Blood astrocyte derived exosome marker and composition for diagnosing Alzheimer's disease

Abstract

The invention provides a blood astrocyte-derived exosome marker for diagnosing Alzheimer's disease and a combination of the blood astrocyte-derived exosome marker. Specifically, the invention provides key nucleic acid markers (including SET, GADME, ADAMTSL3, CP and the like) in exosomes derived from astrocytes in plasma. Compared with classical protein markers in peripheral blood, the nucleic acid markers are more stable in vivo and are not easy to degrade. And the change of the pathological state in the brain of the AD patient can be detected only through blood sampling detection. Compared with a traditional protein marker detection method, the nucleic acid marker can be detected only through RT-qPCR, and a detection platform has universality, is easy to operate and is low in detection cost.

Description

Technical Field

[0001] This invention relates to the field of biotechnology, and more specifically to a blood astrocyte-derived exosome marker and combination for diagnosing Alzheimer's disease. Background Technology

[0002] Traditional diagnostic methods for Alzheimer's disease (AD) mainly rely on magnetic resonance imaging (MRI), positron emission tomography (PET), and cerebrospinal fluid (CSF) biomarker detection. However, most of these methods require invasive lumbar puncture sampling, which has drawbacks such as invasiveness, non-reproducibility, and low patient acceptance, thus limiting the early diagnosis and intervention of AD.

[0003] Therefore, it is crucial to develop novel AD biomarker diagnostic strategies to achieve earlier, more accurate, and non-invasive AD diagnosis, such as blood biomarkers. Summary of the Invention

[0004] The purpose of this invention is to provide a novel method for diagnosing AD using biomarkers, such as blood biomarkers, to achieve earlier, more accurate, and non-invasive diagnosis of AD.

[0005] The first aspect of this invention provides the use of a reagent for detecting target genes or their proteins in exosomes, (i) for use as biomarkers for early Alzheimer's disease (AD) or mild cognitive impairment (MCI); and / or (ii) for use in the preparation of diagnostic reagents, detection reagents, or kits for early Alzheimer's disease (AD) or mild cognitive impairment (MCI). The target genes are selected from the following group: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof.

[0006] In another preferred embodiment, the MOCA score of the patient with early Alzheimer's disease (AD) should be less than 18.

[0007] In another preferred embodiment, the exosomes are astrocyte-derived exosomes.

[0008] In another preferred embodiment, the target gene is derived from a mammal.

[0009] In another preferred embodiment, the target gene is derived from rodents, primates, or humans.

[0010] In another preferred embodiment, the diagnostic reagent includes: a target gene or a protein-specific binding molecule thereof, a target gene or a protein-specific antibody thereof, a target gene-specific primer, a target gene probe, or a chip.

[0011] In another preferred embodiment, the diagnostic reagent is coupled with or carries a detectable marker.

[0012] In another preferred embodiment, the detectable marker is selected from the group consisting of chromophores, chemiluminescent groups, fluorophores, isotopes, enzymes, or combinations thereof.

[0013] In another preferred embodiment, the diagnostic reagent includes: antibodies, primers, probes, sequencing libraries, nucleic acid chips, or protein chips.

[0014] In another preferred embodiment, the diagnostic reagent further includes a pharmaceutically acceptable carrier, diluent, or excipient.

[0015] In another preferred embodiment, the kit also includes a label or instructions.

[0016] In another preferred embodiment, the label or instructions indicate that the kit is used to detect early Alzheimer's disease (AD).

[0017] In another preferred embodiment, the detection is a detection of an ex vivo sample.

[0018] In another preferred embodiment, the diagnosis includes a preliminary assessment.

[0019] In another preferred embodiment, the diagnosis is an early diagnosis.

[0020] In another preferred embodiment, the sample being tested includes a blood sample.

[0021] In another preferred embodiment, the blood sample is selected from the group consisting of blood, plasma, serum, or combinations thereof.

[0022] In another preferred embodiment, the blood sample is a pretreated blood sample.

[0023] In another preferred embodiment, the blood sample is centrifuged plasma.

[0024] In another preferred embodiment, the pretreatment includes the step of separating whole blood to obtain a plasma sample.

[0025] In another preferred embodiment, the plasma sample is collected by centrifugation using an EDTA blood collection tube or a heparin sodium blood collection tube.

[0026] In another preferred embodiment, the detection reagent includes: a reagent for detecting the expression level of target gene mRNA or its protein, wherein the target gene is selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof.

[0027] In another preferred embodiment, the detection reagent comprises: a reagent for detecting the expression level of at least one target gene mRNA or protein in astrocyte-derived exosomes (ADEs), wherein the target gene is selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof.

[0028] In another preferred embodiment, the detection reagent is used to detect the expression level of at least one target gene mRNA or protein in astrocyte-derived exosomes (ADEs) in the plasma of patients with Alzheimer's disease (AD) or mild cognitive impairment (MCI), for the diagnosis and / or monitoring of treatment efficacy for Alzheimer's disease (AD) or mild cognitive impairment (MCI); wherein the target gene is selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof.

[0029] In another preferred embodiment, the diagnosis is to assess the patient's risk of developing early Alzheimer's disease (AD) or mild cognitive impairment (MCI), and to predict whether it is early Alzheimer's disease (AD) or mild cognitive impairment (MCI).

[0030] In another preferred embodiment, the kit is a non-invasive liquid biopsy test kit.

[0031] In another preferred embodiment, the exosomes are separated and purified using a combination of polymer precipitation and immunoaffinity capture.

[0032] In another preferred embodiment, the detection reagent is selected from the group consisting of reagents for detecting the expression level of target gene mRNA or its protein in plasma, wherein the target gene is selected from the group consisting of SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof.

[0033] In a second aspect of the invention, a kit is provided for diagnosing or assisting in the diagnosis of early Alzheimer's disease (AD) or mild cognitive impairment (MCI), the kit comprising reagents for detecting the expression level of at least one target gene mRNA or protein in astrocyte-derived exosomes (ADEs), the target gene being selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof.

[0034] In another preferred embodiment, the kit further comprises reagents for separating ADEs from plasma.

