消化内科 · 本周文献汇编

Autoantibodies neutralising interleukin-10 (IL-10) have recently been described as a mechanism of inflammatory bowel disease (IBD); however, prevalence and clinical impacts are unknown. To determine the prevalence and longitudinal persistence of IL-10 autoantibodies in a population of paediatric-onset IBD and to assess associated clinical features. We analysed plasma IL-10 autoantibodies with a cross-sectional study of a paediatric cohort of patients diagnosed with either IBD (n=239), autoimmune enteropathy (n=37) or congenital diarrhoea and enteropathies (n=42). Functional testing to assess IL-10 response was performed using cell-based assays under biologically relevant stimulation. Five patients with autoantibodies neutralising 2.5 ng/mL of IL-10 were identified, three of whom were also neutralising at 5 ng/mL. All had IBD without a known monogenic cause. Longitudinal samples (available in three of five subjects) showed autoantibody persistence in two cases, 3 and 9 years after baseline. The patient without persistence had undergone haematopoietic stem-cell transplantation (HSCT) 2 years after baseline sampling. The prevalence of neutralising IL-10 autoantibodies was 2.1% (95% CI 0.7% to 4.8%), while the frequency was 2.6% in subjects without IBD-associated monogenic variants (n=192). In paediatric autoimmune and inflammatory gut diseases, IL-10 neutralising autoantibodies were detected only among patients with IBD without known genetic aetiology. Patients had heterogeneous clinical characteristics and IL-10 autoantibodies persisted under standard immunosuppressive treatment but disappeared following HSCT and may have declined with colectomy or ileostomy. Screening for IL-10 autoantibodies via functional assays may identify a subgroup of patients with distinct aetiology and therapeutic needs.

Clinical trials in IBD face difficulties of escalating complexity, high costs and challenges in recruitment. Digital twins are virtual, data-driven replicas of individual patients that model disease trajectories and treatment responses, which offer a potential innovative change in the conduct of clinical trials in IBD. Built from multimodal datasets integrating clinical, molecular, imaging and real-world data, digital twins can generate synthetic control arms, enable adaptive randomisation and predict disease relapse or treatment response. Early studies across oncology, cardiology and endocrinology demonstrate their feasibility and potential to improve statistical power while reducing patient burden. However, the integration of digital twins into clinical trials in IBD will require rigorous validation frameworks, transparent data governance and attention to algorithmic bias and consent. In this review, we explore how digital twins may transform IBD research-from in silico simulation to adaptive, patient-centred trial design-and outline the regulatory, ethical and logistical challenges to be considered in order to successfully integrate them into future trials.

IBD is characterised by recurrent flares, but evidence on whether modifiable dietary factors influence flare risk is limited. The PREdiCCt study was designed to examine demographic, clinical and dietary factors associated with disease flare among patients with IBD in self-reported remission. Multicentre, prospective cohort study conducted across 47 UK centres. Patients with Crohn's disease (CD), ulcerative colitis (UC) or IBD unclassified (IBDU) in self-reported remission were prospectively followed up. The baseline diet was assessed using a validated food frequency questionnaire. The primary outcome was time to patient-reported flare (captured by monthly IBD-Control) and objective flare (clinical flare plus C-reactive protein >5 mg/L and/or faecal calprotectin (FC) >250 µg/g with treatment escalation). Associations were evaluated using Cox frailty models adjusted for demographic, clinical and biochemical variables, including baseline FC. Between November 2016 and March 2020, 2629 participants (1370 CD; 1259 UC/IBDU) were enrolled and followed up for a median of 4.1 years (IQR 3.0-5.0). Baseline FC was strongly associated with patient-reported flares (FC ≥250 µg/g: adjusted HR (aHR) 2.22; FC 50-250 µg/g: aHR 1.52 (reference <50 µg/g)) and objective flares (FC ≥250 µg/g: aHR 3.25; FC 50-250 µg/g: aHR 1.98). In UC, higher total meat intake was associated with increased risk of objective flares (highest versus lowest quartile: aHR 1.95, 95% CI 1.07 to 3.56). No consistent associations were observed for ultraprocessed foods, fibre or polyunsaturated fatty acids and flare. Higher habitual meat intake was associated with increased risk of objective flare in UC, suggesting diet may contribute to flare susceptibility in specific patient groups. NCT03282903.

Flagellin, the structural protein of bacterial flagella, activates the innate immune receptor Toll-like receptor 5 (TLR5). However, the ability of different flagellins to bind and stimulate TLR5 varies widely, suggesting that the composition of an individual's flagellin repertoire, defined as flagellome, may influence host-microbiome interactions and inflammation. Here, we developed FlaPro, a computational pipeline for quantification and functional annotation of human gut flagellomes. Functional categories in FlaPro are derived from a machine learning model trained on experimentally characterized flagellins with defined TLR5-binding and stimulatory activities. Application of FlaPro to a multi-omics inflammatory bowel disease (IBD) cohort revealed a marked depletion of flagellome diversity and a reduced ratio of silent to stimulatory flagellins in Crohn's disease and ulcerative colitis. These alterations were consistent across genomic and transcriptional layers, indicating a disease-associated shift toward more stimulatory flagellome profiles. Our findings suggest that specific features of the gut flagellome contribute to TLR5-mediated immune activation and may serve as functionally interpretable microbiome markers for future microbiome-wide association studies in health and disease. The workflow implemented in Snakemake is openly available at https://github.com/leylabmpi/FlaPro.

Toll-like receptor (TLR) signaling is critical for innate immune system. However, whether it is directly modulated by microbiota-derived metabolites remains unclear. Here, we show that the short-chain fatty acids (SCFAs) propionate and butyrate suppress TLR signaling by directly binding p38α MAP kinase, promoting its interaction with TAB1, thereby activating p38α via autophosphorylation. Activated p38α then phosphorylates TRAF3 at serine 85, inhibiting K63-linked polyubiquitylation of TRAF3 and disrupting TBK1-IRF3 activation, leading to reduced macrophage activation and intestinal inflammation. In ulcerative colitis patients, fecal levels of propionate and butyrate positively correlate with p38α activity and TRAF3 S85 phosphorylation, but inversely correlate with TBK1 activation, and cytokine levels. Notably, oral administration of propionate in three patients with ulcerative colitis markedly improved intestinal inflammation and clinical symptoms. These findings reveal p38α as a direct sensor for microbiota-derived SCFAs that suppress TLR signaling through nonmetabolic functions of propionate and butyrate, providing the first clinical evidence that propionate supplementation represents a practical dietary strategy for ulcerative colitis management.

Anti-tumor necrosis factor (TNF) therapy is widely used for inflammatory bowel disease, yet primary non-response and secondary loss of response remain challenges in clinical practice. In this study, we demonstrate that bacterial phosphoketolase improves the primary response to anti-TNF antibodies during induction therapy by enhancing Treg-mediated immunosuppression and maintaining higher serum drug concentrations. Mechanistically, phosphoketolase acts as a microbial host enzyme in macrophages, increasing phosphoketolase pathway flux and lactate production. Elevated lactate induces histone H4K12 lactylation, leading to the upregulation of the serotonin transporter, which mediates serotonin uptake for subsequent conversion to 5-hydroxyindoleacetic acid, a potential inhibitor of TNF-α-converting enzyme. This increases surface transmembrane TNF levels, thereby enhancing TNFR2 signaling in Tregs and promoting their proliferation and differentiation. In a prospective clinical trial, phosphoketolase-producing Bifidobacterium enhanced anti-TNF antibody efficacy during induction therapy. These findings support phosphoketolase-producing probiotics as an effective adjunct to anti-TNF therapy in inflammatory bowel disease.

