CN105021804A - Application of lung cancer metabolism markers to lung cancer diagnosis and treatment - Google Patents

Abstract

The invention relates to the technical field of medical diagnostics, in particular to the field of tumor diagnosis. The present invention screens out a variety of metabolic markers that can be used as lung cancer curative effect monitoring and lung cancer diagnosis by analyzing the concentration of small molecule metabolites in normal human blood, pre-lung cancer blood, and lung cancer post-operative blood. The method of the invention has the advantages of high sensitivity and high throughput, and is suitable for screening and auxiliary diagnosis of lung cancer.

Description

Use of lung cancer metabolic markers in the diagnosis and treatment of lung cancer

technical field

The invention relates to the technical field of medical diagnostics, in particular to the field of tumor diagnosis. the

Background technique

Cancer remains one of the deadliest threats to human health. Solid tumors are responsible for most of these deaths. Although significant advances have been made in the medical treatment of certain cancers, overall 5-year survival rates for all cancers have only improved by about 10% in the last 20 years. Cancer, or malignancy, grows and metastasizes rapidly in an uncontrolled manner, making timely detection and treatment extremely difficult. In the past 30 years, with the influence of external factors such as air pollution and smoking, the incidence of lung cancer in my country has also increased day by day. In Shanghai, Beijing and Tianjin, it is indeed the first place among various malignant tumors. The premise of treatment is based on a clear diagnosis. Early diagnosis of lung cancer is a difficult and important task. the

At present, the diagnostic methods for lung cancer can be divided into two categories, one is noninvasive diagnosis, mainly including imaging examination and sputum cytology examination, and the other is invasive diagnosis, mainly including bronchoscopy and surgical diagnosis. Among them, imaging examinations mainly include X-ray, CT, MRI and so on. Surgical diagnostic methods include bronchial biopsy (TBLB), percutaneous puncture (NB), mediastinoscopy, video-assisted thoracoscopy, and lymph node biopsy. Although recent studies have found that the detection of superoxide dismutase and monoamine oxidase combined with enzymes has certain significance for the diagnosis of lung cancer, and the serum SOD activity of lung cancer patients is significantly lower than that of benign lung lesions and normal healthy people. Combined with other indicators for diagnosis. Lung cancer is a malignant tumor that seriously endangers human health and life. Screening out new tumor metabolic markers with early diagnosis and prognosis value of lung cancer can be used for early diagnosis and treatment of lung cancer patients, so as to improve the survival rate of lung cancer patients. need to be resolved during treatment. the

Contents of the invention

The purpose of the present application is to provide a blood detection method for early diagnosis of lung cancer and evaluation of postoperative prognosis. the

Research proposal for this application:

(1) Establishment of lung cancer database, sample collection, processing and detection

①Establish and organize the blood sample bank of lung cancer patients and healthy control population;

②Establish a lung cancer postoperative follow-up system and follow-up database, collect and organize patients' pathological files, postoperative treatment plan and effect, postoperative survival and other data;

③Sample collection and processing: Collect 5ml of abdominal blood samples from healthy and normal subjects and lung cancer patients before and 7 days after surgery, put them in sterile coagulation-promoting BD vacuum blood collection tubes, centrifuge at 2500r/min, 4°C for 5min, and take the upper serum , stored in a -80°C refrigerator until use. the

④LC-Q-TOF/MS detection

⑤GC/MS detection

(2) To study the relationship between serum metabolic markers in patients with lung cancer and the diagnosis and curative effect monitoring of lung cancer

①Analyze the data with multivariate statistical methods such as Principal Component Analysis (PCA), Least Squares Discriminant Analysis (PLS-DA) and Orthogonal Least Squares Analysis (OPLS);

② Mining and identification of differential metabolites;

③Relationship between serum metabolic markers and lung cancer diagnosis and curative effect monitoring. the

Description of drawings

Figure 1: The total ion current of normal control group (a), lung cancer preoperative group (b) and lung cancer postoperative group (c) by high performance liquid chromatography-quadrupole-time-of-flight mass spectrometry (LC-Q-TOF/MS) picture

