阅读说明：本技术 利用代谢组分析诊断***的方法 (Method for diagnosing Behcet's disease using metabolome analysis ) 是由 金京宪 车勋锡 金净衍 安重敬 于 2018-04-17 设计创作，主要内容包括：本发明涉及一种利用代谢组分析诊断白塞氏病的方法,并提供一种生物标记,通过所述生物标记可利用代谢组学有效地诊断白塞氏病,并且所述生物标记可用于研发白塞氏病的治疗剂。(The present invention relates to a method for diagnosing Behcet's disease using metabolomic analysis, and provides a biomarker by which Behcet's disease can be effectively diagnosed using metabolomics and which can be used to develop a therapeutic agent for Behcet's disease.)

1. A behcet's disease diagnostic kit comprising a quantification device for one or more blood metabolites selected from the group consisting of capric acid, fructose, tagatose, oleic acid, linoleic acid, L-cysteine, sorbitol, uridine, inosine, plant galactose, glycolic acid, palmitic acid and histidine.

2. The Behcet's disease diagnostic kit of claim 1, wherein the kit comprises one or more selected from the group consisting of capric acid, fructose, tagatose, oleic acid and linoleic acid in the blood metabolites.

3. The Behcet's disease diagnostic kit of claim 1, wherein the quantitative device is a chromatograph/mass spectrometer.

4. the Behcet's disease diagnostic kit of claim 1, wherein in case of an increase in concentration of at least one selected from the group consisting of capric acid, fructose, tagatose, L-cysteine, sorbitol, uridine, inosine, plant galactose and glycolic acid, the case indicates Behcet's disease.

5. The Behcet's disease diagnostic kit of claim 1, wherein in case of a decrease in the concentration of at least one selected from the group consisting of oleic acid, linoleic acid, palmitic acid and histidine, the condition is indicative of Behcet's disease.

6. A method of detecting a metabolite difference between a blood sample obtained from a normal control and a blood sample obtained from a patient with behcet's disease, the method comprising: the metabolite biomarkers from blood were analyzed by sequentially performing the following processes:

(1) Analyzing the metabolites using gas chromatography/time-of-flight mass spectrometry;

(2) Identifying differences between metabolite profiles by partial least squares discriminant analysis of the metabolites identified using gas chromatography/time-of-flight mass spectrometry;

(3) Selecting a metabolite having a variable projection importance value of 1.5 or more than 1.5 among the metabolites obtained by partial least squares discriminant analysis as a metabolite biomarker candidate, and confirming an increase or decrease in the metabolite biomarker candidate by a load value of partial least squares discriminant analysis; and

(4) Validating the metabolite biomarkers using the subject manipulation profile.

7. The method of claim 6, wherein analyzing the metabolites using the gas chromatography/time-of-flight mass spectrometry comprises analyzing the blood sample using a gas chromatography/time-of-flight mass spectrometry system, converting the analysis results into statistically processable values, and then statistically validating the metabolite differences between the blood sample obtained from a normal control and the blood sample obtained from a Behcet's patient using the converted values.

8. The method of claim 7, wherein converting the gas chromatography/time-of-flight mass spectrometry results into statistically processable values comprises determining a maximum value representing an area or height of a chromatographic peak for a unit time as a representative value for the unit time after dividing a total analysis time by a unit time interval.

9. the method of claim 6, wherein a trend of the metabolite increase is indicated in case the load value of partial least squares discriminant analysis is positive, and a trend of the metabolite decrease is indicated in case the load value of PLS-DA is negative.

10. The method of claim 6, wherein the metabolite biomarker comprises one or more selected from the group consisting of capric acid, fructose, tagatose, oleic acid, linoleic acid, L-cysteine, sorbitol, uridine, inosine, plant galactose, glycolic acid, palmitic acid, and histidine.

Technical Field

The present invention relates to a method for diagnosing Behcet's disease by metabolomic analysis (metabolomic analysis).

Background

Behcet's disease is a systemic vasculitis of unknown etiology and characterized by various symptoms such as: oral, genital and anal ulcers, uveitis (uveitis), arthritis, and disease invasion of vital organs (e.g., gastrointestinal tract, blood vessels, central nervous system, etc.). It is reported that Behcet's disease occurs more frequently in regions from the coast of the Mediterranean sea to the far east Asia, especially in Korea, China, Japan, and Turkish.

