Metabolic marker for diagnosing lung function grading in stable phase of chronic obstructive pulmonary disease and application of metabolic marker

Abstract

The invention discloses a metabolic marker for diagnosing lung function grading in a stable phase of chronic obstructive pulmonary disease and application of the metabolic marker. According to the invention, in a training set, the lysophosphatidylcholine (17: 0 / 0: 0) is combined with the dimethylglycine or the lysophosphatidylcholine (17: 0 / 0: 0), the dimethylglycine is combined with the L-pyroglutamic acid or the lysophosphatidylcholine (17: 0 / 0: 0), the dimethylglycine is combined with the 5-hydroxyindoleacetic acid or the lysophosphatidylcholine (17: 0 / 0: 0), and the dimethylglycine is combined with the L-pyroglutamic acid or the lysophosphatidylcholine (17: 0 / 0: 0). The combination of the L-pyroglutamic acid and the 5-hydroxyindoleacetic acid has excellent diagnosis efficiency in diagnosing and distinguishing lung function 1-2 level and 3-4 level patients of the chronic obstructive pulmonary disease in the stable phase; further, verification results in a verification set show that the four combined metabolic markers have excellent accuracy in combined diagnosis and distinguishing of patients with the lung functions of the chronic obstructive pulmonary disease at the 1-2 level and the 3-4 level in the stable phase. Therefore, the four combined metabolic markers have the prospect of being developed into a kit for diagnosing and distinguishing lung function 1-2 level and 3-4 level patients of the stable-phase chronic obstructive pulmonary disease.

Description

Technical Field

[0001] This invention belongs to the field of detection and relates to disease diagnostic biomarkers and their applications, specifically to metabolic biomarkers used for diagnosing lung function classification in the stable phase of chronic obstructive pulmonary disease and their applications. Background Technology

[0002] Chronic obstructive pulmonary disease (COPD) is a chronic, progressive respiratory disease that has become one of the leading causes of death from respiratory illnesses.

[0003] COPD progresses gradually, with persistent airflow limitation, progressive decline in lung function, and repeated acute exacerbations leading to disease progression and ultimately respiratory failure and death. Based on clinical symptoms and pathological characteristics, and in accordance with the Global Initiative for Chronic Obstructive Lung Disease (GOLD), stable COPD is classified into lung function grades 1-4 based on the percentage of predicted forced expiratory volume in one second (FEV1%). According to the Guidelines for the Integrated Traditional Chinese and Western Medicine Diagnosis and Treatment of Chronic Obstructive Lung Disease, patients in different grades of stable COPD have different clinical characteristics. Patients in lung function grades 1-2 have mild clinical symptoms, but their lung function has already declined rapidly; the goal of prevention and treatment is to protect lung function and slow disease progression. Patients in lung function grades 3-4 have more severe clinical symptoms, including significant dyspnea, frequent acute exacerbations, increased hospitalizations, and gradually increasing mortality and disability rates; the goal of prevention and treatment is to reduce acute exacerbations and improve quality of life. Therefore, taking corresponding treatment measures for patients with COPD in different stages of stable condition is of great significance for improving clinical efficacy, patients' quality of life and prognosis.

[0004] Currently, tiered diagnosis mainly relies on lung function tests, but this method still has many limitations in practical applications:

[0005] 1. Insufficient sensitivity: In the early stages of the disease, there may already be significant structural damage to the lung tissue (such as airway remodeling and alveolar destruction), but the changes in FEV1% predicted are not obvious, leading to missed or delayed diagnosis;

[0006] 2. Lagging dynamic monitoring: Changes in lung function indicators lag behind early physiological changes such as inflammation and metabolic abnormalities, making it difficult to reflect the dynamic changes in the condition in a timely manner, resulting in treatment adjustments lagging behind disease progression;

[0007] 3. High operational dependence: Lung function testing is highly dependent on patient cooperation and standardized operation, which poses a challenge, especially for elderly or critically ill patients, and the repeatability and reliability of the test are limited.

[0008] 4. Limited dimensions of reflection: Lung function tests mainly reflect the degree of airflow limitation, which is difficult to fully reveal the pathological mechanisms of systemic inflammation, metabolic disorders, immune imbalance and other diseases, resulting in insufficient ability to identify disease heterogeneity.

[0009] Therefore, relying solely on FEV1% predicted as an indicator for COPD classification and dynamic monitoring is no longer sufficient to meet the development needs of precision medicine. There is an urgent need to introduce a new biological indicator system as an effective supplement to FEV1% predicted, to improve early identification capabilities and the scientific nature of risk stratification, thereby providing a more comprehensive and reliable basis for achieving precise classification and treatment of COPD.

