Plasma autoantibody markers for discriminating lung adenocarcinoma from benign lung disease and uses thereof

Abstract

The application provides a plasma autoantibody marker for identifying lung adenocarcinoma and benign pulmonary diseases and an application thereof, the marker is an autoantibody for any one or more of the following antigens: anti-DSP, anti-AGR2, anti-MDK, anti-ABCC3, anti-MET, anti-ASS1, anti-PYCR1, anti-LGALS4 and anti-CRABP2, and the application further discloses a differential diagnosis model of the plasma autoantibody marker for constructing LUAD and BPD in combination, the model has good differential diagnosis efficiency for lung adenocarcinoma and benign pulmonary disease patients, and has important significance for management and identification of the benign pulmonary disease patients.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of biotechnology, in particular to a plasma autoantibody marker for identifying lung adenocarcinoma and benign lung disease and application thereof. BACKGROUND

[0002] The background section provided herein is merely for information and is not necessarily prior art.

[0003] According to the results of global cancer data in 2024, lung cancer is one of the most common malignant tumors, accounting for 12.4% and 18.7% of all tumor diagnoses and deaths, respectively. Lung cancer is divided into small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC), of which about 85% of lung cancer belongs to NSCLC. Lung adenocarcinoma (LUAD) is the most common type of NSCLC, accounting for about 40% of lung cancer. Although great progress has been made in lung cancer-related research, the incidence rate is still on the rise. The survival prognosis of lung cancer patients has improved significantly in recent years, but the 5-year survival rate is not high. Most patients are in the middle and advanced stages or have distant metastasis at the time of diagnosis, missing the best treatment opportunity. If it can be detected early and treated early, the survival time of patients will be greatly extended. The current X-ray and CT early detection methods have low sensitivity, poor specificity, high cost, non-invasive, and other shortcomings, and it is difficult to meet the early detection needs of large-scale population.

[0004] Tumor markers play a very key role in the early screening, diagnosis, efficacy evaluation and prognosis evaluation of cancer. At present, blood is the main source of tumor markers for lung cancer screening. Due to its easy availability, non-invasiveness and repeatable detection, it has attracted widespread attention. Traditional blood tumor markers have been used in the auxiliary diagnosis of cancer in the clinic. Common tumor markers include carcinoembryonic antigen (CEA), neuron-specific enolase (NSE), and cytokeratin 19 fragment (CYFRA21-1). Although the current tumor markers have been confirmed in the clinic, they cannot be well used for specific diagnosis of lung adenocarcinoma.

[0005] Many studies have shown that cancer patients' plasma contains autoantigens, called tumor associated antigens (TAA), and that autoantibodies against tumor associated antigens (TAAb) can react with them. TAAb has been detected in various tumors. Some autoantibody changes occur early, and can be detected months to years before clinical diagnosis of tumors. In addition, TAAb as an immunodiagnostic marker may have greater advantages because they are amplified by the immune response signal than TAA itself is easier to detect. Therefore, TAAb has great potential in the early diagnosis of cancer, and it is necessary to screen and identify new TAAb for early screening and diagnosis of lung adenocarcinoma. TAAb appears earlier in plasma, has the advantages of early detection, stable existence and dynamic monitoring, and is a good marker for early screening and diagnosis of lung adenocarcinoma. Currently, the research on autoantibodies of lung cancer includes: anti-Annexin1, anti-LAMR1, anti-HSP40, anti-NY-ESO-1 and anti-P53, etc. However, the sensitivity or specificity of the markers related to them in the clinic is low, so it is necessary to find new TAAb with better diagnostic value. SUMMARY

[0006] The technical problem solved by the present application is to provide a plasma autoantibody marker for identifying lung adenocarcinoma and benign lung diseases and its application.

[0007] To solve the above technical problems, the present application discloses the following technical solutions:

[0008] In a first aspect, the present application discloses a lung adenocarcinoma marker, which comprises autoantibodies against any one or more of the following antigens: anti-DSP, anti-AGR2, anti-MDK, anti-ABCC3, anti-MET, anti-ASS1, anti-PYCR1, anti-LGALS4 and anti-CRABP2.

[0009] In some embodiments, the marker is a plasma autoantibody marker.

[0010] In some embodiments, the autoantibody is a plasma autoantibody in peripheral blood, such as an IgG autoantibody in peripheral blood plasma.

