Metabolic marker combination for thyroid cancer diagnosis or auxiliary diagnosis and application thereof

Abstract

The invention provides a metabolic marker combination for thyroid cancer diagnosis or auxiliary diagnosis and application of the metabolic marker combination, and belongs to the technical field of in-vitro diagnosis. According to the invention, phosphatidyl ethanolamine 36: 6e, erucic acid, triglyceride 58: 2, lysophosphatidylcholine 22: 5, cholic acid, L-glutamic acid and L-valyl-L-serine are combined to construct a diagnosis model for distinguishing healthy and benign thyroid diseases and thyroid cancer, and the diagnosis model has high diagnosis efficiency and clinical diagnosis significance. Besides, diagnosis models for distinguishing healthy diseases, thyroid benign diseases and thyroid cancer are respectively constructed for 9 metabolic markers and 11 metabolic markers comprising the metabolic marker combination, and the diagnosis models constructed by the 9 metabolic markers and the 11 metabolic markers are proved to have relatively high diagnosis efficiency; healthy and benign thyroid diseases and thyroid cancer can be effectively distinguished, and clinical diagnosis significance is achieved.

Description

Technical Field

[0001] This invention belongs to the field of in vitro diagnostic technology, and in particular relates to a combination of metabolic markers for the diagnosis or auxiliary diagnosis of thyroid cancer and their applications. Background Technology

[0002] Thyroid cancer (TC) is one of the most common malignant tumors of the endocrine system, originating from the abnormal proliferation of thyroid follicular epithelial cells or parafollicular cells. Its incidence is steadily increasing globally, particularly among women, and the age of onset spans from adolescence to old age, with a significant proportion occurring in young and middle-aged adults aged 30-49. Early-stage thyroid cancer often presents with no obvious symptoms, typically only discovered during physical examinations as thyroid nodules. Common symptoms include painless neck masses, hoarseness, and difficulty swallowing; some patients may also experience cervical lymph node enlargement. In advanced stages, tumor invasion of surrounding tissues or distant metastasis can lead to respiratory distress, bone pain, and organ dysfunction, severely endangering life. Therefore, early screening and diagnosis are crucial for improving the prognosis of thyroid cancer patients and reducing recurrence rates.

[0003] Currently, the diagnosis of thyroid cancer mainly relies on high-resolution ultrasound and fine-needle aspiration biopsy (FNA). Ultrasound has an accuracy rate of only 60%–70% in differentiating between benign and malignant nodules, while FNA still yields 20%–30% of "indeterminate" cytological results. Furthermore, these methods are highly dependent on the experience of sonographers and cytopathologists, and are expensive, invasive, and carry risks of complications (bleeding, recurrent laryngeal nerve injury), leading to poor patient compliance and making them unsuitable for large-scale population screening. In addition, thyroid cancer needs to be differentiated from benign lesions such as nodular goiter, thyroiditis, follicular thyroid adenoma, and parathyroid adenoma. Therefore, there is an urgent clinical need for a highly sensitive, highly specific, cost-effective, and convenient diagnostic strategy for thyroid cancer to compensate for the shortcomings of current technologies. Summary of the Invention

[0004] The purpose of this invention is to provide a combination of metabolic biomarkers for the diagnosis or auxiliary diagnosis of thyroid cancer and its application. This invention provides a combination of metabolic biomarkers that is highly specific, highly sensitive, economical and convenient, and can be used to distinguish between healthy individuals, benign thyroid diseases and thyroid cancer.

[0005] This invention provides a combination of metabolic markers for the diagnosis or auxiliary diagnosis of thyroid cancer, comprising: phosphatidylethanolamine 36:6e, erucic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, cholic acid, L-glutamic acid, and L-valine-L-serine.

[0006] Preferably, the combination of metabolic markers includes: phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, cholic acid, L-glutamate, L-valine-L-serine, and L-octanoylcarnitine.

[0007] Preferably, the combination of metabolic markers includes: phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, decylcarnitine 10:0, triglycerides 58:2, lysophosphatidylcholine 22:5, cholic acid, hypoxanthine, L-glutamic acid, L-valine-L-serine, and L-octylcarnitine.

[0008] Preferably, the metabolic markers include blood metabolic markers.