Behcet's disease has a very diverse clinical presentation and may be associated with mild symptoms (e.g., recurrent oral ulceration) and fatal sequelae (e.g., blindness, intestinal ulceration and perforation, hemoptysis due to aneurysms, deep vein thrombosis, and hemiplegia) caused by invasion of the disease into the eyeball, gastrointestinal tract, blood vessels, central nervous system, and the like. The various symptoms of behcet's disease are considered to have the most severe activity between the ages of 20 and 40 and therefore the disease is expected to cause very large economic and social losses.

Behcet's disease shows different clinical manifestations and prognosis depending on invasion into different organs, and thus it is difficult to diagnose and treat the disease. Therefore, in order to minimize complications and disorders caused by Behcet's disease, it is very important to accurately diagnose Behcet's disease at an early stage.

The diagnosis of Behcet's disease is largely dependent on clinical symptoms, as objective diagnostic biomarkers (biorarker) are lacking to distinguish Behcet's patients from healthy people. However, in fact, behcet's disease is accompanied by various clinical symptoms due to genetic, environmental and immunological abnormalities leading to the invasion of several organs by the disease, and therefore a well-known single biomarker shows low sensitivity and low specificity (specificity). Therefore, accurate diagnosis is difficult due to various clinical symptoms and inaccuracy of existing biomarkers, and thus it takes a long time to make a definite diagnosis after the onset of disease. To overcome these obstacles, it is important to invent objective diagnostic biomarkers.

Thus, by finding objective diagnostic biomarkers capable of diagnosing Behcet's disease, Behcet's disease can be diagnosed at an early stage, and thus the time taken to clearly diagnose Behcet's disease can be reduced, and the disease can be appropriately treated, thereby minimizing complications due to the worsening of patient symptoms. Furthermore, this provides the patient with customized therapy without the need for expensive and unnecessary therapy, and provides the patient with accurate information about disease-related prognosis, and thus a better treatment score (treatment score) is expected. Recently, metabolomics has attracted considerable attention in the discovery of biomarkers for rheumatic diseases (e.g., rheumatoid arthritis, osteoarthritis, psoriatic arthritis, and systemic lupus erythematosus). [ Madson RK (madsen RK) et al, Diagnostic of metabolic disorders in rheumatoid Arthritis (2011) Arthritis research and treatment (Arthritis Res Ther.). 13(1): r19; kapur (kappa) et al, metabolome predicts the response of rheumatoid Arthritis patients to anti-TNF alpha therapy (Metabolic profiling therapy to anti-tumor necrosis factor alpha therapy in patients with Arthritis) (2013) Arthritis and rheumatism (Arthritis Rheum)65: 1448-65; kimms (kims) et al, Global metabolite profiling of synovial fluid is used for specific diagnosis of rheumatoid arthritis and other inflammatory arthritis (Global metabolic profiling of synthetic fluid for the specific diagnosis of rheumatoid arthritis from other inflammatory arthritis). (2014) Public science library (PLos one)9: e97501 ].

The technologies reported so far for discovering biomarkers of behcet's disease are based on genomic or proteomic (genomic or proteomic) approaches, but the results are not clear, or it is difficult to practically apply these technologies to the diagnosis of behcet's disease [ excellent (Yuko) et al, proteomic monitoring with uveitis of behcet's autoimmunity: selenium binding protein is a novel autoantigen in Behcet's disease (bacterial viral infection of autoimmune in Behcet's disease with infection: selenium binding protein is a novel autoantigen in Behcet's disease). (2007) Experimental Eye study (Experimental Eye Research) 84: 823-; honesty (Seido), et al, performed Proteomic monitoring of the patient's autoantigen of Behcet with diseases by a Proteomic approach. (2010) Microbial immunology (Microbiol Immunol) 54: 354-361], and there are no reports on the discovery of biomarkers suitable for diagnosis and prediction of prognosis of Behcet's disease using metabolomics.

