Small extracellular vesicle miRNA and application thereof in early discovery of lung cancer

Abstract

The invention discloses miRNAs (micro Ribonucleic Acid) of small extracellular vesicles of blood, wherein the miRNAs comprise hsa-let-7d-3p, hsa-let-7f-5p, hsa-let-7i-5p, hsa-miR-140-5p, hsa-miR-142-3p, hsa-miR-21-5p, hsa-miR-26a-5p and hsa-miR-99a-5p. The miRNAs can be used as biomarkers for early diagnosis of lung cancer, especially for distinguishing patients with early lung cancer and people with high risk of lung cancer. The specificity of the blood extracellular vesicle miRNA biomarker for distinguishing early-stage lung cancer patients and high-risk people is higher than that of existing clinical imaging and serology methods, excessive diagnosis and treatment can be effectively reduced, and therefore the blood extracellular vesicle miRNA biomarker has good clinical application value.

Description

Technical Field

[0001] The present invention belongs to the field of biomedicine. Specifically, the present invention relates to the diagnosis and differential diagnosis of lung cancer, and in particular to the application of small extracellular vesicle miRNA as a plasma marker in distinguishing high-risk groups and early lung cancer patients. Background Art

[0002] Primary lung cancer is the disease with the highest cancer-related mortality rate in the world, with an estimated 1.6 million deaths from lung cancer each year. It is pathologically divided into two major categories: non-small cell lung cancer (NSCLC, accounting for about 85%) and small cell lung cancer (SCLC, accounting for about 15%). NSCLC can be further divided into adenocarcinoma, squamous cell carcinoma and large cell carcinoma. Due to the limitations of early screening methods for lung cancer and the hidden nature of early symptoms, most patients have entered the late stage of the disease at the time of initial diagnosis, which is the main reason for the poor prognosis of lung cancer patients, and the five-year survival rate is only about 15%. Therefore, screening in people at high risk of lung cancer is beneficial to the detection of early lung cancer, improving prognosis and increasing the cure rate. Therefore, improving the effectiveness of early diagnosis and screening methods is an effective measure to reduce lung cancer mortality and improve the prognosis of lung cancer patients.

[0003] Commonly used techniques for early diagnosis of lung cancer include imaging and serological markers. Imaging techniques include: chest X-ray, chest CT and PET-CT, etc., but the X-ray density resolution is low; chest CT is the main means of diagnosis of lung cancer patients, but CT can only detect small nodules larger than 2 mm, and up to 4% of the normal population detected are early lung cancer patients, with low specificity; PET-CT is one of the best methods for diagnosis, surgical evaluation, efficacy and prognosis evaluation, but it is expensive, and it will produce false negative results when detecting low-grade tumors or false positive results due to inflammation and infection. Serum tumor markers mainly include carcinoembryonic antigen (CEA), neuron-specific enolase (NSE), cytokeratin 19 fragment (CYFRA21-1), progastrin-releasing peptide (ProGRP) and squamous cell carcinoma antigen (SCCAg), etc., but the sensitivity and specificity of these markers are not high, far from meeting the needs of clinical diagnosis. Therefore, it is of great clinical value to explore and develop efficient and accurate tumor markers to supplement imaging results to assist in the early diagnosis of lung cancer.

[0004] At present, there are reports on the use of exosomal miRNA as a potential biomarker for early diagnosis of lung cancer. Li Mengxing et al. used RT-PCR to find that compared with healthy people, miR-128-3p and miR-33a-5p in serum exosomes of NSCLC patients were significantly downregulated, and performed well in NSCLC screening, and have the potential to become a new biomarker for large-scale screening of NSCLC (PMID: 37324093); Zhang Yi et al. used RT-PCR to find that compared with the healthy control group, miR-17-5p was significantly upregulated in serum exosomes of NSCLC patients, and combined with CEA, CYFRA21-1 and SCCA to form a new diagnostic model with clinical value in the diagnosis of NSCLC (PMID: 31146974); Zhang Qian et al. used small RNA sequencing to find that compared with patients with benign lung diseases, the plasma exosomes of patients with early lung cancer had significantly upregulated miR-1290 and significantly downregulated miR-29c-3p, indicating that miR-1290 and miR-29c-3p have important potential as biomarkers in the early detection of lung cancer (PMID: 37916122). The above studies mainly focus on the differential diagnosis of lung cancer, benign lung diseases and healthy people, and there are no studies on plasma exosome miRNA as an efficient and accurate biomarker based on screening of high-risk populations. Summary of the invention

[0005] The purpose of the present invention is to use blood small extracellular vesicle miRNA as a biomarker, establish an efficient diagnostic model through statistical and machine learning methods for early diagnosis of clinical lung cancer, and use auxiliary imaging technology to give each patient a highly accurate prediction result of the risk of lung cancer, which will greatly avoid the mistake or excessive medication of patients caused by unclear diagnosis, and will greatly improve the treatment effect of lung cancer, improve patient prognosis and improve the quality of life of patients.

[0006] To this end, the present invention provides a biomarker that can be used clinically as a biomarker for early diagnosis of lung cancer, especially for distinguishing early lung cancer patients from high-risk groups.

[0007] The present invention also provides a chip and a kit that can be used for early diagnosis of lung cancer, especially for distinguishing early lung cancer patients from high-risk groups.

[0008] The present invention also provides a method for using the biomarker to perform early diagnosis of lung cancer, especially to distinguish early lung cancer patients from high-risk groups.

[0009] In a first aspect, the present invention provides use of blood extracellular vesicle miRNA in preparing a chip, a detection reagent or a detection kit for early diagnosis of lung cancer.

[0010] In a specific embodiment, the blood extracellular vesicle miRNA is:

[0011] (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or

[0012] (ii) a miRNA that is complementary to the miRNA sequence described in (i).

[0013] In a preferred embodiment, the blood extracellular vesicle miRNA is:

[0014] (i) a combination of a miRNA having a sequence as shown in SEQ ID NO:3, a miRNA having a sequence as shown in SEQ ID NO:4, a miRNA having a sequence as shown in SEQ ID NO:6, and a miRNA having a sequence as shown in SEQ ID NO:7, or,

[0015] A combination of a miRNA with a sequence as shown in SEQ ID NO:1, a miRNA with a sequence as shown in SEQ ID NO:2, a miRNA with a sequence as shown in SEQ ID NO:3, a miRNA with a sequence as shown in SEQ ID NO:4, a miRNA with a sequence as shown in SEQ ID NO:5, a miRNA with a sequence as shown in SEQ ID NO:6, a miRNA with a sequence as shown in SEQ ID NO:7, and a miRNA with a sequence as shown in SEQ ID NO:8.

[0016] In a preferred embodiment, the miRNA is isolated from human.

[0017] In a preferred embodiment, the early diagnosis of lung cancer is to distinguish between people at high risk of lung cancer and patients with early lung cancer.

