Prediction model for recurrence-free survival of high-grade serous ovarian cancer advanced patient
Through proteomic analysis and mass spectrometry of urine samples, a recurrence-free survival prediction model is constructed for advanced patients with high-grade serous ovarian cancer, solving the problems of prediction accuracy and monitoring convenience in the existing technology, and realizing the Accurate prediction and long-term monitoring of recurrence.
Patent Information
- Application Number
- CN202410821610.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-24
- Publication Date
- 2024-11-19
AI Technical Summary
The prior art is difficult to accurately predict the recurrence of patients with advanced serous ovarian cancer in high-grade, especially in patients with metal inserts or allergic to contrast agents, and the MRI examination has limitations and cannot be applied to long-term intensive monitoring. .
By obtaining urine samples from advanced patients with high-grade serous ovarian cancer, performing proteomic analysis, screening differential protein expression matrix, selecting candidate biomarkers, and using mass spectrometry technology and LASSO regression to build predictive models to establish suitable for routine and special Patients’ recurrence-free survival predictive model.
Accurate prediction of recurrence-free survival in patients with advanced serous ovarian cancer is achieved. It is suitable for patients with metal inserts or foreign bodies and patients allergic to contrast agents. The urine samples are easy to obtain and preserve, suitable for long-term intensive Monitoring improves prediction accuracy and patient compliance.
Smart Images

Figure CN118983105A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of medical prediction technology, and in particular to a prediction model and method for recurrence-free survival of patients with advanced high-grade serous ovarian cancer. Background Art
[0002] The description of this background technology belongs to the related technology related to the present invention, which is only used to illustrate and facilitate the understanding of the invention content of the present invention, and should not be understood as the applicant explicitly believes or inferred that the applicant believes that it is the prior art of the present invention on the filing date of the first application.
[0003] Ovarian cancer is one of the most common malignancies of the female reproductive system. Due to the lack of specific early symptoms and effective screening methods, more than 75% of ovarian cancer patients are diagnosed in the late stage, resulting in high recurrence and mortality rates. High grade serous ovarian cancer (HGSOC) is one of the most common histological types of ovarian cancer, with invasive characteristics, accounting for about 70% of ovarian cancer cases. Currently, the treatment of HGSOC patients is primary debulking surgery (PDS), followed by platinum-based chemotherapy or neoadjuvant chemotherapy (NAC), or interval debulking surgery (IDS). Although the first treatment of HGSOC patients has a good effect, it is common for recurrence and poor efficacy to occur, especially for advanced patients. Therefore, accurate prediction of recurrence-free survival in advanced HGSOC patients and timely optimization and adjustment of treatment plans based on the prediction results may be the key to improving survival outcomes.
[0004] Existing reports have shown that studies on recurrence-free survival of HGSOC patients mainly focus on single factors, such as LINC00515, SLC7A1 and CXCR4. However, these studies stopped at the correlation between the above factors and recurrence-free survival, and no further predictive studies were conducted. In 2021, the Shanghai Cancer Center of Fudan University constructed a model for predicting recurrence-free survival in patients with advanced HGSOC based on magnetic resonance imaging technology. Research data showed that the model's predictive accuracy index AUC (AreaUnder Curve) value was 0.79, which has a certain predictive value. However, this method has limitations in actual use. For example, MRI examinations are not suitable for patients with metal implants or foreign bodies in the body, patients who are allergic to contrast agents, etc.; the economic burden and patient compliance issues brought about by MRI examinations are not suitable for long-term intensive monitoring. Summary of the invention
[0005] The purpose of the embodiments of the present invention is to provide a prediction model and method for recurrence-free survival of patients with advanced high-grade serous ovarian cancer. The present invention is applicable to conventional advanced HGSOC patients, as well as patients with metal implants or foreign bodies and patients who are allergic to contrast agents. The urine sample involved in the present invention is easy to obtain and preserve, and is suitable for long-term intensive monitoring.
[0006] The purpose of the embodiment of the present invention is achieved through the following technical solutions:
[0007] A method for establishing a prediction model for recurrence-free survival in patients with advanced high-grade serous ovarian cancer, comprising the following steps:
[0008] (1) Urine samples were obtained from patients with advanced high-grade serous ovarian cancer before treatment, and the patients were divided into a recurrence group and a non-recurrence group based on the time interval of 8-16 months (e.g., 12 months) after initial treatment;
[0009] (2) pre-treating the urine sample before treatment to obtain peptide fragments;
[0010] (3) performing proteomic analysis on the peptide segments to obtain a differential protein expression matrix between the relapse group and the non-relapse group;
[0011] (4) Screening of candidate biomarkers;
[0012] (5) Mass spectrometry-based targeted validation of candidate biomarkers;
[0013] (6) Analyze the correlation between candidate biomarkers and recurrence-free survival time, and use LASSO regression and high and low levels of biomarkers to construct a prediction model for recurrence-free survival in patients with advanced high-grade serous ovarian cancer.
[0014] Furthermore, the patient with advanced high-grade serous ovarian cancer has no history of other malignant tumors; the urine sample is the supernatant obtained by centrifugation of the mid-morning urine;
[0015] Step (2) comprises the following steps: taking a urine sample and mixing it with a pre-cooled acetone solution, standing it and then centrifuging it, discarding the supernatant, and dissolving the protein precipitate with a urea solution; determining the protein concentration; alkylating the protein, adding a trypsin solution to cut the protein into free peptides, and desalting and drying the peptides to obtain peptide powder.
[0016] Furthermore, step (3) specifically includes the following steps: dissolving the peptide powder with a formic acid aqueous solution, then taking the peptide solution into a nanofluidic system for chromatographic separation, and finally spraying it into a high-resolution mass spectrometer for signal acquisition and analysis.
[0017] Furthermore, the main parameters of the nanofluid phase are as follows:
[0018] One precolumn: 2 cm in length, 75 μm in inner diameter, filled with 5 μm C18 filler;
[0019] One analytical column: 25 cm in length, 75 μm in inner diameter, filled with 2 μm C18 filler;
[0020] Mobile phase A: 2% acetonitrile, 0.1% formic acid, 98% water;
[0021] Mobile phase B: 98% acetonitrile, 0.1% formic acid, 2% water;
[0022] The peptide solution is first enriched in the pre-column, then eluted into the analytical column under the control of the ratio of mobile phase A and B, and chromatographic separation is performed; finally, the peptide reaches the needle tip at the end of the analytical column, and ionized spray is sent to the mass spectrometer for analysis at a voltage of 2100V;
[0023] Furthermore, the main parameters of mass spectrometry analysis are as follows: mass spectrometry was performed in DIA mode; primary spectrum resolution was 35000, and the maximum ion injection time was 20 ms; (3) secondary spectrum resolution was 17500, fragmentation energy was 27, maximum ion injection time was set to automatic, detection range was 380-980 Da, mass window was 20 Da, and there were 30 windows in total;
[0024] The human FASTA files of the Uniprot database and the DIA-NN software were used to analyze the raw mass spectrometry data, and the differential protein expression matrix between the recurrence group and the non-recurrence group in the urine samples of the discovery set was obtained by statistical analysis.
