Protein markers for predicting sensitivity of adjuvant chemotherapy in ovarian cancer patients and use thereof

Abstract

The application provides a protein marker for predicting the sensitivity of adjuvant chemotherapy of an ovarian cancer patient and an application, based on proteomics data of an ovarian cancer tissue sample, combining a preliminary screening model and a classification model and PRM technology, screening a plurality of protein markers with representative meanings for predicting the sensitivity of adjuvant chemotherapy of an ovarian cancer patient after neoadjuvant therapy, training an ovarian cancer patient chemotherapy sensitivity prediction model by using the screened protein markers, predicting the chemotherapy sensitivity of an ovarian cancer patient who has not undergone adjuvant therapy by using the ovarian cancer patient chemotherapy sensitivity prediction model, if the chemotherapy sensitivity is low, it means that the probability of recurrence of the patient after 6 months of adjuvant therapy is high, so medical staff can adjust the treatment plan accordingly, avoid unnecessary chemotherapy treatment for ovarian cancer patients, and play a preventive effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of biological medicine, and particularly relates to a protein marker for predicting the sensitivity of adjuvant chemotherapy of ovarian cancer patients after neoadjuvant therapy and application. BACKGROUND

[0002] Epithelial ovarian cancer is one of the most common and deadly gynecological cancers, and its high mortality rate is mainly due to late diagnosis and acquired drug resistance of patients to chemotherapy drugs. High-grade serous ovarian cancer is the most common histological subtype, and about 75% of patients are in advanced stage at diagnosis. For patients who are unlikely to achieve optimal cytoreduction or are not suitable for surgery, neoadjuvant chemotherapy can be considered, which is systemic chemotherapy before local treatment methods such as surgery or radiotherapy, aiming to reduce the tumor size and kill invisible metastatic cells early to facilitate subsequent surgery, radiotherapy and other treatments. Platinum-based chemotherapy after cytoreductive surgery is still the standard treatment for patients with advanced EOC.

[0003] However, in the past 30 years, the overall survival of patients with high-grade serous ovarian cancer (HGSC) has hardly improved, and there has been no progress in standard treatment other than platinum-based combination chemotherapy. Currently, some targeted treatment methods such as bevacizumab and / or PARP inhibitors are mainly used as first-line maintenance therapy and single-agent treatment for recurrence, but the treatment effects of different therapies are not good, and the main reason is that 75% of patients with advanced disease still relapse and eventually die of chemotherapy resistance. In particular, the survival rate of patients with advanced high-grade serous ovarian cancer after neoadjuvant chemotherapy has hardly improved, partly due to the lack of a classifier for predicting the response of ovarian cancer to chemotherapy. Therefore, early identification of whether high-grade serous ovarian cancer patients are platinum-resistant after neoadjuvant chemotherapy is beneficial to guide clinicians to adjust the postoperative adjuvant chemotherapy regimen as soon as possible, and has great significance for improving the prognosis of patients.

[0004] For platinum-based chemotherapy after cytoreductive surgery, drug resistance is mainly directed at platinum resistance; generally, patients with platinum resistance are considered to have a recurrence-free survival (RFS) of less than 6 months after platinum-based treatment, and patients with platinum sensitivity are considered to have a RFS of more than 6 months. The current prediction method is based on the clinical diagnostic biomarker CA125 or clinical imaging technology to predict the efficacy of neoadjuvant chemotherapy. For example, CN114846158A provides a series of markers for predicting the prognosis of neoadjuvant chemotherapy, but there is still no mature technology to indicate whether patients have drug resistance after neoadjuvant chemotherapy and whether they will relapse in a short period of time. If it can be predicted that some patients have strong platinum resistance, the adjuvant chemotherapy strategy can be adjusted after neoadjuvant chemotherapy to improve the prognosis of patients and reduce the pain caused by unreasonable chemotherapy to patients.