[0018] In a second aspect, the present invention provides a miRNA chip, wherein the miRNA chip comprises:

[0019] a solid support; and

[0020] Oligonucleotide probes are sequentially fixed on the solid phase carrier, and the oligonucleotide probes specifically bind to miRNA;

[0021] Wherein, the miRNA is:

[0022] (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or

[0023] (ii) a miRNA that is complementary to the miRNA sequence described in (i).

[0024] In a preferred embodiment, the blood extracellular vesicle miRNA is:

[0025] (i) a combination of a miRNA having a sequence as shown in SEQ ID NO:3, a miRNA having a sequence as shown in SEQ ID NO:4, a miRNA having a sequence as shown in SEQ ID NO:6, and a miRNA having a sequence as shown in SEQ ID NO:7, or,

[0026] A combination of a miRNA with a sequence as shown in SEQ ID NO:1, a miRNA with a sequence as shown in SEQ ID NO:2, a miRNA with a sequence as shown in SEQ ID NO:3, a miRNA with a sequence as shown in SEQ ID NO:4, a miRNA with a sequence as shown in SEQ ID NO:5, a miRNA with a sequence as shown in SEQ ID NO:6, a miRNA with a sequence as shown in SEQ ID NO:7, and a miRNA with a sequence as shown in SEQ ID NO:8.

[0027] In a preferred embodiment, the oligonucleotide probe comprises:

[0028] complementary binding region; and / or

[0029] A linker region connected to a solid support.

[0030] In a preferred embodiment, the miRNA chip is used for early diagnosis of lung cancer.

[0031] In a preferred embodiment, the early diagnosis of lung cancer is to distinguish between people at high risk of lung cancer and patients with early lung cancer.

[0032] In a preferred embodiment, the miRNA is isolated from human.

[0033] In a preferred embodiment, the miRNA is blood extracellular vesicle miRNA.

[0034] In a third aspect, the present invention provides a use of the miRNA chip described in the second aspect for preparing an early diagnosis kit for lung cancer.

[0035] In a fourth aspect, the present invention provides a detection kit, wherein the detection kit contains a detection reagent for detecting miRNA;

[0036] Wherein, the miRNA is:

[0037] (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or

[0038] (ii) a miRNA complementary to the miRNA sequence described in (i)

[0039] Alternatively, the detection kit is equipped with the miRNA chip described in the second aspect.

[0040] In a preferred embodiment, the blood extracellular vesicle miRNA is:

[0041] (i) a combination of a miRNA having a sequence as shown in SEQ ID NO:3, a miRNA having a sequence as shown in SEQ ID NO:4, a miRNA having a sequence as shown in SEQ ID NO:6, and a miRNA having a sequence as shown in SEQ ID NO:7, or,

[0042] A combination of a miRNA with a sequence as shown in SEQ ID NO:1, a miRNA with a sequence as shown in SEQ ID NO:2, a miRNA with a sequence as shown in SEQ ID NO:3, a miRNA with a sequence as shown in SEQ ID NO:4, a miRNA with a sequence as shown in SEQ ID NO:5, a miRNA with a sequence as shown in SEQ ID NO:6, a miRNA with a sequence as shown in SEQ ID NO:7, and a miRNA with a sequence as shown in SEQ ID NO:8.

[0043] In a preferred embodiment, the detection kit is used for early diagnosis of lung cancer.

[0044] In a preferred embodiment, the early diagnosis of lung cancer is to distinguish between people at high risk of lung cancer and patients with early lung cancer.

[0045] In a preferred embodiment, the miRNA is isolated from human.

[0046] In a preferred embodiment, the miRNA is blood extracellular vesicle miRNA.

[0047] In a fifth aspect, the present invention provides a miRNA isolated from extracellular vesicles of blood for early diagnosis of lung cancer;

[0048] Wherein, the miRNA is:

[0049] (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or

[0050] (ii) a miRNA that is complementary to the miRNA sequence described in (i).

[0051] In a preferred embodiment, the blood extracellular vesicle miRNA is:

[0052] (i) a combination of a miRNA having a sequence as shown in SEQ ID NO:3, a miRNA having a sequence as shown in SEQ ID NO:4, a miRNA having a sequence as shown in SEQ ID NO:6, and a miRNA having a sequence as shown in SEQ ID NO:7, or,

[0053] A combination of a miRNA with a sequence as shown in SEQ ID NO:1, a miRNA with a sequence as shown in SEQ ID NO:2, a miRNA with a sequence as shown in SEQ ID NO:3, a miRNA with a sequence as shown in SEQ ID NO:4, a miRNA with a sequence as shown in SEQ ID NO:5, a miRNA with a sequence as shown in SEQ ID NO:6, a miRNA with a sequence as shown in SEQ ID NO:7, and a miRNA with a sequence as shown in SEQ ID NO:8.

[0054] In a preferred embodiment, the miRNA is isolated from human.

[0055] In a preferred embodiment, the miRNA is blood extracellular vesicle miRNA.

[0056] In a preferred embodiment, the early diagnosis of lung cancer is to distinguish between people at high risk of lung cancer and patients with early lung cancer.

[0057] In a sixth aspect, the present invention provides an isolated or artificially constructed precursor miRNA, which can be cleaved and expressed in human cells into the miRNA described in the fifth aspect.

[0058] In a seventh aspect, the present invention provides an isolated polynucleotide, which can be transcribed into a precursor miRNA by a human cell, and the precursor miRNA can be cleaved and expressed into the miRNA described in the fifth aspect in a human cell.

[0059] In a preferred embodiment, the polynucleotide has a structure shown in Formula I:

[0060] Seq 正向 -X-Seq 反向 Formula I

[0061] In Formula I,

[0062] Seq 正向 A nucleotide sequence that can be expressed as the miRNA in human cells;

[0063] Seq 反向 is a nucleotide sequence that is substantially complementary or completely complementary to Seq in the forward direction;

[0064] X is located in Seq 正向 and Seq 反向 and the spacer sequence is 正向 and Seq 反向 Not complementary;

[0065] After the structure shown in Formula I is transferred into human cells, a secondary structure shown in Formula II is formed:

[0066]

[0067] In Formula II, Seq 正向 , Seq 反向 and X is defined as above,

[0068] || means in Seq 正向 and Seq 反向 The base pairing relationship formed between them.

[0069] In an eighth aspect, the present invention provides a vector, wherein the vector contains the miRNA described in the fifth aspect or the polynucleotide described in the seventh aspect.

[0070] In a ninth aspect, the present invention provides a method for distinguishing a high-risk group of lung cancer from an early-stage lung cancer patient, comprising the steps of:

[0071] (a) detecting the expression level of miRNA in a sample of a test subject;

[0072] (b) inputting the miRNA expression value obtained in step (a) into the risk prediction model to obtain the probability that the subject is a high-risk population or an early lung cancer patient;

[0073] (c) based on the probability obtained in step (b), providing a prediction result that the subject suffers from lung cancer;

[0074] Wherein, the miRNA is:

[0075] (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or

[0076] (ii) a miRNA that is complementary to the miRNA sequence described in (i).