[0009] The present invention also provides the application of the metabolic biomarker combination described above as a diagnostic biomarker in the preparation of products for the diagnosis or auxiliary diagnosis of thyroid cancer.

[0010] Preferably, the diagnosis or auxiliary diagnosis of thyroid cancer includes distinguishing between healthy individuals, benign thyroid diseases, and thyroid cancer.

[0011] Preferably, the benign thyroid disease includes one or more of nodular goiter, thyroiditis, thyroid follicular adenoma, and parathyroid adenoma.

[0012] Preferably, the product for detecting the combination of metabolic biomarkers according to any one of claims 1 to 4 includes a liquid chromatography-mass spectrometry detection reagent.

[0013] The present invention also provides a thyroid cancer diagnostic system based on the metabolic biomarker combination described above, comprising: a metabolic biomarker combination detection module for detecting the concentration of each metabolic biomarker in the metabolic biomarker combination in the subject's blood; a calculation module for calculating the area under the operating characteristic curve of the subject based on the concentration of each metabolic biomarker; and an evaluation report generation module for evaluating the subject's thyroid cancer based on the area under the operating characteristic curve through an evaluation program.

[0014] The present invention also provides a method for constructing a differential diagnostic model for healthy individuals, benign thyroid diseases, and thyroid cancer based on the combination of metabolic biomarkers described above, comprising the following steps: randomly using 3 / 4 of the blood sample data of the modeling group as the training set and 1 / 4 as the test set for learning, and using a support vector machine for random loop iteration 1000 times, and constructing a differential diagnostic model for healthy individuals, benign thyroid diseases, and thyroid cancer by statistically calculating the average accuracy of the final model.

[0015] This invention provides a combination of metabolic biomarkers for the diagnosis or auxiliary diagnosis of thyroid cancer, comprising: phosphatidylethanolamine 36:6e, erucic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acids, L-glutamate, and L-valine-L-serine. This invention utilizes the combination of phosphatidylethanolamine 36:6e, erucic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acids, L-glutamate, and L-valine-L-serine to construct a diagnostic model that distinguishes between healthy individuals, benign thyroid diseases, and thyroid cancer. This diagnostic model exhibits high diagnostic efficacy and has clinical diagnostic significance. In addition, this invention has constructed diagnostic models for differentiating between healthy individuals, benign thyroid diseases, and thyroid cancer using nine metabolic markers (phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acid, L-glutamate, L-valine-L-serine, and L-caprylcarnitine) and eleven metabolic markers (phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, decylcarnitine 10:0, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acid, hypoxanthine, L-glutamate, L-valine-L-serine, and L-caprylcarnitine), respectively. Verification has shown that the diagnostic models constructed using the aforementioned nine and eleven metabolic markers all possess high diagnostic efficacy and can effectively differentiate between healthy individuals, benign thyroid diseases, and thyroid cancer, thus having clinical diagnostic significance. In summary, this invention constructs a highly sensitive and specific diagnostic model for thyroid cancer based on metabolomics data from blood samples, providing a new and effective tool for the early detection, screening, and precise diagnosis of thyroid cancer. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 Multivariate ROC curve analysis of 11 key biomarkers for thyroid cancer diagnosis in the modeling group; Figure 2 Multivariate ROC curve analysis was performed on 11 key biomarkers for thyroid cancer diagnosis in the validation group; Figure 3 To validate the multivariate ROC curve analysis of nine key biomarkers for thyroid cancer diagnosis in the validation group; Figure 4 To validate the use of multivariate ROC curve analysis for seven key biomarkers for thyroid cancer diagnosis in the validation group. Detailed Implementation

[0018] This invention provides a combination of metabolic markers for the diagnosis or auxiliary diagnosis of thyroid cancer, comprising: phosphatidylethanolamine 36:6e, erucic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, cholic acid, L-glutamic acid, and L-valine-L-serine.

[0019] This invention reveals metabolic reprogramming characteristics in disease states by systematically analyzing the small molecule metabolite profiles in biological samples such as serum, plasma, and urine from healthy individuals, patients with benign thyroid diseases, and patients with thyroid cancer. Furthermore, the diagnostic model constructed based on the aforementioned metabolic biomarker combination helps provide auxiliary identification information when nodules are detected through imaging examinations, improving diagnostic accuracy and reducing unnecessary invasive examinations to some extent. Therefore, this invention, based on a combination of metabolic biomarkers from blood samples, can establish a thyroid cancer diagnostic model with excellent sensitivity and specificity, providing a new and effective tool for the early detection, screening, and accurate diagnosis of thyroid cancer.

