNA extraction module, connected to the miRNA extraction module, for extracting total RNA from tissues; A reverse transcription module, connected to the RNA extraction module, for reverse transcription of miRNA; A fluorescence quantitative PCR module is connected to the reverse transcription module and is used for the fluorescence quantitative PCR module; The model building module is connected with the fluorescence quantitative PCR module for establishing the risk assessment model.

Images