Method for constructing diagnostic model for distinguishing between benign and malignant pulmonary nodules on basis of single-cell immune profile

Abstract

A method for constructing a diagnostic model for distinguishing between benign and malignant pulmonary nodules on the basis of a single-cell immune profile, which method comprises: obtaining PBMCs from a peripheral blood sample, and performing mass cytometry analysis on the PBMCs to obtain a mass cytometry analysis dataset; classifying cells into different phenotypes on the basis of marker expression and by using a PARC clustering algorithm; and using the expression proportion of each cell subgroup as modeling features to obtain a diagnostic model for distinguishing between benign and malignant pulmonary nodules, thereby realizing the screening and early diagnosis of lung cancer. The model features non-invasiveness, high sensitivity and high specificity, improves the diagnostic accuracy of lung cancer screening, and provides patients with earlier treatment opportunities and more suitable surgical approaches.

Description

Construction method of lung nodule benign and malignant differential diagnosis model based on single cell immunogram TECHNICAL FIELD

[0001] The present application belongs to the field of medical treatment, more particularly relates to a method for establishing a lung nodule benign and malignant differential diagnosis model based on single cell immunogram by mass spectrometry flow cytometry technology. BACKGROUND

[0002] Lung cancer is one of the highest incidence and mortality rates of malignant tumors worldwide. The vast majority of patients (48%) have distant metastasis at the time of initial diagnosis, and the 5-year relative survival rate of these patients is only about 8%, while the 5-year survival rate of patients diagnosed as early-stage lung cancer is up to 62%, but its detection rate is only about 20%. In contrast, for stage I cancer, 5-year survival rates increase significantly, from 68% to 92%. There is no doubt that increasing the detection rate of lung cancer at the curable stage (0, I and II) is the most effective way to reduce lung cancer mortality. Improving the early diagnosis rate of lung cancer is of great importance to reduce lung cancer mortality and improve lung cancer prognosis.

[0003] However, due to the lack of clinical symptoms and sensitive techniques, it is difficult to detect lung cancer at these early stages. Randomized controlled trials, including the National Lung Screening Trial (NLST) and the Dutch-Belgian Lung Cancer Screening Trial (NELSON) trial, have shown that screening with low-dose CT (LDCT) can significantly reduce lung cancer mortality, but there is still uncertainty in improving clinical effectiveness and cost-effectiveness of LDCT, especially in the diagnosis of subsolid nodules, which is the most difficult and challenging, and there is a problem with LDCT screening, which is the detection of a large number of false-positive nodules: most of the nodules detected by LDCT in high-risk populations are benign nodules. A large-scale NLST study from the United States showed that while 24.2% of people in the population receiving LDCT examination found lung nodules, after follow-up, as high as 96.4% of positive nodules were diagnosed as benign (false positives); other studies have also shown that the proportion of benign lung nodules after surgical resection is as high as 20%, and after biopsy, it is as high as 38%. Multiple studies have confirmed that the average detection rate of nodules by LDCT screening is about 20%, and >90% of the nodules are benign. Excessive false positives can lead to over-diagnosis, over-treatment, waste of medical resources, and increase anxiety in the examinees.

[0004] Mayo model (Mayo Clinic's clinical prediction model for lung cancer in 1997, which is based on age, smoking history, nodule diameter, spiculation, location, etc. of patients with nodules ranging from 4-30 mm), Veterans Association (VA) model and BROCK model are used to determine the nature of the nodule, and the area under the curve (AUC) is 0.65-0.83, the model accuracy is not high, and the Mayo Clinic's clinical prediction model for lung cancer is based on western patients, which may not be suitable for Chinese patients, and there are still some limitations in practical application.

[0005] Therefore, effectively further identifying and shunting diagnosis of lung nodules, quickly determining their benign or malignant, resecting malignant nodules as early as possible, while avoiding unnecessary overtreatment and reducing the proportion of benign nodules resected by surgery, are the keys to the diagnosis and treatment of lung nodules. For the huge unmet clinical needs, there is an urgent need for a method with high sensitivity and high specificity to accurately identify malignant lung nodules.

