- Email: [email protected]
Mo1001 Prognostic Relevance of Serum Ferritin in Patients With Nonalcoholic Fatty Liver Disease
Mo1000 Gaps in Primary Care Physician Evaluation of Incidentally Identified Hepatic Steatosis Andrew P. Wright, Amit P. Desai, Surabhi Bajpai, Dushyant V. Sahani, Kathleen E. Corey
AASLD Abstracts
Background/Aims: Hepatic steatosis is a frequent incidental finding on abdominal CT scans. Primary care providers (PCPs) commonly perform the evaluation of steatosis however limited data exist regarding current PCP decision-making practice and outcomes in this area. We aimed to characterize and identify variables that predict PCP identification of steatosis. Methods: From January 2008 to October 2011, we identified all patients who underwent abdominal CT scans for evaluation of hematuria or nephrolithiasis at the Massachusetts General Hospital (MGH). We subsequently limited our population to patients with radiographically identified hepatic steatosis. Patient characteristics are described in Table 1. Patient care notes were reviewed for diagnosis of steatosis, alcohol use, medications, and diagnosis of hypertension, hyperlipidemia, and diabetes. Lab values were examined at two time points (Table 2). Baseline data consisted of most recent lab values prior to imaging and evaluation data consisted of the lab values within 14 months after imaging was performed. Lastly, we analyzed radiology reports and recorded whether hepatic steatosis was commented on in the body or the impression. Results: A total of 356 patients with reported steatosis on CT were identified. Patients without MGH PCPs (n=125) or a PCP visit within 14 months of image date (n=10), with prior known liver disease (n=69), prior organ transplant (n=4), or diagnosis of metastatic cancer (n=21) were excluded leaving 127 patients in the study. In the 14 months after identification, steatosis was commented on in PCP follow up in only 23% of patients. HCV screening was performed in 6.3% of patients and 59% underwent insulin resistance screening. Over 80% of patients underwent LFT and lipid screening. Of the 127 patients, the majority of radiology reports commented on fatty liver in the impression vs. body (83 vs. 44). There was a significant difference in the rate of follow up when radiology reports commented on fatty liver in the impression vs. the body (30.1% vs. 9.1%, p = 0.007). Retrospective application of NAFLD fibrosis score to prior data identified 14 patients at high risk of hepatic fibrosis, however this was not predictive of PCP documentation of fatty liver (p=0.51). New cases of insulin resistance were identified in 36% of patients (12% diabetes, 24% pre-diabetes) who underwent screening within 14 months of imaging. Conclusions: We concluded that PCPs infrequently document and follow up steatosis reported on imaging studies. Radiology reporting practices significantly impact PCP documentation rates, and our data suggest that steatosis should be recorded in impression section of reports. Efforts to improve PCP evaluation of steatosis should focus on increased evaluation of alcohol use, insulin resistance, HCV infection, and assessment for advanced fibrosis. Table 2: Laboratory Data Among Patients With Incidentally Identified Hepatic Steatosis
Baseline data refers to prior PCP documentation of presence of hypertension, hyperlipidemia, or diabetes, or active prescription of medication to treat these conditions at time of CT scan. Presence of abnormal LFTs refers to ALT .55 for men and .33 for women, AST .40 for men and .32 for women. Alcohol screening refers to specific quantity documented by PCPs. "Social/occasional" and "moderate" reflect patients where this was language used to characterize alcohol use by PCP in care notes. Mo1001 Prognostic Relevance of Serum Ferritin in Patients With Nonalcoholic Fatty Liver Disease Njideka Momah, Lee Russell, Elisabetta Bugianesi, Jacob George, Paul Angulo
Abnormal values refer to: ALT .55 for men and .33 for women, AST .40 for men and .32 for women, HgbA1c .5.7%, Fasting glucose .100, Total cholesterol .200, HDL ,40 for men and ,50 for women, Triglyceride .150, Ferritin .200, Plt count ,150, and HCV Ab positive. % of patients screened indicates percentage of patients with available lab value within pre-specified time period. Table 1: Baseline Characteristics of Study Population
