WO2015083791A1

WO2015083791A1 - Method for detecting lung cancer and detection kit

Info

Publication number: WO2015083791A1
Application number: PCT/JP2014/082122
Authority: WO
Inventors: 幸嗣植田
Original assignee: 国立研究開発法人理化学研究所
Priority date: 2013-12-04
Filing date: 2014-12-04
Publication date: 2015-06-11

Abstract

A method for detecting lung cancer, said method comprising a step for measuring the amount(s) of CD91 and/or CD317 in a sample collected from a living organism. A detection kit for detecting lung cancer, said detection kit comprising an antibody or peptide probe capable of recognizing CD91 in a sample collected from a living organism and/or an antibody or peptide probe capable of recognizing CD317 in a sample collected from a living organism.

Description

Lung cancer detection method and detection kit

The present invention relates to a detection method and detection kit for lung cancer.

Lung cancer is a leading cause of cancer-related deaths worldwide, with 1,475,117 deaths in 2011 (Global Health Health Observatory Data Repository, World Health Organization). This large number of deaths is largely due to the late diagnosis and lack of effective treatment. In fact, in recent cancer screening tests, only 30% of patients are diagnosed as being in an early stage and capable of being removed by surgery (Non-patent Document 1). Therefore, in order to improve clinical outcomes and overall survival, it is important to develop new biomarkers for lung cancer and establish early detection systems.

In recent years, the biological importance and clinical use of exosomes have been widely discussed. For example, the contribution of tumor-derived exosomes to the formation of a metastatic microenvironment is one of their most basic functions, which provides a better understanding of cancer metastasis and New treatments to prevent this can also be provided (Non-Patent Documents 2 to 4). Delivery of therapeutic RNA via exosomes is already in the pioneering stage of cancer treatment (Non-Patent Documents 5 and 6).

In the field of cancer diagnosis, exosomes are also attractive targets for biomarker searches due to their molecular properties (Non-Patent Documents 7 to 9). Basically, a set of molecules expressed in the original solid tumor cells can be detected as an exosome component in the blood circulation. Theoretically, exosome biomarkers are feasible, but it is difficult to separate exosomes from biological fluids, which significantly hinders efficient search for biomarker candidates. In fact, ultracentrifugation-based methods are the most common method for separating exosomes from serum samples (Non-Patent Document 10), but their reproducibility, processing time and purity quantitatively analyze a large number of clinical samples. Not suitable for biomarker screening studies (Non-patent Document 11).

An object of the present invention is to provide a detection method and detection kit for lung cancer using a novel biomarker for lung cancer.

The detection method according to the present invention is a method for detecting lung cancer, and includes a step of measuring the amount of at least one of CD91 and CD317 in a sample collected from a living body.

The detection kit according to the present invention is a detection kit for detecting lung cancer, an antibody or peptide probe that recognizes CD91 in a sample collected from a living body, and an antibody that recognizes CD317 in a sample collected from a living body Or at least one of the peptide probes.

This specification includes the contents described in the specification and / or drawings of Japanese Patent Application No. 2013-251548, which is the basis of the priority of the present application.

The present invention can provide a detection method and detection kit for lung cancer using a novel biomarker for lung cancer.

FIG. 1 is a schematic diagram of a workflow for searching an exosome biomarker in an example. Exosomes were isolated from serum samples from 46 individuals by anti-CD9 antibody-conjugated monolithic chip (anti-CD9-MSIA chip) on a 12-channel automated pipetting platform. The concentrated exosome fraction was analyzed by LC / MS / MS and subjected to label-free quantitative analysis by RefinerMS software on the Expressionist® proteome server system. The quantified peptides were subjected to a two-step statistical analysis consisting of ANOVA and a feature exclusion method, and finally extracted biomarker candidate peptides were identified using a Sequest database search. The identification threshold was set at a false positive rate (FDR) <1%. FIG. 2 is a diagram illustrating a processing workflow used in biomarker screening analysis in the Expressionist® RefinerMS module in the embodiment. FIG. 3 is a diagram showing the anti-CD9-MSIA chip in the example. (A) is an enlarged view of an anti-CD9-MSIA chip in an example. (B) shows the results of three measurements performed by LC / MS / MS after purifying the exosome fraction from a common serum sample using six independent anti-CD9-MSIA chips. FIG. Shows coefficient of variation (CV) of peak intensity corresponding to CD9 _155-170 peptide (GLAGGVEQFISDICPK, m / z = 845.9266; SEQ ID NO: 1) or CD81 _149-171 peptide (TFHETLDCCGSSTLTALTTSVLK, m / z = 848.0733; SEQ ID NO: 2) It was. FIG. 4 shows a comprehensive overview of the proteome of 1601 identified exosomal proteins in the examples. (A) is a pie chart showing the distribution of localization of proteins in cells. (B) is the figure which used Fisher exact probability statistics (Fisher | Exact | Statistics) in a DAVID system for functional annotation cluster analysis. Ten rich functions detected in 1601 exosomal proteins were demonstrated using Expression® Analysis® Systematic® Explorer (EASE) score. FIG. 5 is a diagram showing a two-stage statistical selection of biomarker candidates in the examples. As a first step, a 3-group ANOVA was performed to compare NC, IP and ADC groups (a) or NC, IP and SCC groups (b). Peptides satisfying the reference value p <0.001 were used for the second ranking selection. The upport-vector-machine-recursive-feature-elimination (SVM-RFE) method based on cross confirmation was used for comparison of NC, IP and ADC groups, and the minimum biomarker set showing the minimum misclassification rate was calculated (c). Similarly, the SVM-SVM method was used for comparison of NC, IP and ADC groups. The number of selected biomarkers and the misclassification rate are shown. NC: healthy control, IP: interstitial pneumonia, ADC: lung adenocarcinoma, SCC: lung squamous cell carcinoma. FIG. 6 is a diagram showing 19 exosomal biomarker candidates identified in the examples. LC / MS / MS signal intensities for 19 biomarker candidates from 46 samples were represented in a box plot. UniProtKB entry protein names are shown at the top of the box plot. N: healthy control, IP: interstitial pneumonia, ADC: lung adenocarcinoma, SCC: lung squamous cell carcinoma. FIG. 7 is a diagram showing a verification test based on an exosome sandwich ELISA for CD91 in Examples. (A) shows the principle of exosome sandwich ELISA. SA-HRP: Streptavidin-horseradish peroxidase. 212 independent serum samples were used to measure exosomal CD91 and CEA concentrations. (B) shows the serum exosome concentration determined by CD9-CD9 sandwich ELISA. (C) shows the concentration of CEA measured by a commercially available ELISA kit. (D) shows the concentration of CD91 in exosomes determined by CD9-CD91 sandwich ELISA. The values are standardized using the exosome concentration calculated in (b). The dashed line shows the cutoff value for CEA (c) at 5.0 ng / mL or exosomal CD91 (d) at 2.04 U / exosome. Sensitivity (Sens) and specificity (Spec) for each small group of lung cancers are shown at the bottom of the box plot. N: Healthy control, IP: Interstitial pneumonia, ADC_1_2: Stage I, II lung adenocarcinoma, ADC_3_4: Stage III, IV lung adenocarcinoma, SCC_1_2: Stage I, II lung squamous cell carcinoma, SCC_3_4: Stage III IV squamous cell carcinoma of the lung. FIG. 8 is a diagram showing ROC curve analysis for exosomal CD91 and CEA in Examples. The ROC curve for CEA (a), the ROC curve for CD91 of exosome (b), and the combined marker CEA + exosome CD91 (c) based on logistic regression are indicated by R, respectively. The diagnostic efficiency between NC + IP (n = 73) and pulmonary ADC patients (n = 105) was evaluated. The cut-off value was set to the point where the distance from (sensitivity, specificity) = (1, 1) reached the minimum value. Sensitivity (Sens), specificity (Spec), positive predictive value (PV +), negative predictive value (PV-) and area under the curve (AUC) are shown in each graph. FIG. 9 is a diagram showing a verification test based on an exosome sandwich ELISA for CD317 in Examples. The OD450 value in CD9-CD317 sandwich ELISA is shown. N: Healthy control, ADC_1: Stage I lung adenocarcinoma, ADC_2: Stage II lung adenocarcinoma, ADC_3: Stage III lung adenocarcinoma, ADC_4: Stage IV lung adenocarcinoma, SCC_1: Stage I lung squamous epithelium Cancer, SCC_2: Stage II lung squamous cell carcinoma, SCC_3: Stage III lung squamous cell carcinoma, SCC_4: Stage IV lung squamous cell carcinoma.