Figure 2: Gas chromatography-mass spectrometry (GC/MS) total ion chromatograms of normal control group (a), lung cancer preoperative group (b) and lung cancer postoperative group (c)

Figure 3: Principal components of normal control group (A), lung cancer preoperative group (B) and lung cancer postoperative group (C) based on high performance liquid chromatography-quadrupole-time-of-flight mass spectrometry (LC-Q-TOF/MS) Analysis (PCA) score plot

Figure 4: The minimum squares of normal control group (A), lung cancer preoperative group (B) and lung cancer postoperative group (C) based on high performance liquid chromatography-quadrupole-time-of-flight mass spectrometry (LC-Q-TOF/MS) Multiplicative Discriminant Analysis (PLS-DA) Score Plot

Figure 5: Orthogonality of normal control group (A), lung cancer preoperative group (B) and lung cancer postoperative group (C) based on high performance liquid chromatography-quadrupole-time-of-flight mass spectrometry (LC-Q-TOF/MS) Analysis of Least Squares (OPLS) Score Plot

Figure 6: Principal component analysis (PCA) score chart of normal control group (A), lung cancer preoperative group (B) and lung cancer postoperative group (C) based on gas chromatography-mass spectrometry (GC/MS)

Figure 7: Least squares discriminant analysis (PLS-DA) score chart of normal control group (A), lung cancer preoperative group (B) and lung cancer postoperative group (C) based on gas chromatography-mass spectrometry (GC/MS)

Figure 8: Orthogonal least squares analysis (OPLS) score chart of normal control group (A), lung cancer preoperative group (B) and lung cancer postoperative group (C) based on gas chromatography-mass spectrometry (GC/MS)

Figure 9: Differences between normal control group (A), lung cancer preoperative group (B) and lung cancer postoperative group (C) based on high performance liquid chromatography-quadrupole-time-of-flight mass spectrometry (LC-Q-TOF/MS) Box plot of maximum metabolic markers

Figure 10: Box plot of metabolic markers with the greatest difference among normal control group (A), lung cancer preoperative group (B) and lung cancer postoperative group (C) based on gas chromatography-mass spectrometry (GC/MS)

Detailed ways

Experimental example:

1. General information: From January 2012 to January 2013, 30 patients with lung cancer diagnosed by pathology or cytology in Huzhou Central Hospital, aged 43-76 years, with an average age of 63.9 years; 21 cases were male and 9 cases were female ; There were 15 cases of adenocarcinoma, 12 cases of squamous cell carcinoma, and 3 cases of large cell lung cancer. The control group included 30 cases, all of whom had normal physical examination in Huzhou City, aged 39-78 years, with an average of 60.4 years; 19 cases were male and 11 cases were female. This study was approved by the Ethics Committee of Huzhou Central Hospital, and all subjects gave informed consent to participate in this study. the

2. Specimen collection: Collect 5ml of fasting blood samples from healthy and normal subjects and lung cancer patients before surgery, put them in sterile coagulation-promoting BD vacuum blood collection tubes, centrifuge at 2500r/min, 4°C for 5min, take the upper serum, and store it in a -80°C refrigerator ready for use. the

3. Pretreatment of LC-Q-TOF/MS serum samples: thaw the cryopreserved serum samples at room temperature and shake well, take 100 μL of samples and add 300 μL of methanol, vortex for 30 seconds, and stand at 4°C for 20 minutes; all samples are refrigerated and centrifuged (12000r/min, 4°C, 15min), take 200μL of the supernatant, transfer it into a sample injection vial, and perform LC-Q-TOF/MS detection and analysis. the

4. GC/MS serum sample pretreatment: thaw the cryopreserved serum sample at room temperature, take 100 μL of the sample and add 300 μL of cold HPLC grade ethanol, vortex for 30 seconds, stand at 4°C for 20 minutes, and refrigerate and centrifuge. Add 80 μL of ice-cold HPLC grade methanol and vortex for 30 seconds. After taking it out, refrigerate and centrifuge (12000rpm, 4°C, 15min), take 150μL of supernatant, transfer it to a glass derivatization vial, add 10μL of internal standard solution (dichlorophenylalanine, the concentration is 0.02mg/mL), and place in 4 Blow dry under mild nitrogen at ℃; add 30 μL of 20 mg/mL methoxyamine hydrochloride pyridine solution to the glass derivatization vial, shake vigorously for 30 s, and react with oximation at 37 °C for 90 min; after taking it out, add 30 μL of BSTFA (containing 1% TMCS) Derivatization reagents, react at 70°C for 60min. The samples were taken out and left at room temperature for 30 min for GC/MS metabolomics analysis. the