[0077] In a preferred embodiment, the sample is a peripheral blood sample.

[0078] In a preferred embodiment, the blood extracellular vesicle miRNA is:

[0079] (i) a combination of a miRNA having a sequence as shown in SEQ ID NO:3, a miRNA having a sequence as shown in SEQ ID NO:4, a miRNA having a sequence as shown in SEQ ID NO:6, and a miRNA having a sequence as shown in SEQ ID NO:7, or,

[0080] A combination of a miRNA with a sequence as shown in SEQ ID NO:1, a miRNA with a sequence as shown in SEQ ID NO:2, a miRNA with a sequence as shown in SEQ ID NO:3, a miRNA with a sequence as shown in SEQ ID NO:4, a miRNA with a sequence as shown in SEQ ID NO:5, a miRNA with a sequence as shown in SEQ ID NO:6, a miRNA with a sequence as shown in SEQ ID NO:7, and a miRNA with a sequence as shown in SEQ ID NO:8.

[0081] In a preferred embodiment, the miRNA is isolated from human.

[0082] In a preferred embodiment, the miRNA is blood extracellular vesicle miRNA.

[0083] In a preferred embodiment, the model formula of the risk prediction model in step (b) is as follows:

[0084]

[0085] Among them, x i represents the marker expression value of sample i, P(x i ) is the probability of the corresponding classification of sample i predicted by the diagnostic model, where 0 represents high-risk people and 1 represents lung cancer patients; the classification with the highest probability is taken as the final prediction result of sample i.

[0086] In a preferred embodiment, the risk prediction model is constructed using a machine learning algorithm; preferably, the machine learning algorithm is K nearest neighbor (KNN), random forest, extremely randomized trees, neural network, gradient boosting machine (GBM) and / or ensemble methods.

[0087] In a preferred embodiment, the method has a sensitivity > 70% and a specificity > 80%.

[0088] In a tenth aspect, the present invention provides a device for distinguishing between a high-risk group of lung cancer and patients with early-stage lung cancer, the device having a storage device storing instructions for executing the method for distinguishing between a high-risk group of lung cancer and patients with early-stage lung cancer described in the ninth aspect.

[0089] In a preferred embodiment, the device comprises the following means:

[0090] (a) a detection device, wherein the detection device detects the expression level of miRNA in a sample of a subject to be tested;

[0091] (b) a computing device, wherein the computing device inputs the expression value of the miRNA detected by the device (a) into a risk prediction model to obtain the probability that the subject is a high-risk population or an early-stage lung cancer patient; and

[0092] (c) an output device, which outputs a prediction result that the subject suffers from lung cancer based on the probability obtained by device (b);

[0093] Wherein, the miRNA is:

[0094] (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or

[0095] (ii) a miRNA that is complementary to the miRNA sequence described in (i).

[0096] In a preferred embodiment, the blood extracellular vesicle miRNA is:

[0097] (i) a combination of a miRNA having a sequence as shown in SEQ ID NO:3, a miRNA having a sequence as shown in SEQ ID NO:4, a miRNA having a sequence as shown in SEQ ID NO:6, and a miRNA having a sequence as shown in SEQ ID NO:7, or,

[0098] A combination of a miRNA with a sequence as shown in SEQ ID NO:1, a miRNA with a sequence as shown in SEQ ID NO:2, a miRNA with a sequence as shown in SEQ ID NO:3, a miRNA with a sequence as shown in SEQ ID NO:4, a miRNA with a sequence as shown in SEQ ID NO:5, a miRNA with a sequence as shown in SEQ ID NO:6, a miRNA with a sequence as shown in SEQ ID NO:7, and a miRNA with a sequence as shown in SEQ ID NO:8.

[0099] In a preferred embodiment, the miRNA is isolated from human.

[0100] In a preferred embodiment, the miRNA is blood extracellular vesicle miRNA.

[0101] In a preferred embodiment, the model formula of the risk prediction model described in device (b) is as follows:

[0102]

[0103] Among them, x i represents the marker expression value of sample i, P(x i ) is the probability of the corresponding classification of sample i predicted by the diagnostic model, where 0 represents high-risk people and 1 represents lung cancer patients; the classification with the highest probability is taken as the final prediction result of sample i.

[0104] In a preferred embodiment, the risk prediction model is constructed using a machine learning algorithm; preferably, the machine learning algorithm is K nearest neighbor (KNN), random forest, extremely randomized trees, neural network, gradient boosting machine (GBM) and / or ensemble methods.

[0105] It should be understood that within the scope of the present invention, the above-mentioned technical features of the present invention and the technical features specifically described below (such as embodiments) can be combined with each other to form a new or preferred technical solution. Due to space limitations, they will not be described one by one here. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Figure 1 The transmission electron microscopic morphology of small extracellular vesicles is shown;

[0107] Figure 2 NTA size distribution of small extracellular vesicles is shown;

[0108] Figure 3 The ROC curve for the training cohort is shown;

[0109] Figure 4 ROC curves for the validation cohort are shown. DETAILED DESCRIPTION

[0110] Lung cancer is the second most common cancer in humans worldwide and has the highest mortality rate among all cancers. With the advancement of technology in recent years, the survival rate of early lung cancer has been effectively improved. However, due to the limitations of lung cancer screening methods and the lack of specific clinical symptoms, most patients are already in the advanced stage of the disease when diagnosed. Therefore, early detection of lung cancer through screening is an effective means to improve the long-term prognosis of patients. At present, the early clinical diagnosis of lung cancer still relies on imaging technology, which has the problems of radiation damage and high false positives. Misjudgment of benign and malignant nodules increases the rate of patients receiving unnecessary surgical treatment. Therefore, if highly sensitive and highly specific biomarkers can be combined as auxiliary diagnostic technology to improve the accuracy of early diagnosis of lung cancer, the prognosis of lung cancer patients will be significantly improved.

[0111] Based on this clinical problem, the inventors conducted the following research work:

[0112] (1) Develop a highly accurate diagnostic model for the early diagnosis of lung cancer by using statistical and machine learning methods to identify blood small extracellular vesicle miRNAs;

[0113] (2) The effectiveness of the constructed diagnostic model was evaluated in the training set and validation set based on the cross-validation method and self-sampling method and using multiple evaluation indicators (accuracy, area under the receiver operating characteristic curve, positive predictive value, negative predictive value, sensitivity and specificity).

[0114] By applying the discovered blood small extracellular vesicle miRNA biomarkers and the constructed diagnostic model to the early diagnosis of clinical lung cancer through (1) and (2), providing each patient with a predicted result of the risk of lung cancer will greatly improve the treatment effect and the quality of life of the patient.