[0011] In some embodiments, the marker is a plasma autoantibody marker for identifying lung adenocarcinoma and healthy people, or for identifying lung adenocarcinoma and benign lung diseases.

[0012] Further, the present application also provides a screening method and a verification method for the plasma autoantibody marker.

[0013] The screening method specifically comprises the following steps:

[0014] First, single-cell transcriptome data of 11 lung adenocarcinoma tissues and 11 para-cancer tissues were downloaded from the GEO database, and the data were reduced in dimension, clustered and grouped; then, proteomic data of 103 lung adenocarcinoma tissues and 103 para-cancer tissues in Cell magazine were used for verification at the protein level, and finally, candidate TAAs with potential diagnostic value were obtained.

[0015] The verification method is a two-stage verification based on ELISA technology, specifically as follows:

[0016] Among them, the first verification stage is a small sample preliminary verification (72 lung adenocarcinoma patients and 72 healthy controls). The difference in expression level of TAAbs between the two groups is analyzed by non-parametric test, and the receiver operating characteristic (ROC) curve of each candidate TAAb is drawn to evaluate its diagnostic value.

[0017] TAAbs with P<0.05 and area under the ROC curve (AUC)>0.5 are subjected to large sample verification (249 lung adenocarcinoma patients and 249 benign lung diseases) in the second verification stage. The difference in level among the three groups is analyzed by non-parametric test, and the value of each TAAb is evaluated according to the epidemiological diagnostic test method. The evaluation indexes include sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio and coincidence rate, etc.

[0018] In the second aspect, the application discloses an application of the marker in the preparation of a product for diagnosing lung adenocarcinoma.

[0019] The diagnosis of lung adenocarcinoma is to distinguish lung adenocarcinoma from healthy people, or to distinguish lung adenocarcinoma from benign lung diseases.

[0020] The product includes a kit, a reagent and a chip.

[0021] In the third aspect, the application discloses a kit for diagnosing lung adenocarcinoma, which comprises the marker of the first aspect.

[0022] The diagnosis of lung adenocarcinoma is to distinguish lung adenocarcinoma from healthy people, or to distinguish lung adenocarcinoma from benign lung diseases.

[0023] The kit is an ELISA detection kit.

[0024] In a fourth aspect, the present application discloses application of the marker in the first aspect in preparation of a diagnostic model for identifying lung adenocarcinoma and benign lung disease.

[0025] When P value is greater than or equal to 0.5, the patient is diagnosed as a lung adenocarcinoma patient; and when P value is less than 0.5, the patient is diagnosed as a benign lung disease patient.

[0026] P = 1 / (1+EXP(-(-4.914+1.329×anti-ASS1+0.793×anti-DSP+1.105×anti-MET+0.990×anti-PYCR1+1.375×anti-MDK)));

[0027] Wherein, anti-ASS1, anti-DSP, anti-MET, anti-PYCR1 and anti-MDK are expression amounts of corresponding autoantibodies in plasma.

[0028] In a fifth aspect, the present application discloses a diagnostic model for identifying lung adenocarcinoma and benign lung disease, and the diagnostic model is as follows:

[0029] When P value is greater than or equal to 0.5, the patient is diagnosed as a lung adenocarcinoma patient; and when P value is less than 0.5, the patient is diagnosed as a benign lung disease patient.

[0030] P = 1 / (1+EXP(-(-4.914+1.329×anti-ASS1+0.793×anti-DSP+1.105×anti-MET+0.990×anti-PYCR1+1.375×anti-MDK)));

[0031] Wherein, anti-ASS1, anti-DSP, anti-MET, anti-PYCR1 and anti-MDK are expression amounts of corresponding autoantibodies in plasma.

[0032] Further, the present application further provides a construction method of the diagnostic model for identifying lung adenocarcinoma and benign lung disease, comprising the following steps:

[0033] In the second validation phase, lung adenocarcinoma and benign lung disease samples were split into training and validation sets in approximately a 7:3 ratio using SPSS Statistics 26.0 software. A logistic differential diagnostic model for lung adenocarcinoma was constructed using the training set samples. The value of each model was evaluated using ROC curves and further validated on the validation set. The De Long test was used to compare the statistical significance of the AUC difference between the training and validation sets: P = 1 / (1 + EXP(-(-4.914 + 1.329 × anti-ASS1 + 0.793 × anti-DSP + 1.105 × anti-MET + 0.990 × anti-PYCR1 + 1.375 × anti-MDK))); where anti-ASS1, anti-DSP, anti-MET, anti-PYCR1, and anti-MDK represent the expression levels of the corresponding autoantibodies in plasma.

