WO2026006451A1 - Compositions and methods for diagnosis and treatment of liver disease, including non-alcoholic fatty liver disease (nafld) - Google Patents

Abstract

Compositions and Methods for detecting a signature biomarker panel of VOCs associated with liver disease are disclosed for diagnosis, management and treatment of liver diseases, particularly NAFLD.

Description

Compositions and Methods for Diagnosis and Treatment of Liver Disease, Including Non-alcoholic Fatty Liver Disease (NAFLD)

By

Audrey Ragan Odom John Amalia Zoraida Berna Perez Jennifer Panganiban

Cross-Reference to Related Application

This application claims priority to US Provisional Application No. 63/663,737 filed June 25, 2024, which is incorporated by reference as though set forth in full.

Field of the Invention

The present invention relates to the fields of non-invasive diagnostic breath tests and the identification of a panel of volatile compounds useful as biomarkers for rapid and accurate diagnosis of subjects having liver disease, e.g., NAFLD.

Background of the Invention

Several publications and patent documents are cited throughout the specification in order to describe the state of the art to which this invention pertains. Each of these citations is incorporated herein by reference as though set forth in full.

Metabolic-dysfunction Associated Steatotic Liver Disease (MASLD), formerly known non-alcoholic fatty liver disease (NAFLD) is the most common chronic liver disease of childhood. NAFLD is a significant public health problem, because of the dramatic increase in the prevalence of obesity over the past decade, as obesity (along with insulin resistance and type II diabetes) is a major risk factor for the pathogenesis of the disease. Children with NAFLD often do not present with symptoms; moreover, there are currently no reliable noninvasive methods to screen for NAFLD. The American Academy of Pediatrics recommends serum alanine aminotransferase (ALT) measurement as a screen for NAFLD in obese children. In both adults and children, however, liver enzymes perform poorly for diagnosing NAFLD — nearly two- thirds of patients with NAFLD will have normal levels of ALT. As a result, pediatric NAFLD is likely unrecognized in a significant proportion of the at-risk population. Because NAFLD may progress during childhood, and interventions (e.g., lifestyle modification and weight reduction) can effectively reverse the course of early disease, it is important to have tools that can easily diagnose NAFLD in children at risk.

Summary of the Invention

In accordance with the present invention, specific breath volatile organic compounds (VOCs) and VOC signatures associated with liver disease, e.g., pediatric MASLD, using gas chromatography- mass spectrometry (GS-MS) have been identified. Relationships between breath VOC's and the intestinal microbiota and MASLD in children have also been analyzed.

In one embodiment, a method for diagnosing or monitoring a subject with non-alcoholic fatty liver disease (NAFLD) is disclosed. An exemplary method entails analyzing a sample of exhaled breath or condensate breath obtained from the subject for levels in a biomarker panel of volatile organic compounds (VOCs) comprising at least 5, 6, 7, 8, 9, 10, 11, 12, or 13 of isoprene, p-xylene, eucalyptol, o-xylene, toluene, m-ethyL, \|/-cumene, D-limonene, phenol, decane, undecane, thiophene, 3-methyl-, nonane, and 2,4-dimethyl-l -heptene, which, when compared to levels observed in healthy control subjects, indicates the subject has NAFLD. The method can also comprise assessment of ALT, AST, GGT, total bilirubin, alkaline phosphatase, total protein, albumin, cholesterol, TAG, HDL and low-density lipoprotein [LDL] and glycated hemoglobin (HbAlc). In yet another aspect, the method comprises collecting a stool sample for metagenomic and transcriptomic analyses of fecal bacteria present in the gut microbiome of pediatric subject suffering from liver disease.

Analysis of the VOC can comprise the use of at least one technique selected from photo ionization detection, flame ionization detection, gas chromatography — mass spectrometry (GC- MS), proton transfer reaction mass spectrometry (PTR-MS), colorimetry, infrared spectroscopy, electrochemical fuel cell sensing, semiconductor gas sensing, quartz tuning fork (QTF) sensors, electronic noses and combinations thereof. In certain approaches, analysis of the VOCs is conducted using a portable, hand-held breathalyzer device. In other approaches, a series of volatile organic compounds (VOCs) comprising each of isoprene, p-xylcnc, eucalyptol, o-xylcnc, toluene, m-ethyl-, \|/-cumene, D-limonene, phenol, decane, undecane, thiophene, 3-methyl-, nonane, and 2, 4-dimethyl- 1 -heptene are detected in the sample of exhaled breath or condensate breath obtained from the subject and a concentration for each of the VOC determined, thereby providing a cumulative abundance based on the concentrations for the VOCs, wherein the cumulative abundance and the concentration of the VOCs indicates the presence of NAFLD. The subject can be a pediatric, adolescent or adult subject. The sample can optionally be condensed or concentrated before analysis.

In certain embodiments, a) the isoprene is increased when compared to the healthy control; b) the decane is decreased when compared to the healthy control; c) the toluene, m- ethyl- is decreased when compared to the healthy control; d) the \|/-cumene is decreased when compared to the healthy control; e) the p-xylene is decreased when compared to the healthy control; f) the 2,4-dimethyl-l -heptene is decreased when compared to the healthy control;g) the limonene is decreased when compared to the healthy control; h) the phenol is decreased when compared to the healthy control; i) the o-xylene is decreased when compared to the healthy control; j) the eucalyptol is decreased when compared to the healthy control; k) the undecane is increased when compared to the healthy control; 1) the thiophene, 3 -methyl- is increased when compared to the healthy control; or m) the nonane is increased when compared to the healthy control. In certain embodiments, 2, 3, 4, 5, 6, 7, 8 , 9, 10, 11, 12, or 13 of a)-m) are present.