[0020] In one implementation, the metabolic biomarker combination comprises: phosphatidylethanolamine 36:6e, erucic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acid, L-glutamate, and L-valine-L-serine. The diagnostic model constructed using the above seven metabolic biomarker combinations in the validation group showed an AUC of 0.867 (sensitivity = 0.710, specificity = 0.875), indicating good discriminative performance.

[0021] In one implementation, the metabolic marker combination includes: phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acid, L-glutamate, L-valine-L-serine, and L-caprylcarnitine; further, the metabolic marker combination is composed of: phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acid, L-glutamate, L-valine-L-serine, and L-caprylcarnitine. The diagnostic model constructed using the above nine metabolic marker combinations in the validation group had an AUC of 0.875 (sensitivity = 0.749, specificity = 0.894), indicating that the diagnostic model exhibits good discriminative performance.

[0022] In one embodiment, the metabolic marker combination includes: phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, decylcarnitine 10:0, triglycerides 58:2, lysophosphatidylcholine 22:5, cholic acid, hypoxanthine, L-glutamate, L-valine-L-serine, and L-caprylcarnitine; further, the metabolic marker combination is composed of: phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, decylcarnitine 10:0, triglycerides 58:2, lysophosphatidylcholine 22:5, cholic acid, hypoxanthine, L-glutamate, L-valine-L-serine, and L-caprylcarnitine. The 36:6e ratio of phosphatidylethanolamine 36:6e indicates that the total number of carbon atoms in the two fatty chains of this phosphatidylethanolamine molecule is 36, and the total number of double bonds is 6 (e represents an ether bond structure); the 10:0 ratio of decanoylcarnitine 10:0 indicates that the fatty chain portion of this decanoylcarnitine molecule contains 10 carbon atoms and has 0 double bonds; the 58:2 ratio of triglycerides 58:2 indicates that the total number of carbon atoms in the three fatty chains of this triglyceride molecule is 58, and the total number of double bonds is 2; the 22:5 ratio of lysophosphatidylcholine 22:5 indicates that the fatty chain of this lysophosphatidylcholine (single chain) contains 22 carbon atoms and has 5 double bonds. The diagnostic model constructed using the above 11 metabolic biomarkers in the validation group had an AUC of 0.888 (sensitivity = 0.794, specificity = 0.918), indicating that the diagnostic model has high discriminative power.

[0023] In one embodiment, the metabolic markers include blood metabolic markers, and more specifically, plasma metabolic markers.

[0024] This invention detects the combination of metabolic markers in blood samples, which can effectively distinguish between patients with thyroid cancer, patients with benign thyroid diseases, and healthy individuals with just one test. It can significantly reduce unnecessary puncture examinations and even excessive surgeries for patients with benign thyroid diseases, reduce the waste of medical resources, and reduce the physical and mental burden and medical expenses of patients, thereby achieving early screening, early diagnosis, and accurate triage.

[0025] The present invention also provides the application of the metabolic biomarker combination described above as a diagnostic biomarker in the preparation of products for the diagnosis or auxiliary diagnosis of thyroid cancer.

[0026] In one implementation, the product includes reagents, kits, diagnostic systems, and devices.

[0027] In one implementation, the thyroid cancer diagnosis or auxiliary diagnosis includes distinguishing between healthy individuals, benign thyroid diseases, and thyroid cancer.

[0028] In one implementation, the benign thyroid disease includes one or more of nodular goiter, thyroiditis, thyroid follicular adenoma, and parathyroid adenoma.

[0029] As one implementation, the product for detecting the combination of metabolic biomarkers described in the above scheme includes a liquid chromatography-mass spectrometry detection reagent.

[0030] The present invention also provides a thyroid cancer diagnostic system based on the combination of metabolic markers described above, comprising: A metabolic biomarker combination detection module is used to detect the concentration of each metabolic biomarker in the said metabolic biomarker combination in the blood of a subject; The calculation module is used to calculate the area under the subject operating characteristic curve based on the concentration of each metabolic biomarker; The assessment report generation module is used to assess the subject's thyroid cancer based on the area under the working characteristic curve through an assessment procedure.

