CN115963271A - Application of protein marker in preparation of detection product for distinguishing early NASH (NASH) from NAFL (non-catalytic artificial membrane replacement) and detection method - Google Patents

Abstract

The invention provides an application of a protein marker in preparation of a detection product for distinguishing early NASH from NAFL and a detection method. Through screening of population queues of NAFL and early NASH, 22 serum protein markers which are obviously related to the early NASH are found, and therefore, the proteins have important clinical application value as markers for effectively detecting the early NASH.

Description

Application of protein marker in preparation of detection product for distinguishing early NASH (NASH) from NAFL (non-catalytic artificial membrane replacement) and detection method

Technical Field

The invention relates to the field of non-alcoholic fatty liver disease detection, in particular to application of a protein marker in preparation of a detection product for distinguishing early NASH from NAFL and a detection method.

Background

Nonalcoholic steatohepatitis (NASH) belongs to nonalcoholic fatty liver disease (NAFLD), a chronic liver disease caused by fat accumulation and inflammation, similar to the symptoms of alcoholic fatty liver disease, but without a history of excessive alcohol consumption. NAFLD disease spectrum typically includes NAFL (simple fatty liver), NASH, cirrhosis and liver cancer. In recent years, the number of patients with NAFLD has continued to increase clinically as obesity, type ii diabetes and cardiovascular disease have spread worldwide. NAFLD has become the most important cause of liver disease worldwide, affecting more than one-fourth of the world's population.

The development of NAFLD is a dynamic process. It is mainly characterized by excessive accumulation of fat in the liver, which acts in combination with other factors such as insulin resistance, so that the liver is subjected to a large metabolic and oxidative stress, thereby causing a persistent inflammatory response and apoptosis of hepatocytes. If NASH is not controlled at this stage, after the death of the injured hepatocytes, the liver initiates its own injury repair mechanism to replenish the dead hepatocytes, and fibrosis is the physiological response that accompanies the liver injury repair process. Metabolic toxicity of the liver due to chronic inflammation and excessive load can continuously lead to death of liver cells, and the process of hepatic fibrosis is continuously accumulated, and finally, cirrhosis is caused. Compared to NAFL, the prognosis for NASH patients is worse. There are studies that indicate that 11% of NASH patients will be diagnosed with cirrhosis, whereas only 1% of NAFL patients will be diagnosed with cirrhosis. Normally NAFL and early NASH are reversible, but NASH and fibrosis occur quite insidiously, with no apparent symptoms at early stages. When overt symptoms appear, the condition may have progressed to mid-late cirrhosis and hepatocellular carcinoma, increasing the risk of liver disease death.

Hepatocellular carcinoma accounts for approximately 75% of all liver cancers, and is the most common primary liver cancer. In 2020, the mortality rate of liver cancer is ranked fifth in all cancers worldwide, the morbidity and the prevalence rate thereof are increased every year, in China, liver cancer is the second largest cancer causing death, and although hepatitis B and hepatitis C are the main causes of liver cancer, NASH is predicted to gradually replace hepatitis virus to become the main cause of liver cancer with the general vaccination of hepatitis B virus vaccine. Therefore, accurate typing of NASH high-risk people and NAFLD diseases in early screening is crucial to early intervention and treatment of the diseases, the cure rate of patients can be effectively improved, and the disease deterioration possibility is reduced.

Liver biopsy is currently the gold standard for differentiating patients NAFL, NASH, liver fibrosis and cirrhosis, and is mainly based on the diagnosis by pathologists on the scores for hepatocellular steatosis, intralobular inflammation, hepatocellular ballooning and fibrosis. However, liver biopsy is an invasive method, expensive to detect, has potential sampling errors and subjective variability of interpretation by pathologists, and is not suitable as the first choice for NASH screening and efficacy assessment. There are studies to assist doctors in evaluating medical image results through artificial intelligence machine learning to achieve more accurate disease diagnosis and reduce time and economic cost, but no related technology is applied to NAFLD diagnosis so far, and the problem of sample sampling errors still cannot be solved by the technology.

In addition, several non-invasive NAFLD diagnostic protocols were also proposed in succession, including serological marker detection, imaging examinations, predictive models, etc. For simple fatty liver, B-ultrasound, MRI and TE are reported to be effective noninvasive methods and can be used for quantifying the fat content of liver, but the application in clinic is not very common at present due to the price and the like.