In yet another aspect, the methods described above can further comprise administering to the subject a pharmaceutical composition comprising a therapeutically effective amount of at least one compound effective for alleviating symptoms of NAFLD.

Brief Description of the Drawings

Figures 1A - IB. Fig. 1A shows breath collection followed by GCxGC MS analysis. Raw data files were processed and analyzed using ChromeSpace software (version 2.0.1, SepSolve, UK), using standard approaches. Fig. IB shows work flow of stool sample collection and laboratory analysis. Next generation sequence consisting of metagenomics and metastranscriptomics were applied.

Figure 2A - 2C. Breathprinting reveals candidate breath VOC biomarkers of pediatric NAFLD. (Fig. 2A) Principal component analysis (PCA) of breath volatile profiles in obese children with and without NAFLD. While control samples cluster tightly and arc highly distinct from NAFLD samples, the larger variability in NAFLD samples likely reflects the variability in clinical severity. (Fig. 2B) Heatmap demonstrating VOC abundances in healthy (obese) children vs. obese children with NAFLD. Each row represents a single patient and the abundances of seven candidate biomarkers are shown. (Fig. 2C) Candidate breath biomarker of pediatric NAFLD. Abundance of a candidate biomarkcr identified from breath volatiles analysis of children with and without NAFLD. Unpublished.

Figures 3A - 3E. (Fig. 3A) Breath biomarkers for NAFLD diagnosed patients compared to healthy controls. (Fig. 3B) Breath biomarkers that discriminate NAFLD diagnosed by liver biopsy from those with high ALT. (Fig. 3C) Table listing of features with VOC names found to be discriminatory between MASLD and healthy controls and between Biopsy proven MASLD and those with high ALT (Figs 3A and 3B). (Fig. 3D) Heat map visualizing abundance of 10 unique breath compounds (represented as z-scores) in obese healthy controls and MASLD cohorts. Breath analysis revealed that obese children with MASLD have dramatically different breath VOC profiles than their healthy obese counterparts. (Fig. 3E) Breath biomarkers that discriminate NAFLD diagnosed by liver biopsy from those with high ALT. Of the 10 discriminatory molecules identified the majority were elevated in obese healthy controls.

Figures 4A - 4C. (Fig. 4A) Correlation coefficient between levels of VOCs and ALT(>60) MASLD patients and lab results. (Fig. 4B) Correlation coefficient between levels of VOCs and ALT (>80) MASLD patients. (Fig. 4C) Graph showing correlation between confirmed and suspected NAFLD.

Figures 5A - 5C. (Fig. 5A) Random Forest of all NAFLD samples vs. Healthy controls. (Fig. 5B) Random Forest NAFLD Confirmed (Biopsy) vs NAFLD (high ALT). (Fig. 5C) XG Boost Model NAFLD Confirmed (Biopsy) vs NAFLD (high ALT).

Figure 6. Example of outputs of ongoing computational analyses of metagenomics and mctatranscriptomics data.

Figure 7A-7B. Scatter plots for each compound identified in Table 2 showing the peak intensities for each compound in healthy controls and MASLD samples.

Detailed Description of the Invention

Knowledge of the risk factors, pathophysiology, management, and prognosis of pediatric nonalcoholic fatty liver disease (NAFLD) has evolved since the disease was first identified in 1983. The roles of insulin resistance and obesity as risk factors in the development of NAFLD are now well established. Unfortunately, the prevalence of children who are overweight, obese, and severely obese is rising in the United States, contributing to the rapid ascent of NAFLD as the most common form of chronic liver disease in children (1). Practice guidelines in the US recommend screening for NAFLD between ages 9 and 11 years for all obese and overweight children with additional risk factors. The recommended screening test is alanine aminotransferase (ALT) using sex-specific upper limits of normal levels (females, 22 U/L; males, 26 U/L). Guidelines recommend against using routine ultrasound as a screening test due to inadequate sensitivity and specificity (2). However, this practice is controversial, as children with evidence of NAFLD on ultrasound may have normal ALT levels and advanced fibrosis can be present when liver enzyme levels are normal. While liver biopsy remains the gold standard for NAFLD diagnosis, this is an invasive test that carries some risk and is inappropriate for screening.

In recent years, we and other have extensively explored the use of volatile organic compounds (VOCs) in exhaled breath as a non-invasive lens to detect perturbated metabolic pathways in children and as a non-invasive approach to diagnose pediatric disease. In data presented herein, we employed state-of-the-art mass spectrometry to identify candidate volatile organic compounds (VOCs) present in the breath from children with NAFLD which distinguish them from normal, healthy subjects with high sensitivity and specificity. Our data show that children with NAFLD have markedly different “breathprints” than their healthy, obese counterparts. See Figures 1A and IB. The state-of-the-art multidimensional gas-chromatography mass-spectrometry (GCxGC-BenchToF) employed coupled to thermal sorption to analyze volatiles from biological specimens includes two separate columns to further separate mixtures that co-elute on the first column, providing an order-of-magnitude increase in compound resolution compared to standard GC. This provides the sensitivity, resolution, and mass accuracy required to accelerate the studies described.

The present invention includes methods for diagnosing or monitoring a subject with MASLD (a.k.a., NAFLD). In some embodiments, the methods comprise analyzing a sample of exhaled breath or condensate breath obtained from the subject for the series of NAFLD- associated VOCs, described herein, wherein the type and concentration of the VOCs indicates the presence of NAFLD. In certain embodiments, the sample is analyzed for VOCs comprising isoprene, p-xylene, eucalyptol o-xylene, toluene, m-ethyl-, \|/-cumene, D-limonene, phenol, decane, undecane, thiophene, 3-methyl-, nonane, and 2,4-dimethyl-l -heptene. The VOCs identified on the heat map shown in Figure 3 A -3E and Tabic 2. The different levels of each NAFLD-associated in healthy controls and MASLD patients are provided in Figure 7A-7B.