[0006] On the other hand, the World Health Organization (WHO) classification of lung tumors (5th edition, 2021) classifies atypical adenomatous hyperplasia (AAH) and adenocarcinoma in situ (AIS) as precancerous lesions of the gland, and does not include lung adenocarcinoma. Microinvasive adenocarcinoma (MIA) and invasive adenocarcinoma (IA) are classified as two types of lung cancer with different degrees of invasion. The International Association for the Study of Lung Cancer / American Thoracic Society / European Respiratory Society Multidisciplinary Classification of Lung Adenocarcinoma indicates that MIA often does not spread to regional lymph nodes or metastasis, and the postoperative disease-free survival rate is almost 100%, so MIA can choose sublobar resection and does not need systematic lymph node dissection, while the surgical method for IA patients needs to be combined with the clinic. Therefore, a method is needed to effectively distinguish MIA and IA before surgery to develop a surgical plan in advance.

[0007] The human immune system is closely related to the occurrence and development of tumors, and the tumor immune microenvironment (TME) is also constantly changing during the process of tumor occurrence and development. The immune changes of the tumor are not only limited to the tumor, but are often accompanied by systemic immune disorders, which highlight the complex interactions between the tumor and the peripheral immune system. Studies have shown that there are many types of infiltrating immune cells in lung tumors, which play an extremely important role in the occurrence and development of tumors. Macrophages and T cell populations have potential interactions in the tumor immune microenvironment, and lung tumors are also rich in other myeloid components, including neutrophils, non-classical monocytes and intermediate monocytes, etc.; at the same time, studies have shown that the presence of B cells is related to the protective immunity of lung cancer patients; some clinical studies have shown that high density of tumor-infiltrating T cells is associated with increased median survival of cancer patients. Therefore, comprehensive and detailed monitoring of the peripheral immune status of the human body can provide great help for the identification of benign and malignant lung nodules.

[0008] In recent years, mass cytometry technology combines traditional flow cytometry methods with mass spectrometry detection methods, uses metal isotope labels instead of fluorescent labels, and uses mass spectrometry to quantify the labels, which can realize the simultaneous detection of more than 40 target proteins at the single cell level, and there is no mutual interference between channels, without complex and tedious compensation calculation, which significantly enhances the ability to evaluate complex cellular systems and processes, and is a multi-parameter, high-throughput single cell protein detection technology platform. By detecting multiple marker combinations, CyTOF technology can distinguish multiple cell subgroups, construct cell atlas of healthy or disease state, comprehensively analyze intracellular signaling network, analyze human basic immune state in high dimension, and obtain comprehensive immune cell composition, phenotype and function information. SUMMARY

[0009] Therefore, the purpose of the present application is to provide a method for constructing a lung nodule benign and malignant differential diagnosis model based on single cell immune atlas, which is a method for establishing a lung nodule benign and malignant differential diagnosis model based on single cell immune atlas of mass cytometry technology. Peripheral blood mononuclear cells (PBMCs) are obtained by treating peripheral blood samples with Ficoll separation method (Ficoll density gradient centrifugation method), and mass cytometry analysis is performed on the PBMCs to obtain mass cytometry analysis data set; using PARC clustering algorithm, the cells are divided into different phenotypes according to marker expression, and the expression proportion of each cell subgroup is used as modeling feature to obtain the lung nodule benign and malignant differential diagnosis model.

[0010] In order to achieve the above purpose, the present application adopts the following technical solutions:

[0011] 1. PBMCs are obtained by treating peripheral blood samples with Ficoll separation method, and they are suspended in 5ml pre-cooled FACS buffer (1xPBS+0.5%BSA), then centrifuged at 400xg for 5min at 4℃ to discard the supernatant, and the cell pellet is resuspended in buffer. Cell counting and quality assessment of PBMCs are performed before mass cytometry (CyTOF, Mass Cytometry) detection to ensure that the count is greater than 3x106 and the survival rate is greater than 85%.

