Antimitochondrial antibody (AMA)-negative primary biliary cholangitis (PBC) is a rare presentation of PBC that comprises 5%–10% of all PBC patients. The pathogenesis of AMA-negative PBC appears to be similar to that of AMA-positive PBC. AMA-negative PBC presents similarly to AMA-positive PBC, with symptoms of cholestasis, fatigue, and pruritus most commonly reported. Defective bicarbonate production, resulting in acidification of bile and bile acids, has been proposed as the primary mechanism of damage to bile ducts and hepatocytes and is reflected in elevations of alkaline phosphatase and aminotransferases. Chronic damage can lead to the development of cirrhosis. The diagnosis is made by the finding of AMA negativity by ELISA or assays of similar sensitivity and a positive PBC-specific antinuclear antibody (ANA; anti-glycoprotein 210 and anti-speckled 100 kDa protein) test. In cases in which anti-glycoprotein 210 and anti-speckled 100 kDa protein assays are also negative, a liver biopsy is required to make the diagnosis after exclusion of other causes of cholestasis by magnetic resonance cholangiopancreatography or endoscopic retrograde cholangiopancreatography. Treatment for AMA-negative PBC is the same as for AMA-positive cases, with ursodeoxycholic acid as the first-line treatment. Current treatment is most effective in early stages, where it slows but does not eliminate progression. Risk stratification by validated tools such as the GLOBE and UK-PBC scores remains useful in AMA-negative PBC.

Metabolic dysfunction-associated fatty liver disease (MAFLD) represents a predominant cause of chronic liver disease, underscoring the demand for accessible, non-invasive diagnostic tools. Tongue diagnosis in Traditional Chinese Medicine provides a distinctive perspective on systemic health, though it remains largely subjective. This study aimed to develop an interpretable multimodal deep learning model for MAFLD screening by integrating quantitative tongue image features with routine clinical data.

From 904 screened candidates, 477 subjects (157 healthy, 320 MAFLD) were included and randomly allocated to training, validation, and test sets in an 8:1:1 ratio. All participants underwent standardized tongue imaging (International Commission on Illumination L*a*b color features) and comprehensive clinical evaluation. We constructed a dual-stream deep learning model, combining a ConvNeXt-Tiny network for tongue images and a multilayer perceptron for clinical variables. Feature fusion was achieved via a Dynamic Affine Feature Transformation module, and the model was trained using weighted cross-entropy loss.

MAFLD patients showed significant metabolic abnormalities compared to healthy controls. A progressive decrease in tongue yellowness (b* value) was observed with advancing fibrosis. On an independent test set (n = 48), the multimodal model achieved 97.92% accuracy, Quadratic Weighted Kappa of 0.9538, and 96.88% sensitivity, and 100% specificity, outperforming single-modality and serological models. Interpretability analyses confirmed the model’s focus on clinically relevant tongue regions and key metabolic drivers.

We developed an accurate and interpretable multimodal model that synergizes tongue image features with metabolic indicators for MAFLD screening. This approach presents a promising, low-cost tool potentially well-suited for resource-limited settings.

The long-term clinical outcomes of patients with hepatitis B virus (HBV)-related cirrhosis receiving nucleos(t)ide analog (NA) therapy according to virological response patterns remain inadequately defined. This study aimed to investigate the association between virological response patterns and clinical outcomes in a large, long-term, real-world cohort.

This retrospective–prospective cohort study enrolled patients with HBV-related cirrhosis receiving NA therapy from 2009 to 2019. According to the serum HBV DNA levels during the initial two years of antiviral treatment, patients were categorized as having a complete (CVR) or partial virological response (PVR). Patients with CVR were further stratified according to their dynamic HBV DNA changes during follow-up into maintained virological response (MVR) or virological breakthrough (VBT) patterns. The primary clinical outcomes included hepatocellular carcinoma (HCC), acute-on-chronic liver failure, and liver-related death. Secondary endpoints included recompensation and progression to decompensation. Cox proportional hazards regression was used to assess the association between virological response patterns and clinical endpoints.

