Application of plasma autoantibodies in early diagnosis of lung adenocarcinoma

Abstract

The application discloses application of autoantibodies or detection reagents thereof in preparation of a kit for early lung adenocarcinoma diagnosis, wherein the autoantibodies include anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG and anti-WWP2-IgM autoantibodies. The application can improve the diagnosis efficiency of early lung adenocarcinoma (TNM stage 0-I stage), and especially when combined with low-dose computed tomography (LDCT), can accurately and effectively diagnose early lung adenocarcinoma and distinguish between benign and malignant lung nodules which are difficult to distinguish by LDCT.

Description

Technical Field

[0001] This invention belongs to the field of biomedical detection technology, specifically involving the application of plasma anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies in the early diagnosis of lung adenocarcinoma. Background Technology

[0002] Lung cancer is the leading cause of cancer-related deaths worldwide, with lung adenocarcinoma accounting for approximately 40% of all lung cancer cases. Early diagnosis of lung cancer is crucial for reducing mortality, especially since the long-term survival rate after surgery for stage I lung cancer patients can approach 90%. Currently, low-dose computed tomography (LDCT) is the primary screening tool, but it has several limitations: approximately 50% of patients are not suitable for screening; the specificity in distinguishing between benign lung disease (BLD) and early-stage lung adenocarcinoma (Early-LUAD) needs improvement; and there is a shortage of equipment and skilled personnel for LDCT testing in underdeveloped regions. Therefore, there is an urgent need to develop a more effective, non-invasive, and readily accessible method for the diagnosis of early-stage lung adenocarcinoma.

[0003] Plasma biomarkers, such as circulating tumor DNA (CTB) and carcinoembryonic antigen (CEA), are of great significance for the early diagnosis of lung cancer. However, achieving the clinical goal of early lung cancer diagnosis through liquid biopsy remains a significant challenge due to the low levels of CTB in early tumor formation and the instability of tumor-associated antigens. Tumor-associated autoantibodies (TAAs) hold great potential as biomarkers for cancer diagnosis. TAs are immune products produced against oncogenes or abnormal proteins. Because they are present at the onset of early tumors, have long half-lives, are stable in the blood, and are easily detectable, they are biomarkers with clinical translational potential for detecting early-stage cancer.

[0004] To date, several enzyme-linked immunosorbent assay (ELISA) kits targeting combinations of tumor-associated autoantibodies have been developed for lung cancer diagnosis. In 2010, a commercially available ELISA kit, EarlyCDT®-Lung, targeting seven autoantibodies (anti-p53, anti-NY-ESO-1, anti-CAGE, anti-GBU4-5, anti-SOX2, anti-HuD / ELAVL4, and anti-MAGEA4), was developed in a European population. This kit demonstrated a sensitivity of 41% and a specificity of 91%, helping to improve the accuracy of lung cancer nodule assessment using LDCT. The Chinese autoantibody diagnostic kit CancerProbe (anti-p53, anti-CAGE, anti-GBU4-5, anti-GAGE7, anti-SOX2, anti-PGP9.5 / UCHL1, anti-MAGEA1) differs from the EarlyCDT®-LungELISA kit only in the composition of anti-GAGE7 and anti-PGP9.5. Furthermore, CancerProbe exhibits a sensitivity of 61% and a specificity of 90% in the Chinese population.

[0005] Nevertheless, new antibodies are needed to enhance clinical diagnostic efficacy, particularly in detecting early-stage (stage 0 and stage I) lung adenocarcinoma (LUAD). However, research on novel autoantibodies is limited. Traditional discovery techniques, such as serum proteomic analysis (SERPA), serum recombinant expression cDNA cloning analysis (SEREX), and phage display, are cumbersome and technically complex, limiting the development of new autoantibodies. Emerging protein microarrays, capable of simultaneously reacting multiple plasma samples with thousands of immobilized recombinant proteins, may be an effective way to meet high-throughput screening needs; however, obtaining a sufficient variety of recombinant proteins remains another technical challenge. TM It is a microarray containing nearly 20,000 proteins, covering approximately 85% of the human proteome, and has been proven effective in detecting novel autoantibodies in many diseases, demonstrating technological innovation and superiority.