As noted above, methods for analyzing a sample of exhaled breath or condensate breath obtained from a subject are also described. Exemplary methods can include the use of at least one technique selected from the group consisting of photo ionization detection, flame ionization detection, gas chromatography — mass spectrometry (GC-MS), proton transfer reaction mass spectrometry (PTR-MS), colorimetry, infrared spectroscopy, electrochemical fuel cell sensing, semiconductor gas sensing, quartz tuning fork (QTF) sensors, electronic noses and combinations thereof. These methods are described in more detail herein.

In certain embodiments, the method comprises use of an electronic nose, or any microarray capable of sensing multiple volatile signatures, particularly one calibrated to the detection of volatile organic compounds (Chang et al, Science Reports, 6(2016): 23970). In further embodiments, the method comprises use of a portable wireless volatile organic compound monitoring device that employs quartz tuning fork (QTF) sensors (Deng et al. Sensors 2016, 16(12), 2060). These techniques involve adsorption of VOCs onto modified (coated) QTFs which alters their resonance frequency and enables quantification of VOC concentration.

The analysis described for the methods herein could also include use of a portable device comprising a sample collection and pre-concentration unit, a sample separation column, and a sensitive, selective and fast sensor (IEEE Sens J. 2013 May; 13(5): 1748-1755).

In some embodiments, the method further comprises the use of solid-phase microextraction fibers to extract and concentrate volatile chemicals in exhaled breath for further analysis. For example, various methods can include the use of micro -extraction fibers alongside GC-MS to detect VOCs in exhaled breath of human patients (Gao et al, J. Breath Res. 10:2 (2016) 027102). In further embodiments, the method comprises use of PTR mass spectrometry to detect VOCs in collected breath of subjects (O'Hara et al., J. Breath Res. 10:4 (2016)). PTR mass spectrometry uses gas phase hydronium (H 30+) ions to ionize trace VOCs in an air sample in order to detect and identify them using mass spectrometry. In still other embodiments, the method comprises use of fast gas chromatography — flame ionization detection (Fast-GC-FID) which is known in the ait to detect VOCs in ambient air samples (Jones et al, Atmos. Meas. Tech, 7, 1259-1275, 2014). Briefly, this method involves separating volatile chemicals on a gas column and using a hydrogen flame to oxidize them for detection.

In various embodiments, the analysis of the NAFLD VOC biomarkers listed above is conducted using a portable, hand-held breathalyzer or electronic nose device. The technique or device used for analysis can also include a display or be in communication with a further device (e.g., monitor or printer) that displays the results of the analysis.

Methods for diagnosing or monitoring a subject with NAFLD comprise analyzing a sample of exhaled breath obtained from the subject for a series of volatile organic compounds (VOCs) comprising: at least one, at least two, at least three, at least 4, at least 5, at least 6, at least 7, at least 8, at least 10, at least 11, at least 12, at least 13, or all 14 of the VOCs shown in Figure 3 or Table 2; and determining a concentration for each of the VOCs; and calculating a cumulative abundance based on the concentrations for the VOCs. VOCs abundance levels correlated with previously identified levels observed in NAFLD patients being indicative of NAFLD in the subject. The different levels of each NAFLD-associated in healthy controls and NAFLD patients are provided in Figure 7A-7B. Other parameters to be assessed include those described in Example 2.

In certain embodiments, concentrations of the series of VOCs in a subject are compared to concentrations in a healthy individual. In other embodiments, concentrations of the series of VOCs in a subject are compared to concentrations in subjects previously diagnosed with NAFLD. In certain embodiments, the NAFLD patient has an increased level of isoprene when compared to a healthy individual. In certain embodiments, the NAFLD patient has a decreased level of decane when compared to a healthy individual. In certain embodiments, the NAFLD patient has a decreased level of toluene, m-ethyl- when compared to a healthy individual. In certain embodiments, the NAFLD patient has a decreased level of \|/-cumene when compared to a healthy individual. In certain embodiments, the NAFLD patient has a decreased level of p- xylene when compared to a healthy individual. In certain embodiments, the NAFLD patient has a decreased level of 2,4-dimethyl- 1 -heptene when compared to a healthy individual. In certain embodiments, the NAFLD patient has a decreased level of limonene when compared to a healthy individual. In certain embodiments, the NAFLD patient has a decreased level of phenol when compared to a healthy individual. In certain embodiments, the NAFLD patient has a decreased level of o-xylene when compared to a healthy individual. In certain embodiments, the NAFLD patient has a decreased level of eucalyptol when compared to a healthy individual. Tn certain embodiments, the NAFLD patient has an increased level of undecane when compared to a healthy individual. In certain embodiments, the NAFLD patient has an increased level of Thiophene, 3-methyl- when compared to a healthy individual. In certain embodiments, the NAFLD patient has an increased level of Nonane when compared to a healthy individual. In certain embodiments, the NAFLD patient has an increased level of the unknown 1 compound identified in Table 2 when compared to a healthy individual.

In additional embodiments, the methods of diagnosing or monitoring a NAFLD can comprise the combination of any of the methods described herein.

