Hepatology International最新文献

Purpose: Predicting hepatic encephalopathy (HE) after transjugular intrahepatic portosystemic shunt (TIPS) is critical for guiding portal hypertension treatment strategies and enabling early intervention. This study aims to employ the Tabular Prior-data Fitted Network (TabPFN) algorithm to develop a machine learning (ML) model that predicts post-TIPS HE.

Methods: This study retrospectively enrolled 218 patients who underwent TIPS across three hospitals. Preoperative contrast enhanced CT (CECT) scans were used to delineate the volumetric region of interest (VOI) for the liver, spleen, abdominal fat, and abdominal muscle. Radiomics and deep transfer learning (DTL) features were extracted from each VOI. Overt HE occurrence during follow-up was divided into two groups. 171 patients (two hospitals) were randomly split (7:3) into training and validation set, 47 patients (third hospital) formed an external test set. After feature selection, we trained and compared multiple ML models. Shapley additive explanation (SHAP) was performed for model interpretability.

Results: The overall incidence of overt HE in the study cohort was 20.6%. The combined TabPFN model with the best predictive performance achieved AUCs of 0.953 (training set), 0.870 (validation set), and 0.942 (external test set), with accuracies of 0.933, 0.846, and 0.872, respectively. SHAP analysis identified the liver radiomics signature as a dominant predictors. Time-dependent AUCs at 90, 180, 365, and 730 days exceeded 0.88 in all cohorts, and high-risk patients had significantly higher HE occurrence (p < 0.01).

Conclusion: A TabPFN-based ML model integrating CECT radiomics, DTL features, and MELD score enables accurate, externally validated prediction of post-TIPS HE, supporting personalized risk stratification and clinical decision-making.

Background: RNA splicing dysregulation plays an important role in hepatocellular carcinoma (HCC). However, the critical functions of zinc finger CCHC type and RNA binding motif 1 (ZCRB1), one of the key components of the minor spliceosome, in HCC remains unclear.

Methods: The mRNA and protein expression of ZCRB1 were evaluated by bioinformatics analysis, western blotting and immunohistochemistry. The role of ZCRB1 on biological functions of HCC were measured in vitro and in vivo. rMATS software analyzes RNA-seq data to identify ZCRB1-related alternative splicing (AS) events. Therapeutic targeting of ZCRB1 using GalNAc-conjugated small-interfering RNA was investigated in the HCC mouse model.

Results: We discovered that ZCRB1 was dramatically upregulated in HCC tissues, and high ZCRB1 expression was associated with poor prognosis in HCC patients. Knockdown of ZCRB1 significantly reduced HCC cell proliferation and promoted cell apoptosis. In addition, high enrichment levels of H3K27ac in the promoter region activated ZCRB1 expression. Mechanistically, AS event analysis revealed that knockdown ZCRB1 resulted in retention of intron 11 of USP21 thereby reducing USP21 expression. USP21 overexpression partially rescued the attenuation of malignant behavior of HCC cells caused by inhibition of ZCRB1, and inhibition of USP21 reduced proliferation in HCC cells. Furthermore, therapeutic targeting of ZCRB1 using GalNAc-conjugated small-interfering RNA significantly reduced tumor burden and increased the infiltration level of CD8+ T cells in the HCC mouse model.

Conclusions: This study identified ZCRB1 as an oncogenic U12-type splicing factor that supports malignant growth through blocking intron 11 retention of USP21 in HCC cells. Thus, ZCRB1 could serve as a new target for HCC therapy and a potential biomarker for HCC prognosis.