WO2013049509A1 - Assay panel for non-alcoholic steatohepatitis - Google Patents

Abstract

The likelihood that a subject has significantly active histological non-alcoholic steatohepatitis (NASH) or a change in histological NASH can be predicted by measuring the level of at least four biomarkers selected from the group consisting of adiponectin, full-length keratin-18, caspase-3-cleaved keratin-18, C-Reactive Protein, insulin-like growth factor binding protein 1, and alanine aminotransferase, in a sample derived from the subject, and analyzing the level of the at least four biomarkers against reference values for those biomarkers. This approach can be used to screen for the presence or absence of NASH, to monitor the progression or regression of NASH over time, or to monitor the efficacy of a therapy for treating NASH.

Description

ASSAY PANEL FOR NON-ALCOHOLIC STEATOHEPATITIS

[0001 ] This application claims priority to U.S. Provisional Application No.

61/540,249, filed September 28, 201 1 , which is herein incorporated by reference in its entirety.

STATEMENT REGARDING GOVERNMENT-SPONSORED RESEARCH

[0002] This invention was made with government support awarded by the Department of Health and Human Services of the United States of America, National Institutes of Health. The government has certain rights in the invention.

FIELD

[0003] The present disclosure relates to methods of detecting and monitoring non-alcoholic fatty liver disease (NAFLD) in a subject. In particular, the present disclosure relates to methods of screening for non-alcoholic steatohepatitis (NASH) and monitoring the effect of therapy on a subject with NASH.

BACKGROUND

[0004] The incidences of metabolic diseases such as diabetes and gout are rising in parallel with increases in body weight in the U.S. population. About 30% of adults and 10% of children or adolescents in the U.S. have excess fat in their livers, a condition called nonalcoholic fatty liver disease (NAFLD), which in itself is relatively benign in most affected people. However, about 10-20% of people with fatty liver have a more serious condition called nonalcoholic steatohepatitis (NASH), which may progress to hepatic fibrosis and cirrhosis, as well as increased risk of cardiovascular disease, leading to increased morbidity and mortality. A majority of patients diagnosed with NAFLD and NASH appear to be asymptomatic.

[0005] Liver biopsy is the current gold standard for diagnosing NASH, however, biopsy may be costly, invasive and subjective. A biomarker or set of biomarkers in blood that correlate with liver histology may identify patients who have no need for a biopsy, and also provide early detection of NASH, as well as information about disease progression or regression over time, which may improve patient treatment and outcome.

[0006] Nonalcoholic fatty liver disease (NAFLD) is the most common form of chronic liver disease in both children and adults. It encompasses a wide spectrum of conditions associated with overaccumulation of fat in the liver ranging from simple steatosis to nonalcoholic steatohepatitis (NASH) and cirrhosis. Simple steatosis is the most common form of NAFLD and typically follows a benign nonprogressive clinical course. In contrast, NASH is a potentially serious condition, since as many as 25% of these patients may progress to cirrhosis and experience complications of portal hypertension, liver failure, and hepatocellulate carcinoma.

[0007] Emerging data suggest that heptocyte apoptosis, a specific form of cell death, may play an important role in liver injury and disease progression in NAFLD. It has been shown that caspase activation and liver cell apoptosis are prominent pathological features of human NAFLD. Moreover, the degree of apoptosis correlated with the severity of steatohepatitis and the stage of fibrosis. (Feldstein, A. et al., "Cytokeratin-18 Fragment levels as noninvasive biomarkers for nonalcoholic steatohepatitis: A Multicenter Validation Study," Am. Assoc. Study Liver Diseases, Wiley InterScience, 2009.)

[0008] The apoptotic pathway is composed of two arms: the intrinsic pathway (initiated by cellular stress) and extrinsic pathway (stimulated through a death receptor-mediated process). Both pathways are suspected to be involved in the pathogenesis of NASH. In the final common step of apoptosis, the effector caspases (in particular caspase-3 and caspase-7) are activated. These specific intracellular proteases are known to cleave several cellular substrates, including keratin-18 (K- 18), the major intermediate filament protein in the liver. Antibodies against caspase- generated K-18 fragments have been shown to specifically label early apoptotic cells. It has been shown that these caspase-generated K-18 fragments could be detected in the blood of patients with chronic hepatitis C infection using a novel enzyme-linked immunosorbent assay kit. (Feldstein, 2009) [0009] Proper diagnosis is crucial to effective treatment and projecting the progression of each disease, which varies greatly between simple steatosis, NASH and NASH-related cirrhosis and fibrosis. However, there are limited options available to establish the proper diagnosis of NASH or NASH-related cirrhosis and fibrosis. Additionally, optimal management of patients with NASH continues to evolve in terms of effective treatment and monitoring for progression of disease versus efficacy of therapy. Common practice includes monitoring of ALT despite its clear limitations with regard to NASH histological activity. Repeat liver biopsy may be recommended for more definitive assessment despite presence of risk and lack of a standard follow-up interval. Recent studies have indicated that markers of cell injury such as keratin 18 fragments and markers of insulin metabolism correlate to histological severity in NASH (Younossi et al., Obes. Surg. 2010), but their utility in monitoring therapy is uncertain.

[0010] Liver biopsy is an invasive procedure associated with various complications but it is currently the only reliable method of differentiating simple steatosis from NASH and evaluating histological severity of the disease. In light of the increasing prevalence of NAFLD there remains a need for non-invasive methods of detecting and screening for NAFLD and NASH. Additionally, there is a need for better methods for diagnosing NAFLD and NASH, monitoring disease progression, and determining efficacy of treatment, as well as methods of differentiating between subjects with NAFLD and NASH, and identifying subjects at risk of transitioning from NAFLD to NASH.

SUMMARY

[001 1 ] Therefore, the present disclosure generally provides methods of screening a subject to predict the likelihood of significantly active histological nonalcoholic steatohepatitis (NASH) in a subject, comprising measuring the level of at least four biomarkers selected from the group consisting of adiponectin (ADPN), full- length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT), in a sample derived from the subject, and analyzing the level of the at least four biomarkers against reference values for those biomarkers. [0012] Other aspects of the present disclosure include methods of screening a subject to identify the presence or absence of non-alcoholic steatohepatitis (NASH) in a subject, comprising measuring the level of at least four biomarkers selected from the group consisting of adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3- cleaved keratin-18 (C3C K-18), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT), in a sample derived from the subject, and analyzing the level of each of the at least four biomarkers against reference values for those biomarkers.

[0013] The present disclosure further relates to methods of measuring the progression of non-alcoholic fatty liver disease (NAFLD) to non-alcoholic

steatohepatitis (NASH) or monitoring the improvement of NASH-like conditions to NAFLD-like conditions over time of NASH in a subject, by measuring the level of at least four biomarkers selected from the group consisting of adiponectin (ADPN), full- length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT), in samples derived from the subject at two or more time points, and analyzing the level of the at least four biomarkers at a second or later point in time against the values for those biomarkers at an earlier point in time.

[0014] Similarly, the present disclosure relates to methods of measuring the efficacy of a therapy for treatment of NASH in a subject by measuring the level of at least four biomarkers selected from the group consisting of adiponectin (ADPN), full- length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT), in samples derived from the subject before and after the subject undergoes the therapy, and analyzing the level of the biomarkers after therapy against the values for those biomarkers before therapy. The present disclosure is also directed to a kit comprising a package, the package containing at least four agents for measuring the level of at least four biomarkers of interest, wherein the at least four biomarkers are selected from the group consisting of adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT). [0015] The present disclosure is based on the finding of a correlation between histological activity (NAS value) or change in NAS score, in patients with NASH and the biomarkers adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT), as described in the examples. This finding provides a predictive model for NAS values, which can be used for effective, non-invasive diagnosis of NASH as well as non-invasive assessment of NASH progression or the outcomes of therapeutic intervention, especially for short-term interval evaluations where re-biopsying is impractical.

[0016] Aside from the subject matter discussed above, the present disclosure includes a number of other exemplary features such as those explained hereinafter. It is to be understood that both the foregoing description and the following description are exemplary only.

BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying table and figures are incorporated in, and constitute a part of this specification.

[0018] FIG. 1 is a collection of box and whisker plots illustrating the empirical distributions of the potential predictors of NASH status of tested subjects (NAS<4, NAS≥4);

[0019] FIG. 2 is a graphical illustration depicting the relationship between the value of the predictor variable and the predicted probability of a NAS<4 value of a subject; and

[0020] FIG. 3 is a graphical illustration depicting the ranking of the predictors of NASH status (NAS<4, NAS≥4) when the ranking criterion is based on the difference between the observed Wald chi-squared statistic and the expected Wald chi-squared statistic under the null hypothesis of no partial association.

DETAILED DESCRIPTION

[0021 ] The present disclosure relates to methods of detecting and monitoring non-alcoholic fatty liver disease (NAFLD) in a subject. The present disclosure also relates to methods of screening for non-alcoholic steatohepatitis (NASH) and monitoring the effect of therapy on a subject with NASH. According to various embodiments, a method for screening a subject to predict the likelihood of significantly active histological non-alcoholic steatohepatitis (NASH) in a subject can comprise the step of measuring the levels of at least four biomarkers in a sample derived from the subject. The method can further comprise analyzing the levels of the at least four biomarkers against references values for those biomarkers.

[0022] In accordance with other embodiments, a method for screening a subject to identify the presence or absence of non-alcoholic steatohepatitis (NASH) in a subject can comprise measuring the level of at least four biomarkers in a sample derived from the subject. The method can further comprise analyzing the levels of the at least four biomarkers against reference values for those biomarkers.

[0023] Examples of suitable biomarkers for screening a subject to predict the likelihood of significantly active histological non-alcoholic steatohepatitis (NASH) and/or to identify the presence or absence of NASH in a subject include, but are not limited to, adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), and insulin-like growth factor binding protein 1 (IGFBP-1 ). Adiponectin and insulin-like growth factor binding protein 1 (IGFBP-1 ) are naturally found in the body of a subject and can be obtained from numerous commercial vendors of antibodies or ELISA kits. In various other embodiments, the antibodies for full-length keratin-18 are monoclonal antibodies, such as, for example, the M6 and M5 Keratin 18 antibodies (PEVIVA AB). In still other embodiments, the antibodies for caspase-3-cleaved keratin-18 are monoclonal antibodies, such as, for example, the M30^® antibody (PEVIVA AB). In various examples described herein, the M30 Apoptosense^® ELISA and M65^® ELISA kits, both manufactured by PEVIVA AB, were used in some of the experiments related to the caspase-3-cleaved K18 fragments and the full-length K-18, respectively.

[0024] It is contemplated that the number of biomarkers measured can also be at least one or at least two or at least three biomarkers. It is further contemplated that additional or alternative biomarkers can be used in addition to or as a substitution for one or more of the biomarkers listed herein. For example, alternative suitable biomarkers include, but are not limited to, resistin, insulin, fetuin, tumor necrosis factor-alpha (TNF-a), interleukin 6 (IL-6), or interleukin 8 (IL-8), C-Reactive Protein (CRP), alanine aminotransferase (ALT), or any combination thereof.

[0025] According to various embodiments, a method for monitoring the progression of non-alcoholic fatty liver disease (NAFLD) to non-alcoholic

steatohepatitis (NASH) or monitoring the improvement of NASH-like conditions to NAFLD-like conditions over time in a subject can comprise measuring the level of at least four biomarkers in a sample derived from the subject at two or more time points. The method can further comprise analyzing the level of the at least four biomarkers at a second or later point in time against the values for those biomarkers at an earlier point in time.

[0026] In accordance with various embodiments, a method for monitoring the efficacy of a therapy for treatment of non-alcoholic steatohepatitis (NASH) in a subject can comprise measuring the level of at least four biomarkers in a sample derived from the subject before and after the subject undergoes the therapy. The method can further comprise analyzing the level of the at least four biomarkers after therapy against the reference values for those biomarkers before therapy.

[0027] Examples of suitable biomarkers for monitoring the progression of NAFLD to NASH or the improvement of NASH-like conditions to NAFLD-like conditions over time and/or monitoring the efficacy of therapy in a subject include, but are not limited to, adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3- cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT). ADPN, CRP, IGFBP-1 , and ALT are naturally found in the body of a subject and can also be obtained from various commercial sources. It is contemplated that the number of biomarkers measured can also be at least one or at least two or at least three biomarkers. It is further contemplated that additional or alternative biomarkers can be used in addition to or as a substitution for one or more of the biomarkers listed herein.

[0028] According to various embodiments, measuring the level of the biomarkers can comprise the steps of reacting with a specific antibody against the biomarkers, or against any fragment of a biomarker containing an antigenic determinant. The antibody can comprise, for example, a whole immunoglobulin molecule, a monoclonal antibody, a polyclonal antibody, a chimeric antibody, a recombinant fragment of antibody, an Fab fragment, an Fab' fragment, an F(ab')2 fragment, an Fv fragment, an scFv fragment, or any combination thereof.

[0029] In accordance with various embodiments of the present disclosure, a variety of samples from a subject can be analyzed. For example, suitable samples include, but are not limited to, serum, blood, plasma, urine, saliva, cerebrospinal fluid, tissue, or a tissue extract. The samples can be diluted prior to analysis. The levels of the biomarkers within the samples can be measured using any conventional quantitative or qualitative assay technique useful for measuring the presence of a biomarker such as competitive and non-competitive immunoassay formats, antigen capture assays, and two-sandwich assays. In particular, acceptable assay methods include, but are not limited to, enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), immunofluorescent assay (IFA), chemiluminescent assay, electrochemiluminescent (ECL) assay, sandwich assay, capillary electrophoresis- based immunoassay (CEIA), magnetic capture, microsphere capture, and/or western blotting.

[0030] In accordance with various embodiments of the present disclosure, the levels of the various biomarkers can be compared to their respective reference values using any conventional technique, for example, ELISA, immunofluorescent assay (IFA), and chemiluminescent assay. Additionally, it has been found that multivariable logistic regression analysis is a powerful tool for analyzing biomarker levels for use in the methods disclosed herein. The reference value ranges for each of the biomarkers can represent, for example, a subject without NASH or a subject without NAFLD. The analysis can calculate, for example, an optimal threshold for detecting NASH. According to various embodiments, a threshold value can be selected to provide a sensitivity and/or a specificity for detecting NASH in a certain range.

[0031 ] In various embodiments of the present disclosure, the methods can further comprise analyzing the Body Mass Index (BMI) of the subject. As is well- understood in the art, BMI is defined as the weight in kilograms of a subject divided by the square of height in meters of the subject (BMI = w/h²). The methods of the present disclosure are suitable for any mammalian subject, such as a human subject.

[0032] The terms "non-alcoholic fatty liver disease" or "NAFLD," "non-alcoholic steatohepatitis" or "NASH," and "simple steatosis" as used herein refer to the definitions as currently used and accepted in the scientific community. For example, simple steatosis is the most common form of NAFLD and typically follows a benign nonprogressive clinical course. On a histological spectrum, the patients exhibit an accumulation of fat in the liver cells without significant inflammation and

hepotocellular damage. In contrast, the patients with NASH exhibit inflammation and hepatocellular damage and sometimes fibrosis, which may then progress to cirrhosis. The NASH patients may experience complications of portal hypertension, liver failure, and hepatocellular carcinoma.

[0033] The term "significantly active histological non-alcoholic steatohepatitis (NASH)" as used herein, represents a NAFLD (non-alcoholic fatty liver disease) Activity Score (NAS) of 4 or higher. The total NAS value represents the sum of scores for steatosis, lobular inflammation, and ballooning, and ranges from 0-8, as presented in Table 1 , depicting NAFLD Activity Score (NAS) and Fibrosis Staging, which is widely accepted and well-known in the art. Diagnosis of NASH (or, alternatively, fatty liver not diagnostic of NASH) should be made first, then NAS is used to grade activity. In the reference study, NAS values of 0-2 occurred in cases largely considered not diagnostic of NASH, values of 3-4 were evenly divided among those considered not diagnostic, borderline, or positive for NASH. Values of 5-8 occurred in cases that were largely considered diagnostic of NASH.