[0035] In another preferred embodiment, the reagent for separating ADEs comprises an anti-GLAST antibody.

[0036] In another preferred embodiment, the reagent for detecting mRNA expression levels comprises one or more of primer pairs, probes, reverse transcriptases, and / or DNA polymerases for RT-qPCR.

[0037] In another preferred embodiment, the kit comprises primer pairs that specifically amplify sequences selected from those shown in SEQ ID NO:1 to SEQ ID NO:8.

[0038] In another preferred embodiment, the kit further includes reagents for detecting internal reference gene mRNA, wherein the internal reference gene is selected from YWHAE and RNPS1.

[0039] In another preferred embodiment, the kit comprises a combination of reagents for detecting the mRNA of the at least two target genes, the combination being selected from the group consisting of: SET and ADAMTSL3; SET and GSDME; SET and CP; ADAMTSL3 and GSDME; ADAMTSL3 and CP; GSDME and CP; SET, ADAMTSL3 and GSDME; SET, ADAMTSL3 and CP; SET, GSDME and CP; ADAMTSL3, GSDME and CP; SET, ADAMTSL3, GSDME and CP.

[0040] In another preferred embodiment, the kit also includes reagents for detecting pTau217 protein.

[0041] In another preferred embodiment, the kit further includes a label or instruction manual indicating that the kit is used for early diagnosis and / or monitoring of treatment efficacy in Alzheimer's disease (AD).

[0042] In another preferred embodiment, the label or instruction manual includes the following information: (1) SET target gene expression level (-Δ rCT) exceeding -2.64, ADAMTSL3 target gene expression level (-Δ rCT) exceeding -2.28, GSDME target gene expression level (-Δ rCT) exceeding -0.65, and CP target gene expression level (-Δ rCT) exceeding -5.97 are interpreted as positive for Alzheimer's disease (AD); SET target gene expression level (-Δ rCT) below -3.14, ADAMTSL3 target gene expression level (-Δ rCT) exceeding -2.22, GSDME target gene expression level (-Δ rCT) exceeding -1.77, and CP target gene expression level (-Δ rCT) exceeding -6.68 are interpreted as negative for Alzheimer's disease (AD); (2) SET target gene expression level (-Δ rCT) exceeding -3.14, ADAMTSL3 target gene expression level (-Δ rCT) exceeding -2.22, GSDME target gene expression level (-Δ rCT) exceeding -1.77, and CP target gene expression level (-Δ rCT) exceeding -6.68 are interpreted as positive for mild cognitive impairment (MCI), while SET target gene expression level (-Δ rCT) below -3.98, ADAMTSL3 target gene expression level (-Δ rCT) exceeding -3.64, GSDME target gene expression level (-Δ rCT) exceeding -2.93, and CP target gene expression level (-Δ rCT) exceeding -7.21 are interpreted as negative for mild cognitive impairment (MCI).

[0043] In another preferred embodiment, the subject of the test is an Alzheimer's disease (AD) patient.

[0044] In another preferred embodiment, the subject of the test is a suspected early-stage Alzheimer's disease (AD) patient.

[0045] In a third aspect of the invention, there is provided the use of a target gene or its protein inhibitor for the preparation of a medicament or pharmaceutical composition for the prevention and / or treatment of early Alzheimer's disease (AD) or mild cognitive impairment (MCI). The target genes are selected from the following group: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof.

[0046] In another preferred embodiment, the inhibitor includes: RNase, DNA expression inhibitor.

[0047] In another preferred embodiment, the RNase comprises an RNase connected to a primer targeting the target gene.

[0048] In another preferred embodiment, the DNA expression inhibitor is an inhibitor that inhibits the gene.

[0049] In a fourth aspect of the invention, a pharmaceutical composition is provided, the pharmaceutical composition comprising: (C1) Target gene or its protein inhibitor; wherein the target gene is selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or a combination thereof; (C2) Optional other medications for treating Alzheimer's disease (AD); and (C3) Pharmaceutically acceptable carriers, diluents or excipients.

[0050] In another preferred embodiment, the other drugs for treating Alzheimer's disease (AD) include donepezil, rivastigmine, galantamine, lencanemab, donepemab, and adukamamarab.

[0051] In another preferred embodiment, the weight ratio of component (C1) to component (C2) is in the range of 100:1-0.01:1, more preferably 10:1-0.1:1, and even more preferably 2:1-0.5:1.

[0052] In another preferred embodiment, the content of component (C1) in the pharmaceutical composition is 1%-99%, more preferably 10%-90%, and even more preferably 30%-70%.

[0053] In another preferred embodiment, the content of component (C2) in the pharmaceutical composition is 1%-99%, more preferably 10%-90%, and even more preferably 30%-70%.

[0054] In another preferred embodiment, the dosage form of the pharmaceutical composition includes: an injectable dosage form and an oral dosage form.

[0055] In another preferred embodiment, the oral dosage form includes: tablets, capsules, films, and granules.

[0056] In another preferred embodiment, the dosage form of the pharmaceutical composition includes: a sustained-release formulation and a non-sustained-release formulation.

[0057] In a fifth aspect of the invention, a medicine box is provided, the medicine box comprising: (1) A first container, and a target gene or its protein inhibitor located in the first container; wherein the target gene is selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or a combination thereof; (2) A second container, and a medication for treating Alzheimer's disease (AD) or mild cognitive impairment (MCI) located in the second container.

[0058] In another preferred embodiment, the first container and the second container may be the same or different containers.

[0059] In another preferred embodiment, the medicine box also includes a label or instructions.

[0060] In another preferred embodiment, the label or instruction manual indicates that the target gene inhibitor is used in combination with a drug for treating Alzheimer's disease (AD) or mild cognitive impairment (MCI) to prevent / slow down Alzheimer's disease (AD) or mild cognitive impairment (MCI).

[0061] In another preferred embodiment, the MOCA score for mild cognitive impairment (MCI) should be between 18 and 25.

