NASH: Diagnosis and Treatment


79

The rs2294918 K434E PNPLA3 Polymorphism is associated with Nonalcoholic Fatty Liver and Steatohepatitis

Luca Valenti1,2, Stefano Romeo3,4, Benedetta M. Motta1,3, Benedetta Del Menico1, Raffaela Rametta1, Giula Buonaiuto1, Anna Alisi5, Anna Ludovica Fracanzani1,2, Enrico Mozzi1,6, Benedetta Donati1, Maria Antonella Burza3, Paolo Dongiovanni2, Valerio Nobili5, Silvia Fargion1,2

1Pathophysiology and Transplanation, Universitá degli Studi di Milano, Milano, Italy; 2Internal Medicine, Fondazione IRCCS Cá Granda Ospedale Policlinico, Milano, Italy; 3Sahlgrenska Center for Cardiovascular ond Metobolic Research/Wallenberg Laboratory, Department of Molecular and Clinical Medicine, University of Gothenburg, Gothenburg, Sweden; 4Clinical Nutrition Unit, Department of Medical and Surgical Sciences, University Magna Graecia, Catanzaro, Italy; 5Hepato-metabolic unit, Ospedale Bambin Gesù, Roma, Italy; 6Surgery, Fondazione IRCCS Cá Granda Ospedale Policlinico, Milano, Italy

Background & aims: The Patatin-like phosholipase domain-containing 3 (PNPLA3) rs738409 polymorphism (I148M) is the major genetic determinant of hepatic fat and predisposes to the full spectrum of liver damage in nonalcoholic fatty liver disease (NAFLD). Aim of this study was to evaluate whether NAFLD is associated with additional PNPLA3 variants. As we detected an association between the rs2294918 polymorphism encoding for the K434E and pediatric NAFLD, which was independent of I148M, we went on to validate these findings in adult patients and to investigate the mechanism by evaluating the association with gene expression. Patients and methods: We considered 142 pediatric and 265 adult patients with histological NAFLD, and 225 healthy controls without NAFLD. Liver damage was assessed by NAFLD activity score, PNPLA3 coding sequence by sequencing, the rs738409 and rs2294918 polymorphisms by Taqman assays, and PNPLA3 expression quantified by qRT-PCR in needle biopsy samples of 60 severely obese patients. Genetic association was tested by Haploview v4.2. Results: Nor novel neither already described rare coding variants of PNPLA3 were detected in pediatric patients and controls. Besides I148M, we identified another variant, rs2294918 A>G (K434E), which was associated with NAFLD (p=9x10-4). The 148I was negatively associated with pediatric NAFLD only in combination with the 434K allele, whereas the 148M was positively associated with NAFLD only in combination with the 434E allele. In pediatric 148I/I patients, the 434E allele at risk of NAFLD was associated with increased ALT (p=0.05). In adult patients, despite being observed at a comparable frequency, only the 148I-434K, but not the 148I-434E, diplotype was negatively associated with NASH, whereas the 148M allele was positively associated with NAFLD only in combination with the 434E allele. At multivariate logistic regression analysis, the K434E variant was associated with NASH independently of I148M status (OR 2.78, 95% c. i. 1.20-6.56). The K434E varaiant was not predicted to alter PNPLA3 function by bioinformatic algorithms, but the 434E at risk allele was associated with higher hepatic PNPLA3 expression (p=0.01). Conclusions: the K434E PNPLA3 variant is associated with predisposition to severe NAFLD in combination with the I148M. Further studies are necessary to clarify the mechanisms, and to assess the effect on other liver-related outcomes.