Disclosure of Invention

Technical problem

Accordingly, in order to find specific biomarkers in a blood sample to rapidly and conveniently diagnose Behcet's disease, research and efforts were made to find metabolomics patterns of metabolites in blood, which can distinguish Behcet's patients exhibiting various symptoms from healthy controls (health controls), and specific metabolome by using gas chromatography/time-of-flight mass spectrometry (GC/TOFMS), and as a result, the present inventors found novel biomarkers for accurately diagnosing Behcet's disease by applying metabolomics to blood, thereby completing the present invention.

Accordingly, it is an object of the present invention to provide a kit for diagnosing Behcet's disease through metabonomic analysis.

It is another object of the present invention to provide a method for analyzing metabolite differences for diagnosing Behcet's disease.

Technical solution

The present invention provides a Behcet's disease diagnostic kit including a quantitative device for one or more blood metabolites selected from the group consisting of capric acid (decanoic acid), fructose (fructose), tagatose (tagatose), oleic acid (olegic acid), linoleic acid (linoleic acid), L-cysteine (L-cysteine), sorbitol (sorbitol), uridine (uridine), inosine (inosine), plant galactose (galactonate), glycolic acid (glycolate), palmitic acid (palmitic acid) and histidine (histidine).

The present invention also provides a method of detecting a metabolite difference between blood obtained from a normal control and blood obtained from a Behcet's patient, the method comprising: the metabolite biomarkers from blood were analyzed by sequentially performing the following processes:

(1) Analyzing the metabolites using gas chromatography/time of flight mass spectrometry (GC/TOF MS);

(2) Confirming a difference between metabolite profiles (metabolite profiles) by partial least squares discriminant analysis (PLS-DA) of the metabolites identified by gas chromatography/time-of-flight mass spectrometry;

(3) Selecting a metabolite having a variable-projection-importance (VIP) value of 1.5 or more than 1.5 among the metabolites obtained by partial least squares discriminant analysis as a metabolite biomarker candidate, and confirming an increase or decrease in the metabolite biomarker candidate by a load value of partial least squares discriminant analysis; and

(4) The metabolite biomarkers were verified using a receiver operating characteristic curve (ROC curve).

Advantageous effects

According to the present invention, it has been found that biomarkers for Behcet's disease can be rapidly and accurately diagnosed by metabolomics methods to specifically diagnose patients with Behcet's disease differently. 104 metabolites have been detected by analyzing metabolites in blood samples of Behcet's patients and normal individuals using GC/TOF MS. By calculating values of partial least squares discriminant analysis (PLS-DA) and variable projection importance (VIP), values of area under the curve (AUC) of the Receiver Operating Characteristics (ROC) curve, fold change, p-value, etc., 13 effective metabolite biomarkers have been proposed. Furthermore, a panel (panel) for the diagnosis of behcet's disease was finally constructed using five biomarkers (decanoic acid, fructose, tagatose, oleic acid, and linoleic acid) and verified its clinical validity using a validation set. According to the present invention, a metabolomics-based blood analysis is first used to determine biomarkers that are capable of specifically diagnosing Behcet's disease. These biomarkers can serve as the basis for studies that reveal the pathogenesis of Behcet's disease, which has not yet been fully discovered. In addition, these biomarkers can also be used to develop therapeutics optimized for various clinical conditions. The discovery of biomarkers that contribute to the diagnosis of Behcet's disease enables rapid and accurate diagnosis of Behcet's patients and significantly reduces the long time required for clinical diagnosis, thereby rapidly providing patients with customized therapy, and thus social and economic linkage effects such as rapid restoration of daily life are expected to be considerable.

Drawings

Fig. 1 shows results obtained using PLS-DA showing metabolite differences in metabolomic profiles between blood obtained from patients with behcet's disease and blood obtained from healthy controls, showing significant differences between patients with behcet's disease and healthy controls [ BD: behcet's disease; HC: healthy control ].

Fig. 2 is a group graph for comparing the levels of the first 9 metabolites (a) showing a significant increase in the behcet's disease group with the levels of the first 4 metabolites (B) showing a significant decrease in the behcet's disease group [ BD: behcet's disease; HC: healthy control ].

Fig. 3a to 3c show PLS-DA results showing metabolite differences between groups of patients with behcet's disease with or without administration of steroids (steroid), colchicine (colchicine) or azathioprine (azathioprine), wherein the difference between each group administered and the group not administered has no reproducibility (reproducibility) due to its very low Q2 value and no statistically significant difference between the groups of metabolites.