[0010] Metabolomics is a discipline that systematically analyzes the composition, relative abundance, and dynamic changes of all small molecule metabolites in an organism, reflecting the combined effects of upstream genotype and environmental exposure by capturing end-phenotype metabolites. In recent years, metabolomics technology has been widely applied to the study of COPD mechanisms. Multiple metabolic pathways (including mitochondrial dysfunction, oxidative stress response, and lipid and amino acid metabolism disorders) are closely related to COPD staging, and abnormal changes in metabolite concentrations often precede detectable declines in lung function, showing significant differences in the early stages of the disease. Despite continuous enrichment of basic research, current work largely remains at the stage of candidate molecule discovery; moreover, detection methods are complex, costly, and have low throughput, making it difficult to meet routine clinical testing needs.

[0011] To overcome the shortcomings of the existing technology, this invention is proposed. Summary of the Invention

[0012] The purpose of this invention is to overcome the shortcomings of the prior art and provide metabolic biomarkers for diagnosing lung function classification in the stable phase of chronic obstructive pulmonary disease and their applications.

[0013] The above-mentioned objective of this invention is achieved through the following technical solution:

[0014] Application of lysophosphatidylcholine (17:0 / 0:0) in combination with dimethylglycine in the preparation of a diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease.

[0015] Preferably, the kit is a serum detection kit.

[0016] Application of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine and L-pyroglutamic acid in the preparation of a diagnostic kit to differentiate between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease.

[0017] Preferably, the kit is a serum detection kit.

[0018] Application of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine and 5-hydroxyindoleacetic acid in the preparation of a diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease.

[0019] Preferably, the kit is a serum detection kit.

[0020] Application of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, L-pyroglutamic acid and 5-hydroxyindoleacetic acid in the preparation of a diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease.

[0021] Preferably, the kit is a serum detection kit.

[0022] Beneficial effects:

[0023] This invention found that, in the training set, the combination of lysophosphatidylcholine (17:0 / 0:0) with dimethylglycine or lysophosphatidylcholine (17:0 / 0:0), dimethylglycine with L-pyroglutamate or lysophosphatidylcholine (17:0 / 0:0), dimethylglycine with 5-hydroxyindoleacetic acid or lysophosphatidylcholine (17:0 / 0:0), and dimethylglycine, L-pyroglutamate, and 5-hydroxyindoleacetic acid has excellent diagnostic efficacy in differentiating stable COPD patients from grade 1-2 to grade 3-4. All combinations achieved an AUC > 0.9, and other indicators were also relatively superior. Further validation results in the validation set showed that the combined use of the above four combinations of metabolic markers to differentiate stable COPD patients from grade 1-2 to grade 3-4 had excellent accuracy, with all combinations exceeding 80%. Therefore, the above four combinations of metabolic biomarkers have the potential to be developed into diagnostic kits that differentiate between stable COPD patients with pulmonary function grades 1-2 and 3-4. Attached Figure Description

[0024] Figure 1 The differences in serum levels of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, L-pyroglutamic acid, and 5-hydroxyindoleacetic acid in patients with stable COPD pulmonary function grades 1-2 and 3-4 were investigated. Detailed Implementation

[0025] The following describes the substantive content of the present invention in detail with reference to embodiments, but this does not limit the scope of protection of the present invention.

[0026] Example 1: Diagnostic efficacy of target metabolites in differentiating stable COPD patients with pulmonary function grades 1-2 from 3-4

[0027] I. Experimental Samples

[0028] Training set: Serum samples were collected from the First Affiliated Hospital of Anhui University of Traditional Chinese Medicine, Jiangsu Provincial Hospital of Traditional Chinese Medicine, Neixiang County People's Hospital, Affiliated Hospital of Shaanxi University of Traditional Chinese Medicine, Songxian County People's Hospital, and Zhumadian Central Hospital. Following strict screening and exclusion criteria, 355 COPD patients with pulmonary function grades 1-2 and 119 COPD patients with pulmonary function grades 3-4 were age- and sex-matched. Baseline characteristics are shown in Table 1. All clinical data are presented as median ± standard deviation.

[0029] Table 1. Baseline characteristics of subjects in the training set (median ± standard deviation)

[0030] GOLD Level 1-2 GOLD 3-4 Age (years) 66 ± 7.64 67 ± 8.60 male 259（72.96%） 100（84.03%） <![CDATA[BMI(kg m -2 )]]> 23.18 ± 3.37 21.95 ± 3.22 Smoking 205（57.75%） 79（66.39%）

[0031] Validation set: Serum samples were collected from the First Affiliated Hospital of Henan University of Traditional Chinese Medicine. Following strict screening and exclusion criteria, 32 COPD patients with pulmonary function grades 1-2 and 37 COPD patients with pulmonary function grades 3-4 were age- and sex-matched. Baseline characteristics are shown in Table 2. All clinical data are presented as median ± standard deviation.

