阅读说明：本技术 诊断肝脂肪变性的方法 (Method for diagnosing hepatic steatosis ) 是由 T·波纳德 于 2019-07-26 设计创作，主要内容包括：本发明涉及用于评估患者中非酒精性脂肪性肝炎(脂肪变性)的存在的新方法,其使用将生物标记物组合的函数,而且未将胆红素和身体质量指数用作函数中的标记物。(The present invention relates to a novel method for assessing the presence of non-alcoholic steatohepatitis (steatosis) in a patient using a function combining biomarkers and without using bilirubin and a body mass index as markers in the function.)

1. An in vitro method for diagnosing the presence of hepatic steatosis in a subject, comprising the steps of:

(a) combining values of at least three biomarkers in a function, the values selected from the group consisting of:

the amount of biochemical marker circulating in the blood, serum or plasma of the subject, and optionally;

the physical characteristics of the patient;

in order to obtain the final value of the value,

(b) optionally comparing the final value with a predetermined value, an

(c) Determining the presence of steatosis based on the final value calculated in (a),

with the proviso that the subject's body mass index and bilirubin level are not used as biomarkers in (a).

2. The in vitro method of claim 1, wherein the circulating biochemical marker is selected from the group consisting of alpha 2-macroglobulin (A2M), GGT (γ -glutamyl transpeptidase), haptoglobin, apolipoprotein a-I (apoA1), alanine Aminotransferase (ALT), aspartate Aminotransferase (AST), triglycerides, total cholesterol, fasting glucose, γ -globulin, albumin, α 1-globulin, α 2-globulin, β -globulin, IL10, TGF- β 1, apoA2, apoB, cytokeratin 18, platelet count, prothrombin level, hyaluronic acid, urea, N-terminal type III procollagen, type 1 tissue inhibitor metalloproteinase (TIMP-1), type IV collagen (Coll IV), osteoprotegerin, miRNA122, cytokeratin 18, serum amyloid a (saa) ((saa)), Alpha-1-antitrypsin (isoform 1), fructose bisphosphate aldolase A, fructose bisphosphate aldolase B, fumarylacetoacetate, transthyretin, PR02275, C-reactive protein (isoform 1), leucine-rich alpha-2-glycoprotein, serine protease inhibitor A11, DNA-directed RNA polymerase I subunit RPA1, obscurin (isoform 1), alpha skeletal muscle actin, aortic smooth muscle actin, alkaline phosphatase, atypical protein C22orf30 (isoform 4), serum amyloid A2 (isoform a), apolipoprotein C-III, apolipoprotein E, apolipoprotein A-II, polymeric immunoglobulin receptor, von Willebrand factor, aminoacylase-1, G protein-coupled receptor 98 (isoform 1), paraoxonase/aryl esterase 1, Complement component C7, hemopexin, complement C1q subcomponent, paraoxonase/lactamase 3, complement C2 (fragment), pluripotency proteoglycan core protein (isoform Vint), extracellular matrix protein 1 (isoform 1), E3 SUMO-protein ligase RanBP2, haptoglobin-related protein (isoform 1), adiponectin, retinol-binding protein, ceruloplasmin, α 2 antiplasmin, antithrombin, thyroxine-binding protein, protein C, α 2 lipoprotein, tetranectin, fucosylated A2M, fucosylated haptoglobin, fucosylated apoA1 and carbohydrate-deficient transferrin.

3. The in vitro method according to any one of claims 1 to 2, wherein said function comprises at least one variable selected from the group consisting of sex and age of the subject.

4. The in vitro method according to any of claims 1 to 3, wherein said function comprises values measured on circulating alpha-2-macroglobulin, ApoA1, GGT, haptoglobin, ALT, AST, fasting glucose, total cholesterol, triglycerides and values corresponding to age and gender of the patient.

5. The in vitro method according to any one of claims 1 to 4, wherein said function is a logistic function obtained by logistic regression.

6. The in vitro method of claim 5, wherein said function is obtained by:

a) assessing the presence of steatosis in a group of patients, wherein the values of circulating biochemical markers of the patients are known;

b) by one-dimensional analysis, circulating biochemical markers with significantly different values between the following groups were identified:

i. patients suffering from steatosis, and

patients without steatosis

c) Performing logistic regression analysis to assess and consider the independent discrimination values of the markers identified in step b) for the occurrence of steatosis;

d) thus by combining these identified independent factors, a function is obtained.

7. The in vitro method according to any one of claims 1 to 6, wherein the function is a0+ a1 × age (years) + A2 × ApoA1(g/l) + a3 × Log (A2M, g/l) + a4 × Log (GGT, IU/l) + a5 × Log (ALT, IU/l) + a6 × Log (AST, IU/l) + a7 × Log (Hapto, g/l) + a8 × Log (triglyceride TG, mmol/l) + a9 × Log (total cholesterol CT, mmol/l) + a10 × Log (fasting glucose, g/l) + a11 × gender (female 0, male 1).

8. The in vitro method of claim 7, wherein:

a) 4.8. ltoreq. a 0. ltoreq.5.5, preferably 5. ltoreq. a 0. ltoreq.5.3;

b) -0.03. ltoreq. a 1. ltoreq.0.015, preferably-0.025. ltoreq. a 1. ltoreq.0.02;

c) 0.9. ltoreq. a 2. ltoreq.1.2, preferably 0.95. ltoreq. a 2. ltoreq.1.05;

d) 1.8. ltoreq. a 3. ltoreq.2.2, preferably 1.95. ltoreq. a 3. ltoreq.2.05;

e) -1.3. ltoreq. a 4. ltoreq.1.1, preferably-1.2. ltoreq. a 4. ltoreq.1.1;

f) -1.45. ltoreq. a 5. ltoreq.1.2, preferably-1.36. ltoreq. a 5. ltoreq.1.3;

g) 0.5. ltoreq. a 6. ltoreq.0.7, preferably 0.55. ltoreq. a 6. ltoreq.0.65;

h) -0.35. ltoreq. a 7. ltoreq.0.22, preferably-0.3. ltoreq. a 7. ltoreq.0.25;

i) -1.45. ltoreq. a 8. ltoreq.1.25, preferably-1.4. ltoreq. a 8. ltoreq.1.3;

j) -0.8. ltoreq. a 9. ltoreq.0.6, preferably-0.75. ltoreq. a 9. ltoreq.0.65;

k) -3.55. ltoreq. a 10. ltoreq.3.35, preferably-3.5. ltoreq. a 10. ltoreq.3.4;

l) 0.35. ltoreq. a 11. ltoreq.0.55, preferably 0.4. ltoreq. a 11. ltoreq.0.5.

9. The in vitro method according to any one of claims 1 to 8, wherein the function is 5.17-0.022 × age (years) +1.02 × ApoA1(g/l) +1.99 × Log (A2M, g/l) -1.16 × Log (GGT, IU/l) -1.33 × Log (ALT, IU/l) +0.6 × Log (AST, IU/l) -0.29 × Log (Hapto, g/l) -1.35 × Log (triglyceride TG, mmol/l) -0.68 × Log (total cholesterol CT, mmol/l) -3.46 × Log (fasting glucose, g/l) +0.46 × sex (female 0, male 1).

10. The in vitro method according to claim 9, wherein the predetermined value is 0.25 and when the final result is higher than or equal to 0.25, the patient suffers from non-alcoholic steatohepatitis (steatosis), and when the final result is lower than 0.25, the patient does not suffer from non-alcoholic steatohepatitis (steatosis).

11. A method of obtaining a function for identifying the presence of steatosis in a patient, wherein said function combines the concentration values of biochemical markers in the blood/serum or plasma of said patient and optionally the age, sex of the patient, said method comprising the steps of:

a) assessing the presence of steatosis in a group of patients, wherein the values of circulating biochemical markers of the patients are known;

b) by one-dimensional analysis, circulating biochemical markers with significantly different values between the following groups were identified:

i. patients suffering from steatosis, and

patients without steatosis

c) Identifying whether the age, gender, of the patients differ significantly between the following groups:

i. patients suffering from steatosis, and

patients without steatosis

d) Performing a logistic regression analysis to assess and consider the independent discriminatory values of the markers identified in steps b) and c) for the occurrence of steatosis, wherein the body mass index and bilirubin levels are not included in the list of markers for which the logistic regression is performed;

e) thereby obtaining a function by combining these identified independent factors, wherein said function can be used to diagnose the presence of non-alcoholic steatohepatitis (steatosis) in a subject and without using values for bilirubin or BMI.