[0005] In summary, although there are related technologies on the market that can predict the prognosis of neoadjuvant chemotherapy, it is not enough for ovarian cancer patients to only predict the prognosis. The chemotherapy sensitivity or drug resistance of different patients will also affect the effect of neoadjuvant chemotherapy. At present, there is no related technology that can predict the chemotherapy sensitivity of ovarian cancer patients after neoadjuvant chemotherapy. SUMMARY

[0006] The purpose of the present application is to provide a protein marker for predicting the sensitivity of ovarian cancer patients to adjuvant chemotherapy and an application. A combination of 40 protein markers, preferably a combination of 6 protein markers, is screened for predicting whether an ovarian cancer individual who has not started adjuvant chemotherapy will relapse within a short time after receiving neoadjuvant chemotherapy. The chemotherapy sensitivity of the patient is evaluated and predicted in a timely manner, which is beneficial for more targeted treatment and medication after surgical tumor reduction, and is of great significance for improving the prognosis and survival of patients. It can also be used as a potential therapeutic target for subsequent research.

[0007] The present scheme can detect proteins in the ovarian cancer tissue sample of the subject, obtain the relative expression amounts of 40 specific protein markers, or obtain the relative expression amounts of specific 6 specific protein markers, and predict the probability of the subject relapsing within a short time after adjuvant therapy based on the relative expression amounts. In the embodiments of the present scheme, it can be predicted whether the subject will relapse within 6 months after neoadjuvant therapy. Generally, the lower the chemotherapy sensitivity of the subject, the worse the effect of the subject receiving neoadjuvant chemotherapy, and the higher the probability of subsequent recurrence. For ovarian cancer subjects with a high probability of short-term recurrence, early changes in diagnosis and treatment strategies, other targeted treatments and medications can be used to improve the prognosis and survival of patients.

[0008] It is particularly pointed out that the present scheme is used to predict the sensitivity of ovarian cancer patients to adjuvant chemotherapy after neoadjuvant therapy. All patients are first treated with neoadjuvant chemotherapy, then surgery is performed to obtain samples for proteomic analysis, then adjuvant chemotherapy is performed, and then follow-up is performed to obtain the time from the start of the last chemotherapy to recurrence to determine drug resistance and sensitivity.

[0009] In a first aspect, the present application provides a protein marker for predicting the sensitivity of adjuvant chemotherapy of an ovarian cancer patient, comprising: a protein marker selected from ovarian cancer tissue samples, wherein the protein marker is selected from one or a combination of IFRD1, UQCR11, SART1, KALRN, TRIM3, BANF1, IGLV3-19, HLA-DRB1, ACYP1, CD48, CKB, HPGD, DES, PZP, GGCX, SLC16A1, KHDRBS1, DGKZ, LRRC41, TOMM34, KIAA1755, MAGI3, ZNF787, TCTN3, ZGPAT, ZCRB1, URGCP, GPKOW, RBMXL1, EYA3, CMSS1, POLR1E, METTL9, LDAH, CTNNBIP1, CGGBP1, PTBP2, NIP7, ST3GAL6 and WDR45, or a combination thereof, and the protein marker is used to predict the probability of recurrence within a short time after neoadjuvant therapy of the ovarian cancer patient. Preferably, the protein marker is selected from a combination of 6 proteins, wherein the protein marker is preferably selected from one or a combination of RBMXL1, DES, MCT1, SART1, GPKOW and PTBP2.

[0010] At this time, the results of the protein marker for predicting the reduction effect of adjuvant therapy and the probability of recurrence within a short time of the ovarian cancer patient are more accurate.

[0011] Preferably, the ovarian cancer tissue sample of the present application is a high-grade serous ovarian cancer tissue sample.

[0012] In a preferred embodiment, the protein marker predicts whether the subject will have a recurrence within 6 months after adjuvant therapy. It should be noted that the protein marker provided by the present application is used to predict the probability of recurrence within 6 months after adjuvant therapy, and is used in a non-diagnostic manner in the clinic. That is, when the probability of recurrence is high, the ovarian cancer patient can be intervened by artificially changing the treatment strategy.

[0013] The present application builds a corresponding prediction model for the sensitivity of adjuvant chemotherapy of an ovarian cancer patient based on the above protein marker, inputs the relative expression amount of the protein marker in the ovarian cancer tissue sample of the subject into the trained prediction model for the sensitivity of adjuvant chemotherapy of an ovarian cancer patient, and obtains the sensitivity of adjuvant chemotherapy. The sensitivity of adjuvant chemotherapy is negatively correlated with the probability of recurrence within a short time after receiving neoadjuvant therapy.