Keratin (CK-18) is a serological marker of NASH, but is still in scientific research and has not been clinically applied. The development of noninvasive diagnosis of fibrosis is more significant, such as APRI (aspartate Aminotransferase (AST) to platelet ratio index), NAFLD fibrosis score (NAFLD fibrosis score), and FIB-4 (fibrosis-4 index) are largely based on clinical routine variables using formulas to predict advanced fibrosis. There are also diagnostic methods designed by detecting new biomarkers, for example the ELF (enhanced liver fibrosis) detection method measures the degree of liver fibrosis by measuring the concentration of 3 matrix-converting proteins (hypouronic acid, tissue inhibitor of osteoproteinase 1and N-terminal procollagen III-peptide). The ELF assay is also in the course of clinical trials and is not diagnostic for patients with early NASH and low levels of fibrosis. VCTE (vibration-controlled transfer imaging; fibroscan) can measure the degree of stiffness of the liver non-invasively and has been approved by the FDA for the detection of children and adults. NIS4 (non-invasive score 4) diagnoses NASH and liver fibrosis patients by measuring the levels of four molecules (miR-34 a-5p, alpha-2-macrobuline, chitinase-like protein 1and HbA1c) in serum, and this detection method has been shown in clinical trials to have high accuracy in distinguishing non-NASH from NASH in the middle and late stages (advanced NASH, which is distinguished from NASH in the early stage by a high degree of fibrosis, i.e., a high score for clinical fibrosis score), and has not been demonstrated in Asian populations.

It is now more difficult to distinguish NAFL from NASH and cannot be used to distinguish NAFL from early NASH in existing serum and in vitro detection methods. NAFL and early NASH differ primarily in the presence of liver damage (inflammation or balloon degeneration) in liver tissue. Furthermore, the early NASH has no obvious symptoms, the symptoms of the two diseases are not obviously distinguished, and NAFL and the early NASH can not be accurately distinguished clinically according to the self-description of patients. The NAFL patient can be accurately detected, and certain intervention treatment is carried out, so that the condition of the patient is easy to reverse. However, if not detected, the treatment difficulty and cost are greatly increased when the disease is exacerbated into NASH.

Therefore, there is an urgent need to develop a highly sensitive and specific protocol that can be used to diagnose patients with early NASH without significant fibrosis.

Disclosure of Invention

The invention mainly aims to provide application of a protein marker in preparation of a detection product for distinguishing early NASH from NAFL and a detection method, so as to solve the problem that the early NASH is difficult to be effectively detected in the prior art.

To achieve the above object, according to one aspect of the present invention, there is provided a use of a protein marker for preparing a test product for distinguishing early NASH from NAFL, the protein marker comprising at least one of: CCL15, SIGLEC10, ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, FBP1, GAS6, GUSB, IL-1ra, KYNU, LILRB4, MVK, and TNF-R2.

Further, the protein markers include at least one of: CCL15 and SIGLEC10; preferably, the protein marker further comprises at least one of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, FBP1, GAS6, GUSB, IL-1ra, KYNU, LILRB4, MVK, and TNF-R2.

Further, the protein markers include at least two of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, and at least one of them is CCL15 or SIGLEC10.

Further, the protein marker is selected from any one of the following groups:

/>

/>

further, the detection product is a detection kit or a detection device.

To achieve the above object, according to one aspect of the present invention, there is provided a kit for distinguishing early NASH from NAFL, the kit comprising a detection reagent for a protein marker, the protein marker comprising at least one of: CCL15 and SIGLEC10.

Further, the protein marker further comprises at least one of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, FBP1, GAS6, GUSB, IL-1ra, KYNU, LILRB4, MVK, and TNF-R2.

Further, the protein markers include at least two of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, and at least one of them is CCL15 or SIGLEC10.

Further, the protein marker is selected from any one of the above groups of protein markers.

In order to achieve the above object, according to one aspect of the present invention, there is provided a detection apparatus for distinguishing early NASH from NAFL, which has a built-in detection model for distinguishing early NASH from NAFL, wherein the detection model is a model for detecting protein markers including a plurality of protein markers in the above application.

Further, the detection model is a logistic regression model.

Further, the detection apparatus comprises a storage medium on which the detection model is stored.

Further, the detection apparatus includes a processor for running the detection model.

Further, the detection device comprises a protein marker expression level receiving module of the sample to be detected, and the receiving module comprises at least one of the following modes: the user manually inputs a mode, an alternative list import mode and a file import mode.

Further, the sample to be tested is a body fluid sample, preferably a serum sample.

Further, the sample to be tested is derived from an obese patient.

Further, the sample to be tested is derived from a patient suffering from NAFLD.

