KR20220076779A - Mitochondrial biomarker for non-invasive diagnosis of hepatic steatosis - Google Patents

Abstract

The present invention relates to a mitochondrial biomarker for non-invasive diagnosis of hepatic steatosis and uses thereof, and specifically, to a mitochondrial biomarker for non-invasive diagnosis of hepatic steatosis, and hepatic steatosis prediction or diagnosis using a non-invasive diagnostic index of hepatic steatosis based thereon. It relates to a method of providing information for

Description

Mitochondrial biomarker for non-invasive diagnosis of hepatic steatosis and use thereof

The present invention relates to a mitochondrial biomarker for non-invasive diagnosis of hepatic steatosis and uses thereof, and specifically, to a mitochondrial biomarker for non-invasive diagnosis of hepatic steatosis, and hepatic steatosis prediction or diagnosis using a non-invasive diagnostic index of hepatic steatosis based thereon. It relates to a method of providing information for

Non-alcoholic fatty liver disease (NAFLD) is one of the most common causes of chronic liver disease worldwide, ranging from simple hepatic steatosis to steatohepatitis (NASH). Simple hepatic steatosis has a benign character, whereas NASH is more likely to progress to cirrhosis and cancer. About 10-29% of NASH patients develop cirrhosis within 10 years, of which 4-27% develop hepatocellular carcinoma. Due to its high prevalence and severe progression, there is a need for reliable diagnostic and prognostic strategies for NAFLD. However, to date, there is no simple blood test panel or evaluation system for non-invasive evaluation of disease severity and treatment plan in NAFLD patients.

The etiology of NAFLD includes metabolic stress and mitochondrial dysfunction associated with obesity and insulin resistance. Therefore, NAFLD is often considered a liver symptom of metabolic syndrome. In the absence of a high-fat diet or physical activity, free fatty acids and triglycerides accumulate in the liver, generating reactive oxygen species (ROS) in the cytoplasm and mitochondria. Consequently, this oxidative stress causes severe mitochondrial dysfunction and endoplasmic reticulum (ER) stress, further exacerbating excessive ROS production in mitochondria and ER. The 'vicious cycle' between oxidative stress and organelle dysfunction leads to side effects including liver inflammation and cytotoxicity.

Adaptive responses to disease-causing mitochondrial stress include increased mitochondrial biogenesis, improved bioenergetic status, upregulated antioxidant defense systems, and management of mitochondrial function. This integrated stress response (ISR) contributes to metabolic and mitochondrial flexibility by reducing or delaying the onset of NAFLD. It is hypothesized that ISR may be induced by a mitochondrial stress-induced humoral factor called mitokine. Fibroblast growth factor 21 (FGF21) and growth differentiation factor 15 (GDF15) produced in the liver and other tissues are induced and secreted by mitochondrial stress and protect against mitochondrial damage and metabolic deterioration. plays a role Despite efforts to restore FGF21 and GDF15, persistent and uncompensated metabolic stress continues to stimulate the induction of these factors and keep their levels in serum high.

A recent serum biomarker analysis showed that FGF21 levels were elevated in patients with obesity and metabolic diseases related to insulin resistance. In particular, it has been reported that NAFLD patients, including fatty liver and steatohepatitis, have significantly higher serum FGF21 values than controls. However, serum FGF21 concentrations vary widely from individual to individual, which may impair their usefulness as a single indicator.

In addition, serum GDF15 levels have been shown to be increased in patients with hepatocellular carcinoma disease associated with cirrhosis and chronic hepatitis B or hepatitis C virus infection. Although the epidemiological role of GDF15 in liver disease is unclear, GDF15 stimulates the expression of transforming growth factor beta 1 (TGF-β1) as well as the SMAD signaling system that plays an important role in the liver fibrosis/carcinogenesis pathway. was found to directly activate

Therefore, as a result of the present inventors' efforts to develop a more accurate and more sensitive diagnostic index than the non-invasive diagnostic index for hepatic steatosis, the serum values of the mitochondrial biomarker panel including FGF21 were obtained using various analysis methods of NAFLD. , by confirming that mitochondrial stress biomarkers and the combination of these factors greatly improve the usefulness and effectiveness of the algorithm for predicting hepatic steatosis, thereby completing the present invention.

An object of the present invention relates to a mitochondrial biomarker for non-invasive diagnosis of hepatic steatosis and a method of providing information for predicting or diagnosing hepatic steatosis using an indicator for non-invasive diagnosis of hepatic steatosis based thereon.

In order to achieve the object of the present invention, the present invention

1) calculating a waist-to-hip ratio (WHR) from waist and hip circumference measurements measured from the subject;

2) fibroblast growth factor 21 (FGF21), fibroblast growth factor 19 (FGF19), adiponectin, leptin, insulin ( insulin), albumin, triglyceride (TG), total choresterol (TC) and alanine-aminotransferase (ALT) concentrations;

3) calculating an adiponectin-to-leptin ratio (A/L) from the adiponectin and leptin concentrations measured above;

4) WHR and A/L calculated above, and FGF21, FGF19, insulin, and albumin concentration values measured above are substituted for [Equation 1] below, and the metabolic stress index for hepetic steatosis; calculating MSI-S); and

[Equation 1]

5) It provides an information providing method for predicting or diagnosing hepatic steatosis, including comparing the MSI-S value with the cut-off value.

Also, the present invention

1) Among biological samples isolated from individuals, growth factor 21 (fibroblast growth factor 21; FGF21), fibroblast growth factor 19 (FGF19), adiponectin, leptin, insulin (insulin), a set of reagents for determining the concentrations of albumin, triglyceride (TG), total choresterol (TC) and alanine-aminotransferase (ALT); and

2) It provides a kit for predicting or diagnosing hepatic steatosis, including a device for measuring the waist and hip circumference of an individual.

In the present invention, waist-hip ratio, GDF15, FGF21, FGF19, adiponectin-to leptin ratio (A/L), insulin, albumin, triglyceride (TG), total cholesterol (TC) and It was confirmed that alanine-aminotransferase (ALT) is an independent predictor of hepatic steatosis, and using this, a metabolic stress index for hepatic steatosis (MSI-S) was derived. . In addition, when the MSI-S is used, the diagnostic accuracy is superior to that of the existing non-invasive indicators of hepatic steatosis, and the liver fat content can be predicted. It can be easily used as a biomarker.