[0012] 2. 40 metal-conjugated antibodies were selected as markers to label cells; PBMCs were washed with PBS buffer, then stained with 0.5 mM cisplatin, and the binding of cells to antibodies was blocked by Fc receptor for 30 min, and unbound antibodies were removed by centrifugation. Then PBMCs were fixed in 200 μL of insertion solution overnight. Cells were washed and resuspended in distilled water, added to 20% EQ beads, and further analyzed on a mass cytometer. The markers of 40 metal-conjugated antibodies include: CD45, CD3, CD56, TCRy / d, CD196 (CCR6), CD14, IgD, CD123 (IL-3Ra), CD85j (ILT2), CD19, CD25 (IL-2Ra), CD274 (PD-L1), CD278 (ICOS), CD39, CD27, CD24, CD45RA, CD86, CD28, CD197 (CCR7), CD11c, CD33, CD152 (CTLA-4), CD161, CD185 (CXCR5), CD66b, CD183 (CXCR3), CD94, CD57, CD45RO, CD127 (IL-7Ra), CD279 (PD-1), CD38, CD194 (CCR4), CD20, CD16, HLA-DR, CD4, CD8a, CD11b.

[0013] 3. The PARC clustering algorithm was used to divide cells into different phenotypes according to marker expression, and the expression ratio of each cell subgroup was used as a modeling feature, and the random forest algorithm and 10-fold cross-validation were used to screen features.

[0014] 4. A lung nodule benign and malignant differential diagnosis model was obtained by random forest modeling.

[0015] Optionally, step 3 further comprises screening 34 immune cell subgroups and markers as modeling features. The 19 features of the lung nodule benign and malignant differential diagnosis model include: CD33-CD14-CD3+CD4+CD28+, CD33-CD14-CD3+CD4+CD274+, CD33-CD14-CD3+CD4+CD197-CD45RA-, CD33-CD14-CD3+CD4+HLA_DR+CD38+, CD33-CD14-CD3+CD4+CXCR5-CD183-CCR6-, CD33-CD14-CD3+CD4+CD25+CD127-CD161-CD45RA+, CD33-CD14-CD3+CD8+CD197-CD45RA+, CD33-CD14-CD3-CD19+CD24+CD38+, CD33-CD14-CD3-CD20-CD38+CD27+, CD33-CD14-CD3-CD56+CD16+CD94+, CD33-CD14-CD3-CD56+CD16+CD161+, CD3-CD19-CD56-CD14-CD123+CD11c+, CD33-CD14-CD3-CD56+CD16-, CD86, CD11c, CD183, CD94, CD4, CD11b; the 15 features of the lung cancer infiltration degree judgment model include: CD33-CD14-CD3+CD8+CD85j+, CD33-CD14-CD3+CD8+CD161-, CD33-CD14-CD3+CD4+, CD33-CD14-CD3+CD4+CD197-CD45RA+, CD33-CD14-CD3+CD4+HLA_DR+CD38+, CD33-CD14-CD3+CD4+HLA_DR+CD38-, CD33-CD14-CD3+CD8+, CD33-CD14-CD3+CXCR5+, CD33-CD14-CD3+CD8+CD197+CD45RA-, CD33-CD14-CD3-CD19+CD24+CD38+, CD33-CD14-CD3-CD56+CD16+, CD33-CD14-CD3-CD56+CD16+CD57+, CD33-CD14-CD3-CD56+CD16+HLA_DR+, CD3-CD19-CD56-CD14-HLA_DR-, CD56.

[0016] Optionally, step 3 further comprises dividing the samples into the training set and the validation set according to the time sequence of entering the group.

[0017] Optionally, step 2 adopts a combination of 40 antibody markers.

[0018] Optionally, in step 1, when the samples are selected for grouping, random grouping is adopted, covering samples of different nodule sizes (such as ≤10 mm, 11-20 mm, 21-30 mm, etc.), different types of nodules (such as solid nodules, partially solid nodules, and pure ground glass density nodules), and different types of adenocarcinoma infiltration (such as atypical adenomatous hyperplasia AAH, carcinoma in situ AIS, microinvasive adenocarcinoma MIA, and invasive adenocarcinoma IA).