In order to achieve the above object, according to an aspect of the present invention, there is provided a detection method for distinguishing early NASH from NAFL, the detection method comprising: detecting the expression quantity of the protein marker in the body fluid of the subject to obtain the expression quantity of the protein to be detected; inputting the expression quantity of the protein to be detected into a detection model for distinguishing early NASH from NAFL, and outputting a detection result; wherein, the detection model is a model for detecting protein markers, and the protein markers comprise the protein markers in the application.

Further, the detection model is a logistic regression model.

Further, the body fluid is serum.

Further, the subject is selected from obese patients.

Further, the subject is selected from a NAFLD patient.

By applying the technical scheme of the invention and screening different crowd queues, 22 serum protein markers which are obviously related to the early NASH are discovered, so that the proteins have important clinical application value as markers for effectively detecting the early NASH. For example, by using the expression level (e.g., concentration) of any one or more protein markers in serum, and combining clinical data, a relevant detection product can be developed, such as preparing a detection kit or establishing a relevant diagnostic model, to detect or diagnose whether the subject is early NASH.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:

fig. 1a to 1b show the differential distribution of 22 protein markers in NAFL and early NASH populations, in particular Boxplot (distribution of reaction sample concentrations), according to an embodiment of the invention.

Detailed Description

It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present invention will be described in detail with reference to examples.

Interpretation of terms:

NASH: non-alcoholic steatohepatitis, clinically classified into 3 stages of NASH, is referred to herein as stage 1 NASH, also known as early NASH. The difference between stage 1and stages 2 and 3 is whether fibrosis (fibrosis) is present in the liver biopsy results, i.e. if only steatosis (steatosis), inflammation (inflammation) in lobule and ballooning of hepatocytes are found, fibrosis (fibrosis) score is 0, it is the early NASH, and fibrosis (fibrosis) score is greater than 0, it is the advanced NASH.

Non-NASH: i.e. not in the non-early NASH population, the meaning of the present application includes healthy people without liver disease as well as people with simple fatty liver.

advanced NASH: the meaning of NAFLD in this application is a generic term referring specifically to liver diseases including NAFL and NASH, and as mentioned above, NASH can be divided into 3 stages, this application mainly studies the difference between stage 1 NASH and non-NASH, advanced NASH generally refers to stage 2, 3 NASH, and stage 1and stage 2, 3 NASH are mainly distinguished by the presence or absence of fibrosis. The sample used in this application was studied for early NASH because the fibrosis score was overall very low (0). A fibrosis score of greater than 0 overall is considered advanced NASH (i.e., middle and advanced NASH).

NAFL: simple fatty liver, non-alcoholic fatty liver disease.

NAFLD: the general term of nonalcoholic liver disease refers to the general term of liver disease including NASH and NAFL.

Uniport: (Universal Protein) is a Protein database containing Protein sequences, functional Information, research paper indices, integrating resources including EBI (European Bioinformatics Institute), SIB (the Swiss Institute of Bioinformatics) and PIR (Protein Information Resource) three major databases.

Based on the current need for more efficient non-invasive detection of NASH as mentioned in the background section, 22 proteins were found to be significantly associated with NASH by detecting serum proteins in the cohort of 135 populations using proximity extension analysis in combination with their well-established clinical information and assembled slice pathology data. Therefore, detection models for distinguishing NASH population from NAFL population are respectively established according to the expression quantity of any one or more of 22 proteins in serum, and clinical data verifies that the different detection models have higher accuracy, so that the method is suitable for rapidly, efficiently and objectively screening related populations.

On the basis of the above research results, the applicant proposed the technical solution of the present application. In a first exemplary embodiment of the present application, there is provided the use of a protein marker for the preparation of a test product for distinguishing early NASH from NAFL, wherein the protein marker comprises at least one of the following 22: CCL15, SIGLEC10, ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, FBP1, GAS6, GUSB, IL-1ra, KYNU, LILRB4, MVK, and TNF-R2.

In some preferred embodiments, the protein marker comprises any one or more of: CCL15 and SIGLEC10.

The two proteins are completely new protein markers which are found in the application and have significant correlation with the early NASH, so that any one of the protein markers can be used for detecting the early NASH. When at least one of the two protein markers is adopted to distinguish the early NASH from the NAFL, the sensitivity and the specificity are high.