1 is a diagram schematically illustrating a flow chart of a test participant recruitment and analysis subgroup according to an embodiment of the present invention.
Figure 2 is a road comparing the metabolic stress index for hepatic steatosis (MSI-S) and other hepatic steatosis indicators according to the present invention, Figure 2A is a non-invasive score for predicting hepatic steatosis It is a diagram showing the ROC curve (* p <0.002, ** p <0.001 vs. MSI-S; ¶ deletion data (n = 2)), and FIG. 2B is a diagram showing the non-invasive predictive score according to the fatty liver grade. where FLI is the fatty liver index, NLFS is the non-alcoholic fatty liver disease (NAFLD) liver fat score, and HSI is the hepatic steatosis index.
3 is a road confirming the predictability of MSI-S for liver fat content, and hepatic fat content and MSI measured by magnetic resonance imaging-proton density fat fraction (MRI-PDFF) -S represents the relationship between predicted values by a multivariate linear regression model using the same biomarkers as used in S. Scatter plots (left) and Bland-Altman plots (right) show that the relationship between measured and predicted values is consistent. The mean difference and 95% concordance limit (SD ± 1.96 of error) are indicated by black and dotted lines, respectively.
4 is a diagram showing the clinical usefulness of a hepatic steatosis predictive index having a sensitivity and specificity of 90%.

Hereinafter, the present invention will be described in more detail.

the present invention

1) calculating a waist-to-hip ratio (WHR) from waist and hip circumference measurements measured from the subject;

2) fibroblast growth factor 21 (FGF21), fibroblast growth factor 19 (FGF19), adiponectin, leptin, insulin ( insulin), albumin, triglyceride (TG), total choresterol (TC) and alanine-aminotransferase (ALT) concentrations;

3) calculating an adiponectin-to-leptin ratio (A/L) from the adiponectin and leptin concentrations measured above;

4) WHR and A/L calculated above, and FGF21, FGF19, insulin, and albumin concentration values measured above are substituted for [Equation 1] below, and the metabolic stress index for hepetic steatosis; calculating MSI-S); and

[Equation 1]

5) It provides an information providing method for predicting or diagnosing hepatic steatosis, including comparing the MSI-S value with the cut-off value.

In the method according to the invention, the subject is, for example, a human, a monkey, a cow, a horse, a sheep, a pig, a chicken, a turkey, a quail, a cat, a dog, a mouse, a rat, a rabbit or a guinea pig, preferably a guinea pig It may be a mammal, more preferably a human, but is not limited thereto.

In the method according to the present invention, the waist and hip circumference measurements in step 1) may be measured by a general waist and hip circumference measurement means. In addition, it is preferable that the WHR is a value obtained by dividing the measured waist circumference by the measured height circumference.

In the method according to the present invention, the biological sample in step 2) may be whole blood, plasma or serum, preferably whole blood. The biological sample may be isolated from the subject by conventional clinicopathological means, preferably obtained by blood collection with an injection needle.

In addition, the method for measuring the concentrations of FGF21, FGF19, adiponectin, leptin, insulin, albumin, TG, TC and ALT in step 2) is not particularly limited, but for example, an immunoassay (eg, chemiluminescence). immunoassay), colorimetric assay, or turbidity assay.

In addition, in step 2), the FGF21 and FGF19 concentration units are pg/mL, the adiponectin concentration unit is μg/mL, the leptin concentration unit is ng/mL, the insulin concentration unit is mU/L, and the albumin concentration unit is g /dL, TG and TC concentration units are mg/dL, and ALT concentration units are preferably IU/L.

In the method according to the present invention, if the MSI-S value is equal to or greater than the cut-off value, it can be determined as a hepatic steatosis patient. Preferably, if the MSI-F value is 23.9 or more, more preferably, if it is 49.43 or more, it may be provided as information for judging a patient with hepatic steatosis. In one embodiment of the present invention, when the cut-off value was 49.43, MSI-S discriminated hepatic steatosis with a specificity of 83%, and excluded hepatic steatosis with a sensitivity of 78%.

In the method according to the present invention, the liver fat content by substituting the WHR and A/L calculated above, the FGF21, FGF19, insulin, albumin, TG, TC, and ALT concentration values measured above into the following [Equation 2] It may further include the step of predicting.

[Equation 2]

.

.

Since the liver fat content value correlates with the liver fat content measured using MRI, the severity of hepatic steatosis can be identified from the liver fat content value, and it is effective in reducing unnecessary MRI scans.

In a specific embodiment of the present invention, the present inventors obtained clinical and laboratory test characteristics for participants in a population-based general cohort, and FGF21 had a high significance coefficient (Wald = 42.2) through univariate logistic regression. , p <0.001), confirming that there is a correlation with hepatic steatosis.

In addition, the present inventors confirmed that WHR, FGF21, FGF19, A/L, insulin, albumin, TG, TC and ALT were independent predictors for hepatic steatosis through multivariate regression analysis, and using this, Metabolic stress index (MSI-S) was derived.

In addition, the present inventors confirmed that the above index had higher AUROC [0.886 (0.85-0.92)] and diagnostic accuracy (81.1%) than other hepatic steatosis indexes, and that the severity of hepatic steatosis could be distinguished using the index.

Therefore, the predictors and the index MSI-S derived using them can be usefully used as strong biomarkers for non-invasive diagnosis of hepatic steatosis.

Also, the present invention

1) Among biological samples isolated from individuals, fibroblast growth factor 21 (FGF21), fibroblast growth factor 19 (FGF19), adiponectin, leptin, insulin (insulin) ), albumin, triglyceride (TG), total choresterol (TC) and alanine-aminotransferase (ALT) concentrations; and

2) It provides a kit for predicting or diagnosing hepatic steatosis, including a device for measuring the waist and hip circumference of an individual.

In the present invention, the kit may include a device for measuring the waist and hip circumference of an individual.

In addition, the kit may include reagents and materials for measuring and analyzing the concentration of the biomarker in a biological sample isolated from an individual, and confirming that the individual has hepatic steatosis. Specifically, a diagnostic kit of the present invention may include a needle, syringe, vial, or other device for obtaining and/or containing a sample from a subject. In addition, the kit may include at least one reagent specifically used to detect or quantify the biomarkers disclosed herein. That is, suitable reagents and techniques can be readily selected by one of ordinary skill in the art for inclusion in a kit for detecting or quantifying a biomarker.

In addition, it is preferable that the reagent measures the concentration of the protein. Specifically, it may include an appropriate reagent (eg, an antibody) for detecting a protein using an immunoassay (eg, a chemiluminescent immunoassay), a colorimetric assay, or a turbidity assay. The reagents may also include extraction buffers or reagents, amplification buffers or reagents, reaction buffers or reagents, hybridization buffers or reagents, immunodetection buffers or reagents, labeling buffers or reagents, and detection means.

