Hepatology

Background & AimsMetabolic dysfunction-associated steatotic liver disease (MASLD), formerly known as NAFLD, is a leading cause of chronic liver disease worldwide. Current diagnostic methods, including liver biopsies, are invasive and have significant limitations, emphasizing the need for non-invasive alternatives. This study aimed to evaluate extracellular vesicles (EV) as biomarkers for diagnosing and staging steatosis in MASLD patients, utilizing machine learning (ML) and explainable artificial intelligence (XAI). MethodsThis prospective, single-center cohort study was conducted at the GI-Liver Unit, Hippocration General Hospital, Athens. It included 76 MASLD patients with ultrasound-confirmed steatosis and at least one cardiometabolic risk factor. Patients underwent transient elastography for steatosis and fibrosis staging and blood sampling for EV analysis using nanoparticle tracking. Twenty machine learning models were developed. Six to distinguish non-steatosis (S0) from steatosis (S1-S3), and fourteen to identify severe steatosis (S3). Models incorporated EV measurements (size and concentration), anthropomorphic and clinical features, with performance evaluated using AUROC and SHAP-based interpretability methods. ResultsThe CB-C1a model achieved, on average on 10 random splits of 5-fold cross validation (5CV) of the train set, an AUROC of 0.71/0.86 (train/test) for distinguishing S0 from S1-S3 steatosis stages, relying on EV alone. The CB-C2h-21 model identified severe steatosis (S3), on average on 10 random splits of 3-fold cross validation (3CV) of the train set, with an AUROC of 0.81/1.00 (train/test), demonstrating superior performance when combining EV with anthropomorphic and clinical features such as diabetes and advanced fibrosis. Key EV features, including mean size and concentration, were identified as important predictors. SHAP analysis highlighted complex non-linear relationships between features and steatosis staging. ConclusionsEV are promising non-invasive biomarkers for diagnosing and staging MASLD. The integration of ML-enhanced EV analysis with clinical features offers a scalable, patient-friendly alternative to invasive liver biopsies, advancing precision in MASLD management. Further research should refine these methods for broader clinical application. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=99 SRC="FIGDIR/small/24318644v1_ufig1.gif" ALT="Figure 1"> View larger version (36K): org.highwire.dtl.DTLVardef@128d63borg.highwire.dtl.DTLVardef@8ed52org.highwire.dtl.DTLVardef@14c356org.highwire.dtl.DTLVardef@1244845_HPS_FORMAT_FIGEXP M_FIG C_FIG

Background & AimsMetabolic dysfunction-associated steatotic liver disease (MASLD), formerly known as NAFLD, is a leading cause of chronic liver disease worldwide. Current diagnostic methods, including liver biopsies, are invasive and have significant limitations, emphasizing the need for non-invasive alternatives. This study aimed to evaluate extracellular vesicles (EV) as biomarkers for diagnosing and staging steatosis in MASLD patients, utilizing machine learning (ML) and explainable artificial intelligence (XAI). MethodsThis prospective, single-center cohort study was conducted at the GI-Liver Unit, Hippocration General Hospital, Athens. It included 76 MASLD patients with ultrasound-confirmed steatosis and at least one cardiometabolic risk factor. Patients underwent transient elastography for steatosis and fibrosis staging and blood sampling for EV analysis using nanoparticle tracking. Twenty machine learning models were developed. Six to distinguish non-steatosis (S0) from steatosis (S1-S3), and fourteen to identify severe steatosis (S3). Models incorporated EV measurements (size and concentration), anthropomorphic and clinical features, with performance evaluated using AUROC and SHAP-based interpretability methods. ResultsThe CB-C1a model achieved, on average on 10 random splits of 5-fold cross validation (5CV) of the train set, an AUROC of 0.71/0.86 (train/test) for distinguishing S0 from S1-S3 steatosis stages, relying on EV metrics alone. The CB-C2h-21 model identified severe steatosis (S3), on average on 10 random splits of 3-fold cross validation (3CV) of the train set, with an AUROC of 0.81/1.00 (train/test), demonstrating superior performance when combining EV with anthropomorphic and clinical features such as diabetes and advanced fibrosis. Key EV features, including mean size and concentration, were identified as important predictors. SHAP analysis highlighted complex non-linear relationships between features and steatosis staging. ConclusionsEV metrics are promising non-invasive biomarkers for diagnosing and staging MASLD. The integration of ML-enhanced EV analysis with clinical features offers a scalable, patient-friendly alternative to invasive liver biopsies, advancing precision in MASLD management. Further research should refine these methods for broader clinical application. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=96 SRC="FIGDIR/small/24318233v1_ufig1.gif" ALT="Figure 1"> View larger version (27K): org.highwire.dtl.DTLVardef@84e6eforg.highwire.dtl.DTLVardef@15591b3org.highwire.dtl.DTLVardef@d086a2org.highwire.dtl.DTLVardef@3713d8_HPS_FORMAT_FIGEXP M_FIG Graphical Abstract C_FIG