5. Liquid phase/mass spectrometry (LC-MS) analysis conditions: high-performance liquid phase-quadrupole-time-of-flight mass spectrometry (Agilent, 1290Infinity LC, 6530UHD and Accurate-Mass Q-TOF/MS) technology is used to analyze lung cancer patients and healthy people Serum metabolic profile of the examinee. the

(1) Chromatographic conditions: Agilent C ₁₈ chromatographic column (100mm×2.1mm, 1.8μm), flow rate 0.4ml/min, column temperature 40°C, mobile phase A: 0.1% formic acid-water, B: 0.1% formic acid-acetonitrile , the injection volume was 4 μl, and the temperature of the autosampler was 4°C. The gradient elution program is shown in Table 1.

Table 1 Chromatographic gradient elution program

(2) Mass spectrometry conditions: ESI ion source, using positive ion scanning mode, using nitrogen as nebulization and cone gas. Capillary voltage 4kV, cone voltage 35kV, ion source temperature 100°C; desolvation temperature 350°C, reverse cone airflow 50L/h, desolvation 600L/h, extraction cone 4V. The ion scanning time is 0.03s, the scanning time interval is 0.02s, and the data acquisition range: 50-1000m/z. the

6. Gas chromatography/mass spectrometry (GC/MS) analysis conditions: Gas chromatography-mass spectrometry (Agilent7890A/5975CGC/MS) technology was used to analyze the serum metabolic profiles of lung cancer patients and healthy healthy subjects. The capillary chromatographic column is HP-5ms (30m×0.25mm×0.25μm) of Agilent J&W Scientific Company. The instrument parameters are set as follows: inlet temperature 280°C, EI ion source temperature 230°C, quadrupole temperature 150°C, high-purity helium (purity greater than 99.999%) as carrier gas, splitless injection, injection volume 1.0 μL. The heating program is as follows: the initial temperature is 80°C, maintained for 2 minutes, then raised to 320°C at a rate of 10°C/min, and maintained for 6 minutes. The mass spectrometry detection is carried out in a full scan mode, and the mass spectrometry detection range is 50-550 (m/z). the

7. LC-Q-TOF/MS data processing: Agilent MassHunter workstation (Qualitative Analysis VB03.01, Agilent, USA) was used to record the total ion current chromatogram (TIC) of each serum sample for visual inspection. Under the R software platform, self-written program codes were used to extract raw data signals (peak identification and integration), then retention time correction, peak alignment and deconvolution analysis (mass spectrum fragment assignment), and finally post-editing in Excel software, Organize the results as a two-dimensional data matrix, including variables (retention time Rt, mass-to-charge ratio m/z), observed quantities (sample), and integrated areas. The qualitative method of differential metabolites is: search the metabolite secondary mass spectrometry database (METLIN), and compare the mass-to-charge ratio m/z or accurate molecular mass mass of the mass spectra. the

8. GC/MS data processing: Under Agilent MassHunter series workstation software, carry out analysis such as retention time correction, peak alignment and mass spectrum fragment attribution, and use self-written program codes for data preprocessing under the R software platform, including baseline filtering, peak Identify and integrate, and then finally edit in the EXCEL2007 software, including impurity peak elimination and quantitative ion selection caused by column loss and sample preparation, etc., organize the final result into a two-dimensional data matrix, including variables (rt_mz, namely retention time_mass-to-charge ratio), observed quantity (sample) and integrated area. The qualitative method of differential metabolites is: search the self-built standard substance database and NIST commercial database (comparison of mass spectrum and chromatographic retention time RT). the