[0034] Beneficial effects:

[0035] This invention uses single-cell transcriptomics data, proteomics data, and enzyme-linked immunosorbent assays to screen and identify novel plasma autoantibody markers (anti-AGR2, anti-MDK, anti-ABCC3, anti-MET, anti-ASS1, anti-PYCR1, anti-LGALS4, anti-DSP, and anti-CRABP2) for the differential diagnosis of lung adenocarcinoma (LUAD) and benign lung disease (BPD). These nine variables are then combined to construct a differential diagnostic model for LUAD and BPD. This model demonstrates good diagnostic efficacy for patients with lung adenocarcinoma and benign lung disease, providing assistance for better clinical management of patients with pulmonary nodules. Attached Figure Description

[0036] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the present invention in the above and / or other aspects will become clearer.

[0037] Figure 1 These are feature dimensionality reduction, cell clustering, and annotation diagrams. (a) t-SNE dimensionality reduction diagram after initial cell clustering, (b) t-SNE dimensionality reduction diagram after cell type annotation.

[0038] Figure 2 This is a scatter plot and ROC curve of the 10 TAAbs in Phase 1 of validation. Note: LUAD: lung adenocarcinoma, NC: healthy controls, *: P<0.05, **: P<0.01, ***: P<0.001, SBI: specific binding index, AUC: area under the ROC curve, 95% CI: 95% confidence interval.

[0039] Figure 3Figure 9 is a scatter plot of the nine TA Abs in the second validation phase. Note: LUAD: lung adenocarcinoma, BPD: benign pulmonary disease, NC: healthy control, SBI: specific binding index, AUC: area under the ROC curve, 95% CI: 95% confidence interval, *: P < 0.05, **: P < 0.01, ***: P < 0.001.

[0040] Figure 4 Figure 10 is a ROC curve of the nine TA Abs in the second validation phase (lung adenocarcinoma and benign pulmonary disease). Note: AUC: area under the ROC curve, 95% CI: 95% confidence interval.

[0041] Figure 5 Figure 11 is a ROC curve of the logistic regression differential diagnosis model in the training set and the validation set. DETAILED DESCRIPTION

[0042] The present application can be better understood according to the following examples. However, it is easily understood by those skilled in the art that the content described in the examples is only for illustrating the present application, and should not and will not limit the present application described in detail in the claims.

[0043] The experimental methods described in the following examples are all conventional methods unless otherwise specified; the reagents and materials described are all commercially available unless otherwise specified.

[0044] The present application provides a plasma autoantibody marker combined to construct a differential diagnosis model of LUAD and BPD, which has good differential diagnosis efficiency for lung adenocarcinoma and benign pulmonary disease patients, and has important significance for the management and identification of benign pulmonary disease patients.

[0045] Example 1 Screening and verification of plasma autoantibody markers

[0046] 1. Collection of clinical data of research subjects

[0047] The clinical information of the patients was consulted and collected in the hospital patient management system by comparing the subject ID, name, gender and age, including smoking history, pathological type, TNM stage (only malignant) and lymph node metastasis and distant metastasis (only malignant). A total of 891 research subjects were included in this study, including 72 lung adenocarcinoma samples and 72 healthy control samples in the first validation phase; 249 lung adenocarcinoma samples, 249 benign pulmonary disease samples and 249 healthy control samples in the second validation phase.

[0048] The basic information, pathological stage and clinical traditional tumor marker levels of all subjects were summarized and arranged through the internal system of the hospital under the condition of patient consent and approval of the institutional review board and hospital ethics committee.

[0049] 2. Collection and processing of plasma samples

[0050] Peripheral venous blood (5 mL) was collected from all subjects in a fasting state using vacuum blood collection tubes containing EDTA-K2 anticoagulant. Within 2 hours after blood collection, the samples were sent to the laboratory and centrifuged at room temperature at 3000 rpm / min for 5 minutes to separate the plasma and blood cells. The upper layer of plasma was pipetted and aliquoted into 1.5 mL Eppendorf tubes according to the 500 mL / tube specification, labeled with disease type, number and date, and stored in an ultra-low temperature freezer at -80°C. The sample storage location and corresponding patient basic information were recorded on the specimen registration book. When used, the required samples were placed in a 4°C refrigerator for slow thawing, and were aliquoted into 96-well plates according to the experimental requirements for use and layout, avoiding repeated freezing and thawing of the original tube.