Certain embodiments can also include treatment of subjects identified as having or being at risk for liver disease, e.g., NAFLD. In various embodiments of diagnosis and monitoring of subjects using the methods described herein, samples from the subjects can be exhaled breath or condensate breath. The concentration of the series of VOCs in breath or breath condensate aids in the diagnosis or the monitoring of NAFLD. In some embodiments, the concentration of the series of VOCs in the breath or breath condensate is compared to levels of the series of VOCs in the breath of other subjects determined to be healthy, optionally obese subjects who do not suffer from NAFLD, (e.g., baseline VOC concentrations).

In various embodiments, the subject exhibits one or more characteristic symptoms or etiology known to be associated with a liver disease, such as NAFLD. These include, but are not limited to, having one or more of being overweight or obese, prediabetes (insulin resistance), Type 2 diabetes, high cholesterol, high triglycerides, high blood pressure, rapid weight loss, poor diet, gastric bypass surgery, bowel disease and use of certain medicines, such as calcium channel blockers and some cancer medicines.

The methods of the present invention can further comprise administering to the subject a pharmaceutical composition comprising a therapeutically effective amount of at least one compound effective for treatment of NAFLD. For example, the compound effective against this disease can include at least one compound selected from Pioglitazone (PIO), GLP-1 receptor agonists (GLP-lRAs) and SGLT2 inhibitors (SGLT2i), however, the preferred treatment is weight loss via exercise and a healthy diet.

The following materials and methods are provided to facilitate the practice of the present invention. Sample collection and volatile analysis:

Sample collection: Pediatric breath collection will be performed as per our previously published methods (19, 21). In brief, exhaled breath will be voluntarily exhaled by subjects through a one-way valve into a 3L SamplePro Flexfilm sample bag (SKC). A total of IL of breath will be collected and breath will be discharged into a labeled concentrating (sorbent) trap for analysis with a GCxGC BenToF (SepSolve) owned and maintained exclusively in our laboratory (22). Samples will be processed using the same analytical conditions and instrument settings, to reduce analytical and technical variability. In addition, internal standards (added directly prior to analysis with inert carrier gas) will be used to control for minor technical variations between runs.

Data processing: Raw data files will be processed and analyzed using ChromSpace software (version 2.0.1, SepSolve, UK), using standard approaches. Files will be aligned and the background from the raw BenchTOF data file will be removed, and the Dynamic Background Compensate (DBC) of 0.2sec peak width and noise factor 6.9 for typical GCxGC data will be applied. DBC files will then be integrated, and GCxGC-MS data will be normalized using internal standards from the sample measured on the day of analysis (i.e., analysis of a mixture of known compounds at a known concentration) to overcome for “day of analysis” effects.

Targeted analysis (NAFLD vs healthy controls): Our previous NAFLD pilot work detected seven NAFLD-associated VOCs (Fig. 6). Therefore, we will specifically query whether these compounds are associated-NAFLD in this new, distinct cohort. The effectiveness of the individual compounds and combinations of candidate compounds to discriminate between NAFLD samples and healthy controls will be evaluated by two common classifiers [Support vector machine (SVM) and k-nearest neighbor (k-NN)], as described below.

Unbiased analysis (NAFLD vs healthy controls): To identify other potential reproducible biomarkers in this new cohort that can deliver the best classification performance, we will use feature selection methods. In brief, univariate methods will be used as a first step to obtain a rough ranking of potential important features before applying more sophisticated techniques. We will use volcano plots that display the fold change (ratio of levels between two groups) and the statistical significance (t-test) for each feature. Multiple testing correction will be used when performing the t-test on multiple features. To evaluate whether the data can support a classification (i.e., if it is possible to correctly classify healthy and NAFLD patients using chosen candidate biomarkcr), wc will use support vector machine (SVM) (24) and k-ncarcst neighbor (k-NN) (25) machine learning algorithms. Leave-one-out cross-validation will be used to train and test classifiers. The full datasets will be partitioned into N pairs of a training set (of N-l data samples) and a test set (of the remaining 1 data sample). This process will be repeated N times to cover each data point in the data set once. A similar approach will be used to identify potential differences between NAFLD with high ALT and NAFLD biopsy-proven breath samples.

The following examples are provided to illustrate certain embodiments of the invention. They are not intended to limit the invention in any way.

Example 1

VOC Panel for the Diagnosis and Management of NAFLD

Human breath is a promising non-invasive sample-type that contains hundreds of VOCs, many of which are reproducibly associated with infectious and non-infectious diseases (17, 18). Previous studies have examined breath samples for biomarkers of liver diseases of different etiologies and various severity (9, 10). The majority of these focused on more advanced conditions or on distinguishing alcohol-linked conditions. These studies lack independently validated results which confirm the presence/absence of the biomarkers. Compounds linked to advanced liver conditions have included dimethyl sulfide (11-13), ethanol (14), acetaldehyde (15), and short chain alkanes such as ethane and pentane (16).

We hypothesized that children with NAFLD would have distinct breath VOC profiles and performed a pilot study to characterize breath volatiles by thermal desorption/GC-MS in obese children ages 4-17 with (n=18) NAFLD and elevated ALT levels, and without (n=18) NAFLD (Table 1). Children with NAFLD were diagnosed by a pediatric gastroenterologist based on ALT levels (ALT >52 U/L for boys and >44 U/L for girls). We find that obese children with NALFD have dramatically different breath VOC profiles than their healthy obese counterparts [as visualized by principal components analysis (PCA), Fig. 2A. A number of specific breath VOCs correlated strongly with the presence or absence of liver injury and NAFLD (see heatmap, Fig. 2B, and candidate biomarker, Fig. 2C). MASLD MASLD Biopsy

Variable High ALT Confirmed p-value ²

N = 14¹ N = 10¹

Demographics

Age 14.50(12.00, 16.75) 16.00(11.00, 13.00(10.00, 0.2

16.75) 14.25)