Intestinal stem cell number or their regeneration ability is crucial for attaining mucosal healing. Deciphering the molecular mechanisms responsible for the impairment of intestinal stem cells in inflammatory bowel disease could yield innovative therapeutic insights. Altered bile acid metabolism is a hallmark feature of inflammatory bowel disease, typically characterized by elevated fecal levels of primary bile acids, such as glycocholic acid. However, the relationship between glycocholic acid and inflammatory bowel disease remains unclear. Here, we report that glycocholic acid accelerates inflammatory bowel disease progression through downregulating TRIB3 expression to disrupt intestinal stem cells self-renewal. TRIB3 is highly expressed in crypt cells and sustains intestinal epithelial stemness by preventing ID1 palmitoylation and AP3D1-mediated lysosomal degradation. Glycocholic acid is found to suppress TRIB3-ID1 axis, thereby compromising intestinal epithelium integrity. Notably, we identify Bergenin, a natural compound, as a potential therapeutic agent against inflammatory bowel disease via upregulating TRIB3. These findings highlight the TRIB3-ID1 axis as a promising therapeutic target for inflammatory bowel disease therapy.

Intestinal barrier is the body's largest immune structure and essential for nutrient absorption. Its dysfunction allows the translocation of pathogenic substances into circulation, thereby driving the pathogenesis of inflammatory bowel disease (IBD). Plant-derived exosome-like nanovesicles (ELNs), recognized for their biocompatibility and ability to traverse biological barriers, hold considerable potential for managing intestinal inflammation. Specially, plants cultivated under extreme environmental conditions typically adapt to be stress resistant with greater accumulation of associated biologics, which may in-turn confer unique bioactivities to their respective ELNs. This study investigated the protective effects and mechanisms of ELNs derived from the extremophyte Rosa roxburghii (R-ELNs) and Artemisia sphaerocephala Krasch (A-ELNs) against intestinal barrier dysfunction. In vitro and in vivo studies indicated that the ELNs, especially R-ELNs, provided enhanced protection against intestinal barrier dysfunction. Specifically, mucus secretion and tight junction protein expression were promoted, and macrophages were polarized toward M2 anti-inflammatory phenotypes. Furthermore, R-ELNs modulated the composition of the intestinal microbiota, thereby promoting a balanced microecological environment. Importantly, the protective effect of R-ELNs was suggested to be through an inhibitory effect on excessive activation of pro-inflammatory signaling proteins (AKT, p38). Notably, exosomes (Exos) derived from R-ELN-treated M2 macrophages had distinct miRNA profiles that can target inflammatory pathway genes (Tgfbr1, Map3k7, Met), enabling anti-inflammatory roles via intercellular communication. These findings suggested that R-ELNs can restore intestinal barrier dysfunction via multiple synergistic mechanisms, positioning R-ELNs as a novel and promising preventive strategy for inflammatory bowel disease.

In inflammatory bowel disease (IBD), complex immune dysregulation induces chronic inflammation. TREM1 is a receptor expressed on myeloid cells and is activated by ligands such as PGLYRP1 to amplify inflammatory responses. Here, we analyze the function of TREM1 signaling in neutrophil-driven inflammation, identifying two positive feedback loops. First, neutrophils activated by phorbol 12-myristate 13-acetate (PMA) or by bacteria release PGLYRP1, which forms a ligand complex with bacterial peptidoglycan and activates TREM1, thereby enhancing neutrophil degranulation, reactive oxygen species production, and NETosis. Second, TREM1 activation on monocytes promotes the secretion of chemokines and cytokines that recruit and activate additional neutrophils, further perpetuating inflammation. In vivo, TREM1 activation exacerbates colitis in mice, with the gut commensal bacterium Blautia faecis potentially contributing to human TREM1 activation in IBD. In biopsy samples from patients with IBD, an elevated TREM1 pathway signature is associated with inflamed tissue and treatment resistance in ulcerative colitis. Our findings thus highlight TREM1 as a regulator of mucosal inflammation, and a potential therapeutic target for IBD.

Adherent-invasive Escherichia coli (AIEC) exhibits proinflammatory properties and has been implicated in the pathogenesis of Crohn's disease (CD), a form of inflammatory bowel disease (IBD). Antibiotic use in CD lacks specificity and may worsen microbiome disruption, prompting interest in bacteriophages (phages) for targeted microbiome editing. Here, we identified HER259, a phage active against clinical AIEC isolates. HER259 ameliorated colitis in gnotobiotic models and attenuated the virulence of AIEC strain NRG857c, including suppression of the FimH adhesin through inversion of the fimS promoter to its "off" orientation. The effects were confirmed in CD-microbiota colitis models. Withdrawal of HER259 treatment led to reversion of the fimS promoter and reactivated colitis. The HER259 phage also enhanced the therapeutic effect of subtherapeutic budesonide independent of microbial drug metabolism. These findings support targeted phage therapy as an adjunct treatment approach in IBD, demonstrating modulation of bacterial virulence and improved response to conventional treatments that may reduce drug-related side effects.

Inflammatory bowel disease is shaped by complex microbial communities, yet the contribution of fungal-bacterial interactions to disease progression remains poorly defined. Here, we identify Cladosporium tenuissimum (C. tenuissimum) as a gut fungus with potent colitis-alleviating activity. Mechanistically, C. tenuissimum restrains Candida albicans (C. albicans) overgrowth through nutrient competition, particularly via the utilization and subsequent limitation of the amino acid ornithine. C. albicans can evade this suppression and potentiate intestinal inflammation through nutrient escape by preferentially exploiting specific amino acids, such as threonine. We further reveal a bacterial-fungal metabolic axis in which threonine-producing Bacteroides fragilis facilitates C. albicans escape from gut microbiome-mediated fungal control, thereby exacerbating colitis. Notably, dietary threonine restriction markedly attenuates C. albicans-driven colitis in mice. Together, our findings uncover a cross-kingdom metabolic network that determines C. albicans homeostasis and, in turn, governs intestinal inflammatory outcomes, offering new conceptual and therapeutic avenues for IBD.

HBV surface antigen (HBsAg) seroconversion indicates the clearance of HBV in patients with chronic HBV infection. We aimed to investigate the predictive value of HBV RNA and HBV mutants on HBsAg seroconversion in patients with chronic HBV infection from childhood to adulthood. We recruited 700 children (409 males and 291 females) with initially hepatitis B e antigen (HBeAg)-positive chronic HBV infection. With the mean initial visit age of 7.28 years (95% CI 7.03 to 7.62 years), they were followed for 15 912 person-years at our institution. Serum samples collected after HBeAg seroconversion in HBeAg seroconverters were analysed for circulating HBV RNA, HBsAg, HBV core-related antigen, HBV DNA levels and the percentage of HBV mutants. 23 subjects (3.29%) experienced HBsAg seroconversion following antiviral therapy, while another 27 (3.86%) achieved spontaneous HBsAg seroconversion in this cohort. The annual probability of HBsAg seroconversion is 0.32% (95% CI 0.30% to 0.33%) per person-year across the entire cohort. After HBeAg seroconversion, the spontaneous annual HBsAg seroconversion rate is 0.47% (95% CI 0.42% to 0.51%) per person-year, and it rises to 1.30% (95% CI 1.10% to 1.51%) per person-year in subjects who had previously received antiviral agents before HBeAg seroconversion. Undetectable circulating HBV RNA and the percentage of A2131C mutants <10% after HBeAg seroconversion are significant predictors of HBsAg seroconversion in both univariate and multivariate survival analyses. In this long-term chronic HBV cohort, we elucidated both the natural course and antiviral-related HBsAg seroconversion. Undetectable circulating HBV RNA and percentage of A2131C mutants <10% after HBeAg seroconversion are novel and independent predictors of HBsAg seroconversion.