The kit may also include a control, which may be a control sample, a reference sample, an internal standard, or previously generated experimental data. Controls may correspond to normal healthy individuals or individuals with a known disease state. Additionally, a control may be provided for each biomarker or a control may be a reference risk score.

The kit may also include one or more containers for each individual reagent. The kit may further comprise instructions for performing the methods described herein and/or for interpreting the results in accordance with any regulatory requirements. Additionally, software for analysis of detected biomarker concentrations, calculation of risk scores, and the like may be included in the kit. Preferably, the kit is packaged in a container suitable for commercial distribution, sale and/or use.

Hereinafter, the present invention will be described in detail by way of Examples and Experimental Examples.

However, the following Examples and Experimental Examples are merely illustrative of the present invention, and the content of the present invention is not limited to the following Examples and Experimental Examples.

<Example 1> Test participants

The subjects were 343 volunteers recruited from the Korean Genome and Epidemiology Study on Atherosclerosis Risk of Rural Areas in the Korean General Population (KoGES-ARIRANG), a general population-based cohort, through stratified random sampling. The trial protocol was approved by the Wonju Severance Christian Hospital Institutional Review Committee (IRB No. CR317131) based on the ethical standards of the Declaration of Helsinki. All trial participants were informed about the trial rationale and possible risks and gave written informed consent prior to participation.

Specifically, as shown in Figure 1, test subjects were recruited from 1894 individuals who completed the third follow-up survey (May 2011 to October 2017) among those already enrolled in the population-based general cohort, KoGES-ARIRANG. did Elderly (85 years and older), alcohol abuse (male, > 30 g/day; female, > 20 g/day), malignancy, chronic viral hepatitis (hepatitis B and/or C), drug-induced liver injury , subjects with a clinical history of secondary causes or alternative diagnoses of hepatic steatosis, including autoimmune liver disease and Wilson disease, were excluded.

From May 8, 2018 to August 8, 2019, a total of 348 subjects were finally enrolled in the order in which they responded to the e-mail or phone recruitment. Of the 348 enrolled individuals, 343 participants (124 males and 219 females) were included in the analysis and 5 subjects were excluded. Based on a comprehensive evaluation of the liver, 135 patients with nonalcoholic fatty liver disease (NAFLD) and 199 control subjects (excluding 9 subjects with hepatic fibrosis only) were selected as hepatic steatosis predictive models. For the predictive model of hepatic fibrosis in the NAFLD population, 22 patients with severe liver fibrosis and 122 patients with hepatic steatosis without fibrosis were selected.

<Example 2> Ultrasonography

All ultrasound examinations were performed by a hepatologist using an Aplio i500 (Canon Medical Systems, Otawara, Japan) with a low frequency convex transducer. Four ultrasound signs (abnormal hepatorenal echo, loss of echogenicity of portal vein, poor diaphragm visualization and posterior beam attenuation) were evaluated to assess the severity of fatty liver.

<Example 3> Magnetic resonance imaging-proton density fat fraction (MRI-PDFF)

MRI was performed on a 3T system (Magnetom Skyra; Siemens, Erlangen, Germany) equipped with a 30-channel body and a 32-channel spine matrix coil. Subjects were examined in the supine position, and all images were acquired while holding exhaled breaths. Multi-echo Dixon mapping was performed to evaluate liver fat and iron content: echo time (TE) = 1.05, 2.46, 3.69, 4.92, 6.15 and 7.38 ms; repetition time (TR) = 9.00 ms; flip angle = 4°; field of view (FOV) = 450 × 393 mm ² ; matrix size = 160 × 111; slice thickness = 3.5 mm; number of slices = 72; Controlled aliasing = 2 × 2 in parallel imaging results at higher acceleration factors; acquisition time = 13 s. Screening Dixon and multi-echo Dixon sequences were performed sequentially. Water image, fat image, fit image, MRI-PDFF map and screening report and Multi-echo Dixon were automatically collected. Hepatic steatosis was graded according to the following criteria: grade 0, PDFF <6.4%; Grade 1, 6.4% ≤ PDFF <16.3%; Grade 2, 16.3% ≤ PDFF <21.7%; Grade 3, PDFF ≥ 21.7%.

<Example 4> Magnetic resonance elastography (MRE)

A continuous longitudinal mechanical wave was generated at 60 Hz using a driver device that delivered sound pressure waves to the anterior chest wall. Spin-echo echo-planar imaging sequences were acquired during apnea in three axial planes: TR = 500 ms; TE = 39 ms; FOV = 380 × 309 mm ² ; matrix size = 100 × 100 mm; slice thickness = 5 mm; Acceleration factor = 2, acquisition time = 6 s for generalized autocorrection of partially parallel acquisitions.

In each MR image obtained from MRE, regions of interest (ROI) were manually set to include only the parenchyma of the liver by two experts with 5 and 6 years of MRE experience, respectively. To evaluate repeatability, two measurement sessions were performed at 2-week intervals. At this time, the patient's biochemical and fat quantitative data information was not known. Corresponding size images, anatomical images, and wave images were simultaneously evaluated to exclude regions directly under the driver, bile ducts, blood vessels, and other regions in which wave propagation was not consistent. The measurement was limited to the area where the reliability parameter of the stiffness restoration was 95% or more. Liver stiffness was averaged from values obtained from three manually selected ROIs (one ROI for each image slice) of the whole liver. After two measurements at 2-week intervals, the fat fraction obtained from the multi-echo Dixon acquisition was recorded for each patient. The reproducibility between the two readings and the repeatability of liver spasticity measurements using MRE were evaluated by discriminating ICC. As shown in Table 1 below, it was confirmed that the repeatability of the liver stiffness values obtained using ICC and MRE for reproducibility was almost perfectly consistent.

Reproducibility and repeatability of measurements obtained from MRE ICC (95% CI) p-value Inter-observer reproducibility 0.955 (0.948-0.962) <.001 repeatability reading 1 0.983 (0.980-0.987) <.001 reading 2 0.936 (0.921-0.949) <.001

Hepatic fibrosis was graded according to the following criteria: normal, stiffness < 2.5 kPa; normal to inflammatory, 2.5 kPa ≤ stiffness < 2.9 kPa; Stage 1-2 hepatic fibrosis, 2.9 kPa ≤ stiffness < 3.5 kPa; Stage 2-3, 3.5 kPa ≤ Stiffness < 4.0 kPa, Stage 3-4, 4.0 kPa ≤ Stiffness < 5.0 kPa.