Table 1 - Components of NAFLD Activity Score (NAS) and Fibrosis Staging

2 >33-66% -

3 >66% -

Acidophil bodies are not included in this assessment,

0 No foci

nor is portal inflammation

Lobular

1 <2 foci/200x - Inflammation

2 2-4 foci/200x -

3 >4 foci/200x -

0 None -

The term "few" means rare but definite ballooned

1 Few balloon cells hepatocytes as well as cases that are diagnostically

Hepatocyte

borderline

Ballooning

Many Most cases with prominent ballooning also had

2 cells/prominent Mallory's hyalin, but Mallory's hyaline is not scored ballooning separately for the NAS

Fibrosis Stage (Evaluated separately from NAS)

0 None -

Perisinusoidal or

1 - periportal

Mild, zone 3,

1A "delicate" fibrosis perisinusoidal

Moderate, zone 3,

1 B "dense" fibrosis perisinusoidal

Fibrosis

This category is included to accommodate cases with

1 C Porta l/periportal portal and/or peri portal fibrosis without

accompanying pericellular/perisinusoidal fibrosis

Perisinusoidal and

2 - portal/peri portal

3 Bridging fibrosis -

4 Cirrhosis -

[0034] The term "NASH-like conditions" as used herein means a condition featuring both fatty liver and inflammation, regardless of etiology. For example, many of the histological signs of NASH can occur in patients with hepatitis C; the assay panel of this disclosure is applicable to detection and monitoring of liver damage in hepatitis C and other conditions overlapping with NASH pathology, including but not limited to liver damage caused by toxic chemicals, infections, autoimmune and autoinflammatory diseases. The term "NAFLD-like conditions" as used herein means liver conditions featuring fatty liver with little or no inflammatory infiltrate or fibrosis, but with variable degrees of hepatic insulin resistance, whether caused by infection, chemical toxins or autoimmune or autoinflammatory processes.

[0035] According to one or more embodiments of the present disclosure, kits for use in the methods of the present teachings are provided. In certain embodiments, a kit can comprise a package containing at least one agent for measuring the levels of selected biomarkers, such as adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), and insulin-like growth factor binding protein 1 (IGFBP-1 ). It is contemplated that the package can contain at least two agents, at least three agents, at least four agents, or more than four agents depending on the number and kinds of biomarkers being measured.

[0036] In various other embodiments, the package can contain at least one agent for measuring the levels of selected biomarkers, such as full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulinlike growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT). It is contemplated that the package can contain at least two agents, at least three agents, at least four agents, or more than four agents depending on the number and kinds of biomarkers being measured.

[0037] According to various embodiments of the kit, the agents can comprise at least one of an antibody of adiponectin (ADPN), an antibody of full-length keratin- 18 (K-18), an antibody of caspase-3-cleaved keratin-18 (C3C K-18), an antibody of C-Reactive Protein (CRP), an antibody of insulin-like growth factor binding protein 1 (IGFBP-1 ), and an antibody of alanine aminotransferase (ALT). It is contemplated that when alternative or additional biomarkers, such as those disclosed herein, are included in the kit, that kit can also contain measuring agents that can comprise an antibody of each of any such alternative or additional biomarkers used. In various embodiments, all of the measuring agents in the kit can be antibodies of the respective biomarkers to be detected. In various other embodiments, the antibodies for full-length keratin-18 are monoclonal antibodies, such as, for example, the M6 and M5 Keratin 18 antibodies (PEVIVA AB). In still other embodiments, the antibodies for caspase-3-cleaved keratin-18 are monoclonal antibodies, such as, for example, the M30^® antibody (PEVIVA AB). It is contemplated that the kit may further comprise one or more control reference samples. It is further contemplated that the kit can contain a package with individual containers for each of the antibodies.

Instructions for using the kit may be provided in the package or separately. The instructions can comprise explanations on using the contents of the kit and on measuring the levels of the biomarkers. The instructions may also include reference levels for comparing to the detected levels of the biomarkers.

[0038] The present disclosure will be better understood by reference to the following examples, which illustrate but do not limit the present teachings described herein. Moreover, it should be understood that various features and/or

characteristics of differing embodiments herein may be combined with one another. It is therefore to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the scope of the present disclosure. Other embodiments will be apparent to those skilled in the art from consideration of the disclosure and practice of the various exemplary embodiments disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope of the invention being indicated by the claims.

[0039] It is also to be understood that, as used herein the terms "the," "a," or "an," mean "at least one," and should not be limited to "only one" unless explicitly indicated to the contrary. Unless otherwise indicated, all numbers used in the specification and claims are to be understood as being modified in all instances by the term "about," whether or not so stated. It should also be understood that the precise numerical values used in the specification and claims form additional embodiments of the invention, and are intended to include any ranges which can be narrowed to any two end points within the exemplary ranges and values provided. Efforts have been made to ensure the accuracy of the numerical values disclosed herein. Any measured numerical value, however, can inherently contain certain errors resulting from the standard deviation found in its respective measuring technique.

[0040] All publications, patents and patent applications mentioned in this specification are herein incorporated by reference in their entirety into the specification to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated herein by reference. Also incorporated by reference is any supplemental information that was published along with any of the aforementioned publications, patents and patent applications. For example, some journal articles are published with supplemental information that is typically available online.

EXAMPLES

[0041 ] The following examples are intended to be non-restrictive and explanatory only.

[0042] The following solutions, diluents and buffers were used throughout some or all of the examples described herein. (1 ) Anti-HAMA Antibody Diluent: (100 mM sodium phosphate, 150 mM NaCI, 0.5% BSA, 0.5% B IgG, 0.1 % Brij 35, 0.1 % MIT, 50 g/mL MAK33-lgG1/lgG Poly, 25 g/mL HBR); (2) Conjugate Diluent, aka Conjugate Buffer. (50 mM Sodium phosphate, 12.7 mM EDTA, 1 % BSA, 0.15% Kathon CG, 0.1 % Tween® 20, 0.01 % Bovine IgG, 0.0025% Mouse IgG); (3) WD- Clean Solution: (0.56% potassium hydroxide, 0.88% sodium chloride, 0.4% Triton X- 100); (4) WD-Store Solution: (0.02% Polidocanol [Thesit], 0.12% Kathon CG-ICP); (5) WD-Diluent Solution aka DILUENT: (0.024% potassium phosphate, monobasic; 0.144% sodium phosphate, dibasic; 0.7% sodium chloride, 0.02% potassium chloride; 0.33% Tween® 20; 0.104% Kathon CG-ICP II); and (6) GLO Solution, aka GLO Plus Solution: (4.1 % potassium phosphate, monobasic; 2.14% tripropylamine [TPA]; 0.88% sodium chloride; 0.02% Polidocanol [Thesit]; 0.1 % Oxaban-A preservative).

[0043] The materials provided below were used throughout some or all of the examples described herein. The abbreviations of "mAb," "MAb," "MoAb," "moAb," or variations thereof as used herein represent "monoclonal antibody" or "monoclonal antibodies." As is understood as common practice, many of the materials and reagents listed below and used throughout the examples were derived from commercially available products and may have been conjugated or otherwise altered for purposes of the examples described herein.

EXAMPLE 1 - Histological Scoring of Biopsies and Statistical Analysis of Biomarkers

[0044] The purpose of this experiment was to develop a predictive model to estimate histological activity in patients with NASH being followed conservatively or undergoing therapeutic intervention. 1 10 NAFLD/NASH patient serum samples were collected (from S. Caldwell, MD at University of Virginia) with biopsies beginning at one year from a diagnosis of NAFLD. Thirty-three of those patients with clinically diagnosed fatty liver disease had two liver biopsies performed. One biopsy was performed prior to an intervention (placebo, omega 3 fatty acid), while the other biopsy was performed post-intervention. Each liver biopsy was evaluated by a professional hepatologist who was blind to the intervention and the biopsy was given scores for fat, inflammation, and ballooning, and similarly scored by a professional liver histopathologist, also blinded to the intervention. Based on these scores, the biopsy was then assigned a NAFLD Activity Scores (NAS). Thirteen patients had a histological "response" to NAS score < 4.