12. The method of claim 11, wherein the biochemical marker of step b) is selected from the group consisting of α 2-macroglobulin (A2M), GGT (γ -glutamyl transpeptidase), haptoglobin, apolipoprotein a-I (apoA1), alanine Aminotransferase (ALT), aspartate Aminotransferase (AST), triglycerides, total cholesterol, fasting glucose, γ -globulin, albumin, α 1-globulin, α 2-globulin, β -globulin, IL10, TGF- β 1, apoA2, apoB, cytokeratin 18, platelet count, prothrombin level, hyaluronic acid, urea, N-terminal type III procollagen, type 1 tissue inhibitor metalloproteinase (TIMP-1), type IV collagen (Coll IV), osteoprotegerin, miRNA122, cytokeratin 18, serum amyloid a (saa) ((saa), Alpha-1-antitrypsin (isoform 1), fructose bisphosphate aldolase A, fructose bisphosphate aldolase B, fumarylacetoacetate, transthyretin, PR02275, C-reactive protein (isoform 1), leucine-rich alpha-2-glycoprotein, serine protease inhibitor A11, DNA-directed RNA polymerase I subunit RPA1, obscurin (isoform 1), alpha skeletal muscle actin, aortic smooth muscle actin, alkaline phosphatase, atypical protein C22orf30 (isoform 4), serum amyloid A2 (isoform a), apolipoprotein C-III, apolipoprotein E, apolipoprotein A-II, polymeric immunoglobulin receptor, von Willebrand factor, aminoacylase-1, G protein-coupled receptor 98 (isoform 1), paraoxonase/aryl esterase 1, Complement component C7, hemopexin, complement C1q subcomponent, paraoxonase/lactamase 3, complement C2 (fragment), pluripotent proteoglycan core protein (isoform Vint), extracellular matrix protein 1 (isoform 1), E3 SUMO-protein ligase RanBP2, haptoglobin-related protein (isoform 1), adiponectin, retinol-binding protein, ceruloplasmin, α 2 antiplasmin, antithrombin, thyroxine-binding protein, protein C, α 2 lipoprotein, tetranectin, fucosylated A2M, fucosylated haptoglobin, fucosylated apoA1, carbohydrate-deficient transferrin, α -fetoprotein (AFP), sccglycosylated AFP, HSP27 (heat shock protein), HSP70, glypican-3 (GPC3), squamous cell carcinoma antigen (a) and in particular as a circulating immune IC complex consisting of a and IgM, Golgi protein 73(GP73), alpha-L-fucosidase (AFU), Des-gamma-carboxyprothrombin (DCP or PIVKA), Osteopontin (OPN), and human carbonyl reductase.

13. An ex vivo method for diagnosing a liver disease in a patient comprising the steps of:

-performing the method of any one of claims 1 to 10 to determine the presence of steatosis in a patient,

-determining the presence of liver fibrosis in a patient,

-determining the presence of liver inflammation in a patient, and

diagnosing a liver disease in the patient based on the result of the above determination.

14. The method of claim 13, wherein liver fibrosis is determined by: liver stiffness is measured ex vivo by combining biomarker values in functions, particularly by fibritest, or by vibration-controlled transient elastography.

15. The method of claim 13 or 14, wherein the presence of liver inflammation in the patient is determined by: the values of the biomarkers are combined in the function by, inter alia, the NASH test.

Technical Field

The present invention relates to a novel non-invasive quantitative test that allows for the detection of hepatic steatosis and facilitates the identification of subjects at risk of non-alcoholic fatty liver disease (NAFLD), in particular subjects suffering from non-alcoholic steatohepatitis (NASH) or fibrosis.

Background

Fatty liver is a reversible disease in which a large number of triglyceride fat vacuoles are accumulated in hepatocytes through the process of steatosis.

Nonalcoholic fatty liver disease (NAFLD) is one of the causes of fatty liver, which occurs when fat is deposited in the liver for reasons other than excessive drinking (steatosis). NAFLD is the most common liver disease in developed countries. NAFLD is a disease commonly associated with a factor in metabolic syndrome.

Although the liver can retain fat without destroying liver function, it may also develop nonalcoholic steatohepatitis (NASH), a state in which steatosis is combined with inflammation and fibrosis (steatohepatitis). NASH is a progressive disease: over a10 year period, up to 20% of NASH patients will develop cirrhosis and 10% will die from liver disease.

Thus, nonalcoholic steatohepatitis (NASH) is the most severe form of NAFLD and is considered to be the leading cause of unexplained cirrhosis.

Biopsy is a common reference for assessment of hepatic steatosis. Whenever there is no imaging tool (ultrasonography, MRE) available or viable (e.g. large epidemiological studies), serum biomarkers and scores are acceptable alternatives to diagnose steatosis.

In a large study of subjects at risk for nonalcoholic fatty liver disease (NALFLD) and NASH, the european guidelines have suggested three blood tests to be performed to diagnose steatosis (EASL J Hepatol 2015). The best validated steatosis scores were the fatty liver index, Steatotest (Poynard et al, 2005, Comparative Hepatology 4: 10) and NAFLD liver fat score (Angulo et al, Hepatology 2007; 45 (4): 846-854); they have been externally validated in the general population or in the obese class 3 population and differentially predicted metabolic, hepatic and cardiovascular outcomes/mortality. These scores correlate with insulin resistance and reliably predict the presence, rather than the severity, of steatosis (Fedchuk et al, Aliment Pharmacol Ther 2014; 40: 1209-1222).

SteatoTest was calculated using a raw combination of nine biochemical markers (which are easily assessed) and combining alpha 2-macroglobulin, haptoglobin, apolipoprotein A1, total bilirubin, Gamma Glutamyl Transpeptidase (GGT), fasting glucose, triglycerides, cholesterol and alanine Aminotransferase (ALT) with parameters adjusted according to the age, sex, weight and height (body mass index) of the patient.

Of the 9 serum components of SteatoTest, two are two sources of variability that may affect the reliability of the results: body Mass Index (BMI) and total bilirubin.

NAFLD liver fat score age, BMI, IGF/diabetes (yes/no), platelets, AST, ALT and albumin were used (NAFLD score-1.675 +0.037 × age (years) +0.094 × BMI (kg/m)²) +1.13 × IFG/diabetes (1, 0+ 0.99 × AST/ALT ratio-0.013 × platelets (× 10)⁹/L) -0.66 × albumin (g/dL)).

Both tests used BMI as a variable in the formula. However, Butler et al (2017, Clinical Nutrition ESPEN 22, 112e115) and Sorkin et al (1999, American Journal of epidemic, 150(9), 969-. In particular, the height of a person may decrease over the years, which is not typically taken into account when querying a patient for such information. In addition, people may not know (or provide) their exact weight. In summary, this means the calculated body mass index (BMI (kg/m)²) Weight (kg)/height (m)²) And is often inaccurate.

Furthermore, Poynard et al (BMC gastroenterol.2011; 11: 39) show that the use of bilirubin may cause false positives, particularly in patients with Gilbert's syndrome or hemolysis.

US20170082603 essentially relates to a method of diagnosing the presence and/or severity of liver fibrosis and/or monitoring the efficacy of a treatment in an individual suffering from liver disease comprising combining the values of various markers to obtain a score by a mathematical function. The methods disclosed in US20170082603 can be used for patients who have been diagnosed with NASH because they enable detection of progression towards fibrosis, but are not intended to determine the presence of NASH. This document does not disclose nor suggest the possibility of detecting hepatic steatosis according to, inter alia, the methods disclosed herein.

WO2014049131 relates to an in vitro non-invasive method for assessing the presence and/or severity of NASH lesions in the liver of a subject, wherein the method comprises the steps of: a. measuring in a sample of said subject at least one biomarker, preferably 1, 2, 3 or 4 biomarkers, selected from at least one of: i. at least one biomarker that reflects apoptosis, and/or ii.at least one biomarker that reflects anthropometry, and/or iii.at least one biomarker that reflects metabolic activity, and/or iv.at least one biomarker that reflects liver condition; combining the at least one biomarker measurement in a mathematical function. The formulas disclosed on pages 23-24 all include BMI. There is no disclosure in this document of an option to specifically deny the use of BMI and bilirubin.

WO2018050804 discloses a new method for assessing NAFLD and NAH in patients, which combines the measurement of serum markers by a mathematical function. The functions disclosed therein all include bilirubin.

Even though Poynard et al (BMC gastroenterol.2011; 11: 39) indicate that the use of bilirubin may induce false positives, there is no suggestion that the test of WO2018050804 can be modified to obtain the methods disclosed herein that are capable of diagnosing liver steatosis.

Munteanu et al (animal Pharmacol Ther 2016; 44: 877-.

Disclosure of Invention

Therefore, there is a need to develop new tests to detect hepatic steatosis, particularly when associated with inflammation (NASH) or fibrosis, to simplify the feasibility of such multi-analyte biomarkers. The test should preferably be of the same quality as the prior art test (steatost).

Typically, the quality of the test is determined by plotting a Receive Operating Characteristic (ROC) curve and measuring the area under the receive operating characteristic (AUROC).

ROC curves were drawn for different thresholds (from 0 to 1) by plotting sensitivity versus (1-specificity) after classifying the patients according to the results obtained from the test.

It is generally accepted that ROC curves with an area under them having a value greater than 0.7 are good prediction curves. It must be recognized that ROC curves are curves that allow prediction of the quality of the test. It is desirable to have AUROC as close to 1 as possible, a value that describes a 100% specific and sensitive test.

To remind that

(1) Sensitivity refers to the probability of diagnosing as positive (true positive detected) in an individual with the phenotype sought: if the patient has this phenotype, the test is positive. When the number of false negatives is high, the sensitivity is low. The sensitivity was calculated as SE ═ number of phenotypical individuals with marker (sign) present)/(number of phenotypical individuals with marker present + number of phenotypical individuals lacking marker present).

(2) Specificity refers to the probability of diagnosing negative (true negative not detected) in an individual without the phenotype sought: if the patient is not diseased, the test is negative. When the number of false positives is high, the specificity is low. The specific calculation formula is SP ═ number of individuals without phenotype lacking the marker)/(number of individuals without phenotype lacking the marker + number of individuals without phenotype with the marker present.

(3) Positive Predictive Value (PPV): refers to the probability of being diseased if the diagnostic test is positive (i.e., the patient is not a false positive): if the test is positive, the patient has a phenotype. The positive predictive value is calculated as PPV ═ number (number of individuals with phenotype with presence marker)/(number of individuals with phenotype with presence marker + number of individuals without phenotype with presence marker).