9. Statistical analysis: Import the edited data matrix into Simca-P11.0 software for principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA) and orthogonal partial least squares analysis (OPLS) . ^The quality of the multivariate statistical model was evaluated by the parameters R2 and ^Q2 . When R ² (X)>0.4, the PCA model is reliable; when Q ² >0.4, the PLS-DA model is reliable. According to the VIP value of PLS-DA to mine differential metabolites, the VIP value indirectly reflects the correlation between metabolites and the studied diseases, which is a widely used variable selection method. Metabolites with a VIP value > 1 were selected as differential metabolites in this study. Two-sample t-test was used to compare the levels of metabolites in the normal control group, lung cancer preoperative group and lung cancer postoperative group. The logarithmic value (base 2) of the ratio of the mean value of the lung cancer preoperative/postoperative group to the normal control group indicates the relative level of the substance in the lung cancer preoperative/postoperative patient, and the positive sign indicates that the substance is in the lung cancer preoperative The negative sign indicates that the substance is decreased in the pre-/post-operative patients of lung cancer.

Effect example

(1) Serum metabolomics analysis: The composition of serum metabolites is complex. The total ion chromatograms of the normal control group, lung cancer preoperative group and lung cancer postoperative group obtained by LC/MS and GC/MS are shown in Figure 1 and Figure 2, a is the normal control group, b is the preoperative group of lung cancer, and c is the postoperative group of lung cancer. It can be seen from the figure that there is a certain difference in the metabolites and their ion intensities in the serum of the normal control population and lung cancer patients before/after surgery. the

(2) LC-Q-TOF/MS multivariate data analysis

① Principal Component Analysis (PCA): As an unsupervised learning method, Principal Component Analysis can truly reflect the clustering of samples. PCA analysis was performed on the data obtained in the positive mode of the three groups of samples, and a total of 7 principal components (PCs) were obtained, R ² X =0.582, Q ² =0.425. Generally speaking, the R ² X value greater than 0.4 indicates that the model is reliable, so the model established in this project can be applied to visually observe the differences in metabolic profiles among the three groups. The PCA score plot (Scores plot) is shown in Figure 3a. It can be seen from the PCA score plot in the mode that most of the samples are in the 95% confidence interval, and the three groups of samples can be clearly distinguished, and the samples in the group are relatively concentrated, and the overall model ideal. PCA analysis was performed on the data obtained in the positive mode of the normal control group (A) and the lung cancer preoperative group (B), and a total of 4 principal components (PC), R ² X = 0.556, Q ² = 0.419, were established in this project The model can be applied to visually observe the differences in metabolic profiles between the two groups, as shown in the PCA score plot (Scores plot) as shown in Figure 3b. PCA analysis was performed on the data obtained in the positive mode of the normal control group (A) and the postoperative lung cancer group (C) group, and a total of 7 principal components (PC) were obtained, R ² X = 0.552, Q ² = 0.319, this project The established model can be applied to visually observe the differences in metabolic profiles between the two groups, as shown in the PCA score plot (Scores plot) as shown in Figure 3c. PCA analysis was performed on the data obtained in the positive mode of the lung cancer preoperative group (B) and the lung cancer postoperative group (C) group, and a total of 7 principal components (PC) were obtained, R ² X = 0.62, Q ² = 0.4, this The model established by the project can be applied to visually observe the differences in metabolic profiles between the two groups. The PCA score plot (Scores plot) is shown in Figure 3d.

② Least squares discriminant analysis (PLS-DA): The supervised method PLS-DA is further used to model and analyze the samples of the normal control group (A) and the lung cancer preoperative group (B). The parameter R ² Y of the model represents the The explanation rate, Q ² represents the prediction rate of the model, generally speaking, this parameter is greater than 0.4, which means that the model is reliable. Results Four principal components were obtained in positive mode, R ² X = 0.527, R ² Y = 0.991, Q ² = 0.938, and the PLS-DA score diagram is shown in Figure 4a. The samples of the normal control group (A) and the postoperative lung cancer group (C) were modeled and analyzed using the supervised method PLS-DA, and the results obtained 4 principal components in the positive mode, R ² X = 0.412, R ² Y = 0.992 , Q ² =0.935, R ² Y and Q ² values greater than 0.4 indicate that the model is reliable, and the PLS-DA score chart is shown in Figure 4b. The samples of preoperative lung cancer group (B) and postoperative lung cancer group (C) were modeled and analyzed using the supervised method PLS-DA, and the results obtained two principal components in the positive mode, R ² X = 0.432, R ² Y = 0.906, Q ² =0.875, R ² Y and Q ² values greater than 0.4 indicate that the model is reliable, and the PLS-DA score diagram is shown in Figure 4c.