[0031] This invention also provides a method for constructing a differential diagnostic model for healthy individuals, benign thyroid diseases, and thyroid cancer based on the metabolic biomarker combination described above, comprising the following steps: Three-quarters of the blood sample data from the modeling group were randomly used as the training set and one-quarter as the test set for learning. The support vector machine was used for random loop iteration 1000 times. By statistically calculating the average accuracy of the final model, a differential diagnostic model for healthy individuals, benign thyroid diseases, and thyroid cancer was constructed.

[0032] Compared with existing technologies, this invention utilizes the combination of metabolic biomarkers described above, combined with machine learning algorithms, to construct a diagnostic model capable of distinguishing between healthy individuals, benign thyroid diseases, and thyroid cancer. This method is simple to operate, low in cost, and readily available for blood samples, making it more suitable for large-scale thyroid cancer screening and long-term patient follow-up monitoring in areas with relatively limited medical resources.

[0033] To further illustrate the present invention, the following detailed description, in conjunction with the accompanying drawings and embodiments, provides a combination of metabolic biomarkers for the diagnosis or auxiliary diagnosis of thyroid cancer and their applications, but these descriptions should not be construed as limiting the scope of protection of the present invention.

[0034] Example 1 Detailed implementation plan: 1. Subject Information 1) Inclusion criteria: Participants must meet all of the following inclusion criteria to be eligible to participate in this study: (1) Male or female aged ≥18 years; (2) Read and fully understand the informed consent form, sign the informed consent form, and be able to provide a blood sample for metabolomics testing; (3) Thyroid cancer patients are diagnosed by biopsy / postoperative pathology or by comprehensive clinical assessment by a clinician; (4) Individuals who have been excluded from thyroid cancer by biopsy / postoperative pathology or by comprehensive clinical assessment by a clinician and are diagnosed by biopsy / pathology as having benign thyroid diseases, including but not limited to nodular goiter, thyroiditis, thyroid follicular adenoma and parathyroid adenoma.

[0035] 2) Exclusion criteria: Subjects who meet any of the following exclusion criteria are ineligible to participate in this study: (1) During pregnancy or lactation; (2) Emergency room visit or resuscitation required; (3) History of blood transfusion within 7 days prior to sampling; (4) People who have received organ transplants or have previously received non-autologous (allogeneic) bone marrow or stem cell transplants; (5) History of malignant tumor within 5 years or any anti-tumor treatment before sampling; (6) Simultaneous co-occurrence of multiple primary malignant tumors.

[0036] 3) Subject information This study collected plasma samples from 230 participants across two medical centers, including 86 healthy controls (HC), 58 participants with benign thyroid disease (BTD), and 86 participants with thyroid cancer (TC). Specifically, the plasma samples used for the modeling group consisted of 64 participants from the healthy controls (HC), 44 participants from the BTD group (including 29 with nodular goiter, 8 with thyroiditis, 6 with follicular thyroid adenomas, and 1 with parathyroid adenoma), and 64 participants from the TC group. The plasma samples used for the validation group consisted of 22 participants from the healthy controls (HC), 14 participants from the BTD group (including 8 with nodular goiter, 3 with thyroiditis, 2 with follicular thyroid adenomas, and 1 with parathyroid adenoma), and 22 participants from the TC group (Table 1).

[0037] Table 1 Subject Information

[0038] 2. Plasma metabolite detection 1) Test reagents: Methanol, acetonitrile, water, acetic acid, methyl tert-butyl ether of mass spectrometry grade, and formic acid of chromatographic (HPLC) grade were all purchased from Sigma-Aldrich, USA.

[0039] 2) Sample preparation: Take 100 μL of plasma and place it in 1000 μL of pre-cooled solution (methyl tert-butyl ether: methanol, volume ratio 3:1). Vortex to mix the extracted blood sample and obtain the sample extract. Add 500 μL of solution (methanol: water, volume ratio 3:1) to the sample extract, sonicate, let stand, vortex and centrifuge to separate the layers. The upper layer is the organic phase and the lower layer is the aqueous phase.

