Conclusion: The Endosleeve offers promising preliminary results at 6 months. It is safe, reduces TEW in a media of 53% ± 20% and is associated with amelioration of fibrosis, HE and dyslipidemia. Long- term follow-up is necessary. FRI107 Use of GLP-1 receptor agonists (GLP1A) and/or SGLT-2 inhibitors (SGLT2I) in populations with NASH or at risk of NASH in US clinical practice Michelle Lai 1 , Jeremy Broestl 2 , Kavita Juneja 3 , Scott Milligan 2 , Janna Radtchenko 2 , Zobair Younossi 4,5 , Nezam Afdhal 1 . 1 Beth Israel Deaconess Medical Center (BIDMC), Boston, United States; 2 Trio Health Analytics, San Diego, United States; 3 Gilead Sciences, Inc., Foster City, United States; 4 Inova Fairfax Hospital, Center for Liver Diseases, Department of Medicine, Falls Church, United States; 5 Inova Health System, Betty and Guy Beatty Center for Integrated Research, Falls Church, United States Email: scott.milligan@triohealth.com Background and Aims: Currently, no drugs are FDA approved for the treatment of NASH. However, some success has been observed with GLP1a in histological resolution of NASH without worsening of fibrosis and with SGLT2i in reduction of body weight, ALT, and FIB4 compared with baseline values. Here, we examined data from type II diabetic (T2DM) and non-T2DM patients with or at risk of NASH to determine use of these therapies in clinical practice. Method: Data collected from a proprietary US EMR database were limited to adult patients with diagnosed NAFLD without NASH (ICD10 K76 and ICD9 equivalent without evidence of NASH), NASH (ICD10 K75.81) or with a risk profile (RISK) of Age >50 + ALT >30 U/L + [BMI >30 or T2DM] and without viral hepatitis or evidence of alcohol abuse. Index date was assigned at the first calculable FIB4 between Jul 2015 to Jun 2017 and for which >1 year history and >2 years follow-up or to death was available. Results: Of 30MM adult patients, 81108 met all study criteria with 22% (17582) NAFLD without NASH,1% (914) NASH, and 77% (62612) RISK (without NAFLD or NASH) [Figure]. At index, 50% (40194/81095) male, 84% (65121/77635) white, 73% BMI >30 (44190/60882), 43% (34730) T2DM,11% (9210) FIB4 ≥ 2.67, and median (IQR) age 60 (55– 65). Use of SGLT2i and GLP1a at index was 2% and 3% respectively overall, but higher in the T2DM subset (4% and 7%) compared to non- T2DM (<1%, p <0.001). Among 105 non-T2DM on SGLT2i, 54% received canagliflozin, 41% dapagliflozin, and 35% empagliflozin. Among 188 non-T2DM on GLP1a at index, 64% received liraglutide, 30% dulaglutide, and 22% exenatide. Non-T2DM patients receiving SGLT2i and/or GLP1a (n = 279) differed from those not on these therapies (n = 46099) in age (mean 57.0 v. 60.5 not treated (NT), p < 0.001), Charlson CI (0.9 v. 1.7 NT, p <0.001), AST>30 IU/L (43% v. 36% NT, p =0.014), blood glucose ≥ 1.1 g/dL (66% v. 27% NT, p < 0.001), HbA1c ≥ 5.6% (91% v. 72% NT, p < 0.001), triglycerides ≥ 150 mg/L (35% v. 25% NT, p < 0.001), LDL cholesterol ≥ 100 mg/dL (22% v. 33% NT, p < 0.001), and eGFR ≥ 90 (41% vs. 31% NT, p =0.001). Study Paents 81,108 NASH 914 (1%) Diabec 457 (50%) On SGLT2i 22 (5%) On GLP1a 30 (7%) No SGLT2i/GLP1a 412 (90%) Non-diabec 457 (50%) On SGLT2i 1 (<1%) On GLP1a 2 (<1%) No SGLT2i/GLP1a 454 (99%) NAFLD without NASH 17,582 (22%) Diabec 6920 (39%) On SGLT2i 350 (5%) On GLP1a 591 (9%) No SGLT2i/GLP1 6103 (88%) Non-diabec 10662 (61%) On SGLT2i 15 (<1%) On GLP1a 47 (<1%) No SGLT2i/GLP1a 10605 (99%) RISK 62,612 (77%) Diabec 27353 (44%) On SGLT2i 953 (3%) On GLP1a 1660 (6%) No SGLT2i/GLP1a 25008 (91%) Non-diabec 35259 (56%) On SGLT2i 89 (<1%) On GLP1a 139 (<1%) No SGLT2i/GLP1a 35040 (99%) Figure: Disposition of