(4) Negative Predictive Value (NPV): refers to the probability of not being diseased if the diagnostic test is negative (i.e., the patient is not false negative): if the test is negative, the patient does not have a phenotype. The negative predictive value is calculated as NPV ═ number (number of unidentified individuals without phenotype)/(number of unidentified individuals without phenotype + number of unidentified individuals with phenotype).

In order to obtain a good diagnostic test, it is important to improve both specificity and sensitivity.

Generally, the diagnostic method comprises:

i. a step of collecting information from the patient;

a step of comparing said information with a threshold;

deriving from the difference between the patient information and the threshold whether the patient has a particular disease, the stage of the patient's disease or whether the patient's state will evolve to a given state.

As an illustration of

i. The information that can be collected from the patient can be collected directly from the patient (e.g., images from NMR, scanners, radiography, contrast enhanced computed tomography), or indirectly from the patient, e.g., from a biological sample (e.g., urine, blood sample) that has been taken from the patient. The information may be the presence (or absence) and/or level of a particular biomarker, whether it is specific for a pathogenic determinant (bacterial or viral DNA/RNA), or elevated patient marker levels;

after the information is obtained, it is compared to different values/standards and the deviation from these standards is evaluated. By way of illustration, the levels of certain biomarkers should be compared to levels typically observed in healthy patients and in patients with the disease (or for prognostic methods, patients known to later evolve to a particular disease stage). There may be a threshold where 95% of patients above the threshold have disease and 95% of patients not above the threshold do not have disease. For diseases where multiple clinical stages can be determined, such a threshold can distinguish between the different stages. In this step ii, the various types of information may be compared to their respective standards to enable diagnosis in step iii (illustratively, values and information obtained from measurements of various blood or plasma markers, images of scanners, and body mass indices may be used).

The final step is in fact a diagnosis (or, alternatively, a determination of the prognosis) by means of a threshold as described above, in particular taking into account the information collected from the patient, i.e. determining whether the patient is suffering from the sought disease (or whether the patient will evolve into a given clinical state). The physician may also consider other factors to make the diagnosis, such as consistency of the collected information, etc.

Certain methods, such as those disclosed in this application, should also include step i.a), which includes the step of modifying the information obtained from the patient in order to obtain a new type of information, which is then compared to the standard in step ii. Such a modification is to combine the values of the variables in the function and obtain the final value.

It should also be noted that, as disclosed herein, it is part of the method to measure and combine in the algorithm only the values of the marker levels in the patient's plasma or serum, but only to provide intermediate results (final values or indices) which are then compared to reference indices (thresholds) to actually enable diagnosis.

It should also be noted that the test disclosed herein is not a "gold standard" test in the sense that the output value (the index calculated by the formula disclosed herein) is not a definite answer to the patient's state. In fact, these tests are statistically based and therefore may have false positive or false negative results, which is why the physician's specific experience in interpreting the index is crucial in making a prognosis for each patient and deciding which follow-up to take.

However, because of the specificity, sensitivity, positive predictive value, and negative predictive value of the tests provided herein for various thresholds of the index, these tests are of great interest in providing assistance to physicians when investigating clinical cases. Thus, step iii as disclosed above is not directly and immediately from step ii, since the physician has to interpret the results according to clinical and general conditions to reach a conclusion.

Accordingly, the present invention relates to an in vitro method for diagnosing the presence of steatosis in a subject, comprising the steps of:

(a) combining values of at least three biomarkers in a mathematical function, the values selected from the group consisting of:

measured amount of biochemical marker circulating in the blood, serum or plasma of the subject, and

physical characteristics of the patient

In order to obtain the final value.

The function may further comprise the steps of:

(b) optionally comparing the final value with a predetermined value, an

(c) Determining the presence of steatosis based on the final value calculated in (a),

provided that the subject's body mass index and bilirubin level are not used as biomarkers in (a).

The method is performed in vitro or ex vivo. It is preferred when combining the values of at least three, preferably at least four, preferably at least five, preferably at least or exactly six, preferably at least or exactly seven, preferably at least or exactly eight, preferably at least or exactly nine biomarkers in a function.

Some functions that may be used to perform the methods disclosed herein are described below. However, it should be noted that following the teachings of the present invention and avoiding the use of bilirubin and BMI in the markers and variables used, there is no technical difficulty in obtaining and developing other functions that are as (or more) effective as the functions disclosed herein.

It should also be noted that although the exemplary functions herein have been obtained by logistic regression, other statistical methods may be used to provide such functions that will use the values of the various markers and provide values indicative of the presence of steatosis.

Although any label may be used in the disclosed functions, the use of bilirubin (total bilirubin) and BMI in the functions should be avoided for the reasons disclosed above.

The predetermined value is calculated and determined by one skilled in the art from the actual function. Preferably has a first predetermined value below which the negative predictive value is higher than 90%, preferably higher than 95%, more preferably higher than 98%. Thus, any final value below this first predetermined value will mean that the patient is not steatosis. The function is preferably normalized so that the first predetermined value is 0.25.

It is also possible to define a second predetermined value above which the positive predictive value is higher than 75%, preferably higher than 85%, more preferably higher than 90%. Thus, any final value above this second predetermined value would mean that the patient suffers from steatosis and that NASH and fibrosis (two serious complications of steatosis) should be diagnosed. The function is preferably normalized so that the second predetermined value is 0.50.

The method uses a semi-quantitative function which makes it possible to provide information about the steatosis in the liver of the patient. This means that the higher the final value, the more severe the steatosis in the liver of the patient.

The method should be used by the doctor in the following way:

when the test is positive (i.e. the final value is above a predetermined threshold), the physician will first look for any other cause of liver disease that is likely to cause steatosis and that is easy to detect, i.e. whether the patient has the most common alcoholic steatosis, viral hepatitis with a specific marker (hepatitis b, c), drug induced hepatitis or genetic steatosis or infectious steatosis. If this potential cause of steatosis is excluded, the physician will consider other metabolic causes (e.g., overweight, type 2 diabetes, dyslipidemia, hypertension, hyperuricemia …). The presence of such other causes makes it possible to finalize the diagnosis, since these causes are related to increased fat in the liver.

It should also be noted that after certain abnormalities are detected (e.g. common liver marker abnormalities or fat in the liver detected by echography), the patient will typically consult a hepatologist to make the diagnosis easier.

Thus, the methods disclosed herein will typically be performed together with other existing tests (fibertest disclosed in WO0216949 and NASH test disclosed in WO 2018050804). Obtaining the results of all these tests will enable the physician to make an accurate diagnosis of the patient's liver disease.

In summary, the general condition and clinical assessment of the patient (which is always a consideration for diagnostic methods) are also taken into account, the final value allowing the presence of steatosis to be determined. Thus, the present method allows conclusions to be drawn as to the presence or absence of steatosis and/or the presence of diagnostically relevant inflammation (NASH) or fibrosis in a patient.

This method is particularly useful for detecting the absence of steatosis (when the final value is lower than the first predetermined value), as it makes it possible to avoid any unnecessary examinations of the patient and to guide the physician to investigate other causes when necessary.

In the present method, one or more (in any combination) of the following are preferred:

(a) bilirubin is not included in the biochemical marker whose concentration is measured in step (a);

(b) the biochemical marker of step (a) is selected from the group consisting of alpha 2-macroglobulin (A2M), GGT (gamma-glutamyltranspeptidase), haptoglobin, apolipoprotein A-I (apoA1), alanine Aminotransferase (ALT), aspartate Aminotransferase (AST), triglyceride, total cholesterol, fasting glucose, gamma-globulin, albumin, alpha 1-globulin, alpha 2-globulin, beta-globulin, IL10, TGF-beta 1, apoA2, apoB, cytokeratin 18, platelet count, prothrombin level, hyaluronic acid, urea, N-terminal type III procollagen, type 1 tissue inhibitor metalloproteinase (TIMP-1), type IV collagen (Coll IV), osteoprotegerin, miRNA122, cytokeratin 18, Serum Amyloid A (SAA), alpha-1-antitrypsin (isoform 1), Fructose bisphosphate aldolase A, fructose bisphosphate aldolase B, fumarylacetoacetate, transthyretin, PR02275, C-reactive protein (isoform 1), leucine-rich alpha-2-glycoprotein, serine protease inhibitor A11, DNA-directed RNA polymerase I subunit RPA1, masking protein (isoform 1), alpha skeletal muscle actin, aortic smooth muscle actin, alkaline phosphatase, atypical (noncharacterized) protein C22orf30 (isoform 4), serum amyloid A2 (isoform a), apolipoprotein C-III, apolipoprotein E, apolipoprotein A-II, polymeric immunoglobulin receptors, von Willebrand factor, aminoacylase-1, G protein-coupled receptor 98 (isoform 1), paraoxonase/arylesterase 1, complement component C7, heme-binding protein (hemipexin), Complement C1q subcomponent, paraoxonase/lactamase 3, complement C2 (fragment), pluripotent proteoglycan (versican) core protein (isoform Vint), extracellular matrix protein 1 (isoform 1), E3 SUMO-protein ligase RanBP2, haptoglobin-related protein (isoform 1), adiponectin, retinol-binding protein, ceruloplasmin (ceruloplastin), α 2 antiplasmin, antithrombin, thyroxine-binding protein, protein C, α 2 lipoprotein, tetranectin, fucosylated A2M, fucosylated haptoglobin, fucosylated apoA1 and carbohydrate-deficient transferrin;