S-1009
Serum ferritin (SF) levels are often elevated in patients with nonalcoholic fatty liver disease (NAFLD). Some cross-sectional studies have reported a relationship between elevated SF in patients with NAFLD and more severe liver fibrosis. However, since all studies reported to date are cross-sectional studies with patients evaluated at a single point in time, the longterm prognostic significance of elevated serum ferritin levels in patients with NAFLD remains unclear. The aim of our study was to determine the association between elevated SF levels and the long-term prognosis of patients with NAFLD. Methods: This is a longitudinal, longterm follow-up study of a cohort of 307 patients with liver biopsy confirmed NAFLD. These patients were followed-up for a median of 8.5 years (range 0.5 - 26.4 years). Extensive clinical and laboratory data were collected at baseline and during follow-up. The upper limit of normal (ULN) for SF used for comparisons was adopted from the hemochromatosis and iron overload screening study [i.e., ,300 μg/L for men and ,200 μg/L for women]. Patients with various degrees of elevation of SF were compared to patients with lower levels using three different cut-off points of SF selected a priori: . ULN, . 1.5 x ULN, and . 2 x ULN. Cumulative liver-related events, non-liver related events, and overall mortality were the outcomes analyzed and compared in the three different groups using Kaplan-Meier analysis. Hazard rate ratio (HR) estimates for outcomes were calculated by the Cox proportional hazard regression analysis to control for the effect of potential confounding factors while taking into consideration different duration of follow-up. Results: The mean (±SE) age was 50±0.6 years, 142 (46%) were men, 211 (69%) were obese, 109 (35%) suffered from diabetes, and 280 (91%) were of white race. SF was .ULN in 89 (29%), .1.5 x ULN in 54 (18%), and .2 x ULN in 33 (11%) patients. During follow-up, 35 (11%) patients developed one or more liver events, 64 (21%) patients developed one or more non-liver events, and 39 (12.7%) patients died or underwent liver transplantation. There was no significant difference in the proportion of cumulative outcomes of liver-events, non-liver events or mortality/liver transplantation in any of the three groups in the unadjusted (KaplanMeir) and adjusted (Cox Regression) analyses as described in the Table. Conclusions: Elevated SF levels are not associated with a higher risk of developing liver-related events, nonliver related events, or mortality or liver transplantation in patients with nonalcoholic fatty liver disease. Cumulative probability (Unadjusted) and multivariate-adjusted hazard ratios and 95% CI's for outcome by serum ferritin status
AASLD Abstracts
AASLD Abstracts
confirm the presence of NASH is a liver biopsy, which is expensive, invasive and subject to sampling error. A noninvasive blood test to detect hepatic fat and fibrosis to replace biopsy and aid in clinical decision-making would represent a significant advance. The goals of this study were to develop a serum proteomic signature to: 1. Detect and monitor severity of steatosis 2. Detect and monitor severity of fibrosis (NASH) Methods: Serum samples were collected and liver biopsies were performed on 443 obese patients who underwent bariatric surgery for weight loss at the Geisinger Clinic. Of these, 125 had normal biopsies, 137 showed varying degrees of steatosis (mild; moderate; severe), and 181 showed various grades of NASH fibrosis (1-4). Biomarker discovery was performed using SomaLogic's highly multiplexed SOMAscan™ proteomic assay, which measured 1129 proteins simultaneously from ~65μL serum. Demographics such as age, BMI and LDL cholesterol levels were balanced across all groups. Candidate markers were selected using a stability selection algorithm with a L1-regularized logistic regression kernel. Results: Two random forest classifiers were developed. The