In total, 1,869 patients were enrolled. During a median follow-up of seven years, the MVR, VBT, and PVR rates were 65.4%, 26.5%, and 8.1%, respectively. The cumulative serum hepatitis B surface antigen (HBsAg) clearance rate was 9.8%. Moreover, 34.9% of patients with HBsAg < 100 IU/mL at baseline experienced HBsAg clearance. Compared with patients with VBT and PVR, those with MVR had a lower five- and ten-year cumulative incidence of HCC in both the compensated (five-year: 10.1% vs. 17.0%; ten-year: 14.2% vs. 33.6%; P < 0.001) and decompensated cirrhosis subgroups (five-year: 19.5% vs. 36.7%; ten-year: 25.7% vs. 49.7%; P < 0.001). Similarly, patients with MVR also had a lower cumulative incidence of liver-related death. Additionally, a higher hepatic recompensation rate was observed in patients with MVR than in those with VBT (34.1% vs. 22.5%, P < 0.001). Importantly, patients achieving HBsAg clearance and undetectable serum HBV DNA levels (“functional cure” during ongoing NA therapy) had the lowest five- and ten-year cumulative incidence of HCC (3.9% and 8.7%, respectively).

Patients with long-term MVR exhibited a lower incidence of HCC and liver-related death in both compensated and decompensated HBV-related cirrhosis subgroups, especially those achieving “functional cure.” However, more than 30% of patients experienced PVR or VBT during long-term NA antiviral therapy. These findings highlight the importance of long-term, rigorous monitoring after initial CVR to optimize outcomes and support clinical decision-making.

The optimal management strategy for adults with immune-tolerant (IT) chronic hepatitis B infection remains undefined. This study aimed to investigate the efficacy and predictive factors of a pegylated interferon (Peg-IFN)-based treatment strategy in IT patients with chronic HBV infection.

In this pilot, open-label, prospective study, 286 patients aged 18 to 60 years with IT characteristics were enrolled and allocated to one of three groups. The combination group received Peg-IFN for 48–96 weeks, with tenofovir disoproxil fumarate (TDF) initiated at week 12 and continued through week 96 (n = 103). The monotherapy group received TDF monotherapy alone (n = 125), and the control group was monitored without therapeutic intervention (n = 58).

No patients in the control group met any predefined efficacy endpoints. Intention-to-treat analysis showed that patients in the combination group achieved significantly higher virological response rates (71.8% vs. 53.6%, p = 0.005), hepatitis B e antigen seroconversion rates (15.5% vs. 1.6%, p < 0.001), and hepatitis B surface antigen (HBsAg) loss rates (10.7% vs. 0%, p < 0.001) compared with those in the monotherapy group at week 96. In the combination group, the cumulative rate of HBsAg loss was 5.4% at week 48 and increased to 11.8% by week 96. Independent predictors of achieving either hepatitis B e antigen seroconversion or HBsAg loss were baseline age under 30 years (odds ratio = 0.217, 95% confidence interval: 0.048–0.976, p = 0.046) and a decline in HBsAg level greater than 1 log10 IU/mL by week 24 (odds ratio = 13.976, 95% confidence interval: 2.506–77.932, p = 0.003).

A Peg-IFN-based treatment strategy significantly increases response rates compared with TDF monotherapy or observation in patients with IT characteristics.

Liver transplant rejection significantly affects patient prognosis. Myeloid-derived suppressor cells (MDSCs), known for their potent immunoregulatory functions, represent a promising target for managing liver transplant rejection. This study aimed to systematically characterize the diversity of MDSC subsets and their context-dependent functions, particularly within the context of transplant tolerance.