Background and AimsThe newly proposed nomenclature for steatotic liver diseases (SLD) aims to reduce the stigma associated with "non-alcoholic fatty liver disease" (NAFLD), increase awareness, and provide a framework for delineating pathogenic pathways. Approach and ResultsWe projected the new nomenclatures diagnostic scheme onto National Health and Nutrition Examination Survey (NHANES) data and determined SLD prevalence, fibrosis risk factors, subtypes, and consistency with previous classifications. Steatosis grade and fibrosis stage were estimated from vibration controlled transient elastography (VCTE). At a threshold of 240 dB/m, 62.1% [95% confidence interval (CI), 59.8-64.3%] of adults ([≥] 20 years) and 30.5% (95% CI, 27.1-34.0%) of adolescents (12-19 years) had SLD. By American Gastroenterological Association criteria, 19.3 million (95% CI, 15.8-22.8) adults with SLD qualify for hepatology referral. Over 98% of adults but only 85% of adolescents with NAFLD met criteria for definite MASLD. Significant fibrosis ([≥] 8.6 kPa) occurred in 13.5 million (95% CI, 10.9-16.2) adults with MASLD; risk factors varied by race and ethnicity. Significant fibrosis occurred in over 1.5 million adults without any identified LD and was associated with lead (Pb) exposure, odds ratio = 3.89 (95% CI, 2.00-7.56). ConclusionsThe overarching term, SLD, changes the diagnostic algorithm and creates an umbrella classification that highlights the extraordinary prevalence of liver steatosis. The more precise nomenclature establishes a valuable patient-centric platform for research and clinical care, clarifying risk groups and risk factors, including adolescents with NAFLD but without definite MASLD and adults without SLD in whom toxic exposures may increase fibrosis risk.

Background & AimsIron corrected T1 (cT1) is an MRI derived biomarker of liver disease activity. Emerging data suggest a change in cT1 of [≥] 80 ms reflects histological improvement. We aimed to validate the association between the [≥] 80 ms decline in cT1 and histological improvement, specifically the resolution of MASH. MethodsA retrospective analysis of study participants from three interventional clinical trials with histologically confirmed MASH (n = 150) who underwent multi-parametric MRI to measure cT1 (LiverMultiScan(R)) and biopsies at baseline and end of study. Histological responders were defined using the four criteria: (1) a decrease in NAFLD Activity score (NAS) [≥] 2 with no worsening in fibrosis, (2) a decrease in fibrosis [≥] 1 stage with no worsening in NAS, (3) both a NAS decrease [≥] 2 and a fibrosis decrease [≥] 1, and (4) MASH resolution with no worsening in fibrosis. Difference in the magnitude of change in cT1 between responders and non-responders was assessed. ResultsSignificant decreases in cT1 were observed in responders for all the histological criteria. The largest decrease was observed for those achieving MASH resolution, and was 119ms, compared to 43ms for non-responders. The optimal reduction in cT1 for separating responders from non-responders for MASH resolution was -74ms (64ms-73ms for the other criteria), in close agreement with the previously predefined threshold of -80ms. Those achieving an [≥] 80 ms reduction in cT1 were substantially more likely to achieve histological response with odds ratios ranging from 2.7 to 6.3. ConclusionsThese results demonstrate that a reduction in cT1 of 80 ms was associated with histological response supporting the utility of cT1 to predict clinical improvement in patients undergoing therapeutic intervention.

Background and AimsChronic hepatitis C (CHC) can progress to cirrhosis and hepatocellular carcinoma (HCC). This study aimed to identify genetic determinants and host-viral interactions that drive this progression to inform risk stratification and personalised treatment strategies. MethodsWe performed a genome-wide association study (GWAS) of cirrhosis (2,829 cases and 1,515 CHC controls), followed by a GWAS of HCC (706 cases and 2,152 cirrhosis controls). We performed cis-eQTL mapping and deconvolution in liver tissue of HCV-infected (136 CHC and 54 cirrhosis) patients to investigate gene expression regulation and cellular heterogeneity. Additionally, ten polygenic risk scores (PRS) for non-viral liver diseases were tested in 3,406 infected individuals. ResultsWe identified the missense risk variant rs738409 in PNPLA3 and a protective variant (rs4386418) in XKR3 in genotype 1-infected patients that were significantly associated with cirrhosis but not HCC progression. HLA fine-mapping identified two amino acids in HLA-DQB1*03:01 and HLA-DRB1*13:01 associated with cirrhosis risk. No genome-wide significant association was observed for HCC, and loci previously linked to non-viral HCC did not replicate. The eQTL analysis revealed 2,060 genes under cis-regulatory control and 129 whose effects were modified by cirrhosis. An intronic eQTL lowered PNPLA3 expression, but was not linked to cirrhosis risk. Deconvolution revealed expansion of plasma cells and macrophages and depletion of hepatocytes in CHC, with further immune-stromal remodelling in cirrhosis. All PRS showed a significant association with cirrhosis risk but not HCC progression. ConclusionCirrhosis in CHC shares genetic architecture with non-viral liver diseases but also displays virus-specific risk variants. Cirrhosis risk involves genetic factors that differ from those underlying progression to HCC.