[0019] According to the technical solution, the application provides a method for establishing a lung nodule benign and malignant differential diagnosis model of single cell immunomapping. The lung cancer (pathologically confirmed CA) and non-cancer group samples (including image confirmed Non-CA and pathologically confirmed Non-CA) have good detection performance in differential diagnosis. The AUC of the training set and the verification set can reach 0.95 and 0.96, which is better than the existing clinical models: Mayo model (the AUC of the training set and the verification set is 0.75 and 0.70), Veterans Association (VA) model (the AUC of the training set and the verification set is 0.73 and 0.65), and BROCK model (the AUC of the training set and the verification set is 0.84 and 0.85). Moreover, in the most difficult to identify pathological non-cancer group and lung cancer group, the AUC of the training set and the verification set is 0.92 and 0.90, which is obviously better than the existing clinical models: Mayo model (the AUC of the training set and the verification set is 0.69 and 0.61), Veterans Association (VA) model (the AUC of the training set and the verification set is 0.68 and 0.61), and BROCK model (the AUC of the training set and the verification set is 0.72 and 0.65).

[0020] Meanwhile, the model established by the present application can effectively distinguish MIA and IA before operation, and the AUC of the training set and the verification set is 0.97 and 0.93. The model improves the sensitivity and specificity of lung cancer screening, reduces the risk of missed diagnosis and misdiagnosis, and can be used for auxiliary diagnosis and screening of lung cancer.

[0021] The application provides a lung cancer screening technology based on mass spectrometry flow cytometry technology, which realizes the screening and early diagnosis of lung cancer by detecting the expression profile of tumor-related immune cells in the peripheral blood of patients, classifying the peripheral blood sample data, combining an artificial intelligence algorithm, constructing a learning model, inputting the peripheral blood sample data into the model, and obtaining the risk score of the patient for lung cancer.

[0022] The application has the advantages that: (1) the sensitivity and specificity of lung cancer screening are improved, and the risk of missed diagnosis and misdiagnosis is reduced; (2) invasive diagnostic operations are avoided, and the clinical risk is reduced; (3) early diagnosis of lung cancer is realized, and patients are provided with earlier treatment opportunities; (4) the infiltration degree of lung cancer is further typed, and more suitable surgical methods are provided for patients; (5) the CyTOF technology is adopted, which has the characteristics of high throughput, high resolution and high sensitivity, and can quickly and accurately detect the expression profile of lung cancer-related immune cells; (6) the artificial intelligence algorithm is combined, and the accuracy and reliability of lung cancer screening are improved. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of the provided drawings.

[0024] Fig. 1 is a flowchart of the lung cancer screening based on the CyTOF cell immunogram of the application.

[0025] Fig. 2 is a lung cancer and non-cancer group sample risk score box plot of the training set and the verification set of the lung cancer diagnosis model of the application.

[0026] Fig. 3 is a training set and verification set AUC graph of the lung cancer diagnosis model of the application.

[0027] Fig. 4 is a lung cancer and non-cancer group sample risk score box plot of the training set and the verification set of the lung cancer diagnosis model of the application in the most difficult to identify pathological non-cancer group and lung cancer group.

[0028] Fig. 5 is a lung cancer and non-cancer group sample AUC graph of the training set and the verification set of the lung cancer diagnosis model of the application in the most difficult to identify pathological non-cancer group and lung cancer group.

[0029] Fig. 6 is a lung cancer and non-cancer group sample risk score box plot of the training set and the verification set of the surgical method decision model of the application.