[0014] In a second aspect, the present application provides a kit for predicting the sensitivity of a patient with ovarian cancer to adjuvant chemotherapy, comprising: reagents for detecting the content of protein markers in a sample of ovarian cancer tissue from the subject, wherein the protein markers are selected from one or more of IFRD1, UQCR11, SART1, KALRN, TRIM3, BANF1, IGLV3-19, HLA-DRB1, ACYP1, CD48, CKB, HPGD, DES, PZP, GGCX, SLC16A1, KHDRBS1, DGKZ, LRRC41, TOMM34, KIAA1755, MAGI3, ZNF787, TCTN3, ZGPAT, ZCRB1, URGCP, GPKOW, RBMXL1, EYA3, CMSS1, POLR1E, METTL9, LDAH, CTNNBIP1, CGGBP1, PTBP2, NIP7, ST3GAL6 and WDR45, or a combination thereof, and preferably from one or more of RBMXL1, DES, MCT1, SART1, GPKOW and PTBP2, or a combination thereof.

[0015] The method for using the kit is as follows: the relative expression amount of the protein markers in the sample of ovarian cancer tissue from the subject is input into the trained prediction model for the sensitivity of a patient with ovarian cancer to chemotherapy to obtain the sensitivity to chemotherapy, and the higher the sensitivity to chemotherapy of the subject, the more likely the subject is a sensitive patient and the lower the probability of recurrence in a short time.

[0016] Preferably, the protein markers are selected from a combination of 40 proteins, and preferably from a combination of 6 proteins, in which case the results of predicting the effect of adjuvant therapy on reducing the tumor size and the probability of recurrence in a short time of a patient with ovarian cancer are more accurate.

[0017] In a third aspect, the present application provides a protein marker for predicting the sensitivity of adjuvant chemotherapy of an ovarian cancer patient, comprising: detecting the relative expression of the protein marker in the ovarian cancer of the subject by a kit, and predicting the sensitivity of the ovarian cancer patient to the chemotherapy according to the relative expression of the protein marker combination, wherein the protein marker combination is selected from one or a combination of IFRD1, UQCR11, SART1, KALRN, TRIM3, BANF1, IGLV3-19, HLA-DRB1, ACYP1, CD48, CKB, HPGD, DES, PZP, GGCX, SLC16A1, KHDRBS1, DGKZ, LRRC41, TOMM34, KIAA1755, MAGI3, ZNF787, TCTN3, ZGPAT, ZCRB1, URGCP, GPKOW, RBMXL1, EYA3, CMSS1, POLR1E, METTL9, LDAH, CTNNBIP1, CGGBP1, PTBP2, NIP7, ST3GAL6 and WDR45, and preferably from one or a combination of RBMXL1, DES, MCT1, SART1, GPKOW and PTBP2.

[0018] Preferably, the protein marker is selected from a combination of 40 proteins, and preferably from a combination of 6 proteins, which can more accurately predict the effect of adjuvant therapy and the probability of recurrence in a short period of time.

[0019] In a fourth aspect, the present application provides a method for constructing a prediction model for the sensitivity of adjuvant chemotherapy of an ovarian cancer patient, comprising: training a machine learning model using the relative expression of the protein marker in the soft ovarian cancer tissue sample of the subject who has not yet received adjuvant therapy and will receive adjuvant therapy in the future as a training sample, wherein the protein marker combination is selected from one or a combination of IFRD1, UQCR11, SART1, KALRN, TRIM3, BANF1, IGLV3-19, HLA-DRB1, ACYP1, CD48, CKB, HPGD, DES, PZP, GGCX, SLC16A1, KHDRBS1, DGKZ, LRRC41, TOMM34, KIAA1755, MAGI3, ZNF787, TCTN3, ZGPAT, ZCRB1, URGCP, GPKOW, RBMXL1, EYA3, CMSS1, POLR1E, METTL9, LDAH, CTNNBIP1, CGGBP1, PTBP2, NIP7, ST3GAL6 and WDR45, and preferably from one or a combination of RBMXL1, DES, MCT1, SART1, GPKOW and PTBP2.