[0062] In a sixth aspect of the invention, a method is provided for determining whether a subject is at risk of Alzheimer's disease (AD) or mild cognitive impairment (MCI), the method comprising the following steps: 1) Detection steps: Detect the subject's target gene or its protein; 2) Comparison step: Compare the expression levels of the target gene or its protein detected in step 1) with the reference value; 3) Judgment step: If the expression level of the target gene or its protein detected in step 1) is significantly changed relative to the reference value, then the conclusion is that the subject has a risk of early Alzheimer's disease (AD) or mild cognitive impairment (MCI). The target genes are selected from the following group: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof; The reference value is the expression level of the corresponding exosomal target gene or its protein in normal individuals.

[0063] In another preferred embodiment, the phrase "the expression level of the detected target gene or its protein has changed significantly relative to the reference value" means that, compared with the reference value, the expression level of the target gene or its protein selected from the following group is increased: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or a combination thereof.

[0064] In another preferred embodiment, the target gene is a combination of SET and ADAMTSL3.

[0065] In another preferred embodiment, the target gene is a combination of SET and GSDME.

[0066] In another preferred embodiment, the target gene is a combination of SET and CP.

[0067] In another preferred embodiment, the target gene is a combination of ADAMTSL3 and GSDME.

[0068] In another preferred embodiment, the target gene is a combination of ADAMTSL3 and CP.

[0069] In another preferred embodiment, the target gene is a combination of GSDME and CP.

[0070] In another preferred embodiment, the target gene is a combination of SET, ADAMTSL3 and GSDME.

[0071] In another preferred embodiment, the target gene is a combination of SET, GSDME and CP.

[0072] In another preferred embodiment, the target gene is a combination of SET, ADAMTSL3 and CP.

[0073] In another preferred embodiment, the target gene is a combination of ADAMTSL3, GSDME and CP.

[0074] In another preferred embodiment, the target gene is a combination of SET, ADAMTSL3, GSDME and CP.

[0075] In a seventh aspect of the invention, a combination of biomarkers for diagnosing early Alzheimer's disease (AD) or mild cognitive impairment (MCI) is provided, the combination comprising at least two mRNA biomarkers selected from astrocyte-derived exosomes (ADEs), the mRNA biomarkers being selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, or CP.

[0076] In another preferred embodiment, the biomarker combination is selected from the group consisting of: SET and ADAMTSL3; SET and GSDME; SET and CP; ADAMTSL3 and GSDME; ADAMTSL3 and CP; GSDME and CP; SET, ADAMTSL3 and GSDME; SET, ADAMTSL3 and CP; SET, GSDME and CP; ADAMTSL3, GSDME and CP; SET, ADAMTSL3, GSDME and CP.

[0077] In another preferred embodiment, the biomarker combination is used for detection via RT-qPCR.

[0078] In another preferred embodiment, the biomarker combination is used in conjunction with the pTau217 protein.

[0079] In an eighth aspect of the invention, a primer pair combination for detecting a combination of biomarkers described in the seventh aspect of the invention is provided, the primer pair combination comprising primer pairs capable of specifically amplifying each mRNA biomarker target sequence in the combination of biomarkers.

[0080] In another preferred embodiment, the primer pair used to amplify the SET comprises the sequences shown in SEQ ID NO:1 and SEQ ID NO:2.

[0081] In another preferred embodiment, the primer pair used to amplify ADAMTSL3 comprises the sequences shown in SEQ ID NO:3 and SEQ ID NO:4.

[0082] In another preferred embodiment, the primer pair used to amplify GSDME comprises the sequences shown in SEQ ID NO:5 and SEQ ID NO:6.

[0083] In another preferred embodiment, the primer pair used to amplify CP comprises the sequences shown in SEQ ID NO:7 and SEQ ID NO:8.

[0084] In a ninth aspect of the invention, a method is provided for constructing a model for diagnosing early Alzheimer's disease (AD) or mild cognitive impairment (MCI), the method comprising: (1) Obtain a set of training samples from individuals with known clinical status, including AD patients, MCI patients and cognitively unimpaired (CU) individuals; (2) For each training sample, the expression level of an mRNA marker detected from plasma astrocyte-derived exosomes (ADEs) was determined, the mRNA marker being selected from one or more combinations of the following: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME or CP; (3) Based on the expression level of the mRNA marker, combined with the individual's age, sex and APOE ε4 genotype information, a classification model is trained by machine learning algorithm to distinguish different clinical states.

[0085] In another preferred embodiment, the machine learning algorithm is logistic regression.

[0086] In another preferred embodiment, the mRNA markers are SET and GSDME.

[0087] In another preferred embodiment, the method further includes assigning class weights to samples of different clinical states during training to address data imbalance.

[0088] In another preferred embodiment, the method further includes using the classification model to generate receiver operating characteristic (ROC) curves and evaluating model performance based on the area under the curve (AUC).

[0089] In another preferred embodiment, the method further includes using the model to predict the risk of AD or MCI in samples from new subjects.

[0090] In a tenth aspect of the invention, a system is provided for diagnosing or assisting in the diagnosis of early Alzheimer's disease (AD) or mild cognitive impairment (MCI), the system comprising: (1) Sample processing module, configured to isolate astrocyte-derived exosomes (ADEs) from plasma samples of subjects. (2) A detection module configured to detect the expression level of mRNA of at least one target gene selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME or CP; (3) An analysis module configured to generate results for assessing the risk or likelihood of the subject having AD or MCI based on the mRNA expression level of the at least one target gene.

[0091] In another preferred embodiment, the sample processing module includes an immunoaffinity purification unit based on an anti-GLAST antibody.

[0092] In another preferred embodiment, the detection module includes a device for performing RT-qPCR.

[0093] In another preferred embodiment, the analysis module is configured to normalize the expression level of the target gene with the expression levels of one or more internal reference genes selected from YWHAE and RNPS1.

[0094] In another preferred embodiment, the analysis module is further configured to perform an assessment in conjunction with the subject's age, sex, and / or APOEε4 genotype information.

[0095] In another preferred embodiment, the analysis module is configured to perform evaluation based on a logistic regression model that includes a combination of biomarkers containing at least one target gene.

[0096] In another preferred embodiment, the system further includes: (4) a reporting module for outputting a diagnostic or risk assessment report.

