Metabolite marker for colorectal cancer diagnosis and application thereof
By combining metabolite biomarkers and machine learning models, the problems of invasiveness and insufficient sensitivity in colorectal cancer screening have been solved, achieving efficient and non-invasive early diagnosis with an AUROC of 0.97.
Patent Information
- Application Number
- CN202511414365.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-29
- Publication Date
- 2026-02-13
AI Technical Summary
Existing colorectal cancer screening methods are highly invasive and lack sufficient sensitivity and specificity, making it difficult to achieve large-scale early diagnosis.
A combination of metabolite biomarkers, such as sulfate, ubiquinone-1, deoxycholate-glycine conjugate, demethyl phloroglucinone, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoic acid ester, and acyl AMP, combined with liquid chromatography-mass spectrometry and machine learning models, was used for the detection and risk assessment of metabolites in blood samples.
It achieves high sensitivity and specificity in non-invasive early diagnosis of colorectal cancer, with high patient compliance and an AUROC of 0.97.
Smart Images
Figure CN121522031A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of biological detection, in particular to a metabolite marker for early diagnosis of colorectal cancer, application thereof and a device. BACKGROUND
[0002] Colorectal cancer (CRC) is a malignant tumor with high morbidity and mortality worldwide. Early detection and diagnosis are crucial for improving patient survival. Traditional screening methods mainly include colonoscopy, fecal occult blood test and fecal immunochemical test. However, colonoscopy is an invasive procedure, which requires bowel preparation and has risks of perforation and bleeding. Patients have low acceptance and poor compliance, and the screening coverage is insufficient, making it difficult to achieve large-scale early screening. Although fecal occult blood test and fecal immunochemical test are non-invasive, they have problems of insufficient sensitivity and specificity.
[0003] In recent years, liquid biopsy techniques, such as blood-based metabolomics, proteomics and epigenetic (such as DNA methylation) analysis, have provided a new way for non-invasive early diagnosis of cancer. Metabolomics can reflect the real-time physiological and pathological state of the body, and abnormal metabolites are closely related to the occurrence and development of various cancers. DNA methylation as an important epigenetic modification, its abnormal pattern is an early event of cancer, and has tissue specificity, existing in cell-free DNA in blood, which is easy to detect. However, there are few studies on metabolite markers for early diagnosis of colorectal cancer, and their sensitivity and specificity need to be improved.
[0004] Therefore, there is an urgent need for a marker and its application that can accurately, quickly and simply realize non-invasive screening and early diagnosis of colorectal cancer. SUMMARY
[0005] The present application aims to at least partially solve one of the problems in the related art.
[0006] To this end, an embodiment of the first aspect of the present application proposes a metabolite marker for early diagnosis of colorectal cancer, the marker being selected from one or more of the group consisting of sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, desmethylphorbol, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoic acid ester and fatty acyl AMP. The above-mentioned marker combination performs well in colorectal cancer detection, with an AUROC of 0.97, which can accurately, quickly and simply realize non-invasive screening and early diagnosis of colorectal cancer.
[0007] In some embodiments, the markers consist of: sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesteryl-docosahexaenoate, and fatty acyl AMP.
[0008] Embodiments of the second aspect of the application propose a use of a reagent for detecting a level of a metabolite marker selected from one or more of the group consisting of: sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesteryl-docosahexaenoate, and fatty acyl AMP in the manufacture of a kit for early diagnosis of colorectal cancer.
[0009] In some embodiments, the markers consist of: sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesteryl-docosahexaenoate, and fatty acyl AMP.
[0010] In some embodiments, the reagent is for detecting the level of the metabolite markers in a blood sample.
[0011] In some embodiments, the blood sample is serum.
[0012] Embodiments of the third aspect of the application propose a device for early diagnosis of colorectal cancer, comprising: a detection unit configured to detect a level of metabolite markers in a sample obtained from a subject; a calculation unit configured to input the level of the metabolite markers of the subject into a model for early diagnosis of colorectal cancer to calculate a risk score of colorectal cancer; and an output unit configured to output an early diagnosis result based on the risk score, wherein the metabolite markers comprise: sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesteryl-docosahexaenoate, and fatty acyl AMP.