first, which was designed to detect liver steatosis, compared controls versus all other groups and yielded an area under the ROC curve (AUC) of 0.90±0.03 (sensitivity 92%; specificity 63%; @ 0.5 cutoff). Based on this 9-protein panel, all 7 NAFLD groups (steatosis and NASH) had elevated probabilities of steatosis (classifier votes). The second considered only groups with hepatic steatosis and distinguished fibrosis (NASH grades 24) from liver without fibrosis (steatosis). This 4-protein classifier had an AUC of 0.83±0.07 (sensitivity 62%; specificity 91%; @ 0.5 cutoff). When applied to NASH grade1 subjects, the average probability of fibrosis vote fell between the average severe steatosis and NASH grade 2 votes. Conclusion: In this initial study we have discovered a set of serum protein biomarkers that are highly predictive of fatty and fibrotic liver. The markers were associated with metabolism, matrix remodelling, apoptosis, and inflammation. These results will be further validated on a larger sample set. Mo1004 a Estimated using Kaplan-Meier analysis. b Estimated using Cox proportional hazard regression analysis. c Adjusted for age, gender, race, body mass index, diabetes, and fibrosis stage CI, confidence intervals
Among Nonalcoholic Fatty Liver Disease Patients Treated With Phlebotomy Does PNPLA3 Genotype Influence Changes in Gene Expression, Adipokines and Liver Histology? Melanie D. Beaton, Kelly Summers, Sarah Woolsey, Yun-Hee Choi, Rommel G. Tirona
Mo1002
Purpose: Variation in PNPLA3 (rs738409, C/G) has been associated with increased NALFD severity, but whether this genotype influences the outcome of NAFLD therapy is unknown. We sought to characterize the impact of this genotype on histology, markers of inflammation, fibrosis, angiogenesis, metabolic dysregulation, oxidative stress and serum adipokines in NAFLD patients treated with phlebotomy with paired liver biopsies. Methods: Thirty-one patients underwent phlebotomy for iron depletion (serum ferritin ≤50 ng/ml or Hgb 10.0 g/ dL). All were genotyped for PNPLA3 and repeat liver biopsy, anthropometric and biochemical measurements obtained following end of treatment. The primary outcome, improvement in NAFLD Activity Score (NAS) was achieved in 38.7%. We evaluated the relationship between PNPLA3 genotype and changes in liver histology, hepatic gene expression of hepcidin, hemeoxygenase-1, connective tissue growth factor, CD34, monocyte chemotactic protein-1, tumor necrosis factor α, PNPLA3 and phosphoenolpyruvate carboxykinase measured by real-time polymerase chain reaction, as well as, serum levels of adiponectin, c-peptide, leptin, resistin and visafatin analyzed by multiplex immunoassay. Results: Pre-phlebotomy, PNPLA3 G allele carriers (C/G and G/G) had higher NAS (p=0.007) and ballooning (p=0.01) but not steatosis (p=0.06), inflammation (p=0.07) or fibrosis (p=0.64) compared with C/C homozygotes. Following therapy, decreased NAS (p=0.01) was seen in PNPLA3 G allele carriers. Phlebotomy did not affect histology in C/C homozygotes. Steatosis (p=0.09), inflammation (p=0.11), ballooning (p=0.06) nor fibrosis (p=0.45) were significantly different between carriers and C/C homozygotes. Phlebotomy significantly reduced liver hepcidin (p=0.01) and heme oxygenase 1 gene expression (p=0.02), a measure of oxidative stress. Treatment did not affect expression of other genes examined. Phlebotomy was significantly associated with increases in leptin (p=0.03) but had no effect on other adipokines. PNPLA3 G allele carriers had significantly decreased heme oxygenase 1 expression in comparison to those not carrying the allele (p ,0.001). Pre and post therapy changes in resistin levels were significantly different between PNPLA3 G allele carriers and non-carriers (p=0.05). Conclusion: Possession of the PNPLA3 G allele is associated with improvement in liver histology with iron reduction therapy. Phlebotomy resulted in decreased liver oxidative stress and increased serum leptin, irrespective of PNPLA3 genotype. Determination of PNPLA3 genotype in individuals with NAFLD may be helpful in guiding therapy and predicting response to treatment for NALFD.