Long-term outcomes data of chronic hepatitis B (CHB) patients treated with high-genetic barrier nucleos(t)ide analogues (NA) beyond year-10 are scarce. We assessed the incidence and predictors of such long-term outcomes. The long-term PAGE-B cohort included 1644 Caucasians with CHB treated with entecavir/tenofovir. Cumulative incidence was estimated using Kaplan-Meier or cumulative incidence function accounting for competing events. Incidence rates (IR) per 100 person-years (/100 PYs) were estimated. Of 1644 patients, 903 were followed beyond year-10 (mean:14±2 years). The 10- and 15-year cumulative incidence of hepatocellular carcinoma (HCC) was 10.9% and 13.2%, respectively, and the HCC IR was 1.25 before and 0.55/100 PYs after year-10 (P<0.001), with similar findings after inverse probability weighting. The IR of death or liver transplantation (LT) was relatively lower before than after year-10 (1.50 vs 1.95/100 PYs, P=0.069) and similar for liver-related death/LT in the two periods (0.74 vs 0.70/100 PYs, P=0.840). HCC development and baseline platelets were independently associated with LT-free liver-related or overall survival, which was also associated with older age and diabetes. The 10- and 15-year cumulative incidence of HBsAg loss on NA(s) was 8.3% and 14.3%, respectively; the IR of HBsAg loss was 0.94 before year-10 increasing to 1.42/100 PYs after year-10 (P=0.025). HBsAg loss was independently associated with older age and baseline HBeAg-positive. NA therapy was discontinued in 125 (7.6%) patients remaining HBsAg-positive. Despite >10 years of NA therapy, CHB patients remain at risk for HCC, but the incidence rate declines significantly. HCC remains the main determinant of mortality. HBsAg loss rate increases after year-10, but it remains low being higher in older, initially HBeAg-positive patients.

PD-1 blockade has emerged as a promising approach for functional cure of chronic hepatitis B (CHB). This study aimed to evaluate the safety profile of anti-PD-1 antibody (αPD-1), as well as its impact on hepatitis B surface antigen (HBsAg) and immune responses in a larger cohort of CHB patients. In this prospective, open-label study, virally suppressed patients with CHB on nucleos(t)ide analogue (NA) were assigned to receive either 24-week NA monotherapy (n=62) or αPD-1 (half-dose sintilimab) add-on therapy (n=59), with both groups subsequently receiving NA monotherapy for an additional 12-week observation period. 93.6% of adverse events (AEs) were grade 1 or 2. Elevations in alanine aminotransferase (ALT) and aspartate aminotransferase were the most common AEs, and they represented the only severe AEs. αPD-1 add-on therapy induced greater HBsAg reductions (mean decline: -0.720 vs -0.034 log10 IU/mL, p<0.001) and higher HBsAg loss rates (6.1% vs 0%, p=0.166) than NA monotherapy at week 24. Notably, significant HBsAg decline and seroclearance exclusively occurred in the initial 12 weeks of αPD-1 treatment. HBsAg levels did not rebound at 12 weeks after discontinuation of αPD-1 therapy. After αPD-1 therapy, the numbers of HBsAg-specific, HBpol-specific, HBx-specific and HBeAg/HBcAg-specific IFN-γ spots all increased (p<0.05), while frequencies of HBsAg-specific B cells remained stable. Furthermore, ALT elevation and enhanced HBsAg-specific T-cell responses following αPD-1 therapy correlated with HBsAg decline (p<0.05). In virally suppressed CHB patients on NA therapy, 24-week half-dose sintilimab treatment demonstrated a favourable safety profile. This regimen can facilitate HBsAg reduction and even HBsAg loss, while concurrently enhancing HBV-specific T-cell responses. These findings support αPD-1 as a potential therapeutic alternative for CHB. NCT05769816.

Withdrawal of prolonged nucleos(t)ide analogue (NA) treatment results in hepatitis B surface antigen (HBsAg) loss in some subjects with chronic hepatitis B (CHB), potentially revealing immune correlates of functional cure. We investigated whether baseline or longitudinal changes in humoral immunity correlated with outcome of discontinuing prolonged NA treatment. Global memory B cells (MBC) and T follicular helper cells (Tfh) were analysed by flow cytometry. HBs (small surface)/HBc (core)-MBC were quantified by ex-vivo bait staining and function assessed by cultured ELISpots (enzyme-linked immunosorbent spots). Immune parameters assessed at end-of-treatment (EOT), 12 and 48 weeks after treatment withdrawal (and at 4-8 years in a subset) were correlated with intrahepatic and longitudinal serum viral markers and alanine transaminase (ALT). Individuals on prolonged NA had comparable frequencies of HBc-MBC and HBs-MBC, although the latter were PD-1hi and functionally defective. Following treatment withdrawal, increases in class-switched HBc-MBC were frequently temporally linked with hepatic flares. Subjects achieving HBsAg loss had an increase in activated global MBC detectable at EOT that become more marked by week 48, accompanied by significant increases in plasmablasts. HBs-MBC in those with HBsAg loss showed significant reductions in PD-1, trends to increased activation (CD71) and function and a more robust correlation with Tfh, compared with HBsAg persistence. MBC changes were maintained 4-8 years after HBsAg seroconversion. Differences in global and HBs-specific B cell immunity associate with HBsAg loss, whereas HBc-MBC temporally associates with flares, following withdrawal of prolonged NA treatment. Our results underscore the need to further explore the potential of B cell targets for monitoring and enhancing HBV functional cure in larger cohorts.

Hepatitis B virus (HBV) infects hepatocytes by specific binding to the cell-surface receptor-sodium taurocholate cotransporting polypeptide (NTCP)-through the preS1 region of its large envelope protein, followed by a less well-understood transport process across the cytoplasm to the nucleus. Here, we report that scavenger receptor class F member 2 (SCARF2), a single-pass transmembrane protein, functions as an intracellular receptor for HBV. SCARF2 binds to a preS1 region downstream of the NTCP binding site through its N-terminal epidermal growth factor (EGF)-like domains 4-6, and its proline-rich C-terminal domain also plays an indispensable role in the infection. The internalized HBV virions are transported to the cytoplasmic side of nuclear pore complexes within the SCARF2-containing endosomes. HBV nucleocapsid release from the endosomal vesicles is impaired by knockdown of SCARF2. These results suggest a model in which SCARF2 conveys HBV to the periphery of nuclear pore complexes (NPCs) and ultimately leads to viral nucleocapsid release for nuclear entry.

Autoimmune hepatitis (AIH) is characterised by death of hepatocytes and chronic inflammation. Patients with deleterious variants in A20 are identified with AIH presentations and immune activation. However, it remains unclear how A20 dysregulation contributes to AIH pathogenesis. This study elucidates the role and mechanism of A20 in AIH progression and its therapeutic potential. A20 expression was analysed in patient sample and AIH mouse model. A20 function was explored in mice with hepatocyte-specific A20 knockout and overactivation and wild type controls. Liver samples were assessed by histology, immunoblot and electron microscopy. Proteomics, mass spectrometry, co-immunoprecipitation and site-specific mutation experiments were used to elucidate underlying mechanisms. AdipoRon was used to identify the effects of pharmacological induction of A20 on AIH. A20 expression is reduced in livers of individuals with AIH. In experimental AIH, hepatocyte-specific A20 depletion aggravates hepatic inflammation and fibrosis, enhances ferroptosis of hepatocytes and increases sensitivity to ferroptosis inhibitors, while A20 overexpression ameliorates AIH pathology and reduces hepatocyte ferroptosis. Mechanistically, A20 is found to interact with KEAP1 and mediate its K48-linked polyubiquitination degradation, promoting NRF2 accumulation and nuclear translocation to enhance antioxidant and antiferroptosis effects. Pharmacological induction of A20 confers protective effects in AIH. The clinical correlation among A20, KEAP1, NRF2 and ferroptosis is validated in human AIH samples. Our findings elucidate a previously unrecognised A20-KEAP1-NRF2 axis that orchestrates hepatocyte ferroptosis in AIH. Targeting A20 may provide a promising therapeutic strategy for AIH.