The inventor of the present application has developed a chip that can purify exosomes in a short time with high reproducibility. In order to identify biomarkers for lung cancer, exosomes were purified using this chip, and an attempt was made to search for biomarkers in exosomes. As a result, it was found that there is a protein specifically expressed in exosomes of lung cancer patients, and the present invention has been completed.

[1. (Lung cancer detection method)
The lung cancer detection method according to the present invention includes a step of measuring the amount of at least one of CD91 and CD317 in a sample collected from a living body. Other specific processes, and instruments and apparatuses to be used are not particularly limited.

The “sample” used in the lung cancer detection method according to the present invention is a sample (biological sample) collected from a living body (subject). Examples of samples collected from a living body include samples derived from body fluids such as blood, urine and saliva. Preferably, it is derived from blood. Examples of the sample derived from blood include whole blood, serum, and plasma. Among these, “sample” is preferably whole blood, serum and plasma, more preferably serum and plasma. One embodiment of the method for detecting lung cancer according to the present invention includes a step of collecting a sample from a living body (subject). Moreover, one embodiment of the lung cancer detection method according to the present invention includes a step of preprocessing a sample collected from a living body (subject). Examples of sample pretreatment include purification of collected whole blood to obtain serum, and purification of collected whole blood to obtain plasma.

Examples of the “living body (subject)” include mammals such as humans, mice, rats, rabbits, monkeys, and preferably humans. Examples of the blood collection site include an axillary skin vein, a cephalic vein, and an ulnar skin vein.

Here, “lung cancer” refers to lung adenocarcinoma, lung squamous cell carcinoma, small cell carcinoma, large cell carcinoma, and a mixture thereof. The lung cancer detection method according to the present invention is particularly suitable for the detection of lung adenocarcinoma and lung squamous cell carcinoma, and is more suitable for the detection of lung adenocarcinoma.

Although the stage of lung cancer is not particularly limited, in the method for detecting lung cancer according to the present invention, when measuring the amount of CD91, it is also possible to detect lung cancer at an early stage, for example, stage I and stage II, which has been difficult to detect conventionally. Can do. When measuring the amount of CD317, it is possible to suitably detect stage III and stage VI lung cancer.

CD91 (Prolow-density lipoprotein-receptor-related-protein-1) and CD317 (Bone-marrow-stromal-antigen-1), whose amounts can be measured in the method for detecting lung cancer according to the present invention, have known amino acid sequences. The accession numbers of UniProtKB for protein and human CD317 protein are shown respectively.

“CD91” includes not only the full-length CD91 protein but also fragments thereof, and also includes the precursor of CD91 protein. That is, for example, “measuring the amount of CD91” means measuring the amount of at least one of full-length CD91 polypeptide, a fragment of CD91 polypeptide, and a precursor of CD91 polypeptide. The CD91 whose amount is measured can be, for example, CD91 _579-605 .

“CD317” includes not only the full-length CD317 protein but also fragments thereof, and also includes the precursor of CD317 protein. Thus, for example, “measuring the amount of CD317” means measuring the amount of at least one of a full length CD317 polypeptide, a fragment of a CD317 polypeptide, and a precursor of a CD317 polypeptide. The CD317 whose amount is measured can be, for example, CD317 _137-147 .

In this specification, “measuring the amount of CD91” is intended to measure the abundance or concentration of CD91 in a biological sample or a sample obtained by purifying it. The “amount of CD91” may be an absolute amount (absolute mass or absolute concentration) or a relative amount (relative mass or relative concentration). Further, for example, it may be a peak area in mass spectrometry, a luminescence intensity in luminescence measurement, or the like, or a value indicating how many times it is with respect to a predetermined standard. “Amount of CD91” is preferably the amount of CD91 present on the exosome surface. Further, it may be preferable that the “amount of CD91” is CD91 per exosome. The same can be said for CD317.

In the present invention, those whose amounts can be measured are collectively referred to as “lung cancer marker” below.