[0030] Figure 7 is a graph of the AUC of the training set and validation set of the surgical modality decision model of the present application. DETAILED DESCRIPTION

[0031] The technical solutions of the present application will be described clearly and completely below in combination with the drawings and embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0032] Example 1: A method for establishing a lung nodule benign and malignant differential diagnosis model of single cell immunomapping based on mass spectrometry flow cytometry technology

[0033] The specific steps are as follows: 1. Collect peripheral blood samples (5 ml per case) and send to the laboratory for further processing within 12 hours at room temperature or within 48 hours at 4°C. The collection subjects should meet the following requirements: 18 years old or older; clear diagnosis of disease (patients with lung nodules planned to undergo surgical resection and further confirmed by histopathology, or lung nodules with no changes after 3 years of follow-up, or lung nodules with a diameter of ≤4 mm); signed informed consent form. The exclusion criteria for collection include: history of cancer-related treatment; in acute infection period; received blood transfusion treatment within 6 months; used drugs affecting peripheral blood components within the past 2 weeks; local recurrent tumor; organ decompensated dysfunction; immunodeficiency syndrome; pre-cancerous blood disease; receiving immunosuppressive treatment; presence of coagulation dysfunction. A total of 1032 peripheral blood samples were collected. As shown in Figure 1, after the collected peripheral blood samples were transferred to the laboratory, they were processed as follows.

[0034] 2. Sample pretreatment, including: using Ficoll-Paque density gradient centrifugation method to separate peripheral blood mononuclear cells (PBMCs) from blood. They were suspended in 5 ml of pre-cooled FACS buffer (1x PBS + 0.5% BSA), then centrifuged at 400xg for 5 min at 4°C to discard the supernatant, and resuspend the cell pellet in buffer. Cell counting and quality assessment of PBMCs were performed before CyTOF detection to ensure that the count was greater than 3x106 and the survival rate was greater than 85%.

[0035] 3. CyTOF staining and data analysis, including: 40 metal-coupled antibodies were selected as markers to label cells; PBMCs were washed with PBS buffer, then stained with 0.5 mM cisplatin, and the binding of cells to antibodies was blocked by Fc receptors for 30 min, and the unbound antibodies were removed by centrifugation. Then the PBMCs were fixed in 200 μL of the insertion solution overnight. The cells were washed and resuspended in distilled water, added to 20% EQ beads, and further analyzed on a mass cytometer. The FCS file was normalized using the bead normalization method. Each sample data was debarcoded using a binary filter scheme with a unique mass marker barcode. FlowJo software was used to exclude debris, dead cells and double cells, leaving only live single immune cells.

[0036] 4. Feature screening: The samples were divided into training set and validation set in chronological order of enrollment, and the training set included 178 lung cancer samples and 218 non-cancer control samples. First, we evaluated the negative and positive expression of each cell marker in the training set. Then, based on the expression profile of these markers, we used random forest algorithm and 10-fold cross-validation to select feature cell subgroups. In each successful random forest model construction, the features with importance level exceeding 0.01 were recorded, and if a feature occurred more than 350 times in 1000 cross-validation processes, it would be counted. Finally, 19 feature cell subgroups were selected for model construction.

[0037] 5. Model establishment: Using the features selected by the above method, a lung cancer diagnosis model was established by random forest method using 178 lung cancer samples and 218 non-cancer control samples in the training set. The risk score of each participant was calculated by the model, which was the average probability of the positive probability of the sample judged by each decision tree in the random forest model, ranging from 0 to 1.

[0038] Example 2 Verification method of a lung nodule benign and malignant differential diagnosis model based on mass cytometry single cell immune atlas

[0039] 1. Using the scheme provided in Example 1, CyTOF staining and data analysis were performed on 251 non-cancer samples and 283 lung cancer samples in the validation set that were not trained.

[0040] 2. The new peripheral blood samples were input into the lung cancer diagnosis model constructed by the combination of 19 cell subgroups, and the blood samples were predicted and evaluated, and whether the sample was a lung cancer patient was determined according to the prediction result of the lung cancer diagnosis model.

[0041] Example 3 Establishment and verification of a lung cancer infiltration degree judgment model based on mass cytometry single cell immune atlas

[0042] 1. Using the protocol provided in Example 1, CyTOF staining and data analysis were performed using 113 lung nodule samples pathologically diagnosed as MIA and 105 lung nodule samples pathologically diagnosed as IA in the training set.

[0043] 2. Using the protocol provided in Example 1, a random forest algorithm and 10-fold cross-validation were used to select characteristic cell subpopulations. A total of 15 cell subpopulations were screened as modeling features to construct a new model for determining the invasive degree of malignant lung nodules.