Disclosures:

The following people have nothing to disclose: Luca Valenti, Stefano Romeo, Benedetta M. Motta, Benedetta Del Menico, Raffaela Rametta, Giula Buonaiuto, Anna Alisi, Anna Ludovica Fracanzani, Enrico Mozzi, Benedetta Donati, Maria Antonella Burza, Paola Dongiovanni, Valerio Nobili, Silvia Fargion

80

Metabomic Signature of Human Non-Alcoholic Fatty Liver Disease Provides Insights into Potential Microbial Contribution to Disease Status

Puneet Puri1, Mohammad S. Siddiqui1, Carol Sargeant1, Sherry L. Boyett1, Larry D. White1, Kalyani Daita1, Faridoddin Mirshahi1, Melanie White1, Tommy Pacana1, Vaishali Patel1, Andrew R. Joyce4, Masoumeh Sikaroodi2, Iliana Bouneva1, Richard K. Sterling1 , R. Todd Stravitz1, Velimir A. Luketic1, Robert Mohney3, Jasmohan S. Bajaj1, Patrick M. Gillevet2, Arun J. Sanyal1

1Virginia Commonwealth University, Richmond, VA; 2Department of Environmental Science and Policy, George Mason University, Prince William Campus, Manassas, VA; 3Metabolon, Durham, NC; 4Venebio, Richmond, VA

BACKGROUND: Non-alcoholic Fatty Liver Disease (NAFLD) afflicts ∼30% of US population. Spectrum of NAFLD includes fatty liver (NAFL) and steatohepatitis (NASH) that can progress to cirrhosis (Cx) and liver cancer. Altered fecal microbiome (FMi) is being implicated in NAFLD pathogenesis. AIMS: To characterize the microbiome and metabolome of entire NAFLD spectrum from NAFL to Cx compared to controls and its putative relevance to pathogenesis. METHODS: 16S RNA multitag pyrosequencing for fecal microbiome (FMi) and mass spectrometry for small molecule metabolomic profiling from plasma (PMe) and feces (FMe) were performed. UniFrac principle coordinates analysis (PCO)and QIIME for FMi and Welch t-test for metabolome were performed. RESULTS: Age and gender matched 83 subjects - control (n=7), [NAFL (n=14), and NASH (n=31) based on NASH-CRN histology criteria] and Cx (n=31) were studied. Bacteroidetes, Firmicutes and Proteobacteria comprised ∼98-99% of phyla with similar distribution in controls, NAFL, NASH and Cx groups. While Bacteroidia and Clostridia were the most abundant classes, Bacteroidaceae, Lachnospiraceae and Ruminococcaceae constituted main families, although no notable differences were found between groups. Metabolomic (PMe and FMe) data were available for all groups except Cx. Interestingly, FMi generated or related metabolites were significantly and differentially changed in NAFL and NASH (FMe changes>> PMe changes). Metabolites of microbial origin or contributions: In plasma, urobilinogen showed 3.2-fold increase in NASH subjects compared to controls, and other heme metabolites also tended to increase both in NAFL and NASH. In contrast, several choline-containing lysolipids had decreasing trends likely reflecting a decrease in PC synthesis. In feces, phenyllactate and 3, 4-dihydroxybenzoate were significantly decreased in both NAFL and NASH compared to controls. Notably, NASH subjects compared to NAFL had marked increase in p-cresol sulphate, 3-(3-hydroxyphenyl) propionate, phenol sulfate and glycodeoxycholate. Taurolithocholate and phenylacetyl glutamine levels were strikingly high in subjects with NASH compared to controls, whereas tricarballylate level was significantly higher in NAFL compared to NASH. CONCLUSION: NAFLD is associated with significant changes in fecal and systemic metabolites of microbial origin/contribution despite no significant differences in microbial biodiversity. Postulated pathophysiologic implications are related to altered membrane permeability, aromatic amino acid metabolism and cellular stress. These findings offer the potential for novel biomarker discovery distinguishing NAFL and NASH.

Disclosures:

Puneet Puri - Advisory Committees or Review Panels: Health Diagnostic Laboratory Inc.; Consulting: NPS Pharmaceuticals Inc.