③ Orthogonal least squares analysis (OPLS): further adopt the supervised method OPLS to model and analyze the samples of normal control group (A) and lung cancer preoperative group (B), and the parameter R ² Y of the model represents the interpretation rate of the model , Q ² represents the prediction rate of the model. Generally speaking, if this parameter is greater than 0.4, it means that the model is reliable. Results One principal component and one orthogonal component were obtained in positive mode, R ² X = 0.468, R ² Y = 0.936, Q ² = 0.898, and the score diagram is shown in Figure 5a. Through OPLS analysis, from the effect of the score map, the two groups of samples can be well separated in the positive ion mode, and the samples of the normal control group are on the left side of the principal component 1 (PC1), while the samples of the lung cancer preoperative group are on the left side of the principal component 1 (PC1). to the right of Principal Component 1 (PC1). The samples of the normal control group (A) and the postoperative lung cancer group (C) were modeled and analyzed using the supervised method OPLS. As a result, 1 principal component and 1 orthogonal component were obtained in the positive mode, R ² X = 0.326, R ² Y=0.937, Q ² =0.899, the score diagram is shown in Figure 5b. Through OPLS analysis, from the effect of the score map, the two groups of samples can be well separated in the positive ion mode, and the samples of the normal control group (A) are on the left side of the principal component 1 (PC1), while the samples of the lung cancer surgery group are on the left side of the principal component 1 (PC1). The sample of the latter group (C) is on the right side of principal component 1 (PC1). The preoperative group (B) and the postoperative group (C) of lung cancer were modeled and analyzed using the supervised method OPLS, and the results obtained in the positive mode were 1 1 principal component and 1 orthogonal component, R ² X = 0.432, R ² Y = 0.906, Q ² = 0.865, the score diagram is shown in Figure 5c. Through OPLS analysis, from the effect of the score map, the two groups of samples can be well separated in positive ion mode, and the samples of the preoperative lung cancer group are on the left side of principal component 1 (PC1), while the samples of the postoperative lung cancer group It is on the right side of the principal component 1 (PC1). In the next section, we will use the values given by the Variable Importance list of this OPLS model, combined with t-test, to find variables that contribute significantly to the significant difference between the two groups (metabolites), and to characterize these variables.

(3) GC/MS multivariate data analysis

①Principal component analysis (PCA): First, evaluate the overall data to check whether it meets the standard of metabolomics analysis. In the Simca-P software, the data are processed with the default UV formatting (Unit Variance Scaling) and mean-centered (Mean-Centered) to obtain more reliable and intuitive results. The software carried out model fitting analysis on the data obtained in the positive mode of the three groups of samples. A total of 10 principal components were obtained for the serum, R ² X = 0.801, Q ² = 0.652 (the abscissa is the score of the first principal component, and t[1] Indicates; the vertical axis is the score of the second principal component, represented by t[2]). The PCA score plot (Scores plot) is shown in 6a; most of the samples are within the 95% confidence interval (Hotelling T ² ellipse), and in order to ensure the originality of the data, this item does not need to be eliminated without affecting the overall analysis of samples. Generally speaking, the R ² X value greater than 0.4 indicates that the model is reliable, so the current PCA model can be reliably used to explain the metabolic differences between samples. PCA analysis was performed on the data obtained in the positive mode of the normal control group (A) and the lung cancer preoperative group (B) group, and a total of 8 principal components were obtained, R ² X =0.787, Q ² =0.64. The PCA score plot (Scores plot) is shown in 6b, and the sample is basically within the 95% confidence interval (Hotelling T ² ellipse). Generally speaking, the R ² X value greater than 0.4 indicates that the model is reliable, so the current PCA model can be reliably used to explain the metabolic differences between two groups of samples. PCA analysis was performed on the data obtained in the positive mode of the normal control group (A) and the postoperative lung cancer group (C) group, and a total of 12 principal components were obtained, R ² X =0.843, Q ² =0.626. The PCA score plot (Scores plot) is shown in 6c, the samples are basically within the 95% confidence interval (Hotelling T ² ellipse), and the current PCA model can be reliably used to explain the metabolic differences between the two groups of samples. PCA analysis was performed on the data obtained in the positive mode of the preoperative lung cancer group (B) and the lung cancer postoperative group (C) group, and a total of 9 principal components were obtained, R ² X = 0.805, Q ² = 0.627. The PCA score plot (Scores plot) is shown in 6d, the samples are basically within the 95% confidence interval (Hotelling T ² ellipse), and the current PCA model can be reliably used to explain the metabolic differences between the two groups of samples.