[0044] 3. Using the protocol provided in Example 2, model verification was performed using 111 lung nodule samples pathologically diagnosed as MIA and 106 lung nodule samples pathologically diagnosed as IA in the validation set.

[0045] Obviously, the above embodiments of the present application are only examples for more clearly illustrating the present application, and are not intended to limit the embodiments of the present application. For those skilled in the art, other different forms of changes or variations can be made on the basis of the above description, and it is impossible to enumerate all the implementation methods here. Any obvious changes or variations derived from the technical solutions of the present application still fall within the protection scope of the present application.

[0046] In combination with FIGS. 2-3, it can be seen that the lung nodule benign and malignant differential diagnosis model disclosed in the present application has good detection efficiency in the differential diagnosis of lung cancer (pathologically confirmed CA) and non-cancer group samples (including image-confirmed Non-CA and pathologically confirmed Non-CA). The AUC of the training set and the validation set can reach 0.95 and 0.96, which is superior to the existing clinically used models: Mayo model (the AUC of the training set and the validation set is 0.75 and 0.70, respectively), Veterans Association (VA) model (the AUC of the training set and the validation set is 0.73 and 0.65, respectively), and BROCK model (the AUC of the training set and the validation set is 0.84 and 0.85, respectively).

[0047] In combination with FIGS. 4-5, it can be seen that the lung nodule benign and malignant differential diagnosis model disclosed in the present application has excellent performance in the most difficult-to-differentiate pathological non-cancer group and lung cancer group (the AUC of the training set and the validation set is 0.92 and 0.90, respectively), which is superior to the existing clinically used models: Mayo model, Veterans Association (VA) model, and BROCK model.

[0048] In combination with FIGS. 6-7, the model scheme established in the present scheme can effectively distinguish MIA and IA before surgery, and the AUC of the training set and the validation set is 0.97 and 0.93, respectively.

[0049] The various embodiments described in this specification are presented by way of example, and each embodiment is presented for the purpose of conveying the novelty and inventive aspects of the present application. Each embodiment is presented in a progressive and explanatory manner, and each embodiment highlights differences from other embodiments. The same or similar parts and / or functions between embodiments are to be understood as mutually interchangeable in any embodiment. The steps disclosed in the embodiments are described in a simple manner, as they correspond to the methods disclosed in the embodiments. The relevant parts are to be understood by referring to the method descriptions.

[0050] The previous description of the disclosed embodiments is not intended to limit the present application, but rather to provide an insight into the inventive concept. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Accordingly, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for constructing a single-cell immunogram-based lung nodule benign and malignant differential diagnosis model, characterized in that, Comprising the following steps: Peripheral blood mononuclear cells are obtained by treating peripheral blood samples with Ficoll separation method, and mass spectrometry flow cytometry analysis is performed on the peripheral blood mononuclear cells to obtain a mass spectrometry flow cytometry analysis data set; the cells are divided into different phenotypes according to marker expression by using a PARC clustering algorithm, and the expression proportion of each cell subgroup is used as a modeling feature to obtain a lung nodule benign and malignant differential diagnosis model.

2. The construction method of claim 1, wherein, Specifically, the following steps are implemented: (1) Peripheral blood mononuclear cells are obtained by treating peripheral blood samples with Ficoll separation method, and they are suspended in 5ml pre-cooled flow cytometry sorting buffer, then centrifuged at 400xg for 5min at 4℃ to discard the supernatant, and the cell pellet is resuspended in buffer solution, and cell counting and quality assessment of peripheral blood mononuclear cells are performed before CyTOF mass spectrometry flow cytometry detection to ensure that the count is greater than 3x106 and the survival rate is greater than 85%; (2) 40 metal-coupled antibodies are selected as markers for labeling cells; PBMCs are washed with PBS buffer, then stained with 0.5mM cisplatin, and the binding of cells to antibodies is blocked by Fc receptors for 30min, and then unbound antibodies are removed by centrifugation, and PBMCs are fixed in 200μL of insertion solution overnight, and cells are washed and resuspended in distilled water, added to 20% EQ magnetic beads, and further analyzed on a mass spectrometry flow cytometer. (3) The cells are divided into different phenotypes according to marker expression by using a PARC clustering algorithm, and the expression proportion of each cell subgroup is used as a modeling feature to screen features by using a random forest algorithm and 10-fold cross-validation; (4) A lung nodule benign and malignant differential diagnosis model is obtained by modeling by random forest method.