Recombinant adeno-associated viruses (rAAVs) are a platform of choice for gene therapy. However, liver-directed transduction has been hindered by immune responses unpredicted in the preclinical models, resulting in therapy failure. Liver immune responses are strictly regulated by the interactions between hepatocytes and non-parenchymal liver cells, such as Kupffer cells (the liver-resident macrophages) but how rAAVs induce such responses remains largely unknown. Therefore, human models recapitulating such interactions are required to address innate immune responses. Here, we developed a human 3D model to characterize the contribution of hepatocytes and Kupffer cells to the innate immune response. We developed a strategy for the differentiation of Kupffer-like cells from circulating monocytes based on cell-cell contact with primary human hepatocytes. We fine-tuned critical co-culture parameters to obtain a Kupffer-like phenotype while retaining hepatocyte viability and identity. Functional assessment of the differentiated 3D co-cultures showed that the model is responsive to classical pathogen-associated molecular pattern molecules, at a gene expression and secretory level. Moreover, we observed increased proinflammatory cytokine expression and secretion when challenged with a rAAV vector. Our data indicate the suitability of the novel model to investigate hepatocyte-Kupffer cell interactions and address innate immune responses within a human liver microenvironment.

Liver transplantation (LT) is a life-saving intervention for patients with end-stage liver disease (ESLD). However, 12-20% of patients listed for LT will die on the waitlist. Modern risk scores used for transplant prioritization cannot encompass the full statistical heterogeneity of patients awaiting LT, disadvantaging women and patients with cholestatic liver disease. Our study objective was to implement more equitable LT prioritization via a more expressive class of statistical models to individualize risk prediction. To do so, we created DynaMELD, a deep machine learning-based model of waitlist prioritization. DynaMELD leverages a neural network to model complex interactions between covariates, and leverages the rate-of-change (velocity) of time-varying laboratory biomarkers to predict a more personalized risk of mortality or dropout. Our study cohort comprised 53,046 patients with ESLD listed for LT from 2016- 2023 from the U.S. Scientific Registry of Transplant Recipients. Using 90-day concordance to measure risk discrimination, DynaMELD achieves 90-day concordance 0.5% higher than MELD 3.0 (p < 0.001). Using pooled group concordance (PGCI) as a measure of fairness, DynaMELD achieves a PGCI 1.2% higher for female patients (p < 0.001), 8.3% higher for patients with primary biliary cholangitis (p < 0.001), 7.2% higher for patients with primary sclerosing cholangitis (p < 0.001), and 1.5% higher for patients with acute-on-chronic liver failure Grade 1 (p < 0.001) compared to MELD 3.0. DynaMELD reclassifies members of these sub-groups into higher risk tiers, suggesting it would improve their access to organ offers. Introspecting upon DynaMELD using the method of SHapley Additive exPlanations (SHAP) values provides an individualized degree of model interpretability. Overall, DynaMELD may provide more accurate, individualized predictions of waitlist mortality or dropout to reduce inequities and fairly prioritize patients for liver transplant.

Hepatocytes derived from human pluripotent stem cells (PSCs) hold great promise for modeling human liver disease, in vitro hepatotoxicity testing, and future cellular therapy. However, current protocols generate hepatocyte-like cells (HLCs) that resemble fetal hepatocytes, and thus do not accurately recapitulate the molecular identity and functions of the adult liver. To address this, we compared the transcriptomes of human fetal and adult liver to PSC-derived HLCs during progressive stages of in vitro differentiation. This revealed that during the final stages of in vitro differentiation the hepatic transcription factors HNF4A and CEBPA were sub-optimally expressed. Computational analyses predicted that ALK5i II (TGF-{beta} receptor inhibitor) and thyroid hormone (T3) would be able to rectify this and improve HLC maturation. We next show that application of these molecules during hepatocyte differentiation indeed increases CEBPA and HNF4A mRNA and protein expression, and that these HLCs show enhanced albumin secretion, a 25-fold increase in CYP3A4 activity, and 10 to 100-fold increased expression of mature hepatic markers. We demonstrate that this improved maturation is effective across different cell lines and HLC differentiation protocols, and exemplifies that our approach provides a tractable template for identifying and targeting additional factors that that will fully mature human liver cells from human pluripotent stem cells.