[0006] Furthermore, different types of autoantibodies, including IgG, IgA, and IgM, can be detected in the serum of cancer patients, reflecting different cancer immune surveillance statuses. IgG is the mainstay of humoral immune responses and is the most frequently studied diagnostic biomarker for lung cancer. IgA is mainly involved in mucosal immunity of the gut and respiratory tract. Many studies have shown that IgA autoantibodies alone are insufficient for diagnosing lung cancer. IgM is the earliest antibody to appear after stimulation by tumor-associated antigens (TAAs) during tumor formation. Theoretically, IgM is the most relevant indicator for early cancer detection. Some studies have shown that a combination of five IgM autoantibodies has the potential to diagnose early-stage lung adenocarcinoma (Early LUAD) (accuracy 83.02%), but larger-scale studies are needed to validate this. Considering the importance of IgM in early tumor immune surveillance and the current limited research on it, combining IgG and IgM may be more helpful in improving the early diagnosis of lung cancer. Summary of the Invention

[0007] To address the aforementioned technical problems and improve the diagnostic efficacy for early-stage lung adenocarcinoma, this invention provides the following technical solution:

[0008] In a first aspect, the present invention provides biomarkers for the diagnosis of early lung adenocarcinoma, said biomarkers being autoantibodies, said autoantibodies including anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies or any combination thereof.

[0009] In a second aspect, the present invention provides the use of autoantibody detection reagents in the preparation of kits for the diagnosis of early lung adenocarcinoma, wherein the autoantibodies include anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies or any combination thereof.

[0010] In a third aspect, the present invention provides a kit for the diagnosis of early lung adenocarcinoma, the kit comprising a detection reagent for autoantibodies, the autoantibodies including anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies or any combination thereof.

[0011] In a fourth aspect, the present invention provides a system for the early diagnosis of lung adenocarcinoma, the system comprising:

[0012] An acquisition module, wherein the acquisition module is used to acquire samples from the subject;

[0013] An evaluation module, connected to an acquisition module, is used to detect autoantibodies in a sample using the kit of the present invention. The autoantibodies include anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies or any combination thereof.

[0014] In a fifth aspect, the present invention provides a Comprehensive Risk Score (CRS) system for the early diagnosis of lung adenocarcinoma, the system comprising:

[0015] The acquisition module is used to acquire samples from the subject, as well as gender information and imaging features;

[0016] An evaluation module, connected to the acquisition module, is used to detect autoantibodies in a sample using the kit of the present invention, and incorporates subject gender information and imaging characteristics to perform a comprehensive risk score (CRS). The autoantibodies include anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies or any combination thereof, and the imaging characteristics are the maximum diameter (IMD) of the image examined by LDCT.

[0017] In a sixth aspect, the present invention provides a method for the early diagnosis of lung adenocarcinoma, the method comprising detecting biomarkers, the biomarkers being autoantibodies, the autoantibodies including anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies or any combination thereof.

[0018] In a seventh aspect, the present invention provides a comprehensive risk scoring method for the diagnosis of early lung adenocarcinoma. The method includes detecting biomarkers, obtaining subject gender information and imaging characteristics, and performing a comprehensive risk score. The biomarkers are autoantibodies, including anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies or any combination thereof. The imaging characteristics are the maximum image diameter (IMD) of LDCT examination.

[0019] In some implementations, the autoantibodies are autoantibodies in peripheral blood. In other implementations, the autoantibodies are autoantibodies in serum or plasma.

[0020] In some embodiments, the biomarkers of the present invention are combinations of autoantibodies anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM.

[0021] In some implementations, the autoantibody detection reagent includes reagents capable of qualitatively or quantitatively detecting autoantibodies (including anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies or any combination thereof).

[0022] In some implementations, the detection reagent for autoantibodies includes a substance (e.g., a protein or fragment thereof) capable of specifically binding to autoantibodies.

[0023] In some implementations, the detection reagents for autoantibodies may be included in tools such as kits, chips, or test strips, for example, reagents capable of qualitatively or quantitatively detecting autoantibodies (e.g., proteins or peptides that specifically bind to autoantibodies and specific labeled secondary antibodies).

[0024] In some implementations, the tool may be a tool for high-throughput protein platforms (protein chips) and / or ELISA methods to detect autoantibody expression levels using reagents for qualitative or quantitative detection of autoantibodies (e.g., proteins or peptides that specifically bind to autoantibodies and specific labeled secondary antibodies) to diagnose early-stage lung adenocarcinoma.

[0025] In some implementations, early-stage lung adenocarcinoma is diagnosed based on biomarker quantification (e.g., high-throughput protein platforms (protein chips) and / or ELISA methods).

[0026] In some implementation schemes, subjects with high levels of autoantibody expression are more likely to be diagnosed with lung adenocarcinoma (e.g., early-stage lung adenocarcinoma).