the study patients. Conclusion: Despite evidence suggesting benefits with GLP1a and SGLT2i in patients with NASH or risk of NASH, real-world use of these therapies in patients with T2DM was relatively lowat 4–7% despite higher rates of insurance drug coverage for this population. Use in non-T2DM patients was <1% overall and more common in younger patients with fewer comorbidities and with lab measures similar to T2DM patients. FRI108 BIO89-100, a novel glycopegylated FGF21 analog, reduces body weight and fat mass in naive CD-1 mice through an increase in energy expenditure despite an increase in food consumption Moti Rosenstock 1 , Thais Rouquet 2 , Bruno Bariohay 2 , Hank Mansbach 1 , Maya Margalit 1 . 1 89bio Ltd.; 2 Biomeostasis Email: moti.rosenstock@89bio.com Background and Aims: BIO89-100, a novel glycoPEGylated analog of FGF21, is being developed for the treatment of nonalcoholic steatohe- patitis (NASH) and cardiometabolic diseases. In CD-1 mice, BIO89-100 reduces body weight (BW) despite increased food consumption (FC). To better understand these findings, we investigated the effect of BIO89-100 on energy expenditure and body composition. Methods: Naïve mice were treated subcutaneously (SC) for 4 weeks at ambient temp. with BIO89-100 at doses of 0, 0.3, 1.0 and 3.0 mg/kg (3qW × 4 weeks) or with liraglutide at 0.2 mg/kg (BID × 4W; SC). BW and FC were measured daily throughout the treatment period. To investigate the impact of BIO89-100 on energy expenditure, the respiratory exchanges (VO 2 and VCO 2 ) were recorded during 3 distinct sessions of METABOpack TM , a multiparametric approach allowing an automated and simultaneous recording of FC, water intake, respiratory exchanges and spontaneous activity. Meal pattern analysis was conducted on the high-resolution recordings of FC. The body composition (fat, lean and fluid masses) was analyzed 3 times during the study by the mean of Time-Domain Nuclear Magnetic Resonance (TD-NMR). Results: Dose-dependent decrease in BW (up to -13.8%) and increase in FC (up to +94%) were observed in BIO89-100-treated mice. Increased FC was related to an increase in mean meal number, with no change in mean meal size. There was no evidence of change in basal metabolism or total spontaneous activity in studied animals, thus the increase in energy expenditure is presumed to be through increased thermogenesis. BIO89-100-treated mice had a marked decrease in fat mass (measured both by TD-NMR during the study and by weighing the epididymal white adipose tissue (eWAT) at sacrifice), without affecting lean masses and body fluid. Liraglutide induced a significant decrease in BW when compared with controls resulting in a final 10% reduction, associated with a decrease in fat mass. However, liraglutide did not affect the energyexpenditure or food consumption. 0 1 2 3 4 5 6 7 8 9 10 Fat mass (g) Energy Expenditure ( (kcal/day/kg lean)) 0 100 200 300 400 500 600 700 DP NP WD Diurnal Period Nocturnal Period Whole Day * * # * # * # * * ** ** Mean ± SEM. * p < 0.05, significantly different vs. Vehicle group; # p < 0.05, significantly different vs. Liraglutide group. © 2019 89bio LTD Conclusion: Naïve CD-1 mice treated with BIO89-100 had a dose- dependent reduction in BWand increased FC. Our results suggest that this BW loss is due, at least in part, to an increase in energy expenditure, resulting in a marked decrease in fat mass, including the eWATweight, without affecting lean masses and body fluid. If these effects translate to humans, they could be beneficial in patients with metabolic diseases. POSTER PRESENTATIONS S452 Journal of Hepatology 2020 vol. 73 | S401–S652