Society: AASLD

Background: Nonalcoholic steatohepatitis (NASH) is a major cause of liver-related morbidity, mortality, and the leading indication for liver transplant in the US. Fibrosis is the best predictor of clinical outcomes in NASH. Obeticholic acid (OCA), a first-in-class farnesoid X receptor agonist, demonstrated efficacy as an antifibrotic agent in the phase 3 REGENERATE trial in NASH. The goal of the new analysis was to confirm the original 18-month biopsy efficacy results and to provide additional safety data from more than 8000 total subject-years, with nearly 1000 subjects receiving OCA for ≥4 years in the REGENERATE study.

Methods: In this multicenter, randomized, double-blind, placebo-controlled phase 3 study, subjects were randomized 1:1:1 to receive once-daily oral placebo, OCA 10 mg, or OCA 25 mg. Safety data were evaluated from 2477 subjects who received ≥1 dose of the study drug. A consensus method using a panel of 3 pathologists for histologic assessment of liver biopsies per NASH Clinical Research Network criteria was employed to confirm the results of the original 18-month interim efficacy analysis performed by individual readers.

Results: Treatment-emergent adverse events (TEAEs), serious TEAEs, and deaths were balanced across treatment groups (N=2477). Pruritus was the most common TEAE in all 3 treatment groups; its incidence was higher in both OCA groups compared to placebo. For adverse events of special interest, incidence of pancreatitis and hepatic disorders was similar between the OCA and placebo groups. The incidence of gallbladder- and gallstone-related TEAEs (including serious events) was higher in both OCA groups compared to placebo, though most were mild to moderate in severity and managed with temporary drug interruption and usual clinical care (Table 1). In the new analysis of the intent-to-treat population (n=931) using a consensus read method, 22.4% of subjects receiving OCA 25 mg experienced ≥1 stage improvement of fibrosis with no worsening of NASH at month 18 vs 9.6% of subjects receiving placebo (primary endpoint), which was consistent with the original analysis performed by individual readers (Figure 1).

Conclusions: In the ongoing REGENERATE trial, subjects receiving OCA 25 mg were twice as likely to experience ≥1 stage improvement in liver fibrosis with no worsening of NASH at month 18 compared to subjects receiving placebo at 18 months using 2 different histologic methodologies. The confirmed antifibrotic effect in liver histology, together with extended exposure within the largest safety database in NASH to date, demonstrates that OCA 25 mg was generally safe and well tolerated, supporting the long-term use of OCA to treat pre-cirrhotic fibrosis due to NASH.

Background: Transmembrane 6 superfamily member 2 (TM6SF2) is predominately expressed in liver and small intestine, and its loss-of-function variant is linked to increased risk of non-alcoholic fatty liver disease (NAFLD). We aim to investigate the role and mechanism of intestinal TM6SF2 dysfunction in the pathogenesis non-alcoholic steatohepatitis (NASH).
Methods: Mice with systematic, liver-specific, or intestine-specific Tm6sf2 knockout (Tm6sf2^ΔIEC) were fed with normal chow, high fat high cholesterol (HFHC), or choline-deficient high fat diet (CD-HFD) for 2 months. Gut microbiota and metabolites were profiled by metagenomic and metabolomic analysis. To determine the role of gut microbiota of Tm6sf2^ΔIEC mice, fecal microbiota transplantation was performed in germ-free mice. Tm6sf2^ΔIEC mice were co-housed with wildtype (WT) mice to evaluate the therapeutic potential of microbiota manipulation.
Results: Hepatocyte-specific Tm6sf2 deletion in mice induced spontaneous hepatic steatosis. while systematic Tm6sf2 ablation showed pronounced hepatic steatosis with necroinflammation, implying that extra-hepatic Tm6sf2 deletion could also contribute to NASH progression. Given that intestine has high endogenous TM6SF2 expression, we established the Tm6sf2^ΔIEC mice. Tm6sf2^ΔIEC mice fed with normal chow spontaneously developed NASH with increased hepatic triglyceride, lipid peroxidation, enhanced pro-inflammatory cytokines and NF-κB. Intestinal Tm6sf2 knockout accelerated NASH development in mice fed HFHC or CD-HFD. The gut barrier in Tm6sf2^ΔIEC mice was impaired with increased permeability and abnormalities of intestinal junctions, suggesting that loss of intestinal TM6SF2 might deregulate the gut-liver axis. Indeed, integrative metagenomic and metabolomic analysis unveiled the enrichment of gut pathogenic bacteria in Tm6sf2^ΔIECmice that was correlated with increased lysophosphatidic acid (LPA) in the portal vein, leading to LPA accumulation in liver of Tm6sf2^ΔIEC mice. Moreover, LPA promoted lipid accumulation and pro-inflammatory cytokines secretion in hepatocytes. Intestine cells isolated from Tm6sf2^ΔIEC mice secreted more free fatty acids compared to the intestine cells isolated from WT mice, which could impair gut barrier in mice. Notably, transplanting stool from Tm6sf2^ΔIEC mice to germ-free mice induced steatohepatitis with microbial dysbiosis and increased LPA level in portal vein in recipient mice; whereas co-housing of Tm6sf2^ΔIEC mice with WT mice significantly restored gut barrier function, and ameliorated hepatic lipid accumulation and inflammation in Tm6sf2^ΔIEC mice.
Conclusion: Intestinal TM6SF2 deficiency induces gut dysbiosis and barrier dysfunction, leading to increased translocation LPA to liver to promote NASH progression. Microbiota manipulation may be an effective strategy for the prevention of NASH driven by down-regulation of TMFSF2.