[0013] In some embodiments, the markers consist of sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesteryl-docosahexaenoate, and acyl AMP.
[0014] In some embodiments, the detecting unit detects the levels of the metabolite markers by liquid chromatography-mass spectrometry (LC-MS) technique.
[0015] In some embodiments, the early-stage colorectal cancer diagnosis model is an SVM diagnosis model, an RF diagnosis model, an XGBoost diagnosis model, and an LR diagnosis model, preferably, the early-stage colorectal cancer diagnosis model is a random forest (RF) diagnosis model.
[0016] In some embodiments, the device further comprises a comparing unit storing levels of reference metabolite markers and corresponding reference colorectal cancer risk scores, the comparing unit being configured to compare the colorectal cancer risk score of the subject with the reference colorectal cancer risk score and / or compare the levels of the metabolite markers of the subject with the levels of the reference metabolite markers.
[0017] In some embodiments, a result that the subject is inclined to have an early-stage risk of colorectal cancer is output based on the following conditions: the level of a first marker in the metabolite markers of the subject is higher than the level of the reference metabolite marker, and the level of a second marker in the metabolite markers of the subject is lower than the level of the reference metabolite marker, wherein the first marker is ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, cholesteryl-docosahexaenoate, acyl AMP, and the second marker is sulfate, sedoheptulose 1,7-bisphosphate; or the colorectal cancer risk score of the subject is higher than the reference colorectal cancer risk score.
[0018] The advantages and technical effects brought by the independent claims according to the embodiments of the present application are as follows: The embodiments of the present application provide a preferred metabolite marker composition based on metabolic abnormalities as an early event of cancer occurrence, which can be used for non-invasive and convenient early-stage diagnosis of colorectal cancer, only blood needs to be drawn in the clinic, and the patient compliance is high; and the AUROC can reach 0.97. BRIEF DESCRIPTION OF DRAWINGS
[0019] Figure 1 FIG. 1 is a schematic diagram of a screening process of metabolite markers according to an embodiment of the present application.
[0020] Figure 2 is a support vector machine (SVM)-based colorectal cancer early diagnosis diagnosis model evaluation result graph of an embodiment of the present application.
[0021] Figure 3 is a random forest (RF)-based colorectal cancer early diagnosis diagnosis model evaluation result graph of an embodiment of the present application.
[0022] Figure 4 is an XGBoost-based colorectal cancer early diagnosis diagnosis model evaluation result graph of an embodiment of the present application.
[0023] Figure 5 is a linear regression (LR)-based colorectal cancer early diagnosis diagnosis model evaluation result graph of an embodiment of the present application. DETAILED DESCRIPTION
[0024] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings. The embodiments described below by reference to the drawings are exemplary and are intended to explain the present application, and cannot be understood as a limitation of the present application.
[0025] An embodiment of the first aspect of the present application proposes a metabolite marker for colorectal cancer early diagnosis, the marker being selected from one or more of the group consisting of sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoic acid ester (also known as CE (22:6 (4Z,7Z,10Z,13Z,16Z,19Z))) and fatty acyl AMP. The above marker combination performs well in colorectal cancer detection, with an AUROC of up to 0.97, enabling accurate, rapid and simple non-invasive screening and early diagnosis of colorectal cancer.
[0026] In some embodiments, the marker consists of sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoic acid ester and fatty acyl AMP.
[0027] In some embodiments, the markers consist of sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoate, and acyl AMP.
[0028] In some embodiments, the markers consist of sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoate, and acyl AMP.
[0029] In some embodiments, the reagent is used for detecting the level of the metabolic markers in a blood sample.
[0030] In some embodiments, the blood sample is serum.
[0031] In some embodiments, the early diagnosis device for colorectal cancer comprises: a detection unit configured to detect the level of metabolic markers in a sample obtained from a subject; a calculation unit configured to input the level of the metabolic markers of the subject into a colorectal cancer early diagnosis model to calculate a colorectal cancer risk score; and an output unit configured to output an early diagnosis result based on the risk score, wherein the metabolic markers comprise: sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoate, and acyl AMP.