Hepatitis E virus (HEV), a leading cause of viral hepatitis globally, is associated with adverse outcomes during pregnancy. Despite its clinical significance, the mechanisms driving enhanced HEV replication during pregnancy remain poorly understood. In this study, we uncover a lipid-mediated pathway that facilitates HEV replication, with potential implications for pregnancy-associated pathogenesis. Lipidomic profiling reveals a marked upregulation of oleic acid during HEV infection in both human liver cells and in HEV-3ra-infected pregnant rabbits. We showed that oleic acid significantly enhances HEV replication, possibly through interaction with a predicted fatty acid-binding domain (FABD) located within the papain-like cysteine protease region of the HEV ORF1 polyprotein. We further demonstrated that phosphatidylethanolamine (PE) levels are elevated during HEV-3ra infection in pregnant rabbits, and that inhibition of PE biosynthesis by CRISPR/Cas9-mediated knockdown and silencing of phosphatidylserine decarboxylase/phosphoethanolamine cytidylyltransferase genes that are responsible for PE synthesis resulted in decreased HEV replication, indicating that PE is important for HEV replication. Additionally, we found that placental lactogen hormone, which is elevated during late pregnancy, stimulated fatty acid accumulation and potentiated HEV replication, therefore providing a potential explanation for pregnancy-associated adverse outcomes. Collectively, our findings reveal an important role of host lipid metabolism in HEV replication and offer mechanistic insights into lipid-dependent enhancement of HEV replication and a potential role of lipid reprogramming in HEV pathogenesis. The results may inform potential future anti-HEV therapeutic strategies targeting lipid pathways.

Clostridioides difficile exemplifies a pathogen that leverages its metabolic plasticity to exploit nutrients that become available during community disruption, including host and microbiota-derived metabolites and substrates enriched in modern diets. These ecological dynamics underpin the high and growing burden of C. difficile infection (CDI), including recurrent disease and the rising prevalence of community-associated CDI. Fecal microbiota transplantation and standardized stool-derived products consistently re-establish colonization resistance through convergent functions that include secondary bile acid restoration, nutrient niche exclusion, and suppression of opportunistic pathogens. These principles have provided a valuable roadmap for rational consortia design. In this review, we synthesize current ecological mechanisms governing C. difficile colonization, persistence, and recurrence, highlight missing dimensions in diet intervention studies and mucosal colonization by C. difficile, and propose an ecology-informed, artificial intelligence-enabled precision framework that integrates host susceptibility, exposures, diet, community function, and pathogen features to guide personalized prevention and treatment.

The human gastrointestinal tract is colonized by trillions of microbes, collectively termed the gut microbiota, which dynamically shape states of health and disease. This occurs through the modulation of host immunity and through metabolites produced by these microbes, which act both locally and at distal organs. Seminal research has defined that the gut microbiota are essential for successful cancer immunotherapy, with specific microbes and associated metabolites linked to therapeutic outcomes. The mechanisms accounting for this remain preliminary, with additional research implicating diet in shaping the composition and functional potential of the gut microbiota to steer host immunity toward fighting or supporting malignant tumors. The goal of this review is to summarize recent advances in a key communication loop between diet, the gut microbiota, and host immunity as it relates to immune checkpoint blockade in cancer. Further, we discuss gaps in knowledge and future opportunities to harness this knowledge to improve therapeutic strategies.

The progression of metabolic dysfunction-associated steatohepatitis (MASH) involves complex crosstalk among hepatocellular injury, inflammation, and ductular reaction-a process characterized by biliary epithelial proliferation and fibrotic activation. Integrin αvβ6, an epithelial-specific receptor largely absent in normal liver but induced under chronic injury, plays a pivotal role in activating latent TGF-β and driving fibrogenesis. Here, we investigated αvβ6 as a molecular imaging biomarker in MASH using a PET-based strategy to non-invasively visualize epithelial fibrotic signaling in vivo. PET/CT imaging and autoradiography studies were performed in MASH mice at different disease stages using [68Ga] DOTA-R01-MG, a radiotracer targeting integrin αvβ6. Tracer uptake was correlated with histopathological changes, immunofluorescent localization of αvβ6, and transcriptional markers of fibrosis, inflammation, and biliary expansion. Human liver biopsy specimens from MASH patients were analyzed as preliminary ex vivo evidence to explore the translational relevance of αvβ6 expression and tracer binding in human fibrotic liver tissue. In both CDAHFD and MCDD models, hepatic [68Ga] DOTA-R01-MG uptake increased progressively with MASH-associated fibrosis. αvβ6 signals were mainly localized to periportal and ductular regions, where they co-localized with expanding CK19+ biliary epithelial cells. Increased tracer uptake and αvβ6 expression were associated with collagen deposition and transcriptional activation of fibrogenic markers, including TGF-β1, α-SMA/Acta2, Col1a1, Vim, and Timp1. In human MASH liver tissue sections, αvβ6 expression and autoradiographic signals also increased with fibrosis stage, supporting the translational relevance of αvβ6-targeted imaging for fibrotic remodeling. This study identifies integrin αvβ6-targeted PET as a potential molecular imaging strategy to map epithelial remodeling in MASH. Ex vivo analyses of human MASH specimens support the translational relevance of this target, although prospective clinical PET studies and longitudinal outcome validation are required before clinical diagnostic application.

Endoscopic submucosal dissection (ESD) enables en bloc resection of early gastrointestinal cancers and provides specimens suitable for precise pathological risk assessment. However, reporting remains variable for key parameters that determine curative resection and the need for additional treatment, including submucosal invasion depth and breadth, margin status, lymphovascular invasion, tumour budding, differentiation and use of ancillary stains. To develop practical international standards for pathology assessment and reporting of invasive carcinoma in ESD specimens. An international panel of 42 experts, including 28 gastrointestinal pathologists and 14 therapeutic endoscopists from 15 countries, participated in a modified Delphi consensus process. Statements addressed measurement of invasion, margin assessment, staining, specimen handling, prognostic histological features and clinically relevant reporting. 56 recommendations reached consensus across seven domains. The panel recommends using Sm1-Sm3 subclassification only when the muscularis propria is present; otherwise, submucosal invasion depth should be reported in micrometres, rounded to the nearest 100 µm. Submucosal invasion breadth should be reported in millimetres as an adjunct metric for future validation. Margin positivity should be defined as direct tumour contact with the inked surface, supported by standardised pinning, inking, complete embedding and parallel sectioning. H&E remains the baseline stain, with selective immunohistochemistry or elastic stains for equivocal lymphovascular invasion, distorted architecture or difficult margin interpretation. Tumour budding should be reported according to International Tumour Budding Consensus Conference criteria, and differentiation, histological subtype, lymphovascular invasion, perineural invasion and margin status should be integrated into composite risk assessment. These consensus standards provide immediately implementable, synoptic-ready pathology reporting criteria after ESD. By standardising measurement landmarks, margin terminology, ancillary stain use and reporting of adverse histological features, they aim to reduce interinstitutional variability, improve multidisciplinary decision-making and support future validation of risk models in early gastrointestinal cancer.