(c) the biochemical marker of step (a) is selected from the group consisting of alpha 2-macroglobulin, AST (aspartate aminotransferase), ALT (alanine aminotransferase), GGT (gamma-glutamyl transpeptidase), total bilirubin, haptoglobin, apoA1, triglyceride, total cholesterol, fasting glucose, gamma globulin, albumin, alpha 1-globulin, alpha 2-globulin, beta-globulin, IL10, TGF-beta 1, apoA2, apoB, cytokeratin 18 and cytokeratin 19 components, platelet number, prothrombin levels, hyaluronic acid, urea, N-terminal type III procollagen, type 1 tissue inhibitor metalloproteinase (TIMP-1), type IV collagen (Coll IV), and osteoprotegerin;

(d) the biochemical markers of step (a) are alpha-2-macroglobulin, ApoA1, GGT, haptoglobin, ALT, AST, fasting glucose, total cholesterol and triglycerides;

(e) using at least 6, more preferably at least 7, more preferably at least 8, more preferably at least or exactly 9 biochemical markers in the mathematical function;

(f) the function of a) further comprises at least one, preferably two physical characteristics of the patient, said physical characteristics being selected from the group consisting of sex and age of the patient;

(g) BMI is not used as a physical characteristic of the patient;

(h) the function of (a) comprises values measured for circulating alpha-2-macroglobulin, ApoA1, GGT, haptoglobin, ALT, AST, fasting glucose, total cholesterol, triglycerides and values corresponding to the age and sex of the patient;

(i) a) combining the measurements of alpha-2-macroglobulin, ApoA1, GGT, haptoglobin, ALT, AST, fasting glucose, total cholesterol and triglycerides and the age and sex of the patient;

(j) the function is a logistic function, i.e. a function obtained by logistic regression, as described below;

(k) the function is a semi-quantitative function.

In a particular embodiment, the function has been obtained by:

a) assessing the presence of steatosis in a group of patients, wherein the values of circulating biochemical markers of the patients are known;

b) by one-dimensional analysis, circulating biochemical markers with significantly different values between the following groups were identified:

i. patients suffering from steatosis, and

a patient without steatosis;

c) performing logistic regression analysis to assess and consider the independent discrimination values of the markers identified in step b) for the occurrence of steatosis;

d) thus a function is obtained by combining these identified independent factors.

The following provides further explanation of this process for obtaining the diagnostic function.

In a preferred embodiment, the function is a0+ a1 × age (years) + A2 × ApoA1(g/l) + a3 × Log (A2M, g/l) + a4 × Log (GGT, IU/l) + a5 × Log (ALT, IU/l) + a6 × Log (AST, IU/l) + a7 × Log (Hapto, g/l) + a8 × Log (triglyceride TG, mmol/l) + a9 × Log (total cholesterol CT, mmol/l) + a10 × Log (fasting glucose, g/l) + a11 × gender (0 for female, 1 for male).

In particular, it is possible to use, for example,

a) 4.8. ltoreq. a 0. ltoreq.5.5, preferably 5. ltoreq. a 0. ltoreq.5.3;

b) -0.03. ltoreq. a 1. ltoreq.0.015, preferably-0.025. ltoreq. a 1. ltoreq.0.02;

c) 0.9. ltoreq. a 2. ltoreq.1.2, preferably 0.95. ltoreq. a 2. ltoreq.1.05;

d) 1.8. ltoreq. a 3. ltoreq.2.2, preferably 1.95. ltoreq. a 3. ltoreq.2.05;

e) -1.3. ltoreq. a 4. ltoreq.1.1, preferably-1.2. ltoreq. a 4. ltoreq.1.1;

f) -1.45. ltoreq. a 5. ltoreq.1.2, preferably-1.36. ltoreq. a 5. ltoreq.1.3;

g) 0.5. ltoreq. a 6. ltoreq.0.7, preferably 0.55. ltoreq. a 6. ltoreq.0.65;

h) -0.35. ltoreq. a 7. ltoreq.0.22, preferably-0.3. ltoreq. a 7. ltoreq.0.25;

i) -1.45. ltoreq. a 8. ltoreq.1.25, preferably-1.4. ltoreq. a 8. ltoreq.1.3;

j) -0.8. ltoreq. a 9. ltoreq.0.6, preferably-0.75. ltoreq. a 9. ltoreq.0.65;

k) -3.55. ltoreq. a 10. ltoreq.3.35, preferably-3.5. ltoreq. a 10. ltoreq.3.4;

l) 0.35. ltoreq. a 11. ltoreq.0.55, preferably 0.4. ltoreq. a 11. ltoreq.0.5.

In a preferred embodiment, the function is 5.17-0.022 × age (years) +1.02 × ApoA1(g/l) +1.99 × Log (A2M, g/l) -1.16 × Log (GGT, IU/l) -1.33 × Log (ALT, IU/l) +0.6 × Log (AST, IU/l) -0.29 × Log (Hapto, g/l) -1.35 × Log (triglyceride TG, mmol/l) -0.68 × Log (total cholesterol CT, mmol/l) -3.46 × Log (fasting glucose, g/l) +0.46 × gender (0 for female, 1 for male).

The coefficients of the formula as provided above have been rounded to the second digit after the decimal point, and any function using coefficients rounded to one digit after the decimal point (further) or lower (lower) decimal places may also be used as disclosed herein, and thus would be equivalent.

As an illustration, a function using coefficients rounded to the fifth decimal place may be 5.17005-0.02202 × age (years) +1.01850 × ApoA1(g/l) +1.98726 × Log (A2M, g/l) -1.16378 × Log (GGT, IU/l) -1.32517 × Log (ALT, IU/l) +0.60003 × Log (AST, IU/l) -0.28781 × Log (Hapto, g/l) -1.34979 × Log (triglyceride TG, mmol/l) -0.68426 × Log (total cholesterol CT, mmol/l) -3.45945 × Log (fasting glucose, g/l) +0.46473 × gender (0 for female, 1 for male). Such a function is also the subject of the present invention.

Thus, the methods disclosed herein utilize ex vivo combinations of values for different variables to obtain a final result indicative of the presence of steatosis in a patient. Thus, any steps applied on the patient's body will be excluded. In the methods disclosed herein, it is understood that the markers whose values are measured and used as variables in the function are circulating proteins or natural components present in the blood, plasma or serum of the patient. The value of the selected marker should be measured on a previously collected blood, plasma or serum sample according to methods known in the art. These values are expressed in units according to the art. However, if one skilled in the art selects other units to represent the measured values, this will only change the coefficients within the logistic function. The method will still be applicable.

However, in another embodiment, the present invention relates to a method for the prognosis of the presence of steatosis in a patient, comprising the steps of: collecting blood, plasma or serum from a patient, measuring the value of the marker present in the collected sample, and combining the measured values by the functions as disclosed above. The marker is as disclosed above. Thus, the method may further comprise the step of obtaining (measuring) different values used in the logistic function, e.g. the step of measuring the concentration/value/amount of the various biomarkers as described above in a blood or plasma sample that has been previously collected from the patient.

The method may further comprise the steps of: the first index is compared to a predetermined threshold, in particular to determine whether the first index is above or below the threshold.

The method may further comprise the steps of: deducing the presence of steatosis in the patient if the first index exceeds the threshold.

The tests designed here are quantitative tests, with final result values of 0 to 1. It is believed that when the end result is equal to or greater than 0.25, the patient is likely to have steatosis (as described above) and therefore needs to be treated or followed up.

In the method described above, the first predetermined value is set to 0.25, and if the final result is higher than or equal to 0.25, the patient suffers from non-alcoholic steatohepatitis (steatosis).

In short, for final values below 0.25, the patient had no steatosis. For final values between 0.25 and 0.50, patients need to be followed up regularly and new assessments are made. For final values between 0.50 and 0.75, of clinical significance, treatment should be performed (starting from diet and exercise). Above 0.75, steatosis is evident and intensive treatment should be initiated and the patient may have other liver diseases.

The present invention also includes an apparatus for diagnosing steatosis in a patient, comprising:

a) a first device, wherein the first device provides a first index by: combining values measured for markers present in serum or plasma of a patient by a logistic function, wherein the first logistic function is

a1+ A2 × Log (A2M, g/l) + a3 × age (years) + a4 × Log (ALT, IU/l) + a5 x (Apoa1, g/l) + a6 × Log (AST, IU/l) + a7 × Log (BILI, μmol/l) + a8 × Log (CT, mmol/l) + a9 × gender (0 for female, 1) + a10 × Log (GGT, IU/l) + a11 × Log (Hapto, g/l) + a12 × Log (TG, mmol/l);

wherein

--8≤a1≤-7；

0.1. ltoreq. a 2. ltoreq.0.6, preferably 0.15. ltoreq. a 2. ltoreq.0.55;

0.02. ltoreq. a 3. ltoreq.0.05, preferably 0.03. ltoreq. a 3. ltoreq.0.04;

1.1. ltoreq. a 4. ltoreq.1.5, preferably 1.2. ltoreq. a 4. ltoreq.1.4;

--0.2≤a5≤1.0；

-1.8. ltoreq. a 6. ltoreq.2.3, preferably 1.95. ltoreq. a 6. ltoreq.2.2;

-0.8. ltoreq. a 7. ltoreq.1.6, preferably 0.9. ltoreq. a 7. ltoreq.1.5;

- -1.7. ltoreq. a 8. ltoreq.1.3, preferably- -1.6. ltoreq. a 8. ltoreq.1.4;

-0.015≤a9≤0.20；

-0.15. ltoreq. a 10. ltoreq.0.25, preferably 0.20. ltoreq. a 10. ltoreq.0.22;

--0.3≤a11≤0.1；

0.9. ltoreq. a 12. ltoreq.1.2, preferably 1.0. ltoreq. a 12. ltoreq.1.1.

Subsequently, the first index may be used to determine the presence of steatosis in the patient and whether it is necessary to initiate treatment or follow-up, in particular with the aid of table 1 above.