We analyzed clinical and murine liver transplants using single-cell RNA sequencing, bulk RNA sequencing, flow cytometry, multiplex immunohistochemistry, and co-culture assays to phenotype MDSC subsets.

Single-cell RNA sequencing analysis of human and murine samples revealed MDSC involvement in transplant rejection. In mice, MDSC scores followed a normal distribution during the first week post-transplant and correlated with clinical flow cytometry data at one month. A distinct LDLR+ monocytic MDSC (M-MDSC) subset was identified and confirmed through spatial mapping by multiplex immunohistochemistry. Flow cytometry demonstrated dynamic changes in LDLR+ M-MDSCs across tissues (liver, spleen, peripheral blood, bone marrow, and lymph nodes), with a peak during acute rejection. Co-culture experiments showed that LDLR−/− M-MDSCs exhibited reduced Arg-1/iNOS expression and an impaired capacity to induce inhibitory receptors (TIGIT, PD1, CTLA-4) or suppress effector molecules (GZMB, IFN-γ, IL-2) in CD8+ T cells.

These findings highlight the critical role of MDSCs in liver transplant rejection. LDLR+ M-MDSCs exhibited enhanced immunosuppressive properties, underscoring their potential clinical relevance in mitigating rejection and promoting immune tolerance.

Unlike the traditional staging treatment of tumors, the core of “Green Tumor Treatment” is to divide the treatment of tumors into three stages: Hegemony (directly targeting the cancer focus), Kingship (supporting the body’s vital energy and eliminating pathogenic factors), and Imperialism (improving the internal environment), based on the urgency of the tumor and the patient’s physical condition. This approach guides the clinical treatment of tumors. Its treatment system incorporates all minimally invasive and low-damage treatment methods, combining internal and external treatments, traditional Chinese medicine and Western medicine, as well as local and systemic treatments. It aims to maximize treatment outcomes while ensuring the patient’s quality of life, which is highly consistent with the treatment goals for primary liver cancer. This review aims to explore the integrated Traditional Chinese and Western medicine treatment model for primary liver cancer under the guidance of the Green Tumor Treatment concept.

Cirrhotic cardiomyopathy (CCM) is a significant complication of cirrhosis, but its progression and underlying mechanisms remain incompletely understood. This study aimed to investigate dynamic changes in cardiac function, pathology, inflammation, and mitochondrial damage in a mouse model of CCM, and to compare echocardiographic characteristics in patients with cirrhosis.

Bile duct ligation was performed in male C57BL/6J mice to induce cirrhosis. Longitudinal analyses were conducted over eight weeks. Cardiac function was assessed using serum biomarkers, echocardiography, and electrocardiography. Pathology was examined with hematoxylin and eosin, Masson’s trichrome, Sirius Red, and wheat germ agglutinin staining. Western blotting and immunohistochemistry were used to detect markers of inflammation, fibrosis, apoptosis, and mitochondrial function. Cardiac and liver function markers were also evaluated in patients with cirrhosis.

Mice subjected to bile duct ligation developed progressive cardiac dysfunction, including reduced cardiac output and diastolic dysfunction (end-diastolic interventricular septal thickness, left ventricular internal diameters, stroke volume, and left ventricular end-diastolic volume decreased, whereas ejection fraction and fractional shortening increased), as well as cardiac atrophy. Myocardial apoptosis, inflammation (elevated tumor necrosis factor, interleukin-6, and p65), and fibrosis worsened over time. Mitochondrial injury was characterized by reduced carnitine palmitoyltransferase 1A and peroxisome proliferator-activated receptor alpha, with increased hexokinase 2, pyruvate kinase M2, and lactate dehydrogenase A. In patients with cirrhosis, impaired cardiac function and elevated brain natriuretic peptide levels correlated with total bilirubin.

The progression of CCM is closely associated with cirrhosis severity and appears to be driven by myocardial atrophy, apoptosis, inflammation, fibrosis, and mitochondrial dysfunction.