② Least squares discriminant analysis (PLS-DA): The supervised method PLS-DA was further used to model and analyze the samples of the normal control group (A) and the lung cancer preoperative group (B), and a total of 3 principal components were obtained, R ² X = 0.533, R ² Y = 0.854, Q ² = 0.747, (the abscissa is the score of the first principal component, represented by t[1]; the ordinate is the score of the second principal component, represented by t[2]), R ² Y and Q ² values greater than 0.4 indicate that the model is reliable, and the score chart is shown in 7a. The supervised method PLS-DA was used to model and analyze the samples of the normal control group (A) and the postoperative lung cancer group (C), and a total of 5 principal components were obtained, R ² X = 0.605, R ² Y = 0.956, Q ² =0.707, R ² Y and Q ² values greater than 0.4 indicate that the model is reliable, and the score diagram is shown in 7b. The samples of preoperative lung cancer group (B) and lung cancer postoperative group (C) were modeled and analyzed using the supervised method PLS-DA, and a total of three principal components were obtained, R ² X = 0.518, R ² Y = 0.883, Q ² = 0.758, R ² Y and Q ² values greater than 0.4 indicate that the model is reliable, and the score diagram is shown in 7c.

③ Orthogonal least squares analysis (OPLS): The supervised method OPLS was further used to model and analyze the samples of the normal control group (A) and the lung cancer preoperative group (B), and a total of 1 principal component and 1 orthogonal Components, R ² X = 0.46, R ² Y = 0.797, Q ² = 0.745, (the abscissa is the score of the first principal component, represented by t[1]; the ordinate is the score of the second principal component, represented by t[2] Indicated), the R ² Y and Q ² values greater than 0.4 indicate that the model is reliable, the score chart is shown in 8a: the samples of the normal control group (A) and the postoperative lung cancer group (C) were modeled and analyzed using the supervised method OPLS , a total of 1 principal component and 1 orthogonal component are obtained, R ² X = 0.457, R ² Y = 0.68, Q ² = 0.57, R ² Y and Q ² values greater than 0.4 indicate that the model is reliable, and the score diagram is shown in 8b shown. The preoperative lung cancer group (B) and the lung cancer postoperative group (C) were modeled and analyzed using the supervised method OPLS, and a total of 1 principal component and 2 orthogonal components were obtained, R ² X = 0.518, R ² Y = 0.883 , Q ² =0.767, R ² Y and Q ² values greater than 0.4 indicate that the model is reliable, and the score diagram is shown in 8c. In the next section, we will use the values given by the Variable Importance list of this OPLS model, combined with t-test, to find variables (metabolites) that contribute significantly to the significant difference between the two groups, and to characterize these variables.