[0025] Furthermore, the top 10 upregulated proteins in the recurrence group of the discovery set were selected as candidate biomarkers; the screening of candidate biomarkers followed the following principles: (1) the candidate biomarker had a higher level in the recurrence group compared with the non-recurrence group; (2) the candidate biomarker had a characteristic peptide segment and a high-responsive signal in the mass spectrometry; (3) the characteristic peptide segment of the candidate biomarker had no post-translational modification in the urine sample.
[0026] Furthermore, step (5) specifically includes the following steps:
[0027] A. Establish a targeted validation method: Import the mass spectrometry raw data and DIA-NN analysis results into Skyline software to select the characteristic peptides of candidate biomarkers; the characteristic peptides meet the screening principles of candidate biomarkers and contain clear amino acid sequence information, mass-to-charge ratio, charge, and chromatographic retention time; export the information of characteristic peptides and the mass spectrometry targeted validation method;
[0028] B. Targeted validation of candidate biomarkers: The mass spectrometry acquisition parameters were set according to the targeted validation method established above, as follows: the mass spectrometry was performed in PRM mode for targeted analysis; the primary spectrum resolution was 35,000, and the maximum ion injection time was 20 ms; the secondary spectrum resolution was 17,500, the fragmentation energy was 27, and the maximum ion injection time was 100 ms; the chromatographic conditions were consistent with step (3); the raw data collected by the mass spectrometry targeted validation was re-imported into the Skyline software for analysis, and the levels of candidate biomarkers in the discovery set and validation set were statistically obtained; finally, the discovery set and validation set data were merged, and the levels of each candidate biomarker were classified into two levels, high and low, according to the cutoff value.
[0029] Furthermore, step (6) specifically includes the following steps: firstly, analyzing the correlation between the candidate biomarker and the recurrence-free survival time, clarifying the clinical value of the candidate biomarker and its potential value as a biomarker to predict recurrence-free survival; finally, constructing a prediction model for recurrence-free survival of patients with advanced HGSOC using LASSO regression and the high and low levels of the biomarker;
[0030] A. Correlation analysis: The present invention uses the survival analysis module of GraphPad Prism 9.0 software to analyze the correlation between candidate biomarkers and recurrence-free survival time, and the parameters are default parameters.
[0031] B. Model construction: The present invention uses the LASSO regression method in R language to construct the model, and the main parameters are as follows: first, the cv.glmnet function (nfolds=5) is used to optimize the important parameter lambda in the LASSO regression model to prevent the model from overfitting; then the optimized lambda value is called in combination with the glmnet function to calculate the variable correlation coefficient, screen the variables, fit the regression formula, and build a prediction model; finally, the predictive accuracy of the model is tested using the predict function.
[0032] Based on the above correlation analysis and LASSO regression, a recurrence-free survival prediction model for patients with advanced HGSOC was constructed.
[0033] A prediction model for recurrence-free survival in patients with advanced high-grade serous ovarian cancer was established using the above method.
[0034] Furthermore, in the prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer, G6PI, SPB3 and SPRR3 were used as biomarkers to establish a prediction model:
[0035]
[0036] Where: P is probability; G6PI bi 、SPB3bi and SPRR3 bi It is binary data, i.e. 0 or 1; the content data can be divided into high: 1 or low: 0 by comparing with the cutoff value in the table below (after log2 transformation).
[0037]
[0038] It should be noted that the above 8-16 months (e.g., 12 months) time node is only an exemplary description, and the specific time range can be adjusted according to actual conditions or needs. This time range is set to statistically distinguish patients in the recurrence group and the non-recurrence group for subsequent biomarker screening and model construction. However, depending on the purpose of the study, patient characteristics, or other clinical needs, researchers can choose different time nodes for grouping to ensure the applicability and accuracy of the prediction model.
[0039] For example, if the study found that most patients with advanced high-grade serous ovarian cancer had signs of recurrence within 6-10 months after initial treatment, the time node could be set to 10 months instead of 12 months. In this way, the recurrence group would include patients who recurred within 10 months after treatment, and the non-recurrence group would include patients who did not relapse within 10 months. Adjusting the time node helps to more accurately reflect the disease progression and treatment effects of a specific patient population, thereby improving the accuracy of the prediction model.
[0040] Therefore, the specific time node should be reasonably set based on actual clinical data and research needs, and should not be limited to the range of 8-16 months (e.g. 12 months).
[0041] Furthermore, the present invention provides the following embodiments for predicting the possibility of recurrence in patients with advanced high-grade serous ovarian cancer.
[0042] The first embodiment of the present invention provides a biomarker combination for prediction and diagnosis of recurrence-free survival in patients with advanced high-grade serous ovarian cancer.
[0043] Further, the biomarker combination includes G6PI, SPB3 and SPRR3, and optionally includes ANAX1.
[0044] The second embodiment of the present invention provides a reagent used in a product for predicting and diagnosing recurrence-free survival of patients with advanced high-grade serous ovarian cancer, for detecting the level of the biomarker combination described in the first embodiment of the present invention in a sample.
[0045] Specifically, the reagent is used to detect the expression amount or level of the biomarker combination in a sample by nucleic acid sequencing technology, nucleic acid hybridization technology, chromatography technology, mass spectrometry technology, digital imaging technology, protein immunoassay technology, dye technology and / or next-generation sequencing technology. Preferred methods include chromatography technology, mass spectrometry technology and protein immunoassay technology.
[0046] In addition, the reagent also includes a reagent for detecting the mRNA expression of the biomarker combination, preferably a reagent for detecting the level of cDNA complementary to the mRNA transcribed by the biomarker combination, more preferably a primer or a probe. Similarly, the reagent also includes a reagent for detecting the protein expression of the biomarker combination, preferably a reagent for detecting the level of the polypeptide or protein encoded by the biomarker combination, more preferably an antibody, an antibody fragment or an affinity protein.
[0047] The primers described in the present invention can be prepared by chemical synthesis, designed and synthesized using known information. Similarly, the probe can also be prepared by chemical synthesis, or by preparing a gene containing the target nucleic acid sequence from biological material, and amplifying the target nucleic acid sequence using the designed primers.