BackgroundDiagnosis of liver cirrhosis in patients with chronic hepatitis B is challenging given rare use of biopsy. In low and middle-income countries, transient elastography (TE), a recommended non-invasive imaging test for cirrhosis is rarely accessible. We therefore investigated the performance of multiple low-cost and more accessible blood-based liver fibrosis markers in patients with chronic hepatitis B infection in Zambia. As alcohol use complicates the assessment and outcomes of hepatitis B, we also considered alcohol use patterns in our evaluation. MethodsWe performed a hospital-based cross-sectional study, in Lusaka, Zambia, among consecutive treatment-naive adults with chronic hepatitis B mono-infection (i.e., HIV-negative) presenting to our hospital. The reference test for cirrhosis was TE of >/=9.6 kPa. Low-cost markers were the AST-to-platelet ratio index (APRI) at recommended threshold >2, as well as lower proposed alternative thresholds for Africa, >0.5 and >0.65, AST/ALT ratio and FIB-4 index >3.25. We evaluated the performance of each marker versus TE. In a secondary analysis, we evaluated marker performance in participants with current alcohol use versus lifetime abstinence. ResultsAmong 239 adults with HBV mono-infection analyzed, the mean age was 34.7 years and 53 (22.2%) reported current alcohol use. The prevalence of cirrhosis by TE was 16.3% (95% CI: 11.87-21.63). The area under the receiver operating characteristic curve was 0.83, 0.80, 0.79 and 0.73 for FIB-4, APRI >0.5, APRI >0.65 and APRI >2 respectively. Virtually all indices performed less well in people with current alcohol use. ConclusionThese data support the adoption of a lower APRI threshold in Africa, and the use of the FIB-4 index, for diagnosis of cirrhosis among patients with chronic hepatitis B infection. The currently-recommended APRI threshold may exclude people with cirrhosis who need antiviral therapy. Clinicians adopting these markers should screen for alcohol use and consider re-assessment of cirrhosis after alcohol reduction.

Co-inhibition of programmed cell death receptor-1 (PD-1) and vascular endothelial growth factor receptor (VEGFR) pathway has shown efficacy in hepatocellular carcinoma (HCC). NLRP3 is a component of the inflammasome involved in the initiation, development, and progression of multiple cancers. We examined whether adding an NLRP3 agonist to dual PD-1/VEGFR inhibitors is feasible and can address treatment resistance in orthotopic HCC in mice with underlying liver damage. Mice with established tumors were treated with an NLRP3 agonist alone, combination of anti-VEGFR2 or the multikinase inhibitor regorafenib with anti-PD1 antibodies, or their combination. In all models tested, NLRP3 agonist therapy showed acceptable toxicity but no effect on tumor growth delay, disease morbidity, or survival. Pharmacodynamic analyses confirmed the effects of NLRP3 agonist therapy on inflammasome, evidenced by a significant elevation in plasma levels of pro-inflammatory cytokines such as IL-1{beta}. However, these changes were not detectable in tumor tissues, where we detected increased expression of immunosuppressive markers IL-6, KC/GRO, CCL9, and IL-18, and immune checkpoint molecules (PD1, PD-L1, and CTLA-4) after NLRP3 agonist therapy. Thus, modulation of the inflammasome with a novel NLRP3 agonist was feasible in mice with orthotopic HCC and liver damage but did not enhance efficacy when combined with anti-PD1/VEGFR therapies.

ObjectivesHepatocellular carcinoma (HCC) is the fourth leading cause of cancer-related death worldwide. Most cases occur in patients with an underlying cirrhosis. The French national guidelines recommend semiannual abdominal ultrasound surveillance for early HCC detection in patients with cirrhosis. The primary goal of our retrospective cohort study was to evaluate compliance with this recommendation. MethodsWe used 2007-2016 general public health insurance program (Regime General) data from the French National Health Data System (Systeme National des Donnees de Sante, or SNDS). Included patients were 18 to 75 years old, diagnosed with liver cirrhosis between 2009 and 2013, and underwent their first ultrasound >4 months after their index date. The number of annual ultrasounds was recorded over a 3-year follow-up period. Compliance was defined as having had at least 2 ultrasounds per year over the follow-up time. ResultsAmong the 66,464 patients included in the analysis, surveillance was optimal (no year with <2 ultrasounds) in 5,082 patients (7.6%), suboptimal (one year with <2 ultrasounds) in 3,928 (5.9%), and failed (remaining cases) in 57,454 (86.4%). Older age, male sex, a high Charlson index, frequent gastroenterologist/hepatologist visits, and viral etiology were associated with better surveillance, whereas low socioeconomic status, despite Frances universal health coverage, was linked to failed surveillance. ConclusionsIn French patients with cirrhosis, most of cancer surveillance is failing when considering recommendation in vigor. In order to improve surveillance, a better understanding of the social determinants of health equity is needed.

