Xia, Ying

Abstract The prevalence of steatotic liver disease (SLD) is rising globally, presenting a significant public health challenge. While non-contrast computed tomography (CT) scans hold promise for the opportunistic detection of SLD, their full potential has not been systematically explored. Clinical indicators, such as serum data, are crucial for monitoring the progression of SLD. We introduce a novel Multimodal AI for Opportunistic Steatotic Liver Disease Screening (MAOSS), which integrates demographic data, serum tests, conventional CT biomarkers, and deep learning-derived CT image features to enhance model performance and explainability. MAOSS employs an innovative learning strategy, trained on a large-scale dataset (n = 1,783) encompassing both histopathologically-confirmed SLD and radiologically-reported steatosis cases. The model underwent extensive retrospective validation using steatosis grades derived from both histology and MRI-PDFF across internal and external datasets (n = 913), as well as real-world patient data (n = 18,504). It accurately detected mild to severe steatosis (≥ S1) with AUCs ranging from 0.903 to 0.929. Notably, MAOSS significantly improved the overall AUC for radiologists (n = 11) from 0.709 to 0.798 (p < 0.001). Furthermore, MAOSS was applied to optimize the standard clinical pathway for a simulated primary risk screening of SLD patients (n = 1,192), resulting in the identification of 31% more patients at risk of steatohepatitis and liver fibrosis progression, highlighting its potential to enhance SLD detection and management.

BackgroundHeavy alcohol consumption disrupts gut epithelial integrity, leading to increased permeability of the gastrointestinal tract and subsequent translocation of microbes. Regenerating islet‐derived protein 3α (REG3α) and Trefoil factor 3 (TFF3) are mainly secreted to the gut lumen by Paneth and Goblet cells, respectively, and are functionally linked to gut barrier integrity. Circulating levels of REG3α and TFF3 have been identified as biomarkers for gut damage in several human diseases. We examined whether plasma levels of REG3α and TFF3 were dysregulated and correlated with conventional markers of microbial translocation (MT) and pro‐inflammatory mediators in heavy drinkers with and without alcoholic hepatitis (AH).MethodsCross‐sectional and longitudinal studies were performed to monitor plasma levels of REG3α and TFF3 in 79 AH patients, 66 heavy drinkers without liver disease (HDC), and 46 healthy controls (HC) at enrollment and at 6‐ and 12‐month follow‐ups. Spearman correlation was used to measure the relationships of REG3α and TFF3 levels with MT, disease severity, inflammation, and effects of abstinence from alcohol.ResultsAt enrollment, AH patients had significantly higher levels of REG3α and TFF3 than HDC and HC. The elevated REG3α levels were positively correlated with the 30‐day fatality rate. Plasma levels of REG3α and TFF3 in AH patients differentially correlated with conventional MT markers (sCD14, sCD163, and LBP) and several highly up‐regulated inflammatory cytokines/chemokines/growth factors. At follow‐ups, although REG3α and TFF3 levels were decreased in AH patients with alcohol abstinence, they did not fully return to baseline levels.ConclusionsCirculating levels of REG3α and TFF3 were highly elevated in AH patients and differentially correlated with AH disease severity, MT, and inflammation, thereby serving as potential biomarkers of MT and gut epithelial damage in AH patients.

BackgroundAlcoholic hepatitis (AH) is a severe inflammatory liver disease that develops in some heavy drinkers. AH patients have intense hepatic infiltration of leukocytes. Up‐regulation of cell adhesion molecules (CAMs) upon endothelial cell (EC) activation plays an important role in leukocyte transendothelial migration. CAMs can shed from EC surface and accumulate in the blood, serving as soluble markers for EC activation. In this study, we examined the impact of heavy drinking on expression of soluble forms of EC activation markers (CD146, ICAM‐1, VCAM‐1, and VEGF‐A) and the effect of alcohol abstinence on the reversal of these abnormalities in heavy drinkers with and without AH.MethodsELISA and multiplex immunoassays were used to measure soluble EC activation markers in plasma samples from 79 AH patients, 66 heavy drinkers without overt liver disease (HDC), and 44 healthy controls (HC) at baseline, 31 AH patients and 30 HDC at 6‐month follow‐up, and 18 AH patients and 25 HDC at 12‐month follow‐up.ResultsAt baseline, the 4 soluble markers were significantly up‐regulated in AH patients compared with HDC and HC, whereas only sVCAM‐1 was elevated in HDC relative to HC. At follow‐ups, plasma levels of CD146, VCAM‐1, and VEGF‐A remained higher in AH patients, even for those who stopped drinking. These dysregulated markers correlated with AH disease severity, clinical parameters, and several soluble inflammatory factors.ConclusionsThe levels of soluble CD146, ICAM‐1, VCAM‐1, and VEGF‐A were highly elevated in AH patients, and alcohol abstinence did not completely reverse these abnormalities.

Alcoholic hepatitis (AH) is a severe inflammatory liver disease that develops in some heavy drinkers. The immune system in patients with AH is hyperactive and yet dysfunctional. Here, we investigated whether this immune‐dysregulated state is related to the alcoholic impact on immune checkpoints (ICPs). We used multiplex immunoassays and enzyme‐linked immunosorbent assay to quantify plasma levels of 18 soluble ICPs (sICPs) from 81 patients with AH, 65 heavy drinkers without liver diseases (HDCs), and 39 healthy controls (HCs) at baseline, 33 patients with AH and 32 HDCs at 6‐month follow‐up, and 18 patients with AH and 29 HDCs at 12‐month follow‐up. We demonstrated that baseline levels of 6 sICPs (soluble T‐cell immunoglobulin and mucin domain 3 [sTIM‐3], soluble cluster of differentiation [sCD]27, sCD40, soluble Toll‐like receptor‐2 [sTLR‐2], soluble herpesvirus entry mediator [sHVEM], and soluble lymphotoxin‐like inducible protein that competes with glycoprotein D for herpes virus entry on T cells [sLIGHT]) were up‐regulated, while 11 sICPs (soluble B‐ and T‐lymphocyte attenuator [sBTLA], sCD160, soluble cytotoxic T‐lymphocyte‐associated protein 4 [sCTLA‐4], soluble lymphocyte‐activation gene 3 [sLAG‐3], soluble programmed death 1 [sPD‐1], sPD ligand 1 [sPD‐L1], sCD28, soluble glucocorticoid‐induced tumor necrosis factor receptor‐related protein [sGITR], sGITR ligand [sGITRL], sCD80, and inducible T‐cell costimulator [sICOS]) were down‐regulated in patients with AH compared to HDCs. The up‐regulated sICPs except sLIGHT and down‐regulated sCD80, sCD160, sCTLA‐4, and sLAG‐3 correlated positively or negatively with AH disease severity, bacterial translocation, and inflammatory factors. At follow‐up, abstinent patients with AH still had higher levels of several sICPs compared to HDCs. We also compared expression of 10 membrane‐bound ICPs (mICPs) on peripheral blood mononuclear cells (PBMCs) from patients with AH and HCs by flow cytometry and found that several mICPs were dysregulated on blood cells from patients with AH. The function and regulation of sICPs and mICPs were studied using PBMCs from patients with AH and HCs. Recombinant sHVEM affected tumor necrosis factor (TNF)‐α and interferon‐γ production by T cells from patients with AH and HCs. Conclusion: Both sICPs and mICPs were dysregulated in patients with AH, and alcohol abstinence did not fully reverse these abnormalities. The HVEM axis plays a role in regulating T‐cell function in patients with AH.