<Example 5> Clinical and laboratory tests

Body mass index (BMI), waist-to-height ratio (WHtR), waist-to-hip ratio; WHR) was calculated.

Participants were measured for height and weight in light clothes and without shoes. Waist circumference was measured midway between the lowest rib and the upper border of the iliac crest. Hip circumference was measured around the widest part of the hip. The circumference was given as the average of the two measurements closest to 0.1 cm. Body mass index (BMI) was calculated by dividing the weight in kg by the square of the height in m.

Blood samples were taken from the antecubital vein for analysis of biochemical parameters and NAFLD-related biomarkers under fasting conditions for at least 10 hours. The isolated serum was immediately stored at -80°C for further testing. Serum concentrations of biochemical analytes were measured using a calibrated Roche Cobas®8000 modular analyzer configured with c702a and e801 modules using the manufacturer's reagents and calibration equipment (Roche, Mannheim, Germany). Triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), γ-glutamyl transferase (γ-GT) and silver urate enzymatic-colorimetric method was measured. Aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), creatinine, albumin, blood urea nitrogen (BUN), protein and total bilirubin were measured by a colorimetric method. Fasting blood glucose was measured by the hexokinase method. Insulin and C-peptide were identified by electro-chemiluminescence immunoassay. Phosphorus and calcium were measured by Molybdate UV and NM-BAPTA methods, respectively. Platelet count was analyzed using an automated blood cell counter (ADVIA 2110I, Bayer, NY, USA).

Routine biochemical tests including liver function parameters were performed using an automated clinical chemistry analyzer. Serum concentrations of 10 metabolic stress-related biomarkers, including FGF21 and GDF15, were quantified using a commercially available ELISA kit according to the manufacturer's procedure. Serum concentrations of FGF21, FGF19, GDF15, adiponectin, leptin, RBP4, IL6, TGF-β1 and myostatin were measured using a human Quantikine ELISA kit (R&D Systems, Minneapolis, MN, USA). Using , decorin concentration was quantified according to the manufacturer's procedure using a Raybio human DCN ELISA kit (RayBiotech, Norcross, GA, USA). Each assay was performed in 4 separate batches. All mean intra- and inter-assay coefficients of variation were less than 10%.

The homeostatic model assessment index (HOMA-IR) was calculated as [Fasting insulin (μU/ml) × Fasting glucose (mg/dl)] / 405]. Diabetes was defined as a fasting blood glucose level of ≥126 mg/dL, a previous diagnosis of diabetes, or the use of antidiabetic drugs. Hypertension was defined as blood pressure ≥140/90 mmHg, previously diagnosed with hypertension and/or using antihypertensive drugs. Dyslipidemia was defined as total cholesterol ≥240 mg/dL, HDL-cholesterol <40 mg/dL (male), <50 mg/dL (female), and/or anti-dyslipidemia treatment. Metabolic syndrome was defined according to the International Diabetes Federation criteria - central obesity (waist circumference ≥90 cm in men and ≥ 80 cm in women) and the presence of two or more of the following components: (1) Serum Triglycerides ≥150 mg/dL or on specific treatment of dyslipidemia; (2) Serum high-density lipoprotein (HDL) < 40 mg/dL for men, <50 mg/dL for women, or if receiving specific treatment for hypoHDL-cholesterolemia; (3) systolic blood pressure (BP) ≥ 130 mmHg or diastolic BP ≥ 85 mmHg, or has received treatment for previously diagnosed hypertension; and (4) fasting plasma glucose > 100 mg/dL or previously diagnosed with type 2 diabetes.

<Example 6> Existing predictive index of hepatic steatosis or liver fibrosis

Several predictive scores derived from clinical and laboratory indicators were calculated to compare diagnostic performance for hepatic steatosis or severe hepatic fibrosis. Fatty liver index (FLI), NAFLD liver fat score (NLFS) and hepatic steatosis index (HIS) were calculated as non-invasive hepatic steatosis index. As indicators of liver fibrosis, AST to platelet ratio index (APRI), fibrosis-4 index (FIB4), and NAFLD fibrosis score (NFLD fibrosis score; NFS) were calculated. All predictive indicators for hepatic steatosis or hepatic fibrosis are freely available because they use clinical and laboratory parameters and have fair or good accuracy in the derived population. Although diagnostic cut-off points for the derivation population were reported, the optimal cut-off values of the scores were recalculated to achieve maximum performance based on this cohort. Each equation is shown in [Table 2] below.

Non-invasive predictive equations for hepatic steatosis or hepatic fibrosis Indicators equation FLI = (e ^{0.953*Loge(TG) + 0.139*BMI + 0.718*Loge(γ-GT) + 0.053*WC - 15.745} ) / (1 + e ^{0.953*Loge(TG) + 0.139*BMI + 0.718*Loge(γ) -GT) + 0.053*WC - 15.745} ) * 100 NLFS = -2.89 + 1.18*Metabolic syndrome (yes =1, no = 0) + 0.45*Diabetes mellitus (yes = 2, no = 0) + 0.15*Insulin + 0.04*AST - 0.94*AST/ALT ratio HSI = (e ^{0.315*BMI + 2.421*ALT/AST ratio + 0.630*Diabetes mellitus (yes = 1, no = 0) - 9.960} ) / (1 + e ^{0.315*BMI + 2.421*ALT/AST ratio + 0.630*Diabetes mellitus (yes = 1, no = 0) - 9.960} ) *100 APRI = AST (/ULN) / platelet (10 ⁹ /L) *100 FIB-4 = age (years) * AST (U/L) / (platelets (10 ⁹ /L) * ALT (U/L) ^1/2 ) NFS = -1.675 + 0.037*Age + 0.094*BMI + 1.13*Diabetes mellitus (yes = 1, no = 0) + 0.99*AST/ALT ratio - 0.013*platelets (10 ⁹ /L) - 0.66*albumin

FLI - fatty liver index; NLFS - non-alcoholic fatty liver disease (NAFLD) liver fat score; HSI - hepatic steatosis index; APRI—AST to platelet ratio index; FIB-4 - Fibrosis-4 index; NFS - NAFLD non-alcoholic fatty liver disease (NAFLD) fibrosis score; AST—aspartate-aminotransferase; ALT - alanine-aminotransferase; γ-GT - γ-glutamyl transferase (γ-glutamyltransferase); ULN - upper limit of normal.