[0115] In the present invention, a content in the extracellular vesicles of the blood, i.e., miRNA, is used to establish a biomarker model to improve the specificity of early diagnosis of lung cancer. However, those skilled in the art know that there are many contents of extracellular vesicles, such as proteins, lipids, small molecule metabolites, lncRNA, circRNA, mRNA, piRNA, etc., and these extracellular vesicle contents of the blood may also construct a biomarker model to achieve an effect similar to miRNA. In addition, various components in the patient's blood, such as CTCs (Circulating tumor cells), ctDNA (Circulating tumor DNA), extracellular free nucleic acids (cfRNA and cfDNA), proteins and metabolites, etc., may be used as biomarkers for early diagnosis of lung cancer, for development into independent liquid biopsy technology.

[0116] definition

[0117] The scientific and technical terms used herein are consistent with the conventional understanding of those skilled in the art. To facilitate understanding of the present invention, the relevant terms are explained and defined as follows:

[0118] Small extracellular vesicles

[0119] The terms "small extracellular vesicles", "small extracellular vesicles", "blood small extracellular vesicles" or "plasma small extracellular vesicles" used in this article have the same meaning, referring to a type of double-layer membrane structured vesicles with a diameter of about 30-150nm, also known as exosomes, which are important members of the extracellular vesicle family. Given that exosomes are rich in various biomacromolecules such as nucleic acids, proteins and complex metabolites, they transfer physiological signals and genetic materials between tumor and matrix cells, and have been valued and widely studied by researchers as an important liquid biopsy material.

[0120] People at high risk of lung cancer

[0121] The terms "high-risk population" or "high-risk population for lung cancer" used in this article have the same meaning, referring to a population with a relatively high incidence of lung cancer or a relatively high risk of developing lung cancer.

[0122] Generally, people at high risk of lung cancer are between (50 and 74) years old and meet at least one of the following conditions: a) have smoked for at least 30 pack-years, including former smokers for at least 30 pack-years but have quit smoking for less than 15 years; b) have lived with or worked in the same room with a person who has been exposed to passive smoking for more than 20 years; c) have chronic obstructive pulmonary disease; d) have a history of occupational exposure for at least one year, including exposure to asbestos, radon, beryllium, chromium, cadmium, silicon, coal smoke and soot; e) have a first-degree relative diagnosed with lung cancer.

[0123] Note 1: Number of pack-years of smoking = number of packs smoked per day (20 cigarettes per pack) × number of years of smoking

[0124] Note 2: First-degree relatives refer to parents, children and siblings (with the same parents).

[0125] Blood extracellular vesicle miRNA of the present invention

[0126] Currently, imaging techniques with low sensitivity and specificity are widely used in the clinic to diagnose lung cancer patients. Even when combined with existing serum tumor markers, the accuracy of diagnosis is limited and far from meeting clinical needs. Therefore, there is an urgent need to develop new biomarkers to assist lung cancer imaging diagnosis results to improve the accuracy of early diagnosis of lung cancer patients, so that lung cancer patients can receive treatment earlier, improve their quality of life, and prolong their survival.

[0127] At present, some researchers have found that serum exosomal miR-21-5p, miR-126-3p and miR-140-5p are significantly upregulated in patients with lung adenocarcinoma, indicating that serum exosomal miR-21-5p, miR-126-3p and miR-140-5p have the potential to be used as early diagnostic biomarkers for lung adenocarcinoma. However, this study used healthy people as the control group, and the diagnostic performance of its public biomarkers was insufficient. In addition, some researchers detected the differential expression of plasma exosomal miRNAs in patients with benign and malignant lung nodules, and established a diagnostic model including miR-30c-5p, miR-30e-5p, miR-500a-3p, miR-125a-5p and miR-99a-5p to distinguish between benign and malignant lung nodules. At the same time, this diagnostic model is closely related to the survival of lung adenocarcinoma. However, the purpose of this study is mainly to distinguish between benign and malignant pulmonary nodules, and cannot be well applied to identify high-risk groups and early lung cancer patients.

[0128] In view of the shortcomings in the art, the inventors used the blood of patients clinically diagnosed with lung cancer as research samples, and used high-risk people as controls, and used the extracellular vesicle extraction reagent L3525 independently developed by Shanghai Siludi Biomedical Technology Co., Ltd. to separate plasma exosomes and extract RNA, and used small RNA sequencing technology to detect the differential expression of plasma small extracellular vesicle miRNA between high-risk people and lung cancer patients, and established and verified the diagnostic model in two independent cohorts, thereby discovering the best small extracellular vesicle miRNA biomarker for early diagnosis of lung cancer. Specifically, the present invention realizes the use of blood small extracellular vesicle miRNA marker combination to identify early lung cancer patients and high-risk people for lung cancer.

[0129] In the invention, small RNA sequencing technology was used to discover 8 plasma small extracellular vesicle miRNAs that were differentially expressed between early lung cancer patients and people at high risk of lung cancer: hsa-let-7d-3p, hsa-let-7f-5p, hsa-let-7i-5p, hsa-miR-140-5p, hsa-miR-142-3p, hsa-miR-21-5p, hsa-miR-26a-5p and hsa-miR-99a-5p. A high-specificity and high-sensitivity lung cancer risk prediction model for high-risk groups was established based on statistical methods and machine learning, and the diagnostic model was confirmed to have good diagnostic efficacy in both the training set and the validation set.

[0130] In a specific embodiment, the blood extracellular vesicle miRNA of the present invention is a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or a miRNA complementary to the above miRNA sequences.

[0131] In a preferred embodiment, the blood extracellular vesicle miRNA is a combination of a miRNA having a sequence as shown in SEQ ID NO: 1, a miRNA having a sequence as shown in SEQ ID NO: 2, a miRNA having a sequence as shown in SEQ ID NO: 3, a miRNA having a sequence as shown in SEQ ID NO: 4, a miRNA having a sequence as shown in SEQ ID NO: 5, a miRNA having a sequence as shown in SEQ ID NO: 6, a miRNA having a sequence as shown in SEQ ID NO: 7, and a miRNA having a sequence as shown in SEQ ID NO: 8.

[0132] Based on the teachings of the present invention, those skilled in the art can understand that the blood extracellular vesicle miRNA of the present invention can be used for the early diagnosis of lung cancer (especially distinguishing between people at high risk of lung cancer and patients with early lung cancer), and can then be used to prepare miRNA chips, detection reagents or detection kits for early diagnosis of lung cancer (especially distinguishing between people at high risk of lung cancer and patients with early lung cancer).