[0097] In an eleventh aspect of the present invention, a method for diagnosing or assisting in the diagnosis of early Alzheimer's disease (AD) or mild cognitive impairment (MCI) is provided, the method comprising: (a) Obtaining blood samples from the subjects; (b) Isolate astrocyte-derived exosomes (ADEs) from the plasma of the blood sample; (c) Detect the expression levels of the mRNA of the target genes in the ADEs; and (d) Assess the risk or likelihood of the subject having AD or MCI based on the mRNA expression level of the target gene; The target genes include one or more combinations selected from the following group: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, or CP.

[0098] In another preferred embodiment, the target gene includes one or more combinations selected from the group consisting of SET, ADAMTSL3, GSDME, and CP.

[0099] In another preferred embodiment, the method is used to diagnose or assist in the diagnosis of early Alzheimer's disease.

[0100] In another preferred embodiment, the blood sample is a peripheral blood sample.

[0101] In another preferred embodiment, the separation of ADEs from plasma includes: immunoaffinity capture using an anti-GLAST antibody.

[0102] In another preferred embodiment, the detection of mRNA expression levels is performed by reverse transcription quantitative polymerase chain reaction (RT-qPCR).

[0103] In another preferred embodiment, the RT-qPCR is performed using primer pairs specific to the target gene.

[0104] In another preferred embodiment, the method further includes comparing the mRNA expression level of the target gene with the mRNA expression level of an internal reference gene, wherein the internal reference gene is selected from YWHAE, RNPS1, or a combination thereof.

[0105] In another preferred embodiment, step (d) of evaluating based on mRNA expression levels includes comparing the expression levels with a predetermined threshold or reference range.

[0106] In another preferred embodiment, the method further includes detecting the level of pTau217 protein in the ADEs and evaluating it in conjunction with the mRNA expression level of the target gene.

[0107] It should be understood that, within the scope of this invention, the above-described technical features of this invention and the technical features specifically described below (such as in the embodiments) can be combined with each other to form new or preferred technical solutions. Due to space limitations, they will not be described in detail here. Attached Figure Description

[0108] Figure 1The figures shown are the characterization and identification results of ADEs isolated in the embodiments of the present invention. A is a TEM image showing the morphological characteristics of ADEs in the NC and AD groups of normal healthy individuals. B is an NTA measurement showing the size distribution range of ADEs in the NC and AD groups. C is a Western blot of exosome positive markers ALIX, CD81, and TSG101, negative marker CNX, and ADE-specific expression markers GLAST and GFAP.

[0109] Figure 2 The figure shown is a graph of the results of detecting the relative expression levels of SET (A), GADME (B), ADAMTSL3 (C), and CP (D) in plasma ADEs in different clinical groups (such as controls with no cognitive impairment (CU), individuals with mild cognitive impairment (MCI), and Alzheimer's disease (AD)) in embodiments of the present invention.

[0110] Figure 3 The figures shown are ROC curve results of SET, GADME, ADAMTSL3 and CP under different combinations in the embodiments of the present invention. AD are ROC curve results of different combinations.

[0111] Figure 4 The figures shown are ROC curve results of SET, GADME, ADAMTSL3 and CP and ptau217 in different combinations in the embodiments of the present invention. AE are ROC curve results of different combinations. Detailed Implementation

[0112] Through extensive and in-depth experiments, the inventors have for the first time discovered key nucleic acid biomarkers (including SET, GADME, ADAMTS L3, CP, or combinations thereof) in astrocyte-derived exosomes in plasma. Compared to classic protein biomarkers in peripheral blood (such as Aβ42, ptau181, ptau217, etc.), these nucleic acid biomarkers are more stable in vivo and less prone to degradation. This makes it possible to detect pathological changes in the brain of AD patients simply by blood sampling. Furthermore, compared to traditional protein biomarker detection methods, the nucleic acid biomarkers in this invention can be detected using only RT-qPCR, and the detection platform is universal (qPCR instrument), simple to operate, and low in detection cost. Based on this, this invention was completed.

[0113] the term Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0114] As used herein, “including” or “containing” includes “comprising,” “mainly composed of,” “substantially composed of,” and “composed of”; “mainly composed of,” “substantially composed of,” and “composed of” are subordinate concepts of “containing,” “having,” or “including.”

[0115] As used herein, the terms "target gene," "gene," and "nucleic acid biomarker" are used interchangeably. They all refer to the biomarkers described in the first aspect of this invention for detecting Alzheimer's disease (AD) or mild cognitive impairment (MCI).

[0116] Early Alzheimer's disease (AD) Clinically, the distinction between mild cognitive impairment (MCI) and early Alzheimer's disease (AD) primarily relies on the degree and progression of cognitive decline. MCI typically presents as mild memory or other cognitive decline (18 ≤ MOCA ≤ 25), but patients can still perform daily activities independently, and symptoms may plateau or improve. Early AD, on the other hand, is characterized by persistent and significant cognitive decline (MOCA < 18), particularly in memory and executive functions, and begins to impact daily life. Early AD symptoms are often accompanied by β-amyloid protein accumulation and abnormally elevated levels of proteins such as pTau217 and GFAP in cerebrospinal fluid / blood, as well as abnormally low levels of Aβ42 protein in cerebrospinal fluid / blood. Imaging and cerebrospinal fluid analysis can aid in diagnosis. Mild cognitive impairment (MCI) may not necessarily show abnormalities in these marker levels. Currently, there is no completely unified standard for distinguishing between MCI and early AD; diagnosis mainly relies on clinical assessment, cognitive testing, and imaging examinations. The use of biomarkers is becoming an important auxiliary diagnostic tool, but certain uncertainties remain.

[0117] Exosomes (EVs) Since Alzheimer's disease (AD) primarily reflects abnormal symptoms of the central nervous system (CNS), there are certain differences between the normal blood environment and the brain's biochemical environment. CNS-derived exosomes can carry disease-related substances across the blood-brain barrier (BBB) ​​into the bloodstream, making them a highly promising biomarker carrier.