[0040] - Organic phase: After the sample is separated into layers, take 500 μL of the upper organic phase into a centrifuge tube, dry it, add 200 μL of solution (acetonitrile:isopropanol, volume ratio 3:1), and incubate at room temperature for 15 min. After incubation, vortex the centrifuge tube, sonicate for 5 min, and then centrifuge at room temperature for 5 min (12000 rpm). Take 180 μL of the supernatant from the centrifuge tube into a 2 mL glass vial, which is the organic phase test solution, and perform LC-MS analysis.

[0041] - Aqueous phase: After sample separation, transfer the lower 400 μL aqueous phase to a centrifuge tube and add 1100 μL of ice-cold methanol to precipitate proteins. After protein precipitation, centrifuge the tube and transfer 1000 μL of the supernatant to a new centrifuge tube, then dry overnight. Add 200 μL of water to the dried centrifuge tube and incubate at room temperature for 15 min. After incubation, vortex the mixture, sonicate for 5 min, and then centrifuge at room temperature for 5 min (12000 rpm). Transfer 180 μL of the supernatant from the centrifuge tube to a 2 mL glass vial as the aqueous phase test solution. Analyze using LC-MS.

[0042] 3) Detection of small molecule metabolites: Organic phase use Columns, aqueous phase use Columns were used for small molecule separation; both liquid chromatography and mass spectrometry used were ACQUITY UPLC I-Class liquid chromatography system (Waters) and Q-Exactive mass spectrometry system (Thermo Fisher Scientific).

[0043] The mobile phase parameters are as follows: Organic phase analyte mobile phase parameters – Mobile phase A is an aqueous solution containing 0.1% acetic acid and 10 mmol / L ammonium acetate; Mobile phase B is an acetonitrile-isopropanol (7:3 v / v) solution containing 0.1% acetic acid and 10 mmol / L ammonium acetate. The separation elution gradient is as follows: 0-12 min is 55%-89% mobile phase B, 12-19.5 min is 100% mobile phase B.

[0044] Mobile phase parameters of the aqueous test solution – Mobile phase A is an aqueous solution containing 0.1% formic acid; Mobile phase B is an acetonitrile solution containing 0.1% formic acid. The separation and elution gradient is as follows: 0-13 min is 1%-70% mobile phase B, 13-18 min is 99% mobile phase B.

[0045] The mass spectrometry parameters are as follows: Mass spectrometry data were acquired using Full MS and Full MS / dd-MS2 (each with both positive and negative modes). The parameters used by QExactive were as follows: Full MS mode had a resolution of 70,000 m / z, a scan range of 100-1500 m / z, an Automatic Gain Control (AGC) of 3E+6, and a Maximum IT of 200 ms; in Full MS / dd-MS2 mode, the resolution of the secondary mass spectrometer was 17,500 m / z, the quadrupole window was 1.5 m / z, the AGC was 1E+5, the maximum ion implantation time was 50 ms, and the Higher Energy Collisional Dissociation (HCD) was 30%.

[0046] 3. Metabolomics data preprocessing and metabolite identification 1) Metabolomics data processing: (1) Extract peaks from the RAW format file of the mass spectrometer and convert it into a FeatureXML format file to reduce the dimensionality of the original mass spectrometry data and improve the signal-to-noise ratio; (2) Use the peak alignment algorithm of OpenMS software to correct and align the retention time of the extracted peak format data between samples, thereby converting the mass spectrometry data into a data matrix; (3) Match and filter the isotope peaks in the data matrix obtained in step 2, and replace abnormal data (0, negative values, background noise, etc.) with missing values; (4) Remove the characteristic peaks with a detection rate of <80% from all the characteristic peaks obtained in step 3, fill the median value of the characteristic peak with the characteristic peaks with a detection rate of >80%, and add 5% random noise (following a standard normal distribution); (5) In order to reduce the difference in metabolite concentration between samples and make the data distribution more symmetrical, use Normalization Autoencoder (NormAE) to perform normalization processing to remove systematic errors such as batch effects.