[0020] Preferably, the protein biomarker is a combination of 40 proteins, more preferably a combination of 6 proteins, in which case its prediction of the tumor reduction effect of adjuvant therapy for ovarian cancer patients and the probability of recurrence in a short period of time is more accurate.

[0021] Fifthly, this solution provides a model for predicting the sensitivity of ovarian cancer patients to adjuvant chemotherapy, which is constructed based on the aforementioned method for constructing a model for predicting the sensitivity of ovarian cancer patients to chemotherapy.

[0022] This patent, based on proteomics data from ovarian cancer tissue samples after neoadjuvant chemotherapy, combines initial screening and classification models with PRM technology to identify 40 representative protein biomarkers that predict the sensitivity of ovarian cancer patients to adjuvant chemotherapy after neoadjuvant chemotherapy. Six protein biomarkers are preferred. These protein biomarkers are used to train a chemotherapy sensitivity prediction model for ovarian cancer patients. This model is then used to predict the sensitivity of ovarian cancer patients who have not received adjuvant therapy. High chemotherapy sensitivity indicates a low probability of recurrence within 6 months after neoadjuvant therapy, allowing medical staff to adjust treatment plans accordingly and prevent unnecessary chemotherapy treatment for ovarian cancer patients, thus achieving a preventative effect. Attached Figure Description

[0023] Figure 1 These are the top 40 proteins selected based on the average decrease in accuracy.

[0024] Figure 2 Is using Figure 1 The diagram shows the AUC of 40 proteins in parallel reaction monitoring (PRM).

[0025] Figure 3 These are six proteins obtained through further screening.

[0026] Figure 4 Is using Figure 3 The diagram shows the AUC obtained from testing the six proteins. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0028] Example 1: The following describes the selection process of the predictive features for the adjuvant chemotherapy sensitivity prediction model in ovarian cancer patients after neoadjuvant chemotherapy:

[0029] Analysis of samples:

[0030] Training cohort: A total of 42 advanced high-grade serous subjects who had not yet started adjuvant therapy but would subsequently undergo adjuvant therapy after neoadjuvant therapy, of which 17 subjects relapsed within 6 months after the last adjuvant therapy, and these subjects were used as the experimental group, and the subjects in the experimental group belonged to the drug-resistant sample with low sensitivity to chemotherapy; 25 subjects relapsed after 6 months after the last adjuvant therapy, and these subjects were used as the control group, and the subjects in the control group belonged to the sensitive sample with high sensitivity to chemotherapy.

[0031] Validation cohort: A total of 29 advanced high-grade serous subjects who had not yet started adjuvant therapy but would subsequently undergo adjuvant therapy after neoadjuvant therapy, of which 15 subjects relapsed within 6 months after the last adjuvant therapy, and these subjects were used as the experimental group; 14 subjects relapsed after 6 months after the last adjuvant therapy, and these subjects were used as the control group.

[0032] Proteomics analysis was performed on the ovarian cancer tissue samples extracted from the subjects in the training cohort before adjuvant therapy. Experimental design was performed on the clinical cohort samples, and different clinical characteristics of the population were considered to make the samples as evenly distributed as possible to avoid batch effects to the greatest extent.

[0033] I. Proteomics analysis of ovarian cancer tissue samples from subjects:

[0034] 1) 0.5-1 mg of tissue was sequentially washed with 70% ethanol, 100% water, 70% ethanol, 85% ethanol, and 100% ethanol at 25°C for 5 minutes to remove the optimal cutting temperature embedding agent. The tissue sample was transferred to a PCT special pressure-resistant tube, and the tissue sample was lysed with 30 μL of lysis buffer (6 M urea, 2 M thiourea, 100 mM ammonium bicarbonate (ABB)) at 30°C using cyclic pressure technology (PCT) in a Barocycler (45000 psi, 90 cycles, each cycle with high pressure lasting for 25 seconds), and then reduced and alkylated with 10 mM tris(2-carboxyethyl)phosphine (TCEP) and 40 mM iodoacetamide (IAA), respectively. Intracellular proteases lysC (enzyme to substrate ratio 1:40) and trypsin (enzyme to substrate ratio 1:50) were added to the above mixture in two steps for PCT-assisted enzymolysis. Then, 1% trifluoroacetic acid (TFA) was used to adjust the pH to 2-3, and the digestion was terminated to obtain the tissue polypeptide sample of the ovarian cancer tissue sample.