Andrew R. Joyce - Independent Contractor: Venebio Group, LLC; Management Position: Venebio Group, LLC

Richard K. Sterling - Advisory Committees or Review Panels: Merck, Vertex, Salix, Bayer, BMS, Abbott; Grant/Research Support: Merck, Roche/Genentech, Pfizer, Medtronic, Boehringer Ingelheim, Bayer, BMS, Abbott

R. Todd Stravitz - Grant/Research Support: Exalenz Biosciences, LTD

Velimir A. Luketic - Grant/Research Support: Intercept, Merck, Idenix, Vertex, Gilead, BMS, Novartis, abbvie, Genfit, Takeda

Robert Mohney - Employment: Metabolon, Inc.

Jasmohan S. Bajaj - Advisory Committees or Review Panels: Salix, Merz, otsuka, ocera, grifols, american college of gastroenterology; Grant/Research Support: salix, otsuka, grifols

Patrick M. Gillevet - Management Position: BioSpherex LLC

Arun J. Sanyal - Advisory Committees or Review Panels: Gore, Gilead, Abbott, Ikaria; Consulting: Salix, Immuron, Exhalenz, Bayer-Onyx, Genentech, Norgine, GalMed, Novartis, Echosens, Takeda; Grant/Research Support: Salix, Genentech, Genfit, Intercept, Ikaria, Takeda, Gilead; Independent Contractor: UpToDate

The following people have nothing to disclose: Mohammad S. Siddiqui, Carol Sargeant, Sherry L. Boyett, Larry D. White, Kalyani Daita, Faridoddin Mirshahi, Melanie White, Tommy Pacana, Vaishali Patel, Masoumeh Sikaroodi, Iliana Bouneva

81

Circulating micro-RNA Profile in Nonalcoholic Fatty Liver Disease: Potential Biomarkers and its role in the modulation of the metabolic syndrome?

Carlos J. Pirola3, Tomas Fernandez Gionotti3, Adriana L. Burguñeo3, Gustavo O. Castaño2, Silvia Sookoian1,2

1Clinical and Molecular Hepatology, IDIM-CONICET, Ciudad Autonoma de Buenos Aires, Argentina; 2Medicine ond Surgery Department, Hospital Abel Zubizarreta, Research Council in Health, Ciudad Autónoma de Buenos Aires, Argentina; 3Moleculor Genetics and Biology of Complex Diseases, IDIM-CONICET, Ciudad Autónomo de Buenos Aires, Argentina

MicroRNAs (miRNAs) are very small non-coding RNA molecules that have emerged as key regulators of gene expression and metabolism. Circulating miRNAs are present in clinical samples in a remarkably stable form that can be used as potential disease biomarkers; circulating miRNAs have also a strong potential as endocrine signalling molecules. We first aimed to characterize the circulating miRNA profile in patients with NAFLD proven through biopsy in a case-control design. We also aimed to explore the differential expression in the liver of candidate miRNAs. Methods: This study was performed in three phases: (i) global serum miRNA profiling using a 84-miRNA array by real-time RT-PCR (n=48 participants), (ii) independent validation of selected miRNAs (n=86), and (iii) evaluation of liver expression of candidate miRNAs (n=52). Phase (i) and (ii) were conducted using independent groups of patients; phase (iii) involved liver tissue of patients recruited in phase (ii). Results: the comparison of global serum miRNA profile among patients with simple steatosis (SS), NASH and controls showed at least three up-regulated miRNAs including, miRNA-122a (7.2 fold change NASH vs. controls and 2.3 fold change NASH vs. SS), miRNA-192 (4.4 fold change NASH vs. controls) and miRNA-19b (2.1 fold change NASH vs. controls and 4.2 fold change SS vs. controls). These results were replicated in a second set of serum samples, including control subjects (n=20) and 65 NAFLD patients. Both, miRNA-122a (involved in lipid metabolism) and miRNA-192 (involved in TGF-β1 signaling) were significantly associated with the histological progression of NAFLD (regression analysis for an ordinal multinomial distribution: p=0.006 and p=0.03, respectively). Serum miRNA-122a levels were significantly correlated with plasma triglycerides (R: 0.32, p=0.006), AST (R: 0.5, p=0.04), ALT (R: 0.4, p=0.0004) levels, and liver fibrosis (R: 0.3, p=0.02). In addition, serum miRNA-19b was significantly up-regulated in NASH patients in comparison with controls, and significantly correlated with plasma glucose levels (R: 0.26, p=0.02), BMI (R: 0.4, p=0.0004) and biomarkers of atherosclerosis (sICAM1 R: 0.43, p=0.04). 〇f note, liver miRNA-122a expression was fold down-regulated in NASH patients in comparison with SS (p=0.03). In conclusion: dysregulated miRNA-122 and 192 expression is associated with NAFLD disease progression. Circulating levels of miRNA-19b, which is a regulator of nuclear factor-қB activity, is a novel candidate involved in the cardiovascular phenotype of NAFLD patients.

