Gut microbiota might affect the severity and progression of metabolic dysfunction-associated steatotic liver disease (MASLD). We aimed to characterize gut dysbiosis and clinical parameters regarding fibrosis stages assessed by magnetic resonance elastography. This study included 156 patients with MASLD, stratified into no/mild fibrosis (F0–F1) and moderate/severe fibrosis (F2–F4). Fecal specimens were sequenced targeting the V4 region of the 16S rRNA gene and analyzed using bioinformatics. The genotyping of PNPLA3, TM6SF2, and HSD17B13 was assessed by allelic discrimination assays. Our data showed that gut microbial profiles between groups significantly differed in beta-diversity but not in alpha-diversity indices. Enriched Fusobacterium and Escherichia_Shigella, and depleted Lachnospira were found in the F2–F4 group versus the F0–F1 group. Compared to F0–F1, the F2–F4 group had elevated plasma surrogate markers of gut epithelial permeability and bacterial translocation. The bacterial genera, PNPLA3 polymorphisms, old age, and diabetes were independently associated with advanced fibrosis in multivariable analyses. Using the Random Forest classifier, the gut microbial signature of three genera could differentiate the groups with high diagnostic accuracy (AUC of 0.93). These results indicated that the imbalance of enriched pathogenic genera and decreased beneficial bacteria, in association with several clinical and genetic factors, were potential contributors to the pathogenesis and progression of MASLD.

Pakanat Decharatanachart, Wirichada Pan-ngum, Thoetchai Peeraphatdit, Natthaporn Tanpowpong, Pisit Tangkijvanich, Sombat Treeprasertsuk, Rungsun Rerknimitr, Roongruedee Chaiteerakij. Gut and Liver 2024;18:135-46. https://doi.org/10.5009/gnl230089

Liver cancer remains a significant global health concern, and early-stage differential diagnosis and treatment can dramatically improve patient survival rates. However, current diagnostic methods are challenging and subjective, highlighting the need for accurate diagnostic support tools. This paper introduces a deep learning-based diagnostic framework for detecting liver tumor malignancy using DCE-MRI, particularly in scenarios with limited well-annotated data. To address the overfitting issue, we propose a Siamese convolutional neural network incorporating metric learning concepts, emphasizing discriminative representation learning. The core principle of our framework is to reorganize the embedding space such that similar sample pairs are closer together while dissimilar ones are further apart. Once the embedding space is learned, unseen samples can be classified based on a distance metric to a few seen samples within the embedding space. Comprehensive experiments on a professionally annotated, self-established dataset demonstrate that our framework outperforms fully supervised approaches in data scarcity situations, achieving a sensitivity of 96%, precision of 82%, and an F1-score of 88%.

Background: Gadolinium diethylenetriamine-pentaacetic acid (Gd-EOB-DTPA) is a developed agent with preferential uptake by hepatocytes. A rapid and specific hepatocyte uptake with biliary excretion was observed of approximately 50% of the injected dose. The amount of contrast uptake is thought to be related to reserve liver function. Objectives: To evaluate correlation between liver signal intensity in the hepatobiliary phase of Gd-EOB-DTPA and reserved liver function by using the model score for end-stage liver disease (MELD). Methods:All patients who underwent magnetic resonance (MR) imaging with Gd-EOB-DTPA were retrospectively collected. The patients with serum creatinine level higher than 1.5 mg/dL or patients without available data to estimate MELD score were excluded. Thirty-six patients were enrolled. A signal-to-noise ratio (SNR) in the liver parenchyma on a fat-suppressed three dimensional fast spoiled-gradient recalled echo sequence images before and 20 minutes after contrast injection were measured and calculated on PACS by two radiologists. The MELD score was determined and interobserver reliability was estimated. Results: Among 36 patients, we found a negative relationship between the percentage enhancement and the MELD score (P < 0.01, r = 0.545). The SNR at 20 minutes after Gd-EOB-DTPA injection also had a negative relationship with the MELD score with statistical significance (P < 0.01, r = 0.460). Interobserver reliability was 0.675. Conclusion: The percentage enhancement in hepatobiliary phase of Gd-EOB-DTPA can predict reserved liver function.