In Non-Alcoholic Fatty Liver Disease, Differential Gene Expression for Phospholipid Metabolism Is Associated With Lower Hepatic Phosphatidylcholine to Phosphatidylethanolamine Ratio, but Not With Lower Choline Intake Bianca M. Arendt, Ian D. McGilvray, Elena Comelli, David W. Ma, Scott Fung, AnaLaura Guzman, Sandra E. Fischer, Johane P. Allard Background: It has been suggested that a lower hepatic ratio of phosphatidylcholine (PC) to phosphatidylethanolamine (PE) may contribute to the pathogenesis of non-alcoholic fatty liver disease (NAFLD), as changes in the phospholipid membrane can affect cell integrity and signal transduction. Lower PC/PE ratio could be due to altered expression of enzymes involved in phospholipid metabolism and/or inadequate choline intake. The aim of this study was to compare hepatic gene expression, choline intake, and hepatic PC/PE ratio between patients with simple steatosis (SS), non-alcoholic steatohepatitis (NASH) and healthy controls (HC). Methods: Hepatic gene expression profiles (Illumina Whole-Genome DASL HT-assay) were determined from liver biopsies taken from NAFLD patients (SS: n= 24, NASH: n=17) or as the first step of a live donor hepatectomy (HC: n=23). In a subgroup of participants, hepatic PC and PE (mass spectrometry) as well as dietary choline intake (7day food records) were assessed. The groups were compared with ANOVA with post-hoc Tukey's HSD test. Differentially expressed genes (p ,0.05) were filtered for at least 2.0-foldchange. Results: Compared to HC, patients with SS were older (45±8 vs. 38±11 y; mean±SD) and those with NASH had higher body mass index (31.2±3.3 vs. 26.2±3.8 kg/m2). We confirmed lower PC/PE ratio in liver tissue of SS (1.1±0.4) and NASH (1.4±0.6) compared to HC (3.1±0.5), due to both lower PC and higher PE in both patient groups. Genes involved in phospholipid metabolism were differentially expressed, mainly between NAFLD patients and controls. Compared to HC, phospholipase D6 (PLD6), which catalyzes the conversion of PC to choline and phosphatidate, was upregulated, whereas MTHFD2 and PLAG10 were downregulated in SS; MOGAT1 was upregulated in SS and MOGAT2, PLCD1, and SPHK2 were upregulated in both SS and NASH vs HC. In contrast, AGPAT9 was downregulated in SS and NASH vs HC. When NASH was compared with SS, only lipoprotein lipase (LPL) was upregulated in NASH vs. SS. Hepatic expression of PE-methyltransferase (PEMT), which converts PE to PC in the liver, was not different among the groups. Dietary intake of the essential nutrient choline was similar between groups but below adequate intake in all participants except for one HC subject. Neither PEMT expression nor choline intake was correlated with PC, PE, or PC/PE ratio. Conclusion: Lower PC/PE ratio in hepatic lipids of patients with NAFLD is associated with different gene expression of enzymes involved in phospholipid metabolism, when compared to HC, but not with altered hepatic PEMT expression or with differences in choline intake. These results identify specific genes related to phospholipid metabolism that are differentially expressed in patients with NAFLD.
Mo1005 Effect of Weight Loss on Novel MRI Quantitative Changes in Liver and Pancreatic Fat in Patients With Biopsy-Proven Nonalcoholic Fatty Liver Disease Iliana Doycheva, Mazen Noureddin, Niraj S. Patel, Michael R. Peterson, David A. Brenner, Claude Sirlin, Rohit Loomba Background: Weight loss has been associated with decrease in hepatic steatosis in patients with nonalcoholic fatty liver disease (NAFLD). Equal or more than 5% reduction in body weight has been linked to improvement in metabolic parameters. However, there are limited data regarding quantitative changes in liver fat and pancreatic fat content measured with a novel magnetic resonance imaging (MRI) biomarker, the proton-density-fat-fraction (PDFF), associated with weight loss in patients with biopsy-proven NAFLD. Aim: The aim of this study is to quantify the effect of weight loss on liver and pancreatic fat content using a novel imaging biomarker, the MRI-PDFF. Methods: This is a secondary analysis of a randomized controlled trial (RCT) assessing the effect of colesevelam versus placebo. We included 43 patients with biopsy-proven NAFLD who were followed for 24 weeks with available MRIPDFF liver and pancreatic fat quantification, anthropometric and biochemical profile at baseline and at 24 weeks. Given that 5% of weight loss has been utilized in previous large studies (Look AHEAD) as key endpoint, we divided the patients into those who had ≥ 5% reduction in BMI versus those who did not between week 0 and week 24. Results: Among the 43 patients with paired liver and pancreatic fat MRI data before and after 24 weeks in this RCT, 33 patients lost less than 5% or gained weight (gp A) and 10 patients lost 5% or