[0032] Table 2. Baseline characteristics of participants in the validation set (median ± standard deviation)

[0033] GOLD Level 1-2 GOLD 3-4 Age (years) 63 ± 6.92 67 ± 7.28 male 24（75.00%） 34（91.89%） <![CDATA[BMI(kg m -2 )]]> 23.18 ± 3.19 22.04 ± 3.59 Smoking 11（65.63%） 79（81.08%）

[0034] COPD diagnostic criteria: According to GOLD (2025), forced expiratory volume in one second (FEV1) / forced vital capacity (FVC) < 0.7 after using a bronchodilator;

[0035] Diagnostic criteria for COPD grade 1-2: FEV1 ≥ 50% predicted;

[0036] Diagnostic criteria for COPD grade 3-4: FEV1 < 50% predicted.

[0037] II. Experimental Instruments and Reagents

[0038] HPLC-grade methanol and isopropanol, HPLC-grade ammonium acetate, and LC-MS-grade acetonitrile and formic acid were purchased from Thermo Fisher Scientific, Shanghai, China. Dimethylglycine, L-pyroglutamic acid, and 5-hydroxyindoleacetic acid standards were purchased from Shanghai Yuanye Biotechnology Co., Ltd. Lysophosphatidylcholine (17:0 / 0:0) standards were purchased from Merck, USA. Ultrapure water was prepared using a laboratory Milli-Q ultrapure water purification system and purchased from Merck Millipore, Shanghai, China. A high-speed benchtop refrigerated centrifuge and centrifugal concentrator were purchased from Thermo Fisher Scientific, USA, and an electronic balance was purchased from Mettler Toledo Instruments (Shanghai) Co., Ltd.

[0039] III. Experimental Methods

[0040] 1. Collection and processing of serum samples

[0041] Fasting peripheral blood was collected from the patient in the morning and placed in a test tube without anticoagulant. The blood was allowed to coagulate naturally at room temperature for 30 minutes. After the blood had coagulated, it was centrifuged at 3000 rpm for 10 minutes. The clear serum liquid at the top was carefully aspirated into a sterile lyophilized tube, labeled, and stored at -80°C for later use.

[0042] 2. Determination of target metabolite content in serum by UHPLC-QQQ MS

[0043] Testing instrument: SCIEX Triple Quad 6500 LC-MS / MS purchased from AB Sciex LLC, USA.

[0044] Liquid chromatography conditions: Column: Synergi TM The column temperature was 35℃, and the mobile phase was 0.05% formic acid in water and 0.05% formic acid in acetonitrile. The flow rate was 0.5 mL / min, the injection volume was 3 μL, and the gradient elution program was: 0% B (0-3 min), 0% B to 100% B (3-15 min), 100% B (15-18 min), 100% B to 0% B (18-19 min), 0% B (19-22 min).

[0045] Mass spectrometry conditions: Electrospray ionization source; spray voltage set to 4500 V in negative ion mode and 5500 V in positive ion mode; ion source auxiliary gas 1 and ion source auxiliary gas 2 both 55 Psi; curtain gas 35 psi; ion source temperature 450 ℃; Multiple Reaction Monitoring (MRM) scanning mode; run time 22 min.

[0046] Sample processing: 80 μL of serum from each patient was added to a 1.5 mL EP tube containing 300 μL of 80% cold methanol (containing ketoibuprofen as an internal standard). Another 10-20 μL was used to prepare a QC sample. The extract was equilibrated at 4 °C for 10 min, then centrifuged (18000 g, 4 °C, 10 min). 300 μL of the supernatant was collected in a 1.5 mL EP tube, concentrated under vacuum, and stored at -80 °C for subsequent analysis. Before analysis, the sample was reconstituted in 100 μL of 50% acetonitrile.

[0047] Lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, L-pyroglutamic acid, and 5-hydroxyindoleacetic acid standards were each dissolved to a 1 mg / mL stock solution. 20 μL of each stock solution was used to prepare mixed standard solutions, which were then diluted to different concentrations. For each concentration, 20 μL of the mixed standard solution was added to 60 μL of blank solvent and then transferred to a 1.5 mL EP tube containing 300 μL of 80% cold methanol (containing an internal standard). The extract was equilibrated at 4 °C for 10 min and then centrifuged (18000 g, 4 °C, 10 min). 300 μL of the supernatant was collected in a 1.5 mL EP tube, concentrated under vacuum, and stored at -80 °C. Before analysis, the sample was reconstituted in 100 μL of 50% acetonitrile.