Fig. 4 shows a multivariate analysis model generated by PCA to generate a metabolite biomarker panel for diagnosing behcet's disease using the first 3 metabolites (decanoic acid, fructose, and tagatose) showing significant increase in the behcet's disease group and the first 2 metabolites (oleic acid and linoleic acid) showing significant decrease in the behcet's disease group, in which when using monoaxial (PC1), R2X value was 0.721, which shows proper classification, and the model showed Q2 value of 0.515, thereby confirming that the model has reproducibility [ [ BD: behcet's disease; HC: healthy control ].

Fig. 5 shows the results of a characteristic curve (ROC) for the subjects of a metabolic diagnostic panel for the diagnosis of behcet's disease using a blood sample, in which a biomarker panel using a combination of 5 metabolites exhibited a sensitivity of 100%, a specificity of 97.1% and an AUC value of 0.993 when behcet's disease was diagnosed.

Fig. 6 shows validation set validation results for a metabolic diagnostic panel for diagnosing behcet's disease using blood samples, where principal component analysis results show that, among 10 blood samples of patients with behcet's disease and 10 blood samples of healthy controls, the panel can accurately predict the relative abundance of 9 blood samples of patients with behcet's disease and 10 blood samples of healthy controls [ BD: behcet's disease; HC: healthy control ].

Detailed Description

Hereinafter, the configuration of the present invention will be described in detail.

The present invention relates to a Behcet's disease diagnostic kit comprising a quantification device for one or more blood metabolites selected from the group consisting of capric acid, fructose, tagatose, oleic acid, linoleic acid, L-cysteine, sorbitol, uridine, inosine, plant galactose, glycolic acid, palmitic acid and histidine.

In order to find biomarkers for Behcet's disease, the present inventors collected blood samples from Behcet's patients, performed methanol extraction on the blood samples, compared and analyzed the differences in metabolite profiles between Behcet's patients and normal individuals using GC/TOF MS, and studied biomarkers found to be capable of diagnosing Behcet's patients using the differences.

as a result, 104 metabolites were identified, which were classified into amines, amino acids, fatty acids, organic acids, phosphates, sugars, and the like. Among these, the largest amount of amino acids is detected, followed by organic acids, fatty acids, sugars, amines, phosphates, and the like.

When blood samples of 35 Behcet's patients (BD) and blood samples of 35 Healthy Controls (HC) were compared, significant differences in metabolite profiles between the blood samples of Behcet's patients and the blood samples of healthy controls were confirmed by partial least squares discriminant analysis (PLS-DA), and VIP values of 1.5 or more than 1.5, fold changes of 1.2, AUC values of 0.800 or more than 0.800, and p values of less than 0.01 were set as reference values for each metabolite, and 13 metabolites were selected as novel biomarker candidates. Each metabolite showed statistically significant differences in behcet patients and healthy controls, thereby confirming that the metabolite is a suitable candidate biomarker. Further, in order to confirm that the specific metabolite profiles and biomarker candidates of Behcet's disease are not due to the effects of drugs administered to treat Behcet's disease, Behcet's patients were grouped according to the administered drugs to perform PLS-DA, and as a result, it was confirmed that the drugs administered to Behcet's patients did not cause metabolite differences.

Furthermore, from the 13 metabolites selected as candidate biomarkers, 3 metabolites (decanoic acid, fructose and tagatose) showing a significant increase in blood samples of patients with Behcet's disease and 2 metabolites (oleic acid and linoleic acid) showing a significant decrease in blood samples of patients with Behcet's disease were selected to generate a metabolite biomarker panel consisting of 5 metabolites for discriminating Behcet's disease. To confirm the potential of the biomarker panel consisting of 5 candidates for the diagnosis of behcet's disease, validation was performed using the ROC curve, and the biomarker panel had a sensitivity of 100%, a specificity of 97.1%, and an AUC value of 0.993, thereby showing very good results for the diagnosis of behcet's disease. In order to confirm the applicability of the model, principal component analysis was performed using 10 blood samples of patients with Behcet's disease and 10 blood samples of healthy controls. As a result, it was verified that the biomarker panel using 5 metabolites found by the inventors of the present invention is suitable for diagnosing behcet's disease.

furthermore, the kit may comprise quantitative information on one or more selected from the group consisting of L-cysteine, sorbitol, uridine, inosine, plant galactose, glycolic acid, palmitic acid and histidine, in addition to one or more selected from the group consisting of capric acid, fructose, tagatose, oleic acid and linoleic acid, which have been newly validated by the inventors of the present invention as indicative metabolites of behcet's disease, and thus enable a more consistent, highly reliable and accurate diagnosis of behcet's disease.

