Serum marker combination for identifying glioma recurrence and radioactive injury and application thereof

Abstract

The invention discloses a serum marker combination for identifying glioma recurrence and radioactive injury and application of the serum marker combination. The serum marker composition is prepared from phenylalanine-proline-lysine, 17 [beta]-trienol ketone, 3, 8, 9-trihydroxy-10-propyl-3, 4, 5, 8, 9, 10-hexahydroxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-desoxyprostaglandin F2 [beta], and is characterized in that the serum marker composition is prepared from phenylalanine-proline-lysine, 17 [beta]-trienol ketone, 3, 8, 9-trihydroxy-10-propyl-3, 4, 5, 8, 9, 10-hexahydroxacyclodecatetraene, 17-allylamino-17-desoxygeldanamycin A differential diagnosis model constructed by using the serum marker combination provided by the invention can be used for accurately distinguishing radioactive injury and glioma recurrence, and a sensitive and reliable marker is provided for identification of malignant glioma progression, especially early differential diagnosis; the method has the characteristics of micro-invasiveness, rapidness, safety, small trauma and the like, is friendly to patients, and effectively solves the problems of low accuracy, low sensitivity, low specificity, high delay and the like in the existing glioma recurrence and radioactive injury differential diagnosis method.

Description

Technical Field

[0001] The present invention relates to biomarkers for distinguishing glioma recurrence and radiation damage, and in particular to a serum marker combination for distinguishing glioma recurrence and radiation damage and an application thereof, belonging to the field of biomarkers for distinguishing glioma recurrence and radiation damage and an application thereof. Background Art

[0002] Glioma refers to a tumor that originates from brain glial cells and is the most common primary intracranial tumor. The prognosis of glioma is still very poor, and the recurrence rate after initial treatment is high, especially for high-grade gliomas. After recurrence, the survival of glioma patients will be significantly shortened, so early diagnosis of glioma recurrence is necessary.

[0003] The diagnosis of tumor recurrence in patients with high-grade glioma (HGG) mainly relies on anatomical imaging, which is inaccurate for early detection of recurrence or disease progression, and radiation-induced lesions often appear similar to recurrence on conventional MRI. At present, the differentiation between glioma recurrence and radiation damage mainly relies on observation for a certain period of time. The diagnosis often depends to a large extent on the experience of the imaging department and clinical physicians. There is no specific examination method to differentiate between radiation necrosis and tumor recurrence (Derlon JM, Chapon F, Noel MH, et al. Non-invasive grading of oligodendrogliomas: correlation between in vivo metabolic pattern and histopathology. Eur J Nucl Med. 2000; 27: 778–787; Ellingson BM, Chung C, Pope WB, Boxerman JL, Kaufmann TJ. Pseudoprogression, radionecrosis, inflammation or true tumor progression? Challenges associated with glioblastoma response assessment in anevolving therapeutic landscape. J Neurooncol. 2017; 134(3): 495–504.). If recurrence and radiation damage are misdiagnosed, patients may be removed from effective treatment or not receive effective treatment early enough, which may result in a poorer prognosis or expose patients to unnecessary surgical intervention.

[0004] Therefore, there is an urgent need for an efficient, accurate, rapid and less invasive method for distinguishing glioma recurrence and radiation damage to improve diagnostic accuracy, reduce risks, and provide patients with earlier, safer and more effective treatment. Summary of the invention

[0005] One of the objectives of the present invention is to identify a combination of serum markers for glioma recurrence and radiation damage;

[0006] The second purpose of the present invention is to apply the serum marker combination to prepare a diagnostic reagent for distinguishing or diagnosing glioma recurrence and radiation damage;

[0007] The third object of the present invention is to use the serum marker combination for distinguishing glioma recurrence and radiation damage to provide a method for distinguishing glioma recurrence and radiation damage for non-diagnostic or therapeutic purposes;

[0008] To achieve the above purpose, the main technical solutions adopted by the present invention include:

[0009] One aspect of the present invention is to provide a serum marker combination for identifying glioma recurrence and radiation damage, wherein the serum marker combination consists of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β.