Sex 0.006

Female 6 (43%) 1 (10%) 14 (70%)

Male 8(57%) 9(90%) 6(30%)

Race <0.001

Asian 1 (7.1%) 1 (10%) 0(0%)

Black or African American 1 (7.1%) 0(0%) 13(65%)

Unknown/Not Reported/Other 6(43%) 5(50%) 1 (5.0%)

White American 6(43%) 4(40%) 6(30%)

Ethnicity 0.026

Hispanic or Latinx 6(43%) 5(50%) 2(10%)

Not Hispanic or LatinX 8(57%) 5(50%) 18(90%)

Lab Values

BMI 34.0(29.2,38.3) 34.4(29.2,39.1) 36.6(31.9,40.8) 0.6

Glucose (mg/dL) 99 (97, 107) 100(98, 109) 89 (81,92) <0.001

(Missing) 5 2 4

ALT or SGPT (units/L) 91 (63, 142) 161 (132, 219) 16 (14, 17) <0.001

AST or SGOT (units/L) 52(46,94) 83(72,91) 22(16,23) <0.001

GGT (U/L) 42 (30,48) 59(48,75) 13 (13, 13) 0.053

(Missing) 2 0 19

Total bilirubin (mg/dL) 0.45 (0.33,0.50) 0.70(0.50,0.90) 0.40(0.25,0.45) 0.017

(Missing) 0 1 5

We found at least 10 discriminatory molecules, the majority of which were elevated in healthy obese controls. These include without limitation, isoprene, p-xylene, eucalyptol o-xylene, toluene, methy-, \|/-cumene, D-limonene, phenol, decane, undecane, thiophene, 3-methyl-, nonane, and 2,4 di-methyl- 1 -heptene. See Figures 3A -3E and Table 2.

Next, Random Forest was applied to the dataset to evaluate the predictive power of the VOCs identified. The results demonstrate that VOCs can predict MASLD with 73% accuracy with 83% sensitivity, and or 65% specificity. See Figure 5. Example 2

Breath VOC changes in NAFLD and Association with Underlying Metabolic Changes.

Obesity is a major risk factor for insulin resistance and development of Type 2 diabetes. Type 2 diabetes mellitus-induced hyperglycemia has an effect on the lipid profile and release of oxidative stress markers and inflammatory mediators in patients with non-alcoholic fatty liver disease, which may in turn accelerate liver disease (26). Insulin resistance leads to hepatocyte fat deposition because the resulting lipolysis and hypcrinsulincmia combine to increase lipid deposition, intracellular fatty acids, and oxidative stress in hepatocytes (27). Free fatty acids and oxidative stress can lead to increased levels of alkenes (28, 29) and aldehydes (30) in the breath of patients. Thus, we hypothesize that some NAFLD volatile biomarkers are directly correlated to metabolic derangements commonly present in these patients.

As mentioned above, ALT value is usually used as a marker of hepatic inflammation and liver injury in patients with NAFLD. In addition, blood levels of routine markers used for diagnosis of liver disease can be correlated with the identified breath biomarkers. Clinical data will include routine liver tests panel (ALT, AST, GGT, total bilirubin, alkaline phosphatase, total protein, albumin), lipid panel (cholesterol, TAG, HDL and low-density lipoprotein [LDL]) and glycated hemoglobin (HbAlc).

Children with and without NAFLD be enrolled in further studies. We will use unbiased thermal desorption-mass spectrometry and machine learning techniques to characterize the exhaled breath volatiles from overweight/obese children with: 1) NAFLD diagnosed by ALT levels > 2ULN; 2) normal ALT levels without NAFLD; and 3) NAFLD diagnosed by imaging (demonstrating hepatic steatosis) with normal ALT and recent (<6 months) US-guided liver biopsy. In addition, we will determine if breath biomarkers are specific to NAFLD by evaluating the breath profile of children with other liver diseases.

To uncover specific breath volatiles that correlate with a clinical feature, we will use Pearson’s correlation analysis. Those correlations identified with a specified feature, will be entered into machine learning regression models to accurately predict levels of a specific clinical feature of NAFLD patients based on the levels of candidate breath NAFLD biomarker/s. Leave- one-out cross-validation will be used to train and test the regression model. Depending on the clinical feature, clinical features will be stratified as “normal” or “elevated” or used as continuous variables and correlated with breath levels. See Figures 4A -4C.

We can also investigate how breath volatiles are impacted by the gut microbiome, which is known to be altered in NAFLD (33-35). For this reason, in independent studies, we will also collect stool samples on all patients and will use standard 16S sequencing to nominate candidate taxa and study the association with VOC production. See Figure 6. Metagenomic and transcriptomic analyses of fecal microbes will identify specific taxa and gcnc/gcnc clusters associated with the VOC panel described herein, thereby further characterizing the MASLD/NAFLD in the children being assessed. These studies will serve to further elucidate the metabolic origin and biological function/s of NAFLD -associated breath volatiles when using the NAFLD “breathalyzer” described herein.

References

1. Shah J, Okubote T, Alkhouri N. Overview of Updated Practice Guidelines for Pediatric Nonalcoholic Fatty Liver Disease. Gastroenterol Hepatol (N Y). 2018;14(7):407-14. PubMed PMID: 30166956; PMCID: PMC6111502.

2. Vos MB, Abrams SH, Barlow SE, Caprio S, Daniels SR, Kohli R, Mouzaki M, Sathya P, Schwimmer JB, Sundaram SS, Xanthakos SA. NASPGHAN Clinical Practice Guideline for the Diagnosis and Treatment of Nonalcoholic Fatty Liver Disease in Children: Recommendations from the Expert Committee on NAFLD (ECON) and the North American Society of Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN). J Pediatr Gastroenterol Nutr.