[0047] 2) Identification of metabolites: After analyzing the raw data using software, the spectral information of the primary precursor ion (MS1) and secondary fragment ion (MS2) of the compound is obtained. This information, such as the mass-to-charge ratio (m / z) of the primary mass spectrometer and the fragment ion data, is matched with the spectral information of primary and secondary metabolites in public databases to qualitatively identify the metabolites. Commonly used metabolite databases include the Human Metabolite Database (HMDB, www.hmdb.ca), the Metabolomics Database (Metlin, metlin.scripps.edu), the Mass Spectrometry Database (www.massbank.jp), and the Lipid Map Database (Lipidmap, www.lipidmaps.org). Metabolites identified based on these databases are then finally validated using retention times, MS1, and MS2 mass spectrometry data obtained from separation of standards under the same chromatographic column and mass spectrometry conditions. The criteria for metabolite identification are a retention time difference within 0.1 min and a theoretical and measured molecular weight difference of less than 10 ppm.

[0048] 4. Data Analysis 1) Biomarker screening Metabolite detection was performed on the above samples, and a total of 570 metabolites were obtained after annotation. LASSO (Least Absolute Shrinkage and Selection Operator) regression analysis was performed on the data of the modeling group. The average error corresponding to each regularization parameter alpha was calculated using 5-fold cross-validation. The optimal alpha with the smallest error was found to be 0.0823. Metabolites with non-zero regression coefficients in their corresponding models were retained. Finally, 11 differential metabolites were selected (Table 2) as important metabolic markers to distinguish between health, benign thyroid diseases and thyroid cancer.

[0049] Table 2. 11 Important Metabolic Markers for Differentiating HC, BTD, and TC

[0050] 2) Development of a diagnostic model to differentiate between healthy individuals, benign thyroid diseases, and thyroid cancer. To verify the discriminative effect of the 11 selected metabolic biomarkers in distinguishing between health, benign thyroid diseases, and thyroid cancer, multivariate ROC curve analysis was performed on the 11 metabolic biomarkers in the modeling group. Three-quarters of the sample data from the HC, BTD, and TC groups in the modeling group were randomly used as the training set and one-quarter as the test set for training. The model was then iterated 1000 times using a support vector machine (SVM) machine learning method. By statistically analyzing the average accuracy of the final model, a diagnostic model that distinguishes between health, benign thyroid diseases, and thyroid cancer was constructed.

[0051] ROC curves are a method for studying the relationship between model sensitivity and specificity. Sensitivity is plotted on the ordinate, and 1-specificity on the x-axis. The evaluation criterion is the area under the curve (AUC). An AUC greater than 0.5, and closer to 1, indicates better model performance and better discrimination. An AUC less than 0.5 indicates poor model accuracy. ROC classification prediction models, in addition to common parameters such as the receiver operating characteristic (ROC) curve and AUC, also include sensitivity and specificity.

[0052] Sensitivity calculation is shown in Formula I: Formula I.

[0053] Specificity is calculated using Formula II: Formula II.

[0054] Among them, TP (True Positive): True positive, the number of samples that are actually positive but were correctly predicted as positive; TN (True Negative): The number of samples that were actually negative but were correctly predicted as negative. FP (False Positive): The number of samples that are actually negative but are incorrectly predicted as positive. FN (False Negative): The number of samples that are actually positive but are incorrectly predicted as negative. The results are as follows Figure 1 As shown, AUC=0.908 (sensitivity=0.795, specificity=0.905), indicating that the constructed diagnostic model has high discriminative power.

[0055] In addition, ROC curve analysis was performed on diagnostic models with different combinations of metabolic markers in the modeling group. Nine metabolic markers were used: phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acid, L-glutamate, L-valine-L-serine and L-octylcarnitine combination. Seven metabolic markers were used: phosphatidylethanolamine 36:6e, erucic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acid, L-glutamate and L-valine-L-serine combination; The results showed that when using 9 metabolic markers, the AUC was 0.878 (sensitivity = 0.777, specificity = 0.905); when using 7 metabolic markers, the AUC was 0.879 (sensitivity = 0.777, specificity = 0.905), both demonstrating stable discrimination ability.

[0056] 3) Validation of diagnostic models used to differentiate between healthy individuals, benign thyroid diseases, and thyroid cancer. To further validate the effectiveness of the diagnostic model for distinguishing between healthy individuals, benign thyroid diseases, and thyroid cancer, built based on the modeling group data, the model was validated using the validation group data. Multivariate ROC curve analysis was performed to evaluate the model's independent validation performance on unknown datasets outside the modeling group dataset. The results showed an AUC of 0.888 (sensitivity = 0.794, specificity = 0.918). These results indicate that the established diagnostic model for distinguishing between healthy individuals, benign thyroid diseases, and thyroid cancer also exhibits good discriminative performance in the validation group. (See attached results). Figure 2 It should be noted that during the validation process, the validation group samples are input into the three-class classification model constructed by the modeling group. The model outputs probability values ​​for the three classes for each sample. According to the "maximum probability principle," the model selects the class with the highest probability value as the final classification result for that sample. Specifically, when the probability value of a sample in a certain class is the highest, it is classified into that class.