[0048] The probe hybridizing with the gene nucleic acid sequence can be DNA, RNA, DNA-RNA chimera, PNA or other derivatives. There is no limit to the length of the probe, as long as it can complete specific hybridization and specifically bind to the target nucleotide sequence. The length of the probe can be as short as 10 base pairs and as long as 300 base pairs or longer. Usually, the length of the probe is greater than 12 base pairs, and the longest is generally no more than 30 base pairs, with 15-25 base pairs being the best. The self-complementary sequence of the probe should be less than 4 base pairs to avoid affecting the hybridization efficiency.
[0049] In one embodiment of the present invention, the sample is a body fluid sample of a subject, such as urine.
[0050] The present invention predicts the recurrence-free survival rate of patients by quantitatively analyzing the expression levels of the biomarker combination G6PI, SPB3 and SPRR3, and optionally ANAX1 in samples of patients with advanced high-grade serous ovarian cancer, and analyzes the expression levels of these biomarkers in the samples to diagnose whether the subjects have a good recurrence-free survival rate.
[0051] The third embodiment of the present invention provides a product for predicting and diagnosing recurrence-free survival in patients with advanced high-grade serous ovarian cancer.
[0052] The product comprises a reagent for detecting the expression of the biomarker combination described in the first embodiment of the present invention. The preferred form is an in vitro diagnostic product, more preferably an in vitro diagnostic kit. These reagents include reagents for detecting the mRNA expression of the biomarker combination and / or reagents for detecting the protein expression. Specifically, these reagents can be primers, probes, antibodies, antibody fragments or affinity proteins. The product can also include total RNA extraction reagents, reverse transcription reagents and / or second-generation sequencing reagents. Preferably, the reagents are antibodies to the markers, primers for PCR, reagents for mass spectrometry analysis or reagents for chromatographic analysis.
[0053] The present invention also provides a use of a reagent for detecting the biomarker of the present invention in preparing a kit or a detection device for predicting recurrence-free survival of patients with advanced high-grade serous ovarian cancer.
[0054] In the present invention, the detection method of the expression amount of the biomarker combination can adopt a known determination method in the field, such as detecting the amount of the RNA transcript of the gene or the polypeptide encoded by the gene. Preferably, the RNA transcript of the gene can be converted into complementary cDNA by a known method, and the amount of the RNA transcript is obtained by measuring the amount of complementary cDNA. These amounts can be standardized for the total RNA or total cDNA in the sample, or for the amount of RNA transcripts or complementary cDNA of a group of housekeeping genes. RNA transcripts can be detected and quantified by hybridization, amplification or sequencing methods, including but not limited to hybridization of RNA transcripts with probes or primers, and detecting the amount of RNA transcripts or corresponding cDNA products by various quantitative PCR techniques or sequencing techniques of polymerase chain reaction (PCR). These quantitative PCR techniques include fluorescent quantitative PCR, real-time PCR or semi-quantitative PCR techniques. Sequencing techniques include Sanger sequencing, second-generation sequencing, third-generation sequencing and single-cell sequencing.
[0055] The fourth embodiment of the present invention proposes a risk assessment model for predicting recurrence-free survival in patients with advanced high-grade serous ovarian cancer.
[0056] Furthermore, the risk assessment model includes the following biomarker combination: G6P I, SPB3 and SPRR3; preferably, the risk assessment model uses the following regression equation to calculate the risk score:
[0057]
[0058] Where: P is probability; G6PI bi 、SPB3 bi and SPRR3 bi It is binary data, i.e. 0 or 1; the content data can be divided into high: 1, or low: 0 by comparing with the cutoff value in Table 2 (after log2 transformation).
[0059] It should be noted that the present invention provides a model method for predicting recurrence-free survival in patients with advanced high-grade serous ovarian cancer. Although the present invention uses the LASSO regression model, those skilled in the art can use other types of models or algorithms to construct different models to achieve similar prediction functions according to specific needs, data characteristics, and different technical means used in the construction process.
[0060] A fifth embodiment of the present invention provides a device or system for predicting recurrence-free survival of patients with advanced high-grade serous ovarian cancer.
[0061] The device or system comprises:
[0062] 1. A data acquisition module, used to obtain the biomarker combination expression profile data in the sample of the subject to be tested;
[0063] 2. A prediction module, which is used to provide the gene group expression profile data obtained by the data acquisition module as input data to the trained prediction model, and the prediction model predicts recurrence-free survival based on the biomarker combination expression profile data of the subject;
[0064] 3. A prediction result acquisition module, used to acquire the output result of the prediction model in the prediction module, and obtain the recurrence-free survival prediction result of the subject. The preferred prediction model is the LASSO regression model, and the most preferred is the risk assessment model described in the fourth embodiment of the present invention.
[0065] A sixth embodiment of the present invention provides a computer device.
[0066] The computer device includes a memory and a processor. The memory stores a program. When the processor executes the program, the following method is implemented:
[0067] 1. Obtaining biomarker combination expression profile data in the sample of the subject to be tested;
[0068] 2. Provide the biomarker combination expression profile data as input data to the trained prediction model;
[0069] 3. Output the recurrence-free survival prediction result of the subject to be tested. The preferred prediction model is the LASSO regression model, and the most preferred is the prediction model described in the fourth embodiment of the present invention.
[0070] A seventh embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a computer, the method described in the sixth embodiment of the present invention is performed.
[0071] The eighth embodiment of the present invention provides the use of the biomarker in predicting recurrence-free survival of patients with advanced high-grade serous ovarian cancer for non-disease diagnosis or treatment purposes.
[0072] The ninth embodiment of the present invention provides the use of any one of G6P I, SPB3, SPRR3 and ANAX1 or any two, three or four of them as biomarkers in the preparation of a preparation for predicting recurrence-free survival in patients with advanced high-grade serous ovarian cancer.
[0073] The tenth embodiment of the present invention provides a method for evaluating whether a drug can enable recurrence-free survival in patients with advanced high-grade serous ovarian cancer, which comprises using any one of G6PI, SPB3, SPRR3 and ANAX1, or any two, three or four of them as biomarkers to evaluate the effect of the drug.
[0074] The eleventh embodiment of the present invention provides a method for screening compounds that can enable patients with advanced high-grade serous ovarian cancer to survive without recurrence, which comprises using any one of G6PI, SPB3, SPRR3 and ANAX1, or any two, three or four of them as biomarkers to evaluate the effect of the compound.