[0010] Another aspect of the present invention is to apply the serum marker combination to prepare a diagnostic reagent for distinguishing glioma recurrence and radiation damage.

[0011] Another aspect of the present invention is to provide a method for distinguishing glioma recurrence and radiation damage for non-diagnostic or therapeutic purposes, comprising:

[0012] (1) Detecting and calculating the relative quantitative values ​​of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β in the serum sample to be tested;

[0013] (2) Calculate the P value according to the following formula:

[0014]

[0015] Among them, β0 = -10.7295, β1 = 3.405 × 10-6, β2 = 4.966 × 10-6, β3 = -1.547 × 10-5, β4 = -8.929 × 10-6, β5 = 1.352 × 10-5; X1 is the relative quantitative value of phenylalanine-proline-lysine, X2 is the relative quantitative value of 17β-trienolone, X3 is the relative quantitative value of methylimidazole acetic acid, X4 is the relative quantitative value of 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecene, and X5 is the relative quantitative value of 11-deoxyprostaglandin F2β;

[0016] e is a natural constant in mathematical definition. It is an infinite non-repeating decimal and a transcendental number. It is the base of natural logarithm and its value is approximately 2.718281828459045.

[0017] (3) According to the calculation results, if P(Y=1|X)>0.5, it is predicted that Y=1, that is, the patient of the sample to be tested has radiation damage; if P(Y=1|X)≤0.5, it is predicted that Y=0, that is, the patient of the sample to be tested has glioma recurrence.

[0018] In a preferred specific embodiment of the present invention, in step (1), a liquid chromatography-mass spectrometry method is used to detect and calculate the relative quantitative values ​​of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β in the serum sample to be detected; more preferably, the method of detecting and calculating the relative quantitative values ​​of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β in the serum sample to be detected by liquid chromatography-mass spectrometry comprises:

[0019] (a) The signal intensity, peak area and other related data of each metabolite in the serum samples were determined by liquid chromatography-mass spectrometry. The molecular formula was predicted by molecular ion peaks and fragment ions and compared with mzCloud (https: / / www.mzcloud.org / ), mzVault and Masslist databases to obtain the original quantitative data of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β; (b) The original quantitative results were standardized according to the following formulas: The relative peak areas of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β were obtained: original sample quantitative value / (sum of sample metabolite quantitative values / sum of QC1 sample metabolite quantitative values); the above relative peak areas are the relative quantitative values ​​of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β.

[0020] QC sample (Quality Control): Take an equal volume of sample from each experimental sample and mix them as QC samples. There are 7 QC samples in this experiment, which are named from QC1 to 7, and QC1 sample is one of them.

[0021] The serum marker combination provided by the present invention has high accuracy in distinguishing patients with radiation-damaged and recurrent gliomas, and provides sensitive and reliable metabolic markers for the identification of malignant progression of gliomas, especially early differential diagnosis; at the same time, the present invention distinguishes patients with radiation-damaged and recurrent gliomas by serum markers, and compared with tissue biopsy, it has the characteristics of minimal invasiveness, rapidity, safety, and less trauma, is patient-friendly, and effectively solves the problems of low accuracy, low sensitivity, low specificity, and high latency in existing differential diagnosis methods for glioma recurrence and radiation damage; the present invention is expected to become an important breakthrough in the differential diagnosis of radiation-damaged and recurrent gliomas, and provide patients with safer and more effective medical services. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The training set was used to verify the ROC curve results of the differential diagnosis model constructed by the present invention using serum markers.

[0023] Figure 2 The validation set is used to verify the accuracy of the differential diagnosis model constructed by the present invention using serum markers. DETAILED DESCRIPTION

[0024] The present invention will be further described below in conjunction with specific embodiments, and the advantages and features of the present invention will become clearer as the description proceeds. However, it should be understood that the embodiments are exemplary only and do not constitute any limitation to the scope of the present invention. It should be understood by those skilled in the art that the details and forms of the technical solution of the present invention may be modified or replaced without departing from the spirit and scope of the present invention, but these modifications or replacements all fall within the protection scope of the present invention.