IntroductionMetabolic dysfunction-associated steatotic liver disease (MASLD) and metabolic dysfunction-associated steatohepatitis (MASH) have risen substantially in prevalence over recent decades, driven by a growing global burden of obesity, diabetes mellitus, and other cardiometabolic risk factors. In response, researchers have intensified efforts to evaluate novel and re-purposed therapies that may prevent or reverse disease progression. Although several pharmacological therapies are under investigation, robust comparative evidence on their relative effectiveness and safety remains limited. We will therefore conduct a systematic review and network meta-analysis (SRNMA) of randomized controlled trials (RCTs) evaluating pharmacological therapies for adults with MASLD or MASH. MethodsWe will search four electronic databases (Ovid MEDLINE, Embase, CINAHL and Cochrane CENTRAL) from inception to August 2025 without language and other restrictions. Eligible studies will include parallel-arm RCTs enrolling [&ge;]10 adults per arm with MASLD or MASH; comparing any pharmacological therapy to standard care, no treatment, lifestyle modifications, placebo or alternative pharmacotherapies; and having a minimum follow-up duration of 12 weeks. Primary clinical outcomes are all-cause mortality, cardiovascular mortality, hospitalization, progression to cirrhosis, hepatic decompensation, hepatocellular carcinoma, and serious treatment-related adverse events. Surrogate outcomes include histological, imaging, biochemical, and metabolic markers of disease activity. Paired reviewers will independently screen identified hits for eligibility, extract data from eligible studies, and assess risk of bias using the Risk Of Bias instrument for Use in SysTematic reviews-for Randomised Controlled Trials (ROBUST-RCT). We will conduct separate NMAs for MASLD and MASH populations using a frequentist graph-theoretic random-effects model. Certainty of evidence will be assessed using GRADE. Subgroup and sensitivity analyses will explore effect modification by comorbidities and study quality. Ethics and DisseminationNo ethics approval is required. Results will be disseminated via peer-reviewed publication and conference presentations to inform clinicians, guideline developers, and health system decision-makers. PROSPERO Registration NumberCRD420251103235.

Background and AimsUnresolved hepatitis B virus (HBV) infection leads to a progressive state of immune exhaustion that impairs resolution of infection, leading to chronic infection (CHB). The immune-competent AAV-HBV mouse is a common HBV preclinical immune competent model, though a comprehensive characterization of the liver immune microenvironment and its translatability to human infection is still lacking. We investigated the intrahepatic immune profile of the AAV-HBV mouse model at a single-cell level and compared with data from CHB patients in immune tolerant (IT) and immune active (IA) clinical stages. MethodsImmune exhaustion was profiled through an iterative subclustering approach for cell-typing analyses of single-cell RNA-sequencing data in CHB donors and compared to the AAV-HBV mouse model 24-weeks post-transduction to assess its translatability. This was validated using an exhaustion flow cytometry panel at 40 weeks post-transduction. ResultsUsing single-cell RNA-sequencing, CD8 pre-exhausted T-cells with self-renewing capacity (TCF7+), and terminally exhausted CD8 T-cells (TCF7-) were detected in the AAV-HBV model. These terminally exhausted CD8 T-cells (expressing Pdcd1, Tox, Lag3, Tigit) were significantly enriched versus control mice and independently identified through flow cytometry. Importantly, comparison to CHB human data showed a similar exhausted CD8 T-cell population in IT and IA donors, but not in healthy individuals. ConclusionsLong term high titer AAV-HBV mouse liver transduction led to T-cell exhaustion, as evidenced by expression of classical immune checkpoint markers at mRNA and protein levels. In both IT and IA donors, a similar CD8 exhausted T-cell population was identified, with increased frequency observed in IA donors. These data support the use of the AAV-HBV mouse model to study T-cell exhaustion in HBV infection and the effect of immune-based therapeutic interventions. Lay SummaryThe AAV-HBV mouse model is used as a research tool to study hepatitis B infection. In this study we evaluated the translation value from mouse to human with regards to T-cell exhaustion. HighlightsO_LIAAV-HBV mice transduced with a high titer vector showed presence of CD8 exhausted T-cells after 24 weeks. C_LIO_LIHigh titer transduced mice, but not lower titer show increased expression of LAG-3, TOX, TIM-3 and TIGIT in CD8 T-cells. PD-1 was increased in CD8 T-cells, independent of HBV transduction titer. C_LIO_LIA similar exhausted CD8 T-cell population could be found in chronic HBV donors, but not in healthy individuals. C_LI Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=75 SRC="FIGDIR/small/552328v1_ufig1.gif" ALT="Figure 1"> View larger version (14K): org.highwire.dtl.DTLVardef@17db34dorg.highwire.dtl.DTLVardef@186a0b1org.highwire.dtl.DTLVardef@19fd39borg.highwire.dtl.DTLVardef@1fd9f64_HPS_FORMAT_FIGEXP M_FIG C_FIG