[0048] Standard curve establishment: MRM detection mode was used to test mixed standard solutions of different concentrations. Data were integrated, and the deviation of the detected concentration was controlled within ±15% to exclude non-compliant mixed standard concentrations. Based on this, a standard curve was established, ensuring a correlation coefficient of ≥0.99. Sample testing: After sample reconstitution, GOLD grade 1-2 and GOLD grade 3-4 patient samples were tested alternately to avoid interference between duplicate samples.

[0049] The target metabolites in the samples were quantified using SCIEX OS 2.2 software (AB SCIEX).

[0050] 3. Data Processing Methods

[0051] Model Training and Evaluation: The model was trained and its performance evaluated using the commonly used random forest model on the training set. Specifically, a 5-fold hierarchical cross-validation method was employed, approximately dividing the 474 subjects in the training set into five equal-sized subsets with 95, 95, 95, 95, and 94 subjects respectively, while ensuring that the proportion of each category in each subset was consistent with the original dataset. The sum of samples from four subsets was used for model training, and the samples from the remaining subset were used for model evaluation. This process was repeated five times, selecting a different subset for evaluation each time, ensuring that each subset was used for model evaluation at least once.

[0052] The importance score of metabolic biomarkers is calculated using a random forest ensemble learning algorithm commonly used in this field. Specifically, a decision tree-based random forest model is constructed, and based on the Gini impurity reduction principle, the contribution of each metabolite feature to diagnosing and differentiating stable COPD patients with pulmonary function grades 1-2 from 3-4 is quantified. Specifically, the algorithm calculates the impurity reduction of each feature at all decision tree nodes and averages the results across all trees to obtain a standardized feature importance score.

[0053] Calculation of diagnostic thresholds for stable COPD patients with pulmonary function grades 1-2 and 3-4: An adaptive threshold optimization algorithm based on decision trees, commonly used in this field, is used to determine clinical diagnostic thresholds for metabolite characteristics in the training set. By constructing a depth-limited decision tree model, the impact of different split points on sample classification purity is systematically evaluated, and a comprehensive scoring function is used to select the optimal threshold. The algorithm first constructs a decision tree with a maximum depth of n layers, extracting the split thresholds of all internal nodes as candidate thresholds; then, it calculates the class purity after each candidate threshold divides the sample into two intervals, and selects the threshold that maximizes the average purity of the two intervals as the final diagnostic threshold.

[0054] Calculation of the probability of metabolic marker diagnosis: Based on the characteristic importance score of each metabolic marker, the predicted probability of each metabolic marker in different combinations of metabolic markers is calculated in the training set according to the following formula, that is, the probability of being diagnosed as stable COPD grade 1-2 and grade 3-4 when the serum metabolic marker content of the subject meets the corresponding diagnostic threshold: Predicted probability (%) = Characteristic importance score of a single marker in the combination / Sum of characteristic importance scores of all markers in the combination × 100%.

[0055] Validation of the diagnostic accuracy of metabolic markers: In the validation set, the content of each metabolic marker in different combinations of metabolic markers was compared with the diagnostic threshold of that metabolic marker to obtain the probability of each metabolic marker in distinguishing between stable COPD pulmonary function grades 1-2 and 3-4. The probability of each metabolic marker was summed to obtain the sum of the probability of each metabolic marker. The conclusion that the sum of the probability of each metabolic marker is >50% was taken as the result of the pulmonary function classification. The accuracy of the combined diagnosis of each metabolic marker in the combination of metabolic markers in distinguishing between stable COPD pulmonary function grades 1-2 and 3-4 was obtained by dividing the number of correctly diagnosed samples by the total number of samples.

[0056] IV. Experimental Results

[0057] 1. Differences in serum levels of target metabolites between patients with stable COPD pulmonary function grades 1-2 and 3-4.

[0058] In the training set, compared with patients with stable COPD pulmonary function grades 1-2, patients with grade 3-4 showed significantly upregulated levels of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and 5-hydroxyindoleacetic acid, and significantly downregulated levels of L-pyroglutamate in serum. Figure 1 As shown.

[0059] 2. Model Training and Evaluation Results

[0060] The model is trained on the training set based on the commonly used random forest model in this field, and the model performance is evaluated. The average of accuracy, precision, recall, FI score and AUC is taken as the final evaluation index of model performance.

[0061] For the pulmonary function grade 1-2 group and the pulmonary function grade 3-4 group, the results of the diagnostic test can be divided into the following categories:

[0062] Positive (TP); indicates the number of samples correctly predicted as positive by the model (consistent with lung function diagnostic results);

[0063] Negative (TN); indicates the number of samples correctly predicted as negative by the model (consistent with lung function diagnostic results);

[0064] False positive (FP): Indicates the number of samples that the model incorrectly predicted as positive (inconsistent with the results of lung function diagnosis);

[0065] False negative (FN): Indicates the number of samples that the model incorrectly predicted as negative (inconsistent with lung function diagnostic results).