[0118] Exosomes (EVs) are important carriers of bioactive substances, participating in homeostasis and intercellular communication by encapsulating biomolecules such as proteins, RNA, and lipids. The contents of exosomes from different cell sources can reflect the physiological state of specific cells, thus characterizing the health status of cells or organisms. Astrocyte-derived exosomes (ADEs) play a key role in the spread of neuropathology and the progression of neurodegenerative diseases, and can even act as carriers of protein aggregates such as Aβ in the course of Alzheimer's disease (AD). ADEs extracted from peripheral blood can more accurately reflect the dysregulation of biomolecule expression under AD pathological conditions.

[0119] While exosome miRNAs and proteins have been extensively studied in AD diagnosis, research on exosome mRNAs remains to be explored. Studies have shown that brain mRNA expression analysis can identify differentially expressed genes between healthy controls and AD patients, aiding in AD disease progression monitoring and biomarker discovery. Since adenosine desensitizers (ADEs) can protect their encapsulated mRNAs from ribonuclease degradation, research on ADEs-mRNA holds promise for advancing the development of novel AD diagnostic tools and treatment strategies.

[0120] The main advantages of this invention include: (1) The nucleic acid markers in the exosomes derived from astrocytes in plasma found by this invention have the protective effect of the exosome membrane and are more stable and less prone to degradation in vivo compared with classic protein markers in peripheral blood (such as Aβ42, ptau181, ptau217, etc.).

[0121] (2) The astrocyte-derived exosomes provided by the present invention can carry some components directly related to AD pathology across the blood-brain barrier into the peripheral system while protecting their contents from degradation, which makes it possible to detect changes in the pathological state in the brain of AD patients by blood sampling alone.

[0122] (3) Compared with traditional protein biomarker detection methods (Simoa, chemiluminescence immunoassay analyzer, etc.), the nucleic acid biomarkers in this invention can be detected by RT-qPCR alone. The detection platform is universal (qPCR instrument), simple to operate, and has low detection cost.

[0123] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Experimental methods in the following embodiments, unless otherwise specified, are generally performed under conventional conditions as described in Sambrook et al., Molecular Cloning: A Laboratory Manual (New York: Cold Spring Harbor Laboratory Press, 1989), or as recommended by the manufacturer. Percentages and parts are by weight unless otherwise stated. Unless otherwise specified, all experimental materials and reagents involved in this invention are commercially available.

[0124] Experimental Materials and Methods 1. Study population and blood collection and processing Based on the clinical characteristics of the subjects shown in Table 1, participants were required to fast for 12 hours before the experiment. Peripheral blood was collected and stored in EDTA tubes. Plasma was first separated by centrifugation at 3000 rpm for 10 minutes at room temperature. The plasma was then transferred to sterile 1.5 mL centrifuge tubes and centrifuged at 10000 g for 20 minutes at 4°C to remove impurities. The supernatant was transferred to new centrifuge tubes, aliquoted into 200 μL portions, and finally stored at -80°C for later use.

[0125] Table 1: Clinical characteristics of the study subjects Note: AD: Clinically diagnosed Alzheimer's Diseases. MCI: Mild Cognitive Impairment. CU: Cognitively unimpaired controls. MOCA: Montreal Cognitive Assessment (MoCA) score, total 30 points. A score ≥26 indicates normal cognitive function, 18-25 indicates mild cognitive impairment (MCI), and <18 indicates moderate or severe cognitive impairment, considered early AD. APOEε4 carrier: The percentage of individuals in each group who are APOEε4 positive. SUVR (Standardized Uptake Value Ratio): A semi-quantitative indicator used in positron emission tomography (PET) to quantify the degree of β-amyloid (Aβ) deposition in the brain. Aβ42: The protein concentration of Aβ42 in the sample plasma, pg / mL. ptau217: Protein concentration of ptau217 in plasma sample, pg / mL. GFAP: Protein concentration of Aβ42 in plasma sample, pg / mL. Expression levels of novel target genes (such as ADAMTSL3, GSDME, or CP) are expressed as -Δ reference CT, where -Δ rCT is Δ reference CT = ΔCT. 目的基因 - ΔCT 参考基因 The abbreviation for .

[0126] 2. Isolation of astrocyte-derived exosomes from plasma After thawing plasma samples stored at -80°C at 37°C, 1 / 5 volume of ExoQuick® reagent (EXOQ; System Biosciences, Inc., Mountain View, CA, USA) was added to every 200 μL of plasma. The mixture was incubated at 4°C for 1 hour, then centrifuged at 1500 × g for 20 min at 4°C, and the supernatant was removed. The exosomes were re-selected using PBS buffer containing protease and phosphatase inhibitors to obtain a total plasma exosome suspension.

[0127] To enrich ADEs, resuspended total plasma exosomes were first incubated with 2.3 μL of mouse anti-human GLAST biotinylated antibody (Miltenyi Biotec) in 50 μL of 3% BSA buffer at 4°C for 1 hour by rotation. Then, 10 μL of room-temperature equilibrated streptavidin-agarose UltraLink resin (Thermo Fisher Scientific) was added, and incubation continued at 4°C for 1.5 hours. After incubation, the supernatant was removed by centrifugation at 800×g for 10 minutes at 4°C, and the precipitate was resuspended in 100 μL of pre-chilled 0.05 M glycine hydrochloride buffer (pH 3). After centrifugation at 4000×g for 10 minutes at 4°C, the supernatant was transferred to a new Eppendorf tube containing 10 μL of 1 M Tris-HCl (pH 8.0), and the mixture was thoroughly mixed to obtain the ADEs suspension.

[0128] 3. Characterization of astrocyte exosomes (ADEs) 3.1 Transmission Electron Microscopy (TEM) 10 μL of appropriately diluted ADEs suspension was added to a copper grid. After standing for 1 minute, excess liquid at the edges was blotted off with filter paper, and the grid was allowed to air dry. Then, a suitable amount of 2% phosphotungstenate solution was added for staining. After absorbing excess stain, the grid was washed with deionized water and blotted dry. Once the copper grid was completely dry, imaging and photographs were taken using a JEM-2100 transmission electron microscope at an accelerating voltage of 80 kV.