[0027] In some implementation schemes, subjects with low levels of autoantibodies are more likely to be diagnosed with benign pulmonary nodules.

[0028] In some implementation schemes, the criteria for determining high autoantibody expression levels are as follows:

[0029] Detection of autoantibodies' OD by ELISA 450 The value is used to build an AdaBoost machine learning model, where

[0030] The AdaBoost model has a combined prediction probability of >0.5.

[0031] In some implementations, the AdaBoost machine learning model is the AdaBoost machine learning model (R version 4.2.2). In some specific implementations, the AdaBoost machine learning model is performed using the following parameters: mfinal=100, K=5 (SMOTE oversampling).

[0032] In some implementations, early-stage lung adenocarcinoma is diagnosed using a comprehensive risk scoring (CRS) method based on biomarker quantification (e.g., high-throughput protein platforms (protein chips) and / or ELISA methods, such as ELISA results based on the AdaBoost model). This method includes detecting biomarkers, obtaining subject gender information and imaging features, performing a comprehensive risk scoring (CRS), wherein the biomarkers are autoantibodies, including anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies or any combination thereof, and the imaging feature is the maximum image diameter (IMD) obtained from LDCT examination.

[0033] In some implementations, the Comprehensive Risk Score (CRS) is constructed using a logistic regression method, mapping the score to the corresponding beta coefficient. The score range is 0-32, with a baseline reference of <10.

[0034] In some implementation schemes, subjects with high composite risk scores are more likely to be diagnosed with lung adenocarcinoma (e.g., early-stage lung adenocarcinoma).

[0035] In some implementation schemes, subjects with low overall risk scores are more likely to be diagnosed with benign pulmonary nodules.

[0036] In some implementation plans, the criteria for determining a high overall risk score are as follows:

[0037] Detection of autoantibodies' OD by ELISA 450 The value is calculated, and the Comprehensive Risk Score (CRS) is determined, where

[0038] The overall risk score is >10, preferably >25.

[0039] In some implementations, the system further includes an output module for outputting results based on the detection data from the evaluation module.

[0040] In some implementations, the assessment module includes assessing the likelihood of a subject being diagnosed with lung adenocarcinoma by detecting the level of autoantibody expression.

[0041] In some implementations, the assessment module includes assessing the likelihood of a subject being diagnosed with lung adenocarcinoma by detecting the level of autoantibody expression and combining subject gender information and imaging characteristics to generate a comprehensive risk score.

[0042] The beneficial effects of this invention are:

[0043] This invention selects autoantibodies that are easily detected and can well reflect the body's immune function, including anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM autoantibodies or any combination thereof. By combining autoantibodies as biomarkers for the diagnosis of early lung adenocarcinoma, the diagnostic efficacy of early lung adenocarcinoma is improved.

[0044] This invention, through the study of autoantibodies against anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM, contributes to the diagnosis of early-stage lung adenocarcinoma. This invention provides a novel diagnostic method for early-stage lung adenocarcinoma in clinical practice. Especially when combined with gender and imaging characteristics, the Comprehensive Risk Score (CRS) has the potential to significantly improve the sensitivity and specificity of early-stage lung adenocarcinoma diagnosis, and has important clinical application value in the early diagnosis of early-stage lung adenocarcinoma and the differentiation between benign and malignant pulmonary nodules. Attached Figure Description

[0045] Figure 1 The ROC curves and mean AUC of the AdaBoost ensemble learning model, calculated 50 times during the ELISA validation phase, are shown to distinguish early-stage lung adenocarcinoma (Early-LUAD) from benign lung disease (BLD) or healthy controls (NHC). Figure 1 A shows the ROC curve and mean AUC of the AdaBoost ensemble learning model in distinguishing between early-stage lung adenocarcinoma and benign lung diseases; Figure 1 B shows the ROC curve and mean AUC of the AdaBoost ensemble learning model in distinguishing early-stage lung adenocarcinoma from healthy controls.

[0046] Figure 2 The positive predictive value of low-dose computed tomography (LDCT) alone and LDCT combined with autoantibodies for diagnosing lung nodules of different sizes is shown. The combination of autoantibodies and LDCT can significantly improve the positive predictive value for diagnosing lung nodules of different sizes.