[0057] Furthermore, diagnostic models using different combinations of metabolic biomarkers were validated in the validation group. Multivariate ROC curve analysis showed that the combination of the nine metabolic biomarkers—phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acid, L-glutamate, L-valine-L-serine, and L-octanoylcarnitine—achieved an AUC of 0.875 in the validation group (sensitivity = 0.749, specificity = 0.894). Figure 3As shown; the above seven metabolic markers, phosphatidylethanolamine 36:6e, erucic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, bile acids, L-glutamate, and the combination of L-valine-L-serine, showed an AUC of 0.867 in the validation group (sensitivity = 0.710, specificity = 0.875), as shown in the results. Figure 4 As shown above, the established diagnostic model also demonstrates good discriminative performance in the validation group.

[0058] Although the above embodiments have provided a detailed description of the present invention, they are only some embodiments of the present invention, and not all embodiments. People can obtain other embodiments based on these embodiments without creative effort, and these embodiments all fall within the protection scope of the present invention.

Claims

1. A combination of metabolic biomarkers for the diagnosis or auxiliary diagnosis of thyroid cancer, characterized in that, include: Phosphatidylethanolamine 36:6e, erucic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, cholic acid, L-glutamic acid and L-valine-L-serine.

2. The metabolic biomarker combination according to claim 1, characterized in that, The combination of metabolic markers includes: phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, triglycerides 58:2, lysophosphatidylcholine 22:5, cholic acid, L-glutamate, L-valine-L-serine, and L-octanoylcarnitine.

3. The metabolic biomarker combination according to claim 1, characterized in that, The combination of metabolic markers includes: phosphatidylethanolamine 36:6e, erucic acid, linolenic acid, decylcarnitine 10:0, triglycerides 58:2, lysophosphatidylcholine 22:5, cholic acid, hypoxanthine, L-glutamate, L-valine-L-serine, and L-octylcarnitine.

4. The metabolic biomarker combination according to claim 1, characterized in that, The metabolic markers include blood metabolic markers.

5. The use of the combination of metabolic markers according to any one of claims 1 to 4 as diagnostic markers in the preparation of products for the diagnosis or auxiliary diagnosis of thyroid cancer.

6. The application according to claim 5, characterized in that, The diagnosis or auxiliary diagnosis of thyroid cancer includes distinguishing between healthy individuals, benign thyroid diseases, and thyroid cancer.

7. The application according to claim 5, characterized in that, The benign thyroid diseases mentioned include one or more of the following: nodular goiter, thyroiditis, thyroid follicular adenoma, and parathyroid adenoma.

8. The application according to claim 5, characterized in that, The product for detecting the combination of metabolic biomarkers according to any one of claims 1 to 4 includes a liquid chromatography-mass spectrometry detection reagent.

9. A thyroid cancer diagnostic system based on a combination of metabolic markers according to any one of claims 1 to 4, characterized in that, include: A metabolic biomarker combination detection module is used to detect the concentration of each metabolic biomarker in the said metabolic biomarker combination in the blood of a subject; The calculation module is used to calculate the area under the subject operating characteristic curve based on the concentration of each metabolic biomarker; The assessment report generation module is used to assess the subject's thyroid cancer based on the area under the working characteristic curve through an assessment procedure.

10. A method for constructing a differential diagnostic model for healthy individuals, benign thyroid diseases, and thyroid cancer based on a combination of metabolic biomarkers according to any one of claims 1 to 4, characterized in that, Includes the following steps: Three-quarters of the blood sample data from the modeling group were randomly used as the training set and one-quarter as the test set for learning. The support vector machine was used for random loop iteration 1000 times. By statistically calculating the average accuracy of the final model, a differential diagnostic model for healthy individuals, benign thyroid diseases, and thyroid cancer was constructed.

Images