Clinically significant portal hypertension (CSPH) drives decompensation and mortality in advanced chronic liver disease (ACLD). Although non-selective β-blockers (NSBB) reduce risk, accurate identification of patients with CSPH requires invasive hepatic venous pressure gradient (HVPG) measurement. The non-invasive Baveno-VII CSPH criteria based on liver stiffness measurement (LSM) and platelet count (PLT)-yield 40-50% indeterminate ("gray-zone") results and vary across etiologies and elastography techniques. Spleen stiffness measurement (SSM) has been proposed to improve the accuracy of the Baveno-VII CSPH criteria. We developed and validated a machine-learning (ML) model integrating pan-elastographic LSM and SSM results with clinical variables to improve CSPH rule-out and rule-in accuracy while minimizing indeterminate cases. We analyzed 1,435 compensated ACLD patients with paired HVPG, LSM, SSM, and clinical parameters. LSM and SSM were obtained by vibration-controlled transient elastography (VCTE), two-dimensional shear-wave elastography (2D-SWE), or point-SWE (p-SWE). Models were trained (n=943) and internally validated (n=150) using harmonized LSM/SSM from different technologies and clinical variables (PLT, Child-Pugh, age, gender, etiology). Cut-offs were selected for 100% negative predictive value (NPV) to rule-out and 100% positive predictive value (PPV) to rule-in CSPH. External validation was conducted in 342 patients across seven centers, comparing ML performance against Baveno VII, Baveno-SSM single- and dual-cut-off criteria, and, in the VCTE subgroup, ANTICIPATE and NICER scores. A Random Forest-based model based achieved the highest performance (external validation: AUC=0.91, Brier Score=0.13), with cut-offs≤0.45 (rule-out) and ≥0.60 (rule-in) yielding NPV=0.90 (95%C.I.0.84-0.94) and PPV=0.96 (95%C.I.0.92-0.98). The ML gray-zone was 12.3%, versus 47.9% (Baveno VII;p<0.001), 38.6% (Baveno-SSM-dual;p<0.001), and 19.6% (Baveno-SSM-single;p<0.05). In the VCTE external-validation subgroup (n=275), ELM achieved comparable rule-in performance to ANTICIPATE and NICER, while reducing the gray zone to 12.0% versus 41.1% and 41.8%, respectively. The ELM Score outperformed current Baveno criteria and markedly reduced gray zones across elastography modalities, supporting broader, safer, and non-invasive identification of ACLD patients with HVPG-defined CSPH who may be candidates for NSBB therapy.

Primary sclerosing cholangitis (PSC) is a chronic, progressing cholestatic disease that often co-occurs with inflammatory bowel disease (PSC-IBD). PSC-IBD affecting the colon (PSC-ulcerative colitis or PSC-UC) resembles clinical UC, but is characterised by less severe disease flares, right-colon predominance, and a greater lifetime risk of colorectal cancer than UC alone. To elucidate differences in the underlying biology between PSC-UC and UC, here we combine single-cell mRNA and antigen receptor sequencing, 16S ribosomal RNA gene analysis and spatial transcriptomics on biopsies from four colon regions of patients with PSC-UC and UC during endoscopic remission or at the time of relapse. We show that the PSC-UC colon, compared to healthy control (HC) or UC colon, harbours distinct and region-specific mucosal-adherent microbial communities and an enrichment of activated CD8 T and γδ T cells at the right colon, even in the absence of histological inflammation. By contrast, a TMEM176B+ mast cell population that may be pro-tumourigenic is enriched in the colon during disease relapse in both PSC-UC and UC colon. These results highlight that the PSC-UC and UC colonic mucosa are fundamentally different while sharing similar cell programmes during active disease. Our data thus provide insights to guide tailored clinical management and precision therapies.

KRASG12D is the predominant oncogenic driver in pancreatic ductal adenocarcinoma (PDAC). While most investigational KRAS-G12D inhibitors are oral small molecules limited by gastrointestinal toxicities and suboptimal tumor exposure, HRS-4642 is a new, high‑affinity, noncovalent KRAS-G12D inhibitor. Formulated as a liposomal nanoparticle for intravenous administration, it is designed to enhance tumor accumulation and prolong the duration of target inhibition. This phase 1b/2 study evaluated HRS-4642 in combination with nab-paclitaxel and gemcitabine (AG) in patients with advanced KRASG12D-mutant PDAC. As of 5 December 2025, 68 patients were screened and 31 (1 previously treated patient and 30 treatment-naive patients) were enrolled and treated. In the phase 1b portion, no dose-limiting toxicities were observed, and the starting dose (500 mg on day 1 and 1,200 mg on day 8, every 3 weeks) was selected as the recommended phase 2 dose. In the phase 2 portion, with a median follow-up of 12.3 months (95% confidence interval (CI) = 12.2-13.0), the primary endpoint was met-the confirmed objective response rate in 30 treatment-naive patients was 63.3% (95% CI = 43.9-80.1). Grade ≥3 treatment-related adverse events (TRAEs) occurred in 90.3% patients, primarily hematologic toxicities consistent with AG chemotherapy. No TRAEs led to treatment discontinuation or death. In conclusion, HRS-4642 combined with AG demonstrates promising antitumor activity and a manageable safety profile in advanced KRASG12D-mutant PDAC, supporting further investigation. Clinicaltrials.gov registration: NCT06520488 .

Locoregional therapy (LRT) is frequently used as bridging therapy to transplantation/resection or palliative treatment for hepatocellular carcinoma (HCC). Multiple prediction models have been developed for prognosis and treatment response among patients undergoing LRTs. We aimed to systematically review the methodological quality and performance of clinical risk scores predicting outcomes in patients with HCC treated with LRT. EMBASE and PubMed were searched from inception to 18 March 2026. Our main outcome was the concordance statistic, or area under the receiver operating characteristic curve (AUROC), to predict survival and other tumour-related and liver-related outcomes. 130 studies met the inclusion criteria, resulting in 179 individual scoring systems. The total population was 70 061 patients, 21% female (4-51%), most with Barcelona Clinic Liver Cancer Stage B (39%). Risk scores commonly incorporated tumour parameters, liver function tests, cirrhosis staging and comorbidities. AUROC values ranged from 0.56 to 0.94. 16 studies (12.3%) had low risk of bias, while most had high risk of bias. Of these, the Y-scoring system, Cheng et al nomogram and Li et al nomogram showed the highest discrimination for overall survival in palliative LRT (AUROC >0.87). Separate pooled analyses showed that the Hepatoma Arterial-embolisation Prognostic (HAP) and modified Hepatoma Arterial-embolisation Prognostic II (mHAP-II) scores had the highest performance (pooled AUROC >0.72), while the Six-and-Twelve (0.68) and Albumin-Bilirubin (ALBI) scores (0.60) demonstrated lower performance. Several outcome scoring systems are promising for specific LRTs and populations. However, most models have high risk of bias, underscoring the need for further development and validation.

Patients with advanced biliary tract cancers (aBTC) are in urgent need of additional/new treatment options. We aimed to assess the efficacy and safety of ivonescimab plus chemotherapy in patients with aBTC. In this multicenter, open-label, phase II study, 30 patients with treatment-naive unresectable locally advanced or metastatic BTC received ivonescimab (20 mg/kg or 30 mg/kg) combined with gemcitabine (1000 mg/m2) and cisplatin (25 mg/m2) every 3 weeks for up to eight cycles, followed by ivonescimab maintenance. The primary endpoint was the investigator-assessed objective response rate (ORR) and safety. Pretreatment tumor specimens available from the trial were subjected to a post hoc exploratory proteomic analysis. The correlation between MAP2K7 levels and ivonescimab efficacy was assessed by BTC tumor cell-T cell co-culture and BTC organoids-T cell. At data cutoff, 1 patient achieved complete response and 19 patients achieved partial response yielding an ORR of 66.7% (95% Confidence Interval [CI]: 47.2-82.7). The disease control rate was 100%. The median progression-free survival (mPFS) was 8.5 months (95% CI: 7.6-10.5) and the median overall survival (mOS) was 16.8 months (95% CI: 11.1-22.5). Treatment-related adverse events occurred in 100.0% of patients with the most common being anemia (25, 83.3%), neutrophil count decreased (23, 76.7%), white blood cell count decreased (22, 73.3%), and platelet count decreased (22, 73.3%). No treatment-related deaths occurred. Additionally, exploratory proteomic and functional analyses identified MAP2K7 as a resistance-associated biomarker. MAP2K7 was upregulated in non-responders, and MAP2K7 suppression enhanced ivonescimab-mediated antitumor activity in immune co-culture models. Ivonescimab plus chemotherapy showed potential anti-tumor activity and tolerable safety as first-line treatment of aBTC patients. Exploratory analyses suggest that MAP2K7 may serve as a candidate biomarker of resistance and a potential therapeutic target for optimizing ivonescimab-based therapy of aBTC patients. Currently, the standard of care for first-line therapy in patients with aBTC is the addition of immune checkpoint inhibitors to chemotherapy based on the TOPAZ-1 and KEYNOTE-966 trials. However, this new regimen only improved OS by less than 2 months. The ORR of 66.7%, DCR of 100% and the median overall survival of 16.8 months were observed with ivonescimab plus chemotherapy. This study provides evidence supporting the potential role of ivonescimab plus chemotherapy as a first-line therapy for patients with treatment-naive unresectable locally advanced or metastatic BTC. NCT05214482 and NCT06048289. NCT05214482 and NCT06048289.