[0066] This can be represented by the following table:

[0067] Diagnostic test Model predicts positive class Model predicts negative class total Positive Class A (Number of TPs) B (Number of FPs) A+B negative class C (number of FN) D (Number of TNs) C+D total A+C B+D A+B+C+D

[0068] Accuracy = (A+D) / (A+B+C+D);

[0069] Accuracy = A / (A+C);

[0070] Recall rate = A / (A+B);

[0071] F1 score = 2 × (precision × recall) / (precision + recall).

[0072] AUC: The ROC curve is plotted with each test result as a possible diagnostic cutoff value. The area under the curve (AUC) indicates the accuracy of the diagnostic test. AUC = [(number of times the positive score > negative score) + 0.5 × (number of times the positive score = negative score)] / total number of positive and negative sample pairs. AUC is widely recognized as an inherent accuracy indicator for evaluating the validity of diagnostic tests. An AUC of 0.5 indicates no diagnostic significance; an AUC between 0.5 and 0.7 indicates low diagnostic accuracy; an AUC between 0.7 and 0.9 indicates moderate diagnostic accuracy; and an AUC > 0.9 indicates high diagnostic accuracy.

[0073] Table 3 shows the diagnostic efficacy of different combinations of metabolic markers in distinguishing between stable COPD patients with pulmonary function grades 1-2 and 3-4.

[0074] Table 3. Diagnostic efficacy of different combinations of metabolic markers in differentiating stable COPD patients with pulmonary function grades 1-2 and 3-4.

[0075] Logo combination Accuracy Sensitivity Recall rate F1 score AUC Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + L-pyroglutamic acid + 5-hydroxyindoleacetic acid 0.8592 0.8569 0.8592 0.8578 0.9344 Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + 5-hydroxyindoleacetic acid 0.8662 0.865 0.8662 0.8656 0.923 Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + L-pyroglutamic acid 0.838 0.8366 0.838 0.8373 0.9178 Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine 0.8451 0.8425 0.8451 0.8436 0.9059 Dimethylglycine + L-pyroglutamic acid + 5-hydroxyindoleacetic acid 0.831 0.8281 0.831 0.8294 0.8899 Lysophosphatidylcholine (17:0 / 0:0) + L-pyroglutamic acid + 5-hydroxyindoleacetic acid 0.7887 0.776 0.7887 0.7794 0.881 Lysophosphatidylcholine (17:0 / 0:0) + 5-hydroxyindoleacetic acid 0.7958 0.7849 0.7958 0.788 0.8526 Dimethylglycine + L-pyroglutamic acid 0.7606 0.7748 0.7606 0.7663 0.8343 Lysophosphatidylcholine (17:0 / 0:0) + L-pyroglutamic acid 0.8099 0.8082 0.8099 0.809 0.8328 Dimethylglycine + 5-hydroxyindoleacetic acid 0.838 0.8322 0.838 0.8339 0.8286 L-pyroglutamic acid + 5-hydroxyindoleacetic acid 0.7817 0.7726 0.7817 0.7761 0.7622

[0076] As can be seen from the results in Table 3, the combined diagnostic efficacy of two, three or four metabolic markers based on lysophosphatidylcholine (17:0 / 0:0) and dimethylglycine was the highest, with all reaching AUC>0.9, and other indicators were also relatively better.

[0077] Therefore, subsequent experiments were conducted using combinations of two, three, or four metabolic markers based on lysophosphatidylcholine (17:0 / 0:0) and dimethylglycine.

[0078] 3. Calculation results of importance scores for metabolic biomarkers

[0079] As mentioned earlier, the combined diagnostic efficacy of two, three, or four metabolic markers based on lysophosphatidylcholine (17:0 / 0:0) and dimethylglycine was the highest, with all achieving AUC > 0.9, and other indicators were also relatively superior. In the training set, we calculated the characteristic importance scores of the four metabolic markers involved in the two, three, or four metabolic marker combinations based on lysophosphatidylcholine (17:0 / 0:0) and dimethylglycine for diagnosing and differentiating stable COPD pulmonary function grades 1-2 from 3-4, as shown in Table 4.

[0080] Table 4. Characteristic Importance Scores of Metabolic Biomarkers

[0081] Metabolic markers Feature Importance Score Lysophosphatidylcholine (17:0 / 0:0) 0.057 dimethylglycine 0.038 L-pyroglutamic acid 0.036 5-Hydroxyindoleacetic acid 0.036

[0082] 4. Calculation results of diagnostic thresholds and probabilities for classifying patients with stable COPD pulmonary function grades 1-2 and 3-4.