[0051] 3. Screening of lung adenocarcinoma candidate TAAs based on single-cell transcriptome data and proteome data

[0052] 3.1 Data preprocessing

[0053] The single-cell transcriptome sequencing data came from the GSE131907 dataset in the GEO database, including 11 lung adenocarcinoma tissues and 11 adjacent tissues. The Seurat package was used to standardize all the data and perform quality control, which identified and filtered low-quality cells and genes in the original data, including cells that had died or were damaged, and genes with low expression. The quality control standards were as follows: (1) filter out cells with less than 3 cells expressing a single gene and less than 300 genes expressing; (2) calculate the percentage of mitochondrial genes and ribosome genes in each cell through the PercentageFeatureSet function, and only keep cells with mitochondrial gene percentage less than 20% and ribosome gene percentage greater than 3%; (3) filter out 13 mitochondrial genes (MT-CO1, MT-CO2, MT-CO3, MT-ATP6, MT-ATP8, MT-ND1, MT-ND2, MTND3, MT-ND4, MT-ND4L, MT-ND5, MT-ND6 and MT-CYB) and the housekeeping gene MALAT1. After quality control, there were 87880 cells remaining with 27113 expressed genes.

[0054] 3.2 Dimensionality reduction of data

[0055] In this study, PCA dimension reduction and t-SNE method were used for data dimension reduction. In this study, the process of dimension reduction was carried out according to the following steps: (1) Calculate the cycle score of S phase and G2 / M phase of each cell. The cell cycle was verified by the CellCycleScoring function; (2) Identify the top 2000 genes with high variation rate by the FindVariableFeatures function; (3) Centralize the candidate 2000 genes by the ScaleData function, and eliminate the differences between cells caused by mitochondrial genes and cell cycle scores; (4) Linear dimension reduction by RunPCA function, using the top 2000 genes with high variation rate obtained in the above steps for calculation; Select the number of PC number, visualize and nonlinearly reduce the data by RunTSNE function; (5) Cell clustering by FindNeighbors function, continue to adjust the resolution parameter (resolution = 0.8) by FindClusters function to identify cell clusters; At the same time, also use DimPlot and FeatureScatter functions to draw the corresponding dimension reduction map, etc.

[0056] 3.3 Annotation of cell clustering types

[0057] Manual annotation of cell types was performed using classic Maker genes (CellMarker database). For example: Epithelial cell adhesion molecule (EPCAM) is a marker gene of normal epithelial cells and tumor cells; Platelet derived growth factor receptor alpha (PDGFRA) is a marker gene of fibroblasts; Von willebrand factor (VWF) is a marker gene of endothelial cells; Protein tyrosine phosphatase receptor type C (PTPRC) is a marker gene of different types of immune cells, etc., as shown in the results. Figure 1

[0058] (4) Analysis of differences in cell distribution to find differential genes of potential tumor cell sources

[0059] Based on the analysis of cell subtypes of epithelial origin, the cell subtypes with significantly high proportion or specificity in tumors were identified, and then the COSG package was used to find the marker genes (100 each) of each cell subtype. The marker genes of potential malignant epithelial cell subtypes were extracted for subsequent analysis.

[0060] ​(5) Enrichment analysis of differential genes and determination of candidate TAAs

[0061] Based on the potential 200 marker genes of malignant epithelial cells in single-cell transcriptome data, then verified at protein level from proteomic data of 103 lung adenocarcinoma samples and 103 paracancerous tissue samples in Cell magazine, performed differential analysis and calculated the average fold change (avg_log2FC) of protein relative expression between groups, extracted 65 potential protein level high expression of malignant epithelial cell subtype marker genes. The ClusterProfiler function package was used for GO and KEGG analysis of the above differential genes, and the ggplot2 function package was used for data visualization. GO analysis can comprehensively describe the characteristics of genes and gene products in organisms, including biological processes, cellular components and molecular functions. KEGG analysis can systematically study gene functions, link genomic information and functional information, and help to analyze the whole network of genes and their expression information. Finally, 10 candidate TAAs were determined by combining literature and protein accessibility.