[0129] 3.2 Nanoparticle Tracking Analysis (NTA) Exosome particle size and concentration were determined using a NanoSight NS300 nanoparticle tracking analyzer (Marvin Panaco, UK) from Shanghai Youmei Biotechnology Co., Ltd., in conjunction with NTA 3.4.4 software. The isolated exosome samples were appropriately diluted with 1×PBS buffer (Israel Bioindustrial Corporation) before loading and analysis. The temperature was maintained within the range of 20-30℃ throughout the experiment.

[0130] 3.3 Western blotting to identify characteristic proteins of ADEs The ADEs suspension was mixed with an equal volume of M-PER mammalian protein extraction reagent (containing protease and phosphatase inhibitors), lysed on ice for 5 minutes, and the protein concentration in the lysate was determined using the Omni-Easy™ BCA Protein Quantification Kit (EpiZyme, Shanghai) to prepare the sample. 15 μg of protein sample was separated by SDS-PAGE electrophoresis and transferred to a PVDF membrane (Millipore, USA). After blocking with rapid blocking buffer (EpiZyme, Shanghai), the membrane was incubated overnight at 4°C with diluted primary antibodies ALIX, CD81, cadherin (ABclonal, Wuhan), TSG101 (Servicebio, Wuhan), GLAST (ABclonal, Wuhan), and GFAP (Abcam, ab4674, UK). After washing three times with TBST, the membrane was incubated with HRP-labeled secondary antibody at room temperature for 1 hour, washed again, and developed using the Omni-ECL™ High-Sensitive Chemiluminescence Kit (EpiZyme, Shanghai). Finally, images were acquired using a Tanon chemiluminescence imaging system.

[0131] 4. RNA isolation and cDNA synthesis The ADEs suspension was lysed with an appropriate volume of Trizol reagent (Invitrogen), followed by extraction of total RNA according to the manufacturer's instructions. RNA concentration was determined using a NanoDrop 2000 (Thermo Fisher Scientific, USA), and purity was assessed by the 260 / 280 nm absorbance ratio. RNA integrity was detected using a Qsep100 assay. Reverse transcription was performed using the TransScript® One-Step gDNA Removal and cDNA Synthesis Kit (TransGen Biotech).

[0132] 5. RT-qPCR polymerase chain reaction The mRNA expression levels of candidate reference genes were detected using PerfectStart Green qPCR SuperMix (containing Universal Passive ReferenceDye, Transgene Biotech). Detection was performed using a LightCycler@480 real-time quantitative PCR system. The reaction program was as follows: 95℃ pre-denaturation for 30 seconds; followed by 50 cycles of 95℃ denaturation for 5 seconds, 60℃ annealing for 30 seconds, and 72℃ extension for 10 seconds; a final extension at 72℃ for 5 minutes, followed by storage at 4℃. All samples were performed in triplicate, and a template-free control was used to exclude nucleic acid contamination and primer dimer interference.

[0133] The gene primer sequences used in this experiment are shown in Table 2 below: Table 2 6. Data Analysis Use 2^(-Δ reference CT, Δ reference CT=ΔCT 目的基因 - ΔCT 参考基因 The corresponding genes are compared to the internal reference (the internal reference genes are RNPS1 and YWHAE, ΔCT). 参考基因 =(ΔCT RNPS1 +ΔCT YWHAE Statistical analysis was performed on the expression levels of AV45 (Aβ) and Tau (Tau), but the gene expression levels exhibited an exponential distribution. Subsequent analysis used log(expression level) for normalization, i.e., -Δ reference CT. A logistic regression model was constructed using a stratified five-fold cross-validation strategy and a class weight adjustment method to predict Aβ and Tau PET imaging status. The purpose of introducing class weights was to address data imbalance by assigning higher weights to minority classes, thereby enhancing the model's ability to identify AV45-positive cases.

[0134] New models were constructed by adding different metrics to the basic model (age + sex + APOE genotype). Cross-validation was used to evaluate the area under the curve (AUC), accuracy, sensitivity, specificity, F1 score, and recall of each model, and the corresponding 95% confidence intervals were calculated. Receiver operating characteristic (ROC) curves were plotted for model comparison and clinical utility assessment, while forest plots were used to display the AUC (95% confidence interval) of each model. Given the large sample size, these continuous variables were assumed to follow a normal distribution, and parametric analysis methods were employed. All statistical analyses were performed using Python 3.11.4, R 4.3.2, and SPSS version 27.

[0135] Example 1: Extraction and Identification of ADEs After successfully enriching ADEs from total EVs in the plasma of healthy controls and AD patients using the aforementioned method, their typical exosome morphology was confirmed by transmission electron microscopy (TEM). Figure 1 A). Nanoparticle tracking analysis (NTA) showed that the peak particle size distribution of ADEs in the healthy group and the AD group were 82 nm and 106 nm, respectively. Figure 1 B). Western blot analysis showed that ADEs were positively expressed with exosome markers ALIX, CD81, and TSG101, while cadherin (CNX) was negative, consistent with exosome characteristics. Figure 1 C). Notably, the astrocyte marker proteins GLAST and GFAP were specifically enriched in the ADEs, confirming the successful isolation of astrocyte-derived exosomes from plasma.

[0136] Example 2: Gene expression of SET, ADMATSL3, GADME, and CP in different plasma ADEs In this embodiment, quantitative RT-PCR technology was used to quantify the plasma expression levels of four astrocyte-enriched exosomal transcripts (GSDME, SET, ADAMTSL3, and CP) in different clinical groups (such as controls with no cognitive impairment (CU), patients with mild cognitive impairment (MCI), and patients with Alzheimer's disease (AD)).

[0137] Each transcript exhibited a unique expression pattern across different clinical groups. GSDME was the most discriminative marker, with significantly higher expression levels in the AD (-0.65±1.40) and MCI (-1.77±1.82) groups compared to the CU (-2.93±2.69) group (p values ​​<0.0001 and 0.0154, respectively). Figure 2 B). SET transcript levels showed a similar trend, gradually increasing from CU (-3.98±1.88) to AD patients (-2.64±2.01; p=0.0047; overall p=0.0067). Figure 2 A). ADAMTSL3 expression was also lowest in the CU group (-3.64±3.13), compared to the AD group (-2.28±0.73; p=0.0214) and the MCI group (-2.22±0.70; p=0.0181; overall p=0.0062). Figure 2 Compared to C), all were significantly lower. Similarly, CP transcripts gradually increased from CU (-7.21±1.66) to AD patients (-5.97±1.53; p=0.0004; overall p=0.0006). Figure 2 D).