The term "diagnosing" as used herein is intended to encompass determining a subject's susceptibility to a particular disease or disorder, determining whether a subject is currently suffering from a particular disease or disorder (e.g., recognizing Behcet's disease), determining the prognosis of a subject with a particular disease or disorder, or a measure of therapy (therametrics) (e.g., monitoring the subject's condition to provide information about the efficacy of a therapy).

The quantitative device included in the diagnostic kit of the present invention may be a chromatograph/mass spectrometer.

Chromatographs used in the present invention include those for gas chromatography, liquid-solid chromatography (LSC), Paper Chromatography (PC), thin-layer chromatography (TLC), gas-solid chromatography (GSC), liquid-liquid chromatography (LLC), Foam Chromatography (FC), Emulsion Chromatography (EC), gas-liquid chromatography (GLC), Ion Chromatography (IC), Gel Filtration Chromatography (GFC) or gel permeation chromatography (GFC), but are not limited to any of the chromatography fields to which GPC belongs. Preferably, the chromatograph used in the present invention is a gas chromatograph. Further, the mass spectrometer used in the present invention is MALDI-TOF MS or TOF MS, more preferably TOF MS.

The components of the blood metabolite of the present invention are separated by gas chromatography and identified not only by accurate molecular weight information but also by elemental composition using information obtained by Q-TOF MS.

According to an exemplary embodiment of the present invention, the case where the concentration of at least one selected from the group consisting of capric acid, fructose, tagatose, L-cysteine, sorbitol, uridine, inosine, plant galactose and glycolic acid is increased is indicative of behcet's disease, and the case where the concentration of at least one selected from the group consisting of oleic acid, linoleic acid, palmitic acid and histidine is decreased is indicative of behcet's disease.

The term "increase in blood metabolite concentration" as used herein means that the blood metabolite concentration of a Behcet's patient is measurably significantly increased compared to a healthy individual, preferably by 70% or more than 70%, and more preferably by 30% or more than 30%.

the term "reduction of blood metabolite concentration" as used herein means that the blood metabolite concentration of a Behcet's patient is measurably significantly reduced compared to a healthy individual, preferably by 40% or more than 40%, and more preferably by 20% or more than 20%.

according to the present invention, a significant increase in the concentration of at least one selected from the group consisting of capric acid, fructose, tagatose, L-cysteine, sorbitol, uridine, inosine, plant galactose and glycolic acid was exhibited in Behcet's disease patients as compared to healthy individuals, and a significant decrease in the concentration of at least one selected from the group consisting of oleic acid, linoleic acid, palmitic acid and histidine was exhibited in Behcet's disease patients as compared to healthy individuals (see Table 1).

The present invention also provides a method of detecting a metabolite difference between a blood sample obtained from a normal control and a blood sample obtained from Behcet's disease, the method comprising: the metabolite biomarkers from blood were analyzed by sequentially performing the following processes:

(1) Analyzing the metabolites using gas chromatography/time of flight mass spectrometry (GC/TOF MS);

(2) Confirming differences between metabolite profiles by performing partial least squares discriminant analysis (PLS-DA) on the metabolites identified using gas chromatography/time-of-flight mass spectrometry;

(3) Selecting a metabolite having a variable projection importance (VIP) value of 1.5 or more than 1.5 among the metabolites obtained by partial least squares discriminant analysis as a metabolite biomarker candidate, and confirming an increase or decrease in the metabolite biomarker candidate by a load value of partial least squares discriminant analysis; and

(4) The metabolite biomarkers were validated using a subject operating characteristic curve (ROC curve).

A method of analyzing a metabolite difference between two biological sample groups (i.e., a blood sample obtained from a patient with behcet's disease and a blood sample obtained from a normal control) according to the present invention will be described in detail as an example.