Alpinia calcarata (A. calcarata) Roscoe (Family: Zingiberaceae) is a rhizomatous perennial herb used in traditional medicine to treat inflammatory conditions. This study aimed to develop a topical emulgel dosage form by incorporating the essential oil of A. calcarata rhizome and to investigate it’s in vitro anti-inflammatory activity. A thin-layer chromatographic fingerprint of the essential oil of A. calcarata rhizome was developed. Then, an emulsion base containing plant oil was formulated and incorporated within a Carbopol gel base. The physical characteristics of this formulation were evaluated subsequently. The anti-inflammatory mechanism of the emulgel was determined by in vitro blood cell membrane stabilization assay and thrombolytic activity assay. The results were statistically analyzed by one-way analysis of variance. The thin-layer chromatographic fingerprint of the test oil demonstrated several bands with unique retention factor values. The formulated herbal emulgel was white, viscous, and homogeneous in appearance. The spreadability was 118 g·cm/M, and the pH of the emulgel was 6.30 at 25°C. The A. calcarata emulgel significantly (p < 0.05) inhibited heat-induced in vitro hemolysis, with the highest activity at a 50 µg/mL dose (87.68 ± 0.35%) compared to the placebo. Furthermore, this activity was found to be dependent on the essential oil concentration (r2 = 0.99) of the emulgel. Therefore, it was concluded that the essential oil of A. calcarata rhizome is an effective active ingredient to be used in a topical emulgel formulation, whereas the diverse phytochemicals present in the essential oil would be the underlying source of its anti-inflammatory activity.

Colorectal polyp detection from endoscopic images is critical for the early diagnosis of colorectal cancer. However, traditional deep learning methods often suffer from limited generalization when deployed across datasets containing different polyp morphologies. This work aimed to investigate whether vision-language foundation models can facilitate zero-shot generalization across multiple polyp datasets without target-domain fine-tuning.

We introduced a zero-shot colorectal polyp detection framework based on Contrastive Language-Image Pretraining (CLIP) to improve cross-dataset detection performance. Key innovations include: (1) a background patch contrastive loss using pseudo-normal tissue patches to teach the model to distinguish normal mucosa from polyps; (2) attribute-enhanced text prompts that incorporate domain-specific descriptors of polyp appearance, improving the model’s semantic generalization to novel polyp morphologies; and (3) an enhanced CLIP visual adapter with per-layer adaptive feature fusion and generalized mean pooling to capture multi-scale features for better polyp localization. During training, we use one annotated colorectal polyp dataset (e.g., CVC-ColonDB) to learn patch-level image-text correspondence. The model is then evaluated in a zero-shot manner on different polyp datasets (CVC-ClinicDB, Kvasir-SEG, and CVC-300), where we evaluate pixel-level anomaly detection performance.

The framework demonstrated robust zero-shot generalization on unseen test cohorts. Without any dataset-specific fine-tuning, the model achieved a mean pixel-level AUROC of 0.94 and a mean average precision of 0.81 across the 12 leave-one-dataset-out zero-shot transfer settings. In the CVC-ColonDB-source benchmark, the model achieved a mean Dice coefficient of 0.84 across CVC-ClinicDB, Kvasir-SEG, and CVC-300. This high level of performance was consistent across datasets with distinct visual characteristics, underscoring the ability of the model to detect diverse polyp morphologies that it had not been explicitly trained to recognize.

Our findings demonstrate that an anomaly-aware vision-language model significantly improves cross-dataset polyp detection generalization without requiring normal images for training. This multimodal strategy may facilitate the robust deployment of artificial intelligence-based colorectal screening systems by enabling reliable detection of diverse polyp morphologies across different clinical settings. Extension to non-polyp colorectal pathologies (e.g., ulcerative colitis and colorectal tumors) remains an important direction for future work, pending the availability of pixel-level annotated datasets for these lesion categories.