[0062] 4. Verification of plasma differential protein indicators by ELISA method

[0063] 4.1 Experimental materials and reagents:

[0064] (1) 10 tumor-related antigen proteins: purchase LGALS4, ASS1, CRABP2, DSP, MET, ABCC3, AGR2, MDK, IGFBP3 and PYCR1 corresponding recombinant proteins;

[0065] (2) 96-well enzyme-labeled plate (8 rows x 12 columns);

[0066] (3) Coating solution: aqueous solution containing 0.15% sodium carbonate (Na2CO3) and 0.29% sodium bicarbonate (NaHCO3);

[0067] (4) Blocking solution: 0.2% Tween 20 PBST buffer containing 2% bovine plasma albumin (BSA);

[0068] (5) Plasma sample diluent: PBST buffer containing 1% BSA;

[0069] (6) Enzyme-labeled secondary antibody: horseradish peroxidase (HRP) labeled mouse anti-human immunoglobulin antibody (hereinafter referred to as HRP labeled mouse anti-human IgG antibody);

[0070] (7) Antibody diluent: PBST buffer containing 1% BSA;

[0071] (8) washing solution: PBST buffer solution containing 0.2% Tween 20;

[0072] (9) developing solution: the developing solution is composed of developing solution A and developing solution B, wherein the developing solution A is a 20% tetramethyl benzidine dihydrochloride aqueous solution, and the developing solution B is a 3.7% Na2HPO4·12H2O, 0.92% citric acid, 0.75% hydrogen peroxide urea aqueous solution; when used, the developing solution A and the developing solution B are mixed uniformly at a volume ratio of 1:1, and are prepared on site;

[0073] (10) termination solution: 10% sulfuric acid.

[0074] (11) quality control plasma: the quality control plasma is prepared by mixing 100 samples of healthy controls in equal amounts.

[0075] 4.2 Experimental method:

[0076] (1) Preparation of 10 tumor-related antigen coated enzyme-labeled plates.

[0077] Taking the preparation of the tumor-related antigen AGR2 coated enzyme-labeled plate as an example, the specific operation steps are as follows:

[0078] 1) Preparation of tumor-related antigen AGR2 protein solution: the AGR2 protein is prepared into a 0.25 μg / mL AGR2 protein solution using the coating solution.

[0079] 2) Coating of enzyme-labeled plate: the AGR2 protein solution prepared in step 1) is added to each reaction well of the 96-well enzyme-labeled plate, and the sample amount is 50 μL / well, which is coated at 4°C overnight, then the remaining coating solution is shaken off and dried.

[0080] 3) Blocking: add blocking solution to the reaction wells of the coated 96-well enzyme-labeled plate, the sample amount is 100 μL / well, block in a 37°C water bath for 2 h, then remove the blocking solution, wash with washing solution (sample amount is 300 μL / well) for 3 times and dry, to obtain the tumor-related antigen AGR2 coated enzyme-labeled plate.

[0081] The operation steps for preparing the other 9 tumor-related antigen coated enzyme-labeled plates are basically the same as those for preparing the tumor-related antigen AGR2 coated enzyme-labeled plate, and the difference lies in that the tumor-related antigens used in step 1) are different, and the concentrations of the prepared tumor-related antigen solutions are different, wherein the concentrations of the tumor-related antigen MDK, ABCC3, ASS1, LGALS4, DSP and CRABP2 solutions are 0.25 μg / mL, the concentrations of the tumor-related antigen MET, PYCR1 solutions are 0.125 μg / mL, and the concentration of the tumor-related antigen IGFBP3 solution is 0.5 μg / mL.

[0082] (2) Detection of the expression levels of 10 kinds of autoantibodies against tumor-related antigens in the plasma samples.

[0083] The same plasma sample was used to detect the expression levels of autoantibodies against tumor-related antigens AGR2, MDK, ABCC3, ASS1, LGALS4, DSP, CRABP2, MET, PYCR1 and IGFBP3 in the plasma sample by ELISA method using the above-prepared 10 kinds of tumor-related antigen-coated enzyme plates.