First, blood samples taken from normal individuals and patients with Behcet's disease were extracted with 100% methanol and then derivatized (derivitized) using known techniques for use in GC/TOF MS.

Methods for analyzing metabolites of blood samples using GC/TOF MS include: the blood extract was analyzed using a GC/TOF MS system, the analysis results were converted into statistically processable values, and then the differences in metabolites between the two groups of biological samples were statistically validated using the converted values.

The converting of the GC/TOF MS result into a statistically processable value may be determining a maximum value representing an area or height of a chromatographic peak for a unit time as a representative value for the unit time after dividing the total analysis time by the unit time interval.

According to one embodiment of the present invention, as a result of GC/TOF MS, 104 metabolites are identified, which can be classified into amines, amino acids, fatty acids, organic acids, phosphates, sugars, etc., and wherein the maximum amount of amino acids is detected, followed by organic acids, fatty acids, sugars, amines, phosphates, etc., in that order.

Each metabolite was normalized by dividing the intensity of the metabolite obtained by GC/TOF MS by the sum of the intensities of all identified metabolites and subjected to PLS-DA.

Creating a V-plot (V-plot) consisting of PLS-DA load values and VIP values of the metabolites, selecting a metabolite having a VIP value of 1.5 or more than 1.5 as a metabolite biomarker candidate, and confirming an increase or decrease in the load value of PLS-DA, and in this regard, a positive load value indicates that the metabolite tends to increase, and a negative load value indicates that the metabolite tends to decrease.

The increase or decrease of the metabolite can be confirmed using the intensity of the metabolite of the blood sample analyzed by GC/TOF MS.

Metabolite biomarkers were verified by ROC curves.

According to one embodiment of the present invention, as the biomarker for diagnosing Behcet's disease, one or more selected from the group consisting of capric acid, fructose, tagatose, oleic acid, linoleic acid, L-cysteine, sorbitol, uridine, inosine, plant galactose, glycolic acid, palmitic acid, and histidine may be used.

According to the present invention, the method of analyzing the metabolite difference between the blood sample obtained from the normal group and the blood sample obtained from the Behcet's patient enables a more consistent, highly reliable and accurate diagnosis of Behcet's disease, and is applicable to the development of therapeutic agents.

Modes for the invention

Hereinafter, the present invention will be described in more detail with reference to the following examples, which are not intended to limit the scope of the present invention.

Examples of the invention

Example 1: identification of metabolites using GC/TOF MS

20 microliters of each of the blood of the patient group with Behcet's disease and the healthy control was mixed with 980 microliters of pure methanol, and the mixture was centrifuged to extract metabolites.

The derivatization process of GC/TOF MS is as follows.

Each extracted sample was dried in a speed bag (speed bag), and then 5 microliters of O-methylhydroxylamine hydrochloride (O-methylhydroxylamine hydrochloride) in pyridine (40%) was added thereto to allow a reaction therebetween at 30 ℃ and 200 revolutions per minute (rpm) for 90 minutes. Then, 45. mu.l of N-methyl-N- (trimethylsilyl) trifluoroacetamide (N-methyl-N- (trimethylsilyl) trifluoroacetamide) was added thereto to allow a reaction therebetween at 37 ℃ and 200 rpm for 30 minutes.

The instrument conditions for GC/TOF MS are as follows.

The column used for the analysis was a RTX-5Sil MS capillary column (length 30 m, film thickness 0.25 mm and internal diameter 25 mm) and the gas chromatography column (GC column) temperature conditions were as follows: first, the temperature was maintained at 50 ℃ for 5 minutes, and then the temperature was raised to 330 ℃ and maintained for 1 minute. Mu.l of each sample was injected in a splitless mode. The transfer line temperature and the ion source temperature were maintained at 280 ℃ and 250 ℃, respectively. 104 metabolites were found in the library with GC/TOF MS results and identified (see Table 1).

As shown in table 1 below (104 metabolites identified by analysis of metabolites using blood samples from patients with behcet's disease and healthy controls), when sorted by metabolome group, 26% amino acids, 19% organic acids, 17% fatty acids, 15% sugars, 11% amines, 5% phosphates and 7% others were present.