[0025] Experimental Example 1 Screening of serum marker combinations for distinguishing glioma recurrence and radiation damage, construction of differential diagnosis model, and verification of accuracy

[0026] 1. Screening of serum marker combinations for distinguishing glioma recurrence and radiation damage, and construction of differential diagnosis model

[0027] 1. Collect samples

[0028] (1) Blood specimens from glioma patients who underwent lumbar puncture from May 2021 to December 2023 were collected, and the supernatant was extracted after centrifugation at 2000 rpm for 10 min at 4°C and stored at -80°C in time. The characteristics of the included patients are summarized in Table 1.

[0029] Table 1 Clinical information of collected samples

[0030]

[0031] (2) Inclusion and exclusion criteria:

[0032] Inclusion criteria:

[0033] 1) Patients with a clear pathological diagnosis, including histopathological diagnosis and molecular pathological diagnosis.

[0034] 2) Patients with confirmed recurrence or radiation damage after multiple magnetic resonance imaging (MRI) reexaminations.

[0035] 3) Those who sign the informed consent form for specimen collection.

[0036] Exclusion criteria:

[0037] 1) Those who have had malignant tumors or metastatic tumors in other parts of the body.

[0038] 2) Patients with clear contraindications who cannot complete lumbar puncture examination.

[0039] 3) Request to withdraw as a researcher during the course of the project.

[0040] (3) Sample grouping

[0041] The classification of the disease status of the tumor (including recurrence and radiation injury) was completed by experienced neuro-oncologists based on comprehensive consideration of symptoms, signs and imaging findings. The diagnostic criteria for recurrence were mainly based on the modified RANO (modified radiographic response assessment in neuro-oncology) criteria, including enhancement outside the radiation field (any new measurable lesions need to be confirmed in a repeat study after 4 weeks), an increase of 25% or more in the sum of the perpendicular diameter products between the first post-radiotherapy (RT) scan and the scan after 12 weeks (based on at least 2 consecutive studies, 4 weeks apart), or obvious clinical deterioration. The diagnostic criteria for radiation injury were PWI (Perfusion Weighted Imaging), DWI (Aiffusion Weighted Imaging) and MRS (Magnetic Resonance Spectroscopy) indicating non-tumor progression and no progression after 6 months of observation.

[0042] 2. Detection and identification of metabolites

[0043] (1) Metabolite extraction and mass spectrometry analysis:

[0044] Pretreat the blood sample to obtain a blood sample to be tested. The specific method of pretreatment is: 1. Extract the supernatant after centrifugation at 2000rpm for 10 minutes at 4°C, and store it at -80°C in time. 2. Take 100μL of sample and place it in an EP tube, add 400μL of 80% methanol aqueous solution, vortex and shake, let stand in an ice bath for 5 minutes, centrifuge at 15000g and 4°C for 20 minutes, and take the supernatant. 3. Add mass spectrometry grade water to dilute to a methanol content of 53%, centrifuge at 15000g and 4°C for 20 minutes, and take the supernatant to obtain the pretreated blood sample to be tested.

[0045] The pretreated blood sample (100 μL) was placed in an EP tube and resuspended with pre-cooled 80% methanol through a vortex. Then incubated on ice for 5 minutes and centrifuged at 4°C, 15000g for 20 minutes. An equal volume of sample was taken from each experimental sample and mixed to prepare serum QC (quality control) samples, which were used to balance the chromatography-mass spectrometry system and monitor the instrument status, and to evaluate the system stability during the entire experiment. The supernatant was diluted with LC-MS grade water to a final concentration of 53% methanol. The sample was then transferred to a fresh Eppendorf tube and then centrifuged at 15000g, 4°C for 20 minutes. Finally, the supernatant was injected into the LC-MS / MS system for analysis.