<Example 7> Statistical analysis

Data were analyzed using SPSS 25.0 software (IBM Corp., Armonk, NY, USA). A 2-sided P-value of less than 0.05 was considered statistically significant. A univariate descriptive statistic was used to compare subjects with/without hepatic steatosis or significant hepatic fibrosis. The continuity data were tested for normality using the Shapiro-Wilk test and expressed as the median of the interquartile range (IQR). Categorical data are expressed as proportioned frequencies. Comparison of continuity data between subgroups was appropriately performed by Student's t-test, Mann-Whitney's U test, or Kruskal Wallis test followed by Dunnett's T3 hoc test. Categorical data were analyzed using a chi-square test. Correlation between continuity data was evaluated by Spearman's correlation coefficient (r).

All variables were included in multivariate forward stepwise logistic regression analysis to identify variables independently related to the presence or absence of NAFLD or hepatic fibrosis. Variables with P < 0.05 by multivariate logistic regression analysis were used to construct the predictive scoring system. Non-parametric data were used as independent variables after natural log transformation. The whole model was calibrated using the Hosmer-Lemeshow goodness-of-fit test, and global performance and predictability were tested using Nagelkerke R ² . The contribution of each variable to the multivariate model was evaluated as a Wald chi-square value (Wald χ ² ) calculated by squaring the ratio of the regression coefficient divided by the standard error.

The area under the curve of the receiver-operator characteristics curve (AUROC) was calculated to evaluate the diagnostic performance of the predictive model. Other measures such as sensitivity, specificity, positive likelihood ratio (LR+), negative likelihood ratio (LR-), positive predictive value (PPV), and negative predictive value (NPV) also affect model performance. used for evaluation. The diagnostic accuracy of the scoring system was evaluated as follows: Accuracy = sensitivity × prevalence + specificity × (1-prevalence).

AUROC was compared using DeLong's method. The optimal cut-off was identified by maximizing the sum of sensitivity and specificity in the Youden index: the value corresponding to the highest Youden index maximizing sensitivity and specificity and ≥ 90% sensitivity (ruling- low threshold for out) and values corresponding to ≥90% specificity (high threshold for ruling-in). To construct predictive equations for liver fat percentage and level of liver stiffness, multivariate forward stepwise linear regression analyzes were performed using the same variables used in the scoring system for NAFLD or hepatic fibrosis. To evaluate the reliability of quantitative prediction, intraclass correlation coefficient (ICC) was calculated. ICC values of 0.90-1.00, 0.75-0.90, 0.5-0.75, and 0-0.50 were considered good, good, average and poor, respectively. A Bland-Altman plot was constructed to evaluate the equivalence between the measured and predicted values.

<Experimental Example 1> Confirmation of test participant characteristics

It was confirmed that the average age and BMI of the test participants were 66 years (interquartile range; IQR: 61-72) and 25.0 kg/m ² (IQR 23.2-27.5), respectively. It was found that 21.6%, 46.9%, 39.4% and 6.4% of the test subjects had type 2 diabetes mellitus, metabolic syndrome, NAFLD (MRI-PDFF ≥ 6.4%) and hepatic fibrosis (MRE ≥ 2.9 kPa), respectively. The clinical and laboratory test characteristics of the subjects are shown in Table 3.

Clinical characteristics and biochemical values of subjects with/without hepatic steatosis variable Subjects without hepatic steatosis
(MRI-PDFF < 6.4) Subjects with hepatic steatosis
(MRI-PDFF ≥6.4) P-value n (M/F) 199 (63/136) 135 (56/79) 0.066 Liver fat (%, MRI-PDFF) 2.9 (2.2, 4.4) 10.4 (8.3, 14) < 0.001 Liver fat (ln %, MRI-PDFF) 1.1 (0.8, 1.5) 2.3 (2.1, 2.6) < 0.001 Age (years) 66 (61, 72) 65 (61, 71) 0.426 BMI (kg/m ² )* 24.1 (22.3, 25.8) 26.8 (24.9, 28.4) < 0.001 Waist (cm) 81 (76, 88) 89 (84, 95) < 0.001 WHtR* 0.52 (0.49, 0.55) 0.55 (0.53, 0.59) < 0.001 WHR* 0.88 (0.85, 0.92) 0.93 (0.9, 0.96) < 0.001 Systolic blood pressure (mmHg) 126 (112, 140) 129 (120, 140) 0.241 Diastolic blood pressure (mmHg) 80 (70, 84) 80 (76, 86) 0.127 Triglyceride (mg/dL) 112 (86, 165) 165 (123, 242) < 0.001 Total cholesterol (mg/dL) 173 (151, 198) 185 (154, 209) 0.031 High-density lipoprotein (mg/dL) 54 (44, 65) 48 (40, 58) 0.001 Fasting glucose (mg/dL) 99 (93, 105) 102 (95, 113) 0.012 Insulin (mU/L) 6 (4, 10) 11 (7, 21) < 0.001 HOMA-IR 1.5 (0.9, 2.6) 2.8 (1.7, 5.3) < 0.001 AST (IU/L) 22 (19, 25) 23 (20, 29) 0.040 ALT (IU/L) 17 (13, 21) 21 (17, 28) < 0.001 ALT/AST 0.77 (0.64, 0.92) 0.94 (0.77, 1.12) < 0.001 GGT (IU/L) 18 (13, 25) 25 (17, 46) < 0.001 ALP (IU/L) 69 (58, 82) 67 (55, 83) 0.869 Albumin (g/dL) 4.6 (4.5, 4.8) 4.6 (4.6, 4.8) 0.071 Uric acid (mg/dL) 4.5 (3.9, 5.3) 5.1 (4.1, 5.8) < 0.001 Total bilirubin (mg/dL) 0.4 (0.31, 0.55) 0.43 (0.32, 0.56) 0.375 Protein (g/dL) 7.4 (7.1, 7.6) 7.5 (7.3, 7.7) 0.002 Blood urea nitrogen (mg/dL) 16 (13, 19) 15 (13, 18) 0.210 Creatinine (mg/dL) 0.72 (0.63, 0.86) 0.77 (0.67, 0.93) 0.029 Phosphorus (mg/dL) 3.7 (3.4, 4) 3.8 (3.5, 4.1) 0.063 Calcium (mg/dL) 9.5 (9.3, 9.8) 9.7 (9.4, 9.9) < 0.001 C-Peptide (ng/mL) 1.9 (1.4, 2.7) 3 (2.1, 4.4) < 0.001 GDF15 (pg/mL) 829.3 (672, 1142.8) 1015.9 (708.5, 1387.7) 0.002 FGF21 (pg/mL) 171.4 (112.3, 275.8) 306.8 (192.4, 486.4) < 0.001 FGF19 (pg/mL) 197.7 (119.4, 325.5) 148.4 (86.2, 248) < 0.001 Adiponectin (μg/mL) 7.08 (4.43, 12.6) 4.66 (2.57, 7.82) < 0.001 Leptin (ng/mL) 6.93 (3.65, 11.68) 10.13 (6.18, 16.35) < 0.001 A/L (103) 1.09 (0.55, 2.38) 0.46 (0.26, 0.76) < 0.001 RBP4 (μg/mL) 29.9 (25.1, 34.7) 31.6 (27.2, 38.9) 0.008 IL6 (pg/mL) 1.44 (1.04, 2.26) 1.91 (1.26, 2.72) < 0.001 TGF-β1 (ng/mL) 23.6 (19.1, 29.3) 25.4 (20.5, 30.8) 0.025 Myostatin/GDF8 (ng/mL) 2.6 (1.97, 3.19) 2.58 (2, 3.17) 0.841 Decorin (ng/mL) 6.62 (5.45, 7.69) 6.33 (4.99, 7.72) 0.452 Fatty liver index 19.2 (9.8, 35.1) 49.7 (31.5, 68) < 0.001 NAFLD Liver Fat Score -1.8 (-2.6, -0.57) -0.1 (-1.35, 1.4) < 0.001 Hepatic steatosis index 0.41 (0.24, 0.61) 0.7 (0.5, 0.84) < 0.001 Central obesity (%) 86 (43) 102 (76) < 0.001 Dyslipidemia (%) 79 (40) 56 (42) 0.745 Hypertension (%) 89 (45) 73 (54) 0.093 Type 2 diabetes (%) 35 (18) 37 (27) 0.032 Metabolic syndrome (%) 67 (34) 88 (57) < 0.001 Drugs Antidiabetics 34 (17) 37 (27) 0.024 Lipid-lowering agents 68 (34) 46 (34) 0.985 Antihypertensive agents 88 (44) 73 (54) 0.077 NAFLD, non-alcoholic fatty liver disease; MRI-PDFF, magnetic resonance imaging-proton density fat fraction; BMI, body mass index; WHtR, waist-to-height ratio; WHR, waist-to-hip ratio; HOMA-IR, homeostatic model assessment of insulin resistance; AST, aspartate-aminotransferase; ALT, alanine-aminotransferase; γ-GT, γ-glutamyltransferase; ALP, alkaline-phosphatase; A/L, adiponectin-to-leptin ratio; GDF15, growth differentiation factor 15; FGF21, fibroblast growth factor 21; FGF19, fibroblast growth factor 19; RBP4, retinol binding protein 4; IL6, interleukin 6; TGF-β1, transforming growth factor-beta 1. Continuous variables are presented as median (IQR) and compared using Mann-Whitney U test or independent t -test*. Categorical data are expressed as sample size (proportion) and analyzed by Pearson's chi-squared test.