[0045] Additionally, each patient had two blood-draws. One draw was taken before the intervention and one draw was taken post-intervention. Fasting blood samples at study entry and follow-up were coded for blinded testing of adiponectin (ADPN), caspase-3-cleaved keratin-18 (C3C K-18), full-length keratin-18 (K-18), C- Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT). (ALT was assayed for using the Architect Chemical Analyzer (Abbott). The reaction detects conversion of alanine to pyruvate and then lactate.) The patient data are summarized in Tables 2A-2C. Table 2A - Combined Assay Results Pre-lntervention provides the combined results generated from each of the assays completed (ADPN, C3C K-18, K-18, IGFBP-1 , CRP and ALT), and includes the initial NAS value for each patient clinically diagnosed with fatty liver disease tested prior to the intervention. Table 2B - Combined Assay Results Post- Intervention with Response provides data generated from the tested patients that exhibited a change in NAS >2 after the intervention resulting in an average NAS value below 4, regardless of the starting NAS value. Table 2C - Combined Assay Results Post-Intervention with Non-Response provides data generated from those patients that exhibited a change in NAS <2. While many of these patients did actually have a response to the intervention, for purposes of the study, the change in the NAS value <2 did not drop the average NAS value below 4, thereby indicating that these patients still were affected with NASH following the intervention.

Tables 2A-2C - Combined Assay Results

Table 2A - Combined Assa Results Pre-lntervention re n = 33

37 6 5.87 1241 2278 50.28 2.43 133

19 6 5.15 932 1756 64.2 12.0 48

25 6 6.68 1014 1709 6.58 4.02 88

35 6 6.51 1899 6771 58.5 6.52 109

41 6 3.87 662 504 3.66 4.26 63

11 5.5 6.19 1052 1484 9.96 5 87

15 5.5 7.59 557 1032 4.31 8.35 48

27 5 7.64 2000 1298 9.88 2.18 173

29 5 4.65 510 1217 51 .7 5.62 46

26 5 5.88 830 480 5.99 1 .54 42

8 4.5 4.18 739 886 16.3 0.44 49

12 4.5 4.15 777 709 8.12 1 .32 46

23 4.5 8.11 894 1150 4.94 0.77 96

24 4.5 4.43 791 526 25.4 0.5 27

36 4.5 10.27 803 502 22.9 2.33 42

38 4.5 4.43 690 970 6.3 9.93 44

6 4.5 4.99 751 954 23.4 1 .34 63

10 4.5 7.52 689 937 1.02 18.2 89

22 4.5 5.98 642 595 0.97 0.23 66

28 4.5 4.53 987 1366 20.7 0.90 71

32 4.5 4.43 856 1079 5.87 0.53 88

5 4 7.04 739 682 8.7 3.56 58

17 4 7.28 767 765 3.84 0.92 48

31 4 4.68 738 500 8.41 1 .31 17

14 4 5.98 653 592 3.15 1 1 .4 27

4 4 3.32 718 719 0.95 2.65 77

878 + 1196 +

18 + 18 5.5 + 6 68 + 36

5.8 +1 .5 316 1098

12 2.5 4.58 681 679 10.6 0.43 38

567 +

6.7 + 1 .9 651 + 84 141 8 + 6 3.9 + 5.4 38 + 18 p = .13 p=.02 p=.05 p=0.06 0=0.41

Table 2C - Combined Assa Results Post-Intervention with Non-Res onse

Normal Human Serum Samples - Reference Ranges

[0046] Table 3 provides data generated from twenty randomly selected normal human serum samples received from BioReclamation and includes reference ranges for some of the assays completed (ADPN, IGFBP-1 , CRP). Adiponectin in the normal human serum samples were quantitated using an eight-point calibrator curve with an analytical range from 0 - 100 ng/mL, and quality controls (0.18, 1 .8, and 18 μg/mL) diluted in DILUENT. IGFBP-1 in the normal human serum samples were quantitated using a nine-point calibrator curve with an analytical range from 0 to 500 μg/L, and quality controls (0.75, 3.50, 35.0 & 350 μg/L) diluted in Porcine Diluent. CRP in the normal human serum samples were quantitated using an eight-point calibrator curve with an analytical range from 0 - 6 μg/mL. Bio-Rad controls were used as the assay quality controls. Results from the normal serum sample testing showed that the reference range for total adiponectin was from 3.23 - 25.77 μg/mL. Results from the normal serum sample testing showed that the reference range for total IGFBP-1 was from 0.03-284 μg/L with an average concentration of 41 .9 μg/L. The serum pool prepared for testing the spike recovery of the IGFBP-1 assay had a concentration of 28.2 μg/L and the spiked serum sample tested at 34.4 μg/L. For IGFBP-1 , the measured spike concentration was 6.2 μg/L, which was 62% of the 10 μg/L spike concentration.

Table 3 - Normal Human Serum Samples - Reference Ranges

EXAMPLE 2 - Adiponectin (ADPN)

[0047] A BV-TAG Plus anti-Human Adiponectin monoclonal antibody was prepared and diluted to a working concentration of 10 μg/mL in anti-HAMA diluent. The prebound monoclonal antibody beads were prepared at a concentration of 0.25 mg/mL. The pre-bound beads were diluted to a working concentration of 0.25 mg/mL in anti-HAMA antibody diluent. Prior to analysis all control and test samples were diluted to 1 :300 in DILUENT with 0.5% BSA. Specifically, adiponectin quality controls were prepared at trilevels of 0.18, 1 .8, and 18 μg/mL in DILUENT with 0.5% BSA. Prior to adding to the microplate, the controls were diluted to 1 :300 dilution in DILUENT with 0.5% BSA. Calibrators were prepared in DILUENT with 0.5% BSA to the following levels of 0.0, 0.41 , 1 .02, 2.56, 6.4, 16, 40, and 100 ng/mL. An antibody master mix was prepared by adding equal volumes of biotinylated monoclonal antibody beads and BV-TAG Plus labeled monoclonal antibody into a microfuge tube (3.0 mL/each), prior to adding to the microplate.

[0048] With reference to the assay, prior to analysis, all test samples were diluted to 1 :300 in DILUENT with 0.5% BSA. Then 50 μί of each calibrator, control, and test sample was added to the plate wells following the plate layout. 50 μί of antibody master mix was then added to wells containing calibrator, control, and test samples per plate layout. The plate was sealed and shaken at room temperature for 60±5 minutes.

[0049] Following the assay incubation, each assay plate was washed 2X with 150 μίΛ/νβΙΙ of WD-DILUENT solution. The microplate was evaluated on one M1 MR Analyzer (Wellstat Diagnostics, LLC) using the plate protocol Adiponectin: 100 μί assay volume and 50 μί draw volume. The assay plates were analyzed and data collected. The concentrations generated from the testing were added to the

Combined Assay Results Tables 2A-2C. The normal patient serum samples were added to Table 3.

EXAMPLE 3 - Caspase-3-Cleaved K-18 Assay

[0050] The biotinylated monoclonal antibody coated beads were prepared using Dynabeads^® Streptavidin (SA) beads that were pre-bound with biotinylated monoclonal antibody M5 at 8 μg of biotinylated monoclonal antibody per mg of beads. The prebound beads were diluted to a working concentration of 0.4 mg/mL in Conjugate Buffer. The BV-TAG Plus Conjugated Anti-CK18 monoclonal antibody clone M30 (0.73 mg/mL) was diluted to a working concentration of 5 μg/mL in Conjugate Diluent.

[0051 ] For the caspase-3-cleaved K-18 assay, the K-18 standards and controls were diluted in Porcine Diluent. The Standards, Samples and Controls were tested undiluted in duplicate at RT with the 60±5 minute incubation. Then 50 μί of the Standards, Controls, and patient samples were added to the wells of the plate.