Disclosures:

Carlos J. Pirola - Advisory Committees or Review Panels: Merck Sharp and Dohme; Grant/Research Support: Merck Sharp and Dohme

The following people have nothing to disclose: Tomas Fernandez Gianotti, Adriana L. Burgueño, Gustavo O. Castaño, Silvia Sookoian

82

Extended treatment with pioglitazone improves liver histology in patients with prediabetes or type 2 diabetes mellitus and NASH

Kenneth Cusi1,3, Beverly Orsak3, Romina Lomonaco1, Fernando Bril1, Carolina Ortiz-Lopez3, Joan Hecht2, Amy Webb2,3, Fermin Tio2,3, Celia M. Darlana2, Jeon Hardies3

1Division of Endocrinology, Diabetes and Metabolism, University of Florida, Gainesville, FL; 2Diabetes Division, Audie Murphy VAMC, San Antonio, TX; 3Diabetes Division, Univ of Texas HSC of San Antonio, San Antonio, TX

Background and Aim: Nonalcoholic steatohepatitis (NASH) is increasingly common in patients with prediabetes or type 2 diabetes (T2DM). In this population, pioglitazone (PIO) was reported to be effective in a small 6-month proof-of-concept randomized controlled trial (RCT) (Belfort et al, NEJM 2006). We performed a long-term RCT to fully assess the efficacy/safety of PIO in patients with prediabetes or T2DM. Methods: We randomized 101 patients with biopsy-proven NASH (age=51±1 yr, BMI: 34.4±0.5 kg/m2, A1c: 6.3±0.1%, 49/51% prediabetes/T2DM) to PIO vs. placebo. We assessed at before and after 18 months of treatment: 1) Liver histology (primary endpoint); 2) Liver fat (LFAT) by MRS; 3) Measures of adipose tissue, liver and muscle insulin sensitivity, both fasting (hepatic-insulin resistance [IR] or HIRi= hepatic glucose production [HGP] x fasting plasma insulin [FPI] and adipose tissue-IR or AdipoIRi= FFA x FPI) and during an insulin clamp with glucose turnover. Results: Placebo- and PIO-treated patients were wellmatched at baseline for clinical/metabolic and histological characteristics (all NS). More patients on PIO (55%) vs. placebo (19%; p<0.001) had improvement in steatohepatitis (primary outcome), defined as an improvement by ≥1 points in the hepatocellular ballooning score; no increase in the fibrosis score; and either a decrease in the nonalcoholic fatty liver disease activity score (NAS) to ≤ 3 or a decrease in the activity score of at least 2 points (with 1-point decrease in either the lobular inflammation or steatosis score). More subjects had improvement with PIO vs. placebo in the scores for steatosis (75% vs. 29%, p<0.001), ballooning (55% vs. 26%, p<0.01), and inflammation (53% vs. 26%, p=0.01), as well as in the NAS (as defined above) (83% vs. 38%, all p<0.001). All mean scores improved compared to baseline with PIO, but were unchanged with placebo (all p<0.001 vs. placebo). Fibrosis decreased from baseline with PIO (p<0.01) and this change was significant vs. placebo (p=0.03). On secondary clinical/ metabolic outcomes, PIO decreased 60% LFAT and improved liver aminotransferases, A1c, FPI, HOMA-IR and insulin sensitivity in adipose tissue (AdipoIRi; insulin-suppression of FFA), liver (HIRi: insulin-suppression of HGP) and insulininduced muscle glucose uptake (all p<0.001). No patient on PIO discontinued the study due to an adverse event. Conclusion: Treatment of patients with prediabetes/T2DM and NASH with PIO for 18 months was safe and effective in improving liver histology and insulin resistance. Longer-term studies are needed to assess its overall safety and whether PIO modifies the natural history of NASH in prediabetes/T2DM.