(4) Mining and identification of differential metabolites between the two groups based on LC-Q-TOF/MS detection: This project uses the VIP (Variable Importance in the Projection) value of the PLS-DA model (threshold > 1), and Combined with the p-value of the t-test (threshold value 0.05) to look for differentially expressed metabolites. The qualitative method of differential metabolites is: search the online database ( http://metlin.scripps.edu/ ) (comparing the mass-to-charge ratio m/z or accurate molecular mass mass of the mass spectrum). There are 32 differential metabolites in the normal control group (A) and the lung cancer preoperative group (B). As shown in Table 2, there are 30 differential metabolites in the normal control group (A) and the lung cancer postoperative group (C). As shown in Table 3, there are 27 differential metabolites in the lung cancer preoperative group (B) and lung cancer postoperative group (C), as shown in Table 4. The differential metabolites of the three groups were comprehensively analyzed, and five metabolic markers with the largest differences were obtained. The differences in peak intensities are shown in Figure 9. It can be seen from the figure that the concentration of sphingosine in the serum of the lung cancer preoperative group was significantly lower than that of the normal control group and the lung cancer postoperative group. The concentration of choline chloride, anandamide and γ-linoleic acid were significantly higher than those in the normal control group and the postoperative lung cancer group; Compared with the control group, it was significantly lower, and the differences were statistically significant. It is speculated that sphingosine, foscholine chloride, anandamide and γ-linoleic acid can be used as metabolic markers for monitoring the efficacy of lung cancer, and prasterone sulfate can be used as metabolic markers for lung cancer diagnosis.

Table 2 Potential metabolic markers of normal control group and lung cancer preoperative group based on LC-Q-TOF/MS detection (p<0.05)

^a The logarithmic value of the ratio of the mean value of the lung cancer preoperative group to the normal control group (base 2).

Table 3 Potential metabolic markers of lung cancer preoperative group and lung cancer postoperative group based on LC-Q-TOF/MS detection (p<0.05)

^b The logarithmic value of the ratio of the postoperative lung cancer group to the mean value of the lung cancer preoperative group (base 2).

Table 4 Potential metabolic markers of normal control group and lung cancer postoperative group based on LC-Q-TOF/MS detection (p<0.05)

^c The logarithmic value (base 2) of the ratio between the normal control group and the lung cancer postoperative group. The positive sign indicates that the lung cancer postoperative group increased compared with the normal control group, and the negative sign indicates a decrease

(5) Mining and identification of differential metabolites between the two groups based on GC/MS detection: This project uses the VIP (Variable Importance in the Projection) value of the first principal component of the PLS-DA model (threshold > 1), and Combined with the p-value of the t-test (threshold value 0.05) to look for differentially expressed metabolites. The qualitative method is: search self-built reference material database and NIST commercial database (comparison of mass spectrum and chromatographic retention time RT). There are 23 differential metabolites in the normal control group (A) and the lung cancer preoperative group (B), as shown in Table 5. There are 20 differential metabolites in the normal control group (A) and the lung cancer postoperative group (C). As shown in Table 6, there are 21 differential metabolites between the lung cancer preoperative group (B) and the lung cancer postoperative group (C), as shown in Table 7. The differential metabolites of the three groups were comprehensively analyzed, and five metabolic markers with the largest differences were obtained. The differences in peak intensities are shown in Figure 10. It can be seen from the figure that the concentration of 9,12-octadecadienoic acid and oleic acid in the serum of the lung cancer preoperative group was significantly higher than that of the normal control group and the lung cancer postoperative group, and the serine concentration was significantly higher than that of the normal control group and the lung cancer postoperative group. Compared with the normal control group, the concentration of α-hydroxyisobutyric acid in the serum of the lung cancer preoperative group and the lung cancer postoperative group was significantly higher than that of the normal control group, and the concentration of 2,3,4-trihydroxybutyric acid was significantly higher than that of the normal control group. decreased significantly, and the differences were statistically significant. It is speculated that 9,12-octadecadienoic acid, oleic acid and serine can be used as metabolic markers for monitoring the efficacy of lung cancer, and α-hydroxyisobutyric acid and 2,3,4-trihydroxybutyric acid can be used as metabolic markers for lung cancer diagnosis things. the

Table 5 Potential metabolic markers of normal control group and lung cancer preoperative group based on GC/MS detection (p<0.05)

^d The logarithmic value (base 2) of the mean ratio between the lung cancer preoperative group and the normal control group, the positive sign indicates that the preoperative lung cancer group is increased compared with the normal control group, and the negative sign indicates a decrease

Table 6 Potential metabolic markers of lung cancer preoperative group and lung cancer postoperative group based on GC/MS detection (p<0.05)

^eThe logarithmic value of the mean ratio between the lung cancer postoperative group and the lung cancer preoperative group (base 2).