[0075] The present invention constructs a prediction model for recurrence-free survival in patients with advanced HGSOC and its application, which is not only applicable to conventional patients with advanced HGSOC, but also to patients with metal implants or foreign bodies and patients who are allergic to contrast agents. In addition, the urine sample in the present invention is easy to obtain and preserve, and is suitable for long-term intensive monitoring. BRIEF DESCRIPTION OF THE DRAWINGS
[0076] Figure 1 The present invention is used for the prediction model of recurrence-free survival of patients with advanced high-grade serous ovarian cancer, showing the differential protein expression matrix of the non-recurrence group (n=16) and the recurrence group (n=15) in the urine samples (p value <0.01, |log2(difference fold)|≥1);
[0077] Figure 2 The volcano plot in the prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer of the present invention shows upregulated (red) and downregulated (blue) proteins in urine samples of patients in the recurrence group, and also shows 10 candidate biomarkers screened by the present invention;
[0078] Figure 3The results of the targeted validation of candidate biomarkers ANAX1, G6PI, SPB3 and SPRR3 in the prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer of the present invention are shown in Figure 1, where: (A) the level of candidate biomarkers in the discovery set; (B) the level of candidate biomarkers in the validation set; (C) the level of candidate biomarkers after the discovery set and the validation set are merged. *p value < 0.05; **p value < 0.01; ***p value < 0.001;
[0079] Figure 4 This is a correlation analysis diagram of the candidate biomarkers in the prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer of the present invention and the recurrence-free survival time of patients with advanced HGSOC;
[0080] Figure 5 This is the accuracy test result of the 12-month recurrence-free survival prediction model for advanced HGSOC patients in the prediction model for recurrence-free survival of advanced high-grade serous ovarian cancer patients used in the present invention. DETAILED DESCRIPTION
[0081] The present application is further described below in conjunction with embodiments.
[0082] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, in the following description, different "one embodiment" or "embodiment" does not necessarily refer to the same embodiment. Different embodiments can be replaced or combined, and for ordinary technicians in this field, other implementation methods can be obtained based on these embodiments without creative work.
[0083] A method for establishing a prediction model for recurrence-free survival in patients with advanced high-grade serous ovarian cancer, comprising the following steps:
[0084] (1) Urine samples were obtained from patients with advanced high-grade serous ovarian cancer before treatment, and the patients were divided into recurrence group and non-recurrence group based on the 12-month mark after initial treatment;
[0085] (2) pre-treating the urine sample before treatment to obtain peptide fragments;
[0086] (3) performing proteomic analysis on the peptide segments to obtain a differential protein expression matrix between the relapse group and the non-relapse group;
[0087] (4) Screening of candidate biomarkers;
[0088] (5) Mass spectrometry-based targeted validation of candidate biomarkers;
[0089] (6) Analyze the correlation between candidate biomarkers and recurrence-free survival time, and use LASSO regression and high and low levels of biomarkers to construct a prediction model for recurrence-free survival in patients with late-stage high-grade serous ovarian cancer.
[0090] The present invention proposes the idea of "mining recurrence-free survival-related biomarkers based on urine samples combined with proteomics technology, and constructing a 12-month recurrence-free survival prediction model for patients with advanced HGSOC". Body fluid samples have always been the most commonly used biological samples for clinical medical testing. Among them, urine samples have the advantages of easy acquisition, easy preservation, good patient compliance, low cost, and suitability for long-term intensive monitoring. At the same time, urine samples are rich in proteins, small molecule metabolites, etc. Therefore, the present invention selects urine samples of patients with advanced HGSOC as the research objects.
[0091] The proteomics technology of mass spectrometry analysis is booming. With the continuous iteration of proteomics technology, the detection speed is constantly accelerating, and the detection cost is also constantly decreasing. Based on the above advantages, the present invention selects proteomics technology based on mass spectrometry analysis for 12-month recurrence-free survival biomarker mining of urine samples of patients with advanced HGSOC. Finally, the L1 regularization algorithm of Lasso regression is used to screen biomarkers and construct a prediction model.
[0092] The present invention is applicable to conventional HGSOC patients in the late stage, and is also applicable to patients with metal implants or foreign bodies and patients who are allergic to contrast agents; the urine sample involved in the present invention is easy to obtain and preserve, and is applicable to long-term intensive monitoring.
[0093] The technical solution of the present invention includes 8 parts: research ethics, experimental design principles, urine sample collection, urine sample pretreatment, mass spectrometry-based urine proteomics analysis, screening of candidate biomarkers, mass spectrometry-based targeted verification of candidate biomarkers, correlation analysis and model construction.
[0094] Part I: Research Ethics
[0095] The urine sample research involved in the present invention was approved by the Medical Ethics Committee of Zhejiang Cancer Hospital and conducted in accordance with the Declaration of Helsinki. All patients provided written informed consent to participate in this study.
[0096] Part II: Principles of Experimental Design
[0097] (1) The patients with advanced HGSOC (III and IV) included in the study had no history of other malignancies; (2) After the urine sample was collected, the patients completed the initial treatment (tumor reduction surgery or neoadjuvant chemotherapy combined with intermittent tumor reduction surgery); (3) The patients' clinical information was collected and recorded, especially the specific time of recurrence-free survival, and the patients were divided into recurrence group and non-recurrence group with 12 months as the boundary; (4) The data-independent acquisition (DIA) mode of mass spectrometry was used for proteomic analysis of urine samples to screen candidate biomarkers; (5) The parallel reaction monitoring (PRM) mode of mass spectrometry was used for targeted validation of candidate biomarkers; (6) The object of targeted validation was the characteristic peptide segment of each candidate biomarker; (7) The abundance of the characteristic peptide segment was used to characterize the content of the candidate biomarker.
[0098] Part III: Urine Sample Collection and Grouping
[0099] Urine samples were collected before patients with advanced HGSOC (stages III and IV) began treatment. The collection process is as follows: collect 2 mL of mid-morning urine into a sample tube and centrifuge at 4°C and 1000g for 5 minutes; then take 1 mL of supernatant and transfer it to a new sample tube and store it in a -80°C refrigerator. A total of 56 urine samples were collected in the present invention, including 31 in the recurrence group and 25 in the non-recurrence group. In order to improve the rationality of biomarker mining and verification, the present invention further grouped the above 56 urine samples into a discovery set and a validation set. Among them, the discovery set contains 16 urine samples from patients without recurrence and 15 urine samples from patients with recurrence, and the validation set contains 9 urine samples from patients without recurrence and 16 urine samples from patients with recurrence.