[0032] In some embodiments, the markers consist of sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, demethylphytonadiene, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoate, and acyl AMP.
[0033] In some embodiments, the detection unit detects the level of the metabolic markers using liquid chromatography-mass spectrometry (LC-MS) technology. It should be noted that the specific detection means is not limited thereto, as long as it can detect the level of the above-mentioned metabolic markers In some embodiments, the colorectal cancer early diagnosis model is a SVM diagnosis model, a RF diagnosis model, an XGBoost diagnosis model, and a LR diagnosis model, preferably, the colorectal cancer early diagnosis model is a random forest diagnosis model.
[0034] In some embodiments, the apparatus further comprises a comparison unit storing levels of reference metabolite markers and their corresponding reference colorectal cancer risk scores, the comparison unit being configured to compare the colorectal cancer risk score of the subject and the reference colorectal cancer risk score and / or compare the levels of the metabolite markers of the subject and the levels of the reference metabolite markers.
[0035] In some embodiments, a result that the subject is inclined to have an early risk of colorectal cancer is outputted based on the condition that: the level of a first marker in the metabolite markers of the subject is higher than the level of the reference metabolite marker, and the level of a second marker in the metabolite markers of the subject is lower than the level of the reference metabolite marker, wherein the first marker is ubiquinol-1, deoxycholic acid glycine conjugate, demethylphyquionone, 3-(3,5-diiodo-4-hydroxyphenyl)lactate, vitamin K1, O-decanoyl-L-carnitine, cholesterol-docosahexaenoic acid ester, fatty acyl AMP, and the second marker is sulfate, sedoheptulose 1,7-bisphosphate; or the colorectal cancer risk score of the subject is higher than the reference colorectal cancer risk score.
[0036] In some embodiments, the reference colorectal cancer risk score is 0.5.
[0037] In some embodiments, the reference colorectal cancer risk score (i.e. threshold or cut-off value) is shown in Table 1 below.
[0038] Table 1
[0039] The experimental methods in the following examples are routine methods, which are carried out according to the techniques or conditions described in the literature in the art or according to the product instructions, unless otherwise specified. The materials, reagents, etc. used in the following examples can be obtained commercially, unless otherwise specified. The quantitative analysis tests in the following examples are set up with three repeated experiments, and the results are averaged, unless otherwise specified.
[0040] Example Serum samples of 248 patients with colorectal cancer (CRC) diagnosed by histopathology were collected from Zhuhai People's Hospital from 2020 to 2023. At the same time, 467 serum samples of non-cancer controls (NCC) were collected during the same period of colonoscopy. All study subjects were collected 5 mL of peripheral venous blood after fasting for more than 8-16 hours, and the serum was separated by centrifugation at 3000 r / min for 10 minutes at room temperature within 2 hours after collection. After aliquot, it was stored in a -80℃ ultra-low temperature refrigerator for standby. This study was approved by the Ethics Committee and Animal Ethics Committee of Zhuhai People's Hospital (Ethics Approval No: (2022) Lunsan
Research
[0041] Example 1 Screening of metabolite markers In this example, metabolite detection was performed on colorectal cancer (248 cases) and non-cancer control (467) samples. The samples were randomly divided into training set and test set at a ratio of 7:3. Differential metabolite analysis was performed on CRC samples (174 cases) and NCC samples (327 cases) in the training set, and 10 core colorectal cancer metabolite markers were obtained.