[0083] (1) Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + L-pyroglutamic acid + 5-hydroxyindoleacetic acid

[0084] Table 5 shows the diagnostic thresholds and probabilities for using four biomarkers—lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, L-pyroglutamic acid, and 5-hydroxyindoleacetic acid—to differentiate between stable COPD patients with pulmonary function grades 1-2 and 3-4.

[0085] For ease of understanding, an example is given below: In the combination of lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + L-pyroglutamic acid + 5-hydroxyindoleacetic acid, the probability of determining lysophosphatidylcholine (17:0 / 0:0) is calculated as follows: 0.057 ÷ (0.057 + 0.038 + 0.036 + 0.036) × 100% = 34.13%.

[0086] Table 5 Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + L-pyroglutamic acid + 5-hydroxyindoleacetic acid

[0087] Metabolic markers Lung function grade 1-2 Lung function grade 3-4 Probability of judgment Lysophosphatidylcholine (17:0 / 0:0) ≤ 4194.73 ng / mL > 4194.73 ng / mL 34.13% dimethylglycine ≤ 96.906 ng / mL > 96.906 ng / mL 22.75% L-pyroglutamic acid > 2282 ng / mL ≤ 2282 ng / mL 21.56% 5-Hydroxyindoleacetic acid ≤ 112.219 ng / mL > 112.219 ng / mL 21.56%

[0088] (2) Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + 5-hydroxyindoleacetic acid

[0089] Table 6 shows the diagnostic thresholds and probabilities for using three biomarkers—lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and 5-hydroxyindoleacetic acid—to differentiate between stable COPD patients with pulmonary function grades 1-2 and 3-4.

[0090] For ease of understanding, an example is given below: In the combination of lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + 5-hydroxyindoleacetic acid, the probability of determining lysophosphatidylcholine (17:0 / 0:0) is calculated as follows: 0.057 ÷ (0.057 + 0.038 + 0.036) × 100% = 43.51%.

[0091] Table 6 Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + 5-hydroxyindoleacetic acid

[0092] Metabolic markers Lung function grade 1-2 Lung function grade 3-4 Probability of judgment Lysophosphatidylcholine (17:0 / 0:0) ≤ 4194.73 ng / mL > 4194.73 ng / mL 43.51% dimethylglycine ≤ 96.906 ng / mL > 96.906 ng / mL 29.00% 5-Hydroxyindoleacetic acid ≤ 112.219 ng / mL > 112.219 ng / mL 27.49%

[0093] (3) Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + L-pyroglutamic acid

[0094] Table 7 shows the diagnostic thresholds and probabilities for using three biomarkers—lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and L-pyroglutamate—to differentiate between stable COPD patients with pulmonary function grades 1-2 and 3-4.

[0095] For ease of understanding, an example is given below: In the combination of lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + L-pyroglutamic acid, the probability of determining lysophosphatidylcholine (17:0 / 0:0) is calculated as follows: 0.057 ÷ (0.057 + 0.038 + 0.036) × 100% = 43.51%.

[0096] Table 7 Lysophosphatidylcholine (17:0 / 0:0) + Dimethylglycine + L-pyroglutamic acid

[0097] Metabolic markers Lung function grade 1-2 Lung function grade 3-4 Probability of judgment Lysophosphatidylcholine (17:0 / 0:0) ≤ 4194.73 ng / mL > 4194.73 ng / mL 43.51% dimethylglycine ≤ 96.906 ng / mL > 96.906 ng / mL 29.00% L-pyroglutamic acid > 2282 ng / mL ≤ 2282 ng / mL 27.49%

[0098] (4) Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine

[0099] The diagnostic thresholds and probabilities for using two biomarkers, lysophosphatidylcholine (17:0 / 0:0) and dimethylglycine, to differentiate between stable COPD patients with pulmonary function grades 1-2 and 3-4 are shown in Table 8.

[0100] For ease of understanding, here is an example: In the combination of lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine, the probability of determining lysophosphatidylcholine (17:0 / 0:0) is calculated as follows: 0.057 ÷ (0.057 + 0.038) × 100% = 60%.