2017;64(2):319-34. doi: 10.1097/MPG.0000000000001482. PubMed PMID: 28107283; PMCID: PMC5413933.

3. Sinha R, Lockman KA, Homer NZM, Bower E, Brinkman P, Knobel HH, Fallowfield JA, Jaap AJ, Hayes PC, Plevris JN. Volatomic analysis identifies compounds that can stratify nonalcoholic fatty liver disease. JHEP reports : innovation in hepatology. 2020;2(5): 100137. Epub 2020/08/11. doi: 10.1016/j.jhepr.2020.100137. PubMed PMID: 32775974; PMCID: PMC7397704.

4. Alkhouri N, Cikach F, Eng K, Moses J, Patel N, Yan C, Hanouneh I, Grove D, Lopez R, Dweik R. Analysis of breath volatile organic compounds as a noninvasive tool to diagnose nonalcoholic fatty liver disease in children. Eur J Gastroen Hepat. 2014;26(l):82-7. doi: 10.1097/MEG.0b013e3283650669. PubMed PMID: WGS:000327763900013.

5. del Rio RF, O'Hara ME, Holt A, Pemberton P, Shah T, Whitehouse T, Mayhew CA. Volatile Biomarkers in Breath Associated With Liver Cirrhosis - Comparisons of Pre- and Postliver Transplant Breath Samples. Ebiomedicine. 2015;2(9): 1243-50. doi:

10.1016/j.ebiom.2015.07.027. PubMed PMID: 181:000365957600040.

6. Schaber C, Katta N, Bollinger LB, Mwawi M, Mlotha-Mitole R, Trehan I, Raman B, Odom John A. Breathprinting reveals malaria-associated biomarkers and mosquito attractants. J Infect Dis. 2018;217(10): 1553-60. doi: 10.1093/infdis/jiy072.

7. Berna AZ, Schaber CL, Bollinger LB, Mwale M, Mlotha-Mitole R, Trehan I, John ARO. Comparison of breath sampling methods: a post hoc analysis from observational cohort studies. Analyst. 2019;144(6):2026-33. doi: 10.1039/c8an01823e. PubMed PMID: WOS:000460765600017.

8. Tangerman A, Meuwese- Arends MT, Jansen JB. Cause and composition of foetor hepaticus. Lancet. 1994;343(8895):483. Epub 1994/02/19. doi: 10.1016/s0140-6736(94)92729-4. PubMed PMID: 7905979.

9. Millonig G, Praun S, Netzer M, Baumgartner C, Dornauer A, Mueller S, Villinger J, Vogel W. Non-invasive diagnosis of liver diseases by breath analysis using an optimized ionmolecule reaction-mass spectrometry approach: a pilot study. Biomarkers. 2010;15(4):297-306. Epub 2010/02/16. doi: 10.3109/13547501003624512. PubMed PMID: 20151876. 10. Qin T, Liu H, Song Q, Song G, Wang HZ, Pan YY, Xiong FX, Gu KS, Sun GP, Chen ZD. The screening of volatile markers for hepatocellular carcinoma. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2010;19(9):2247-53. Epub 2010/09/10. doi: 10.1158/1055-9965.epi-10-0302. PubMed PMID: 20826831.

11. Khalid TY, Costello BDL, Ewen R, White P, Stevens S, Gordon F, Collins P, McCune A, Shenoy A, Shetty S, Ratcliffe NM, Probert CS. Breath volatile analysis from patients diagnosed with harmful drinking, cirrhosis and hepatic encephalopathy: a pilot study. Metabolomics. 2013;9(5):938-48. doi: 10.1007/sll306-013-0510-4.

12. Hanouneh IA, Zein NN, Cikach F, Dababneh L, Grove D, Alkhouri N, Lopez R, Dweik RA. The Breathprints in Patients With Liver Disease Identify Novel Breath Biomarkers in Alcoholic Hepatitis. Clin Gastroenterol H. 2014;12(3):516-23. doi: https://doi.Org/10.1016/j.cgh.2013.08.048.

13. Eng K, Alkhouri N, Cikach F, Patel N, Yan C, Grove D, Lopez R, Rome E, Dweik RA. Analysis of breath volatile organic compounds in children with chronic liver disease compared to healthy controls. J Breath Res. 2015;9(2):026002. doi: 10.1088/1752-7155/9/2/026002.

14. Netzer M, Millonig G, Osl M, Pfeifer B, Praun S, Villinger J, Vogel W, Baumgartner C. A new ensemble-based algorithm for identifying breath gas marker candidates in liver disease using ion molecule reaction mass spectrometry. Bioinformatics. 2009;25(7):941-7. doi:

10.1093/bioinformatics/btp093.

15. Hanouneh IA, Zein NN, Cikach F, Dababneh L, Grove D, Alkhouri N, Lopez R, Dweik RA. The breathprints in patients with liver disease identify novel breath biomarkers in alcoholic hepatitis. Clin Gastroenterol Hepatol. 2014; 12(3):516-23. Epub 09/10. doi:

10.1016/j.cgh.2013.08.048. PubMed PMID: 24036050.

16. Probert CS, Ahmed I, Khalid T, Johnson E, Smith S, Ratcliffe N. Volatile organic compounds as diagnostic biomarkers in gastrointestinal and liver diseases. J Gastrointestin Liver Dis. 2009;18(3):337-43. Epub 2009/10/02. PubMed PMID: 19795029.