As an example of a method for detecting lung cancer according to the present invention, the amount of CD91 in a biological sample derived from a subject is measured, the amount of CD91 in a biological sample derived from a control subject not having lung cancer is measured, By comparing these, lung cancer can be detected. Specifically, when the amount of CD91 in the sample derived from the subject is compared with the amount of CD91 in the sample derived from the control subject, and the amount of CD91 in the sample derived from the subject is significantly increased. Can be determined that the subject has lung cancer or is likely to have lung cancer. The amount of CD91 in the sample derived from the control subject can be determined in advance. Alternatively, the amount of CD91 in a biological sample derived from a control subject who does not have lung cancer is measured in advance, and a normal concentration range is determined based on this, and the amount of CD91 in the biological sample derived from the subject is normal. It can also be determined that the patient has lung cancer or is likely to have lung cancer when increasing outside the concentration range. That is, in one embodiment of the method for detecting lung cancer according to the present invention, when the measured amount of CD91 is increased compared to a control subject not having lung cancer, Or the process of determining with the high possibility of having lung cancer is further included. In another embodiment, the control subjects may include both control subjects and lung cancer patients who do not have lung cancer, or only lung cancer patients.

When comparing the amount of CD91 in a subject with the amount of CD91 in a control subject, the significant difference is determined by a statistical method such as t-test, F-test, chi-square test, or Mann-Whitney's U test. I can judge.

Further, another example of the method for detecting lung cancer according to the present invention is to measure the amount of CD317 instead of CD91, and the same can be said. Therefore, in another embodiment of the method for detecting lung cancer according to the present invention, whether the measured amount of CD317 is increased compared to a control subject not having lung cancer, has lung cancer? Or a step of determining that the patient is likely to have lung cancer.

The method for measuring the amount of the lung cancer marker is not particularly limited as long as the abundance of the lung cancer marker can be quantitatively or semi-quantitatively determined. For example, an antibody that recognizes the lung cancer marker to be measured was used. A method using an immunological technique, a method using a peptide probe that recognizes a lung cancer marker to be measured, a liquid column chromatography method, a mass spectrometry method, and the like can be used. Examples of the method using an antibody include an ELISA method, a quantitative western blotting method, a radioimmunoassay method, an immunochromatography method, and an immunoprecipitation method. The type of ELISA method is not particularly limited, but is a so-called antigen measurement system (measurement of the amount of antigen contained in a biological sample), ELISA by direct adsorption method, ELISA by competition method, ELISA by sandwich method, and micro flow An ELISA specialized for the measurement of a very small amount of sample using a road system or microbeads can be mentioned.

The antibody that recognizes the lung cancer marker to be measured may be a monoclonal antibody or a polyclonal antibody, but is preferably a monoclonal antibody. Based on information from public databases such as UniProt, those skilled in the art can easily determine an amino acid sequence suitable as an antigen for producing an antibody that recognizes CD91 and an antibody that recognizes CD317. The antibody that recognizes CD91 is preferably an antibody specific to CD91. The antibody that recognizes CD317 is preferably an antibody specific for CD317.

In the present invention, “antibody” is intended to mean a form including all classes and subclasses of immunoglobulins and functional fragments of antibodies. The antibody is a concept including both natural antibodies of a polyclonal antibody and a monoclonal antibody, and includes an antibody produced using a gene recombination technique and a functional fragment of the antibody. The “functional fragment of an antibody” refers to one having a partial region of the above-described antibody and having an antigen-binding ability (synonymous with a binding fragment). Natural antibodies can be derived from any species including, but not limited to, humans, mice, rats, monkeys, goats, rabbits, camels, llamas, cows and chickens. The antibody produced using gene recombination technology is not particularly limited, but chimeric antibodies such as humanized antibodies and primatized antibodies obtained by genetic modification of natural antibodies, synthetic antibodies, recombinant antibodies, Mutagenized antibodies and graft-bound antibodies (for example, antibodies to which other proteins and radioactive labels are conjugated or fused), and antibodies already produced using genetic recombination techniques are described above. These also include antibodies that have been modified in the same manner as when genetically modifying natural antibodies. Specific examples of functional fragments of antibodies include F (ab ′) ₂ , Fab ′, Fab, Fv (variable fragment of antibody), sFv, dsFv (disulphide stabilized Fv) and dAb (single domain antibody). Etc.

Furthermore, in the present invention, the binding fragment includes an antibody fragment mutated in a range that maintains reactivity with the target lung cancer marker as a concept of the binding fragment. The aforementioned mutation introduction is performed using a known technique such as a gene modification technique, which is appropriately selected by those skilled in the art.

The detection of the above-described antibody can be performed by forming a complex with a secondary antibody labeled with a reporter molecule capable of attracting a detectable signal. Alternatively, the antibody described above may be pre-labeled with a reporter molecule. The reporter molecule is, for example, an enzyme, a fluorophore-containing molecule, or a radionuclide-containing molecule. Various reporter molecules are known and may be appropriately selected.

When an enzyme is used as the reporter molecule, for example, peroxidase, β-galactosidase, alkaline phosphatase, glucose oxidase, or acetylcholinesterase can be used as the enzyme. These enzymes and antibodies can be bound by a known method using a crosslinking agent such as a maleimide compound. As the substrate, a known substance can be used according to the type of enzyme used. For example, when peroxidase is used as the enzyme, 3,3 ′, 5,5′-tetramethylbenzidine and the like can be used. Moreover, when using an alkaline phosphatase as an enzyme, paranitrol phenol etc. can be used. As the radionuclide, ¹²⁵ I, ³ H, or the like can be used. As the fluorescent dye, fluoroless isothiocyanate (FITC) or tetramethylrhodamine isothiocyanate (TRITC) can be used.

A simple measurement system such as immunochromatography for detecting by aggregation of colloidal gold on which the secondary antibody is immobilized can also be used.

In the method using a peptide probe, for example, a peptide probe labeled with the above-mentioned reporter molecule can be used. The peptide probe that recognizes CD91 is preferably a peptide probe specific for CD91. The peptide probe that recognizes CD317 is preferably a peptide probe specific for CD317.

Thus, in the measurement of the amount of the lung cancer marker, a luminescence measurement method including color development or fluorescence measurement, or a radiometric method can be used.

Mass spectrometry may be performed using a known mass spectrometer. Since mass spectrometry using a mass spectrometer is excellent in sensitivity and accuracy, accurate determination can be made. Furthermore, by using a multi-channel mass spectrometer capable of simultaneous multi-component analysis, the amount of two or more lung cancer markers (for example, CD91 and CD317) can be measured at a time. Furthermore, biomarkers relating to other diseases other than lung cancer can be simultaneously measured, and detection of various diseases can be attempted at once. In order to perform detection with higher accuracy, it is preferable to use a tandem mass spectrometer (MS / MS). The mass spectrometer used in the detection method of the present invention is not particularly limited as long as it can be quantified, and conventionally known types of mass spectrometers such as a quadrupole type and a time-of-flight type can be used.