Disclosures:

Kenneth Cusi - Consulting: Merck, Daichi-Sankyo, Roche, Janssen; Grant/Research Support: Takeda, Novartis, Mannkind

Beverly Orsak - Employment: UTHSCSA

The following people have nothing to disclose: Romina Lomonaco, Fernando Bril, Carolina Ortiz-Lopez, Joan Hecht, Amy Webb, Fermin Tio, Celia M. Darland, Jean Hardies

83

Modest Alcohol Consumption Decreases the Risk of Having Nonalcoholic Fatty Liver Disease: A Meta-Analysis of 43, 175 Individuals

Silvia Sookoian1,2, Gustavo O. Castaño2, Carlos J. Pirola3

1Cinical and Molecular Hepatology, IDIM-CONICET CABA, Argentina; 2Medicine and Surgery Department, Hospital Abel Zubizarreta, CABA, Argentina; 3Molecular Genetics and Biology of Complex Diseases, IDIM-CONICET, CABA, Argentina

Epidemiological studies suggested that the prevalence and histological disease severity of NAFLD are lower for people who drink modest amounts of alcohol than those who are abstainers. Nevertheless, the final answer of whether or not patients with NAFLD should be allowed modest alcohol consumption is still open. Hence, our primary purpose was to estimate from published data the effect of light or modest alcohol consumption (MAC) on the risk of having NAFLD by performing a metaanalysis to give a quantitative assessment of this relationship. In addition, we aimed to identify the gender-specific effect and also the relationship with co-morbidities, such as obesity. Due to the presence of limited published data about the association between modest alcohol consumption and NAFLD histological disease severity, as a secondary purpose, we collected our own data and revised current literature. Literature searches identified nine studies. In addition to published research, we included data from a case-control study conducted with 414 Argentinean adults. The meta-analysis included 43,175 adult individuals (30,791 nondrinkers and 12,384 modest drinkers). Our results showed that MAC is associated with a significant protection on the risk of NAFLD in both fixed (OR 0.688, 95% CI: 0.646-0.733, p<10-8) and random model (O R 0.684, 95% CI: 0.580-0.806, p<10-5). Meta-regression analysis showed that this association was independent of BMI (slope= 0.01, p<0.44). We observed a significant higher protective effect of MAC on NAFLD in female sex (n: 12, 459, p< 0.0002). Pooled data of liver histology from 822 patients (550 nondrinkers and 272 modest drinkers) showed that MAC has a significant protective effect on the development of NASH in both fixed and random models (OR 0.501, 95% CI: 0.340-0.740, p<0.0005). In conclusion, quantitative evidence showed a significant protective effect of about 31% of the risk of having NAFLD associated with MAC. This beneficial effect was independent of covariates such as BMI, but was influenced by sex. Among females, the protective effect of MAC on NAFLD was higher (about 53%) than males (about 30%). MAC was associated with an average protective effect of about 50% on the risk of developing an advanced disease stage.