[0047] Figure 3The study demonstrated that a comprehensive risk score (CRS) based on sex, maximum image diameter (IMD) on LDCT, and a combination of 10 autoantibodies helps in assessing the risk of early-stage lung adenocarcinoma (Early-LUAD). Patients with CRS scores of 10–25 and >25 had a significantly increased risk of developing Early-LUAD compared to patients with CRS scores <10. Detailed Implementation

[0048] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. These embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Experimental methods in the following embodiments, unless otherwise specified, are generally performed under conventional conditions in the art or as recommended by the manufacturer. Unless otherwise specified, all methods are conventional. Unless otherwise defined, technical and scientific terms used herein have the same meaning as those familiar with the art.

[0049] abbreviation:

[0050] ELAVL4: ELAV-like protein 4;

[0051] GDA: Guanine deaminase;

[0052] GIMAP4: GTPase, IMAP family member 4;

[0053] MGMT: O-6-methylguanine-DNA methyltransferase;

[0054] UCHL1: Ubiquitin carboxyl-terminal hydrolase L1;

[0055] DCTPP1: dCTP pyrophosphatase 1, nucleoside triphosphate pyrophosphatase

[0056] KCMF1: potassium channel modulatory factor 1;

[0057] WWP2: WW domain containing E3 ubiquitin protein ligase 2.

[0058] The reagents used in the embodiments of this invention are sourced from:

[0059] The high-throughput protein chip (HuProt™) was purchased from CDI LABS, catalog number CDIHP-004; the small chip was purchased from CDI LABS, catalog number CDIHP-005.PC; the ELISA kit was a self-made kit (not a commercially available kit) from the Beijing Key Laboratory for Clinical Research of Antitumor Molecular Targeted Drugs.

[0060] Samples were collected from 58 patients with early-stage lung adenocarcinoma (Early-LUAD), 30 patients with benign lung disease (BLD), and 24 healthy controls (NHC) for Huprot. TM Chip screening. All samples were collected between December 2019 and March 2021. Early-LUAD patients were those with surgically confirmed stage 0, IA, and IB. Lung cancer staging was determined according to the 8th edition of the TNM tumor staging system developed by the American Joint Committee on Cancer (AJCC) and the International Union for Cancer Control (UICC). Benign lung disease (BLD) patients were those with histopathologically confirmed benign pulmonary nodules, including tuberculosis and suspected tuberculous nodules, organizing pneumonia, hamartomas, sclerosing alveolar cell tumors, and inflammatory pseudotumors. Healthy control (NHC) samples had no history of malignancy during routine cancer screening, and no suspicious findings were found in a series of imaging and laboratory tests. Huprot was used. TM The chip detects autoantibodies in blood, and the "limma" package in R is used to screen for differentially expressed autoantibodies. The specific experimental procedure is as follows:

[0061] 1.1 Preparation of patient serum samples

[0062] Plasma samples were collected from patients with LUAD and BLD at initial diagnosis and before any surgical intervention. NHC samples were collected during routine cancer screening. All whole blood samples underwent initial separation by centrifugation at 3000 r / min and 4°C for 10 minutes on the day of collection. The supernatant was immediately stored at -80°C for subsequent analysis and thawed once before microarray analysis.

[0063] 1.2 Differential Protein Screening Steps

[0064] 1.2.1 Chip Preparation: Remove the HuProt chip from the -80°C freezer. TM Place the chip in a chip box and let it warm up to room temperature for 20 minutes.

[0065] 1.2.2 Blocking: Place the chip protein side up in the container, add 5 mL of blocking solution (5% bovine serum albumin phosphate buffer solution), gently shake to cover the chip protein side, and cover the container. Place it on a shaker and shake it horizontally slowly (70 times / min) for 1.5 h at room temperature.

[0066] 1.2.3 Sample preparation: Take out the subject's serum sample from -80°C in advance, thaw it in an ice bath or at room temperature, shake it to mix, and then centrifuge it at high speed (12000 rpm) for 10 min; take 5 μL of the supernatant and mix it with 5 mL of blocking solution by vortexing at a ratio of 1:1000.

[0067] 1.2.4 Sample addition: Discard the sealing solution in the incubation tank, add the sample prepared in 1.2.3 directly, and gently shake to cover the chip; cover the tank and incubate at room temperature for 1 h.

[0068] 1.2.5 Rinsing: Wash 6 times with PBST buffer. The last 3 rinsing processes should be placed on a shaker and shaken rapidly left and right for 10 minutes each time, with a reciprocating frequency of 100 times / min.

[0069] 1.2.6 Secondary antibody preparation: Turn off the lights and remove the blocking buffer and goat anti-human fluorescent Alexa fluor647 IgG secondary antibody from the freezer 10 minutes in advance. Take a 5 mL cryovial, wrap it with aluminum foil, add 10 μL of secondary antibody IgG and 5 mL of blocking buffer, and vortex to mix.