[0052] Equal volumes (3,500 μί) of capture and detector reagents were combined to form a master mix. Then 75 μί of the master mix was added to a 96- well plate. The plate was then covered with a plate cover and incubated for 60±5 minutes at room temperature with shaking. Following the assay incubation, the plates were washed 2X with 150 μΙ_ WD-DILUENT. The beads were resuspended with 100 μΐ WD-DILUENT. The plates were evaluated on one M1 MR analyzer. The raw data was analyzed using SOFTmax Pro GxP with a 5 parameter logistic curve fit. The concentrations generated from the testing were added to the Combined Assay Results Tables 2A-2C.

EXAMPLE 4 - Full-Length K-18 Assay

[0053] The CK-18 M65 Standard Curve was prepared in Porcine Diluent at concentrations of 0.0, 62.5, 125, 250, 500, 1000, and 2000 U/L from a Stock = 41 ,800 Units/L*. (*Recombinant Antigen stock concentration adjusted to 41 ,800 U/L by applying offset of 2.09to vendor-supplied concentration of 20,000 U/L). The standard curve was prepared fresh for each assay.

[0054] The CK-18 Controls were prepared in Porcine Diluent at concentrations of 150, 750 and 1500 U/L from a stock of 41 ,800 Units/L*. (*Recombinant Antigen stock concentration adjusted to 41 ,800 U/L by applying offset of 2.09 to vendor- supplied concentration of 20,000 U/L). The controls were prepared fresh for each assay.

[0055] To complete the detector reagent dilution, the stock BV-TAG Plus labeled anti-CK-18 monoclonal antibody clone M5 was diluted to 1 :279 with Conjugate Buffer to a working concentration of 2 μg/mL. For all of the ECL Assay runs, the stock biotinylated monoclonal antibody anti-CK-18 Clone M6 coated beads were diluted to a working concentration of 0.2 mg/mL. Reagent Master Mixes were made by adding the same volume of diluted anti-CK18 monoclonal antibody Clone M5 to the tubes containing the diluted biotinylated monoclonal antibody anti-CK-18 Clone M6 coated beads as noted above. 25 μΙ_ of each CK-18 M65 Standard, Control, Porcine Diluent for 0, and noted serum samples were added to their respective wells.

[0056] 75 μΙ_ of Master Mix was added to all plate wells. The plate was covered with a plate sealer and the plate was shaken for 1 hour at room temperature with setting Form 8 and Amp 6. At the end of the hour, the plate was placed on a plate magnet for bead concentration. The supernatant was discarded and the samples were washed 2X with 150 μΙ_ of WD-DILUENT solution. The washing consisted of adding WD-DILUENT, shaking for approximately two minutes to mix, placing on a magnet for approximately two minutes to concentrate, and then discarding the supernatant. 100 μΙ_ of DILUENT was added and the beads were re- suspended by shaking for approximately two minutes. The samples were detected on the one M 1 MR analyzer. The concentrations generated from the testing were added to the Combined Assay Results Tables 2A-2C.

EXAMPLE 5 - C-Reactive Protein (CRP)

[0057] The biotinylated anti-Human CRP monoclonal antibody C3 was pre- bound to Dynabeads^® M280 streptavidin (SA) beads at 10 μg per mg of beads. The pre-bound beads were then diluted to 0.3 mg/mL in anti-HAMA Antibody Diluent. The BV-TAG Plus conjugated anti-Human CRP monoclonal antibody CRP135 was diluted to 10 μg/mL in anti-HAMA Antibody Diluent. The master-mix was prepared just prior to addition to the assay plate by adding equal volumes of the biotinylated monoclonal antibody/beads (0.30 mg/mL) and BV-TAG Plus monoclonal antibody (10 μg/mL) at a final volume sufficient for addition of 50 μίΛ/νβΙΙ in each assay plate based on the plate layout.

[0058] The CRP calibrators were prepared at concentrations of 0.0, 0.0043, 0.013, 0.04, 0.123, 0.37, 1 .1 1 , and 6 μg/mL in anti-HAMA Antibody Diluent. Bio-Rad controls were used as the assay quality controls. The material was supplied as a liquid serum sample containing a target concentration of CRP. For the assay, each control was diluted 1 :50 in anti-HAMA Antibody Diluent. Each test sample was diluted 1 :50 in anti-HAMA Antibody Diluent prior to addition to the assay plates.

[0059] Each assay plate had one set of eight calibrators and one set of three controls along with test samples in duplicate. 50 μΙ-Λ/vell of each calibrator, diluted (1 :50) quality control and test sample was added to the microplate according to the plate layout. 50 μΙ-Λ/vell of master mix was added to the microplate. The plate was sealed and shaken for 60 ±5 minutes at room temperature on the Micromix 5 shaker (DPC/Siemens) set at Form 8 Amp 6. Following incubation, each assay plate was washed 2X with 125 μΙ-Λ/vell of WD-DILUENT. The washed bead complex was resuspended in 100 μΙ-Λ/vell of WD-DILUENT. Each of the four plates from the first test day were evaluated on different M 1 MR Analyzers using the following plate protocol including CRP: 100 μΙ_ assay volume; 50 μΙ_ draw volume. Each of the two plates from the second test day were evaluated on two M1 MR Analyzers used on the initial test day with the exact same assay protocol as above. The raw data was analyzed using SOFTmax Pro GxP with a 4 parameter logistic curve fit. The concentrations generated from the testing were added to the Combined Assay Results Tables 2A-2C. The concentrations generated from normal patient serum samples were added to Table 3.

EXAMPLE 6 - Insulin-Like Growth Factor Binding Protein 1 (IGFBP-1 )

[0060] The biotinylated anti-human IGFBP-1 monoclonal antibody was prebound to Dynabeads^® M280 streptavidin (SA) beads at a ratio of 10 μg biotinylated antibody per mg of beads. The beads were diluted to a working concentration of 0.15 mg/mL in anti-HAMA Antibody Diluent. The BV-TAG Plus conjugated anti-Human IGFBP-1 monoclonal antibody was diluted to a working concentration of 10 μg/mL in anti-HAMA Antibody Diluent.

[0061 ] Master mix was prepared just prior to addition to the assay plate by adding equal volumes of the biotinylated monoclonal antibody/beads (0.15 mg/mL) and BV-TAG Plus monoclonal antibody (10 μg/mL) at a final volume sufficient for addition of 50 μίΛ/νβΙΙ in each assay plate based on the plate layout. The IGFBP-1 calibrators were prepared at concentrations of 0.0, 0.10, 0.25, 1 .00, 5.00, 25.0, 100, 250, and 500 μ9/ιηΙ_ in Porcine Diluent.

[0062] The IGFBP-1 antigen (Hytest Cat. # 8IGB1 ) was supplied as a lyophilized powder (0.1 mg/tube) stored at -20°C. The IGFBP-1 powder was reconstituted with 1 ml. of Dl water to prepare a 100 μg/mL stock solution (Stock I). The stock was aliquoted and stored at -20°C. The 100 μg/mL (Stock I) was diluted to 2000 μg/L (IGFBP-1 Intermediate) in Porcine Diluent. The IGFBP-1 controls were prepared at levels of 0.75, 3.5, 35.0, and 350 μg/mL in Porcine Diluent. The IGFBP- 1 controls were prepared from the IGFBP-1 Intermediate (2000 μ9/Ι). The volume prepared was sufficient for two assay plates.

[0063] The test samples were prepared and then each test sample was added to the assay plate undiluted with 50 μΙ_ per well in duplicate. Each assay plate had one set of nine calibrators and one set of four controls along with test samples analyzed in duplicate.

[0064] 50 μΙ-Λ/vell of each of calibrator, quality control and test sample was added to the microplate according to the plate layout. 50 μΙ-Λ/vell of master mix was added to the microplate. The plate was sealed and shaken for 60±5 minutes at room temperature on the Micromix shaker set at Form 8 Amp 6. Following incubation, each assay plate was washed 2X with 150 μΙ-Λ/vell of WD- DILUENT. The washed bead complex was resuspended in 100 μΙ-Λ/vell of WD-DILUENT.