[0084] Taking the detection of the expression level of autoantibodies against tumor-related antigen AGR2 as an example, the specific operation steps are as follows:

[0085] 1) Plasma sample incubation: The 891 plasma samples to be detected in step 1 were diluted with the plasma sample diluent at a volume ratio of 1:100. The diluted plasma samples were added to the reaction wells of columns 1-11 of the AGR2 protein-coated 96-well enzyme plate prepared in step (1) at a sample amount of 50 μL / well; 1:100 diluted quality control plasma was added to the first 6 reaction wells of column 12 of the AGR2 protein-coated 96-well enzyme plate at a sample amount of 50 μL / well, which was used as a quality control for standardization between different enzyme plates; 50 μL / well of antibody diluent without plasma was added to the 6-8 reaction wells of column 12 of the AGR2 protein-coated 96-well enzyme plate as a blank control; then the 96-well enzyme plate was incubated in a 37°C water bath for 1 h, and then the liquid in the reaction wells was discarded, washed with washing solution (300 μL / well) for 5 times and patted dry.

[0086] 2) Secondary antibody incubation: The HRP-labeled mouse anti-human IgG antibody was diluted with the antibody diluent at a ratio of 1:5000 v / v, and then the diluted HRP-labeled mouse anti-human IgG antibody was added to the corresponding reaction wells of the 96-well enzyme plate at a sample amount of 50 μL / well, which was incubated in a 37°C water bath for 1 h, and then the liquid in the reaction wells was discarded, washed with washing solution (300 μL / well) for 5 times and patted dry.

[0087] 3) Color development and reaction termination: Color developing solution A and color developing solution B were mixed at a ratio of 1:1, and then the mixed color developing solution was quickly added to the reaction wells of the 96-well enzyme plate at a sample amount of 50 μL / well, which was incubated at room temperature for 5-15 min in the dark, and then 25 μL of termination solution was added to each reaction well to terminate the color developing reaction; the enzyme marker was used to read the absorbance OD450 and OD620 at 450 nm and 620 nm wavelengths, respectively, wherein the absorbance OD620 at 620 nm wavelength was the background value, and the difference between OD450 and OD620 was taken as the final result of the detected absorbance value.

[0088] The specific operation steps for detecting the expression levels of other 9 tumor-associated antigen autoantibodies in the plasma sample are basically the same as those for detecting the anti-tumor-associated antigen AGR2 autoantibody described above, and the difference is that in step 1), the enzyme-labeled plates used for detection are tumor-associated antigen MDK, ABCC3, ASS1, LGALS4, DSP, CRABP2, MET, PYCR1 and IGFBP3 protein-coated enzyme-labeled plates; in step 2), for the reaction wells coated with tumor-associated antigens AGR2, MDK, ABCC3, ASS1, LGALS4, DSP, CRABP2, MET and PYCR1, the HRP-labeled mouse anti-human IgG antibody is diluted at a ratio of 1:5000 by volume; for the reaction wells coated with tumor-associated antigen IGFBP3, the HRP-labeled mouse anti-human IgG antibody is diluted at a ratio of 1:2500 by volume.

[0089] 5. Experimental results

[0090] The results in the verification phase 1 showed that the expression levels of 9 TA Abs in the plasma of lung adenocarcinoma were higher than those of healthy controls, including anti-AGR2, anti-MDK, anti-ABCC3, anti-DSP, anti-MET, anti-ASS1, anti-PYCR1, anti-LGALS4 and anti-CRABP2, with AUC (95% CI) ranging from 0.618 (0.536-0.709) to 0.768 (0.689-0.847), as shown in Figure 2 .

[0091] The results in the verification phase 2 showed that, except for anti-DSP, the expression levels of the remaining 8 TA Abs in the plasma of lung adenocarcinoma were higher than those of healthy controls, with AUC (95% CI) ranging from 0.571 (0.521-0.621) to 0.780 (0.740-0.820); the expression levels of 9 TA Abs in lung adenocarcinoma were higher than those in benign lung diseases, with AUC (95% CI) ranging from 0.622 (0.573-0.671) to 0.726 (0.682-0.770), as shown in Figure 3 and Figure 4 .

[0092] Therefore, the 9 anti-tumor-associated antigen autoantibodies can be used for the auxiliary diagnosis of lung adenocarcinoma.

[0093] Example 2: Evaluation of the ability of 9 anti-tumor-associated antigen autoantibodies to diagnose lung adenocarcinoma

[0094] 1. Experimental samples.

[0095] The expression levels of anti-tumor associated antigen MDK, ABCC3, ASS1, LGALS4, DSP, CRABP2, MET, PYCR1 and AGR2 autoantibodies in each plasma sample in the second verification stage were detected by ELISA, and the ROC curve was drawn using GraphPad Prism 8.0 to analyze the diagnostic value of the nine kinds of anti-tumor associated antigen autoantibodies for lung adenocarcinoma.