[0070] 1.2.7 Adding secondary antibody: After rinsing, add 5 mL of secondary antibody to the incubation tank and gently shake to ensure it is completely submerged; cover the tank and wrap it with aluminum foil to protect it from light. Place it on a shaker and gently shake horizontally (70 times / min) for 1 h at room temperature.

[0071] 1.2.8 Rinsing: Wash 6 times with PBST buffer, with the last 3 rinses performed on a shaker, rapidly shaking left and right for 10 minutes each time. Then wash 3 times with 10× PBST buffer, and shake rapidly left and right for 10 minutes each time on a shaker.

[0072] 1.2.9 Drying the chips: After rinsing, remove the chips from the incubation tank, hang them vertically on absorbent paper to drain, and place them in the chip box.

[0073] 1.2.10 Scanning and image reading:

[0074] 1.2.10.1 Warm up the device 20 minutes in advance: Turn on the computer power button, host switch, and scanner switch; open the Genepix pro-Device software.

[0075] 1.2.10.2 Place the chip with the protein side of the barcode facing down—set the laser excitation to 635nm (power = 95, photomultiplication = 700) for scanning—click the green triangle start button to begin scanning—after scanning, click the white envelope, the third bar, and save in .tif format.

[0076] 1.2.11 Data Extraction / Marking:

[0077] Open Genepix software – drag the TIFF and GAL files together into a small window – settings (top left image and bottom right blue settings PMT-GAN) – data processing: select the region, adjust the region's position based on fixed positive points (two adjacent points with strong signals are considered positive), only operate to ensure that the quasi-positive points (intensity about 10 times that of the background) are exactly circled (move up, down, left, and right, use Ctrl+up and down to adjust the size) – save as a GPS format, with the same name as the TIFF format.

[0078] 1.3 Experimental Results

[0079] All raw intensity data obtained in section 1.2.11 were normalized by Loess and transformed by log2 before differential analysis. Differential autoantibodies were screened among 58 Early-LUAD, 30 BLD, and 24 NHC samples using the R software limma (version 4.1.1). 229 IgG and 163 IgM autoantibodies were found to be significantly higher in Early-LUAD than in BLD / NHC. The sensitivity of these IgG autoantibodies ranged from 10.34% to 34.48%, with a specificity exceeding 83.3%. Similarly, the sensitivity and specificity of IgM autoantibodies ranged from 7.89% to 36.84% and 90%, respectively. A total of 418 candidate autoantibodies were selected for microarray preparation, including 392 newly discovered autoantibodies and 26 previously reported autoantibodies (see Table 1).

[0080] Table 1. Results of autoantibody screening on large-chip microarrays

[0081]

[0082]

[0083]

[0084] Example 2: Preliminary Validation of the Diagnostic Efficacy of Autoantibodies Using a Small Chip

[0085] Samples were collected from 342 patients with early-stage lung adenocarcinoma (Early-LUAD), 150 patients with benign lung disease (BLD), and 193 healthy controls (NHC) for microarray validation. All samples were collected between December 2019 and March 2021. Early-LUAD patients were defined as those with surgically confirmed stage 0, IA, and IB. Lung cancer staging was determined according to the 8th edition of the TNM tumor staging system developed by the American Joint Committee on Cancer (AJCC) and the International Union for Cancer Control (UICC). BLD patients were defined as those with histopathologically confirmed benign pulmonary nodules, including tuberculosis and suspected tuberculous nodules, organizing pneumonia, hamartomas, sclerosing alveolar cell tumors, and inflammatory pseudotumors. Healthy controls (NHC) samples had no history of malignancy in their annual cancer screening, and no suspicious findings were found in a series of imaging and laboratory tests. The microarray data were used to screen for differentially expressed autoantibodies among Early-LUAD, BLD, and NHC using the R software limma (version 4.1.1). The specific experimental procedure is as follows:

[0086] 2.1 Fabrication of small chips

[0087] The small chip fabrication process involves expressing a protein with a GST tag using a yeast expression system, and then dividing it into 2×7 subarrays using a 14-chamber rubber gasket (GraceBio Corp, Bend, OR).

[0088] 2.2 Verification of the small chip:

[0089] 2.2.1 Preservation of protein chips: Store the plastic chip cartridges containing the targeted protein chips in a -80°C freezer. After removing the chip cartridges, place them in PE gloves and leave them at room temperature for 20 minutes.