TOPAZ-1 (NCT03875235) demonstrated improved overall survival (OS) with durvalumab plus gemcitabine and cisplatin vs. placebo plus gemcitabine and cisplatin with manageable safety in advanced biliary tract cancer (aBTC). The phase IIIb TOURMALINE study (NCT05771480) is evaluating the safety and efficacy of first-line durvalumab plus seven gemcitabine-based chemotherapy regimens in a close-to-real-world aBTC population, including participants with poor prognosis. Participants received 1500 mg IV durvalumab (first infusion: 60 minutes; subsequent infusions: 30 minutes) plus investigator's choice of one of seven gemcitabine-based chemotherapy regimens. The primary endpoint was incidence of Grade 3/4 adverse events (AEs) possibly related to treatment (PRAEs) within 6 months of first durvalumab dose. Secondary endpoints included progression-free survival (PFS), OS, objective response rate (ORR), and safety. As of 17 September 2025, 142 participants (median age 68.0 years) had follow-up ≥6 months and received ≥1 treatment dose; most had metastatic disease (72.5%). Participants received a median of 9 cycles of durvalumab; median (IQR) follow-up was 11.63 (6.87-17.25) months. Overall, 50.7% (95% confidence interval [CI] 42.19-59.19) of participants experienced Grade 3/4 PRAEs within 6 months of durvalumab initiation, with no serious AEs of special interest with an outcome of death. Median PFS was 7.39 months (95% CI 6.74-9.07); median OS was 13.50 months (95% CI 11.24-20.53); ORR was 33.1% (95% CI 25.44-41.48). Safety profiles of durvalumab plus seven gemcitabine-based chemotherapy regimens were manageable, and shorter durvalumab infusion duration (30 minutes vs. 60 minutes) did not increase infusion-related reactions. Efficacy data are indicative of potential clinical activity across treatment arms. These findings demonstrate the feasibility of combining durvalumab with alternative gemcitabine-based regimens in a broad aBTC population. The choice of gemcitabine-based chemotherapy backbones for the treatment of advanced biliary tract cancer (aBTC) varies globally and their use in conjunction with immunotherapy has not been fully investigated. The TOURMALINE phase IIIb study demonstrated manageable safety and efficacy data showed potential clinical activity in a close-to-real-world population of participants with aBTC receiving durvalumab plus one of seven gemcitabine-based chemotherapy regimens. These findings point towards the feasibility of combining durvalumab with different gemcitabine-based chemotherapy regimens, even in participants with less favourable disease characteristics. NCT05771480 TRIAL REGISTRATION: ClinicalTrials.gov: NCT05771480; https://clinicaltrials.gov/study/NCT05771480.

The Oral Rheumatoid Arthritis Trial Surveillance trial identified increased malignancy risk with tofacitinib vs tumour necrosis factor inhibitors (TNFis) in rheumatoid arthritis (RA), leading to class-wide regulatory warnings for Janus kinase inhibitors (JAKis). Comparative cancer risk vs other advanced therapies and across immune-mediated inflammatory diseases (IMIDs) remains uncertain. The objective was to estimate relative and absolute cancer risk associated with JAKi vs other advanced therapies across IMIDs. MEDLINE, Embase, Cochrane Library, and ClinicalTrials.gov were searched from inception to February 2026 for phases II to IV randomised trials and long-term extension studies of licensed advanced therapies in adults with RA, psoriasis (PsO)/psoriatic arthritis (PsA), or inflammatory bowel disease (IBD) reporting malignancy outcomes. Bayesian class-level network meta-analyses using Poisson models estimated incidence rate ratios (RRs) vs standard care (placebo or methotrexate monotherapy). It includes all malignancies including nonmelanoma skin cancer. A total of 305 studies (164,824 participants; 264,100 person-years exposure) were included, comprising 174 studies across IMIDs, 123 in RA, 129 in PsO/PsA, and 54 in IBD. In the combined IMID network, JAKi were associated with higher malignancy risk than TNFi (RR: 1.60; 95% credible interval [CrI]: 1.27-2.02) and standard care (RR: 1.85; 95% CrI: 1.38-2.47). Similar directional findings were observed across RA, PsO/PsA, and IBD networks. Interleukin-6 (IL-6), Interleukin-17 (IL-17), Interleukin-23 (IL-23), Interleukin-12/23 (IL-12/23), Cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), B-lymphocyte antigen CD20 (CD20)-targeting therapies showed risks broadly comparable with TNFi. Absolute excess cancer risk with JAKi compared with TNFi was negligible in standard-risk populations (+0.037 cancers per 1000 person-years exposure [PYE]; ∼1 additional cancer per 27,000 PYE) but meaningful in higher-risk populations (+6.814 cancers per 1000 PYE; ∼1 additional cancer per 147 PYE). Across IMIDs, JAKi were consistently associated with higher malignancy risk relative to TNFi and standard care, extending regulatory warnings beyond RA and emphasising the need for risk-stratified treatment decisions.

In the TOPAZ-1 trial's primary analysis, durvalumab plus gemcitabine and cisplatin (GemCis) showed statistically significant improved overall survival (OS) vs placebo plus GemCis with comparable safety between treatment groups in participants with advanced biliary tract cancer (aBTC). Durvalumab plus GemCis was recently established as the first-line standard of care among patients with aBTC. To evaluate 4-year OS and safety of durvalumab plus GemCis in participants with aBTC. This post hoc analysis of the global, double-blind, placebo-controlled, phase 3 TOPAZ-1 randomized clinical trial included participants 18 years and older with histologically confirmed unresectable, locally advanced, or metastatic biliary tract adenocarcinoma. In the TOPAZ-1 trial, patients were enrolled from April 2019 to December 2020 at 105 sites in 17 countries. Data cutoff was February 28, 2025. Participants received intravenous durvalumab, 1500 mg, or placebo plus gemcitabine, 1000 mg/m2, and cisplatin, 25 mg/m2, on days 1 and 8 every 3 weeks for up to 8 cycles, followed by durvalumab or placebo monotherapy every 4 weeks. OS, duration of treatment exposure, serious adverse events, and adverse events resulting in discontinuation were assessed approximately 48 months after the last participant was randomized. Overall, 685 participants were randomized, with 341 receiving durvalumab plus GemCis (median [range] age, 64 [20-84] years; 172 [50.4%] female) and 344 receiving placebo plus GemCis (median [range] age, 64 [31-85] years; 168 [48.8%] female). Median (range) follow-up in censored participants was 56.9 (1.7-67.2) months for participants who received durvalumab plus GemCis and 50.7 (0.9-62.6) months for participants who received placebo plus GemCis. Median OS was 13.0 (95% CI, 11.6-14.1) months for durvalumab plus GemCis and 11.4 (95% CI, 10.1-12.5) months for placebo plus GemCis (hazard ratio, 0.75; 95% CI, 0.64-0.88); 48-month OS rate was 11.8% vs 4.3%, respectively. The rate of serious adverse events possibly related to treatment was similar between arms (52 of 338 participants [15.4%] in the durvalumab plus GemCis arm vs 59 of 342 participants [17.3%] in the placebo plus GemCis arm). In the durvalumab plus GemCis and placebo plus GemCis arms, 21 of 338 participants (6.2%) and 18 of 342 participants (5.3%), respectively, experienced adverse events leading to study drug discontinuation. In this post hoc analysis of the phase 3 TOPAZ-1 randomized clinical trial, durvalumab plus GemCis demonstrated long-term survival benefit and a clinically manageable safety profile, supporting its use as a first-line treatment for aBTC. ClinicalTrials.gov Identifier: NCT03875235.