In a particular embodiment, the first device is computerized. It may be a spreadsheet in which a formula is recorded, which provides a first index as an output when the above-mentioned various factors are input. It may also be a computer program that, upon receiving the various factors described above, provides as an output the first index.

The first device may exhibit one or more of the following (in any combination):

-operating in a private or public network;

-receiving inputs (values of the various factors mentioned above) from a sender at a remote location (i.e. sending these inputs to the first device from a location different from where the first device is located);

-requiring the sender to identify himself before sending the input;

-receiving input (values of the above-mentioned various factors) in a secure manner;

-sending an output (first index) to a sender of the input;

storing the output in a database (possibly with a unique identifier to which the input, output can be assigned);

providing further information (e.g. sensitivity and/or specificity and/or positive predictive value and/or negative predictive value relating to prevalence in a population to which the patient belongs) for the first index.

A non-transitory computer-readable storage medium is also foreseen, on which a computer program comprising program instructions is stored, which computer program is loadable into a data-processing unit and adapted to cause the data-processing unit to execute a method for calculating a first index by: the values measured for the markers present in the serum or plasma of a patient are combined by a logistic function as disclosed above.

In another embodiment, the invention relates to a microprocessor comprising a computer algorithm for performing a method of determining the presence of steatosis in a patient: providing a value of a blood marker of the patient and optionally an age and a value assigned to the sex of the patient; and performing a mathematical function (as disclosed above, or obtained as disclosed above) to combine the values to obtain a score for the presence of steatosis in the patient.

The invention also relates to the use of a function, device and/or microprocessor to diagnose the presence of steatosis in a patient by using the function and comparing the result with a predetermined threshold to determine the presence of steatosis in the patient.

The present invention also relates to an ex vivo method for diagnosing a liver disease in a patient, comprising the steps of:

performing the method disclosed herein to determine the presence of steatosis in a patient,

-determining the presence of liver fibrosis in a patient,

-determining the presence of liver inflammation in a patient, and

diagnosing a liver disease in the patient based on the result of the above determination.

It is preferred when liver fibrosis is determined by combining the values of the biomarkers in a function, in particular via Fibrotest. In other embodiments, liver fibrosis is determined by ex vivo measurement of liver stiffness by vibration-controlled transient elastography.

It is preferred when the presence of liver inflammation in a patient is determined by combining the values of the biomarkers in the function, in particular via a NASH test.

An apparatus for carrying out the method is also subject of the invention.

In particular, the invention also relates to a method and a device for detecting liver diseases in a patient, comprising the steps of:

performing the method for detecting steatosis as disclosed herein,

performing the method for determining liver fibrosis (Fibrotest) disclosed in WO0216949,

performing a method for determining inflammation (NASH test disclosed in WO 2018050804),

conclusions about liver disease were drawn from the results obtained with each method.

Five markers of alpha 2-macroglobulin, haptoglobin, apolipoprotein A1, total bilirubin and beta-glucosidaseGamma-glutamyl transpeptidase (GGT) combinations, and adjusted for gender and age.The algorithm of (1) is as follows: 4.467 × Log (α 2 macroglobulin (g/l)) -1.357 × Log (haptoglobin (g/l)) +1.017 × Log (GGT (IU/l)) +0.0281 × age (years) +1.737 × Log (bilirubin (μmol/l)) -1.184 × ApoA1(g/l) +0.301 × gender (female ═ 0, male ═ 1) -5.540. It allows the status of fibrosis to be determined (classified as F2 or higher according to METAVIR).

The NASH assay is in the form of a1+ A2 XLog (A2M, g/l) + a3 Xage (years) + a4 XLog (ALT, IU/I) + a 5X (Apoa1, g/l) + a6 XLog (AST, IU/l) + a7 XLog (BILI, μmol/l) + a8 XLog (CT, mmol/l) + a9 Xgender (0 for female, 1) + a10 XLog (GGT, IU/l) + a11 XLog (Hapto, g/l) + a12 XLog (TG, mmol/l);

wherein

--8≤a1≤-7；

0.1. ltoreq. a 2. ltoreq.0.6, preferably 0.15. ltoreq. a 2. ltoreq.0.55;

0.02. ltoreq. a 3. ltoreq.0.05, preferably 0.03. ltoreq. a 3. ltoreq.0.04;

1.1. ltoreq. a 4. ltoreq.1.5, preferably 1.2. ltoreq. a 4. ltoreq.1.4;

--0.2≤a5≤1.0；

-1.8. ltoreq. a 6. ltoreq.2.3, preferably 1.95. ltoreq. a 6. ltoreq.2.2;

-0.8. ltoreq. a 7. ltoreq.1.6, preferably 0.9. ltoreq. a 7. ltoreq.1.5;

- -1.7. ltoreq. a 8. ltoreq.1.3, preferably- -1.6. ltoreq. a 8. ltoreq.1.4;

-0.015≤a9≤0.20；

-0.15. ltoreq. a 10. ltoreq.0.25, preferably 0.20. ltoreq. a 10. ltoreq.0.22;

--0.3≤a11≤0.1；

0.9. ltoreq. a 12. ltoreq.1.2, preferably 1.0. ltoreq. a 12. ltoreq.1.1.

The NASH test allows inflammation to be detected when the final result is above 0.25.

These tests are known in the art and their use has not been further developed.

As described above, the function used in the methods disclosed herein may be a function obtained by logistic regression (may be referred to as a logistic function).

Methods of performing logistic regression are known in the art. In summary, they include assessing individual differences in markers between a population of subjects (one with an output such as steatosis and the other without an output). The most variable markers can be selected and subjected to logistic regression analysis to account for the independent discrimination of the selected markers. If some markers do not vary between groups, the coefficients of these markers in the function will be low (thus indicating that the weights of these markers have no real effect on the presence of the trait).

The function may then be normalized (e.g., such that the final result is always between 0 and 1).

Thus, the present invention also relates to a method for obtaining a function for identifying the presence (or absence) of (or diagnosing) steatosis in a patient, wherein said function combines the concentration values of biochemical markers in the blood/serum or plasma of said patient and optionally the age, sex of the patient, said method comprising the steps of:

a) assessing the presence of steatosis in a group of patients, wherein the values of circulating biochemical markers of the patients are known;

b) by one-dimensional analysis, circulating biochemical markers with significantly different values between the following groups were identified:

i. patients suffering from steatosis, and

patients without steatosis

c) Identifying whether the age, gender, of the patients differ significantly between the following groups:

i. patients suffering from steatosis, and

patients without steatosis

d) Performing a logistic regression analysis to assess and consider the independent discriminatory values of the markers identified in steps b) and c) for the occurrence of steatosis, wherein the body mass index and bilirubin levels are not included in the list of markers for which the logistic regression is performed;

e) thereby obtaining a function by combining these identified independent factors, wherein said function can be used to diagnose the presence of non-alcoholic steatohepatitis (steatosis) in a subject and without using bilirubin or BMI values.

In developing such a diagnostic test, one or more of the following (in any combination) are preferred:

(a) in step a), a test group comprising at least 100 patients is used;

(b) in group a), at least 50% of the patients should exhibit at least one factor of the metabolic syndrome (and therefore be at risk of suffering from steatosis), without excluding the factor (chronic or acute liver disease);

(c) in the group of a), at least 10% of patients are patients without steatosis and without activity (activity);

(d) the biochemical marker of step (a) is selected from the group consisting of alpha 2-macroglobulin (A2M), GGT (gamma-glutamyltranspeptidase), haptoglobin, apolipoprotein A-I (apoA1), alanine Aminotransferase (ALT), aspartate Aminotransferase (AST), triglyceride, total cholesterol, fasting glucose, gamma-globulin, albumin, alpha 1-globulin, alpha 2-globulin, beta-globulin, IL10, TGF-beta 1, apoA2, apoB, cytokeratin 18, platelet count, prothrombin level, hyaluronic acid, urea, N-terminal type III procollagen, type 1 tissue inhibitor metalloproteinase (TIMP-1), type IV collagen (Coll IV), osteoprotegerin, miRNA122, cytokeratin 18, Serum Amyloid A (SAA), alpha-1-antitrypsin (isoform 1), Fructose bisphosphate aldolase A, fructose bisphosphate aldolase B, fumarylacetoacetate, transthyretin, PR02275, C-reactive protein (isoform 1), leucine-rich alpha-2-glycoprotein, serine protease inhibitor A11, DNA-directed RNA polymerase I subunit RPA1, masking protein (isoform 1), alpha skeletal muscle actin, aortic smooth muscle actin, alkaline phosphatase, atypical protein C22orf30 (isoform 4), serum amyloid A2 (isoform a), apolipoprotein C-III, apolipoprotein E, apolipoprotein A-II, polymeric immunoglobulin receptor, von Willebrand factor, aminoacylase-1, G protein-coupled receptor 98 (isoform 1), paraoxonase/arylesterase 1, complement component C7, heme-binding protein, complement C1q subcomponent, Paraoxonase/lactamase 3, complement C2 (fragment), pluripotency proteoglycan core protein (isoform Vint), extracellular matrix protein 1 (isoform 1), E3 SUMO-protein ligase RanBP2, haptoglobin-related protein (isoform 1), adiponectin, retinol binding protein, ceruloplasmin, α 2 antiplasmin, antithrombin, thyroxine binding protein, protein C, α 2 lipoprotein, tetranectin, fucosylated A2M, fucosylated haptoglobin, fucosylated apoA1 and carbohydrate deficient transferrin;