In some preferred embodiments, the protein marker further comprises at least one of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, FBP1, GAS6, GUSB, IL-1ra, KYNU, LILRB4, MVK, and TNF-R2. On the basis of the two completely new protein markers CCL15 and SIGLEC10, at least one of the two markers is further randomly combined with at least one of the other 20 markers, so that the marker has higher sensitivity and specificity when used for distinguishing NASH from NAFL in early stage.

It should be noted that the protein markers may be randomly combined in different numbers according to actual needs, for example, the combinations may be random 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or 21, or even all 22 protein markers. In some preferred embodiments, the protein markers include at least two of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, and at least one of them is CCL15 or SIGLEC10. When the protein markers are used for distinguishing the NASH from the NAFL, the sensitivity and the specificity are higher.

In some preferred embodiments, the protein marker is selected from any one of the group of protein markers in table 1.

Table 1:

/>

/>

/>

among the combinations of the various protein markers in the above table, any of the following combinations is further preferred:

combination No. 1: ALCAM and SCF;

combination No. 17: CCL15, PON3, ROBO1, SCF and FGF-21;

combination of sequence number 20: ALCAM, CCL15, PON3, SIGLEC10 and CDCP1;

combination No. 45: insulin, ALCAM, CCL15, PON3, SIGLEC10, FCER2, CDCP1, SCF, FGF-21, and PTS;

combination of sequence number 54: insulin, ALCAM, CCL15, PON3, SIGLEC10, FCER2, CDCP1, SCF, FGF-21, CCL28, PTS and SYND1;

combination No. 55: insulin, ALCAM, CCL15, PON3, SIGLEC10, ROBO1, FCER2, SCF, FGF-21, CCL28, PTS and ADAM-TS 15.

The combined protein markers have more accurate diagnosis results predicted according to different clinical classification models.

The detection product is a detection kit or a detection device. It should be noted that the detection product can be any product that can be used for clinical detection, such as a detection kit, a detection device, a polypeptide chip, and the like. Specific application methods include, but are not limited to, the use of clinical mass spectrometry, chemiluminescence, or single-molecule immunization of Simoa.

In a second exemplary embodiment, a kit for distinguishing early NASH from NAFL is provided, the kit comprising a detection reagent for a protein marker, the protein marker comprising at least one of: CCL15 and SIGLEC10. Because the protein markers have obvious correlation with early NASH, the detection kit designed for the protein markers not only ensures that the detection is convenient, simple and rapid, but also has high detection sensitivity and specificity.

In some preferred embodiments, the protein marker further comprises at least one of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, FBP1, GAS6, GUSB, IL-1ra, KYNU, LILRB4, MVK, and TNF-R2. The kit containing the protein marker has relatively higher detection sensitivity and specificity.

In some preferred embodiments, the protein markers include at least two of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, and at least one of them is CCL15 or SIGLEC10. The kit for distinguishing the early NASH from the NAFL by adopting the protein markers has higher detection sensitivity and specificity.

In some preferred embodiments, the protein marker is selected from any one of the groups in table 1 above. Among the combinations of the various protein markers in table 1, any of the following combinations is further preferred:

combination No. 1: ALCAM and SCF;

combination No. 17: CCL15, PON3, ROBO1, SCF and FGF-21;

combination No. 20: ALCAM, CCL15, PON3, SIGLEC10 and CDCP1;

combination No. 45: insulin, ALCAM, CCL15, PON3, SIGLEC10, FCER2, CDCP1, SCF, FGF-21, and PTS;

combination of sequence number 54: insulin, ALCAM, CCL15, PON3, SIGLEC10, FCER2, CDCP1, SCF, FGF-21, CCL28, PTS and SYND1;

combination No. 55: insulin, ALCAM, CCL15, PON3, SIGLEC10, ROBO1, FCER2, SCF, FGF-21, CCL28, PTS, and ADAM-TS 15.

The diagnostic results predicted by the kit formed by the combined protein markers are relatively more accurate.

For the configuration of the detection kit, according to specific needs, a plurality of different types of detection kits can be prepared, and the form of the specific kit is not limited, for example, the specific kit can be an ELISA kit, and can also be an immunofluorescence kit or an immune colloidal gold kit, etc. The detection mode of the kit is not limited, and all clinical protein detection methods are applicable to the application, for example, the detection can be carried out in a polypeptide chip form, mass spectrometry detection or unimolecular immune Simoa detection.

In a preferred embodiment, in the kit, the reagent for detecting each protein marker is an antibody against the corresponding protein marker. In some preferred embodiments, the antibodies are disposed (e.g., coated) on a solid support; preferably, the solid phase carrier is selected from an enzyme label plate, a membrane carrier or a microsphere, and further preferably, the membrane carrier is selected from a nitrocellulose membrane, a glass cellulose membrane or a nylon membrane; preferably, the antibody of the protein marker is a monoclonal antibody or a polyclonal antibody.