In the case of measuring the amount of lung cancer marker present on the exosome surface, the ELISA method is preferable, and the sandwich ELISA is more preferable. As an example, a sandwich ELISA for measuring the amount of CD91 present on the exosome surface will be described (see also FIG. 7 (a)). First, an antibody (capture antibody) against a protein (excluding CD91) present on the exosome surface. ) Is bound, and then contacted with a sample containing exosomes to form a capture antibody-exosome complex. Further, it is brought into contact with a labeled anti-CD91 antibody (detection antibody) to form a capture antibody-exosome-detection antibody complex. Then, the amount of CD91 present on the exosome surface can be measured by measuring the color development intensity and the like. The capture antibody is preferably directed to a protein that has no significant difference in the expression level on the exosome surface between lung cancer patients and healthy individuals who do not have lung cancer. Examples of such proteins include CD9 (5H9 antigen, cell-growth-inhibiting gene-2 protein, leukocyte antigen-MIC3, Motility-related protein protein MRP-1, Tetraspanin-29; p24), CD63, CD81, and other molecules belonging to the tetraspanin family. , Anti-CD9 antibody, anti-CD63 antibody, anti-CD81 antibody and the like (reference document: International Publication WO2013 / 099925). For CD317, the amount existing on the exosome surface can be measured by the same method as described above.

In the method for detecting lung cancer according to the present invention, it is preferable to measure the amount of CD91 and / or CD317 per exosome. In this case, the amount of exosome in the biological sample or a sample obtained by purifying the same is measured, and the result is used to normalize the separately measured amount of CD91 and / or CD317 to obtain CD91 and / or per exosome. The amount of CD317 is calculated. The “amount of exosome” may be an absolute amount (absolute mass or absolute concentration) or a relative amount (relative mass or relative concentration). Since the amount of exosome varies depending on the subject, lung cancer can be detected more accurately by standardization. That is, in one embodiment of the lung cancer detection method according to the present invention, the method includes a step of measuring the amount of exosome in a sample collected from a living body. In another embodiment of the lung cancer detection method according to the present invention, the method includes a step of calculating the amount of CD91 per exosome using the measured amount of exosome and the measured amount of CD91. Further, in another embodiment of the lung cancer detection method according to the present invention, a step of calculating the amount of CD317 per exosome using the measured amount of exosome and the measured amount of CD317 is included.

As a method for measuring the amount of exosome, sandwich ELISA can be mentioned. An example of a sandwich ELISA for measuring the amount of exosome will be described (see also FIG. 7 (a)). First, a substrate to which a first antibody (capture antibody) against a protein existing on the exosome surface is prepared is prepared. And then contacted with a sample containing exosomes to form a capture antibody-exosome complex. Furthermore, it is contacted with a labeled second antibody (detection antibody) to form a capture antibody-exosome-detection antibody complex. Then, the amount of exosome is measured by measuring the color intensity. The first antibody and the second antibody are preferably against a protein that has no significant difference in the expression level on the exosome surface between a lung cancer patient and a healthy person who does not have lung cancer. Examples include CD9, CD63, CD81, and other molecules belonging to the tetraspanin family, and examples of the antibody include anti-CD9 antibody, anti-CD63 antibody, and anti-CD81 antibody. The first antibody and the second antibody may be the same type of antibody, or may be different types of antibodies. In addition, the first antibody and the second antibody may be of the same type or different types from those used in the sandwich ELISA for measuring the amount of CD91 and / or CD317 present on the exosome surface. May be.

In the method for detecting lung cancer according to the present invention, exosomes may be purified from a sample collected from a living body. That is, one embodiment of the lung cancer detection method according to the present invention includes a step of purifying exosomes from a sample collected from a living body. Examples of the method for purifying exosomes include a method using an antibody against a protein existing on the exosome surface, a method using an ultracentrifuge, and a method using a commercially available purification kit. In an example of a method using an antibody against a protein present on the exosome surface, a mass spectrometry immunoassay pipette chip to which the antibody (for example, anti-CD9 antibody) is immobilized is used (see also Examples). By using this method, exosomes can be purified from a sample in a short time with high purity and good reproducibility.

The method for detecting lung cancer according to the present invention preferably further includes a step of measuring the amount of CEA (carcinoembryonic antigen) in a sample collected from a living body. By combining the amount of CD91 and / or CD317 with the amount of CEA, sensitivity and specificity are improved, and lung cancer can be detected more accurately (see also Examples). CEA has a known amino acid sequence, and the accession number of UniProtKB is P06731.

In addition, the method for detecting lung cancer according to the present invention may include a step of measuring the amount of CD91 and the amount of CD317 in a sample collected from a living body. Combining the amount of CD91 and the amount of CD317 may improve sensitivity and specificity and enable more accurate detection of lung cancer.

“CEA” includes not only the full length of the CEA protein but also a fragment thereof, and also includes a precursor of the CEA protein. That is, for example, “measuring the amount of CEA” means measuring the amount of at least one of a full-length CEA polypeptide, a fragment of CEA polypeptide, and a precursor of CEA polypeptide.

In addition, “sensitivity” refers to a ratio (true positive ratio) indicating a positive (abnormal subject value) when a test is performed on a population developing a specific disease. In addition, “specificity” refers to a ratio (a ratio of true negative) indicating a negative (normal value) when a test is performed on a population not suffering from a specific disease. “Positive predictive value” refers to the proportion of individuals who actually have a disease among subjects who show a positive result in a test. Further, the “negative predictive value” means the proportion of individuals who do not actually have a disease among subjects who show a negative test result.

When measuring the amounts of multiple types of lung cancer markers, they may be measured using the same sample or different samples.

When combining the amount of CD91 and the amount of CEA, for example, the logistic regression score is calculated using the amount of CD91 and the amount of CEA. As a calculation method, for example, the methods described in the literature: Pepe, M. S., Cai, T., and Longton, G. (2006), Biometrics 62, 221-229. If there is a significant difference between the two for this logistic regression score, it can be determined that the subject has lung cancer or is likely to have lung cancer. The amount of CEA in the sample derived from the control subject can be determined in advance. Alternatively, the amount of CEA and the amount of CD91 in a biological sample derived from a control subject are measured in advance, a logistic regression score is calculated, and based on this, a normal value range of the logistic regression score is determined. It can also be determined that the patient has lung cancer when the logistic regression score is out of the normal value range in the biological sample. The same method as described above is used when combining the amount of CD317 and the amount of CEA, when combining the amount of CD91 and the amount of CD317, and when combining the amount of CD91, the amount of CD317 and the amount of CEA. be able to.