3. The construction method of claim 2, wherein, When the samples are selected for grouping in step (1), random grouping is used to cover samples of different nodule sizes, different types of nodules, and different types of adenocarcinoma infiltration.

4. The construction method according to claim 3, characterized in that, The different nodule sizes are ≤10mm, 11-20mm, and 21-30mm, the different types of nodules are solid nodules, partial solid nodules, and pure ground glass density nodules, and the different types of adenocarcinoma infiltration samples are atypical adenomatous hyperplasia AAH, adenocarcinoma in situ AIS, microinvasive adenocarcinoma MIA, and invasive adenocarcinoma IA.

5. The construction method of claim 2, wherein, The flow cytometry sorting buffer in step (1) is 1xPBS+0.5%BSA.

6. The method of construction of claim 2, wherein, The 40 metal-coupled antibody markers in step (2) include CD45, CD3, CD56, TCRγ / δ, CD196, CD14, IgD, CD123, CD85j, CD19, CD25, CD274, CD278, CD39, CD27, CD24, CD45RA, CD86, CD28, CD197, CD11c, CD33, CD152, CD161, CD185, CD66b, CD183, CD94, CD57, CD45RO, CD127, CD279(PD-1), CD38, CD194, CD20, CD16, HLA-DR, CD4, CD8a, CD11b.

7. The method of construction of claim 2, wherein, The step (2) adopts a combination of 40 metal conjugated antibody markers.

8. The construction method of claim 2, wherein, The step (3) screens out 34 immune cell subgroups and markers as modeling features, and 19 features of the lung nodule benign and malignant differential diagnosis model include: CD33-CD14-CD3+CD4+CD28+, CD33-CD14-CD3+CD4+CD274+, CD33-CD14-CD3+CD4+CD197-CD45RA-, CD33-CD14-CD3+CD4+HLA_DR+CD38+, CD33-CD14-CD3+CD4+CXCR5-CD183-CCR6-, CD33-CD14-CD3+CD4+CD25+CD127-CD161-CD45RA+, CD33-CD14-CD3+CD8+CD197-CD45RA+, CD33-CD14-CD3-CD19+CD24+CD38+, CD33-CD14-CD3-CD20-CD38+CD27+, CD33-CD14-CD3-CD56+CD16+CD94+, CD33-CD14-CD3-CD56+CD16+CD161+, CD3-CD19-CD56-CD14-CD123+CD11c+, CD33-CD14-CD3-CD56+CD16-, CD86, CD11c, CD183, CD94, CD4, CD11b; The 15 features of the lung cancer infiltration degree judgment model include: CD33-CD14-CD3+CD8+CD85j+, CD33-CD14-CD3+CD8+CD161-, CD33-CD14-CD3+CD4+, CD33-CD14-CD3+CD4+CD197-CD45RA+, CD33-CD14-CD3+CD4+HLA_DR+CD38+, CD33-CD14-CD3+CD4+HLA_DR+CD38-, CD33-CD14-CD3+CD8+, CD33-CD14-CD3+CXCR5+, CD33-CD14-CD3+CD8+CD197+CD45RA-, CD33-CD14-CD3-CD19+CD24+CD38+, CD33-CD14-CD3-CD56+CD16+, CD33-CD14-CD3-CD56+CD16+CD57+, CD33-CD14-CD3-CD56+CD16+HLA_DR+, CD3-CD19-CD56-CD14-HLA_DR-, CD56.

9. The construction method of claim 2, wherein, The step (3) adopts the training set and the verification set, and the samples are divided into the training set and the verification set according to the time sequence of entering the group. The step (3) adopts the training set and the verification set, and the samples are divided into the training set and the verification set according to the time sequence of entering the group.

Images