17. Belizario JE, Faintuch J, Malpartida MG. Breath Biopsy and Discovery of Exclusive Volatile Organic Compounds for Diagnosis of Infectious Diseases. Front Cell Infect Microbiol. 2020;10:564194. doi: 10.3389/fcimb.2020.564194. PubMed PMID: 33520731; PMCID: PMC7839533.

18. Beauchamp J, Davis C, Pleil J. Breathborne Biomarkers and the Human Volatilome Beauchamp J, Davis C, Pleil J, editors. Boston: Elsevier; 2020 2020/01/01/. 663-75 p.

19. Schaber CL, Katta N, Bollinger LB, Mwale M, Mlotha-Mitole R, Trehan I, Raman B, Odom John AR. Breathprinting Reveals Malaria-Associated Biomarkers and Mosquito Attractants. J Infect Dis. 2018;217(10): 1553-60. Epub 2018/02/08. doi: 10.1093/infdis/jiy072. PubMed PMID: 29415208; PMCID: PMC6279169.

20. Berna AZ, Schaber CL, Bollinger LB, Mwale M, Mlotha-Mitole R, Trehan I, Odom John AR. Comparison of breath sampling methods: a post hoc analysis from observational cohort studies. Analyst. 2019;144(6):2026-33. Epub 2019/02/01. doi: 10.1039/c8an01823e. PubMed PMID: 30702091; PMCID: PMC6868761.

21. Berna AZ, DeBosch B, Stoll J, Odom John AR. Breath Collection from Children for Disease Biomarker Discovery. J Vis Exp. 2019(144):e59217. Epub 2019/03/05. doi: 10.3791/59217. PubMed PMID: 30829338; PMCID: PMC6596991. l. Berna AZ, Akaho EH, Harris RM, Congdon M, Korn E, Neher S, M’Farrej M, Bums J, John ARO. Reproducible breath metabolite changes in children with SARS-CoV-2 infection. Accepted at ACS- Infectious diseases. 2021. doi: 10.1101/2020.12.04.20230755.

23. Vos MB, Abrams SH, Barlow SE, Caprio S, Daniels SR, Kohli R, Mouzaki M, Sathya P, Schwimmer JB, Sundaram SS. NASPGHAN clinical practice guideline for the diagnosis and treatment of nonalcoholic fatty liver disease in children: recommendations from the Expert Committee on NAFLD (ECON) and the North American Society of Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN). Journal of pediatric gastroenterology and nutrition. 2017;64(2):319.

24. Cortes C, Vapnik V. Support vector networks. Mach Learn. 1995;20(3):273-97. doi: 10.1023/a: 1022627411411. PubMed PMID: WOS:A1995RX35400003.

25. Russell S.J. NP. Artificial Intelligence: A Modern Approach: Prentice Hall 2009.

26. Shams ME, ALGayyar MM, Barakat EA. Type 2 Diabetes Mellitus-Induced Hyperglycemia in Patients with NAFLD and Normal LFTs: Relationship to Lipid Profile, Oxidative Stress and Pro-Inflammatory Cytokines. Sci Pharm. 2011;79(3):623-34. doi: 10.3797/scipharm.1104-21. PubMed PMID: 21886908; PMCID: PMC3163367.

27. Raz I, Elder R, Cemea S, Shafrir E. Diabetes: insulin resistance and derangements in lipid metabolism. Cure through intervention in fat transport and storage. Diabetes/metabolism research and reviews. 2005;21(l):3-14. Epub 2004/09/24. doi: 10.1002/dmrr.493. PubMed PMID: 15386813.

28. Van Gossum A, Decuyper J. Breath alkanes as an index of lipid peroxidation. Eur Respir J. 1989;2(8):787-91. Epub 1989/09/01. PubMed PMID: 2680586.

29. Paredi P, Kharitonov SA, Barnes PJ. Analysis of expired air for oxidation products. Am J Respir Crit Care Med. 2002;166(12 Pt 2):S31-7. Epub 2002/12/10. doi: 10.1164/rccm.2206012. PubMed PMID: 12471086.

30. Ratcliffe N, Wieczorek T, Drabinska N, Gould O, Osborne A, De Lacy Costello B. A mechanistic study and review of volatile products from peroxidation of unsaturated fatty acids: an aid to understanding the origins of volatile organic compounds from the human body. J Breath Res. 2020;14(3):034001. doi: 10.1088/1752-7163/ab7f9d. PubMed PMID: 32163929.

31. Shan B, Broza YY, Li W, Wang Y, Wu S, Liu Z, Wang J, Gui S, Wang L, Zhang Z, Liu W, Zhou S, Jin W, Zhang Q, Hu D, Lin L, Zhang Q, Li W, Wang J, Liu H, Pan Y, Haick H. Multiplexed Nanomaterial-Based Sensor Array for Detection of COVID- 19 in Exhaled Breath. Acs Nano. 2020; 14(9): 12125-32. doi: 10.1021/acsnano.0c05657.

32. de Vries R, Vigeveno RM, Mulder S, Farzan N, Vintges DR, Goeman JJ, Bruisten S, van den Corput B, Geelhoed JJM, Visser LG, van der Lubben M, Sterk PJ, in ’t Veen JCCM, Groeneveld GH. Ruling out SARS-CoV-2 infection using exhaled breath analysis by electronic nose in a public health setting. medRxiv. 2021:2021.02.14.21251712. doi: 10.1101/2021.02.14.21251712.

33. Stanislawski MA, Lozupone CA, Wagner BD, Eggesbp M, Sontag MK, Nusbacher NM, Martinez M, Dabelea D. Gut microbiota in adolescents and the association with fatty liver: the EPOCH study. Pediatric research. 2018;84(2):219-27.