[0133] In a specific embodiment, the miRNA chip of the present invention comprises: a solid phase carrier; and oligonucleotide probes fixed in order on the solid phase carrier, wherein the oligonucleotide probes specifically bind to the blood extracellular vesicle miRNA of the present invention. Based on conventional technical means in the art, those skilled in the art know how to design the oligonucleotide probes. For example, the oligonucleotide probes contain: a complementary binding region; and / or a connection region connected to the solid phase carrier.

[0134] In a specific embodiment, the detection kit contains a detection reagent for detecting the miRNA of the present invention; or, the miRNA chip.

[0135] Based on the teachings of the present invention and the common knowledge in the field of miRNA, a person skilled in the art can think of an isolated or artificially constructed precursor miRNA, which can be cleaved and expressed in human cells as the miRNA of the present invention. Further, a person skilled in the art can also think of an isolated polynucleotide, which can be transcribed into the precursor miRNA by human cells, which can be cleaved and expressed in human cells as the miRNA of the present invention.

[0136] In a specific embodiment, the polynucleotide has a structure shown in Formula I:

[0137] Seq 正向 -X-Seq 反向 Formula I

[0138] In Formula I, Seq 正向 is a nucleotide sequence that can be expressed as the miRNA in human cells; Seq 反向 is a nucleotide sequence substantially complementary or completely complementary to Seq; X is a nucleotide sequence located at Seq 正向 and Seq 反向 and the spacer sequence is 正向 and Seq 反向 Not complementary;

[0139] After the structure shown in Formula I is transferred into human cells, a secondary structure shown in Formula II is formed:

[0140]

[0141] In Formula II, Seq 正向 , Seq 反向 and X are defined as above, || means in Seq 正向 and Seq 反向 The base pairing relationship formed between them.

[0142] Based on the teachings of the present invention and the common knowledge in the field of miRNA, those skilled in the art can also conceive of a vector containing the miRNA or polynucleotide of the present invention.

[0143] Method of the present invention and device for implementing the method of the present invention

[0144] The blood small extracellular vesicle miRNA marker combination of the present invention can be used to identify early lung cancer patients and people at high risk of lung cancer. Therefore, the present invention also provides a method for distinguishing high-risk lung cancer people and early lung cancer patients using the blood small extracellular vesicle miRNA of the present invention, comprising the steps of:

[0145] (a) detecting the expression level of miRNA in a sample of a test subject;

[0146] (b) inputting the miRNA expression value obtained in step (a) into the risk prediction model to obtain the probability that the subject is a high-risk population or an early lung cancer patient;

[0147] (c) Based on the probability obtained in step (b), a prediction result is given that the subject suffers from lung cancer.

[0148] The sensitivity of the method of the present invention is >70% and the specificity is >80%.

[0149] On the basis of understanding the technical content of the present invention, that is, specific blood extracellular vesicle miRNA can be used to identify early lung cancer patients and high-risk groups for lung cancer, those skilled in the art can use various methods, such as various algorithms to construct a risk prediction model. In a specific embodiment, the risk prediction model is constructed using a machine learning algorithm; preferably, the machine learning algorithm is K-nearest neighbor (KNN), random forest, extremely randomized trees, neural network, gradient boosting machine (GBM) and / or ensemble methods.

[0150] In a specific implementation, the model formula of the risk prediction model is as follows:

[0151]

[0152] Among them, x i represents the marker expression value of sample i, P(x i ) is the probability of the corresponding classification of sample i predicted by the diagnostic model, where 0 represents high-risk people and 1 represents lung cancer patients; the classification with the highest probability is taken as the final prediction result of sample i.

[0153] In order to implement the method of the present invention for distinguishing between people at high risk of lung cancer and patients with early lung cancer, the present invention also provides a device for early diagnosis of lung cancer, wherein the device has a storage device storing instructions for executing the method of the present invention for distinguishing between people at high risk of lung cancer and patients with early lung cancer.

[0154] In a specific embodiment, the device includes a detection device for detecting the expression level of miRNA in a sample of a subject to be tested; a calculation device for inputting the expression value of the detected miRNA into a risk prediction model to obtain the probability that the subject is a high-risk population or an early lung cancer patient; and an output device for giving a prediction result that the subject suffers from lung cancer based on the obtained probability.

[0155] Advantages of the present invention:

[0156] 1. The present invention is a biomarker that can be used clinically to distinguish early-stage lung cancer patients from high-risk groups. It has higher specificity than existing clinical imaging and serological methods and can effectively reduce overdiagnosis and treatment.

[0157] 2. The present invention is a non-invasive diagnostic technology, which is easy to operate, non-invasive to patients, and can be tested repeatedly. Compared with the risks of radiation and fiberoptic bronchoscopy in LDCT, the present invention has the advantages of being convenient, non-invasive, and painless.

[0158] The present invention will be further described below in conjunction with specific examples. It should be understood that these examples are intended to illustrate the present invention only and are not intended to limit the scope of the present invention. The experimental methods in the following examples where specific conditions are not specified are usually performed under conventional conditions, such as those described in Sambrook et al., Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Laboratory Press, 2001), or under conditions recommended by the manufacturer. Unless otherwise stated, percentages and parts are calculated by weight.

[0159] Example

[0160] Material

[0161] The materials used in the following examples are all commercially available.

[0162] method

[0163] 1. Study cohort and clinical information

[0164] This study included 71 patients clinically diagnosed with early lung cancer (stage I-II) and 77 high-risk people in the study cohort. Blood samples were collected from lung cancer patients before surgery. After the above 148 samples were enrolled, 30% of the samples were randomly selected from the population as the validation cohort, and the rest were used as the training cohort.

[0165] 2. Extraction and Characterization of Small Extracellular Vesicles from Patient Blood

[0166] 1) Blood collection and plasma separation

[0167] Blood samples from lung cancer patients and high-risk groups were collected and stored in 10 ml vacuum blood collection tubes (REF367525, BD, USA). After gentle inversion and mixing, plasma was separated by a two-step centrifugation method: first, the blood collection tube was centrifuged at 1600 g and 4 °C for 10 min, and the plasma hemolysis level was determined and recorded. If the level was less than 4, it was used for subsequent research; the plasma supernatant was taken and continued to be centrifuged at 16000 g and 4 °C for 15 min. The supernatant was aspirated into a new 1.5 ml centrifuge tube and stored in a -80 °C refrigerator for subsequent research.

[0168] 2) Extraction of small extracellular vesicles

[0169] The self-developed exosome separation reagent L3525 was used to extract plasma exosomes from lung cancer patients and high-risk groups: the plasma samples frozen in a -80°C refrigerator were taken out and placed in a 37°C water bath until thawing. First, they were centrifuged at 12000g and 4°C for 10 min, then the plasma supernatant was carefully aspirated and added to a 0.45μm filter column (Costar, CLS8163-100EA, Corning, USA), and centrifuged under the same centrifugation conditions for 5 min to collect the plasma filtrate. Repeat the above steps to pass through a 0.22 μm filter column (Costar, CLS8161-100EA, USA) and transfer to a new 1.5 ml centrifuge tube. Measure and record the plasma volume. Add L-type exosome precipitant (L3525, 3DMed, Shanghai, China) according to 1 / 4 volume of plasma. Vortex mix and incubate in a 4°C refrigerator for more than 30 min. Subsequently, centrifuge at 4700g and 4°C for 30 min and discard the supernatant. Add 200 μl PBS (phosphate buffer saline) to resuspend the small extracellular vesicles precipitated at the bottom of the tube.