[0138] The above data suggest that plasma ADE transcripts can distinguish AD and MCI patients from cognitively healthy individuals, supporting their application value as a biomarker of disease state.

[0139] Example 3: ROC curves of different targets and combinations To evaluate whether ADE biomarkers can improve the accuracy of diagnostic classification, this embodiment combines demographic factors (age, sex), genetic risk (APOE genotype), and multiple biomarker combinations to construct a baseline model (age, gender, APOE) (abbreviated as Model 0), and 15 logistic regression models containing different target genes (such as SET, GADME, ADAMTSL3, CP, or combinations thereof). The specific model numbers and their components are shown below: Model 0: age, gender, APOE; Model 1: age, gender, APOE, SET; Model 2: age, gender, APOE, ADAMTSL3; Model 3: age, gender, APOE, GADME; Model 4: age, gender, APOE, CP; Model 5: age, gender, APOE, SET+ADAMTSL3; Model 6: age, gender, APOE, SET+ GSDME; Model 7: age, gender, APOE, SET+ CP; Model 8: age, gender, APOE, ADAMTSL3+ GSDME; Model 9: age, gender, APOE, ADAMTSL3+ CP; Model 10: age, gender, APOE, GSDME + CP; Model 11: age, gender, APOE, SET+ADAMTSL3+GSDME; Model 12: age, gender, APOE, SET+ADAMTSL3+CP; Model 13: age, gender, APOE, SET+GSDME+CP; Model 14: age, gender, APOE, ADAMTSL3+GSDME+CP; Model 15: age, gender, APOE, SET+ADAMTSL3+GSDME+CP; Model 16: age, gender, APOE, GFAP.

[0140] ROC analysis results are as follows: Figure 3 As shown in AD, the AUC values, clinical sensitivity, clinical specificity, and corresponding P values ​​of different targets and their combinations are summarized in Table 3.

[0141] Table 3 Note: AUC is the area under the ROC curve, typically used to measure the overall performance of a classification model. AUC is a dimensionless value ranging from 0 to 1. Specificity reflects the model's ability to identify negative samples, ranging from 0 to 1. Sensitivity reflects the model's ability to identify positive samples, ranging from 0 to 1. The P-value is the between-group P-value between the CU group and the mixed MCI and AD group.

[0142] Table 3 shows that, except for the CP single-target (see Model 4) with an AUC value of 0.797, the other targets (such as SET, ADAMTSL3, or GSDME) and their combinations exhibited good performance in terms of AUC values, clinical sensitivity, and clinical specificity, with AUC values ​​all >0.80. These results indicate that ADE biomarkers (such as SET, ADAMTSL3, or GSDME, or their combinations and their groups with CP, such as any of the biomarkers shown in Models 1-3 and 5-15) have good diagnostic capabilities for early AD, while also possessing high specificity and sensitivity.

[0143] Example 4: ROC curves of combined diagnosis of different targets and combinations with plasma p-tau217 To explore whether new targets and their combinations can improve the diagnostic efficacy of classic clinical indicators, this invention still uses age, gender, and APOE as the baseline Basic model (Model 0p). The combination of 15 logistic regression models from Example 3 was used in conjunction with plasma p-tau217 to plot diagnostic efficacy curves. The combined model numbers are: Model 0: age, gender, APOE; Model 0p: age, gender, APOE, p-tau217; Model 1p: age, gender, APOE, SET+ p-tau217; Model 2p: age, gender, APOE, ADAMTSL3+ p-tau217; Model 3p: age, gender, APOE, GADME+ p-tau217; Model 4p: age, gender, APOE, CP+ p-tau217; Model 5p: age, gender, APOE, SET+ADAMTSL3+ p-tau217; Model 6p: age, gender, APOE, SET+ GSDME+ p-tau217; Model 7p: age, gender, APOE, SET+ CP+ p-tau217; Model 8p: age, gender, APOE, ADAMTSL3+ GSDME+ p-tau217; Model 9p: age, gender, APOE, ADAMTSL3+ CP+ p-tau217; Model 10p: age, gender, APOE, GSDME + CP + p-tau217; Model 11p: age, gender, APOE, SET+ADAMTSL3+GSDME+ p-tau217; Model 12p: age, gender, APOE, SET+ADAMTSL3+CP+ p-tau217; Model 13p: age, gender, APOE, SET+GSDME+CP+ p-tau217; Model 14p: age, gender, APOE, ADAMTSL3+GSDME+CP+ p-tau217; Model 15p: age, gender, APOE, SET+ADAMTSL3+GSDME+CP+ p-tau217;.

[0144] Model 16: age, gender, APOE, GFAP.

[0145] Its ROC curve is shown below. Figure 4 AE. We summarize the AUC, clinical sensitivity, clinical specificity, and corresponding P-value for each combination that can be combined with p-tau217 in Table 4.

[0146] Table 4 Note: AUC is the area under the ROC curve, typically used to measure the overall performance of a classification model. AUC is a dimensionless value ranging from 0 to 1. Specificity reflects the model's ability to identify negative samples, ranging from 0 to 1. Sensitivity reflects the model's ability to identify positive samples, ranging from 0 to 1. The P-value is the between-group P-value between the CU group and the mixed MCI and AD group.

[0147] As shown in Table 4, the results indicate that the combination of new targets (i.e., ADE biomarkers, such as SET, ADAMTSL3, GSDME or CP, or combinations thereof, such as any of the biomarkers shown in Model 1-15) with plasma p-tau217 can improve the AUC value of clinical diagnosis and also play a role in improving the sensitivity and specificity of diagnosis.

[0148] In addition, compared with the traditional AD inflammatory marker GFAP, the new targets and their different combinations also showed higher AUC values, clinical sensitivity, and clinical specificity.