When comparing a subject's logistic regression score to a control subject's logistic regression score, the significant difference is determined by a statistical method such as t-test, F-test, chi-square test, or Mann-Whitney's U test. I can judge.

The method for measuring the amount of CEA is not particularly limited as long as these abundances can be determined quantitatively or semi-quantitatively. For example, an immunological technique using an antibody recognizing CEA was used. A method, a method using a peptide probe recognizing CEA, a liquid column chromatography method, a mass spectrometry method and the like can be used (the details are the same as described above). The sample for measuring the amount of CEA may be the same sample as the sample for measuring the amount of CD91 and / or CD317, or may be a different sample. The antibody that recognizes CEA is preferably an antibody specific for CEA. The peptide probe that recognizes CEA is preferably a peptide probe specific for CEA.

[2. (Lung cancer detection kit)
The detection kit for lung cancer according to the present invention is a detection kit for detecting lung cancer, and recognizes an antibody or peptide probe that recognizes CD91 in a sample collected from a living body and CD317 in a sample collected from a living body. At least one of an antibody or a peptide probe.

If the lung cancer detection kit according to the present invention is used, the amount of CD91 and / or CD317 can be measured. Using the amount of CD91 and / or CD317, for example, [1. Lung cancer can be detected according to the method described in the section “Lung cancer detection method”.

The lung cancer detection kit according to the present invention preferably further comprises an antibody or peptide probe that recognizes CEA in a sample collected from a living body. By combining the amount of CD91 and / or CD317 and the amount of CEA, sensitivity and specificity are improved, and lung cancer can be detected more accurately.

The peptide probe refers to a peptide probe that recognizes CD91, CD317 or CEA, preferably a peptide probe that specifically binds to CD91, CD317 or CEA. Specific examples include peptide sequences that specifically bind to CD91, CD317, or CEA. The peptide probe included in the detection kit may include a non-natural amino acid in addition to a natural amino acid.

The lung cancer detection kit according to the present invention preferably further includes a substrate to which an antibody (capture antibody) for capturing an exosome in the sample is bound. Examples of antibodies for capturing exosomes include antibodies to proteins (excluding CD91 and CD317) present on the exosome surface. The antibody is preferably directed to a protein having no significant difference in the expression level on the exosome surface between a lung cancer patient and a healthy person who does not have lung cancer. Examples of such a protein include CD9, CD63, CD81. And other molecules belonging to the tetraspanin family, and examples of the antibody include anti-CD9 antibody, anti-CD63 antibody, and anti-CD81 antibody. In one embodiment of the detection kit for lung cancer according to the present invention, the capture antibody and the substrate may be separately included. In another embodiment of the lung cancer detection kit according to the present invention, either one of the capture antibody and the substrate may be included.

Examples of the base material include materials such as protein highly adsorbed polypropylene. The substrate is preferably a microwell plate (eg, 96 well). This is because a large number of samples can be processed in parallel.

The lung cancer detection kit according to the present invention further includes various reagents (secondary antibodies, reporter molecules, buffers, etc.) and instruments (plates and pipettes) for immunologically detecting antibodies or peptide probes as necessary. Etc.), at least one of a use instruction of the detection kit, a control sample used at the time of measurement, control data used when analyzing the measurement result, and the like. In addition, in the instruction manual of the detection kit, the above [1. The contents of the lung cancer detection method according to the present invention described in the section “Lung cancer detection method” are recorded.

Although not particularly limited, the detection kit for lung cancer according to the present invention is preferably, for example, a detection kit based on the ELISA method, and more preferably a detection kit based on the sandwich ELISA method.

[3. Others]
[1. The result obtained by carrying out the detection method described in the section “Lung cancer detection method” can be used as one of diagnostic materials for diagnosis by a doctor. The above [1. For subjects whose results indicate that they may have lung cancer by performing the detection method described in the section `` Lung cancer detection method '', With this, treatment can be performed. Here, examples of treatment include chemotherapy, radiation treatment, and surgery performed by a doctor, and in some cases, a specialist other than the doctor. In particular, since the detection method according to the present invention can detect early lung cancer, early diagnosis and early treatment can be realized. Further, according to the detection method according to the present invention, it can be detected that the cancer is stage I or II, or stage III or IV lung cancer, so that the result obtained by the doctor performing the method according to the present invention is used. Based on this, diagnosis or treatment suitable for each stage can be adopted.

More specifically, for example, when a doctor determines that “possibility of having lung cancer” based on the result obtained by performing the method according to the present invention, other examination methods (X-rays) Examination, CT examination, endoscopy, etc.), smoking history, subject age and family history, etc. can be taken into account for biopsy. Usually, a definitive diagnosis of the presence or absence of lung cancer is made based on the result of a biopsy, and a treatment policy is established. When a doctor determines that "there is a possibility of having lung cancer", a predetermined reference value is set in advance, and if the measured value of the subject exceeds the reference value, It may be determined that there is a possibility that

[4. (Summary)
As described above, the detection method according to the present invention is a method for detecting lung cancer, and includes a step of measuring the amount of at least one of CD91 and CD317 in a sample collected from a living body.

The detection method according to the present invention has lung cancer when the amount of at least one of the measured CD91 and CD317 is increased compared to a control subject not having lung cancer, Alternatively, it is preferable to further include a step of determining that there is a high possibility of having lung cancer.

In the detection method according to the present invention, the amount of CD91 is the amount of CD91 present on the exosome surface, and the amount of CD317 is preferably the amount of CD317 present on the exosome surface.

The detection method according to the present invention preferably further includes a step of measuring the amount of CEA in a sample collected from a living body.

In the detection method according to the present invention, the lung cancer may be stage I or II.

In the detection method according to the present invention, the lung cancer may be stage III or IV.

In the detection method according to the present invention, it is preferable to measure the amount of at least one of CD91 and CD317 by ELISA.

In the detection method according to the present invention, the sample is preferably whole blood, serum or plasma.