[0170] The small extracellular vesicle precipitation agent used in this aspect is L3525 independently developed by Shanghai Siludi Biomedical Technology Co., Ltd. Based on the teachings of the present invention and conventional technical means in the prior art, those skilled in the art know that precipitation reagents from other manufacturers can also be used. For example: Invitrogen TM Total Exosome Isolation Kits (Invitrogen, NYC, USA), etc.

[0171] Those skilled in the art may also use other separation methods to extract extracellular vesicles, such as centrifugation, filtration, immunomagnetic beads, etc.

[0172] 3) Characterization of plasma small extracellular vesicles

[0173] Transmission electron microscopy (TEM) was used to detect the morphological structure of small extracellular vesicles resuspended in PBS. The plasma small extracellular vesicles were fixed with freshly prepared 4% paraformaldehyde, then placed on the surface of the carbon-coated copper grid of the electron microscope and washed twice with PBS, followed by washing the copper grid with PBS containing glycine (50mM) for 3min, then washed with PBS containing 0.5% BSA for 10min, and stained with 2% uranyl acetate. Finally, the morphological structure of the stained small extracellular vesicles was detected using a transmission electron microscope (H-7650, Hitachi High-Technologies, Japan).

[0174] The particle size distribution of plasma extracellular vesicles was detected by nanoparticle tracking analysis (NTA): the resuspended plasma extracellular vesicles were diluted with PBS (1*10^ 7 -1*10^ 9 / ml) and pipetted to mix. The above samples were added to the sample chamber of the NTA instrument (NanoSight NS300, Malvern, UK), and the 488nm excitation module was used. The camera lens parameters (shutter value was 890, gain value was 146, and detection threshold was 7) were set and then detected. At least 200 complete tracks were analyzed and obtained for each video, and the nanoparticle tracking data of plasma small extracellular vesicles were analyzed by NTA software (version 2.3).

[0175] The present invention uses TEM and NTA to identify the morphological characteristics of exosomes. Based on the teachings of the present invention and conventional technical means in the prior art, those skilled in the art know other alternative methods, such as NanoView, DLS (dynamic light scattering), etc.

[0176] 3. Extraction and expression detection of plasma small extracellular vesicle miRNA

[0177] 1) Extraction of plasma small extracellular vesicle miRNA

[0178] The miRNeasy Serum / Plasma Kit (217184, QIAGEN, Shanghai, China) was used to extract plasma extracellular vesicle miRNA from lung cancer patients and high-risk groups, and the 2100 analyzer and supporting chips and reagents (5067-1548, Agilent, USA) were used to detect the distribution and total amount of extracellular vesicle miRNA fragments. For detailed experimental operations, please refer to the corresponding manufacturer's user manual.

[0179] 2) Detection of plasma small extracellular vesicle miRNA expression

[0180] Small RNA sequencing technology was used to detect the expression level of plasma exosome miRNA in patients with early lung cancer and high-risk populations: the plasma small extracellular vesicle miRNA library was prepared using the NEBNext, Multiplex Small RNA Library Prep Set for Illumina (E7300L, NEB, USA) kit, and then purified using the NucleoSpin Gel and PCR Clean-up (740609.50, MN, Germany) kit. The detailed steps of library construction and purification are shown in the product manual, which is briefly described as 3' adapter ligation, hybridization reverse transcription primer, 5' adapter ligation, reverse transcription, PCR amplification and library purification for a total of 100 ng of miRNA in a total volume of no more than 6 μl. Then, the library was tested for concentration using the Qubit 4.0 fluorescence quantifier and the supporting reagent Qubit dsDNA HS Assay Kit (Q32854, Thermofisher, USA), and the peak shape was tested using the 2100 analyzer and the supporting chip and reagent High Sensitivity DNA Kit & Reagents (5067-4626, Agilent, USA). Finally, the Illumina NovaSeq sequencing platform was used for sequencing (sequencing strategy was PE150).

[0181] In the present invention, small RNA sequencing technology is used to detect the expression of small extracellular vesicle miRNA. Based on the teachings of the present invention and conventional technical means in the prior art, those skilled in the art know other alternative methods, such as using miRNA chips, using third-generation sequencing technology and RT-PCR.

[0182] 4. Sequencing data analysis process

[0183] Based on small RNA second-generation sequencing technology, the expression levels of plasma small extracellular vesicle miRNAs in patients with early lung cancer and high-risk populations were obtained. The sequencing data analysis process is as follows:

[0184] 1) Sequencing data alignment. After removing the sequencing adapters of the Small RNA-seq data, the sequencing data were aligned to the human reference genome hg19 (genome download link: http: / / hgdownload.soe.ucsc.edu / goldenPath / hg19 / bigZips / ) using BWA software (version: 0.7.12-r1039), and the number of reads aligned to miRNA was counted.

[0185] 2) miRNA annotation. The miRNAs were annotated using the Gencode v25 and miRBase v21 databases, and those annotated as known mature miRNAs were retained for subsequent analysis.

[0186] 3) miRNA filtering. For the training cohort, mature miRNAs with a length of less than or equal to 30 nt and covered by at least g samples in the training cohort data with no less than 10 reads were retained for subsequent analysis, where g is the minimum number of samples in the group after the cohort was grouped; for the validation cohort, miRNAs screened by the training cohort were retained for subsequent analysis.

[0187] 4) Normalization of miRNA expression. The original expression of miRNA in the training cohort samples was normalized using the trimmed mean of M-values ​​(TMM) method, and the same parameters were used to process the validation cohort samples.

[0188] 5. Discovery of biomarkers

[0189] Based on the expression of miRNAs in the training cohort, the samples were grouped according to the pathological test results, and statistical and machine learning methods were used to discover miRNAs that can be used to distinguish high-risk groups and lung cancer patients as biomarkers. The process is as follows:

[0190] 1) Grouping of training cohort: Based on the results of sample pathology tests, the patient samples in the training cohort were divided into two groups: a high-risk group and a lung cancer group.

[0191] 2) Statistical methods for screening biomarkers. First, the statistical methods U test and T test were used to analyze the expression differences of all miRNAs between the two groups, and the mutual information (MI) between all miRNA expressions and sample groups was calculated. The miRNAs with inter-group statistical difference p value <= 0.1 and mutual information scores in the top 10% of all markers were screened. Then, the correlation coefficients between these miRNAs were calculated, and only the miRNAs with the largest variance were retained as candidate biomarkers among the miRNAs with correlation coefficients > 0.8 for subsequent analysis.