(e) the biochemical marker of step (a) is selected from the group consisting of alpha 2-macroglobulin, AST (aspartate aminotransferase), ALT (alanine aminotransferase), GGT (gamma-glutamyl transpeptidase), total bilirubin, haptoglobin, apoA1, triglyceride, total cholesterol, fasting glucose, gamma globulin, albumin, alpha 1-globulin, alpha 2-globulin, beta-globulin, IL10, TGF-beta 1, apoA2, apoB, cytokeratin 18 and cytokeratin 19 components, platelet number, prothrombin levels, hyaluronic acid, urea, N-terminal type III procollagen, type 1 tissue inhibitor metalloproteinase (TIMP-1), type IV collagen (Coll IV), and osteoprotegerin;

(f) the biochemical markers of step (a) are alpha-2-macroglobulin, ApoA1, GGT, haptoglobin, ALT, AST, fasting glucose, total cholesterol and triglycerides;

(g) using at least 6, more preferably at least 7, more preferably at least 8, more preferably at least or exactly 9 biochemical markers in the mathematical function;

(h) the finally obtained function comprises at least one, preferably two physical characteristics of the patient, said physical characteristics being selected from the group consisting of the sex and the age of the patient;

(i) the function includes values measured for circulating alpha-2-macroglobulin, ApoA1, GGT, haptoglobin, ALT, AST, fasting glucose, total cholesterol, triglycerides and values corresponding to the age and sex of the patient;

(j) the function combines the measurements of alpha-2-macroglobulin, ApoA1, GGT, haptoglobin, ALT, AST, fasting glucose, total cholesterol and triglycerides and the age and sex of the patient;

(k) the method further comprises the following steps: the logistic function is validated in a validation set comprising at least 100 patients.

In order to obtain a function that is as accurate as possible, the number of patients in the group should be as large as possible, which means that it preferably comprises more than 50 patients, preferably more than 100 patients, preferably more than 200 patients, more preferably more than 500 patients, or even more than 1000 patients. As stated, there is no upper limit on the number of patients, more being better.

After diagnosis of steatosis in a patient, the physician can initiate and provide an appropriate diagnosis of the associated severe disease, inflammation (NASH) or fibrosis. The invention also relates to a method for treating NASH patients in need of confirmation of the presence of steatosis, comprising the steps of: the diagnostic method as disclosed above is performed and the patient is provided with an appropriate treatment according to its degree of steatosis. In another embodiment, the present invention provides a method for treating a patient comprising the step of providing an appropriate steatosis treatment to the patient depending on the degree of steatosis. Such treatment should be particularly carried out when the final value calculated by the method disclosed above is higher than 0.25, preferably higher than 0.50.

The primary treatment for patients with steatosis is dietary therapy in order to control obesity, hyperglycemia, hyperlipidemia, and excess iron levels (if present). Drinking should also be avoided. It is also intended to reduce the weight of the patient.

Drugs may also be provided to patients when steatosis is associated with NASH or fibrosis. By way of illustration, pioglitazone, a type 2 diabetic drug, may be cited, particularly in NASH patients without diabetes.

Vitamin E can also be used in people with NASH and without diabetes or cirrhosis.

Obeticholic acid (obetichicholic acid) may also be cited to improve the histological features of non-alcoholic steatohepatitis. (Neuschwender-Tetri et al, Lancet.2015Mar 14; 385 (9972): 956-65).

Other drugs such as ethyl-eicosapentaenoic acid or ethyl eicosapentaenoate (ethysiposantetate) (clinical trial NCT01154985), semuzumab (simtuzumab) (GS 6624) (clinical trial NCT01672866 and NCT01672879), GFT 505 (clinical trial NCT01694849), liraglutide (clinical trial NCT01237119), losartan (clinical trial NCT01051219), cenicriviroc (clinical trial NTC002217475), seletracertib (selonsertib) and aramchol can also be used.

A particular benefit of this approach is that it allows the follow-up of patients and the extent of steatosis in patients to be managed, and that further diagnostic tests (e.g. biopsies) are only proposed for a part of patients that do not improve over time.

The following examples are intended to illustrate one aspect of the invention, but should not be construed as limiting the invention.

Drawings

FIG. 1: for constructing, validating and targeting populations.

FIG. 2: for the diagnosis of steatosis > -5%, SteatoTest-2 was non-inferior compared to the original SteatoTest.

Drawing notes:¹NASH patients were analyzed at 24-week biopsy according to inclusion criteria of the trial, while there was no steatosis at baseline biopsy<5% (72 cases). Of the 63 cases analyzed at 24 weeks, only 3 did not have steatosis ((R))<5%). Since the sample size is very small, the comparison performance is weak. For the integrated analysis, all 72 baseline biopsy cases were included.

²AUROC is greatly reduced in chronic hepatitis c by a spectrum effect, since the case with the lowest steatosis (1-4%) represents 732/918 (80%) of S0 as defined by the CRN score. Unfortunately, these minimal steatosis grades are not described in metabolic diseases.

³AUROC was increased by the spectral effect of the control group, as the median number in the steatosis test of the donor was much lower than the biopsy control, thereby increasing specificity. The prevalence of steatosis does not increase significantly.

⁴AUROC is greatly affected by the greatly increased prevalence of controls, although higher in the median,while the prevalence of steatosis compared to the integrated biopsy population was divided by two.

⁵The addition of a subset of donors did not significantly alter AUROC.

FIG. 3: reduction of steatosis, NASH and fibrotic biomarkers in the selongotuber's test. Patients with NAS improvement vs patients with no improvement at 24 weeks: stepote-2 decreased by-0.02 vs +0.04(P ═ 0.03), stepote decreased by-0.014 vs +0.014(P ═ 0.02), NashTest-2 decreased by-0.06 vs +0.01, and fibritest decreased by-0.02 vs +0.02(P ═ 0.04).

Detailed Description

Examples

Example 1 summary

The method comprises the following steps: five different subgroups of 2997 biopsy patients were evaluated for test construction and validation, and four for assessing the prevalence of steatosis in the target population with increased risk of steatosis. The performance of SteatoTest-2 was compared to the reference test using a non-inferiority test (0.10 cutoff) and a Lin consistency factor.

As a result: the AUROC of SteatoTest-2 is no worse than the reference test (P < 0.001). AUROC varied in SteatoTest-2 and reference tests according to subgroup and prevalence of steatosis, with 0.772 (95% CI 0.713-0.820) vs 0.786(0.729-0.832) in 2997 biopsy cases and 0.822(0.810-0.834) vs 0.868(0.858-0.878) in 5776 cases (including healthy subjects without steatosis risk factors as controls), respectively. The Lin coefficients are highly consistent (P <0.001), from 0.74(0.74-0.74) in the hypothetical NAFLD to 0.91(0.89-0.93) in the building subgroups.

And (4) conclusion: SteatoTest-2 is simpler and not inferior to the first generation SteatoTest in diagnosing steatosis, and is not limited by body mass index and bilirubin.

Example 2 patients and methods

Design of research

Prospective analysis of the study of a patient subgroup (figure 1) the main objective was to construct a simplified novel SteatoTest-2, which is highly consistent with and not inferior to the first generation SteatoTest, with higher applicability and lower risk of false positives in the general population. SteatoTest-2 was also evaluated on paired biopsies before and after treatment in a prospective trial of Seron Christian for treatment of NASH.

The second objective was to evaluate the performance of SteatoTest-2 when used in combination with NashTest-2 (as disclosed in WO 2018050804) in a non-invasive algorithm that replicates the histological NASH algorithm (CRN or FLIP) that requires the presence of steatosis for diagnosing NASH (Poynard et al, Eur JGastroenterol Heapatol.2018; 30: 384-.

The variability of the SteatoTest-2 AUROC was evaluated, relating to prevalence of steatosis, spectral effects, degree of inflammation, stage of fibrosis, fasting glucose and BMI. Finally, SteatoTest-2 was validated for the CRN grade of moderate and significant steatosis (European liver research Association (EASL); European diabetes research Association (EASD); European obesity research Association (EASO); EASL-EASD-EASO Clinical Practice Guidelines for the management of non-alcoholic fat liver disease: J hepatol.2016; 64: 1388-.

Patients and controls

A total of 9 different subsets of individual data were included. All data was published previously (fig. 1). The first was a subset of the constructs originally tested (C1), including 307 cases with different causes of steatosis, and controls (6.Poynard et al, Comp Hepatol 2005; 4: 10). Four subgroups for validation, (V1) included 600 NAFLD patients (Munteanu et al, Aliment Pharmacol Ther 2016; 44: 877-; (V2) included 481 obese patients (Poynard et al, PLoS one.2012; 7: e 30325); (V3) included 72 NASH patients (Loomba et al, hepatology.2017); and (V4) included 1537 patients with chronic hepatitis C (Poynard et al, J hepatol.2011; 54: 227-. These five subgroups were biopsied. Finally, four biopsy-free subgroups were used to assess the prevalence of steatosis presumed by SteatoTest-2 and SteatoTest in target subjects with an increased risk of steatosis; (T1)327 blood donors (Jacquenet et al, Clin Gastroenterol hepatol.2008; 6: 828-31); (T2)7416 healthy volunteers (Poynard et al, BMC Gastroenterol 2010; 10: 40); (T3)359 patients with type 2 diabetes (jacquenet et al, supra) and (T4)133045 patients with NAFLD (muntanu et al, 2016, supra). Controls with low or very low risk are described in the statistical section.