[0042] 1.1 Sample pretreatment Take out the serum samples stored at -80℃ and thaw on ice. Take 100 μL serum and add 400 μL methanol (chromatographically pure), vortex mix for 30 seconds, then centrifuge at 4℃, 14,000 rpm for 10 minutes. Transfer 200 μL supernatant to a new Eppendorf tube, dry with a vacuum centrifugal concentrator (37℃) for 150 minutes. After drying, seal the sample and store it back at -80℃. Before UPLC-MS analysis, take out the sample and reconstitute with 50 μL ultrapure water, vortex mix for 30 seconds, then place in a water bath for ultrasonic treatment for 30 seconds. Then centrifuge at 4℃, 14,000 rpm for 10 minutes, take 20 μL supernatant immediately for subsequent LC-MS analysis 1.2 LC-MS analysis The ultra-performance liquid chromatography-electrospray ion source-quadrupole time-of-flight mass spectrometry (UPLC-ESI-QTOF-MS) system was used for analysis, wherein the UPLC system was a Waters ACQUITY UPLC I-Class system, and the mass spectrometer was a Waters Synapt G2-Si tandem mass spectrometer. Specifically, an ACQUITY UPLC HSS T3 chromatographic column (1.8 μm, 2.1×100 mm i.d., Waters Corp.) was used; the column temperature was controlled at 30°C; the temperature of the autosampler was maintained at 4°C to prevent sample degradation; and the injection volume was 2 μL. The mobile phase A was ultrapure water containing 0.1% formic acid, and the mobile phase B was acetonitrile containing 0.1% formic acid. Each sample was injected twice, and data were collected in positive and negative ion full scan modes; the scan mass range was m / z 50 to the upper limit (resolution 10,000); the electrospray ion source (ESI) parameters were as follows: capillary voltage 2.0 kV, cone hole voltage 20 V; ion source temperature 100°C, desolvation temperature 200°C; desolvation gas flow rate 500 L / h. A lock mass solution (leucine enkephalin) was used for real-time mass correction during the experiment.
[0043] 1.3 Data processing (1) The original data were preprocessed using the XCMS package based on R language, including peak extraction, peak optimization, retention time alignment, feature formation, feature grouping, aggregation, spectrum extraction, spectrum aggregation, compound identification, and quantitative data generation, and the parameter settings were as follows: peak width = c(5, 20), noise = 1000, snthresh = 3, ppm = 20, binSize = 6, minFraction = 0.4, and bw = 20.
[0044] (2) Metabolite annotation was performed by matching the Human Metabolome Database (HMDB) (version 5.0) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) (version 2023) databases through the metID package, and the parameter settings were as follows: ms1.match.ppm = 15, rt.match.tol = 30, threads = 30, and column = rp.
[0045] (3) The serum metabolome peak intensity data were standardized by log2 conversion, data normalization based on QC-RLSC, and feature filtering with a relative standard deviation (RSD) of > 35%.
[0046] 1.4 Differential metabolite screening and marker determination The normalized data in the training set was subjected to multivariate statistical analysis and modeling using the R package ropls, including principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA). Further, Wilcoxon rank-sum test was used to compare the metabolite levels between tumor patients and healthy controls, and to screen significantly different metabolites (criteria: FDR < 0.05; VIP > 1; |log2FC| > 1). LASSO regression analysis was performed by the R package glmnet to screen non-zero coefficient features, and RF algorithm was used to screen stable features by the R package randomForest, in order to identify important and significant different metabolites. Then, the features obtained by LASSO and RF algorithms were subjected to intersection analysis using a Venn diagram, and finally 10 metabolites common to both were determined as core metabolite markers.
[0047] The results are shown in Figure 1 , and Figure 1 is a schematic diagram of the screening process of the metabolite markers of the embodiments of the present application, wherein part A is the features screened based on the RF algorithm, part B is the features screened based on the LASSO regression algorithm, and part C is the Venn diagram of the metabolites obtained by the two methods.
[0048] Through this embodiment, 75 significantly different metabolites were identified, and by consulting the HMDB, MCE database and related literature, 26 endogenous metabolites were screened from the above 75 metabolites for subsequent analysis according to the biological source of the metabolites (i.e. produced by the human body or intestinal flora, not exogenous intake). Further, 11 metabolites were obtained by RF algorithm, and 21 metabolites were obtained by LASSO regression. At the same time, 10 core metabolite markers were identified in the above two methods, namely sulfate, ubiquinone-1, deoxycholic acid glycine conjugate, desmethylphorbol, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-diphosphate, cholesterol-eicosapentaenoate and fatty acyl AMP.
[0049] Example 2 Construction and evaluation of early diagnosis model for colorectal cancer 2.1 Feature definition In the training set, the input features of the model were the relative expression levels of the 10 core metabolites obtained in Example 1, and the output label was the pathological diagnosis result (CRC was marked as "1" and NCC was marked as "0").