[0192] 3) Screening biomarkers using machine learning methods. The miRNAs screened in 2) were used as initial markers and first screened using three machine learning feature screening methods. Method 1: Evaluate the classification performance of each marker on the sample separately, and mark the markers with a performance score <0.8 as candidate discarded markers; Method 2: First, take all the initial markers as a whole to evaluate the classification performance of the sample, then randomly shuffle the expression of the markers in the sample, calculate the degree of reduction in the classification performance score after the expression of each marker was shuffled, and mark the markers with a performance score reduction of <0.01 as candidate discarded markers; Method 3: First, take all the initial markers as a whole to evaluate the classification performance of the sample, and sort the markers by marker importance, then add one marker at a time starting from the marker with the highest importance and evaluate the improvement of the model score, and mark the markers with a score improvement of <0.01 as candidate discarded markers. The intersection markers of the candidate discarded markers obtained by the above three methods were discarded to obtain 8 markers. Subsequently, six machine learning algorithms, including K-nearest neighbor (KNN), random forest, extremely randomized trees, neural network, gradient boosting machine (GBM) and ensemble methods, were used to train and evaluate the classification effect and feature weight of the eight markers as a whole on the training set data, and markers with feature weights <5% were marked as candidate discarded markers. If a marker was marked as discarded by all algorithms, it was not retained. Combining the above methods, all eight markers were used as candidate markers for subsequent analysis.

[0193] 4) Evaluation of candidate marker combinations. The eight candidate markers were trained using six machine learning algorithms, including K-nearest neighbor (KNN), random forest, extremely randomized trees, neural network, gradient boosting machine (GBM), and ensemble methods. The classification effect and feature weight of all marker combinations were evaluated using a 5-fold cross-validation method on the training set data. Based on the training results, markers with low feature weights and no improvement in model classification performance were removed.

[0194] 6. Construction of risk scoring model for high-risk population and early-stage lung cancer patients

[0195] Taking high-risk population and early lung cancer patients as the classification prediction targets, the eight markers discovered in (5) were used, and six machine learning algorithms were adopted, including K-nearest neighbor (KNN), random forest, extremely randomized trees, neural network, gradient boosting machine (GBM) and ensemble methods. Different hyper-parameters were preset for each algorithm, and the diagnostic models of multiple different machine learning algorithms were trained using the training set sample data. The model formula is as follows:

[0196]

[0197] x i represents the marker expression value of sample i, P(x i ) is the probability of the corresponding classification of sample i predicted by the diagnostic model, where 0 represents high-risk people and 1 represents lung cancer patients. The classification with the highest probability is taken as the final prediction result of sample i. The trained model is saved in the form of a file on the hard disk. When calling the model, the sample marker expression value is input to obtain the model prediction result.

[0198] In the present invention, the model building methods used are extreme randomized trees, K-nearest neighbors (KNN), gradient boosting machine (GBM), neural network, random forest and ensemble methods. Based on the teachings of the present invention and conventional technical means in the prior art, those skilled in the art know other alternative methods, such as complex feature engineering, support vectors and deep learning.

[0199] 7. Evaluation and Validation of Risk Scoring Model Performance

[0200] In the training set, the 5-fold cross-validation method was used to evaluate the classification effect and feature weight of the model. In the validation set, the self-sampling method (bootstrap, 1000 times) was used to evaluate the classification effect and feature weight of the model. The model evaluation and validation indicators included accuracy (range 0-1), area under the receiver operating characteristic curve (AUC, range 0-1), positive predictive value (PPV, Positive Predictive Value, range 0-1), negative predictive value (NPV, Negative Predictive Value, range 0-1), sensitivity (range 0-1) and specificity (range 0-1). The higher the value, the better the model classification performance.

[0201] 8. Application of risk scoring model in high-risk population and lung cancer patients

[0202] 1) Collect peripheral blood from patients suspected of lung cancer in clinical diagnosis, obtain peripheral blood small extracellular vesicles, and detect the expression of biomarkers using small RNA sequencing;

[0203] 2) Using the risk scoring model, the expression values ​​of blood extracellular vesicles hsa-let-7d-3p, hsa-let-7f-5p, hsa-let-7i-5p, hsa-miR-140-5p, hsa-miR-142-3p, hsa-miR-21-5p, hsa-miR-26a-5p and hsa-miR-99a-5p were obtained, and the values ​​were put into the trained model formula to calculate the probability of each patient being predicted as a high-risk population or lung cancer patient;

[0204] 3) Select the output result with the highest probability and give the predicted result of lung cancer risk for each patient.

[0205] Example 1.

[0206] In this embodiment, a total of 71 patients clinically diagnosed with lung cancer and 77 high-risk people were included in the study cohort, and blood samples were collected from lung cancer patients before surgery. After the above 148 samples were enrolled, 30% of the samples were randomly selected from the population as the validation cohort, and the rest were used as the training cohort. The training cohort consisted of 103 cases, including 54 high-risk samples and 49 lung cancer samples. The validation cohort consisted of 45 cases, including 23 high-risk samples and 22 lung cancer samples. Table 1 shows the clinical information of the selected samples, showing that there is no significant difference in the proportion of samples in different classifications.

[0207] Table 1. Clinical information of high-risk groups for lung cancer and patients with lung cancer.

[0208]

[0209] Example 2. Characterization of plasma small extracellular vesicles

[0210] In this example, transmission electron microscopy was used to detect the morphology of small extracellular vesicles. The results showed that the extracted exosomes showed a typical "horseshoe-shaped" morphology (see Figure 1 ). At the same time, the particle size distribution of small extracellular vesicles was detected by NTA, and the average particle size was 108nm, indicating that the average particle size of plasma small extracellular vesicles is consistent with the particle size range of exosomes (see Figure 2 ).