<Experimental Example 2> Confirmation of metabolic stress index for hepatic steatosis

Independent predictors derived from logistic regression analysis using NAFLD as a dependent variable are shown in Table 4.

Univariate and Multivariate Step Forward Logistic Regression for Prediction of Hepatic Steatosis univariate multivariate variable Coefficient (95% CI) S.E. Wald P-value Coefficient (95% CI) S.E. Wald P-value Body mass index (kg/m ² ) 0.271 (0.188-0.353) 0.04 41.06 < 0.001 Ln [waist (cm)] 8.772 (6.15-11.393) 1.34 43.02 < 0.001 WHtR 14.86 (10.051-19.668) 2.45 36.69 < 0.001 WHR 16.17 (11.365-20.974) 2.45 43.51 < 0.001 14.079 (7.731-20.427) 3.24 18.89 < 0.001 Ln [triglyceride (mg/dL)] 1.508 (1.021-1.996) 0.25 36.82 < 0.001 0.808 (0.171-1.445) 0.33 6.18 0.013 Ln [total cholesterol (mg/dL)] 1.151 (0.036-2.267) 0.57 4.10 0.043 1.878 (0.32-3.436) 0.80 5.58 0.018 Ln [high-density lipoprotein (mg/dL)] -1.564 (-2.42--0.708) 0.44 12.82 < 0.001 Ln [fasting glucose (mg/dL)] 1.472 (0.261-2.684) 0.62 5.67 0.017 Ln [fasting insulin (mU/L)] 1.143 (0.797-1.488) 0.18 42.00 < 0.001 0.652 (0.166-1.138) 0.25 6.90 0.009 Ln (HOMA-IR) 0.959 (0.659-1.259) 0.15 39.24 < 0.001 Ln [AST (IU/L)] 1.269 (0.459-2.079) 0.41 9.42 0.002 Ln [ALT (IU/L)] 1.837 (1.175-2.499) 0.34 29.60 < 0.001 0.833 (0.018-1.648) 0.42 4.01 0.045 Ln [ALT/AST] 2.221 (1.336-3.105) 0.45 24.22 < 0.001 Ln [γ-GT (IU/L)] 1.284 (0.855-1.714) 0.22 34.37 < 0.001 Ln [albumin (g/dL)] 6.146 (1.455-10.838) 2.39 6.59 0.010 8.206 (1.816-14.596) 3.26 6.34 0.012 Ln [uric acid (mg/dL)] 1.391 (0.522-2.26) 0.44 9.85 0.002 Ln [protein (g/dL)] 5.679 (1.492-9.865) 2.14 7.07 0.008 Ln [calcium (mg/dL)] 11.478 (5.135-17.82) 3.24 12.58 < 0.001 Ln [C-Peptide (ng/mL)] 1.373 (0.928-1.818) 0.23 36.52 < 0.001 Ln [GDF15 (pg/mL)] 0.837 (0.337-1.337) 0.26 10.76 0.001 Ln [FGF21 (pg/mL)] 1.269 (0.886-1.652) 0.20 42.18 < 0.001 0.888 (0.418-1.358) 0.24 13.72 < 0.001 Ln [FGF19 (pg/mL)] -0.542 (-0.837--0.248) 0.15 13.01 < 0.001 -0.579 (-0.967--0.191) 0.20 8.56 0.003 Ln [adiponectin (μg/mL)] -0.87 (-1.196--0.544) 0.17 27.39 < 0.001 Ln [leptin (ng/mL)] 0.693 (0.4-0.985) 0.15 21.57 < 0.001 Ln [A/L (10 ³ )] -0.843 (-1.096--0.591) 0.13 42.74 < 0.001 -0.469 (-0.808--0.13) 0.17 7.36 0.007 Ln [RBP4 (μg/mL)] 0.892 (0.1-1.685) 0.40 4.87 0.027 Ln [IL6 (pg/mL)] 0.463 (0.135-0.791) 0.17 7.67 0.006 Constant -45.426 (-61.192--29.66) 8.04 31.89 < 0.001 CI, confidence interval; S.E., standard error; ln, natural logarithm; WHtR, waist-to-height ratio; WHR, waist-to-hip ratio; HOMA-IR, homeostatic model assessment of insulin resistance; AST, aspartate-aminotransferase; ALT, alanine-aminotransferase; γ-GT, γ-glutamyltransferase; ALP, alkaline-phosphatase; GDF15, growth differentiation factor 15; FGF21, fibroblast growth factor 21; FGF19, fibroblast growth factor 19; A/L, adiponectin-to-leptin ratio; RBP4, retinol binding protein 4; IL6, interleukin 6.