[0065] The assay plates were evaluated on the M1 MR Analyzer. The raw data was analyzed using SOFTmax Pro GxP with a 4 parameter logistic curve fit. The concentrations generated from the testing were added to the Combined Assay Results Tables 2A-2C. The concentrations generated from normal patient serum samples were added to Table 3.

EXAMPLE 7 - Statistical Analyses of Predictive Information of Biomarkers

[0066] The principle objective of the statistical analyses was to determine if biomarkers of cell injury and insulin metabolism provide unique predictive information about patient NASH status (absent, present) beyond the predictive information provided by common measures of liver disease progression and obesity. In the statistical analyses, three biomarkers of cell injury (adiponectin, full-length keratin-18, and caspase-3-cleaved keratin-18) and one biomarker of insulin metabolism (insulinlike growth factor binding protein 1 ) from the assay panel for NASH were evaluated along with alanine aminotransferase and body mass index as predictors of patient NASH status (NAS<4, NAS≥4). (See Combined Assay Results Tables 2A-2C)

[0067] Univariate analyses were conducted to evaluate if the measurement distributions of adiponectin (ADPN), caspase-3-cleaved keratin-18, full-length keratin-18, insulin-like growth factor binding protein 1 (IGFBP-1 ), alanine

aminotransferase (ALT), and body mass index (BM I) differed between the patients who had a biopsy NAS< 4 and the patients who had a biopsy NAS>4. Univariate analyses were also conducted to determine if ADPN, caspase-3-cleaved keratin-18, full-length keratin-18, IGFBP-1 , ALT, and BMI individually predict a NASH status (NAS< 4, NAS≥4).

[0068] Linear mixed models (McCulloch C.E, Searle, S.R. (2001 ),

Generalized, Linear, and Mixed Models, John Wiley & Sons Inc. New York, New York.) were used to conduct the first set of analyses. The mean of the distribution of the each outcome measured was compared between the patients who had a NAS<4 and the patients who had a NAS>4 after rescaling the measurements of the outcome to the natural logarithmic scale. Measurement rescaling was required so that the linear mixed model assumptions were satisfied. Since each patient contributed two measurements to any particular analysis, a subject-specific random effect was incorporated into the linear mixed model to account for intra-subject measurement correlation in measurement variance estimation. With regard to hypothesis testing, a F-test was used to test the null hypothesis that mean of the distribution of the outcome measurements was the same irrespective of patient NASH status (NAS<4, vs. NAS>4). A two-sided p≤0.05 rejection rule was used as the null hypothesis rejection criterion.

[0069] Binomial generalize estimating equation (GEE) models (Hardin J.W., Hilbe, J.M. (2003), Generalized Estimating Equations, Chapman & Hall/CRC, New York, New York.) were used to conduct the second set of analyses. The outcome variable for each GEE analysis was binary (i.e. 0 or 1 ). The value 1 was assigned if the patient's biopsy NAS was less than 4 (NAS<4), and the value 0 assigned if the patient's biopsy NAS was greater than or equal to 4 (NAS>4). The biomarkers

related to cell injury (ADPN, caspase-3-cleaved K-18 (C3C K-18 or M30 antigen), and full-length K-18 (K-18 or M65 antigen), the biomarker related to insulin

metabolism (IGFBP-1 ), and the ALT and BMI measurements were evaluated

individually as predictors of patient NASH status (NAS<4, vs. NAS>4). Since each patient contributed two measurements to any particular analysis, to account for intra- subject measurement correlation, the variance-covariance matrix of the GEE model was estimated via the Huber and White sandwich estimator (Huber P.J. 1967, The behavior of maximum under nonstandard conditions, In Proceedings of the Fifth

Berkeley Symposium on Mathematical Statistics and Probability, Vol. 1 , pp. 221 -233, Berkeley, CA, University of California Press.). With regard to hypothesis testing, the robust version of the Wald chi-squared test was used to test the null hypothesis that there is no association between the predictor variable values and the patient NASH status. A two-sided p≤0.05 rejection rule was utilized as the null hypothesis rejection criterion.

[0070] Partial associations between the set of predictor variables and patient

NASH status (NAS<4, vs. NAS>4) were examined via multivariate binomial GEE

regression. The dependent variable in the regression analysis was a binary indicator of patient NASH status. The value 1 was assigned if the patient's biopsy NAS was less than 4 (NAS<4), and the value 0 assigned if the patient's biopsy NAS was

greater than or equal to 4 (NAS>4). Six predictor variables were included in the

regression model. Three of the predictors were the biomarkers of cell injury (ADPN, C3C K-18 (M30 antigen), and K-18 (M65 antigen)). One predictor was the biomarker of insulin metabolism ([IGFBP-1 ). One predictor was the biomarker of liver disease progession (ALT), and the remaining predictor was BMI. In symbolic notation, the multivariate regression model is written as equation 1 :

1) logit(^) = β₀ + β_χ (ADPN) + β₂ (M30 antigen) + β₃ (M65 antigen) + β₄ (IGFBP-1)

+ ?₅(ALT) + ?₆(BMI), where logit(^ ) = log_e(^. / l - ^.) , ^. denotes the probability that subject ί has a NAS<4, β₀ is the regression intercept, and β is the regression coefficient associated with predictor j.

[0071 ] Since each patient contributed two measurements to the regression analysis, to account for intra-subject measurement correlation, the variance- covariance matrix of the GEE model was estimated via the Huber and White sandwich estimator. With regard to hypothesis testing, partial associations were tested via the robust version of the Type III Wald chi-squared test. For each predictor we tested the null hypothesis that there is no unique association (i.e., partial association) between the values of the predictor and patient NASH status after taking into account the information that is explained by the other regression predictors. A two-sided p≤0.05 rejection rule was utilized as the null hypothesis rejection criterion.

[0072] A receiver operator characteristic curve (ROC) (Huber, 1967) was generated based on the multivariate binomial GEE regression predicted probabilities for NAS<4. An optimal probability cut-point classification rule was then established to classify the liver biopsies according to NASH status (NAS<4 versus NAS≥4). The level of diagnostic agreement between the model-based liver biopsy NASH status classifications and the hepatologist's biopsy NASH status classifications was evaluated in terms of classification sensitivity and specificity, as well as in terms of the false positive and false negative error rates of the model based NASH status classifications.

[0073] The model predictions from a multivariate binomial GEE regression model in which ADPN, C3C K-18, K-18, IGFBP-1 and BM I were the predictor of NAS<4 were used in developing a nomogram, which is a simple tool that is based on a point system that allows one to predict a patient's probability of having a liver biopsy NAS<4 based on the patient's set of predictor values.

Results

[0074] Patient Demographics and Characteristics: The study-group consisted of thirty-three patients with clinically diagnosed fatty liver disease. Twenty one of the patients were female (63.6%) and 12 were male (36.4). The average age of the patients was 46.1 + 1 1 .2 years. The median age was 47 years and the minimum and the maximum ages were 25, and 66 years respectively. Thirty-one of the 33 patients were Caucasian (93.9%). The average of the liver biopsy NAS was 4.8 ± 1 .1 units. The median NAS was 4.5 units and the minimum and maximum NAS were 2.5 units and 7.0 units, respectively. Thirteen liver biopsies had a NAS<4 and 53 had a NAS≥4.

[0075] The empirical distributions of the biomarker measurements of cell injury (ADPN, C3C K-18, and K-18) and the empirical distribution for the biomarker measurements of insulin metabolism (IGFBP-1 ) are presented as box and whisker plots in FIG. 1 , along with box and whisker plots of the ALT and the BMI empirical measurement distributions. Corresponding descriptive statistics are presented in Table 4.

Table 4 - Descriptive statistics for the potential predictors of NASH status (NAS<4, NAS>4).

† the p value is for the test that the geometric means are equal for those patients with NAS<4 and those patients with NAS≥4.