Mo1003 Novel Serum Protein Signature Associated With Non-Alcoholic Fatty Liver Disease Malti P. Nikrad, Stuart G. Field, Robert Mehler, Stephen Williams, Christopher D. Still, George Argyropoulos, Xin Chu, Glenn Gerhard Non alcoholic fatty liver disease (NAFLD) - the accumulation of hepatic fat without significant alcohol consumption - has become a worldwide epidemic. In the US NAFLD is present in 20-40% of the population, and nonalcoholic steatohepatitis (NASH) is present in about 25% of the obese population. 10-29% of the NASH patients develop cirrhosis, and 4-27% of those develop hepatic cancer. The major risk factors for NAFLD are visceral obesity, type 2 diabetes mellitus, dyslipidemia, and metabolic syndrome. NAFLD may be associated with elevated blood levels of the liver enzymes AST and ALT. The current gold standard to
AASLD Abstracts
S-1010
AASLD Abstracts
Background/Aims: Hepatic steatosis is a frequent incidental finding on abdominal CT scans. Primary care providers (PCPs) commonly perform the evaluation of steatosis however limited data exist regarding current PCP decision-making practice and outcomes in this area. We aimed to characterize and identify variables that predict PCP identification of steatosis. Methods: From January 2008 to October 2011, we identified all patients who underwent abdominal CT scans for evaluation of hematuria or nephrolithiasis at the Massachusetts General Hospital (MGH). We subsequently limited our population to patients with radiographically identified hepatic steatosis. Patient characteristics are described in Table 1. Patient care notes were reviewed for diagnosis of steatosis, alcohol use, medications, and diagnosis of hypertension, hyperlipidemia, and diabetes. Lab values were examined at two time points (Table 2). Baseline data consisted of most recent lab values prior to imaging and evaluation data consisted of the lab values within 14 months after imaging was performed. Lastly, we analyzed radiology reports and recorded whether hepatic steatosis was commented on in the body or the impression. Results: A total of 356 patients with reported steatosis on CT were identified. Patients without MGH PCPs (n=125) or a PCP visit within 14 months of image date (n=10), with prior known liver disease (n=69), prior organ transplant (n=4), or diagnosis of metastatic cancer (n=21) were excluded leaving 127 patients in the study. In the 14 months after identification, steatosis was commented on in PCP follow up in only 23% of patients. HCV screening was performed in 6.3% of patients and 59% underwent insulin resistance screening. Over 80% of patients underwent LFT and lipid screening. Of the 127 patients, the majority of radiology reports commented on fatty liver in the impression vs. body (83 vs. 44). There was a significant difference in the rate of follow up when radiology reports commented on fatty liver in the impression vs. the body (30.1% vs. 9.1%, p = 0.007). Retrospective application of NAFLD fibrosis score to prior data identified 14 patients at high risk of hepatic fibrosis, however this was not predictive of PCP documentation of fatty liver (p=0.51). New cases of insulin resistance were identified in 36% of patients (12% diabetes, 24% pre-diabetes) who underwent screening within 14 months of imaging. Conclusions: We concluded that PCPs infrequently document and follow up steatosis reported on imaging studies. Radiology reporting practices significantly impact PCP documentation rates, and our data suggest that steatosis should be recorded in impression section of reports. Efforts to improve PCP evaluation of steatosis should focus on increased evaluation of alcohol use, insulin resistance, HCV infection, and assessment for advanced fibrosis. Table 2: Laboratory Data Among Patients With Incidentally Identified Hepatic Steatosis
Baseline data refers to prior PCP documentation of presence of hypertension, hyperlipidemia, or diabetes, or active prescription of medication to treat these conditions at time of CT scan. Presence of abnormal LFTs refers to ALT .55 for men and .33 for women, AST .40 for men and .32 for women. Alcohol screening refers to specific quantity documented by PCPs. "Social/occasional" and "moderate" reflect patients where this was language used to characterize alcohol use by PCP in care notes. Mo1001 Prognostic Relevance of Serum Ferritin in Patients With Nonalcoholic Fatty Liver Disease Njideka Momah, Lee Russell, Elisabetta Bugianesi, Jacob George, Paul Angulo