From the viewpoint of convenience in detection and convenience in determination of the detection result, the antibody for each protein marker in the kit is preferably provided in a pre-coated form. Preferably, the pre-coated antibody is coated on a solid support; the specific solid phase carrier is reasonably designed according to the requirement. More preferably, the solid phase carrier comprises an enzyme label plate (mostly made of polystyrene material), a membrane carrier or microspheres; further preferably, the membrane carrier comprises a nitrocellulose membrane (most widely used), a glass cellulose membrane or a nylon membrane, and further preferably, the membrane carrier is further coated with a positive control, and the corresponding protein marker and the positive control are sequentially arranged on the nitrocellulose membrane according to the detection sequence.

According to different detection methods of the kit, specific matching reagents in the kit are different correspondingly, but the matching reagents can be combined according to the preparation mode of the known kit. Preferably, the kit further comprises at least one of the following components: (1) An enzyme-labeled secondary antibody, more preferably an HRP-labeled secondary antibody (corresponding to an ELISA detection kit); (2) A colloidal gold conjugate pad coated with a specific conjugate of a colloidal gold-labeled antibody and a positive control (corresponding to an immune colloidal gold assay kit); (3) The kit comprises a labeling pad, wherein the labeling pad is coated with microspheres which are fluorescently labeled, and specific binders (corresponding to an immunofluorescence detection kit) of a positive control are loaded on the microspheres.

The immune colloidal gold detection kit and the immune fluorescent detection kit are relatively convenient to detect, and only a positive control C line and a detection sample T line are required to be established. The positive control pre-coated at the C-line of the positive control is not particularly limited as long as it can be bound with the specific binding substance with the detection label carried along with the chromatography of the solution to be detected in the sample to be detected, and the specific antibody of the positive control is not particularly limited. Preferably, the positive control is selected from a mouse immunoglobulin, a human immunoglobulin, a sheep immunoglobulin or a rabbit immunoglobulin, and correspondingly, the specific binding substance of the positive control is selected from an anti-mouse immunoglobulin, an anti-human immunoglobulin, an anti-sheep immunoglobulin or an anti-rabbit immunoglobulin.

The anti-mouse immunoglobulin may be a goat anti-mouse immunoglobulin, a rabbit anti-mouse immunoglobulin, or other immune animal anti-mouse immunoglobulin, depending on the subject to be immunized. Similarly, the anti-human, anti-sheep or anti-rabbit immunoglobulin may be derived from different species depending on the animal to be immunized. The immunoglobulin may be any of IgM, igG, igA, igD or IgE. These anti-immunoglobulin antibodies may be monoclonal or polyclonal.

In the kit, the specifications of the used ELISA plates are different according to the number of samples to be detected, and the ELISA plates can be reasonably selected from 12-384-hole ELISA plates.

The kits with different forms can realize the quantification of the protein, namely the quantification of each protein marker in the serum. For example, protein markers in serum, such as CDCP1, react with CDCP1 antibodies on the surface of a solid support, as measured using an ELISA kit. Then, an enzyme-labeled antibody is added thereto, and the resulting product is bound to a solid phase carrier by a reaction. At this time, the amount of enzyme on the solid phase is in a certain ratio to the amount of CDCP1 protein in the serum. After the substrate of the enzyme reaction is added, the substrate is catalyzed by the enzyme to form a colored product, and the amount of the colored product is directly related to the amount of the CDCP1 protein in serum, so that qualitative or quantitative analysis can be carried out according to the shade of color. Because the catalytic efficiency of the enzyme is high, the result of the immune reaction is indirectly amplified, and the determination method achieves high sensitivity.

In addition, the detection kit may also be in the form of a detection chip, for example, a plurality of or all of the antibodies of protein markers may be simultaneously disposed on the chip to achieve more efficient detection.

In a third exemplary embodiment of the present application, a detection apparatus for distinguishing early NASH from NAFL is provided, the detection apparatus is provided with a detection model for distinguishing early NASH from NAFL, wherein the detection model is a model for detecting protein markers, and the protein markers include a plurality of protein markers detected in any one of the above application schemes. According to the research result of the application, the found 22 protein markers are utilized, different detection models can be established according to different model construction methods according to any multiple protein markers in the 22 protein markers according to different clinical classification standards, and the detection models can well distinguish early NASH from NAFL.