[0046] UHPLC-MS / MS analysis was performed using a Vanquish UHPLC system (ThermoFisher, Germany), combined with an Orbitrap Q ExactiveTM HF mass spectrometer (Novogene Co., Beijing) and an Orbitrap QExactiveTM HF-X mass spectrometer (ThermoFisher, Germany). The sample was injected into a Hypersil Gold column (100×2.1 mm, 1.9 μm) at a flow rate of 0.2 mL / min with a 12 min linear gradient. The eluents in the positive and negative polarity modes were eluent A (0.1% FA in Water) and eluent B (Methanol), respectively. The solvent gradient was set as: 2% B, 1.5 min; 2-85% B, 3 min; the QExactiveTM HF mass spectrometer operated in positive and negative polarity modes, with a spray voltage of 3.5 kV, a capillary temperature of 320°C, a sheath gas flow rate of 35 psi, an auxiliary gas flow rate of 10 L / min, an S-lens RF level of 60, and an auxiliary gas heater temperature of 350°C.

[0047] (2) Metabolite identification and data analysis

[0048] The offline data (.raw) file was imported into the CD3.1 search software for processing. The retention time, mass-to-charge ratio and other parameters of each metabolite were simply screened. Then, the retention time deviation was set to 0.2 min and the mass deviation was set to 5 ppm to align the peaks of different samples to make the identification more accurate. Then, the mass deviation was set to 5 ppm, the signal intensity deviation was set to 30%, the signal-to-noise ratio was set to 3, the minimum signal intensity, the summed ion and other information were set for peak extraction. At the same time, the peak area was quantified, and the target ion was integrated. The molecular formula was predicted by the molecular ion peak and the fragment ion and compared with the mzCloud (https: / / www.mzcloud.org / ), mzVault and Masslist databases. The blank sample was used to remove the background ions. The original quantitative results were standardized according to the formula: sample original quantitative value / (sum of sample metabolite quantitative values / sum of QC1 sample metabolite quantitative values) to obtain the relative peak area. The compounds with a relative peak area CV greater than 30% in the QC sample were deleted, and finally the identification and relative quantitative results of the metabolites were obtained. The data processing part is based on the Linux operating system (CentOS version 6.6) and software R and Python.

[0049] QC sample (Quality Control): Take an equal volume of sample from each experimental sample and mix them as QC samples. There are 7 QC samples in this experiment, which are named from QC1 to 7, and QC1 sample is one of them.

[0050] (3) Metabolite annotation

[0051] The metabolites were annotated using the KEGG database (https: / / www.genome.jp / kegg / pathway.html), the HMDB database (https: / / hmdb.ca / metabolites), and the LIPIDMaps database (http: / / www.lipidmaps.org / ).

[0052] 3. PLS-DA preliminary screening of differential metabolites

[0053] Principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) were performed in metaX. We used univariate analysis (t-test) to calculate statistical significance (p-value). The risk of overfitting of the partial least squares model was verified by sorting validation. Metabolites with VIP>1, p-value<0.05, differential fold (FC) ≥1.2 or differential fold ≤0.8 were considered differential metabolites. The specific results of the screened differential metabolites are shown in Table 2.

[0054] Table 2 Results of preliminary screening of differential metabolites

[0055]

[0056]

[0057] 4. The data is balanced by using the Synthetic Minority Over-sampling Technique (SMOTE) to generate new, virtual minority class samples. 80% of the balanced data is randomly divided into a test set and 20% as a validation set for subsequent analysis. The number and grouping of balanced data are shown in Table 3 below.