BackgroundSodium-Glucose Cotransporter Protein 2 Inhibitors (SGLT2Is) are known for their cardiovascular and renoprotective benefits and are efficacious in managing Metabolic-Associated Steatotic Liver Disease (MASLD). However, limited data exist on their use in advanced liver disease, particularly liver cirrhosis. AimsTo synthesize existing evidence on the efficacy and safety of SGLT2Is in patients with liver cirrhosis and to provide clinical guidance. MethodsA systematic review and meta-analysis were conducted following the PRISMA 2020 Statement. Searches in major health databases identified studies where SGLT2Is were used in patients with cirrhosis. The analysis focused on prospective trials in decompensated cirrhosis and retrospective studies in compensated cirrhosis. Primary outcomes included the need for large-volume paracentesis (LVP) and mortality. Secondary outcomes assessed weight loss, loop diuretic dose reduction, residual ascites, acute kidney injury (AKI), hyponatremia, hepatic encephalopathy (HE), and urinary tract infections (UTIs). ResultsTen studies (8 peer-reviewed) from 2020-2024 were included: 2 randomized controlled trials, 4 single-arm prospective trials, and 4 retrospective studies. SGLT2I use was associated with reduced LVP (RR 0.45, CI 0.31-0.66, p<0.001) and mortality (aHR 0.46, CI 0.38-0.55, p<0.001). Benefits included a 39 mg reduction in loop diuretic dose, 7 kg weight loss, and no significant increase in residual ascites, AKI, hyponatremia, HE, or UTIs. ConclusionsSGLT2Is show promise in managing diuretic-resistant ascites and reducing mortality in liver cirrhosis without causing significant adverse events. Larger, randomized controlled trials are needed to validate these findings.

Introduction and ObjectivesMetabolic dysfunction-associated steatotic liver disease (MASLD) affects about one-quarter of adults worldwide, and liver fibrosis is its strongest predictor of liver-related morbidity and mortality. Using combined in-silico screening and clinical validation, we aimed to identify circulating biomarkers associated with fibrosis progression. Fibulin-3 was identified, and its diagnostic performance was evaluated in biopsy-proven MASLD cohorts. Materials and MethodsThe GSE125251 RNA-seq dataset was reanalyzed to compare liver transcriptomes from MASLD subjects with minimal (F0-F1) versus moderate to advanced fibrosis (F2/F3-F4). Differentially expressed genes (DEGs) were filtered to retain plasma-secreted, protein-coding candidates. Top-ranked genes were evaluated in liver biopsies from a morbidly obese cohort (n = 65) stratified by fibrosis stage, and their plasma levels were measured via ELISA in an independent bariatric cohort (n = 225). ResultsAmong 106 DEGs, 22 encoded plasma-circulating proteins. Six top candidates (EFEMP1, LTBP2, LUM, DPT, CHI3L1, CCL20) were prioritized. EFEMP1 (Fibulin-3) showed the strongest association with fibrosis, with significantly higher hepatic mRNA and protein expression in F2/F3-F4 versus F0-F1 (p < 0.005). Plasma Fibulin-3 levels correlated with fibrosis stage ({rho} = 0.40, p < 0.0001), increasing from 9.4 ng/mL in F0-F1 to 21.7 ng/mL in F2/F3-F4. Its diagnostic performance for F [&ge;] 2 (AUROC = 0.78) exceeded that of APRI, FIB-4, NFS, and HSI. A combined index including Fibulin-3, HSI, platelets, and GGT increased the AUROC to 0.87 (CI: 0.79-0.92). ConclusionsPlasma Fibulin-3 is notably higher in individuals with advanced MASLD and represents a promising non-invasive biomarker for liver fibrosis stratification in metabolically unhealthy obese populations.

Background & AimsLiver cancer, the third leading cause of cancer-related mortality worldwide, has two main subtypes: hepatocellular carcinoma (HCC), accounting the majority of the cases, and cholangiocarcinoma (CAA). Notch pathway primarily regulates the intrahepatic development of bile ducts, which are lined with cholangiocytes, but it can also be upregulated in 1/3 of HCCs. To better understand the role of NOTCH1 in HCC, we developed a novel mouse model driven by activated Notch1 intracellular domain (NICD1) and MYC overexpression in hepatocytes. MethodsUsing the hydrodynamic tail-vein injection method for establishing primary liver tumors, we generated a novel murine model of liver cancer harboring MYC overexpression and NOTCH1 activation. We characterized this model histopathologically as well as transcriptomically, utilizing both bulk and single cell RNA-sequencing. We also performed functional experiments using monoclonal antibodies. ResultsMYC;NICD1 tumors displayed a combined HCC-CCA phenotype with temporal plasticity. At early time-points, histology was predominantly "cholangiocellular", which then progressed to mainly "hepatocellular". The "hepatocellular" component was enriched in mesenchymal genes and gave rise to lung metastasis. Metastatic cells were enriched in the TGFB and VEGF pathways and their inhibition significantly reduced the metastatic burden. ConclusionsOur novel mouse model uncovered NOTCH1 as a driver of temporal plasticity and metastasis in HCC, the latter of which is, in part, mediated by angiogenesis and TGF{beta} pathways. Impact and ImplicationsThis study develops a novel murine model of NOTCH1-driven liver cancer, an understudied oncogene in HCC. Using this model, we show that NOTCH1 drives plasticity in HCC and metastasis to the lungs that can be therapeutically targeted through inhibition of VEGF and TGF{beta} pathways. HighlightsO_LINOTCH1 activation in combination with MYC overexpression drives combined HCC-CCA. C_LIO_LINOTCH1 activation in hepatocytes drives temporal plasticity. C_LIO_LINOTCH1 activation drives metastasis of HCC cells to the lungs, but not of CCA cells. C_LIO_LIAngiogenesis and TGF{beta} pathways mediate NOTCH1-induced lung metastasis. C_LI Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=93 SRC="FIGDIR/small/619856v2_ufig1.gif" ALT="Figure 1"> View larger version (39K): org.highwire.dtl.DTLVardef@143c866org.highwire.dtl.DTLVardef@119ddb6org.highwire.dtl.DTLVardef@12ad951org.highwire.dtl.DTLVardef@2160b8_HPS_FORMAT_FIGEXP M_FIG C_FIG