To investigate whether detectable C-peptide levels in type 1 diabetes is associated with a lower risk of diabetic ketoacidosis (DKA). We analyzed the Diabetes Control and Complications Trial publicly available data repository for the association between detectable stimulated C-peptide (>0.2 nmol/mL, measured annually) and DKA incidence over an average 6.5-year follow-up. We used crude and adjusted Andersen-Gill models for recurrent DKA events with time-dependent covariates. Of the 1,441 participants (53% male, median age 27 years), 129 (9%) experienced 180 DKA events. Of these events, 179 (99.44%) occurred after C-peptide was ≤0.2 nmol/L in the prior year and only 1 event occurred with C-peptide >0.2 nmol/L. C-peptide >0.2 nmol/L was associated with a DKA hazard ratio of 0.07 (95% CI 0.01-0.48; P = 0.007), consistent across adjusted models. Endogenous insulin production was significantly associated with lower DKA risk, suggesting that treatments preserving insulin production could decrease long-term DKA risk.

Despite significant advances in lipid-lowering therapeutics, residual lipid risk persists in patients with or at risk of atherosclerotic cardiovascular disease (ASCVD), even after optimizing low-density lipoprotein (LDL) cholesterol. Emerging evidence highlights the role of non-LDL cholesterol fractions, such as remnant cholesterol, lipoprotein(a) [Lp(a)], apolipoprotein B (apoB), and non-high-density lipoprotein (HDL) cholesterol, as key contributors to residual ASCVD risk. Remnant cholesterol, Lp(a), and LDL cholesterol represent three independent lipoprotein species causing ASCVD, while apolipoprotein B (apoB) and non-HDL cholesterol integrate the other three variables. Thus, clinically interpreting elevated apoB and non-HDL cholesterol is potentially complicated since remnants, Lp(a), and LDL cause ASCVD by different mechanisms and by varying proportions in different patients. Indeed, recent research into the pathophysiology of lipid-driven atherogenesis and development of ASCVD has revealed novel mechanisms that in turn suggest new therapeutic strategies targeting non-LDL lipid components. Elevated remnant cholesterol jointly with elevated LDL cholesterol contributes to arterial wall cholesterol deposition, plaque development, and ASCVD endpoints. Furthermore, the additional triglyceride content in remnant particles may theoretically promote intimal inflammation and possibly plaque rupture and erosion, independently contributing to atherogenesis and ASCVD. The lipid component and pro-inflammatory properties of Lp(a) could similarly contribute directly to atherosclerotic plaque development and ASCVD. In addition, the homology with plasminogen of the defining apolipoprotein(a) moiety of Lp(a) has long been speculated to confer anti-fibrinolytic and pro-thrombotic properties that could produce more severe ASCVD outcomes independent of atherogenesis. This review explores the evolving understanding of residual lipid risk in ASCVD, practical guidance for clinicians today, recent advances in therapeutic interventions, and their implications for clinical practice, aiming to optimize lipid management beyond LDL cholesterol reduction today and in the future.

Image-based machine learning tools are powerful resources for analyzing medical images, with deep learning-based semantic segmentation commonly utilized to enable the spatial quantification of structures visible in images. However, dataset generation and training of segmentation algorithms requires advanced programming skills and intricate workflows, limiting their accessibility to scientists without prior coding expertise. Here we present the step-by-step instructions to carry out automatic segmentation of medical images guided by a graphical user interface using the CODAvision algorithm. This workflow simplifies the process of semantic segmentation of microanatomical structures by enabling users to train highly customizable deep learning models without extensive coding expertise. The protocol outlines best practices for creating robust training datasets, configuring model parameters and optimizing performance across diverse biomedical image modalities. CODAvision enhances the usability of the CODA algorithm by streamlining parameter configuration, model training and performance evaluation, automatically generating quantitative results and comprehensive reports. We show the use of CODA to serial histology by demonstrating robust performance across numerous medical image modalities and diverse biological questions. We provide sample results in data types, including histology, magnetic resonance imaging and computed tomography. We demonstrate the diverse use of this tool in applications, including quantification of metastatic burden in in vivo models and deconvolution of spot-based spatial transcriptomics datasets. This protocol is designed for researchers with interest in rapid design of highly customizable semantic segmentation algorithms and a basic understanding of programming and anatomy.

The efficacy of pharmacological glycogen synthase kinase-3β (GSK3β) inhibition is currently being investigated in unselected cohorts of metastatic pancreatic ductal adenocarcinoma (PDAC). Here, we sought to determine the clinical significance of nuclear GSK3β accumulation in patients with resectable PDAC. This study aimed to explore the therapeutic potential and underlying mechanisms of GSK3β pathway disruption in PDAC with enriched nuclear GSK3β levels. We investigated the activation and function of GSK3β and its downstream transcription factor NFATc1 in tumour recurrence, growth and resistance using human PDAC tissues, patient-derived organoids and tumour cells, PDAC explants, cell lines and murine models. GSK3β signalling was disrupted using genetic and pharmacological approaches. Live-cell imaging, proliferation, homologous recombination (HR) repair and comet assays, messenger RNA sequencing and chromatin immunoprecipitation were used to explore GSK3β-NFATc1 signalling-mediated target gene regulation in DNA repair, growth and resistance. Nuclear GSK3β accumulates in a subset of resected PDAC and promotes proliferation and DNA repair through NFATc1. The GSK3βhigh/NFATc1high subtype accounts for 14% of resected PDAC and is associated with rapid tumour recurrence and poor survival. The GSK3β-NFATc1 signalling pathway contributes to cisplatin resistance by inducing BRCA genes transcription, which facilitates HR-mediated DNA double-strand breaks (DSBs) repair. Disruption of the GSK3β-NFATc1 axis impairs HR-driven DSB repair, increasing cisplatin sensitivity in vitro and in preclinical PDAC models. We have identified a highly aggressive GSK3βhigh/NFATc1high subtype that predicts early recurrence, poor survival and cisplatin resistance in PDAC. This subtype reveals new treatment vulnerabilities suggesting that patients with PDAC may benefit from stratification-based tailored treatment strategies.