Background: Patients with obesity are at increased risk of non-alcoholic steatohepatitis (NASH), which is associated with serious liver-related complications, cardiovascular (CV) disease, and mortality. Biopsy-confirmed liver fibrosis is a predictor of severe outcomes, but liver biopsies are not scalable. This real-world study investigated the prognostic potential of six non-invasive fibrosis scores (liver stiffness measurement, enhanced liver fibrosis score, fibrosis-4 index [FIB4], nonalcoholic fatty liver disease fibrosis score, fibrometer V2G and fibrometer V3G) on clinical outcomes in people with obesity; the score of focus in this abstract is FIB4.
Methods: In this longitudinal cohort study, adults with obesity, ≥ 1 fibrosis score calculable from the UK Clinical Practice Research Datalink after Jan 1, 2001, no alcohol-related disorders and/or other chronic liver diseases in Hospital Episodes Statistics, and/or no prescriptions of drugs inducing liver disease were included. Patients were followed from inclusion date until time of first event (liver-related hospitalization or death [liver event], CV hospitalization or death [CV event], or all-cause death), database migration, 10 years follow-up, or Jan 1, 2020, whichever came first. FIB4 was grouped into low (<1.30), indeterminate (1.30–2.67), or high (>2.67) risk. Cumulative incidence functions were calculated. Hazard ratios (HRs) were estimated using Cox proportional hazard models with calendar time as timescale.
Results: There were 715 liver events among 35,348 eligible patients (44% male, median age 55.8 years) with obesity with a FIB4 measure. Cumulative incidence proportions after 10 years follow-up for an event in the high FIB4 group were 17% (liver), 32% (CV), and 53% (death). HRs (95% confidence intervals) for a liver event for the indeterminate and high vs low FIB4 group were 3.32 (2.80–3.93) and 21.29 (17.58–25.77), respectively. HRs remained significantly higher for high vs low FIB4 after adjustment for sex and age. Higher risks were also seen for CV events and death and for all endpoints using the other five scores.
Conclusion: High FIB4 score is associated with significantly higher risk of severe clinical events in patients with obesity.

*Presenting author (added only for presentation purposes): Joshua Neuman

Introduction
Weight gain among liver transplant (LT) recipients is common and associated with an increased risk of cardiovascular disease and reduced survival. Despite, the oversized impact of weight gain in LT recipients, strategies aimed at weight loss are not successful, underscoring the need to better understand the physiology of weight gain among LT recipients. Metabolic flexibility is the ability to match biofuel availability to utilization and is inversely associated with metabolic burden among LT recipients. The present study evaluated the impact of metabolic flexibility on the likelihood of post-LT weight gain.