[0100] Part IV: Urine sample pretreatment
[0101] Take 200 μL of urine sample thawed at room temperature, add 800 μL of pre-cooled acetone solution, mix thoroughly and let stand at 4°C for 12 hours. Subsequently, centrifuge at 4°C and 10,000g for 10 minutes, discard the supernatant, and dissolve the protein precipitate with 8M urea solution. The protein concentration was determined using a BCA kit. Take 50 μg of protein for conventional alkylation treatment, and add 1 μg of trypsin solution for enzyme digestion at 37°C overnight to cut the protease into free peptides. Subsequently, the peptides were desalted using a commercial C18 column, and the peptides were dried using a freeze dryer. Finally, the resulting peptide powder was stored in a -80°C refrigerator. Using the above method, the present invention will discover and verify that the proteases in the urine samples of the collection are cut into peptides.
[0102] Part V: Mass spectrometry-based urine proteomics analysis
[0103] The present invention uses the "nanofluidic liquid combined with Q-Exactive high-resolution mass spectrometer" system of Thermo Fisher Scientific, USA to perform proteomics analysis on the peptides of the above discovery set. Take 30 μL of formic acid aqueous solution to dissolve the peptide powder, then take 3 μL of the peptide solution into the nanofluidic system for chromatographic separation, and finally spray into the Q-Exactive high-resolution mass spectrometer for signal acquisition and analysis. The main parameters of the nanofluidic phase are as follows: the nanofluidic phase body includes a pre-column (length 2 cm, inner diameter 75 μm, filler is 5 μm C18 filler), an analytical column (length 25 cm, inner diameter 75 μm, filler is 2 μm C18 filler), mobile phase A (containing 2% acetonitrile, 0.1% formic acid, 98% water) and mobile phase B (98% acetonitrile, 0.1% formic acid, 2% water). The peptide solution is first enriched in the pre-column, and then eluted into the analytical column under the ratio control of mobile phase A and B, and chromatographic separation is performed. Finally, the peptides reached the needle tip at the end of the analytical column and were ionized at 2100 V to enter the mass spectrometer for analysis. The main parameters of mass spectrometry analysis are as follows: (1) The mass spectrometer was detected and analyzed in DIA mode; (2) the primary spectrum resolution was 35,000, and the maximum ion injection time was 20 ms; (3) the secondary spectrum resolution was 17,500, the fragmentation energy was 27, the maximum ion injection time was set to automatic, the detection range was 380-980 Da, the mass window was 20 Da, and there were 30 windows in total. Finally, the human FASTA file of the Uniprot database and the DIA-NN software (default parameters) were used to analyze the raw mass spectrometry data, and statistical analysis was performed to obtain the differential protein expression matrix between the recurrence group and the non-recurrence group in the urine samples.
[0104] Part VI: Screening of candidate biomarkers
[0105] The screening of candidate biomarkers of the present invention follows the following principles: (1) Compared with the non-recurrence group, the candidate biomarker has a higher level in the recurrence group, log2 (recurrence group / non-recurrence group) ≥ 1, that is, a two-fold difference. Because upregulated high-level proteins are more suitable as biomarkers than low-level proteins; (2) Candidate biomarkers have characteristic peptide segments and have mass spectrometry peaks that are significantly different from background signals in the mass spectrometry; (3) The characteristic peptide segments of the candidate biomarkers are not post-translationally modified in urine samples. Because post-translational modifications usually affect the accuracy of proteins in mass spectrometry quantification. Based on the above principles, the present invention screens the protein expression matrix of the discovery set urine samples, that is, the top 10 upregulated proteins in the recurrence group of the discovery set are selected as candidate biomarkers.
[0106] Part VII: Mass spectrometry-based targeted validation of candidate biomarkers
[0107] This section includes two steps: (1) establishing a targeted validation method; and (2) targeted validation of candidate biomarkers.
[0108] (1) Establish a targeted validation method: Import the mass spectrometry raw data and DIA-NN analysis results obtained in Part 5 into the Skyline software to select the characteristic peptides of the candidate biomarkers. The characteristic peptides meet the screening principles in Part 6 and contain clear amino acid sequence information, mass-to-charge ratio, charge, and chromatographic retention time. Then export the information of the characteristic peptides and the mass spectrometry targeted validation method.
[0109] (2) Targeted verification of candidate biomarkers: The mass spectrometry acquisition parameters were set according to the targeted verification method established above, as follows: the mass spectrometry was performed in PRM mode for targeted analysis; the primary spectrum resolution was 35,000, and the maximum ion injection time was 20 ms; the secondary spectrum resolution was 17,500, the fragmentation energy was 27, and the maximum ion injection time was 100 ms. The chromatographic conditions were consistent with those in Part 5. Subsequently, the raw data collected by the mass spectrometry targeted verification was re-imported into the Skyline software for analysis, and the levels of candidate biomarkers in the discovery set and validation set were statistically obtained. Finally, the discovery set and validation set data were merged, and the levels of each candidate biomarker were classified into two levels, high and low, according to the cutoff value.
[0110] Part 8: Correlation Analysis and Model Building
[0111] This part first analyzed the correlation between candidate biomarkers and recurrence-free survival time, clarified the clinical value of candidate biomarkers and their potential value as biomarkers to predict recurrence-free survival. Finally, a prediction model for recurrence-free survival in advanced HGSOC patients was constructed using LASSO regression and high and low levels of biomarkers.
[0112] (1) Correlation analysis: The present invention uses the survival analysis module of GraphPad Prism 9.0 software to analyze the correlation between candidate biomarkers and recurrence-free survival time, and the parameters are default parameters.
[0113] (2) Model construction: The present invention uses the LASSO regression method in R language to construct the model, and the main parameters are as follows: first, the cv.glmnet function (nfolds = 5) is used to optimize the important parameter lambda in the LASSO regression model to prevent the model from overfitting; then the optimized lambda value is called in combination with the glmnet function to calculate the variable correlation coefficient, screen the variables, fit the regression formula, and build a prediction model; finally, the predictive accuracy of the model is tested using the predict function.
[0114] The present invention constructs a recurrence-free survival prediction model for HGSOC advanced patients based on the above correlation analysis and LASSO regression.
[0115] The present invention is implemented according to the first to fifth parts of the technical solution, and statistical analysis is performed to obtain a differential protein expression matrix (p value <0.01, |log2(difference fold)|≥1) between the non-recurrence group (n=16) and the recurrence group (n=15) in the urine samples collected. Figure 1 As shown. According to proteomic analysis, the present invention found a total of 51 differential proteins. Compared with the urine samples of patients in the non-recurrence group, 39 proteins were upregulated in the urine samples of patients in the recurrence group, and another 12 proteins were downregulated in the urine samples of patients in the recurrence group. According to the experimental design principles and differential protein matrix of the second part of the technical solution, the present invention defines the range of candidate biomarkers for recurrence-free survival to the above 39 proteins.