HistologyReference to

All biopsies are scored by an experienced pathologist (with no knowledge of historical biopsy reports, test results, and other clinical data). For the primary endpoint, the FLIP-CRN scoring system known in the art was used. In addition to the foam microvesicles (microviscles), the steatosis score (S) was evaluated and the amount of large or medium lipid droplets was estimated from 0 to 3 (S0: < 5%; S1: 5-33%, mild; S2: > 33-66%, moderate; S3: > 66%, significant). The activity scale (A, from 0 to 4) is an unweighted sum of hepatocyte ballooning (0-2) and lobular inflammation (0-2). The case of a0(a ═ 0) had no activity; a1(a ═ 1) mild range of activity; a2(a ═ 2) moderate activity; a3(a ═ 3) severe activity, a4(a ═ 4) very severe activity. Fibrosis stage (F) was assessed by the following score: stage 0(F0) none; stage 1(F1) sino-3 or 1c portal fibrosis with 1a or 1 b; stage 2(F2) was perisinus and periportal fibrosis with no bridging; stage 3(F3) is bridging fibrosis, and stage 4(F4) is cirrhosis. To reduce inter-observer variation and normalize readings according to the new SAF-FLIP histological classification, reports reviewed by members of the FLIP pathology association (Frederic Charlotte for C1, V1, V2, pierce Bedossa for C1, V1, V4, Dina tianaakos for V1) or CRN (Zack goodmadman for V1 and V3) were used.

Blood testing

FibroTest, ActiTest, and pristine SteatoTest were patented as "in vitro diagnostic multiplex index assays" for diagnosing the METAVIR fibrosis stage (including cirrhosis), SAF equivalent activity, and SAF equivalent steatosis grade, respectively (Bedossa et al, Hepatology 2012; 56: 1751-9). Quantitative NashTest-2 was constructed and validated internally in 1081 patients at risk for metabolic liver disease (Poynard et al, Eur J Gastroenterol hepatol.2018; 30: 569-. These tests are only available online and include clinical safety algorithms. The recommended cut-off value is the same regardless of chronic liver disease. The analyzer and kit were BioPredictive validated and recommended, and all control assays were performed at the reference biochemistry department of Piti e Salp E trie Hospital.

The original SteatoTest recommended in the recent NAFLD guideline was used as a comparison subject (european liver research association (EASL), supra). SteatoTest-2 was constructed by regression analysis, without BMI or total bilirubin as a component, but using AST. Thus, the new test included the following ten ingredients in its proprietary formulation: alpha 2-macroglobulin, apolipoprotein a1, haptoglobin, GGT, ALT, AST, total cholesterol and fasting glucose, and the new assay was adjusted for age and gender.

Statistical analysis

Protocols and analyses followed the FibroStaRD recommendation for steatosis. The limitations defined by the FLIP and CRN standards for metabolic liver disease and their effects on the construction of non-invasive tests were previously discussed (Poynard et al, Eur J Gastroenterol Heapatol.2018; 30: 384-. These limitations include the presence of appropriate histological controls in only 2.2% (13/576) cases without steatosis and inflammatory activity in the reference study of the CRN group.

SteatoTest-2 was constructed retrospectively from a sample size of 307 cases using all components, according to the same subset used for construction of the original SteatoTest (C1). To ensure proper performance of the non-inferiority test between the new test and the AUROC of the reference test, and to ensure proper performance of the concordance correlations of different subsets of data, more than 2000 subjects at risk for metabolic or viral steatosis (performing focused biopsies) were included. Two validation subgroups, V3 and V4, have not previously been published for evaluating the performance of SteatoTest.

Empirically estimated AUROC was compared directly for the primary endpoint of non-inferiority, and a non-binary AUROC was not used, since the same patient received both new and reference tests in each data subset. The primary endpoint was to compare the test results of all the counted biopsy cases for diagnosis of steatosis (CRN grade from S1 to S3) to all grades without steatosis (S0).

To prevent the limitations of previous biomarker studies, which included less than 30 cases without steatosis, a number of controls were discussed, verifying that 158 cases of the subgroup (V1-V4) had histologically confirmed CRN-S0 grade, and 2779 biopsy-free controls of the T1 (207) and T2(n 2562) subgroups (fig. 2). These biopsy-free controls were defined by the following, according to the same criteria as the controls studied, which were used to evaluate local proton Magnetic Resonance Spectroscopy (MRS) performance: triglyceride content of liver (HTGC): no recognizable risk factor, BMI less than 25kg/m²No diabetes, fasting glucose below 6.1mmol/L, minimal alcohol consumption (20 g in women, 30g in men), and no known liver disease. The second endpoint was a specific AUROC assessed in three control groups (biopsy + T1-control, biopsy + T2-control, biopsy + T1+ T2-control).

The cutoff value is based on the method used for MRS. In this study, the 95% percentile of HTGC evaluated in 345 controls without risk of steatosis was 5.56%. This corresponds to a liver level of 55.6mg/g, considered as a cut-off value for the upper normal limit (ULN) of MRS, a reference value without steatosis (S0 rating). In our study, the selected cut-off optimized the high negative predictive value (at least 90%) of ST2 to diagnose steatosis ≧ 5%. Based on the usual range of prevalence of steatosis in adults (17-46%), 18.1% (95% CI 17.2% -18.9%; 1336/7395) was selected as the predetermined prevalence to determine the negative predictive value of the new cutoff value for SteatoTest-2. This prevalence was previously assessed in continuous healthy volunteers representing a french population aged 40 or over 40.

The performance of SteatoTest-2 was compared to a reference SteatoTest in subgroups C1 to V4, using a non-inferiority test (0.10 cutoff) that predicts the difference between the AUROC of NAS-CRN steatosis grades 1 to 3 and grade 0 (no steatosis or less than 5% hepatocytes). Sensitivity analysis was performed to assess the impact of variability factors (inflammation grade, fibrosis stage, obesity and fasting glucose (with two cut-offs of 6.1 for insulin resistance and 7.0 for type 2 diabetes)).

To ensure that the predictions of Steatotest-2 and Steatotest on steatosis grade were similar in all subgroups, the level and significance of the Lin consistency factor was evaluated between the two tests.

The median and quartile distributions for all tests were graphically represented according to histological score and control subgroups. The Tukey-Kramer test compares all pairs simultaneously using mean-difference confidence intervals and P values. The notched boxplot was constructed using the following formula: median ± (1.57xIQR/√ n). If the grooves of two box lines do not overlap, the median is significantly different.

In patients in the sementotal trial, liver biopsies and SteatoTest, SteatoTest-2, NashTest-2, and FibroTest including serum markers were performed at baseline and 24 weeks of treatment. The difference in paired test results at baseline and 24 weeks of treatment was compared between patients with no (non-responder) and with (responder) histological improvement (defined as an improvement of ≧ 1 in the NAFLD activity score). All analyses were not aware of the therapeutic effects described in the protocol.

The effect of the definition of steatosis (extrapolated from SteatoTest 2) on the NASH prevalence extrapolated from NashTest-2 in a subgroup of non-biopsied patients was evaluated based on either the standard CRN algorithm (steatosis ≧ 5%) or the previously published FLIP algorithm for significant activity (FLIP score A2). The purpose of use in a large population is: determining a sensitivity cutoff for SteatoTest-2 for subjects at least at steatosis grade S1; and those cases with clinically significant NASH (i.e., at least a grade of N2 in the CRN or FLIP scoring system).

All statistical analyses were performed using NCSS-12.0 and R.

Example 3 results

Characteristics of the subject involved

A total of 2997 biopsy and histologically scored patients were evaluated by the SAF scoring system. Although the characteristics of the included subjects have been published in previous publications, the value of the present integrated database lies in its broad characteristics, which allows the robustness of blood tests to be assessed in terms of variability factors. IQR at age is between 34 and 61 years. The prevalence of histological steatosis ranges from 7.3% to 24.5%, NASH ranges from 13.7% to 100%, cirrhosis ranges from 0% to 42.6%, the prevalence of T2 diabetes ranges from 7.5% to 70.8%, and obesity (BMI ≧ 30) ranges from 13.6% to 100%. A subgroup of obese subjects was younger, with a higher percentage of women and a lower prevalence of advanced fibrosis than the other subgroups.

New Steatotest-2

Unlike SteatoTest, SteatoTest-2 does not include BMI or total bilirubin, but includes AST, with the remaining 9 components having different independence factors.

Clinical features in four target subgroups with increased risk of steatosis were recorded as well as the grade and fibrosis stage of steatosis and NASH as predicted by blood tests. As expected, the prevalence of steatosis predicted by SteatoTest increased from 15.5% in donors to 84.4% in subjects receiving fibritest.

AUROC of SteatoTest-2 is not worse than that of the reference SteatoTest as a comparison object. All non-inferiority tests were significant (P <0.001) (fig. 2). The AUROC results for SteatoTest-2 and SteatoTest vary depending on the prevalence of control and steatosis: 0.772 (95% CI 0.713-0.820) and 0.786(0.729-0.832) respectively in 2997 biopsy cases (prevalence of steatosis 62%) and 0.822(0.810-0.834) and 0.868(0.858-0.878) respectively in 5776 control cases (prevalence of steatosis 32%) that included no steatosis risk factor and no biopsy (FIG. 2).

The effect of control selection on SteatoTest-2 AUROC was evaluated. AUROC was changed from 0.734 (95% CI 0.715-0.751) for the control-S0-biopsy to 0.822(0.810-0.834) for the control-S0-biopsy-T1-T2. Since the spectrum of the steatosis case is always defined by the biopsy (stages S1 to S3), there is no change in the sensitivity of the test. Therefore, AUROC was affected by both the spectrum of the control group directly related to the specificity of the test and the change in prevalence of steatosis from 32% to 62%. When control-T1 (n-207, SteatoTest-2 median 0.15) was added to the control-S0-biopsy, AUROC increased slightly from 0.734(0.715-0.751) to 0.767(0.750-0.782) and the prevalence of steatosis decreased from 62% to 58%. AUROC 0.815(0.803-0.827) increased significantly and the prevalence of steatosis decreased significantly from 62% to 34% when control-T2 (n-2562, SteatoTest-2 median 0.30) was added to control-S0-biopsy.