[0090] 2.2.2 Install the fence: With the chip face up, attach the matching chip fence to the chip according to the position of each array on the chip, and secure it to ensure that the fence is completely attached to the chip. Then place it in the incubation box.

[0091] 2.2.3 Blocking: Add 60 μL of blocking solution (3% BSA (w / v), PBS-T) to each array of the chip, incubate at room temperature for 1 hour, and gently shake on a shaker (60 rpm).

[0092] 2.2.4 Sample preparation: Take out the serum in advance, dissolve it, centrifuge at 12000 rpm for 10 minutes, dilute the serum with blocking buffer (3% BSA (w / v), PBS-T) at a ratio of 1:2000, mix well, and serial dilution is recommended.

[0093] 2.2.5 Hybridization: Discard the blocking solution from the chip array, and carefully add the prepared serum sample (60 μL) to each array, avoiding the formation of air bubbles. Incubate at room temperature for 1 hour.

[0094] 2.2.6 Washing: After incubation, discard the liquid in the chip array, remove the baffles, and place the chips into the incubation chamber. Add 10.0 mL of PBS-T to each cell, and wash by shaking at room temperature for 10 minutes (40 rpm). Repeat three times.

[0095] 2.2.7 Drying the chip: Use flat-tipped tweezers to remove the chip from the incubation box and place it vertically in a 50 mL centrifuge tube. Centrifuge at 1000 rpm for 2 minutes. Place absorbent paper on the work surface, remove the chip from the centrifuge tube, and place it vertically in contact with the absorbent paper to absorb any remaining moisture from the edges of the chip.

[0096] 2.2.8 Install fences and add fluorescently labeled secondary antibody for incubation: Install fences according to the position of each array on the chip, ensuring proper fit. Dilute the secondary antibody (Alexa 647-labeled goat anti-human IgG antibody) with blocking buffer 1:2000 BSA, vortex to mix, add 60 μL of the diluted secondary antibody to each array, and incubate at room temperature for 1 hour, protected from light.

[0097] 2.2.9 Washing: After incubation, discard the liquid in the chip array, remove the enclosure, and place the chips into the incubation chamber. Add 10.0 mL of PBST to each cell, gently shake at room temperature for 10 minutes (40 rpm), repeat three times, and protect from light. Rinse the protein chips with 10 mL of ddH2O for 10 minutes each time, repeating 3 times.

[0098] 2.2.10 Drying the slide: Using flat-tipped forceps, remove the chip from the incubation box and place it vertically in a 50 mL centrifuge tube. Centrifuge at 1000 rpm for 2 minutes. Place absorbent paper on the work surface, remove the chip from the centrifuge tube, and place it vertically in contact with the absorbent paper to absorb any remaining moisture from the edges of the chip. Transfer the dried chip to a new, clean slide container.

[0099] 2.2.11 Scanning and Saving Data

[0100] The specific process is the same as steps 1.2.10 and 1.2.11 of Example 1.

[0101] 2.3 Experimental Results

[0102] Autoantibody data from 342 Early-LUAD patients, 150 BLD patients, and 193 NHC controls obtained in section 2.2.11 were analyzed using the limma package for differential analysis, which further narrowed the scope to autoantibodies targeting 32 tumor-associated antigens (TAAs). These differentially expressed autoantibodies were selected based on their fold change (FC) and sensitivity values ​​and will be used for further validation by subsequent ELISA (see Table 2).

[0103] Table 2 Preliminary validation of small chip autoantibodies

[0104]

[0105] Example 3: ELISA was used to finally verify the diagnostic efficacy of the autoantibody.

[0106] Samples were collected from 234 patients with early-stage lung adenocarcinoma (Early-LUAD), 100 patients with benign lung disease (BLD), and 115 healthy controls (NHC) for ELISA validation. All samples were collected between December 2019 and March 2021. Early-LUAD patients were defined as those with surgically confirmed stage 0, IA, and IB. Lung cancer staging was determined according to the 8th edition of the TNM tumor staging system developed by the American Joint Committee on Cancer (AJCC) and the International Union for Cancer Control (UICC). Patients with benign lung disease (BLD) were defined as those with histopathologically confirmed benign pulmonary nodules, including pulmonary tuberculosis and suspected tuberculous nodules, organizing pneumonia, hamartomas, sclerosing alveolar cell tumors, and inflammatory pseudotumors. Healthy controls (NHC) samples had no history of malignancy during routine cancer screening, and no suspicious findings were found in a series of imaging and laboratory examinations. An AdaBoost machine learning model was used, with the SMOTE algorithm applied for oversampling in the smaller sample size group. Using the Caret package (v6.0-86), the samples were randomly divided into training and validation sets at a ratio of 75%:25%, and the training-validation process was repeated 50 times to ensure the robustness of the model. The area under the curve (AUC) of receiver operating characteristic (ROC) analysis was used to evaluate the model performance, and sensitivity and specificity were calculated. Subsequently, a comprehensive risk score (CRS) was constructed using logistic regression. The specific experimental procedure is as follows:

[0107] 3.1 ELISA Steps

[0108] 3.1.1 Protein coating:

[0109] 3.1.1.1 Preparation of coating solution: Weigh 1.59 g of Na2CO3 and 2.93 g of NaHCO3, dissolve them in 900 mL of deionized water, adjust the pH to 9.6, add water to 1 L, and store at 4°C.

[0110] 3.1.1.2 Remove the target protein, thaw it on ice, and dilute it to 1 μg / mL using coating buffer.

[0111] 3.1.1.3 Add 50 μL of diluted protein to each well, seal the plate, and incubate overnight at 4°C.

[0112] 3.1.2 Closure

[0113] 3.1.2.1 Prepare at least 30 mL of 5% milk in advance using PBST and place it at 4°C, and 500 mL of PBST at room temperature.

[0114] 3.1.2.2 Remove the coated plate, discard the coating solution, and wash with PBST 5 times for 3 minutes each time.

[0115] 3.1.2.3 Add 50 μL of 5% milk to each well, seal the plate, and incubate at room temperature for 2 h.

[0116] 3.1.3 Plasma dilution

[0117] 3.1.3.1 Half an hour before the end of the sealing process, remove the plasma sample, thaw it on ice for 20 minutes, and remove the plasma sample immediately before opening the cap.

[0118] 3.1.3.2 Dilute the plasma with 5% milk to 1:100, 1:300, and 1:600, mix well, and store at 4°C for later use (the primary antibody concentration was set at 1:300 based on the preliminary experimental results).

[0119] 3.1.4 Incubate primary antibody (with plasma)

[0120] 3.1.4.1 Discard the milk in the plate, wash 5 times with PBST for 5 minutes each time, and pat the liquid in the plate dry on paper after the last wash.

[0121] 3.1.4.2 Add 50 μL of diluted plasma to each well, 50 μL of 5% milk to each negative control well, and anti-GST protein-tagged antibody to each positive control well. Detect anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM, respectively. Seal the plate and incubate on a shaker at room temperature for 1 h.

[0122] 3.1.5 Add secondary antibody

[0123] 3.1.5.1 Dilution of secondary antibody: Take out the goat anti-human IgG secondary antibody (Jackson) and dilute it with 5% milk to determine a dilution concentration of 1:8000.

[0124] 3.1.5.2 After the primary antibody incubation is complete, discard the primary antibody liquid in the plate, wash 5 times with PBST for 5 minutes each time, and pat the liquid in the plate dry on paper after the last wash.

[0125] 3.1.5.3 Add 50 μL of diluted HRP-labeled secondary antibody to each well, seal the plate, and incubate on a shaker at room temperature for 1 h.

[0126] 3.1.6 Color Development

[0127] 3.1.6.1 After the secondary antibody incubation begins, take out an appropriate amount of the colorimetric solution 3,3',5,5'-tetramethylbenzidine (TMB) and place it in a dark place at room temperature.

[0128] 3.1.6.2 After the secondary antibody incubation is complete, discard the secondary antibody liquid in the plate, wash 5 times with PBST for 5 minutes each time, and pat the liquid in the plate dry on paper after the last wash.

[0129] 3.1.6.3 Add 100 μL of TMB to each well. Start timing when the TMB is completely added.

[0130] 3.1.6.4 Place at room temperature for 25-30 minutes, the specific time is when the positive well turns blue.

[0131] 3.1.6.5 Add 50 μL of 0.5 M sulfuric acid to each well to terminate the colorimetric reaction.

[0132] 3.1.7 Detection

[0133] Detect immediately after adding sulfuric acid to stop the reaction, and after a brief shake, detect at a wavelength of 450 nm.