[0116] According to the above 39 proteins and the principle of candidate biomarker screening in Part VI of the technical solution, the present invention selects the top 10 upregulated proteins in the recurrence group of the discovery set as candidate biomarkers. They are ANXA1, G6PI, ELAF, SPB3, S10A7, SPB4, FABP5, SPRR3, CE350 and MUC5B, as follows Figure 2 Subsequently, the Skyline software described in the seventh part of the technical solution is used in the present invention to select characteristic peptide segments of candidate biomarkers, including clear amino acid sequence information, mass-to-charge ratio, and chromatographic retention time, as shown in Table 1 below.
[0117] Table 1. Characteristic peptide information of candidate biomarkers
[0118]
[0119] The present invention carried out targeted validation of candidate biomarkers based on the above characteristic peptide information and the targeted validation method described in Part VII of the technical scheme. The results showed that 4 of the 10 candidate biomarkers (ANAX1, G6PI, SPB3 and SPRR3) could be well validated in both the discovery set and the validation set, as follows Figure 3 A and 3B. Subsequently, the present invention combined the levels of the four candidate biomarkers (the values were log2 transformed), as shown in Figure 3 As shown in C.
[0120] Subsequently, the present invention uses the Youden index to calculate the cutoff values of the four candidate biomarkers. The cutoff values of ANAX1, G6PI, SPB3 and SPRR3 are 20.34, 19.69, 20.79 and 24.50 respectively (all log2-converted values). The detailed data are shown in Table 2 below.
[0121] Table 2. ANAX1, G6PI, SPB3 and SPRR3 targeting validation results and their cutoff values
[0122]
[0123] SD: standard deviation; the values in the table are all log2-converted values
[0124] According to the eighth part of the technical solution, the present invention analyzed the correlation between ANAX1, G6PI, SPB3 and SPRR3 levels and recurrence-free survival time of patients with advanced HGSOC. The results are as follows Figure 4 The results showed that the above candidate biomarkers were correlated with the recurrence-free survival time of patients with advanced HGSOC, and the increase in the levels of candidate biomarkers in urine samples was associated with a shorter recurrence time of patients (log-rank p<0.05).
[0125] The present invention constructs the model according to the eighth part of the technical solution, calculates the intercept and the correlation coefficient of each candidate biomarker in the model, and the results are shown in Table 3 below.
[0126] Table 3. Specific parameters of the prediction model
[0127]
[0128] From the results, it can be found that the coefficients of ANAX1, G6PI, SPB3 and SPRR3 in the model are 0.0000, 0.0872, 1.5501 and 0.5955, respectively. That is, LASSO regression finally selected G6PI, SPB3 and SPRR3 as biomarkers for 12-month recurrence-free survival in advanced HGSOC patients, and established a prediction model. The 12-month recurrence-free survival probability model for advanced HGSOC patients is:
[0129]
[0130] Where P is the probability; G6PI bi 、SPB3 bi and SPRR3 bi It is binary data, i.e. 0 or 1. Note: The content data can be divided into high (1) or low (0) by comparing with the cutoff value in Table 2.
[0131] According to the eighth part of the technical solution, the present invention uses the R language predict function to predict the 12-month recurrence-free survival probability of 56 patients with advanced HGSOC to evaluate the prediction accuracy of the model. The results are as follows Figure 5 As shown in the figure, it can be found that the prediction accuracy of the model is 87.2%.
[0132] Example: E l i sa kit
[0133] 1. Kit Composition
[0134] 1. Capture antibodies: Specific capture antibodies against G6PI, SPB3, SPRR3 and ANAX1, which are optionally coated in different wells of a 96-well plate.
[0135] 2. Detection antibodies: Biotinylated detection antibodies against G6PI, SPB3, SPRR3 and ANAX1.
[0136] 3. Enzyme conjugate: Horseradish peroxidase (HRP) conjugated to streptavidin (Streptavidin-HRP), used to bind to the biotinylated detection antibody.
[0137] 4. Standards: Recombinant protein standards of G6PI, SPB3, SPRR3 and ANAX1, used to draw the standard curve.
[0138] 5. Sample diluent: buffer used to dilute urine samples.
[0139] 6. Washing solution: 10X washing solution, used to clean the microplate.
[0140] 7. Substrate solution: TMB substrate solution, used for color development with HRP reaction.
[0141] 8. Stop solution: 2M sulfuric acid solution, used to stop the enzyme reaction.
[0142] 9.96-well plate: high-binding ELISA plate pre-coated with capture antibody.
[0143] 10. Sealing film: used to seal the 96-well plate during the incubation step to prevent evaporation and contamination.
[0144] 2. Operation steps
[0145] 1. Sample processing: Collect the mid-morning urine sample, centrifuge it and take the supernatant; dilute the urine sample with sample diluent at a ratio of 1:10.
[0146] 2. Preparation of standard curve: Use the standard solution to prepare a series of standard solutions with a concentration gradient (e.g. 0, 0.1, 0.5, 1, 5, 10 ng / mL) by doubling dilution.
[0147] 3. Sample addition: Add 100 μL of diluted standard solution or urine sample to each well of a 96-well plate; incubate for 1 hour at 37°C.
[0148] 4. Washing: Add 300 μL of washing solution to each well, let it stand for 1 minute and then discard it. Repeat the washing process 3 times.
[0149] 5. Detection antibody: Add 100 μL of biotinylated detection antibody solution to each well and incubate for 1 hour at 37°C; repeat washing step 4.
[0150] 6. Enzyme conjugate: Add 100 μL HRP-conjugated streptavidin to each well and incubate for 30 minutes at 37°C; repeat washing step 4.
[0151] 7. Substrate color development: Add 100 μL TMB substrate solution to each well, incubate for 15-30 minutes, and protect from light; when the color development in the standard wells reaches the ideal intensity, add 50 μL stop solution to each well.
[0152] 8. Plate reading: Read the absorbance value (OD value) of each well using a microplate reader at a wavelength of 450 nm.
[0153] 9. Data analysis: Calculate the concentrations of G6PI, SPB3, SPRR3, and ANAX1 in the samples using the standard curve.
[0154] 3. Interpretation of results
[0155] 1. Standard curve: Draw a standard curve based on the OD value of the standard, and calculate the concentration of the target protein in the sample using the standard curve equation.
[0156] 2. Sample results: According to the OD value of the sample, check the standard curve or calculation formula to obtain the specific concentrations of G6PI, SPB3, SPRR3 and ANAX1.
[0157] 3. Data verification: Verify the reliability and repeatability of the data through repeated experiments and negative and positive controls.
[0158] The kit of this embodiment can be used in the following aspects:
[0159] 1. Clinical diagnosis: Detection of G6PI, SPB3, SPRR3 and ANAX1 levels in urine of patients with advanced high-grade serous ovarian cancer for prediction of recurrence-free survival.