Abnormal values refer to: ALT .55 for men and .33 for women, AST .40 for men and .32 for women, HgbA1c .5.7%, Fasting glucose .100, Total cholesterol .200, HDL ,40 for men and ,50 for women, Triglyceride .150, Ferritin .200, Plt count ,150, and HCV Ab positive. % of patients screened indicates percentage of patients with available lab value within pre-specified time period. Table 1: Baseline Characteristics of Study Population
S-1009
Serum ferritin (SF) levels are often elevated in patients with nonalcoholic fatty liver disease (NAFLD). Some cross-sectional studies have reported a relationship between elevated SF in patients with NAFLD and more severe liver fibrosis. However, since all studies reported to date are cross-sectional studies with patients evaluated at a single point in time, the longterm prognostic significance of elevated serum ferritin levels in patients with NAFLD remains unclear. The aim of our study was to determine the association between elevated SF levels and the long-term prognosis of patients with NAFLD. Methods: This is a longitudinal, longterm follow-up study of a cohort of 307 patients with liver biopsy confirmed NAFLD. These patients were followed-up for a median of 8.5 years (range 0.5 - 26.4 years). Extensive clinical and laboratory data were collected at baseline and during follow-up. The upper limit of normal (ULN) for SF used for comparisons was adopted from the hemochromatosis and iron overload screening study [i.e., ,300 μg/L for men and ,200 μg/L for women]. Patients with various degrees of elevation of SF were compared to patients with lower levels using three different cut-off points of SF selected a priori: . ULN, . 1.5 x ULN, and . 2 x ULN. Cumulative liver-related events, non-liver related events, and overall mortality were the outcomes analyzed and compared in the three different groups using Kaplan-Meier analysis. Hazard rate ratio (HR) estimates for outcomes were calculated by the Cox proportional hazard regression analysis to control for the effect of potential confounding factors while taking into consideration different duration of follow-up. Results: The mean (±SE) age was 50±0.6 years, 142 (46%) were men, 211 (69%) were obese, 109 (35%) suffered from diabetes, and 280 (91%) were of white race. SF was .ULN in 89 (29%), .1.5 x ULN in 54 (18%), and .2 x ULN in 33 (11%) patients. During follow-up, 35 (11%) patients developed one or more liver events, 64 (21%) patients developed one or more non-liver events, and 39 (12.7%) patients died or underwent liver transplantation. There was no significant difference in the proportion of cumulative outcomes of liver-events, non-liver events or mortality/liver transplantation in any of the three groups in the unadjusted (KaplanMeir) and adjusted (Cox Regression) analyses as described in the Table. Conclusions: Elevated SF levels are not associated with a higher risk of developing liver-related events, nonliver related events, or mortality or liver transplantation in patients with nonalcoholic fatty liver disease. Cumulative probability (Unadjusted) and multivariate-adjusted hazard ratios and 95% CI's for outcome by serum ferritin status
AASLD Abstracts
AASLD Abstracts
confirm the presence of NASH is a liver biopsy, which is expensive, invasive and subject to sampling error. A noninvasive blood test to detect hepatic fat and fibrosis to replace biopsy and aid in clinical decision-making would represent a significant advance. The goals of this study were to develop a serum proteomic signature to: 1. Detect and monitor severity of steatosis 2. Detect and monitor severity of fibrosis (NASH) Methods: Serum samples were collected and liver biopsies were performed on 443 obese patients who underwent bariatric surgery for weight loss at the Geisinger Clinic. Of these, 125 had normal biopsies, 137 showed varying degrees of steatosis (mild; moderate; severe), and 181 showed various grades of NASH fibrosis (1-4). Biomarker discovery was performed using SomaLogic's highly multiplexed SOMAscan™ proteomic assay, which measured 1129 proteins simultaneously from ~65μL serum. Demographics such as age, BMI and LDL cholesterol levels were balanced across all groups. Candidate markers were selected using a stability selection algorithm with a L1-regularized logistic regression kernel. Results: Two random forest classifiers were developed. The first, which was designed to detect liver steatosis, compared controls versus all other groups and yielded an area under the ROC curve (AUC) of 0.90±0.03 (sensitivity 92%; specificity 63%; @ 0.5 cutoff). Based on this 9-protein panel, all 7 NAFLD groups (steatosis and NASH) had elevated probabilities of steatosis (classifier votes). The second considered only groups with hepatic steatosis and distinguished fibrosis (NASH grades 