[0058] SMOTE is a synthetic data augmentation technique specifically designed to deal with class imbalance problems. In imbalanced classification problems, minority class samples are often insufficient to train an effective classification model. SMOTE solves this problem by generating new samples from minority class samples. Specifically, for each minority class sample, SMOTE increases the number of minority class samples by randomly selecting a sample in its neighborhood and generating new samples along the line between the two samples. Advantages: (1) It can effectively increase the number of minority class samples, balance the distribution of the dataset, and reduce the model's bias towards the majority class. (2) By increasing the diversity and number of minority class samples, SMOTE can improve the classifier's learning ability for the minority class and improve the precision and recall of the classification model. (3) Compared with simple oversampling methods (such as copying samples), SMOTE enhances the diversity of minority class samples by generating new samples, reducing the risk of overfitting. (4) SMOTE can be applied to a variety of data types, especially performing well on numerical data. The specific steps are: (1) Determine minority class samples: Identify minority class samples in the dataset. That is, in the class imbalance problem, the class with a smaller number. (2) Select neighbors: For each minority class sample, use the K nearest neighbor algorithm (KNN) to find its k nearest neighbors. (3) Generate synthetic samples: For each minority class sample, randomly select k neighbors, calculate the difference between them and the current minority class sample, and generate a new sample based on the difference. Specifically, the synthetic sample is generated by interpolating along the line between the original sample and the neighbor sample. (4) Repeatedly generate samples: According to the number of minority class samples that need to be generated, repeat the above steps until the predetermined number of minority class samples is reached. (5) Merge data sets: Combine the generated synthetic minority class samples with the original minority class samples to form a balanced data set for classifier training.

[0059] Table 3 Blood sample data distribution

[0060] Radiation injury group Glioma recurrence group total Test Set 72 72 144 Validation set 18 18 36 total 90 90 180

[0061] 5. Recursive feature elimination to select features and binary logistic regression to build the model

[0062] 80% of all serum samples were randomly divided into a test set and 20% as a validation set. The test set data was used to build the model.

[0063] (1) Recursive feature elimination is a feature selection method that aims to find the most influential features by recursively reducing the set of features considered. This process can help improve the performance of machine learning models or make the models more concise. In order to reduce the risk of overfitting, feature selection is performed through recursive feature elimination. The specific steps are:

[0064] 1) Model training: First, the model is trained using the initial complete feature set.

[0065] 2) Feature importance evaluation: After the model training is completed, the importance of each feature is evaluated. Different types of models may use different methods to evaluate the importance of features, such as based on the size of the coefficient, the number of feature splits in the tree model, etc.

[0066] 3) Remove the least important features: Based on the evaluation results, remove one or more of the least important features. The specific number of features to be removed depends on the preset strategy or parameter settings.

[0067] 4) Repeat step: Repeat steps 1 to 3 using the remaining feature sets until the preset number of features is reached or other stopping conditions are met.

[0068] 5) Final model training: When the feature number requirement or other stopping conditions are reached, the final model is trained using the selected feature set.

[0069] The final number of features determined in this test example is 5, and the 5 serum markers screened are shown in Table 4.

[0070] Table 4 Five serum markers for distinguishing glioma recurrence from radiation damage

[0071]

[0072] (2) Constructing a binary logistic regression model

[0073] A binary logistic regression model was constructed with the relative values ​​of the selected serum metabolites as independent variables and the groups of recurrence and radiation damage as dependent variables:

[0074] The specific formula of the model is:

[0075]

[0076] Among them, β0 = -10.7295, β1 = 3.405 × 10-6, β2 = 4.966 × 10-6, β3 = -1.547 × 10-5, β4 = -8.929 × 10-6, β5 = 1.352 × 10-5. X1 is the relative quantitative value of phenylalanine-proline-lysine, X2 is the relative quantitative value of 17β-trienolone, X3 is the relative quantitative value of methylimidazole acetic acid, X4 is the relative quantitative value of 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecene, and X5 is the relative quantitative value of 11-deoxyprostaglandin F2β.

[0077] According to the model results, if P(Y=1|X)>0.5, then Y=1 is predicted, that is, the patient has radiation injury. If P(Y=1|X)≤0.5, then Y=0 is predicted, that is, the patient has recurrence.

[0078] The ROC curve was used to validate the model. The P values ​​of all selected metabolites were less than 0.05, and the AUC of the constructed model was 1, and the accuracy was 0.97 ( Figure 1 ).