[0035] 2) Nanoliter liquid chromatography-high resolution mass spectrometry analysis:

[0036] Prior to mass spectrometry analysis, polypeptides were cleaned up with a Nest Group C18 column. The reconstituted tissue polypeptide samples were injected into a nanflow DIONEX UltiMate 3000 RSLCnano system (Thermo Fisher Scientific, San Jose, USA) coupled with a QE-HF high-resolution mass spectrometer (Thermo Fisher Scientific, San Jose, USA) for PulseDIA analysis.

[0037] Briefly, 0.25 pg of polypeptide sample was first loaded onto a pre-column (3 pm, 20 mm*75 pm i.d.) at a speed of 6 ul / min, and then the sample loaded on the pre-column was flushed into an analytical column (1.9 pm, 120a, 150 mm*75 pm diameter) at a flow rate of 300 nL / min for further online separation, with a 5%-28% buffer B (buffer A: 2% ACN, 98% H2O (containing 0.1% FA), buffer B: 98% ACN (containing 0.1% FA)) gradient separation for 30 minutes.

[0038] The m / z range of MS1 was 390-1210, the resolution was 60,000, the AGC was 3e6, and the max IT was 80 ms. Each sample was injected twice, and data were collected using 2 complementary discontinuous isolation windows. The MS2 resolution was 30000, the AGC was 1e6, the max IT was 50 ms, and 10100 proteomic data were obtained.

[0039] II. Screening of preliminary protein markers using the primary screening model:

[0040] The 10100 proteomic data of 42 subjects in the training cohort were used as a data set to construct the corresponding machine learning model of the primary screening model to distinguish between sensitive patients with high chemotherapy sensitivity and drug-resistant patients with low chemotherapy sensitivity.

[0041] First, the drug-resistant group and the sensitive group in the training data set were subjected to Student's t-test, and the t-test standard was: p-value less than 0.05, and the protein abundance difference between groups was greater than 1.5-fold, resulting in 145 significantly different proteins between groups.

[0042] Subsequently, the protein quantification data of the 145 significantly different proteins between groups were used as input features of the primary screening model, and the R software package of random forest (version 4.6.14) was used to first screen proteins with an average reduction accuracy greater than 1.5, a total of 54 different proteins;

[0043] ​The 54 differential proteins were used as input features of the initial screening model. Further, 100 times 5-fold cross-validation (i.e., 34 samples in the training set and 8 samples in the validation set) was performed, the ntree of the initial screening model was set to 1000, and the proteins screened based on the above machine learning were ranked according to the average reduction in accuracy. The top 40 proteins were screened, and the top 40 proteins are shown in Table 1. Figure 1 The 40 proteins are IFRD1, UQCR11, SART1, KALRN, TRIM3, BANF1, IGLV3-19, HLA-DRB1, ACYP1, CD48, CKB, HPGD, DES, PZP, GGCX, SLC16A1, KHDRBS1, DGKZ, LRRC41, TOMM34, KIAA1755, MAGI3, ZNF787, TCTN3, ZGPAT, ZCRB1, URGCP, GPKOW, RBMXL1, EYA3, CMSS1, POLR1E, METTL9, LDAH, CTNNBIP1, CGGBP1, PTBP2, NIP7, ST3GAL6, and WDR45. The 40 proteins can be used to predict the sensitivity of adjuvant chemotherapy in ovarian cancer patients after neoadjuvant chemotherapy.

[0044] It is worth mentioning that the training set and the validation set were tested by 5-fold cross-validation, and the area under the curve (AUC) is 1 as shown in FIG. 1. Model A trained by the above 40 proteins was tested on an independent test set (29 samples), and the AUC is 0.867 as shown in FIG. 2. Figure 2 Figure 2 The results show that the 40 proteins can better represent the sensitivity of chemotherapy.