Background & AimsIntrahepatic cholestasis of pregnancy (ICP) is the most common and high-risk liver disorder during the critical period of human reproduction. Despite varying prevalence across populations, a mechanistic understanding of this phenomenon is lacking. This study delves into the genetic etiology of ICP in East Asians, drawing comparisons with Europeans to comprehend ICP etiology in the context of genetic background and evolution. MethodsWe conducted the hitherto largest-scale genome-wide association studies (GWAS) on total bile acid concentration (TBA) and ICP among 101,023 Chinese pregnancies. The findings were subsequently replicated in two cohorts and compared with European populations. Additionally, phenome-wide association and spatio-temporal evolution analyses were employed to understand the function and explore evolutionary pattern of sites associated with ICP. ResultsWe identified eight TBA and five ICP loci, including ten novel loci. Notably, we found an East-Asian-specific genetic locus at 14q24.1, contributing to a 6.41 {micro}mol/L increase in TBA and a 15.23-fold higher risk of ICP per risk allele (95% CI: 15.10 to 15.36, P = 9.23x10-375). Phenome-wide association studies and spatial-temporal evolution analyses revealed that the 14q24.1 ICP risk locus exhibits resistance to hepatitis B infection and has become prevalent only within the last 3,000 years in East and Southeast Asia. ConclusionsOur investigations have unraveled a distinct etiology of ICP between Europeans and East Asians, and has linked ICP etiology in East Asians to a historical HBV epidemic in East and Southeast Asia within the last 3,000 years. These findings lay the groundwork for an improved biological understanding of ICP pathophysiology. Further exploration and utilization of these variations hold the potential for more precise detection, assessment, and treatment of ICP. Lay summaryIntrahepatic cholestasis of pregnancy (ICP) is a prevalent and high-risk liver disorder that occurs during pregnancy, a critical period in human reproduction. It affects approximately 1% to 6.06% pregnancies and has been associated with severe adverse outcomes such as preterm birth and stillbirth. While rare and common variants associated with ICP have been identified in the European population, the genetic basis of ICP in East Asian population remains uncharacterized. Here, we conducted the largest-scale genome-wide association studies to date for TBA and ICP among 101,023 Chinese pregnant women, including 4,703 cases and 96,320 controls from two hospitals in Shenzhen, China. We replicated our findings in two independent Chinese cohorts and compared them with ICP genetic studies in the European population. We identified eight and five genome-wide significant loci for TBA and ICP, respectively, including ten novel loci. Notably, we identified an East-Asian-specific genetic locus contributing to a 6.41 {micro}mol/L increase in TBA per risk allele and a 15.23-fold higher risk of ICP. Further exploration through phenome-wide association studies and spatial-temporal evolution analyses revealed that the 14q24.1 ICP risk locus exhibits resistance to hepatitis B infection and has become prevalent only within the last 3,000 years in East and Southeast Asia. These findings suggest a historical HBV epidemic in East and Southeast Asia within 3,000 years may have contributed to the increased prevalence of ICP and TBA risk alleles among East Asians. Our study unravels a distinct genetic etiology of ICP between Europeans and East Asians. These findings lay the foundation for an improved understanding of ICP pathophysiology and emphasize the need for integrating population evolution into genetic medicine for personalized genomics and clinical guidance. Highlights(1) In the most powerful genome-wide association studies on TBA and ICP in East Asians to date, we identified eight and five genetic loci, respectively, of which, 7 and 3 were novel discoveries. (2) One of the novel loci, the 14q24.1 locus, stands out as it contains unique causal genetic variants specific to East-Asians. These variants demonstrate large effects, contributing to an average increase of 6.41 {micro}mol/L in TBA per risk allele and a 15.23-fold higher risk of ICP. (3) The risk mutations associated with ICP at the 14q24.1 exhibit resistance to hepatitis B infection and has only become prevalent within the last 3000 years in East and Southeast Asia. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=98 SRC="FIGDIR/small/24309754v1_ufig1.gif" ALT="Figure 1"> View larger version (35K): org.highwire.dtl.DTLVardef@1e56492org.highwire.dtl.DTLVardef@12dbc93org.highwire.dtl.DTLVardef@1b7774corg.highwire.dtl.DTLVardef@1f94984_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOAbstract Figure:C_FLOATNO Genetic basis and evolutionary history of intrahepatic cholestasis of pregnancy in East Asia. TBA: Total bile Acid. ICP: Pregnancy intrahepatic cholestasis. Refer to the main text for the illustration. C_FIG