The Lin coefficients were highly consistent across all subgroups (P <0.001) and ranged between 0.74(0.74-0.74) in a4 to 0.91(0.89-0.93) in C1.

Cut-off value of SteatoTest-2

The upper 95% percentile (ULN) in Steato-Test-2 was 0.40 in 177 donors with BMI <25 and fasting glucose < 6.1. This cut-off value was chosen as the ULN of ST2 to predict the presence or absence of steatosis of at least 5%. The following cutoff values were recommended based on median and 95% confidence intervals: s1> is 0.40, S2> is 0.55, S3> is 0.62.

The sensitivity of the integrated biopsy database using a 0.40 cutoff value was 79% (77-85) with 92% predictive value when adjusted for a predetermined prevalence of 18%. The corresponding specificity was 50% (47-53).

Comparison between NASH prevalence inferred from CRN or FLIP simplified algorithms

In the NAFLD subgroup, the prevalence of steatosis was 84.7%. The prevalence of clinically significant NASH (moderate or severe) using the CRN algorithm was 72.3% (434/600), whereas the prevalence using the FLIPA2 algorithm was 80.3% (482/600), which was significantly (P ═ 0.001) different by 8.0% (95% CI 3-13%).

The prevalence of steatosis in a subgroup of obese subjects was 65.5%. The prevalence of clinically significant NASH using the CRN algorithm was 16.6% (80/481), whereas the prevalence using the FLIP-a2 algorithm was 18.1% (87/481), which differed non-significantly by 1.5% (-4%; 6%; P ═ 0.55).

All patients enrolled in the NASH trial subgroup had histological steatosis and NASH at enrollment, and all had steatosis as predicted by SteatoTest-2, regardless of the simplified definition of CRN or FLIP.

The prevalence of steatosis in the donor subgroup was 15.5%. The prevalence of clinically significant NASH (moderate or severe) using the CRN algorithm was 3.1% (10/322), while the prevalence using the FLIPA2 algorithm was 5.3% (17/322), which was not significant (P ═ 0.17) with a 2.2% (-1.2; 6.8%) difference.

The prevalence of steatosis in a subset of healthy volunteers was 50.1%. The prevalence of clinically significant NASH (moderate or severe) using the CRN algorithm was 22.5% (1667/7416), whereas the prevalence using the FLIPA2 algorithm was 29.9% (2220/7416), which differed significantly (P <0.001) by 7.5% (6.0; 8.9%).

The prevalence of steatosis in a subgroup of diabetic patients was 85.2%. The prevalence of clinically significant NASH (moderate or severe) using the CRN algorithm was 41.8% (150/359), whereas the prevalence using the FLIPA2 algorithm was 44.6% (160/359), which was not significant (P ═ 0.45) with a 2.8% (-4.7; 10.3%) difference.

Improvements in SteatoTest-2, NashTest-2, FibroTest in the Stachromo test (FIG. 3)

All 72 patients with NASH and stage 2 (35%) or 3 (65%) fibrosis were included. At baseline, all patients had NAS scores ≧ 5 (100%). At 24 weeks, 49% of patients had at least a1 point improvement in NAS. The patients who improved were-0.02 vs +0.04(P ═ 0.03) for SteatoTest-2, 0.014vs +0.014(P ═ 0.02) for SteatoTest-2 (B in fig. 3), 0.06vs +0.01 for NashTest-2 (C in fig. 3), and 0.02vs +0.02(P ═ 0.04) for fibritest-3, respectively, as compared with the patients who did not improve. Although the range of NASH disease severity is narrow (only grade 2 and grade 3), NashTest-2 predicts that NAS improves AUROC to 0.700 (95% CI 0.543, 0.809; p ═ 0.003).

Sensitivity analysis

The AUROC of SteatoTest-2 is no worse than the reference SteatoTest, regardless of the associated inflammatory activity, stage of fibrosis, fasting glucose levels, or obesity. The only exception was a small subgroup of hepatitis c patients with no mobility, of which only 11 had steatosis. Statistical comparisons could not be made in both subgroups due to the prevalence of steatosis of 98% (i.e. in obese patients with advanced fibrosis and fasting glucose > -7 mmol/L).

Correlation with grade of CRN steatosis

Histological grading with SteatoTest-2 revealed significant differences at all different stages of the disease. The median of S0, S1, S2 and S3 was 0.40, 0.53, 0.60, respectively.

Example 4 discussion

The construction and characterization of a new blood test for the diagnosis of steatosis was described in this study and verified that the results were not inferior to the reference SteatoTest, which is a well-established comparison subject (european liver research institute, supra). The advantage of this new test is that its composition does not include body mass index or total bilirubin, which are two causes of significant differences.

Non-inferiority and consistency with the comparison object

Steato-test2 has been shown to be less aberrant than comparable to the most common cause of hepatic steatosis in biopsy patients with metabolic liver disease (overweight, type 2 diabetes, dyslipidemia) and chronic hepatitis C. To assess their specificity, this new test was also demonstrated to be non-inferiority in both the low risk (including general population) and very low risk groups of steatosis (e.g. blood donors, healthy volunteers and hepatitis c without steatosis). Furthermore, a highly significant quantitative agreement between the two tests was confirmed in all subgroups. The AUROC comparison demonstrates non-inferiority after taking into account, among others, the following major sources of variability: prevalence of steatosis, stage of fibrosis, grade of inflammatory activity, and prevalence of diabetes, obesity, increase in fasting glucose with standard cut-off values (6.1 and 7.0 mmol/L).

Selection and Effect of controls

This study underscores the importance of the selection of controls (no steatosis) and cases (with steatosis) in the study of AUROC to evaluate biomarkers. The construction of this new test takes into account the methodological limitations of previous studies, relating to sample size and the definition and control of steatosis and its effect on AUROC.

Due to ethical limitations in biopsy and normal liver function testing in healthy controls or in cases of metabolic liver disease, the same incorporated standards as used for MRS were taken as our reference to define the presence or absence of steatosis and to define the ULN in SteatoTest-2. In constructing a steatosis biomarker, it is believed that these criteria should be recommended in the guidelines to standardize the methodology to prevent human divergence.

Whether biopsy based on MRS to select patients with steatosis significantly affects AUROC in biomarker studies. If a study was designed to use MRS first, only subjects with steatosis according to MRS would be biopsied, and the controls used to assess specificity would be those patients with steatosis according to MRS rather than according to biopsy. This is obviously rare and not in accordance with the frequent use case. Therefore, these studies cannot reliably assess the specificity and AUROC for steatosis. This strategy is also problematic for assessing the value of NASH biomarkers, as ballooning and lobular inflammation may be present with or without steatosis at biopsy < 5%. Due to these limitations, it has been decided to assess the specificity of the steatosis biomarkers in large samples. Like the first generation SteatoTest, all available biopsies from 498 patients with chronic viral hepatitis were included, 1537 subjects without risk of steatosis served as controls, and the ULN of MRS served as the basis for our ULN (Szczepaniak et al, Am J Physiol Endocrinol Metab.2005; 288: 462-8).

In this study, the effect of any selection of controls when evaluating the steatosis non-invasive test by AUROC was confirmed. AUROC "artificially" changed from 0.734 for biopsy-control to 0.822 for biopsy + T1+ T2-control without changing the sensitivity of the test. The results were affected by both the spectrum of the control group and the prevalence of steatosis varying from 32% to 62% depending on the choice of cases without steatosis. Since the prevalence of steatosis varies between 20% and 50% (by MRS speculation), the performance of new steatosis biomarkers must be assessed according to the source of these variability, avoiding indirect comparisons of AUROC.

Performance for significant (grade S2) and severe (grade S3) CRN steatosis

SteatoTest-2 was also shown for the first time to have a significant difference between the four stages of CRN scoring. Although the significant difference between the semi-quantitative correlation and the histological stage in our study is not similar to the results for MRS (which has a better correlation with the percentage of hepatocellular steatosis than any blood test), the objective was to validate a robust test without practical MRS limitations. Although the main value of SteatoTest-2 is its sensitivity and the associated high negative predictive value (92%) for diagnosing at least 5% steatosis, there is also a significant correlation between the test value of steatosis and the histological grade. Furthermore, AUROC (0.603; 0.577-0.229; P <0.001 vs. randon) has limited clinical value for the diagnosis S2, S3 vs. S1. Indeed, the AUROC was evaluated within a narrow range of steatosis spectra, without the S0 control or without a control without the risk of steatosis.

Validation in NASH paired biopsy (FIG. 3)

The changes in the three non-invasive tests, SteatoTest-2, NashTest-2, and FibroTest, were first demonstrated in a second phase trial of Stamperoth to reliably predict improvement in NAS scores in NASH patients. This confirms that paired SteatoTest-2 detected an improvement in steatosis in patients with moderate and severe grade steatosis, which was also observed in biopsies and MRS.

Despite retrospective design, SteatoTest-2 is clearly no worse than and highly consistent with the reference SteatoTest, which has been widely used for diagnosis of steatosis and validated by guidelines.

In summary, the new multi-analyte SteatoTest-2 simplifies the reference SteatoTest for diagnosing steatosis and is considered to be no worse than the reference SteatoTest and without variability due to body mass index and the risk of false positives associated with unbound bilirubin.