[0101] Table 8 Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine

[0102] Metabolic markers Lung function grade 1-2 Lung function grade 3-4 Probability of judgment Lysophosphatidylcholine (17:0 / 0:0) ≤ 4194.73 ng / mL > 4194.73 ng / mL 60% dimethylglycine ≤ 96.906 ng / mL > 96.906 ng / mL 40%

[0103] 5. Validation results of the diagnostic accuracy of metabolic markers

[0104] (1) Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + L-pyroglutamic acid + 5-hydroxyindoleacetic acid

[0105] In the validation set, the levels of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, L-pyroglutamic acid, and 5-hydroxyindoleacetic acid in each sample were compared with the diagnostic thresholds in Table 5 to obtain the diagnostic probabilities of these four metabolic markers in differentiating stable COPD pulmonary function grades 1-2 and 3-4. The probabilities of these four metabolic markers were summed to obtain the sum of the probabilities. A conclusion with a sum of probabilities > 50% was taken as the result of pulmonary function classification. The accuracy of the combined diagnosis of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, L-pyroglutamic acid, and 5-hydroxyindoleacetic acid in differentiating stable COPD pulmonary function grades 1-2 and 3-4 was obtained by dividing the number of correctly diagnosed samples by the total number of samples.

[0106] For ease of understanding, consider the following example: If in a sample of the validation set, lysophosphatidylcholine (17:0 / 0:0) > 4194.73 ng / mL (pulmonary function grade 3-4), dimethylglycine ≤ 96.906 ng / mL (pulmonary function grade 1-2), L-pyroglutamic acid ≤ 2282 ng / mL (pulmonary function grade 3-4), and 5-hydroxyindoleacetic acid ≤ 112.219 ng / mL (pulmonary function grade 1-2), then from the perspective of pulmonary function grade 3-4, the sum of the probabilities of classifying this sample as pulmonary function grade 3-4 is 34.13% + 21.56% = 55.69%; from the perspective of pulmonary function grade 1-2, the sum of the probabilities of classifying this sample as pulmonary function grade 1-2 is 22.75% + 21.56% = 44.31%. If the sum of the probabilities of judgment is greater than 50%, the lung function classification result is taken as lung function grade 3-4. If the sample does indeed belong to lung function grade 3-4, the classification is correct; if the sample does not belong to lung function grade 3-4, the classification is incorrect.

[0107] Ultimately, in the validation set, the combined diagnostic accuracy of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, L-pyroglutamic acid, and 5-hydroxyindoleacetic acid in differentiating stable COPD pulmonary function grades 1-2 from 3-4 was 82.61%.

[0108] (2) Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + 5-hydroxyindoleacetic acid

[0109] In the validation set, the levels of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and 5-hydroxyindoleacetic acid in each sample were compared with the diagnostic thresholds in Table 6 to obtain the diagnostic probabilities of these three metabolic markers in differentiating stable COPD pulmonary function grades 1-2 and 3-4. The probabilities of these three metabolic markers were summed to obtain the sum of the probabilities. A conclusion with a sum of probabilities > 50% was taken as the result of pulmonary function classification. The accuracy of the combined diagnosis of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and 5-hydroxyindoleacetic acid in differentiating stable COPD pulmonary function grades 1-2 and 3-4 was obtained by dividing the number of correctly diagnosed samples by the total number of samples.

[0110] For ease of understanding, consider the following example: If in a sample of the validation set, lysophosphatidylcholine (17:0 / 0:0) > 4194.73 ng / mL (pulmonary function grade 3-4), dimethylglycine ≤ 96.906 ng / mL (pulmonary function grade 1-2), and 5-hydroxyindoleacetic acid ≤ 112.219 ng / mL (pulmonary function grade 1-2), then from the perspective of pulmonary function grade 3-4, the sum of the probabilities of classifying this sample as pulmonary function grade 3-4 is 43.51%; from the perspective of pulmonary function grade 1-2, the sum of the probabilities of classifying this sample as pulmonary function grade 1-2 is 29.00% + 27.49% = 56.49%. Taking the conclusion that the sum of the probabilities is > 50% as the pulmonary function grading result, then this sample is classified as pulmonary function grade 1-2. If the sample does indeed belong to lung function grade 1-2, the judgment is correct; if the sample does not belong to lung function grade 1-2, the judgment is incorrect.

[0111] Ultimately, in the validation set, the combined diagnostic accuracy of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and 5-hydroxyindoleacetic acid in differentiating stable COPD pulmonary function grades 1-2 from 3-4 was 84.06%.

[0112] (3) Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine + L-pyroglutamic acid

[0113] In the validation set, the levels of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and L-pyroglutamate in each sample were compared with the diagnostic thresholds in Table 7 to obtain the diagnostic probabilities of these three metabolic markers in differentiating stable COPD pulmonary function grades 1-2 and 3-4. The probabilities of these three metabolic markers were summed to obtain the sum of the probabilities. A conclusion with a sum of probabilities > 50% was taken as the result of pulmonary function classification. The accuracy of the combined diagnosis of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and L-pyroglutamate in differentiating stable COPD pulmonary function grades 1-2 and 3-4 was obtained by dividing the number of correctly classified samples by the total number of samples.