[0160] 2. Scientific research: used for basic research and new drug development of biomarkers for high-grade serous ovarian cancer.
[0161] 3. Personalized medicine: Develop personalized treatment plans and follow-up plans for patients based on test results.
[0162] The Elisa kit of this example provides an efficient, accurate and convenient detection method for predicting recurrence-free survival of patients with advanced high-grade serous ovarian cancer, and has broad clinical application prospects.
[0163] Example: Computer system and computer program for predicting recurrence-free survival in patients with advanced high-grade serous ovarian cancer Preparation
[0164] 1. Computer system composition
[0165] 1. Data acquisition module: used to obtain the biomarker combination expression spectrum data in the urine sample of the test subject, including the concentration data of G6PI, SPB3, SPRR3 and ANAX1. It includes the following hardware and software.
[0166] Hardware: Equipped with a mass spectrometer or ELISA detector and connected to a computer system.
[0167] Software: Data acquisition software used to read and store data from a mass spectrometer or detector.
[0168] 2. Prediction module: used to input the biomarker expression profile data obtained by the data acquisition module into the trained prediction model to predict recurrence-free survival.
[0169] Hardware: High-performance computing servers or cloud computing platforms.
[0170] Software: Predictive model implementation software.
[0171] Algorithm: A LASSO regression model based on a combination of biomarkers was used, which was trained and validated to accurately predict relapse-free survival.
[0172] 3. Prediction result acquisition module: used to obtain the output results of the prediction model in the prediction module and obtain the recurrence-free survival prediction results of the subjects.
[0173] Hardware: A computer monitor or mobile device to display the prediction results.
[0174] Software: Results display and analysis software that can generate reports and charts.
[0175] 2. System Workflow
[0176] 1. Sample Collection and Data Acquisition
[0177] By collecting urine samples from subjects, using the mass spectrometer or ELISA in the data acquisition module
[0178] The detector measures the concentrations of G6P I, SPB3, SPRR3 and ANAX1.
[0179] Input the concentration data into the data acquisition software for preliminary processing and storage.
[0180] 2. Data input and model prediction
[0181] The data acquisition software inputs the processed biomarker expression profile data into the prediction model in the prediction module.
[0182] The prediction model processes the input data and uses the trained LASSO regression model to calculate and predict the recurrence-free survival of the subjects.
[0183] 3. Result output and analysis
[0184] The prediction result acquisition module acquires the prediction result from the prediction model.
[0185] The results display and analysis software generates detailed prediction reports, including biomarker concentrations, predicted recurrence-free survival, risk assessment charts, etc.
[0186] Display the results on a computer monitor or mobile device for reference by doctors and patients.
[0187] The computer equipment consists of the following:
[0188] 1. Memory: stores programs for implementing the above method, including modules for data collection, prediction model processing and result display.
[0189] 2. Processor: Executes the program in the memory to implement the following steps:
[0190] Obtaining biomarker combination expression profile data in the sample of the subject to be tested;
[0191] providing the biomarker combination expression profile data as input data to the trained prediction model; and,
[0192] Output the recurrence-free survival prediction results of the tested subjects.
[0193] The computer system or device of this embodiment is suitable for conventional HGSOC patients in the late stage, as well as patients with metal implants or allergic to contrast agents, because samples are easy to obtain, and urine samples are used, which are convenient for collection and preservation and suitable for long-term monitoring. In addition, the LASSO regression model based on the present invention has been fully trained and verified, and the prediction results are accurate and reliable.
[0194] It should be noted that the above embodiments can be freely combined as needed. The above description is only the preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for establishing a prediction model for recurrence-free survival in patients with advanced high-grade serous ovarian cancer, characterized in that: The steps include: (1) Urine samples were obtained from patients with advanced high-grade serous ovarian cancer before treatment, and the patients were divided into recurrence group and non-recurrence group based on the 12-month mark after initial treatment; (2) pre-treating the urine sample before treatment to obtain peptide fragments; (3) performing proteomic analysis on the peptide segments to obtain a differential protein expression matrix between the relapse group and the non-relapse group; (4) Screening of candidate biomarkers; (5) Mass spectrometry-based targeted validation of candidate biomarkers; (6) Analyze the correlation between candidate biomarkers and recurrence-free survival time, and use LASSO regression and high and low levels of biomarkers to construct a prediction model for recurrence-free survival in patients with late-stage high-grade serous ovarian cancer.
2. The method for establishing a prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer according to claim 1, characterized in that: The patient with advanced high-grade serous ovarian cancer has no history of other malignant tumors; the urine sample is the supernatant obtained by centrifugation of the mid-morning urine; Step (2) comprises the following steps: taking a urine sample and mixing it with a pre-cooled acetone solution, standing it and then centrifuging it, discarding the supernatant, and dissolving the protein precipitate with a urea solution; determining the protein concentration; alkylating the protein, adding a trypsin solution to cut the protein into free peptides, and desalting and drying the peptides to obtain peptide powder.
3. The method for establishing a prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer according to claim 1, characterized in that: Step (3) specifically includes the following steps: dissolving the peptide powder with a formic acid aqueous solution, then taking the peptide solution into a nanofluidic system for chromatographic separation, and finally spraying it into a high-resolution mass spectrometer for signal acquisition and analysis.
4. The method for establishing a prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer according to claim 3, characterized in that: The main parameters of nanofluidic phase are as follows: One precolumn: 2 cm in length, 75 μm in inner diameter, filled with 5 μm C18 filler; One analytical column: 25 cm in length, 75 μm in inner diameter, filled with 2 μm C18 filler; Mobile phase A: 2% acetonitrile, 0.1% formic acid, 98% water; Mobile phase B: 98% acetonitrile, 0.1% formic acid, 2% water; The peptide solution is first enriched in the pre-column, then eluted into the analytical column under the control of the ratio of mobile phase A and B, and chromatographic separation is performed; finally, the peptide reaches the needle tip at the end of the analytical column and is ionized and sprayed into the mass spectrometer for analysis at a voltage of 2100V.
5. The method for establishing a prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer according to claim 4, characterized in that: The main parameters of mass spectrometry analysis are as follows: mass spectrometry was performed in DIA mode; primary spectrum resolution was 35000, and the maximum ion injection time was 20 ms; (3) secondary spectrum resolution was 17500, fragmentation energy was 27, maximum ion injection time was set to automatic, detection range was 380-980 Da, mass window was 20 Da, and there were 30 windows in total; The human FASTA files of the Uniprot database and the DIA-NN software were used to analyze the raw mass spectrometry data, and the differential protein expression matrix between the recurrence group and the non-recurrence group in the urine samples of the discovery set was obtained by statistical analysis.