[0050] 2.2 Model construction Four classification models were constructed based on the training set, and the hyperparameters were optimized by 5-fold cross-validation. Specifically, the support vector machine (SVM) model was implemented using the R package e1071, with a radial basis function as the kernel function and the optimization parameters including the penalty coefficient C and the kernel coefficient gamma; the random forest (RF) model was implemented using the R package randomForest, with the optimization parameters including the number of decision trees n_estimators and the maximum depth of the tree max_depth; the XGBoost model was implemented using the R package xgboost, with the optimization parameters including the learning rate learning_rate, the number of iterations n_estimators, and the tree depth max_depth; and the logistic regression (LR) model was implemented using the R package caTools, with the optimization parameter being the regularization coefficient C. To avoid model optimization bias caused by data partition deviation, the process was repeated 5 times, each time using a different random split seed to generate 5 independent optimized models for validation robustness.
[0051] 2.3. Model evaluation The performance of the four models was evaluated for colorectal cancer (74 cases) and non-cancer control (140 cases) samples in the test set, with metabolite determination and data processing as described in embodiments 1.1-1.3, and the models obtained in embodiments 2.1-2.2. The evaluation indicators included AUROC, sensitivity (also known as sensitivity), and specificity, etc.
[0052] The training results and test results are shown in Table 2 and Figures 2-5 .
[0053] Figure 2 is a support vector machine (SVM) based colorectal cancer early diagnosis diagnosis model evaluation result figure of the embodiment of the application, wherein part A is the evaluation confusion table of the SVM based colorectal cancer early diagnosis diagnosis model of the embodiment of the application in the training set, part B is the evaluation confusion table of the SVM based colorectal cancer early diagnosis diagnosis model of the embodiment of the application in the test set, and part C is the ROC curve of the SVM based colorectal cancer early diagnosis diagnosis model of the embodiment of the application in the test set and the training set.
[0054] Figure 3 is a random forest (RF) based colorectal cancer early diagnosis diagnosis model evaluation result figure of the embodiment of the application, wherein part A is the evaluation confusion table of the RF based colorectal cancer early diagnosis diagnosis model of the embodiment of the application in the training set, part B is the evaluation confusion table of the RF based colorectal cancer early diagnosis diagnosis model of the embodiment of the application in the test set, and part C is the ROC curve of the RF based colorectal cancer early diagnosis diagnosis model of the embodiment of the application in the test set and the training set.
[0055] Figure 4is an evaluation result diagram of a colorectal cancer early diagnosis diagnosis model based on XGBoost of an embodiment of the present application, wherein part A is an evaluation confusion table of the colorectal cancer early diagnosis diagnosis model based on XGBoost of the embodiment of the present application in the training set, part B is an evaluation confusion table of the colorectal cancer early diagnosis diagnosis model based on XGBoost of the embodiment of the present application in the test set, and part C is an ROC curve of the colorectal cancer early diagnosis diagnosis model based on XGBoost of the embodiment of the present application in the test set and the training set.
[0056] Figure 5 is an evaluation result diagram of a colorectal cancer early diagnosis diagnosis model based on linear regression (LR) of an embodiment of the present application, wherein part A is an evaluation confusion table of the colorectal cancer early diagnosis diagnosis model based on LR of the embodiment of the present application in the training set, part B is an evaluation confusion table of the colorectal cancer early diagnosis diagnosis model based on LR of the embodiment of the present application in the test set, and part C is an ROC curve of the colorectal cancer early diagnosis diagnosis model based on LR of the embodiment of the present application in the test set and the training set.
[0057] Table 2
[0058] The results show that the RF model has the best performance, and the AUROC of the test set is 0.97. The above results show that the classification model based on 10 kinds of core metabolites has excellent diagnostic performance for colorectal cancer.
[0059] In addition, the terms "first", "second", "third", etc. are used only for descriptive purposes and are not to be construed as indicating or implying relative importance or an indicated number of technical features. Thus, a feature defined with "first", "second" or "third" can include at least one of the feature explicitly or implicitly. In the description of the present application, the meaning of "a plurality of" is at least two, for example, two, three, etc., unless otherwise explicitly and specifically limited.