The detection kit according to the present invention preferably further includes a base material to which an antibody for capturing the exosome in the sample is bound.

In the detection kit according to the present invention, the antibody for capturing the exosome is preferably an anti-CD9 antibody.

The detection kit according to the present invention preferably further comprises an antibody or peptide probe that recognizes CEA in a sample collected from a living body.

Examples will be shown below, and the embodiments of the present invention will be described in more detail. Of course, the present invention is not limited to the following examples, and it goes without saying that various aspects are possible in detail. Further, the present invention is not limited to the above-described embodiments, and various modifications can be made within the scope shown in the claims, and the present invention is also applied to the embodiments obtained by appropriately combining the disclosed technical means. It is included in the technical scope of the invention. Moreover, all the literatures described in this specification are used as reference.

[Experimental procedure]
In this example, biomarkers were searched and verified. The outline of biomarker search is as shown in FIG.

(Serum sample)
Serum samples from lung cancer patients (n = 165) and interstitial pneumonia patients (n = 29) were collected at the Hiroshima University Hospital upon study approval. Healthy control serum samples (n = 64) were collected at Hiroshima NTT Hospital. Documented informed consent was obtained from all collaborators. This study was approved by the RIKEN Ethics Committee (approval code: Yokohama H20-12) and the Hiroshima University Hospital Ethics Committee.

(Exosome purification by CD9-MSIA chip)
All procedures were performed on a Novus i 12 channel electronic pipette and an adjustable pipette stand (Thermo Fisher Scientific, USA). MSIA DART's, Protein G chip (Thermo Fisher Scientific) was equilibrated by pipetting (300 μL × 10 cycles) in 500 μL of PBS, and then 1 μg of anti-CD9 antibody (Shiono Pharmaceutical Co., Ltd.) in 50 μL of PBS. The product was pipetted (25 μL × 100 cycles) to immobilize the anti-CD9 antibody. After cross-linking (100 μL × 100 cycles) using 0.25 mM BS3 (Thermo Fisher Scientific) in PBS, the reaction was stopped using 100 mM ethanolamine-HCl (pH 8.0) (100 μL × 100). Cycle). Equilibrated in 500 μL PBS (300 μL × 10 cycles) and then the CD9-MSIA chip was incubated with 350 μL of a 7-fold diluted serum sample (by pipetting 300 μL × 100 cycles). The chip was washed 3 times by pipetting (300 μL × 25 cycles) in 500 μL PBS. Finally, pipetting (20 μL × 100 cycles) in 30 μL 8M urea + 50 mM ammonium bicarbonate solution was performed to elute the captured exosomes.

Reduced for 30 minutes at 37 ° C. with 5 mM TCEP and alkylated with 25 mM iodoacetamide for 45 minutes at room temperature. The sample was then diluted 7-fold with 50 mM ammonium bicarbonate, placed in a 96-well filter plate, and digested by shaking with 5 μL of immobilized trypsin (Thermo Fisher Fisher Scientific) at 37 ° C. for 6 hours. The trypsin digest was desalted using Oasis HLB 96-well Elution Plate (Waters Corporation, USA) and subjected to LC / MS / MS analysis.

(LC / MS / MS analysis)
LTQ-Orbitrap-Velos mass resuspended dry peptide sample in 2% acetonitrile solution containing 0.1% trifluoroacetic acid and equipped with Ultimate 3000 RSLC nano-flow HPLC system (DIONEX, USA) Analysis was performed using a spectrometer (manufactured by Thermo Fisher Scientific). In a 75 μm × 150 mm C18 tip-column (manufactured by Nikyo Technos), solvent A (0.1% formic acid) and solvent B (0.1% formic acid in acetonitrile) were used at a flow rate of 250 nL / min. Samples were separated by a multi-step linear gradient with solvent B 6.4-30% in 95 minutes and 30-95% solvent B in 10 minutes.

The eluted peptide was ionized with a spray voltage of 2000 V, and MS data was acquired by the data-dependent fragment method. The measurement scan was performed at m / z 400-1600, resolution 60,000, AGC target value 1.0 × 10 ⁶ ion count. The top 20 intensities of precursor ions in each measurement scan were subjected to low resolution MS / MS acquisition using a normal CID scan mode with an AGC target value of 5000 ion counts in a linear ion trap.

Protein identification was attempted by performing a SEQUEST database search using Proteome Discoverer 1.3 software (Thermo Fischer Scientific). MS / MS spectra were searched against the human protein database SwissProt 2013_03 (20,255 sequences). The search parameters are: Enzyme Name = Semitrypsin, Precursor Mass Tolerance = 3 ppm, Fragment Mass Tolerance = 0.8 Da, Dynamic Modification = Oxidation (Met), Static Modification = Carbamidomethyl (Cys). When FDR (false discovery rate) by Peptide Validator activity in SEQUEST Decoy Database Search was less than 1%, the peptide was identified.

(Label-free quantitative analysis on Expressionist server)
The LC / MS / MS dataset was loaded into an Expressionist RefinerMS module (Genedata AG, Switzerland) for data processing and free quantitative analysis. The overall workflow of RefinerMS software is shown in Figure 2. A Spectrum Grid was set for every 10 data points on the 2D MS chromatogram plane (x = m / z, y = RT). Subsequently, Structure Removal was sequentially performed at RT = 2 scans and m / z = 6 points in the first and second chemical noise subtraction. Subsequently, in the third chemical noise subtraction, structure removal was performed at RT Window = 500 scans and Quantile = 90%. In the fourth Chemical Noise Subtraction, Structure Removal was performed at RT = 2 scans. Thereafter, signals of intensity less than 2000 were excluded using Intensity Thresholding. Finally, Structure Removal was performed at RT = 2 scans and m / z = 5 points in the fifth and sixth Chemical Noise Subtraction, respectively.

Next, Chromatogram Grid was set every 10 scans on the noise subtraction data, and Chromatogram RT Alignment was performed using the following parameters: m / z Window = 11 points, RT Window = 11 scans, Gap Penalty = 1, RT Search Interval = 2 minutes, Alignment Scheme = pairwise alignment based tree.