6. The method for establishing a prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer according to claim 5, characterized in that: Step (4) specifically includes the following steps: selecting the top 10 upregulated proteins in the recurrence group of the discovery set as candidate biomarkers; the screening of candidate biomarkers follows the following principles: (1) the candidate biomarker has a higher level in the recurrence group compared with the non-recurrence group; (2) the candidate biomarker has a characteristic peptide segment and a high-responsiveness signal in the mass spectrometry; (3) the characteristic peptide segment of the candidate biomarker has no post-translational modification in the urine sample.
7. The method for establishing a prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer according to claim 6, characterized in that: Step (5) specifically includes the following steps: A. Establish a targeted validation method: import the mass spectrometry raw data and DIA-NN analysis results into Skyline software to select the characteristic peptides of candidate biomarkers; The characteristic peptides conform to the screening principles of candidate biomarkers and contain clear information on amino acid sequence, mass-to-charge ratio, charge, and chromatographic retention time; Methods for exporting characteristic peptide information and mass spectrometry targeted validation; B. Targeted validation of candidate biomarkers: The mass spectrometry acquisition parameters were set according to the targeted validation method established above, as follows: the mass spectrometry was performed in PRM mode for targeted analysis; the primary spectrum resolution was 35,000, and the maximum ion injection time was 20 ms; the secondary spectrum resolution was 17,500, the fragmentation energy was 27, and the maximum ion injection time was 100 ms; the chromatographic conditions were consistent with step (3); the raw data collected by the mass spectrometry targeted validation was re-imported into the Skyline software for analysis, and the levels of candidate biomarkers in the discovery set and validation set were statistically obtained; finally, the discovery set and validation set data were merged, and the levels of each candidate biomarker were classified into two levels, high and low, according to the cutoff value.
8. The method for establishing a prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer according to claim 7, characterized in that: Step (6) specifically includes the following steps: first, analyzing the correlation between the candidate biomarker and the recurrence-free survival time, clarifying the clinical value of the candidate biomarker and its potential value as a biomarker to predict recurrence-free survival; finally, constructing a prediction model for recurrence-free survival of patients with advanced HGSOC using LASSO regression and the high and low levels of the biomarker; A. Correlation analysis: The present invention uses the survival analysis module of GraphPad Prism 9.0 software to analyze the correlation between candidate biomarkers and recurrence-free survival time, and the parameters are default parameters; B. Model construction: The present invention uses the LASSO regression method in R language to construct the model, and the main parameters are as follows: first, the cv.glmnet function (nfolds=5) is used to optimize the important parameter lambda in the LASSO regression model to prevent the model from overfitting; then the optimized lambda value is called in combination with the glmnet function to calculate the variable correlation coefficient, screen the variables, fit the regression formula, and build a prediction model; finally, the predictive accuracy of the model is tested using the predict function.
9. A prediction model for recurrence-free survival in patients with advanced high-grade serous ovarian cancer, characterized in that: The method is established by any one of claims 1 to 8.
10. The prediction model for recurrence-free survival of patients with advanced high-grade serous ovarian cancer according to claim 9, characterized in that: G6PI, SPB3 and SPRR3 were used as biomarkers to establish a prediction model: Where: P is probability; G6PI bi 、SPB3 bi and SPRR3 bi It is binary data, that is, 0 or 1; the content data is compared with the cutoff value, high is 1, and low is 0.
11. A biomarker or a combination thereof for predicting and diagnosing recurrence-free survival in patients with advanced high-grade serous ovarian cancer, wherein the biomarker combination comprises any one, any two, any three or all of G6PI, SPB3, SPRR3 and ANAX1.
12. A kit comprising reagents for detecting the level of the biomarker combination of claim 11, preferably, the reagents are antibodies to the markers, primers for PCR, reagents for mass spectrometry analysis or reagents for chromatography analysis.
13. Use of a reagent for detecting the biomarker defined in claim 11 in the preparation of a kit or a detection device for predicting recurrence-free survival in patients with advanced high-grade serous ovarian cancer.
14. A device or system for predicting recurrence-free survival in patients with advanced high-grade serous ovarian cancer, comprising: A data acquisition module, used to acquire the biomarker combination expression profile data defined in claim 11 in the sample of the subject to be tested; A prediction module, used to provide the gene group expression profile data obtained by the data acquisition module as input data to a trained prediction model, wherein the prediction model predicts recurrence-free survival based on the biomarker combination expression profile data of the subject, preferably, the prediction model is the prediction model defined in claim 10; and The prediction result acquisition module is used to obtain the output result of the prediction model in the prediction module to obtain the recurrence-free survival prediction result of the subject.
15. A computer device, comprising a memory and a processor, wherein a program is stored in the memory, and when the processor executes the program, the following method is implemented: obtaining the biomarker combination expression spectrum data defined in claim 11 in the sample of the subject to be tested; providing the biomarker combination expression spectrum data as input data to a trained prediction model; and outputting the recurrence-free survival prediction result of the subject to be tested, preferably, the prediction model is the prediction model defined in claim 10.
16. A computer-readable storage medium having a computer program stored thereon, which, when executed by a computer, performs the method defined in claim 15.
17. Use of the biomarker according to claim 11 in predicting recurrence-free survival in patients with advanced high-grade serous ovarian cancer for non-disease diagnosis or treatment purposes.
18. Use of any one of G6PI, SPB3, SPRR3 and ANAX1, or any two, three or four of them as biomarkers in the preparation of a preparation for predicting recurrence-free survival in patients with advanced high-grade serous ovarian cancer.
19. A method for evaluating whether a drug can improve recurrence-free survival in patients with advanced high-grade serous ovarian cancer, comprising using any one of G6PI, SPB3, SPRR3 and ANAX1, or any two, three or four of them as biomarkers to evaluate the effect of the drug.
20. A method for screening a compound capable of improving recurrence-free survival in patients with advanced high-grade serous ovarian cancer, comprising using any one of G6PI, SPB3, SPRR3 and ANAX1, or any two, three or four thereof as biomarkers to evaluate the effect of the compound.
Citation Information
Patent Citations
Compositions, methods and kits for diagnosis of lung cancer
CA2860298A1
Ovarian cancer urine marker combination and application thereof
CN113391072A
TC disease risk prediction modeling method based on cell proliferation markers
CN115705928A
Personalized treatment decision-making method and system for intestinal cancer and storage medium comprising personalized treatment decision-making method and system
CN115954045A
Prostate cancer prognosis prediction model construction method and device and storage medium
CN116741378A