Table 7 Potential metabolic markers of normal control group and postoperative lung cancer group based on GC/MS detection (p<0.05)

^f The logarithmic value (base 2) of the mean ratio between the normal control group and the postoperative lung cancer group. A positive sign indicates an increase in the postoperative lung cancer group compared to the normal control group, and a negative sign indicates a decrease.

Claims (10)

1. identify the metabolic indicator compositions of lung cancer, it comprises more than one metabolic markers in following compound group, 8-hydroxy guanine, phenylglyoxylic acid, L-nipecotic acid, VBT, 3-sulfuric acid caffeic acid, fumaric acid dimethyl ester, fumaric acid dimethyl ester, L-adrenaline, Reichstein's compound G, 3-hydroxyl-3-methylglutaric acid, acetylcarnitine, Valine, uric acid, oxalosuccinic acid, cortisol, Prasterone sulfate, uridine, dihydrosphingosine, sphingol, Phosphorylcholine, phosphatid ylcholine (15:0), phosphoglycerol-N-anandamide, PLC, phosphatid ylcholine (16:0), oleamide, phosphoglycerol-N-Palmitylethanolamide, phosphatid ylcholine (17:0), 1,25-hydroxycholecalciferol glycocholic acid, peanut carnitine, phosphatid ylcholine (18:0), glycocholic acid, gamma-linoleic acid, alpha-tocopherol, creatinine, acetyl, 2-hydroxyl glutaric acid, dodecene diacid, adrenalone, forulic acid, uridine, galactitol phosphoric acid, lauroyl, cholerythrin, 1-linolenyl choline glycerophosphatide, anandamide, gamma-Linolenic acid, twenty carnitine, tetradecanoic acid.

2. the metabolic indicator compositions of qualification lung cancer according to claim 1, is characterized in that, comprises the metabolin of more than two in described composition group or three kind or five kinds.

3. the metabolic indicator compositions of qualification lung cancer according to claim 1, is characterized in that, the serum-concentration of compound in blood of patients with lung cancer in described compound group obviously changes.

4. the metabolic indicator compositions of qualification lung cancer according to claim 1, is characterized in that, the compound in described compound group is detected by LC-Q-TOF/MS.

5. the metabolic indicator compositions of qualification lung cancer according to claim 1, is characterized in that, preferably comprise in composition the one in sphingol, Phosphorylcholine, anandamide and g-linoleic acid, two kinds, three kinds or four kinds.

6. the metabolic indicator compositions of the qualification lung cancer according to any one of claim 1-5, is characterized in that, preferably comprise Prasterone sulfate in composition.

7. the purposes of metabolic indicator compositions in the kit preparing pulmonary cancer diagnosis or lung cancer examination of curative effect or reagent of the qualification lung cancer according to any one of claim 1-6.

8. identify the metabolic indicator compositions of lung cancer, it comprises more than one metabolic markers in following compound group, lactic acid, alanine, alpha-hydroxybutyric dehydrogenase, beta-hydroxy-butanoic acid, Valine, urea, serine, L-Leu, phosphate, ILE, glycocoll, L-threonine, amidomalonic acid, pyroglutamic acid, 2, 3, 4-trihydroxy-butyric acid, L-Phe, tetradecylic acid, glucose, palmitic acid, inositol, 9, 12-octadecadienoic acid, oleic acid, cholesterol, oxalic acid, Alpha-hydroxy valeric acid, glutamine, L-Orn, tyrosine, stearic acid, α-D-glucopyranoside, 2-deoxidation-galactopyranose.

9. the metabolic indicator compositions of qualification lung cancer according to claim 8, is characterized in that, comprises the metabolin of more than two in described composition group or three kind or five kinds.

10. the metabolic indicator compositions of qualification lung cancer according to claim 1, is characterized in that, the serum-concentration of compound in blood of patients with lung cancer in described compound group obviously changes.