[0060] In the present application, the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" and the like mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the skilled person in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples, without contradiction.
[0061] Although the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are exemplary and are not to be construed as limiting the present application, and that changes, modifications, substitutions and variations can be made by those skilled in the art without departing from the scope of the present application.
Claims
1. A metabolite biomarker for early diagnosis of colorectal cancer, characterized in that, The marker is selected from one or more of the following groups: sulfate, ubiquinone-1, deoxycholate-glycine conjugate, demethyl chlorophyllin, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoic acid ester, and acyl AMP.
2. The marker according to claim 1, wherein the marker comprises the following: Sulfate, ubiquinone-1, deoxycholate glycine conjugate, demethyl chlorophyllin, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoic acid ester and acyl AMP.
3. The use of a reagent for detecting the level of a metabolite marker in the preparation of a kit for the early diagnosis of colorectal cancer, characterized in that, The metabolite markers are selected from one or more of the following groups: sulfate, ubiquinone-1, deoxycholate-glycine conjugate, demethyl chlorophyllin, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoic acid ester, and acyl AMP.
4. The use according to claim 2, characterized in that, The markers consist of the following: sulfate, ubiquinone-1, deoxycholate-glycine conjugate, demethyl chlorophyllin, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose 1,7-bisphosphate, cholesterol-docosahexaenoic acid ester, and acyl AMP.
5. The use according to claim 2, characterized in that, The reagent is used to detect the levels of the metabolite markers in blood samples. Optionally, the blood sample is serum.
6. A device for early diagnosis of colorectal cancer, characterized in that, include: A detection unit configured to detect the level of metabolite markers in a sample obtained from a subject; A calculation unit configured to input the levels of the aforementioned metabolite markers of the subject into an early colorectal cancer diagnosis model to calculate a colorectal cancer risk score; and An output unit, configured to output an early diagnostic result based on the risk score, The metabolite markers mentioned include: sulfate, ubiquinone-1, deoxycholate-glycine conjugate, demethyl chlorophyllin, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose-1,7-bisphosphate, cholesterol-docosahexaenoic acid ester, and acyl AMP.
7. The apparatus according to claim 6, characterized in that, The markers consist of the following: sulfate, ubiquinone-1, deoxycholate-glycine conjugate, demethyl phloroglucinone, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, sedoheptulose-1,7-bisphosphate, cholesterol-docosahexaenoic acid ester, and acyl AMP. Optionally, the detection unit uses liquid chromatography-mass spectrometry (LC-MS) to detect the level of the metabolite marker.
8. The apparatus according to claim 6, characterized in that, The early diagnosis model for colorectal cancer is a support vector machine diagnostic model, a random forest diagnostic model, an XGBoost diagnostic model, and a logistic regression diagnostic model. Preferably, the early diagnosis model for colorectal cancer is a random forest diagnostic model.
9. The apparatus according to claim 6, characterized in that, The device further includes a comparison unit that stores the levels of reference metabolite markers and their corresponding reference colorectal cancer risk scores. The comparison unit is configured to compare the subject's colorectal cancer risk score with the reference colorectal cancer risk score and / or compare the subject's levels of the metabolite markers with the levels of the reference metabolite markers.
10. The apparatus according to claim 9, characterized in that, Based on the following criteria, output results are generated indicating that the subjects are at an early risk of colorectal cancer: The level of the first biomarker among the metabolite biomarkers in the subject was higher than the level of the reference metabolite biomarker, and the level of the second biomarker among the metabolite biomarkers in the subject was lower than the level of the reference metabolite biomarker. The first biomarker was ubiquinone-1, deoxycholate-glycine conjugate, demethyl phloroglucinone, 3-(3,5-diiodo-4-hydroxyphenyl) lactate, vitamin K1, O-decanoyl-L-carnitine, cholesterol-docosahexaenoic acid ester, and acyl AMP. The second biomarker was sulfate and sedoheptulose 1,7-bisphosphate. The subject's colorectal cancer risk score was higher than the reference colorectal cancer risk score.