[0149] discuss The nuclear protein encoded by the SET gene, through aberrant phosphorylation (e.g., at the Ser9 site) or hematoxylinization modification, retains cytoplasm, inhibits PP2A activity, drives tau hyperphosphorylation and Aβ toxicity, and promotes neuronal apoptosis. Furthermore, SET phosphorylation (e.g., S9E mutation) can activate the p53 pathway, inducing neuronal apoptosis and disrupting synaptic function. GSDME (Gasdermin E), a member of the Gasdermin family, encodes a protein that participates in the pathological process of Alzheimer's disease (AD) by mediating pyroptosis. Studies have shown that Aβ deposition in the AD brain can induce active caspase-3 in neurons to cleave GSDME, releasing the N-terminal domain with pore-forming activity, transforming apoptosis into inflammatory pyroptosis, leading to neuroinflammation (e.g., IL-1β and IL-18 release) and neuronal damage. Animal models show that GSDME knockdown can alleviate cognitive impairment and microglia activation in APP23 / PS45 mice. In addition, aberrant GSDME expression may exacerbate neurodegeneration by regulating axonal and mitochondrial damage. The secreted protein encoded by the ADAMTSL3 (ADAMTS-like 3) gene participates in the neuroinflammatory process by regulating extracellular matrix (ECM) homeostasis.

[0150] Studies have shown that ADAMTSL3 may affect the activation state of astrocytes by regulating the TGF-β signaling pathway, promoting the release of pro-inflammatory factors such as IL-1β and TNF-α, and exacerbating Aβ-induced neuroinflammation in Alzheimer's disease (AD). Furthermore, the interaction between ADAMTSL3 and the complement component C1q may amplify the inflammatory response through microglia-astroglia crosstalk, accelerating neuronal damage. The CP gene encodes ceruloplasmin, which participates in neuroinflammation and the pathological process of AD by regulating iron metabolism and antioxidant function. Studies have shown that the loss of CP expression in astrocytes leads to abnormal iron accumulation in the brain, causing oligodendrocyte maturation disorders and myelin damage, thereby exacerbating neurodegeneration.

[0151] Regarding the diagnostic criteria for Alzheimer's disease (AD), the updated NIA-AA guidelines in 2024 clearly state that the ATN framework has been updated to ATX(N), where X includes other biomarkers such as I (inflammation / immunity), V (vascular brain injury), and S (α-synuclein pathology), which helps to make more accurate judgments on pathological changes in different individuals. The new targets involved in this patent—SET, GADME, ADAMTS L3, and CP—belong to the I (inflammation / immunity) class of biomarkers and are superior to plasma GFAP in terms of current clinical diagnostic efficacy, operational procedures, and cost. They are expected to become new potential I (inflammation / immunity) biomarkers to replace GFAP.

[0152] All documents mentioned in this invention are incorporated herein by reference as if each document were individually incorporated by reference. Furthermore, it should be understood that after reading the foregoing teachings of this invention, those skilled in the art can make various alterations or modifications to this invention, and these equivalent forms also fall within the scope defined by the appended claims.

Claims

1. The use of a reagent for detecting a target gene or its protein in an exosome, (i) for use as a biomarker for early Alzheimer's disease (AD) or mild cognitive impairment (MCI); and / or (ii) for use in the preparation of diagnostic reagents, detection reagents, or kits for early Alzheimer's disease (AD) or mild cognitive impairment (MCI). in, The target genes are selected from the following group: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof.

2. The use as described in claim 1, characterized in that, The exosomes are astrocyte-derived exosomes.

3. A kit for diagnosing or assisting in the diagnosis of early Alzheimer's disease (AD) or mild cognitive impairment (MCI), characterized in that, The kit contains reagents for detecting the expression level of at least one target gene mRNA or protein in astrocyte-derived exosomes (ADEs), wherein the target gene is selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof.

4. The reagent kit according to claim 3, characterized in that, The kit is a non-invasive liquid biopsy detection kit.

5. Use of a target gene or its protein inhibitor for the preparation of a medicament or pharmaceutical composition for the prevention and / or treatment of early Alzheimer's disease (AD); wherein, The target genes are selected from the following group: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or combinations thereof.

6. A medicine box, the medicine box comprising: (1) A first container, and a target gene or its protein inhibitor located in the first container; wherein the target gene is selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, CP, or a combination thereof; (2) A second container, and a medication for treating Alzheimer's disease (AD) or mild cognitive impairment (MCI) located in the second container.

7. A combination of biomarkers for diagnosing early Alzheimer's disease (AD) or mild cognitive impairment (MCI), characterized in that, The combination comprises at least two mRNA markers selected from astrocyte-derived exosomes (ADEs) and said mRNA markers are selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME, or CP.

8. A primer pair combination, characterized in that, The primer pair combination comprises primer pairs capable of specifically amplifying the target sequence of each mRNA biomarker in the biomarker combination of claim 7.

9. A method for constructing a model for diagnosing early Alzheimer's disease (AD) or mild cognitive impairment (MCI), characterized in that, The method includes: (1) Obtain a set of training samples from individuals with known clinical status, including AD patients, MCI patients and cognitively unimpaired (CU) individuals; (2) For each training sample, the expression level of an mRNA marker detected from plasma astrocyte-derived exosomes (ADEs) was determined, the mRNA marker being selected from one or more combinations of the following: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME or CP; (3) Based on the expression level of the mRNA marker, combined with the individual's age, sex and APOE ε4 genotype information, a classification model is trained by machine learning algorithm to distinguish different clinical states.

10. A system for diagnosing or assisting in the diagnosis of early Alzheimer's disease (AD) or mild cognitive impairment (MCI), characterized in that, The system includes: (1) Sample processing module, configured to isolate astrocyte-derived exosomes (ADEs) from plasma samples of subjects. (2) A detection module configured to detect the expression level of mRNA of at least one target gene selected from the group consisting of: SET, MTTP, TAF3, SLC27A6, PLCG2, PAPOLA, ADAMTSL3, RTN1, GSDME or CP; (3) An analysis module configured to generate results for assessing the risk or likelihood of the subject having AD or MCI based on the mRNA expression level of the at least one target gene.

Images