34. Abu-Shanab A, Quigley EM. The role of the gut microbiota in nonalcoholic fatty liver disease. Nature reviews Gastroenterology & hepatology. 2010;7(12):691-701.

35. Li D-Y, Yang M, Gong S, Ye SQ. Nonalcoholic Fatty Liver Disease in Children: Role of the Gut Microbiota. The Gut Microbiome-Implications for Human Disease: IntechOpen; 2016. While certain of the preferred embodiments of the present invention have been described and specifically exemplified above, it is not intended that the invention be limited to such embodiments. Various modifications may be made thereto without departing from the scope and spirit of the present invention, as set forth in the following claims.

Claims

What is claimed is:

1. A method for diagnosing or monitoring a test subject with non alcoholic fatty liver disease (NAFLD), the method comprising detecting in a sample of exhaled breath or condensate breath obtained from the test subject for levels in a biomarker panel of volatile organic compounds (VOCs) comprising at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, or 13 of isoprene, p-xylcnc, eucalyptol, o-xylene, toluene, m-ethyl-, \|/-cumene, D-limonene, phenol, decane, undecane, thiophene, 3-methyl-, nonane, and 2,4-dimethyl-l -heptene, detection of altered levels similar to those observed in previously diagnosed patients with NAFLD rather than those observed in healthy control subjects being indicative that said test subject has NAFLD.

2. The method of claim 1, further comprising assessment of ALT, AST, GGT, total bilirubin, alkaline phosphatase, total protein, albumin, cholesterol, TAG, HDL and low-density lipoprotein [LDL] and glycated hemoglobin (HbAlc).

3. The method of claim 1 or claim 2, comprising collection of a stool sample for metagenomic and transcriptomic analyses of fecal microbes in gut microbiota which associate with the NAFLD phenotype.

4. The method claim 1 wherein analysis of the VOC comprises the use of at least one technique selected from the group consisting of photo ionization detection, flame ionization detection, gas chromatography — mass spectrometry (GC-MS), proton transfer reaction mass spectrometry (PTR-MS), colorimetry, infrared spectroscopy, electrochemical fuel cell sensing, semiconductor gas sensing, quartz tuning fork (QTF) sensors, electronic noses and combinations thereof.

4. The method of claim 1 wherein analysis of the VOCs is conducted using a portable, hand-held breathalyzer device.

5. The method of claim 1 comprising: a) analyzing the sample of exhaled breath or condensate breath obtained from the subject for a series of volatile organic compounds (VOCs) comprising each of isoprene, p-xylene, eucalyptol o-xylene, toluene, methy-, \|/-cumene, D-limonene, and phenol VOCs: b) determining a concentration for each of the VOCs; and c) calculating a cumulative abundance based on the concentrations for the VOCs, wherein the cumulative abundance and the concentration of the VOCs indicates NAFLD, wherein said subject is a pediatric, adolescent or adult subject.

6. The method of claim 5 wherein the analysis of the series of VOCs comprises the use of at least one technique selected from the group consisting of photo ionization detection, flame ionization detection, gas chromatography — mass spectrometry (GC-MS), proton transfer reaction mass spectrometry (PTR-MS), colorimetry, infrared spectroscopy, electrochemical fuel cell sensing, semiconductor gas sensing, quartz tuning fork (QTF) sensors, electronic noses and combinations thereof.

7. The method of claim 6 wherein the analysis of the series of VOCs comprises is conducted using a portable, hand-held breathalyzer device.

8. The method of claim 1 wherein the sample is exhaled breath.

9. A method of detecting a combination of VOCs in a subject selected from isoprene, p-xylene, eucalyptol, o-xylene, toluene, m-ethyl-, y-cumene, D-limonene, phenol, decane, undecane, thiophene, 3-methyl-, nonane, and 2,4-dimethyl- 1 -heptene, the method comprising analyzing a sample of exhaled breath or condensate breath obtained from the subject for said VOCs.

10. The method of claim 9, wherein analysis of the at least one monoterpene comprises the use of at least one technique selected from the group consisting of photo ionization detection, flame ionization detection, gas chromatography - mass spectrometry (GC-MS), proton transfer reaction mass spectrometry (PTR-MS), colorimetry, infrared spectroscopy, electrochemical fuel cell sensing, semiconductor gas sensing, quartz tuning fork (QTF) sensors, electronic noses and combinations thereof.

11. The method of claim 9 wherein analysis of the VOCs is conducted using a portable, handheld breathalyzer device.

12. The method of claim 1, further comprising condensing or concentrating the sample before analysis.

13. The method of claim 1 , further comprising administering to the subject a pharmaceutical composition comprising a therapeutically effective amount of at least one compound effective for alleviating symptoms of NAFLD.

14. The method of any one of claims 1-13, wherein a) the isoprene is increased when compared to the healthy control; b) the decane is decreased when compared to the healthy control; c) the toluene, m-ethyl- is decreased when compared to the healthy control; d) the y-cumene is decreased when compared to the healthy control; e) the p-xylene is decreased when compared to the healthy control; f) the 2,4-dimethyl-l -heptene is decreased when compared to the healthy control; g) the limonene is decreased when compared to the healthy control; h) the phenol is decreased when compared to the healthy control; i) the o-xylene is decreased when compared to the healthy control; j) the eucalyptol is decreased when compared to the healthy control; k) the undecane is increased when compared to the healthy control; l) the thiophene, 3-methyl- is increased when compared to the healthy control; or m) the nonane is increased when compared to the healthy control.

15. The method of claim 14, wherein 2, 3, 4, 5, 6, 7, 8 , 9, 10, 11, 12, or 13 of a)-m) are present.