The model may be constructed using a variety of known classification models, such as logistic regression models, random forest models, or SVC models. In some preferred embodiments, the detection model is a logistic regression model.

In some preferred embodiments, the detection means comprises a storage medium on which the detection model is stored.

In some preferred embodiments, the detection means comprises a processor for running the detection model.

It should be noted that, when the 22 protein markers of the present application are determined, different detection models are established according to different algorithms, which can be easily implemented by those skilled in the art using the existing model establishment methods, and therefore, any detection model established using the 22 protein markers of the present application is suitable for the present application. After the detection model is established, the detection model can be built in any electronic device. In particular, the detection device can be arranged on a storage medium or a processor, and can realize the distinguishing of the early NASH and the NAFL no matter what arrangement mode is.

Depending on the specific device, in some preferred embodiments, the detection device comprises a module for receiving the expression level of the protein marker in the sample to be tested, and the receiving module comprises at least one of the following modes: the user manually inputs a mode, an alternative list import mode and a file import mode. Different acceptance modes provide diversified choices for the subjects, and improve the convenience of user detection.

The above-mentioned sample that awaits measuring is the body fluid sample, specifically, can be various body fluids, for example, serum, urine, saliva, pleural effusion, peritoneal cavity hydrops etc.. In the present application, the sample to be tested is preferably a serum sample.

In addition, the test sample can be used to screen obese patients (who may be early NASH, or NAFL, or a healthy person), or NAFLD patients. Note that NAFLD patients herein do not include liver fibrosis and cirrhosis patients, but include only NASH and NAFL.

In a fourth exemplary embodiment of the present application, there is provided a detection method for distinguishing early NASH from NAFL, the detection method comprising: detecting the expression quantity of the protein marker in the body fluid of the subject to obtain the expression quantity of the protein to be detected; inputting the expression quantity of the protein to be detected into a detection model for distinguishing early NASH from NAFL, and outputting a detection result; wherein the detection model is a model for detecting protein markers, and the protein markers comprise protein markers detected in any one of the kits. By utilizing the 22 protein markers discovered by the application, different detection models can be established according to any multiple of clinical classification standards, and the detection models can well distinguish early NASH from NAFL.

The detection model can adopt various known classifier models, such as a random forest model and an SVC model, and can well classify the two types of patients. In some preferred embodiments, the detection model is a logistic regression model.

Such subject body fluids include, but are not limited to, serum, urine, saliva, pleural effusion, peritoneal effusion, and the like. Serum is preferred in this application.

The subject may be any human, such as an obese patient or a NAFLD patient. In practice, the sample to be tested is usually a number of obese patients, some of whom may suffer from liver diseases and some of whom may be healthy, so that the scheme of the application is also suitable for early NASH screening of obese patients. Note that NAFLD patients herein do not include liver fibrosis and cirrhosis patients, but include only NASH and NAFL.

The advantageous effects of the present application will be described in detail with reference to specific examples.

Example 1 screening of protein markers

Queue samples

The invention utilizes the serum samples of 135 fat metabolic disorder patients who are obese and have metabolic surgery, and the corresponding perfect clinical information and tissue slice pathological data. According to three indexes in liver biopsy pathological section: scoring results for steatosis (steatosis), inflammation (inflammation) and balloon-like deformation (balloon) all samples (n = 135) were grouped.

NAFL group, also called simple fatty liver group: steatosis score greater than 0, with both other items scoring 0 (n = 55); and

NASH group: also known as the nonalcoholic steatohepatitis group: the scores for all three pathological indicators were greater than 0 (n = 80).

Since this cohort was designed for the early NASH population, there was no significant fibrosis in all liver biopsy samples. Clinical information includes sex, age, body Mass Index (BMI), waist-to-hip ratio (WHR), insulin resistance index (HOMAIR, C-peptide (C-peptide), alanine transferase (Alt), aspartate aminotransferase (Ast), and total bile acid (Tba), among others.

(II) protein detection

This example mainly uses Proximity extension assay (PEA; olink platform) technology to measure the concentration of the corresponding protein marker in blood samples. The PEA technology is an immunoassay method for detecting protein biomarkers in a liquid sample in a high-throughput manner, and is characterized in that a pair of matched antibodies coupled with DNA oligonucleotide single chains is used for detecting proteins (namely when two antibodies coupled with the DNA oligonucleotide single chains are only combined on the same target protein, the two DNA oligonucleotide single chains are combined to form double-stranded DNA molecules and are detected through real-time PCR extension amplification), and the detection result is an npx (normalized protein expression) value.