Subsequently, peaks in the temporal averaged chromatogram were detected by Summed Peak Detection Activity using the following parameters: Summation Window = 20 scans, Overlap = 10, Minimum Peak Size = 6 scans, Maximum Merge Distance = 1 data points, Gap / Peak Ratio = 5, Method = curvature-based peak detection, Peak Refinement Threshold = 5, Consistency Filter Threshold = 1. Finally, using Ismed Isotope Clustering Activity, isotopic peaks derived from one molecule were grouped into isotope clusters. The parameters at this time are as follows: Minimum Charge = 1, Maximum Charge = 6, Maximum Missing Peaks = 0, First Allowed Gap Position = 10, Ionization = protonation, RT Tolerance = 0.1 minute, m / z Tolerance = 0.01 Da, Minimum Cluster Size Ratio = 0.5.

(Statistical analysis in Expressionist Analyst)
A two-step statistical selection was performed for efficient identification of biomarkers. In the first stage, analysis of variance (ANOVA) of 3 groups was performed to roughly extract candidates showing expression levels significantly separated between the 3 clinical groups (p <0.001). Next, the minimum combination of biomarker candidates was estimated according to the instruction manual by the Support Vector Machine-Recursive Feature Elimination (SVM-RFE) algorithm or SVM-SVM algorithm in Expressionist Analyst module (Genedata AG).

(Exosome sandwich ELISA assay 1)
An exosome capture antibody solution (5 μg / mL anti-CD9 antibody in PBS, 50 μL / well) was loaded onto a Nunc MaxiSorp flat-bottom 96 well plate (Thermo Fischer Scientific) and incubated at 4 ° C. overnight. Blocking solution (5% BSA in PBS, 150 μL / well) was added and incubated on a plate shaker for 60 minutes at ambient temperature. After three washes with PBS, (5 μL serum + 95 μL PBS) was loaded into the upper 48 wells and (30 μL serum + 70 μL PBS) were loaded into the lower 48 wells, respectively. After 5 hours incubation, the plates were washed 3 times with PBS. Biotinylated anti-CD9 antibody (125 ng / mL) in 1% BSA was loaded into the upper 48 wells and biotinylated anti-CD91 antibody (500 ng / mL) in 1% BSA was loaded into the lower 48 wells (100 μL). / Well). After 60 minutes incubation, the plates were washed 3 times with PBS and subsequently covered with 1 × HRP-streptavidin (Abcam) in 1% BSA (100 μL / well). After incubation for 30 minutes, the plate was washed 3 times with PBS, followed by loading One-step Ultra TMB-ELISA Substrate Solution (Thermo Fischer Scientific) into the well (100 μL / well). The reaction was stopped after 10 minutes incubation with 2N HCl (100 μ / well). The OD at 450 nm was measured immediately. CD9-CD9 and CD9-CD91 ELISA assays were standardized using gradient curves (5, 10, 15, 20, and 25 μL) obtained from common pleural effusion samples. Finally, the concentration of CD91 was normalized using the exosome concentration calculated using the CD9-CD9 ELISA described above.

(Exosome sandwich ELISA assay 2)
Exosome capture antibody solution (5 μg / mL anti-CD9 antibody in PBS, 50 μL / well) was loaded onto a Nunc MaxiSorp flat-bottom 96 well plate (Thermo Fischer Scientific) and incubated for 60 minutes at ambient temperature. Blocking solution (5% BSA in PBS, 150 μL / well) was added and incubated on a plate shaker for 30 minutes at ambient temperature. After washing 3 times with PBS, (10 μL serum + 90 μL PBS) was loaded into the wells. After 3 hours incubation, the plates were washed 3 times with PBS. HRP-labeled anti-CD317 antibody (250 ng / mL) in 1% BSA was loaded into the wells (100 μL / well). After incubation for 60 minutes, the plate was washed 3 times with PBS, followed by loading One-step Ultra TMB-ELISA Substrate Solution (Thermo Fischer Scientific) into the well (100 μL / well). After 15 min incubation, the reaction was stopped with 2N HCl (100 μL / well). The OD at 450 nm was measured immediately.

(Box plot analysis and ROC curve analysis)
The intensity of the mass spectral peak corresponding to the biomarker peptide candidate was displayed by a box plot using the R algorithm. For each survey, the box shows the distribution of data points contained in the first to third quartiles. The line running through the box shows the median value. The length of the line above and below the box is defined by the maximum data value and the minimum data value, respectively. The data value exists within a range of 1.5 times the box. The ROC curve is indicated by R. The cut-off value was set to the point where the distance from (sensitivity, specificity) = (1, 1) reached the minimum value. Sensitivity (Sens), specificity (Spec), positive predictive value (PV +), negative predictive value (PV−) and area under the curve (AUC) are shown on each graph.

〔result〕
(Serum exosome isolation with anti-CD9-MSIA chip)
Separation of serum-derived exosomes with reproducible and high purity is essential for efficient biomarker identification. Therefore, a low back pressure monolith chip with immobilized antibody was used in an automated 12-channel pipette system ((a) of FIG. 3). This allows exosome isolation from 12 serum samples simultaneously in 45 minutes. Here, CD9 which is a tetraspanin molecule was selected as a target of the exosome capture antibody. This is because CD9 is strongly expressed on the surface of exosomes secreted from various types of cells (Reference: Keller S, Sanderson MP, Stoeck A, Altevogt P (2006) Exosomes: from biogenesis and secretion to biological. function. Immunol Lett 107: 102-108.).

Serum samples from 46 people (10 NCs, 10 IPs, 14 advanced stage ADCs, 12 advanced stage SCCs), LC / MS / MS for use of this new device And was subjected to statistical biomarker screening (FIG. 1, Table 1). Importantly, it has been confirmed that exosomes can be collected from human serum with high reproducibility by using an anti-CD9-MSIA chip (FIG. 3 (b)). Using 6 independent chips, the exosome fraction was purified from a common serum sample and analyzed by LC / MS / MS three times. The coefficient of variation (CV) of the peak region corresponding to the CD9 _155-170 peptide (GLAGGVEQFISDICPK, m / z = 845.9266; SEQ ID NO: 1) and the CD81 _149-171 peptide (TFHETLDCCGSSTLTALTTSVLK, m / z = 848.0733; SEQ ID NO: 2) is They were 2.49% and 2.87%, respectively. This indicates that the error in relative quantitative analysis based on the anti-CD9-MSIA chip is small enough and sufficient to identify a reliable biomarker.