In univariate logistic analysis, the mitochondrial stress biomarker showed a high significance coefficient (Wald c ² for FGF21; 42.2, p < 0.001), followed by central obesity (Wald c ² for WHR; 43.5, p <0.001). was confirmed. It was confirmed that serum FGF19 and adiponectin to leptin ratio (adiponectin-to-leptin ratio; A/L) showed a negative correlation with hepatic steatosis. In multivariate logistic analysis, natural logarithm of WHR and FGF21, FGF19, A/L, insulin, albumin, total triglyceride (TG), total cholesterol (TC) and aminotransferase (ALT) was selected as a significant independent predictor (the factors are listed in order of Wald c ² values), and the metabolic stress index for steatosis (MSI-S) for hepatic steatosis of the following [Equation 1] was derived. was used to

Hosmer-Lemeshow statistic was not significant (χ ² = 4.24, p = 0.835), and Nagelkerke R ² was 0.547, confirming the excellent fit of the MSI-S model. AUROC of predicted probability values for MSI-S was 0.886 (95% CI 0.85-0.92), with FLI (AUROC [95 % CI]; 0.807 [0.76 -0.85]), NLFS (0.755 [0.70 - 0.81]) and HSI ( 0.770 [0.72-0.82]), and it was confirmed that it showed excellent diagnostic performance compared to the existing indicators for hepatic steatosis (FIG. 2A). Compared to USG (0.825 [0.78-0.87]), it was found that MSI-S also exhibited a higher AUROC (0.884, [0.85-0.92]) in participants suffering from USG. At the optimal cutoff value of 49.43, MSI-S excluded NAFLD with a sensitivity of 78% (95% CI 70-83) and a negative likelihood ratio of 0.27 (95% CI 0.19-0.37), and a specificity of 83% It was confirmed that NAFLD was discriminated with (95% CI 78-88) and a positive likelihood ratio of 4.6 (95% CI 3.4-6.5). Therefore, it can be seen that MSI-S has significantly better diagnostic accuracy (81.1%) than other hepatic steatosis indicators (Table 5).

Diagnostic performance of a non-invasive predictive score for hepatic steatosis in the test population. prediction models SN (%) SP (%) LR+ LR- PPV (%) NPV (%) Accuracy (%) MSI-S 77.8 (69.8-84.5) 83.4 (77.5-88.3) 4.6 (3.4-6.5) 0.27 (0.19-0.37) 76.1 (69.7-81.5) 84.7 (80.1-88.4) 81.1 (76.5-85.2) FLI 70.4 (61.9-77.9) 78.9 (72.6-84.4) 3.3 (2.5-4.5) 0.38 (0.29-0.49) 69.3 (62.9-75.1) 79.7 (75.0-83.7) 75.5 (70.5-80.0) NFLS 86.7 (79.8-91.9) 51.3 (44.1-58.4) 1.8 (1.5-2.1) 0.26 (0.17-0.41) 54.7 (50.8-58.5) 85.0 (78.3-89.9) 65.6 (60.2-70.7) HSI 68.9 (60.4-76.6) 77.9 (71.5-83.5) 3.1 (2.3-4.1) 0.40 (0.31-0.52) 67.9 (61.4-73.7) 78.7 (74.0-82.7) 74.3 (69.2-78.9) Data are presented as percentage (95% CI).
MSI-S, metabolic stress index for liver steatosis; FLI, fatty liver index; NLFS, non-alcoholic fatty liver disease (NAFLD) liver fat score; HSI, hepatic steatosis index; SN, sensitivity; SP, specificity; LR+, positive likelihood ratio; LR-, negative likelihood ratio; PPV, positive predictive value; NPV, negative predictive value.

Comparing analysis according to hepatic steatosis grade, it was confirmed that the Kruskal-Wallis chi-square value (KW χ ² = 146.5, p = 2e-32) of MSI-S was higher than that of other indicators. MSI-S values for severe hepatic steatosis were significantly higher than those for mild grades (median [95 % CI]; 68.6 [59.6-69.4] for grade 1, 86.9 [71.6-90.7] for grade 2, p = 0.009) (Fig. 2B).

<Experimental Example 3> Prediction of liver fat content

Equations for predicting liver fat content were derived from multivariate stepwise linear regression analysis using the same variables used for MSI-S (Table 6 and Equation 2).

Multivariate Linear Regression Model for Predicting Liver Fat Content variable B (95% CI) SE β t P-value VIF WHR 2.631 (1.494-3.769) 0.578 0.201 4.551 <0.001 1.237 Ln [triglyceride (mg/dL)] 0.288 (0.152-0.424) 0.069 0.196 4.157 <0.001 1.411 Ln [total cholesterol (mg/dL)] 0.344 (0.021-0.667) 0.164 0.089 2.094 0.037 1.134 Ln [fasting insulin (mU/L)] 0.102 (0.003-0.202) 0.051 0.106 2.020 0.044 1.761 Ln [ALT (IU/L)] 0.27 (0.113-0.427) 0.080 0.147 3.383 <0.001 1.194 Ln [albumin (g/dL)] 2.039 (0.728-3.35) 0.666 0.128 3.059 0.002 1.108 Ln [FGF21 (pg/mL)] 0.18 (0.098-0.263) 0.042 0.187 4.289 <0.001 1.206 Ln [FGF19 (pg/mL)] -0.131 (-0.212--0.051) 0.041 -0.135 -3.206 0.001 1.124 Ln [A/L (10 ³ )] -0.125 (-0.189--0.061) 0.033 -0.191 -3.849 <0.001 1.563 Constant -8.426 (-11.253--5.6) 1.437 -5.865 <0.001 model summary F(9, 324) = 34.494; p = 2.2E-42; Durbin-Watson score: 1.913
R ² =0.489; Adj R ² = 0.475, SEE = 0.565 B, unstandardized regression coefficient; SE, standard error; SEE, standard error of estimate; β, standardized regression coefficient; VIF, variance inflation factor; ln, natural logarithm; WHR, waist-to-hip ratio; ALT, alanine-aminotransferase; A/L, adiponectin-to-leptin ratio; FGF21, fibroblast growth factor 21; FGF19, fibroblast growth factor 19.
Dependent variable: liver fat content (ln %, MRI-PDFF)

