Liver International最新文献

Idiosyncratic hepatotoxicity induced by prescribed drugs has been known since the early 20th century. Identifying risk factors, including genetic factors, that trigger this drug-induced liver injury (DILI) has been an important priority for many years, both to prevent drugs that cause liver injury being licensed and as a potential means of preventing at-risk patients being prescribed causative drugs. Improved methods for genomic analysis, particularly the development of genome-wide association studies, have facilitated the identification of genomic risk factors for DILI, but, to date, there are only two main examples, liver injury caused by amoxicillin-clavulanate (AC) and by flucloxacillin, where genetic risk factors causing the injury have been identified and replicated with understanding of the underlying mechanism. There has also been progress on identifying genetic risk factors for liver injury caused by other anti-infective agents, herbal remedies and nonsteroidal anti-inflammatory drugs. The majority of genetic risk factors identified to date are specific human leucocyte antigen (HLA) alleles and evidence that these alleles preferentially present self-peptides inappropriately to T cells in the liver has been obtained. Non-HLA genes also contribute to genetic susceptibility, both as co-factors in T-cell responses and, in the case of isoniazid-only, drug metabolism. Polygenic risk scores to predict DILI have been developed, both a simple score that predicts AC injury and complex scores that may be applied to DILI more generally and provide evidence that additional risk factors other than HLA genes exist.

Background and aims: This study aims to develop and validate a machine learning (ML) model predicting hepatocellular carcinoma (HCC) in chronic hepatitis B (CHB) patients after the first 5 years of entecavir (ETV) or tenofovir (TFV) therapy.

Methods: CHB patients treated with ETV/TFV for > 5 years and not diagnosed with HCC during the first 5 years of therapy were selected from two hospitals. We used 36 variables, including baseline characteristics (age, sex, cirrhosis, and type of antiviral agent) and laboratory values (at baseline, at 5 years, and changes between 5 years) for model development. Five machine learning algorithms were applied to the training dataset and internally validated using a test dataset. External validation was performed.

Results: In years 5-15, a total of 279/5908 (4.7%) and 25/562 (4.5%) patients developed HCC in the derivation and external validation cohorts, respectively. In the training dataset (n = 4726), logistic regression showed the highest area under the receiver operating curve (AUC) of 0.803 and a balanced accuracy of 0.735, outperforming other ML algorithms. An ensemble model combining logistic regression and random forest performed best (AUC, 0.811 and balanced accuracy, 0.754). The results from the test dataset (n = 1182) verified the good performance of the ensemble model (AUC, 0.784 and balanced accuracy, 0.712). External validation confirmed the predictive accuracy of our ensemble model (AUC, 0.862 and balanced accuracy, 0.771). A web-based calculator was developed (http://ai-wm.khu.ac.kr/HCC/).

Conclusions: The proposed ML model excellently predicted HCC risk beyond year 5 of ETV/TFV therapy and, therefore, could facilitate individualised HCC surveillance based on risk stratification.

The discovery of PNPLA3 as a genetic risk factor for liver disease has transformed our understanding of the pathogenesis of alcohol-related liver disease (ALD). The recent reclassification of fatty liver disease as steatotic liver disease (SLD), introducing metabolic dysfunction and alcohol-related liver disease (MetALD), has highlighted how genetic and environmental factors synergistically drive liver damage. The PNPLA3 rs738409 variant stands as a paradigmatic example of gene–environment interaction, where its effect on liver disease is dramatically amplified by alcohol consumption, obesity and type 2 diabetes. Understanding these interactions has revealed novel pathogenic mechanisms. The robust genetic evidence and a growing understanding of molecular mechanisms have made PNPLA3 an attractive therapeutic target. Several compounds targeting PNPLA3 are now in clinical development. While initial trials have focused on metabolic dysfunction-associated SLD, the recognition that almost all individuals with excessive alcohol consumption have metabolic comorbidities provides an unprecedented opportunity to evaluate these genetically informed therapies in MetALD. In this review, we examine the role of PNPLA3 in ALD, focusing on gene–environment interactions and therapeutic implications in the context of the new disease classification framework.