The present invention uses a total of 12 panels of detection (panel) containing 1066 candidate protein markers. Each test hole contains internal reference quality control substances (2 incubation quality controls, 1 extension quality control and 1 detection quality control), each detection panel is provided with a negative control, and a positive control and a plate-to-plate control sample are used for experimental quality control. The detection method conforms to the instruction manual of the kit detection. All samples were tested separately in 3 batches (see in particular tables 2 and 3 below).

Table 2:

/>

table 3:

(III) data preprocessing

Aiming at Olink protein npx data, firstly, carrying out quality control on a single sample according to the detection result of an internal reference quality control product. When the difference between the detection value of the reference quality control in a single test hole and the detection median value of the reference quality control in the whole plate is larger than a threshold value (0.35 NPX), the test hole is considered to be failed in quality control, the detection result of the sample is removed, and the detection result is recorded as a missing value NA. In addition, the lowest detection Limit (LOD) of each protein index was determined from the detection results of the negative control samples, and when the detection value was lower than the LOD, the protein index was regarded as undetected and was regarded as the deletion value ND. When the deletion ratio of the protein is >25%, the protein is rejected. When the protein deletion ratio in a single sample is >80%, the sample is rejected. Filling missing values NA in the numerical matrix by using a KNN method (a known algorithm for filling the missing values); the LOD is used to fill in for the missing value ND. And (3) carrying out batch correction on the data of a plurality of batches by using a median method to obtain a final data analysis matrix.

Missing values in the clinical data are filled in using the method of KNN for the clinical data.

(IV) protein marker screening

Data were randomly divided into a training set (70%) and a test set (30%), and protein markers were selected in the training set. The protein variables were first filtered by collinearity calculation, i.e. the pearson correlation coefficient matrix was calculated for all variables and clustering was performed using hclust. Only the most important variable in each cluster, i.e. the one with the smallest p-value in the t-test, is retained (t-test = Student's t-test, p-value = p-value). The coefficients for each variable in the logistic regression were then calculated by univariate logistic regression analysis (using the glm method), and the variable was retained when the p-value of the coefficient was significant (significant when p < 0.05). This resulted in 22 significantly different proteins (as shown in table 4) as protein markers to distinguish NAFL from early NASH. Among the 22 protein markers, one or more of ADAM-TS 15, ALCAM, CCL15, CCL28, CDCP1, FCER2, FGF-21, insulin, PON3, PTS, ROBO1, SCF, SIGLEC10, and SYND1 are more preferable.

Table 4:

/>

note: the expression "increase" or "decrease" in the above table means that the expression level in the population with NASH at the early stage is significantly increased or decreased compared to the expression level of the protein in the population with NFAL.

Example 2 evaluation of protein markers for the differentiation of NAFL from early NASH Performance

(1) Calculating AUC value of each protein marker region NAFL and early NASH to determine the accuracy of the protein marker.

The AUC of all the individual proteins obtained from the preliminary screening showed more than 0.639 (0.639-0.759) in differentiating NAFL from early NASH (1-AUC) when the expression was decreased in the population with early NASH) (see Table 4). This result demonstrates the ability of each of the 22 proteins to distinguish between early NASH and NAFL. Of these 22 protein markers, CDCP1 performed best in the test set (AUC = 0.759). The concentration of most proteins will gradually increase with disease progression, while the concentrations of CCL28, PON3 and SCF gradually decrease with disease progression (see fig. 1 a-1 b).

(2) Calculating the discriminative power of various protein combinations

Based on the protein markers found, random combinations of different numbers (n = 2-22) of proteins were used to establish the detection model. Various modes are tried to verify the detection capability of the random protein combination, for example, a logistic regression model, a random forest model and an SVC model are utilized, and the prediction results are found to be good. The effect will be described below by taking a logistic regression model as an example. In the following examples, logistic regression models were established for different combinations of protein markers, the models were trained in the training set, and the models were validated in the test set to calculate AUC and other values. Random arbitrary combinatorial proteins were found to have higher AUC and other values (see table 1). This result indicates that any two or more of these 22 proteins in combination have a high ability to discriminate NAFL from early NASH.

Among the preferred 14 proteins are: CCL15, SIGLEC1, ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF and SYND1, any two or more of these 14 proteins are combined, and the combination has good capacity of distinguishing early NASH from NAFL, and the combination with the serial numbers 1, 17, 20, 45, 54 and 55 in the table 1.