24) from liver without fibrosis (steatosis). This 4-protein classifier had an AUC of 0.83±0.07 (sensitivity 62%; specificity 91%; @ 0.5 cutoff). When applied to NASH grade1 subjects, the average probability of fibrosis vote fell between the average severe steatosis and NASH grade 2 votes. Conclusion: In this initial study we have discovered a set of serum protein biomarkers that are highly predictive of fatty and fibrotic liver. The markers were associated with metabolism, matrix remodelling, apoptosis, and inflammation. These results will be further validated on a larger sample set. Mo1004 a Estimated using Kaplan-Meier analysis. b Estimated using Cox proportional hazard regression analysis. c Adjusted for age, gender, race, body mass index, diabetes, and fibrosis stage CI, confidence intervals
Among Nonalcoholic Fatty Liver Disease Patients Treated With Phlebotomy Does PNPLA3 Genotype Influence Changes in Gene Expression, Adipokines and Liver Histology? Melanie D. Beaton, Kelly Summers, Sarah Woolsey, Yun-Hee Choi, Rommel G. Tirona
Mo1002
Purpose: Variation in PNPLA3 (rs738409, C/G) has been associated with increased NALFD severity, but whether this genotype influences the outcome of NAFLD therapy is unknown. We sought to characterize the impact of this genotype on histology, markers of inflammation, fibrosis, angiogenesis, metabolic dysregulation, oxidative stress and serum adipokines in NAFLD patients treated with phlebotomy with paired liver biopsies. Methods: Thirty-one patients underwent phlebotomy for iron depletion (serum ferritin ≤50 ng/ml or Hgb 10.0 g/ dL). All were genotyped for PNPLA3 and repeat liver biopsy, anthropometric and biochemical measurements obtained following end of treatment. The primary outcome, improvement in NAFLD Activity Score (NAS) was achieved in 38.7%. We evaluated the relationship between PNPLA3 genotype and changes in liver histology, hepatic gene expression of hepcidin, hemeoxygenase-1, connective tissue growth factor, CD34, monocyte chemotactic protein-1, tumor necrosis factor α, PNPLA3 and phosphoenolpyruvate carboxykinase measured by real-time polymerase chain reaction, as well as, serum levels of adiponectin, c-peptide, leptin, resistin and visafatin analyzed by multiplex immunoassay. Results: Pre-phlebotomy, PNPLA3 G allele carriers (C/G and G/G) had higher NAS (p=0.007) and ballooning (p=0.01) but not steatosis (p=0.06), inflammation (p=0.07) or fibrosis (p=0.64) compared with C/C homozygotes. Following therapy, decreased NAS (p=0.01) was seen in PNPLA3 G allele carriers. Phlebotomy did not affect histology in C/C homozygotes. Steatosis (p=0.09), inflammation (p=0.11), ballooning (p=0.06) nor fibrosis (p=0.45) were significantly different between carriers and C/C homozygotes. Phlebotomy significantly reduced liver hepcidin (p=0.01) and heme oxygenase 1 gene expression (p=0.02), a measure of oxidative stress. Treatment did not affect expression of other genes examined. Phlebotomy was significantly associated with increases in leptin (p=0.03) but had no effect on other adipokines. PNPLA3 G allele carriers had significantly decreased heme oxygenase 1 expression in comparison to those not carrying the allele (p ,0.001). Pre and post therapy changes in resistin levels were significantly different between PNPLA3 G allele carriers and non-carriers (p=0.05). Conclusion: Possession of the PNPLA3 G allele is associated with improvement in liver histology with iron reduction therapy. Phlebotomy resulted in decreased liver oxidative stress and increased serum leptin, irrespective of PNPLA3 genotype. Determination of PNPLA3 genotype in individuals with NAFLD may be helpful in guiding therapy and predicting response to treatment for NALFD.
In Non-Alcoholic Fatty Liver Disease, Differential Gene Expression for Phospholipid Metabolism Is Associated With Lower Hepatic Phosphatidylcholine to Phosphatidylethanolamine Ratio, but Not With Lower Choline Intake Bianca M. Arendt, Ian D. McGilvray, Elena Comelli, David W. Ma, Scott Fung, AnaLaura Guzman, Sandra E. Fischer, Johane P. Allard Background: It has been suggested that a lower hepatic ratio of phosphatidylcholine (PC) to phosphatidylethanolamine (PE) may contribute to the pathogenesis of non-alcoholic fatty liver disease (NAFLD), as changes in the phospholipid membrane can affect cell integrity and signal transduction. Lower PC/PE ratio could be due to altered expression of enzymes involved in phospholipid metabolism and/or inadequate choline intake. The aim of this study was to compare hepatic gene expression, choline intake, and hepatic PC/PE ratio between patients with simple steatosis (SS), non-alcoholic steatohepatitis (NASH) and healthy controls (HC). Methods: Hepatic gene expression profiles (Illumina Whole-Genome DASL HT-assay) were determined from liver biopsies taken from NAFLD patients (SS: n= 24, NASH: n=17) or as the first step of a live donor hepatectomy (HC: n=23). In a subgroup of participants, hepatic PC and PE (mass spectrometry) as well as dietary choline intake (7day food records) were assessed. The groups were compared with ANOVA with post-hoc Tukey's HSD test. Differentially expressed genes (p ,0.05) were filtered for at least 2.0-foldchange. Results: Compared to HC, patients with SS were older (45±8 vs. 38±11 y; mean±SD) and those with NASH had higher body mass index (31.2±3.3 vs. 26.2±3.8 kg/m2). We confirmed lower PC/PE ratio in liver tissue of SS (1.1±0.4) and NASH (1.4±0.6) compared to HC (3.1±0.5), due to both lower PC and higher PE in both patient groups. Genes involved in phospholipid metabolism were differentially expressed, mainly between NAFLD patients and controls. Compared to HC, phospholipase D6 (PLD6), which catalyzes the conversion of PC to choline and phosphatidate, was upregulated, whereas MTHFD2 and PLAG10 were downregulated in SS; MOGAT1 was upregulated in SS and MOGAT2, PLCD1, and SPHK2 were upregulated in both SS and NASH vs HC. In contrast, AGPAT9 was downregulated in SS and NASH vs HC. When NASH was compared with SS, only lipoprotein lipase (LPL) was upregulated in NASH vs. SS. Hepatic expression of PE-methyltransferase (PEMT), which converts PE to PC in the liver, was not different among the groups. Dietary intake of the essential nutrient choline was similar between groups but below adequate intake in all participants except for one HC subject. Neither PEMT expression nor choline intake was correlated with PC, PE, or PC/PE ratio. Conclusion: Lower PC/PE ratio in hepatic lipids of patients with NAFLD is associated with different gene expression of enzymes involved in phospholipid metabolism, when compared to HC, but not with altered hepatic PEMT expression or with differences in choline intake. These results identify specific genes related to phospholipid metabolism that are differentially expressed in patients with NAFLD.
Mo1005 Effect of Weight Loss on Novel MRI Quantitative Changes in Liver and Pancreatic Fat in Patients With Biopsy-Proven Nonalcoholic Fatty Liver Disease Iliana Doycheva, Mazen Noureddin, Niraj S. Patel, Michael R. Peterson, David A. Brenner, Claude Sirlin, Rohit Loomba Background: Weight loss has been associated with decrease in hepatic steatosis in patients with nonalcoholic fatty liver disease (NAFLD). Equal or more than 5% reduction in body weight has been linked to improvement in metabolic parameters. However, there are limited data regarding quantitative changes in liver fat and pancreatic fat content measured with a novel magnetic resonance imaging (MRI) biomarker, the proton-density-fat-fraction (PDFF), associated with weight loss in patients with biopsy-proven NAFLD. Aim: The aim of this study is to quantify the effect of weight loss on liver and pancreatic fat content using a novel imaging biomarker, the MRI-PDFF. Methods: This is a secondary analysis of a randomized controlled trial (RCT) assessing the effect of colesevelam versus placebo. We included 43 patients with biopsy-proven NAFLD who were followed for 24 weeks with available MRIPDFF liver and pancreatic fat quantification, anthropometric and biochemical profile at baseline and at 24 weeks. Given that 5% of weight loss has been utilized in previous large studies (Look AHEAD) as key endpoint, we divided the patients into those who had ≥ 5% reduction in BMI versus those who did not between week 0 and week 24. Results: Among the 43 patients with paired liver and pancreatic fat MRI data before and after 24 weeks in this RCT, 33 patients lost less than 5% or gained weight (gp A) and 10 patients lost 5% or
Mo1003 Novel Serum Protein Signature Associated With Non-Alcoholic Fatty Liver Disease Malti P. Nikrad, Stuart G. Field, Robert Mehler, Stephen Williams, Christopher D. Still, George Argyropoulos, Xin Chu, Glenn Gerhard Non alcoholic fatty liver disease (NAFLD) - the accumulation of hepatic fat without significant alcohol consumption - has become a worldwide epidemic. In the US NAFLD is present in 20-40% of the population, and nonalcoholic steatohepatitis (NASH) is present in about 25% of the obese population. 10-29% of the NASH patients develop cirrhosis, and 4-27% of those develop hepatic cancer. The major risk factors for NAFLD are visceral obesity, type 2 diabetes mellitus, dyslipidemia, and metabolic syndrome. NAFLD may be associated with elevated blood levels of the liver enzymes AST and ALT. The current gold standard to
AASLD Abstracts
S-1010