[0076] The geometric mean of the distribution of the C3C K-18 measurements of cell injury differed between the patients who had a NAS<4 and those who had a NAS≥4 (p<0.001 ). Similarly, the geometric mean of the distribution of the K-18 measurements of cell injury differed between patients who had a NAS<4 and those who had a NAS≥4 (p<0.001 ). Additionally, the geometric mean of the distribution of the IGFBP-1 measurements of insulin metabolism differed between the patients who had a NAS<4 and those who a had NAS≥4 (p=0.047), and finally, the geometric mean of the distribution of the BMI measurements differed between the patients who had a NAS<4 and those who had a NAS≥4 (p=0.001 ) (Table 4).

[0077] The odds ratios from the univariate binomial GEE analyses are presented in Table 5. For each predictor variable, the odds ratio compares the odds of a NAS<4 for patients who have measurements at the 3^rd quartile of the predictor variable measurement distribution relative to the odds of a NAS<4 for patients who have measurements at the 1 ^st quartile of predictor variable measurement distribution. NASH status was significantly associated with all predictor variables other than ALT. With the exception of the association between ADPN and patient NASH status, all associations were in an inverse direction, meaning that patients who have large values for the predictor variable are less prone to have a NAS<4. The relationship between the value of the predictor variable and the predicted probability of a NAS<4 is graphically summarized in FIG. 2 for each predictor variable.

Table 5 - Odds ratios for estimating the direction and the magnitude of the association between the values of predictor and patient NASH status (NAS<4, NAS>4)

[0078] With reference to the multivariate analyses, in Table 6, the regression parameter estimates are presented for the multivariate binomial GEE regression model that is specified in symbolic notation in equation 1 of the Statistical Methods. The Type III Wald chi-squared tests are presented in Table 7 for testing the null hypothesis that there is no unique association between the values of the predictor and patient NASH status after taking into account the information that is explained by the other regression predictors. With the exception of the null hypothesis associated with ALT, all of the remaining null hypotheses were rejected based on a priori defined two-side p≤0.05 rejection rule.

Table 6 - Regression coefficient estimates and associated standard error for the multivariate binomial GEE regression model specified in equation 1 of the statistical methods

Table 7 - Type III Wald chi-squared statistics for testing the null hypothesis of no predictor versus NASH status (NAS<4, NAS>4) partial association

[0079] The adjusted odds ratios for determining the direction and the magnitude of the partial association between the values of predictor and patient NASH status (NAS<4, NAS≥4) are present in Table 8. Table 8 - Adjusted odds ratio for determining the direction and the magnitude of the association between the values of predictor and patient NASH status (NAS<4, NAS>4)

[0080] Odds ratios compare the odds of NAS<4 for a patient who has a value at the 3^rd quartile of the predictor variable measurement distribution relative to the odds of NAS<4 for a patient who has a value at the 1 ^st quartile of the predictor variable measurement distribution.

[0081 ] In FIG. 3, the individual predictor variables are ranked according to the strength of the partial association between the predictor variable and patient NASH status (NAS<4, NAS≥4). The criterion for the ranking of predictors is based on the magnitude of difference between the observed Wald chi-squared statistic (see Table 7) and the expected Wald chi-squared statistic under the null hypothesis of no partial association. Based on this ranking criterion, BM I and ADPN are the predictor variables that rank the highest in relative importance, followed by C3C K-18, IGFBP- 1 , and K-18. ALT, which was determined to be only marginally associated with NASH status (p=0.071 ), ranks last among the six predictors in terms of relative importance.

[0082] In regard to the adequacy of the multivariate binomial GEE regression model to separate the patients according to their clinically determined patient NASH status, the model C-statistic is 0.95. The C-statistic is a measure of the concordance between the model predictions and actual risk. The value of the C-statistic ranges from 0.5 to 1 . A C-statistic equal to 0.5 indicates that the model classifies the observation units into their true groups no better than a classification scheme that is based on the flip of a fair coin, while a C-statistic equal to 1 indicates the model can classify the observation units into their true groups perfectly.

[0083] The area under the receiver operator characteristic curve (ROC) that was generated based on the regression predictions for the probability of having a

NAS<4 was 0.95. The optimal cut-point probability was 0.174 for correctly

classifying the patients according to their clinically determined NASH status (NAS<4 versus NAS≥4). The sensitivity of the model-based patient NASH status

classification rule was 1 .0. Out of the 13 biopsies that had a NAS<4, all were

correctly classified. The specificity of the model-based patient NASH status

classification rule was 0.83. Out of the 53 liver biopsies in which the hepatologist gave the biopsy a NAS≥4, 9 biopsies were misclassified. The false positive error rate of the model-based patient NASH status classification rule was 0.17, while the false positive error rate was 0.

[0084] Since the variable ALT was of only marginal associated with patient

NASH status, a second multivariate binomial GEE regression analysis was

conducted using the five remaining variables (ADPN, C3C K-18 (M30 antigen), K-18 (M65 antigen), IGFBP-1 , and BMI) as the predictors of patient NASH status. In symbolical notation, the new multivariate GEE regression model is written as

equation 2:

2) logit^. ) = β₀ + β₁ (ADPN) + β₂ (M30 antigen) + β₃ (M65 antigen) + β₄ (IGFBP-1)

+ β(ΒΜΙ), where logit(^ ) = log_e(^. / l - ^.) , ^.denotes the probability that subject z^' has a NAS<4, β₀ is the regression intercept, and β. is the regression coefficient associated with predictor j.

[0085] In Table 9, the regression model parameter estimates are presented for the multivariate regression model that is specified in equation 2. In regard to the adequacy of this model to separate the patients according to their clinically

determined patient NASH status, the model C-statistic is 0.95. This value is equal to the value of the C-statistic of the model that included the set of predictors, plus ALT. Table 9 - Regression coefficient estimates and associated standard error for the multivariate binomial GEE regression model specified in equation 2 of the statistical methods

[0086] For the model in equation 2 the area under the receiver operator characteristic curve (ROC) that was generated based on the multivariate GE regression predictions for the probability of having a NAS<4 was 0.95. The optimal cut-point probability was 0.174 for correctly classifying the patients according to their clinically determined patient NASH status (NAS<4 versus NAS≥4). The sensitivity of the model-based patient NASH status classification rule was 1 .0. Out of the 13 biopsies that had a NAS<4, all were correctly classified. The specificity of the model- based patient NASH status classification rule was 0.87. Out of the 53 liver biopsies in which the hepatologist gave the biopsy a NAS≥4, 7 biopsies were misclassified. The false positive error rate of the model-based patient NASH status classification rule was 0.13, while the false positive error rate was 0.

[0087] A nomogram for predicting the probability that a patient has a NAS< 4 is presented in Table 10. The nomogram was developed based on the model parameter estimates of the model that is specified in equation 2. The nomogram in Table 10 is based on a points system. To use the nomogram you simply add up the points associated with each predictor variable (ADPN, C3C K-18, K-18, IGFBP-1 , and BMI) to determine the patient's total number of points. In the lower right portion of the nomogram, two columns of numbers are underlined. In the left column of the underlined area, the total number of points are listed, and in right column of the underlined area, the corresponding predicted probabilities for NAS <4 are listed. It should be noted that since no external validation of the model in equation 2 was performed, at this time the utility of the nomogram is specific to this particular set of patients.

Table 10 - Nomogram for predicting a patient's probability of having a NAS<4 based on the patient's ADPN, C3C K-18, K-18, 1 GFBP-1 and BMI values.

EXAMPLE 8 - Changes in NAS Value

[0088] An alternative framework for evaluating the NASH clinical trial data shown in the Combined Assay Results Tables 2A-2C is to correlate the change in NAS with changes in serum analytes, regardless of the starting or ending NAS value. Table 1 1 below divides patients into three groups according to the change in NAS: (1 ) Reduction in NAS <1 , (2) Reduction in NAS≥1 , and (3) Reduction in NAS≥2, with the absolute and percent change (from baseline) of each measured serum biomarker. Table 1 1 indicates that downward percent changes in caspase-3-cleaved K-18, full-length K-18, IGFBP-1 , CRP and ALT increase as a function of the downward change in NAS value.