The 22 protein markers and the detection model established by using the protein markers can distinguish early NASH diseases from NAFL diseases under different application scenes, including NASH screening of healthy people, especially obese people. Specifically, by inputting the relative serum protein level of the patient (the input method is not limited to manual input, alternative list input or file import, etc.), and performing the corresponding algorithm model operation, the evaluation can be made as to whether the subject has NASH.

From the above description, it can be seen that the embodiments of the present application achieve the following technical effects: the NAFL disease and NASH disease treatment scheme is greatly different, so that the NASH and NAFL patients at early stage can be accurately distinguished, the risk of the disease becoming hepatocellular carcinoma is reduced, and the survival rate of the patients is improved. However, a commercially available noninvasive detection scheme capable of effectively distinguishing NAFLs from NASH at an early stage is lacked at present, and the protein marker and the detection product derived from the protein marker can fill up the diagnosis vulnerability to a certain extent, so that the distinguishing of NAFLD disease spectra is effectively improved. The clinical information acquisition of the patient and the measurement of the concentration of the protein marker in the serum are relatively simple, noninvasive and accurate, and the patient acceptance is high, so the method has important clinical application value.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (22)

1. Use of a protein marker for the manufacture of a test product for distinguishing early NASH from NAFL, wherein the protein marker comprises at least one of: CCL15, SIGLEC10, ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, FBP1, GAS6, GUSB, IL-1ra, KYNU, LILRB4, MVK, and TNF-R2.

2. The use of claim 1, wherein the protein markers comprise at least one of: CCL15 and SIGLEC10;

preferably, the protein marker further comprises at least one of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, FBP1, GAS6, GUSB, IL-1ra, KYNU, LILRB4, MVK, and TNF-R2.

3. The use according to claim 1 or 2, wherein the protein markers comprise at least two of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, and at least one of them is CCL15 or SIGLEC10.

4. The use according to claim 1, wherein the protein marker is selected from any one of the group consisting of:

/>

/>

/>

5. the use according to claim 1, wherein the test product is a test kit or a test device.

6. A kit for differentiating early NASH from NAFL, comprising a detection reagent for a protein marker comprising at least one of: CCL15 and SIGLEC10.

7. The kit of claim 6, wherein the protein markers further comprise at least one of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, FBP1, GAS6, GUSB, IL-1ra, KYNU, LILRB4, MVK, and TNF-R2.

8. The kit of claim 6, wherein the protein markers comprise at least two of: ALCAM, PON3, ROBO1, CDCP1, PTS, insulin, ADAM-TS 15, CCL28, FCER2, FGF-21, SCF, SYND1, and at least one of them is CCL15 or SIGLEC10.

9. The kit according to claim 6, wherein the protein markers are selected from any group of protein markers for the use according to claim 4.

10. A detection apparatus for distinguishing early NASH from NAFL, wherein a detection model for distinguishing early NASH from NAFL is built in the detection apparatus, wherein the detection model is a model for detecting protein markers, and the protein markers comprise a plurality of the protein markers in the application of any one of claims 1 to 4.

11. The detection apparatus of claim 10, wherein the detection model is a logistic regression model.

12. The inspection device of claim 11, wherein the inspection device comprises a storage medium on which the inspection model is stored.

13. The detection apparatus of claim 11, wherein the detection apparatus comprises a processor configured to run the detection model.

14. The test device according to any one of claims 10 to 13, wherein the test device comprises a module for accepting the expression level of the protein marker in the sample to be tested, wherein the module for accepting comprises at least one of the following modes: the user manually inputs a mode, an alternative list import mode and a file import mode.

15. The test device according to claim 14, wherein the sample to be tested is a body fluid sample, preferably a serum sample.

16. The test device of claim 14, wherein the test sample is derived from an obese patient.

17. The test device of claim 14, wherein the test sample is derived from a NAFLD patient.

18. A detection method for distinguishing early NASH from NAFL is characterized by comprising the following steps:

detecting the expression quantity of the protein marker in the body fluid of the subject to obtain the expression quantity of the protein to be detected;

inputting the expression quantity of the protein to be detected into a detection model for distinguishing early NASH from NAFL, and outputting a detection result;

wherein the detection model is a model for detecting a protein marker comprising the protein marker for use of any one of claims 1 to 4.

19. The detection method of claim 18, wherein the detection model is a logistic regression model.

20. The method of detecting according to claim 18, wherein the body fluid is serum.

21. The test method of claim 18, wherein the subject is selected from obese patients.

22. The assay of claim 18, wherein the subject is selected from a patient with NAFLD.

Images