[0045] III. Parallel reaction monitoring targeted quantification

[0046] Based on the 40 proteins screened by the previous machine learning, further parallel reaction monitoring (PRM) targeted quantification was performed based on polypeptide samples. 15 polypeptides were selected for retention time correction, and a total of 54 polypeptides were obtained.

[0047] Briefly, 0.5 μg of polypeptide sample was first loaded into the pre-column (3 μm, 300 A, 4.6*50 mm) at a speed of 6 ul / min. ​20mm*75pm i.d.) and then the sample loaded on the pre-column was flushed into the analytical column (1.9pm, 120a, 150mm*75pm i.d.) for further online separation at a flow rate of 300nL / min with a gradient of 5%-30% buffer B (buffer A: 2% ACN, 98%H2O (containing 0.1% FA), buffer B: 98% ACN (containing 0.1% FA)) for 45 min. 54 polypeptides were timed collected in a ±3 min time window, the MS1 m / z range was 400-2000, the resolution was 60,000, the AGC was 3e6, the max IT was 55 ms, the target precursor ion isolation window was set to 1.6, and the normalized collision energy was 27%. The MS2 resolution was 30000, the AGC was 2e5, and the max IT was 80 ms. The PulseDIA mass spectrometry result file was analyzed using DIA-NN (1.8). The PRM data file and the reverse analysis were performed using skyline (20.1.0.31) with the default DIA-NN settings, and quantitative data of 30 proteins were obtained.

[0048] The PRM technology was used to optimize the acquisition of quantitative data of 30 proteins in the 40 proteins in the screening model based on the PulseDIA data, which was more sensitive, had higher quantitative repeatability, higher throughput, and was more suitable for clinical use. The differential proteins were further verified and the classification model was optimized.

[0049] III. Screening and determining protein markers using the classification model

[0050] Student's t-test was performed on the drug-resistant and sensitive groups in the training data set of 42 samples of the ovarian cancer tissue proteomic data based on 30 proteins, and 12 differential proteins were obtained.

[0051] The protein quantitative data of the 12 differential proteins were then used as the input features of the machine learning model of the classification model, and the R software package (version 4.6.14) of random forest was used. First, proteins with an average reduction in accuracy greater than 3 were screened, a total of 8; these 8 differential proteins were used as input features of the classification model and further subjected to 100 times of 5-fold cross-validation (i.e., 34 samples in the training set and 8 samples in the validation set), and the ntree was set to 1000. Based on the machine learning of the above classification model, the proteins screened were sorted according to the average reduction in accuracy, and the top 6 proteins as shown in Table 1 were screened. Figure 3 The 6 proteins were RBMXL1, DES, MCT1, SART1, GPKOW, and PTBP2. Subsequently, these 6 proteins were used to construct a prediction model for the sensitivity of ovarian cancer patients to chemotherapy.

[0052] V. Construction of a prediction model for chemotherapy sensitivity of ovarian cancer patients:

[0053] The machine learning model is trained using the ovarian cancer tissue samples in the training cohort and the labeled recurrence as training samples, wherein the protein marker combination is selected from one or a combination of RBMXL1, DES, MCT1, SART1, GPKOW, and PTBP2. The type of machine learning model is not limited, and the training set and the validation set are tested by 5-fold cross-validation to obtain the area under the curve (AUC) as shown in Figure 4 The AUC of the model A trained using the above 6 proteins is 0.762 when tested on an independent test set (29 samples).

[0054] That is, the present scheme is based on PulseDIA data to pre-screen 40 proteins, and the quantitative data of 30 proteins are optimized by PRM technology which has higher quantitative repeatability, higher sensitivity, higher throughput and is more suitable for clinical use, and then the final 6 proteins are selected from the 30 proteins as markers for predicting the chemotherapy sensitivity of ovarian cancer patients through the classification model, and the 6 protein molecules can effectively predict whether the ovarian cancer individual who has not started adjuvant chemotherapy will relapse within 6 months after surgery and receiving adjuvant chemotherapy

[0055] The present application is not limited to the above best mode, and anyone can derive other various forms of products under the inspiration of the present application, but regardless of any changes in shape or structure, any technical solution with the same or similar technical solutions as the present application falls within the protection scope of the present application.