[0211] Example 3. Biomarker Discovery

[0212] In this embodiment, small RNA sequencing technology and statistical methods U test and T test were used to analyze the expression differences of all miRNAs in lung cancer patients and high-risk groups and the correlation coefficients between miRNAs, and three machine learning feature screening methods were used to screen miRNAs that can be used to distinguish lung cancer patients from high-risk groups of lung cancer as candidate biomarker molecular markers, including hsa-let-7d-3p (SEQ ID NO: 1, CUAUACGACCUGCUGCCUUUCU), hsa-let-7f-5p (SEQ ID NO: 2, UGAGGUAGUAGAUUGUAUAGUU), hsa-let-7i-5p (SEQ ID NO: 3, UGAGGUAGUAGUUUGUGCUGUU), hsa-miR-140-5p (SEQ ID NO: 4, CAGUGGUUUUACCCUAUGGUAG), hsa-miR-142-3p (SEQ ID NO: 5, CAGUGGUUUUACCCUAUGGUAG), hsa-miR-144-3p (SEQ ID NO: 6, CAGUGGUUUUACCCUAUGGUAG), hsa-miR-145-3p (SEQ ID NO: 7, CAGUGGUUUUACCCUAUGGUAG), hsa-miR-147-3p (SEQ ID NO: 8, CAGUGGUUUUACCCUAUGGUAG), hsa-miR-148-3p (SEQ ID NO: 9, CAGUGGUUUUACCCUAUGGUAG), hsa-miR-149-5p (SEQ ID NO: 5, UGUAGUGUUUCCUACUUUAUGGA), hsa-miR-21-5p (SEQ ID NO: 6, UAGCUUAUCAGACUGAUGUUGA), hsa-miR-26a-5p (SEQ ID NO: 7, UUCAAGUAAUCCAGGAUAGGCU) and hsa-miR-99a-5p (SEQ ID NO: 8, AACCCGUAGAUCCGAUCUUGUG), a total of 8, were used for subsequent analysis. Subsequently, six machine learning algorithms, including K nearest neighbor (KNN), random forest, extremely randomized trees, neural network, gradient boosting machine (GBM) and ensemble methods, were used to mark markers with feature weights < 5% as candidate discarded markers. If a marker was marked as discarded by all algorithms, it was not retained. All eight markers, including hsa-let-7d-3p, hsa-let-7f-5p, hsa-let-7i-5p, hsa-miR-140-5p, hsa-miR-142-3p, hsa-miR-21-5p, hsa-miR-26a-5p and hsa-miR-99a-5p, were used as candidate markers for the subsequent construction of risk scoring models for lung cancer patients and people at high risk of lung cancer.

[0213] Example 4. Risk scoring model for lung cancer patients and high-risk population

[0214] In order to construct a classification model for lung cancer patients and high-risk populations, the classification effect and feature weight of the model were evaluated in the training set using the 5-fold cross-validation method based on the miRNA expression data of the training cohort. Combined with the results of pathological detection, a risk scoring model for lung cancer patients and high-risk populations was constructed. The model evaluation indicators include accuracy (range 0-1), area under the receiver operating characteristic curve (AUC, range 0-1), positive predictive value (PPV, Positive Predictive Value, range 0-1), negative predictive value (NPV, Negative Predictive Value, range 0-1), sensitivity (range 0-1) and specificity (range 0-1). The higher the value, the better the effect, which are 0.7873, 0.8763, and 0.8189 ( Figure 3 ), 0.82, 0.76 and 0.8145 (Table 2). The results showed that in the training cohort, this risk prediction model had high accuracy, AUC, sensitivity and specificity, and the model prediction efficiency was better.

[0215] Table 2. Cross-validation evaluation results of biomarkers in the training set

[0216]

[0217]

[0218] Example 5. Validation of the predictive efficacy of the risk scoring model for lung cancer patients and high-risk populations for lung cancer

[0219] In order to verify the predictive efficacy of the established risk score model for lung cancer, another group of independent lung cancer patients and high-risk lung cancer populations were selected as the validation cohort to evaluate the classification effect and feature weights of the model. Taking the pathological test results as the true value, the model evaluation indicators include accuracy (range 0-1), area under the receiver operating characteristic curve (AUC, range 0-1), positive predictive value (PPV, Positive Predictive Value, range 0-1), negative predictive value (NPV, Negative Predictive Value, range 0-1), sensitivity (range 0-1) and specificity (range 0-1). The higher the value, the better the effect, which are 0.7767, 0.8804 and 0.8 ( Figure 4 ), 0.76, 0.7273 and 0.8261 (Table 3). The results showed that in the validation cohort, this risk prediction model had high accuracy, AUC, sensitivity and specificity, and the model prediction efficiency was better.

[0220] Table 3. Cross-validation evaluation results of biomarkers in the training set

[0221]

[0222]

[0223] All documents mentioned in the present invention are cited as references in this application, just as each document is cited as reference individually. In addition, it should be understood that after reading the above teachings of the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the claims attached to this application.

Claims

1. Use of blood extracellular vesicle miRNA in the preparation of a chip, a detection reagent or a detection kit for early diagnosis of lung cancer.

2. The use according to claim 1, characterized in that The blood extracellular vesicle miRNA is: (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or (ii) a miRNA that is complementary to the miRNA sequence described in (i).

3. A miRNA chip, comprising: Solid phase carrier; as well as Oligonucleotide probes are sequentially fixed on the solid phase carrier, and the oligonucleotide probes specifically bind to miRNA; Wherein, the miRNA is: (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or (ii) a miRNA that is complementary to the miRNA sequence described in (i).

4. Use of the miRNA chip according to claim 3 for preparing an early diagnosis kit for lung cancer.

5. A detection kit containing a detection reagent for detecting miRNA; in, The miRNA is: (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or (ii) a miRNA complementary to the miRNA sequence described in (i) Alternatively, the detection kit is equipped with the miRNA chip according to claim 3.

6. A miRNA isolated from extracellular vesicles in blood for early diagnosis of lung cancer; in, The miRNA is: (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or (ii) a miRNA that is complementary to the miRNA sequence described in (i).

7. An isolated or artificially constructed precursor miRNA, which can be cleaved and expressed in human cells to become the miRNA according to claim 6.

8. An isolated polynucleotide, which can be transcribed into a precursor miRNA by a human cell, and which can be cleaved and expressed into the miRNA of claim 6 in a human cell.

9. A vector comprising the miRNA according to claim 6 or the polynucleotide according to claim 8.

10. A method for distinguishing a high-risk group of lung cancer from an early-stage lung cancer patient, comprising the steps of: (a) detecting the expression level of miRNA in a sample of a test subject; (b) inputting the miRNA expression value obtained in step (a) into the risk prediction model to obtain the probability that the subject is a high-risk population or an early lung cancer patient; (c) based on the probability obtained in step (b), providing a prediction result that the subject suffers from lung cancer; Wherein, the miRNA is: (i) a miRNA with a sequence as shown in SEQ ID NO: 1, a miRNA with a sequence as shown in SEQ ID NO: 2, a miRNA with a sequence as shown in SEQ ID NO: 3, a miRNA with a sequence as shown in SEQ ID NO: 4, a miRNA with a sequence as shown in SEQ ID NO: 5, a miRNA with a sequence as shown in SEQ ID NO: 6, a miRNA with a sequence as shown in SEQ ID NO: 7, and / or a miRNA with a sequence as shown in SEQ ID NO: 8; or (ii) a miRNA that is complementary to the miRNA sequence described in (i).

11. A device for distinguishing between a high-risk group of lung cancer and patients with early-stage lung cancer, the device comprising a storage device storing instructions for executing the method for distinguishing between a high-risk group of lung cancer and patients with early-stage lung cancer as claimed in claim 10.

Images