(Overview of human serum exosome proteome)
To assess the isolation efficiency of anti-CD9-MSIA chips, 1601 proteins identified from LC / MS / MS analysis of 46 serum samples in the biomarker search stage were classified based on subcellular localization (FIG. 4). Cell component distribution by DAVID GO analysis showed very rich 826 intracellular proteins (51.6%) and 333 plasma membrane proteins (20.8%), whereas 146 Only extracellular protein (9.1%) was identified from the eluate of the anti-CD9-MSIA chip. These numbers clearly showed high yield enrichment of exosomes related to the cellular components from the original cells. Importantly, most of the serum proteins that significantly interfered with sensitive detection of a small number of exosomal proteins were effectively removed during the MSIA purification step. Furthermore, in order to comprehensively elucidate the physiological functions of serum exosomes, the functional characteristics of 1601 identified proteins were evaluated based on the Expression Analysis Systematic Explorer (EASE) score (Fig. 4 ( b)). This functional estimation suggested that exosomes may be associated with immune regulation, cell-cell interactions and stimulatory responses in addition to vesicular trafficking. These data will contribute to new discoveries not only about the biological function of tumor-derived exosomes, but also about the biological function of normal exosomes.

(Statistical identification of exosome biomarkers for lung cancer)
113,582 non-overlapping peptides from 46 serum samples were quantified by label-free quantitative analysis (Figures 1 and 2) in the Expressionist RefinerMS module. In the first statistical selection, using ANOVA, characteristic peptides specific to ADC patients (230 peptides, p <0.001) or SCC patients (316 peptides, p <0.001) Was roughly extracted. These are shown in FIG. 5 (a) and FIG. 5 (b), respectively. As a second step, a minimum biomarker set showing a minimum misclassification rate was calculated using a feature exclusion method based on cross validation. Here, according to the support vector machine recursive feature elimination (SVM-RFE) method or SVM-SVM method, the true prediction rate in the ADC patient group is 181 peptides and the true prediction rate in the SCC patient group is 90.9% 32 peptides with 100% were determined as final candidate biomarkers (FIG. 5 (c) and FIG. 5 (d), respectively).

Referring to 1601 protein identification data, 20 peptides derived from 19 proteins were identified (FIG. 6 and Table 2). Among these, CD91, ITA2B (Integrin alpha-IIb) and CD317 are expressed on the surface of exosomes and can be easily measured by exosome sandwich ELISA. Therefore, preferred exosome biomarker candidates in subsequent large-scale verification tests Met.

(Large-scale verification test for CD91 using exosome sandwich ELISA)
To assess the quantitative reproducibility of the results of label-free quantification in the single run screening analysis and clinical utility of candidate marker CD91 as described above, 212 additional serum samples (Table 1) were used to determine the exosome sandwich. A verification test was performed by ELISA. In this assay, an anti-CD9 antibody was used as an exosome capture antibody, and a biotinylated anti-CD9 antibody or anti-CD91 antibody was used as a detection antibody ((a) of FIG. 7). The concentration of serum exosomes measured by CD9-CD9 sandwich ELISA varied widely (FIG. 7 (b)), and the measurement results of CD9-CD91 sandwich ELISA were normalized by the concentration of exosomes (FIG. 7 (( d) in U / exosomes). In addition, CEA, an existing clinical biomarker, was tested on the same sample set (FIG. 7 (c)), and compared with that of exosome CD91 for diagnostic effectiveness.

Sensitivity in stage III and IV ADC patients was 66.3% for CEA and CD91 at a cutoff value of 2.04 U / exosome for exosome CD91 and 5.0 ng / mL for CFA. In 61.4%, CEA was slightly higher. However, the power in stage I and II ADC patients was 22.7% for CEA and 54.5% for CD91, which was significantly higher compared to CEA. The exosome CD91 false positive rate in the control group (NC and IP, n = 73) was 11.0%, and the CEA false positive rate was 8.2%. These results indicated that exosomal CD91 had a better potential for early detection of lung ADC compared to CEA.

Further, using all the control samples (n = 73) and all the ADC samples (n = 105), the improvement in the effect by combining exosomal CD91 and CEA was evaluated. The ROC curve analysis in (a) to (c) of FIG. 8 shows that the combined biomarkers are each sensitive (71.4%), specificity (91.8%), and under the curve compared to the case of each alone. It has been found that all areas (0.882) are effectively improved.

(Large-scale verification test for CD317 using exosome sandwich ELISA)
To assess the quantitative reproducibility of the results of label-free quantification in the single run screening analysis and clinical utility of candidate marker CD317 as described above, 89 additional serum samples (Table 3) were used to determine the exosome sandwich. A verification test was performed by ELISA. In this assay, anti-CD9 antibody was used as an exosome capture antibody, and HRP-labeled anti-CD317 antibody was used as a detection antibody (FIG. 9).

As a result, the power of detection was particularly high in ADC patients. Furthermore, it was found that the amount of blood exosome CD317 increases depending on the stage of lung adenocarcinoma.

The present invention can be used for detection of lung cancer. Therefore, the present invention can be widely used in the diagnostic medical field and the health medical field.

Claims

A method for detecting lung cancer, comprising:
A detection method comprising a step of measuring the amount of at least one of CD91 and CD317 in a sample collected from a living body.
May have lung cancer or have lung cancer if the measured amount of at least one of CD91 and CD317 is increased compared to a control subject not having lung cancer The detection method according to claim 1, further comprising a step of determining that the property is high.
The detection method according to claim 1 or 2, wherein the amount of CD91 is the amount of CD91 present on the exosome surface, and the amount of CD317 is the amount of CD317 present on the exosome surface.
The detection method according to any one of claims 1 to 3, further comprising a step of measuring the amount of CEA in a sample collected from a living body.
The detection method according to any one of claims 1 to 4, wherein the lung cancer is stage I or II.
The detection method according to any one of claims 1 to 4, wherein the lung cancer is stage III or IV.
The detection method according to any one of claims 1 to 6, wherein the amount of at least one of the CD91 and CD317 is measured by an ELISA method.
The detection method according to any one of claims 1 to 7, wherein the sample is whole blood, serum or plasma.
A detection kit for detecting lung cancer,
A detection kit comprising at least one of an antibody or peptide probe that recognizes CD91 in a sample collected from a living body and an antibody or peptide probe that recognizes CD317 in a sample collected from a living body.
The detection kit according to claim 9, further comprising a base material to which an antibody for capturing exosomes in the sample is bound.
The detection kit according to claim 10, wherein the antibody for capturing the exosome is an anti-CD9 antibody.
The detection kit according to any one of claims 9 to 11, further comprising an antibody or peptide probe that recognizes CEA in a sample collected from a living body.