Background and AimsInflammation drives progression of chronic liver disease. However, the triggers of inflammation remain undefined during chronic hepatitis B (CHB) because hepatic flares are spontaneous and difficult to capture. We used nucleoside analogue (NA) withdrawal to investigate early inflammatory events because liver damage after stopping therapy occurs in a predictable time frame. 11 CHB patients underwent 192 weeks of NA therapy before a protocol defined stop. Liver fine-needle aspirates (FNAs) were collected at baseline and 4-weeks post-withdrawal and analyzed using flow cytometry and single-cell RNA sequencing (scRNA-seq). Intrahepatic mononuclear cells (IHMCs) from uninfected livers were used to validate transcriptomic findings. At 4 weeks post NA-withdrawal, HBV DNA rebounded but alanine aminotransferase (ALT) levels remained normal, 7/11 patients developed ALT elevations (>2xULN) at later timepoints. There were no changes in cell frequencies between baseline and viral rebound. ScRNA-seq revealed upregulation of IFN stimulated genes (ISGs) and pro-inflammatory cytokine MIF upon viral rebound. In vitro experiments confirmed the type I IFN-dependent ISG profile whereas MIF was induced primarily by IL-12. MIF exposure further amplified inflammatory cytokine production by myeloid cells. Our data show that innate immune activation is detectable in the liver before clinically-significant liver damage is detectable in the serum.

BackgroundHepatocellular carcinoma represents a significant global health challenge, affecting over a million patients annually, arising mainly from chronic liver diseases, with a majority being related to viral infections. However, despite the groundbreaking clinical results of immune checkpoint blockades and adoptive cell therapies, we still face non-responders accompanied by high rebound rates after resection. Considering that the main concern is to overcome a highly specialized immunosuppressive tumor microenvironment of the individual patient, the characterization of the particular tumor microenvironment and the source of the immune cells used in ACT is of immense importance. Approved ACT therapies mainly use modified peripheral blood cells from individuals. At the same time, tumor-infiltrating lymphocytes are underrepresented even if they have garnered interest due to their potential to target tumor-specific antigens more effectively. MethodsIn this study, we employed allogenic and autologous immune cell sources for expansion and stimulation, resulting in adoptive T-cell transfer experiments determining the effector cell differentiation and the related anti-tumor effects by the possible implementation of re-stimulation. ResultsWe determined a high success rate in expanding and stimulating tumor-infiltrating lymphocytes with consistent CD8 T-cell fractions from HCC patients. To showcase the effectiveness of stimulated T-cells from different sources, we generated cell lines derived from the margin and center of an HBV-induced HCC with a highly immune-suppressive TME. We found effective immune responses supported by cell death induction, ferroptosis, proptosis and apoptosis triggered by all sources of T-cells depending on the area of derived tumor cells. ConclusionEffector T-cell fractions derived from tumor-infiltrating lymphocytes present a viable source for cell-based immune therapy combined with immune checkpoint inhibitors in HCC patients, especially after resection, to suppress rebound strategies of the parental tumor on an individual level.

Immune dysfunction, both depression and hyperactivation, in liver disease contributes to significant morbidities and mortalities, depending on liver damage severity and etiology. The underlying causes of immune dysfunction in advanced liver disease, whether pathogen or host-mediated, remain unclear. We reported lasting generalized CD8+ T cell hyperfunction in individuals with advanced liver fibrosis in chronic HCV infection. The separation of viral and fibrosis-driven effects or the association of this phenomenon with clinical outcomes of advanced liver fibrosis remains to be determined. Here, a hepatotoxic murine model of liver fibrosis was used to decouple liver fibrosis from viral infection. Carbon tetrachloride (CCl4)-treated mice presented progressive liver fibrosis within {approx}12 weeks, resulting in severe diffuse fibrosis, focal necrosis and surrounding mixed inflammation; pathology similar to that of chronic HCV infection. Taking advantage of this model, we investigated if liver fibrosis caused systemic CD8+ T cell hyperfunction and evaluated its impact on host immune response. At peak liver fibrosis, circulating CD8+ T cells presented increased expression of IFN-{gamma} and granzyme B (GrzB) in comparison to control animals. CD8+ T cell hyperfunction arose by 8 weeks of CCl4 treatment and was sustained with continued liver insult. As a result, fibrotic mice were unable to resist an ectopic tumour challenge and were less responsive to immunotherapy. Furthermore, CD8+ T cell dysfunction was observed in other contexts of chronic liver insult such as high fat diet-induced liver steatosis, even in the absence of significant fibrosis. Collectively, this study shows the impact of chronic liver insult on systemic CD8+ T cell function and its association with impaired immune response, such as tumour surveillance.