It was confirmed that the derived liver fat content had a good correlation with the PDFF measured using MRI and was consistent. The adjusted R ² of the model was 0.475 and the ICC was 0.79 (95% CI 0.74-0.83) (Table 6 and Figure 3). The Bland-Altman plot confirmed that 94.9% of the test subjects were within the limit of the mean difference (MD) (FIG. 3).

To improve the clinical applicability of predictive indicators, additional analyzes were performed under the assumption that liver biopsies were not required in patients with TP (true positive) or TN (true negative). It was confirmed that the cutoff values of the MSI-S with sensitivity and specificity of 90% or more were 23.9 and 60.8, respectively. Thus, it can be seen that 75.5% of cases would have avoided a liver biopsy or further evaluation. In conclusion, it can be seen that MSI-S is more effective in reducing unnecessary invasive tests than other currently available indicators (Fig. 4, Table 7).

Sensitivity and specificity of predictive indicators to identify hepatic steatosis using a threshold of 90% sensitivity and specificity iTrsodSN (%) (%) + - V (%) V (%) curacy (%) and TN (%) 390 (84-95) 67 (60-73) 2.7 (2.2-3.3) 0.14 (0.09-0.24) 65 (60-79) 91 (86-95) 76 (71-81) 65.9 64 (56-73) 91 (86-94) 6.8 (4.3-10.5) 0.39 (0.31-0.50) 82 (75-88) 79 (75-83) 80 (75-84) 1.90 (84-95) 47 (40-54) 1.7 (1.5-2.0) 0.20 (0.12-0.35) 54 (50-57) 88 (81-93) 65 (59-70) 44.6 41 (32-50) 91 (86-94) 4.3 (2.7-6.9) 0.66 (0.57-0.76) 74 (64-82) 69 (66-72) 70 (65-75) 2090 (84-95) 41 (34-48) 1.5 (1.4-1.8) 0.23 (0.14-0.40) 51 (48-54) 86 (79-92) 61 (56-66) 39.2 36 (28-45) 91 (86-94) 3.8 (2.4-6.2) 0.70 (0.62-0.81) 72 (61-81) 68 (65-71) 69 (63-74) 3.90 (84-95) 41 (34-48) 1.5 (1.4-1.8) 0.23 (0.14-0.40) 51 (48-54) 86 (79-92) 61 (56-66) 37.7 33 (25-41) 91 (86-94) 3.4 (2.1-5.6) 0.75 (0.66-0.85) 70 (59-79) 66 (64-69) 67 (62-72) Data are presented as percentage (95% CI) MSI-S, metabolic stress index for liver steatosis; FLI, fatty liver index; NLFS, non-alcoholic fatty liver disease (NAFLD) liver fat score; HSI, hepatic steatosis index; SN, sensitivity; SP, specificity; LR+, positive likelihood ratio; LR-, negative likelihood ratio; PPV, positive predictive value; NPV, negative predictive value; TP, true positive; TN, true negative; FP, false positive; FN, false negative.

Claims (11)

1) calculating a waist-to-hip ratio (WHR) from waist and hip circumference measurements measured from the subject;
2) fibroblast growth factor 21 (FGF21), fibroblast growth factor 19 (FGF19), adiponectin, leptin, insulin ( insulin), albumin, triglyceride (TG), total choresterol (TC) and alanine-aminotransferase (ALT) concentrations;
3) calculating an adiponectin-to-leptin ratio (A/L) from the adiponectin and leptin concentrations measured above;
4) WHR and A/L calculated above, and FGF21, FGF19, insulin, and albumin concentration values measured above are substituted for [Equation 1] below, and the metabolic stress index for hepetic steatosis; calculating MSI-S); and
[Equation 1]

5) An information providing method for predicting or diagnosing hepatic steatosis, comprising comparing the MSI-S value and the cut-off value.
The method according to claim 1, wherein, in step 1), WHR is a value obtained by dividing a measured waist circumference by a hip circumference measurement.
The method according to claim 1, wherein the biological sample in step 2) is whole blood, plasma or serum.
The method of claim 1, wherein the concentrations of FGF21, FGF19, adiponectin, leptin, insulin, albumin, TG, TC and ALT in step 2) are measured by an immunoassay, a colorimetric assay, or a turbidity assay.
The method according to claim 1, wherein in step 2), the FGF21 and FGF19 concentration units are pg/mL, the adiponectin concentration unit is μg/mL, the leptin concentration unit is ng/mL, the insulin concentration unit is mU/L, and albumin The method of claim 1, wherein the concentration unit is g/dL, the TG and TC concentration units are mg/dL, and the ALT concentration unit is IU/L.
The method of claim 1, wherein A/L in step 3) is a value obtained by dividing an adiponectin concentration value by a leptin concentration value.
The method according to claim 1, wherein if the MSI-F value is equal to or greater than the cut-off value, the patient is determined to be a hepatic steatosis patient.
Method according to claim 7, characterized in that the cut-off value is 23.9.
The method according to claim 8, characterized in that the cut-off value is 49.43.
According to claim 1, wherein the WHR and A/L calculated above, the FGF21, FGF19, insulin, albumin, TG, TC and ALT concentration values measured above are substituted into the following [Equation 2] to predict the liver fat content A method, characterized in that it further comprises the step of:
[Equation 2]

.
1) Among biological samples isolated from individuals, growth factor 21 (fibroblast growth factor 21; FGF21), fibroblast growth factor 19 (FGF19), adiponectin, leptin, insulin (insulin), a set of reagents for determining the concentrations of albumin, triglyceride (TG), total choresterol (TC) and alanine-aminotransferase (ALT); and
2) A kit for predicting or diagnosing hepatic steatosis including a device for measuring the waist and hip circumference of an individual.

Images