[ Table 1]

example 2: differences in metabolite profiles in blood samples of Behcet's patients obtained using PLS-DA and blood samples of healthy controls

The intensity (intensity) of each metabolite identified according to example 1 was divided by the sum of the intensities of all identified metabolites to normalize each metabolite. Subsequently, it was PLS-DA performed using SIMCA-P + (version 12.0).

As shown in FIG. 1, it was confirmed that there was a significant difference in metabolite profiles between the blood samples of Behcet's patients and the blood samples of healthy controls.

table 2 shows VIP values and load values, which indicate the degree and directionality of the influence of 104 metabolites used in the PLS-DA model on the model.

[ Table 2]

example 3: selection of metabolites of Behcet's patient-specific biomarkers

To find biomarkers that show specific increase or decrease in behcet's patients, VIP values, fold changes, AUC values and p-values that affect the metabolomic profile differences for each metabolite obtained from example 2 were obtained. A VIP value of 1.5 or more than 1.5, a fold change of 1.2, an AUC value of 0.800 or more than 0.800 and a p value of less than 0.01 were set as reference values for each metabolite, and 13 metabolites showed applicability to the diagnosis of behcet's disease (see table 3). In addition, the absolute intensities of these metabolites were compared by group (see FIG. 2).

Table 3 below shows the VIP, AUC, fold change (fold change) and p-values (BD: Behcet's disease; control: healthy individuals) for the 13 metabolites selected as potential biomarkers for the diagnosis of Behcet's disease.

[ Table 3]

AUC, area under ROC curve; BD, behcet's disease; VIP, variable projection importance

Example 4: use of PLS-DA to verify whether a drug has an effect on the expression of an increase or decrease in a metabolite in Behcet's disease patients

In order to verify that the biomarkers showing specific increase or decrease in Behcet's disease patients were not affected by the drug, comparison of each drug-administered group and non-drug-administered group was performed using PLS-DA, and as a result, the level of segregation was found to be insufficient and non-reproducible. There was no reproducibility and no statistically significant difference according to the drugs in the 3 groups to which steroids, colchicine or azathioprine were administered.

Therefore, since the increase or decrease of the metabolites of the Behcet's disease group exhibited in example 3 was due to changes caused by the disease itself, it was confirmed that the metabolites are suitable as biomarkers (see FIGS. 3 a-c).

Example 5: establishment of Metabolic diagnostic Panels for diagnosis of Behcet's disease by blood samples Using 5 metabolites

Among the 13 biomarkers selected in example 3 for the diagnosis of Behcet's disease, the first 3 metabolites (decanoic acid, fructose and tagatose) showing specific increase in Behcet's disease and the first 2 metabolites (oleic acid and linoleic acid) showing specific decrease in Behcet's disease were selected to establish a metabolic diagnostic panel capable of diagnosing Behcet's disease. Therefore, on the basis of principal component analysis, a multivariate classification model capable of distinguishing BD from HC based on 5 metabolites was established. When using a single axis (i.e., PC1), BD was shown to be completely distinct from HC, and the model showed R2X values of 0.721 and Q2 values of 0.515, showing that patients with behcet's disease were properly and reproducibly distinguished from healthy controls (see fig. 4).

Example 6: validation set validation by model validation of ROC and metabolic diagnostic panel for diagnosis of Behcet's disease using blood samples

To examine whether the metabolite biomarker panel for diagnosing Behcet's disease using blood samples produced according to example 5 was suitable for diagnosis, a Receiver Operating Characteristic (ROC) curve was plotted in the model using the PC1score (PC1score) of each sample. As a result, the model showed a sensitivity of 100%, a specificity of 97.1% and an AUC value of 0.993, which showed that the model was very suitable for diagnosing behcet's disease (see fig. 5). Furthermore, to check whether the panel could use the validation set to predict the diagnosis of Behcet's disease, 10 blood samples from Behcet's patients and 10 blood samples from healthy controls were used, i.e. 20 samples in total. The results show that the panel was able to accurately predict whether 19 out of a total of 20 samples were Behcet disease patients or healthy controls, indicating that the biomarker panel with 5 metabolites is also suitable for the confirmation set of Behcet disease diagnosis (see FIG. 6).