Disclosures:

Carlos J. Pirola - Advisory Committees or Review Panels: Merck Sharp and Dohme; Grant/Research Support: Merck Sharp and Dohme

The following people have nothing to disclose: Silvia Sookoian, Gustavo O. Castaño

84

Hedgehog Pathway Targeted by Vitamin E Therapy in NASH

Cynthia D. Guy1, Ayako Suzuki2, Manal F. Abdelmalek3, James Burchette1, Anna Mae Diehl3

1Pathology, Duke University Medical Center, Durham, NC; 2Medicine, University of Arkansas for Medical Science, Little Rock, AR; 3Medicine, Duke University Medical Center, Durham, NC

INTRODUCTION AND AIMS Hedgehog (Hh) ligand production by ballooned hepatocytes has been implicated in NASH progression in mice. The NIDDK-sponsored PIVENS trial (NCT00063622) showed that Vitamin E (VitE) improved NASH. Because Hh signaling promotes fibrogenesis and is upregulated in NASH, we aimed to determine if VitE decreased Hh pathway activity. METHODS Immunohistochemistry (IHC) was performed on both pre- and post-treatment liver biopsies of 59 PIVENS patients randomized to VitE (n=29) or placebo (n=30). Hh ligand expression was evaluated by IHC for Sonic Hh (Shh). Response to Hh ligand, glioblastoma 2 (gli2) positivity, was evaluated by IHC for Hh-responsive (gli2+) progenitors (SRY-related high-mobility group box 9 [sox-9]-positive cells) and myofibroblastic hepatic stellate cells (a-smooth muscle actin [a-SMA]-positive HSCs). Ballooning injury was evaluated by keratin 8/18 and ubiquitin (K8/18/Ub) IHC. RESULTS Regardless of the treatment arm, changes in the number of Shh+ hepatocytes were correlated with K8/18/Ub foci (r2=0.47, p<0.001) and AST (r2=0.15, p=0.002). Compared to the placebo group, the VitE group showed a greater reduction in the number of Shh+ hepatocytes (p<0.05) and K/8/1 8/Ub foci (p=0.08). After adjusting for baseline numbers of Shh+ hepatocytes or K8/18/しb foci, the VitE group showed a greater decrease in Shh+ hepatocytes (p<0.04) and K8/18/Ub foci (p<0.04) by multiple linear regression analysis. Similarly, changes in serum aminotransferases, AST (partial r2=0.75, p<0.0001) and ALT (partial r2=0.26, p<0.0001), as well as changes in ballooning (p=0.004) and fibrosis stage (p=0.02), were also associated with changes in Shh+ hepatocytes. Furthermore, changes in the numbers of Shh+ hepatocytes were associated with changes in the numbers of Shh-responsive progenitors (gli2+/sox9+ cells) (p=0.03), i. e., ductular reaction resolution, by multiple ordinal logistic regression after adjusting for baseline values. Activated HSCs (α SMA+ cells) tended to be associated (p=0.10) with changes in the numbers of Shh+ hepatocytes. A response to treatment (as established by the PIVENS study design) was associated with a greater decrease in Shh+ hepatocytes compared to nonresponse (p=0.007). CONCLUSIONS Reduction of NASH with VitE therapy decreased production of Hh ligand and accumulation of Hh-responsive progenitors (i. e., decreased the ductular reaction). Inhibition of Hh pathway activity was accompanied by improved liver injury, fibrosis stage, and treatment response, implicating Hedgehog as a therapeutic target for VitE. Further study of mechanistic interrelationships between NASH and Hedgehog may reveal novel therapeutic targets in NASH.

Disclosures:

Manal F. Abdelmalek - Grant/Research Support: Mochida Pharmaceuticals; Speaking and Teaching: Takeda Pharmaceuticals

Anna Mae Diehl - Consulting: Bristol Myers Squibb, Synergy, GlaxoSmithKline, Norgine; Grant/Research Support: GlaxoSmithKline

The following people have nothing to disclose: Cynthia D. Guy, Ayako Suzuki, James Burchette

Ancillary