[0134] 3.1.8 Statistical Analysis

[0135] To evaluate the combined diagnostic potential of 10 identified autoantibodies (anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM), an AdaBoost ensemble learning model using ELISA data was employed. The AdaBoost machine learning model used the following parameters: mfinal=100, K=5 (SMOTE oversampling). First, the ELISA data were randomly divided into training and validation sets at a ratio of 75%:25%. In the training set, a model was built through 10 cross-validations. Then, the model was validated and evaluated using the validation set. The entire modeling and validation process was randomly repeated 50 times, generating a total of 500 AdaBoost models. Subsequently, a comprehensive risk score (CRS) for early-stage lung adenocarcinoma (Early LUAD) was established. The score was constructed using logistic regression, with significant factors identified by the β coefficient. These factors included gender, maximum intramural diameter (IMD) of the LDCT image, and the ELISA results of 10 autoantibody combinations for binary classification (obtained from the AdaBoost ensemble learning model mentioned above; a predicted probability >0.5 for the 10 autoantibody combinations was considered positive, and <0.5 was considered negative). Males with IMD ≤8mm and negative results for the 10 autoantibody combinations served as the control group. The CRS score ranged from 0 to 32, with higher scores indicating a greater risk of Early-LUAD.

[0136] 3.2 Experimental Results

[0137] The AdaBoost model's overall prediction probability > 0.5 tends to indicate a diagnosis of early-stage lung adenocarcinoma. Evaluation of the validation set showed that the AUC distinguishing Early-LUAD from BLD or NHC was 0.70–0.85 (mean 0.77) or 0.74–0.91 (mean 0.80) (see [link to validation data]). Figure 1 Compared with single autoantibodies (sensitivities of 11.06%-24.68%) and EarlyCDT®-Lung (sensitivity of 41%), the combination of 10 autoantibodies significantly improved the sensitivity (70.5%) for the diagnosis of Early-LUAD (see Table 3). Figure 1 ).

[0138] Table 3. Performance of single autoantibodies: sensitivity 11.06% ~ 24.68%, specificity > 90%.

[0139]

[0140] In addition, the positive predictive value (PPV) of 10 autoantibody combinations combined with LDCT was evaluated to differentiate between benign and malignant nodules of different sizes reported on LDCT. To distinguish between benign and malignant nodules in the Early-LUAD and BLD groups, nodule size was determined by the image diameter (IMD) on LDCT. Compared with LDCT alone, the PPV of LDCT combined with autoantibody combinations increased from 47.3% to 79.2% for nodules with IMD ≤ 8 mm, from 52.0% to 71.1% for nodules with IMD ≤ 20 mm, and from 62.9% to 87.9% for nodules > 20 mm (see [link to relevant documentation]). Figure 2 Subsequently, we incorporated ELISA results from 10 autoantibody combinations categorized by sex, IMD, and binary classification to construct a Comprehensive Risk Score (CRS). Patients with early-stage lung adenocarcinoma (Early-LUAD) and benign lung disease (BLD) were assessed based on the CRS score, categorizing them into three groups: <10, 10-25, and >25. Compared to patients with a CRS <10, patients in the 10-25 and >25 groups had a significantly increased risk of Early-LUAD, with odds ratios (ORs) of 5.28 (95% CI: 3.18–8.76) and 9.05 (95% CI: 5.40–15.15), respectively (see [link to relevant documentation]). Figure 3 ).

[0141] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present invention, various simple modifications can be made to the technical solution of the present invention, and these simple modifications all fall within the protection scope of the present invention.

Claims

1. The application of autoantibody detection reagents in the preparation of kits for the diagnosis of early lung adenocarcinoma, wherein the autoantibodies include combinations of anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM.

2. The application according to claim 1, wherein the kit is used to detect the expression level of autoantibodies for the diagnosis of early lung adenocarcinoma.

3. The application according to claim 2, wherein the detection reagent includes reagents for protein chips and / or ELISA, and the method for detecting the expression level of the autoantibody includes protein chips and / or ELISA.

4. A system for early diagnosis of lung adenocarcinoma, wherein the system comprises: An acquisition module, wherein the acquisition module is used to acquire samples from the subject; An evaluation module, connected to an acquisition module, includes a kit containing detection reagents for detecting autoantibodies in a sample. The autoantibodies include combinations of anti-ELAVL4-IgM, anti-GDA-IgM, anti-GIMAP4-IgM, anti-GIMAP4-IgG, anti-MGMT-IgM, anti-UCHL1-IgM, anti-DCTPP1-IgM, anti-KCMF1-IgM, anti-UCHL1-IgG, and anti-WWP2-IgM.

5. The system according to claim 4, wherein the assessment module is used to detect autoantibodies in the sample and incorporate subject gender information and imaging characteristics to perform a comprehensive risk score.

Images