[0114] For ease of understanding, consider the following example: If a sample in the validation set contains lysophosphatidylcholine (17:0 / 0:0) > 4194.73 ng / mL (pulmonary function grade 3-4), dimethylglycine ≤ 96.906 ng / mL (pulmonary function grade 1-2), and L-pyroglutamate ≤ 2282 ng / mL (pulmonary function grade 3-4), then from the perspective of pulmonary function grade 3-4, the sum of the probabilities of classifying this sample as pulmonary function grade 3-4 is 43.51% + 27.49% = 71%; from the perspective of pulmonary function grade 1-2, the sum of the probabilities of classifying this sample as pulmonary function grade 1-2 is 29%. Taking the conclusion that the sum of the probabilities is > 50% as the pulmonary function grading result, then this sample is classified as pulmonary function grade 3-4. If the sample does indeed belong to pulmonary function grade 3-4, the classification is correct; if the sample does not belong to pulmonary function grade 3-4, the classification is incorrect.

[0115] Ultimately, in the validation set, the combined diagnostic accuracy of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and L-pyroglutamate in differentiating stable COPD pulmonary function grades 1-2 from 3-4 was 82.61%.

[0116] (4) Lysophosphatidylcholine (17:0 / 0:0) + dimethylglycine

[0117] In the validation set, the levels of lysophosphatidylcholine (17:0 / 0:0) and dimethylglycine in each sample were compared with the diagnostic thresholds in Table 8 to obtain the diagnostic probabilities of these two metabolic markers in differentiating between stable COPD pulmonary function grades 1-2 and 3-4. The probabilities of these two metabolic markers were summed to obtain the sum of the probabilities. The conclusion that the sum of the probabilities is >50% was taken as the result of the pulmonary function classification. The accuracy of the combined diagnosis of lysophosphatidylcholine (17:0 / 0:0) and dimethylglycine in differentiating between stable COPD pulmonary function grades 1-2 and 3-4 was obtained by dividing the number of correctly diagnosed samples by the total number of samples.

[0118] For ease of understanding, consider the following example: If a sample in the validation set has lysophosphatidylcholine (17:0 / 0:0) > 4194.73 ng / mL (pulmonary function grade 3-4) and dimethylglycine ≤ 96.906 ng / mL (pulmonary function grade 1-2), then from the perspective of pulmonary function grade 3-4, the sum of the probabilities of classifying this sample as pulmonary function grade 3-4 is 60%; from the perspective of pulmonary function grade 1-2, the sum of the probabilities of classifying this sample as pulmonary function grade 1-2 is 40%. Taking the conclusion that the sum of the probabilities is > 50% as the pulmonary function grading result, then this sample is classified as pulmonary function grade 3-4. If the sample does indeed belong to pulmonary function grade 3-4, the classification is correct; if the sample does not belong to pulmonary function grade 3-4, the classification is incorrect.

[0119] Ultimately, in the validation set, the combined diagnostic accuracy of lysophosphatidylcholine (17:0 / 0:0) and dimethylglycine in differentiating stable COPD pulmonary function grades 1-2 from 3-4 was 81.59%.

[0120] Example 2: Diagnostic kit for differentiating stable COPD pulmonary function grades 1-2 from 3-4

[0121] 1. A diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease, the kit containing standards of lysophosphatidylcholine (17:0 / 0:0) and dimethylglycine.

[0122] 2. A diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease, the kit containing standards of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and L-pyroglutamic acid.

[0123] 3. A diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease, the kit containing standards of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, and 5-hydroxyindoleacetic acid.

[0124] 4. A diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease, the kit containing standards of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, L-pyroglutamic acid and 5-hydroxyindoleacetic acid.

[0125] The purpose of the above embodiments is to specifically illustrate the substantive content of the present invention, but those skilled in the art should know that the scope of protection of the present invention should not be limited to the specific embodiments.

Claims

1. Application of lysophosphatidylcholine (17:0 / 0:0) in combination with dimethylglycine in the preparation of a diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease.

2. The application according to claim 1, characterized in that: The kit is a serum detection kit.

3. Application of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine and L-pyroglutamic acid in the preparation of a diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease.

4. The application according to claim 3, characterized in that: The kit is a serum detection kit.

5. Application of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine and 5-hydroxyindoleacetic acid in the preparation of a diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease.

6. The application according to claim 5, characterized in that: The kit is a serum detection kit.

7. Application of lysophosphatidylcholine (17:0 / 0:0), dimethylglycine, L-pyroglutamic acid and 5-hydroxyindoleacetic acid in the preparation of a diagnostic kit for differentiating between stable pulmonary function grades 1-2 and 3-4 in chronic obstructive pulmonary disease.

8. The application according to claim 7, characterized in that: The kit is a serum detection kit.

Images