[0079] The ROC curve is a method to study the relationship between model sensitivity and specificity, with sensitivity as the vertical axis and 1-specificity (1-specificity, which refers to the value obtained by subtracting the specificity value from 1) as the horizontal axis. The evaluation basis is to compare the area under the curve (AUC). When AUC is greater than 0.5, the closer AUC is to 1, the better the model performance is, indicating a better diagnostic effect. If it is less than 0.5, it means that the model accuracy is poor.

[0080] The AUC value obtained from the test results is close to 1, indicating that the serum marker composition screened by the present invention has an excellent effect in distinguishing glioma recurrence and radiation damage.

[0081] 2. Validation of the differential diagnosis model using the validation set

[0082] The accuracy of the differential diagnosis model constructed above was verified by the validation set; the accuracy of the differential diagnosis model constructed by the present invention using the screened serum marker combination was verified in the validation set. The validation result showed that the model showed AUC=0.97 and accuracy=0.89 ( Figure 2 ), indicating that the constructed differential diagnosis model also has good diagnostic performance in the validation set and can accurately distinguish glioma recurrence from radiation damage.

Claims

1. A serum marker combination for distinguishing glioma recurrence and radiation damage, characterized in that: The cerebrospinal fluid metabolite combination consists of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β.

2. Use of the serum marker combination according to claim 1 in the preparation of a diagnostic reagent for distinguishing or diagnosing glioma recurrence and radiation damage.

3. A method for distinguishing glioma recurrence and radiation damage for non-diagnostic or therapeutic purposes, characterized in that: include: (1) Detecting and calculating the relative quantitative values ​​of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β in the serum sample to be tested; (2) Calculate the P value according to the following formula: Among them, β0 = -10.7295, β1 = 3.405 × 10-6, β2 = 4.966 × 10-6, β3 = -1.547 × 10-5, β4 = -8.929 × 10-6, β5 = 1.352 × 10-5; X1 is phenylalanine-proline-lysine, X2 is 17β-trienol ketone, X3 is methylimidazole acetic acid, X4 is 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecene, and X5 is 11-deoxyprostaglandin F2β; (3) According to the calculation results, if P(Y=1|X)>0.5, it is predicted that Y=1, that is, the patient of the sample to be tested has radiation damage; if P(Y=1|X)≤0.5, it is predicted that Y=0, that is, the patient of the sample to be tested has glioma recurrence.

4. The method according to claim 3, characterized in that In step (1), a liquid chromatography-mass spectrometry method is used to detect and calculate the relative quantitative values ​​of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β in the serum sample to be tested.

5. The method according to claim 4, characterized in that The method for detecting and calculating the relative quantitative values ​​of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β in the serum sample to be detected by liquid chromatography-mass spectrometry comprises: (a) The signal intensity and peak area data of each metabolite in the serum sample were determined by liquid chromatography-mass spectrometry, and the molecular formula was predicted by molecular ion peaks and fragment ions and compared with the database to obtain the original quantitative data of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β; (b) The original quantitative results were standardized according to the following formula to obtain the original quantitative data of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β. The relative peak areas of 8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β: original sample quantitative value / (sum of sample metabolite quantitative values / sum of QC1 sample metabolite quantitative values); the above relative peak areas are the relative quantitative values ​​of phenylalanine-proline-lysine, 17β-trienolone, 3,8,9-trihydroxy-10-propyl-3,4,5,8,9,10-hexahydrooxacyclodecatetraene, 17-allylamino-17-demethoxygeldanamycin and 11-deoxyprostaglandin F2β.

6. The method according to claim 5, characterized in that The databases include mzCloud, mzVault and Masslist.

7. The method according to claim 3, characterized in that The e is a natural constant in mathematical definition, an infinite non-repeating decimal, a transcendental number, and the base of natural logarithms.

8. The method according to claim 7, characterized in that The value of e is 2.718281828459045.

9. The method according to claim 5, characterized in that The meaning of the QC1 sample is: take an equal volume of sample from each experimental sample and mix it as a QC sample. There are 7 QC samples in total, which are named from QC1 to 7, and the QC1 sample is one of them.

Images