[0096] 2. The ability of single anti-tumor associated antigen autoantibody to diagnose and distinguish lung adenocarcinoma patients from normal people.

[0097] The subjects in the second verification stage (249 lung adenocarcinoma, 249 benign lung disease and 249 healthy controls) were randomly divided into a training set (183 lung adenocarcinoma, 183 benign lung disease and 183 healthy controls) and a verification set (66 lung adenocarcinoma, 66 benign lung disease and 66 healthy controls) in a ratio of about 7:3.

[0098] The ability of each anti-tumor associated antigen autoantibody to diagnose and distinguish lung adenocarcinoma patients from benign lung disease was evaluated by ROC curve. The ROC curve of the nine autoantibodies in the second verification stage for diagnosing and distinguishing lung adenocarcinoma patients from benign lung disease is shown in Figure 2. Figure 4

[0099] The AUC of the nine TAAs for differentiating lung adenocarcinoma from healthy controls ranged from 0.622 (0.573-0.671) to 0.726 (0.682-0.770). The SBI value corresponding to a specificity greater than 90% and a maximum Youden index was taken as the cutoff value, and the evaluation indexes of different TAAs were calculated, with a sensitivity range of 14.9%-31.7%, a specificity range of 90.0%-92.0%, a positive predictive value and a negative predictive value range of 64.9%-77.5% and 51.9%-57.1%, respectively, a positive likelihood ratio and a negative likelihood ratio range of 1.850-3.435 and 0.752-0.926, respectively, and a coincidence rate range of 53.4%-61.2%. Among them, the anti-DSP autoantibody had the highest predictive value, with an AUC (95% CI) of 0.726 (0.682-0.770), and the anti-LGALS4 autoantibody had the lowest diagnostic value, with an AUC (95% CI) of 0.622 (0.573-0.671).

[0100] Example 3: Construction of a diagnostic model for differentiating lung adenocarcinoma from benign lung disease

[0101] 1. Model construction using the training set

[0102] ​The expression amounts of anti-ASS1, anti-DSP, anti-MET, anti-PYCR1 and anti-MDK in the plasma samples of 183 lung adenocarcinoma patients and 183 patients with benign lung diseases in the training set were used as independent variables, and whether the patients were lung adenocarcinoma patients or not was used as dependent variable. Logistic regression analysis was performed on the expression amounts of anti-ASS1, anti-DSP, anti-MET, anti-PYCR1 and anti-MDK in the plasma samples of lung adenocarcinoma patients and patients with benign lung diseases to construct a diagnostic model for distinguishing lung adenocarcinoma patients from patients with benign lung diseases.

[0103] The diagnostic model is: P = 1 / (1+EXP(-(-4.914+1.329x anti-ASS1+0.793x anti-DSP+1.105x anti-MET+0.990x anti-PYCR1+1.375x anti-MDK))).

[0104] In the diagnostic model, P represents the predicted probability, and anti-ASS1, anti-DSP, anti-MET, anti-PYCR1 and anti-MDK represent the expression amounts of the corresponding autoantibodies in the plasma of the subject (the expression amounts are measured as SBI values by the ELISA method described in Example 2).

[0105] The expression amounts of anti-ASS1, anti-DSP, anti-MET, anti-PYCR1 and anti-MDK in the plasma samples were substituted into the diagnostic model to obtain the predicted probability (i.e. P value) of each plasma sample, and the predicted probability PRE = 0.5 was used as the optimal cutoff value for distinguishing lung adenocarcinoma patients from patients with benign lung diseases (if the PRE value is greater than or equal to the cutoff value, the subject is determined to be a lung adenocarcinoma patient, and if the PRE value is less than the cutoff value, the subject is determined to be a patient with benign lung disease), and the corresponding sensitivity and specificity were calculated.

[0106] The ROC curve was plotted according to the predicted probability, and the ROC curve is shown in Figure 5 The AUC (95% CI) of the differential diagnosis model in the training set for distinguishing lung adenocarcinoma from benign lung diseases was 0.789 (0.744-0.834), the sensitivity was 79.1%, the specificity was 70.6%, and the coincidence rate was 64.7%.

[0107] 2. The value of the combination of autoantibodies in diagnosing lung adenocarcinoma was verified by using the validation set.