Claims

1. A kit for predicting the sensitivity of adjuvant chemotherapy in a patient with high-grade serous ovarian cancer, characterized in that, include: The reagents for detecting the levels of protein markers in ovarian cancer tissue samples from subjects included a combination of the following protein markers: IFRD1, UQCR11, SART1, KALRN, TRIM3, BANF1, IGLV3-19, HLA-DRB1, ACYP1, CD48, CKB, HPGD, DES, PZP, GGCX, SLC16A1, KHDRBS1, DGKZ, LRRC41, TOMM34, KIAA1755, MAGI3, ZNF787, TCTN3, ZGPAT, ZCRB1, URGCP, GPKOW, RBMXL1, EYA3, CMSS1, POLR1E, METTL9, LDAH, CTNNBIP1, CGGBP1, PTBP2, NIP7, ST3GAL6, and WDR45.

2. A kit for predicting the sensitivity of adjuvant chemotherapy in a patient with high-grade serous ovarian cancer, characterized in that, A reagent for detecting the levels of protein markers in ovarian cancer tissue samples from subjects, wherein the protein markers are a combination of RBMXL1, DES, MCT1, SART1, GPKOW and PTBP2.

3. The kit for predicting the sensitivity of adjuvant chemotherapy of a patient with high-grade serous ovarian cancer according to any one of claims 1 or 2, characterized in that, The relative expression levels of protein markers in the ovarian cancer tissue samples of the subjects were input into a trained ovarian cancer patient chemotherapy sensitivity prediction model to obtain chemotherapy sensitivity. Chemotherapy sensitivity is negatively correlated with the probability of recurrence in a short period of time after receiving adjuvant therapy.

4. A protein marker for predicting the sensitivity of adjuvant chemotherapy in a patient with high-grade serous ovarian cancer, characterized by, include: Protein biomarkers selected from ovarian cancer tissue samples, wherein the protein biomarkers are combinations of IFRD1, UQCR11, SART1, KALRN, TRIM3, BANF1, IGLV3-19, HLA-DRB1, ACYP1, CD48, CKB, HPGD, DES, PZP, GGCX, SLC16A1, KHDRBS1, DGKZ, LRRC41, TOMM34, KIAA1755, MAGI3, ZNF787, TCTN3, ZGPAT, ZCRB1, URGCP, GPKOW, RBMXL1, EYA3, CMSS1, POLR1E, METTL9, LDAH, CTNNBIP1, CGGBP1, PTBP2, NIP7, ST3GAL6, and WDR45, are used to predict the recurrence probability of ovarian cancer patients within a short period of time after neoadjuvant therapy.

5. A method for constructing a prediction model for predicting the sensitivity of adjuvant chemotherapy of a high-grade serous ovarian cancer patient, characterized by, include: The relative expression amount of the protein marker in the ovarian cancer tissue sample of the subject with ovarian cancer who has not yet received adjuvant therapy at the present stage and will receive adjuvant therapy in the future is used as a training sample to train a machine learning model, wherein the protein marker is a combination of IFRD1, UQCR11, SART1, KALRN, TRIM3, BANF1, IGLV3-19, HLA-DRB1, ACYP1, CD48, CKB, HPGD, DES, PZP, GGCX, SLC16A1, KHDRBS1, DGKZ, LRRC41, TOMM34, KIAA1755, MAGI3, ZNF787, TCTN3, ZGPAT, ZCRB1, URGCP, GPKOW, RBMXL1, EYA3, CMSS1, POLR1E, METTL9, LDAH, CTNNBIP1, CGGBP1, PTBP2, NIP7, ST3GAL6 and WDR45.

6. A method for constructing a prediction model for predicting the sensitivity of adjuvant chemotherapy of a high-grade serous ovarian cancer patient, characterized by, Comprise: The relative expression amount of the protein marker in the ovarian cancer tissue sample of the subject with ovarian cancer who has not yet received adjuvant therapy at the present stage and will receive adjuvant therapy in the future is used as a training sample to train a machine learning model, wherein the protein marker is a combination of RBMXL1, DES, MCT1, SART1, GPKOW and PTBP2.

Images