Methods
LT recipients were enrolled prospectively (n=47) and followed for 6 months. Metabolic flexibility was measured using whole-room calorimetry and expressed as a respiratory quotient (RQ). Peak RQ represents maximal carbohydrate metabolism and occurs in the post-prandial state. In contrast, trough RQ represents maximal fatty acid metabolism occurring in the fasted state and each component of metabolic flexibility has been annotated in Figure 1A. All patients received standard-of-care advice for weight loss as recommended by societal guidelines. Patients on weight loss medications such as GLP-1 receptor agonists were excluded.

Results
The clinical, metabolic, and laboratory characteristics of the study cohort of lost weight (n=14) and gained weight (n=33) were similar at baseline including body mass index (BMI). Patients who lost weight were more likely to reach maximal RQ peak early and rapidly transition to trough RQ (Figure 1B). In contrast, patients who gained weight had delayed time to peak RQ and trough RQ. In multivariate modeling, time to peak RQ (β-coefficient 0.509, p=0.01), time from peak RQ to trough RQ (β-coefficient 0.634, p=0.006), and interaction between time to peak RQ to trough RQ and fasting RQ (β-coefficient 0.447, p=0.02) directly correlated with the severity of weight gain. An inverse relationship between future weight gain and the interaction between time to peak RQ and the difference between peak RQ and trough RQ (β-coefficient - 0.262, p=0.05) was also demonstrated. No statistically significant relationship between peak RQ, trough RQ and weight change was demonstrated.

Conclusion
The inefficient transition between biofuel (carbohydrates and fatty acids) is associated with weight gain in LT recipients that is independent of clinical metabolic risk. These data offer novel insight into the physiology of obesity post-LT with the potential to develop new diagnostics and therapeutics.

Background:
The primary treatment recommendation for patients with non-alcoholic fatty liver disease (NAFLD) is weight loss. Both endoscopic sleeve gastroplasty (ESG) and the incretin mimetic semaglutide are effective in sustaining weight loss. Herein, we investigate and compare the effects of ESG and semaglutide on patients with NAFLD.

Methods:
Retrospective data were collected from patients prospectively recruited to the Innovative Center for Health and Nutrition (ICHANGE) Program at Weill Cornell Medicine. Patients in ESG and semaglutide treatment groups were matched by age and gender. The primary outcomes were percent total body weight loss (%TBWL) and change in NAFLD as assessed by weight loss, Hepatic Steatosis Index (HSI), and NAFLD fibrosis score.

Results:
Our study included 33 matched patients with a mean age of 54.4 ± 13.9 years and mean baseline BMI of 38.2 ± 6.8 kg/m²for those in the ESG cohort, and mean age of 54.0 ± 14.1 years and mean baseline BMI of 36.6 ± 8.2 kg/m²for those in the semaglutide cohort. In multivariable analysis controlling for age, sex, and baseline BMI, mean %TBWL at 3, 6, and 12 months was significantly higher in patients undergoing ESG than those receiving semaglutide (p<0.05, p<0.05, and p<0.05, respectively). Though a similar trend was observed at the 24 month timepoint, there was no significant difference in %TBWL between groups. The ESG group demonstrated significant improvement in HSI at the 6, 12, and 24 month timepoints (p<0.001, p<0.001, and p<0.01, respectively). In the semaglutide group, significant improvement in HSI was observed at 6 months (p<0.05) and 12 months (p<0.01), but not at 24 months. There was significant improvement in NAFLD fibrosis score at 12 months (p<0.05) and 24 months (p<0.01) for the ESG group, while the semaglutide group showed no significant improvement of NAFLD fibrosis score at any timepoint. Although there was no significant difference when comparing treatment groups, there were trends of worse NAFLD fibrosis score at the 6 and 12 month timepoints in the semaglutide group. At all timepoints, the ESG group trended toward more improvement in HSI compared to the semaglutide group despite not reaching statistical significance.

Conclusion:
Treatment with ESG may be superior to semaglutide therapy for patients with NAFLD due to a greater reduction in %TBWL at 3, 6, and 12 months, with similar trends shown at later time points. Trends showed relatively greater improvement in HIS with ESG treatment when compared to semaglutide. Similarly, improvement in NAFLD fibrosis score at 12 months and 24 months was significant only for the ESG group. Though this study does not demonstrate significant between-group differences in NAFLD fibrosis score or HSI, our preliminary findings can guide future assessment of weight loss modalities to improve clinicians’ recommendations for patients.

Sonal Kumar

Weill Cornell Medical College

Appearances