[0108] The expression levels of anti-ASS1, anti-DSP, anti-MET, anti-PYCR1, and anti-MDK autoantibodies in plasma samples from 66 patients with lung adenocarcinoma and 66 patients with benign lung diseases in the validation set were substituted into the diagnostic model constructed above, P = 1 / (1 + EXP(-(-4.914 + 1.329 × anti-ASS1 + 0.793 × anti-DSP + 1.105 × anti-MET + 0.990 × anti-PYCR1 + 1.375 × anti-MDK))) to obtain the predicted probability of each plasma sample. The optimal cutoff value for diagnosing and distinguishing between patients with lung adenocarcinoma and benign lung diseases was set at a predicted probability of PRE = 0.5, and the corresponding sensitivity and specificity were calculated.

[0109] Plot the ROC curve based on the predicted probability (e.g.) Figure 5 As shown in the figure, the diagnostic value of this model for lung adenocarcinoma was verified. The model's AUC for distinguishing between lung adenocarcinoma and benign lung diseases in the validation set was 0.776 (0.698-0.853), with sensitivity, specificity, and concordance rates of 91.2%, 66.2%, and 54.2%, respectively. Compared with other models, the model provided in this invention can incorporate fewer variables, while exhibiting better logistic regression results and being simple and convenient.

[0110] To facilitate comparative analysis, the AUC values, sensitivity, and specificity of the ROC curves of single anti-tumor-related antigen autoantibodies or combinations of multiple anti-tumor-related antigen autoantibodies in the training and validation sets for diagnosing and distinguishing lung adenocarcinoma from healthy controls were statistically analyzed, as shown in Table 1.

[0111] Table 1. Evaluation of the value of 9 TAAbs and their differential diagnosis in lung adenocarcinoma during Phase II of validation.

[0112]

[0113]

[0114] This model significantly outperforms single anti-tumor-associated antigen autoantibodies; moreover, when this model differentiates between lung adenocarcinoma patients and normal individuals, its ROC curve AUC reaches a maximum of 0.789; at this point, the diagnostic sensitivity reaches 91.2% and the specificity reaches 66.2%, indicating that the combined diagnostic effect of this model is optimal. Therefore, this invention demonstrates that the model for diagnosing lung adenocarcinoma using a combination of multiple anti-tumor-associated antigen autoantibodies possesses good stability.

[0115] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the patent scope of the present application. It should be noted that for ordinary skilled persons in the art, without departing from the concept of the present application, several modifications and improvements can be made, which are within the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A lung adenocarcinoma marker, characterized in that, The biomarkers are autoantibodies against anti-DSP, anti-MDK, anti-MET, anti-ASS1, and anti-PYCR1 antigens.

2. The marker according to claim 1, characterized in that, The autoantibodies mentioned are autoantibodies found in peripheral blood plasma.

3. The use of the biomarker according to claim 1 or 2 in the preparation of products for diagnosing lung adenocarcinoma.

4. The application according to claim 3, characterized in that, The diagnosis of lung adenocarcinoma is to differentiate lung adenocarcinoma from healthy individuals, or to differentiate lung adenocarcinoma from benign lung diseases.

5. The application according to claim 3, characterized in that, The products include reagent kits, reagents, and chips.

6. A reagent kit for diagnosing lung adenocarcinoma, characterized in that, Includes the marker as described in claim 1 or 2.

7. The reagent kit according to claim 6, characterized in that, The diagnosis of lung adenocarcinoma is to differentiate lung adenocarcinoma from healthy individuals, or to differentiate lung adenocarcinoma from benign lung diseases.

8. The reagent kit according to claim 6, characterized in that, The kit is an ELISA detection kit.

9. The use of the biomarker according to claim 1 or 2 in the preparation of a diagnostic model for differentiating lung adenocarcinoma and benign lung diseases.

10. The application according to claim 9, characterized in that, The diagnostic model is as follows: when the P value is ≥0.5, the patient is diagnosed with lung adenocarcinoma; when the P value is <0.5, the patient is diagnosed with benign lung disease. P =1 / (1+EXP(-(-4.914+1.329×anti-ASS1+0.793×anti-DSP+1.105×anti-MET+0.990×anti-PYCR1+1.375×anti-MDK))); Among them, anti-ASS1, anti-DSP, anti-MET, anti-PYCR1, and anti-MDK represent the expression levels of the corresponding autoantibodies in plasma.

Images