Table 11 - Changes in NAS Value

^* C3C K-18 = caspase-3-cleaved K-18

[0089] The data generated by the assays completed for each of the different biomarkers was broken down further in Tables 12-17 based on the same patient serum samples from Example 1 . In particular, each of Tables 12-17 provides results following the change in NAS values for each of the biomarkers assayed on the panel: Adiponectin (Table 12); caspase-3-cleaved keratin-18 (Table 13); full-length keratin-18 (Table 14); IGFBP-1 (Table 15); C-Reactive Protein (Table 16); and ALT (Table 17). NAS values have been left in the tables to corroborate the collected data from Tables 2A-2C and Table 1 1 .

[0090] With regard to Tables 12-17, the data generated from these analyses demonstrate that the assay panel of the disclosure is useful not only for assigning a likely NAS value at a single time point, but can be used to detect changes in liver histology over time, regardless of whether the final result is an NAS value <4. For example, a patient with an initial NAS of 6.5 and a later (e.g., 1 year) value of 4.5 can be identified by the panel of the disclosure to provide guidance for continued treatment, as the trajectory of liver disease is improving even though the patient still has a clinical diagnosis of NASH (NAS>4). Use of the diagnostic panel to detect changes in liver histology (correlating with changes in NAS numerical values) without requiring serial liver biopsies provides an important tool for conducting efficient clinical studies to identify effective therapies, or disprove ineffective therapies, and for providing information to encourage patients to persist in lifestyle changes with beneficial effects on liver status.

Table 12 - Chan es in NAS Values and Serum Adi onectin in NASH Patients

Table 13 - Chan es in NAS Values and Serum C3C K-18 in NASH Patients

Table 14 - Chan es in NAS Values and Serum K-18 in NASH Patients

Table 15 - Chan es in NAS Values and Serum IGFBP-1 in NASH Patients

Table 16 - Chan es in NAS Values and Serum CRP in NASH Patients

Table 17 - Chan es in NAS Values and Serum ALT in NASH Patients

Claims

CLAIMS What is claimed is:

1 . A method for screening a subject to predict the likelihood of significantly active histological non-alcoholic steatohepatitis (NASH) in a subject, comprising measuring the level of at least four biomarkers selected from the group consisting of adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), , insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT), in a sample derived from the subject, and analyzing the levels of the at least four biomarkers against reference values for those biomarkers.

2. The method of claim 1 , wherein the levels of adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), and insulin-like growth factor binding protein 1 (IGFBP-1 ) are measured.

3. The method of claim 1 , wherein the sample comprises serum, blood, plasma, urine, saliva, cerebrospinal fluid, tissue, or tissue extract.

4. The method of claim 1 , wherein the level of one or more of the biomarkers is measured using electrochemiluminescence (ECL).

5. The method of claim 1 , wherein the analyzing the levels of the biomarkers against reference values for those biomarkers comprises performing multivariable logistic regression analysis.

6. The method of claim 1 , wherein the analysis further comprises analyzing Body Mass Index of the subject.

7. The method of claim 1 , wherein the subject is a human subject.

8. A method for screening a subject to identify the presence or absence of nonalcoholic steatohepatitis (NASH) in a subject, comprising measuring the level of at least four biomarkers selected from the group consisting of adiponectin (ADPN), full- length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT), in a sample derived from the subject, and analyzing the level of the at least four biomarkers against reference values for those biomarkers.

9. The method of claim 8, wherein the levels of adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), and insulin-like growth factor binding protein 1 (IGFBP-1 ) are measured.

10. The method of claim 8, wherein the sample comprises serum, blood, plasma, urine, saliva, cerebrospinal fluid, tissue, or tissue extract.

1 1 . The method of claim 8, wherein the level of one or more of the biomarkers is measured using electrochemiluminescence (ECL).

12. The method of claim 8, wherein the analyzing the levels of the biomarkers against reference values for those biomarkers comprises performing multivariable logistic regression analysis.

13. The method of claim 8, wherein the analysis further comprises analyzing Body Mass Index of the subject.

14. The method of claim 8, wherein the subject is a human subject.

15. A method for monitoring the progression of non-alcoholic fatty liver disease (NAFLD) to non-alcoholic steatohepatitis (NASH) or improvement of NASH-like conditions to NAFLD-like conditions over time of non-alcoholic steatohepatitis (NASH) in a subject, comprising measuring the level of at least four biomarkers selected from the group consisting of adiponectin (ADPN), full-length keratin-18 (K- 18), caspase-3-cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulinlike growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT), in samples derived from the subject at two or more time points, and analyzing the level of the at least four biomarkers at a second or later point in time against the values for those biomarkers at an earlier point in time

16. The method of claim 15, wherein the levels of full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT) are measured.

17. The method of claim 15, wherein the levels of adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), and insulin-like growth factor binding protein 1 (IGFBP-1 ) are measured.

18. The method of claim 15, wherein the sample comprises serum, blood, plasma, urine, saliva, cerebrospinal fluid, tissue, or tissue extract.

19. The method of claim 15, wherein the level of one or more of the biomarkers is measured using electrochemiluminescence (ECL).

20. The method of claim 15, wherein the analyzing the levels of the biomarkers against reference values for those biomarkers comprises performing multivariable logistic regression analysis.

21 . The method of claim 15, wherein the analysis further comprises analyzing Body Mass Index of the subject.

22. The method of claim 15, wherein the subject is a human subject.

23. A method for monitoring the efficacy of a therapy for treatment of nonalcoholic steatohepatitis (NASH) in a subject, comprising measuring the level of at least four biomarkers selected from the group consisting of adiponectin (ADPN), full- length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT), in samples derived from the subject before and after the subject undergoes the therapy, and analyzing the level of the at least four biomarkers after therapy against the value for those biomarkers before therapy.

24. The method of claim 23, wherein the levels of full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT) are measured.

25. The method of claim 23, wherein the levels of adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), and insulin-like growth factor binding protein 1 (IGFBP-1 ) are measured.

26. The method of claim 23, wherein the sample comprises serum, blood, plasma, urine, saliva, cerebrospinal fluid, tissue, or tissue extract.

27. The method of claim 23, wherein the level of one or more of the biomarkers is measured using electrochemiluminescence (ECL).

28. The method of claim 23, wherein the analyzing the levels of the biomarkers against reference values for those biomarkers comprises performing multivariable logistic regression analysis.

29. The method of claim 23, wherein the analysis further comprises analyzing Body Mass Index of the subject.

30. The method of claim 23, wherein the subject is a human subject.

31 . A kit comprising a package, the package containing at least four agents for measuring the level of at least four biomarkers of interest, wherein the at least four biomarkers are selected from the group consisting of adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), C-Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT).

32. The kit of claim 31 , wherein the package contains agents for measuring the levels of adiponectin (ADPN), full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), and insulin-like growth factor binding protein 1 (IGFBP-1 ).

33. The kit of claim 31 , wherein the package contains agents for measuring the levels of full-length keratin-18 (K-18), caspase-3-cleaved keratin-18 (C3C K-18), C- Reactive Protein (CRP), insulin-like growth factor binding protein 1 (IGFBP-1 ), and alanine aminotransferase (ALT).

34. The kit of claim 31 , wherein the at least four agents comprise at least one of an antibody of adiponectin (ADPN), an antibody of full-length keratin-18 (K-18), an antibody of caspase-3-cleaved keratin-18 (C3C K-18), an antibody of C-Reactive Protein (CRP), an antibody of insulin-like growth factor binding protein 1 (IGFBP-1 ), and an antibody of alanine aminotransferase (ALT).

35. The kit of claim 31 , wherein the antibody of full-length keratin-18 is

monoclonal antibody M65.

36. The kit of claim 31 , wherein the antibody of caspase-3-cleaved keratin-18 is monoclonal antibody M30.

37. The kit of claim 31 , further comprising one or more control reference samples.

38. The kit of claim 31 , wherein the package further contains instructions for using the kit.