Pancreatic ductal adenocarcinoma has a unique tumor microbiome, and the depletion of gut bacteria or fungi using antibiotic/antifungal cocktails has been shown to decrease pancreatic tumor burden in mice. However, functional studies evaluating the role of tumor-associated microbes are few due to the limited availability of clinically relevant microbiota. Here, we describe in detail an effective workflow for the isolation of bacteria and fungi from the duodenum and tumor of pancreatic cancer patients, specifically optimized for cryopreserved, low biomass samples. Using this workflow we also isolated microbiota from normal pancreatic tissue and duodenum from organ donors, and we confirmed the presence of bacteria and fungi isolated from tissue samples with 16S and ITS sequencing analysis. Isolation and sequencing results show distinct similarities between the pancreatic and duodenal microbiomes and highlight unique bacterial strains that survive in the tumor microenvironment. As a proof of concept, we characterized a select Klebsiella oxytoca strain (UMKO1) isolated from a pancreatic tumor, using whole genome sequencing, metabolomics, and ex- vivo tumor cultures to determine its potential impact on the pancreatic tumor microenvironment. In summary, this optimized workflow allows for the isolation of a variety of bacteria and fungi from low biomass, cryopreserved pancreatic and duodenal tissues, which can then be used for functional studies characterizing clinically relevant tumor-associated microbiota.

Nature killer (NK) cell plays a critical role in cancer immunosurveillance and is considered a potent immunotherapeutic tool for many cancers, including pancreatic ductal adenocarcinomas (PDACs). Increasing evidence suggests that cancer occurs with systemic immune perturbations. However, the effects of PDAC tumor burden on systemic NK cells remain poorly understood. PDAC tumor-bearing mice display decreased frequency and dysfunction of NKs the spleens. We identified an increase in Gr-1+ myeloid cells within the spleens, which negatively impacts both endogenous and adoptively transferred NK cell frequency and function. Apolipoprotein E (ApoE), a lipid metabolism regulator, is upregulated in Gr-1+ myeloid cells of tumor-bearing mice, promoting lipid oxidation and reactive oxygen species (ROS) generation. Genetic knockout of Apoe in Gr-1+ myeloid cells abrogate their suppressive effects on NK cell function. Furthermore, treatment with lipid metabolism inhibitors restores endogenous and adoptively transferred NKs' effector function in the spleens and tumor microenvironment. These studies underscore the importance of understanding preexisting systemic alterations in PDAC patients before applying NK cell-based immunotherapies.

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive malignancy characterised by remarkable cellular heterogeneity, which emerges early from the interplay of oncogenic KRAS signalling and inflammatory injury. However, the transcriptional, metabolic and functional properties of these pre-malignant cell states that initiate and drive PDAC progression remain elusive. This study aimed to identify and functionally characterise the critical premalignant cell states that arise from this heterogeneity, to define novel biomarkers and targets for early intervention. Public and in-house scRNA-seq data of pancreatic tumour models were analysed to identify key subpopulations in early cellular heterogeneity. Genetic perturbation in KrasG12D-driven models was performed to assess functional impact. Mechanistic studies used TurboID proximity proteomics, epigenetic profiling and metabolic assays. Clinical relevance was validated in human PDAC cohorts. We identified LY6D as a marker of a distinct, gastric-like cell state that emerges early and persists throughout tumourigenesis. The LY6D+ population exhibits conserved stemness and a unique, pan-stage dependency on oxidative phosphorylation (OXPHOS). Genetic ablation of Ly6d specifically impaired the gastric lineage and delayed tumourigenesis, while its overexpression enhanced tumourigenic and metastatic potential. Mechanistically, the glycosylphosphatidylinositol (GPI)-anchored LY6D protein scaffolds a lipid raft-associated kinase network that drives FOSL1-dependent epigenetic-transcriptional reprogramming. In human PDAC, LY6D+ cells harbour stemness and Epithelial-Mesenchymal Transition (EMT) signatures, and high LY6D expression is an independent prognostic marker of poor survival. Our work defines the LY6D+ gastric-like cell state as a key driver linking early pre-malignant heterogeneity to PDAC initiation and progression. LY6D represents a pan-stage therapeutic target and a candidate biomarker for early detection and therapeutic targeting.

The biological significance of forkhead box A1 (FOXA1) in non-steroid-driven malignancies, such as pancreatic ductal adenocarcinoma (PDAC), has garnered increasing recognition. It assumes a pivotal role in regulating critical processes such as PDAC cell lineage, metabolism, and metastasis. However, its regulatory mechanisms remain elusive. Here, we demonstrate that histone deacetylase 5 (HDAC5) mediates the deacetylation of FOXA1 at lysine residue 270 (K270), leading to repression of FOXA1's global chromatin occupancy. In HDAC5-loss PDAC, K270 hyper-acetylated FOXA1 is reprogrammed to the transcription start sites (TSSs) of HIF1α-targeted genes, functioning as a pioneer factor of HIF1α signaling. Additionally, we show that HDAC5 antagonizes LSD1-mediated FOXA1 activation by converging on the dynamic equilibrium of acetylation-methylation transition at K270. Pharmacological inhibition of HIF1α/LSD1 suppresses the growth and progression of HDAC5-deficient PDAC in in vivo and in vitro models. Our study reveals the role of FOXA1 as a pioneer factor of HIF1α in PDAC, providing potential therapeutic strategies for HDAC5-deficient PDAC.

Pancreatic ductal adenocarcinoma (PDAC) exhibits profound therapy resistance driven by lysosome-dependent nutrient recycling, metabolic adaptation, and stress tolerance. Current lysosome targeting agents such as chloroquine (CQ)/hydroxychloroquine (HCQ) show limited efficacy due to transient activity and dose-limiting-toxicities. To overcome these limitations, we developed lysostilbenes, a new class of hybrid small molecules combining the CQ pharmacophore with lysosome-disrupting stilbene analogs. Stilbene pharmacophore is the core structural component of resveratrol. Among the synthesized hybrids, lysostilbene-4 emerged as the lead candidate, demonstrating ~30-40-fold greater cytotoxicity against PDAC cells than parent compounds, while sparing nonmalignant cells. At nanomolar concentrations, lysostilbene-4 induced rapid, irreversible lysosomal membrane permeabilization (LMP), initiating a lysosome mitochondria apoptotic cascade via CTSB (cathepsin B) release, BID cleavage, BAX activation, and caspase-mediated apoptosis. In parallel, it abrogated lysosomal recovery by significantly reducing repair, lysophagy, autophagosome maturation, and uncoupling TFEB-driven transcriptional programs from effective lysosome biogenesis. Reduced TFEB mRNA expression correlated with poor overall-survival and disease-free-survival across multiple cancer patients, with a particularly strong association in pancreatic cancer patients. Using TFEB+/+ and TFEB-/- knockout pancreatic cancer cells we establish that lysostilbene-4 exerts severe cytotoxicity by inducing persistent lysosomal-damage and disrupting autophagosome-lysosome assembly, with vulnerability further amplified in TFEB-deficient cells. This finding underscores TFEB as a key determinant of lysosomal-resilience and a potential predictive biomarker. Importantly, lysostilbene-4 was well tolerated in preclinical mouse-models at supra-therapeutic doses without systemic-toxicity. These findings position lysostilbene-4 as a first-in-class lysosome-targeting therapeutic that enforces sustained lysosomal collapse while compromising adaptive recovery-mechanisms, providing a mechanistically precise and safe strategy against PDAC.Abbreviations: ALG: autophagy-lysosome genes; AMPK: AMP-activated protein kinase; CASM: conjugation of ATG8s to single membranes; CTSB: cathepsin B; LGALS3: galectin 3; LMP: lysosomal membrane permeabilization; LS: lysostilbene; MTOR: mechanistic target of rapamycin kinase; PDAC: pancreatic ductal adenocarcinoma; TCGA: The Cancer Genome Atlas; TFEB: transcription factor EB; ULK1: unc-51 like autophagy activating kinase 1.

Pancreatic ductal adenocarcinoma (PDAC) remains one of the deadliest cancers due to late diagnosis, early metastasis and resistance to therapy. Cancer stem cells (CSCs) have been implicated in PDAC aggressiveness and treatment failure. MAP kinase-interacting kinase 1 (MNK1) is overexpressed in PDAC and plays a critical role in tumor progression and CSC maintenance. Here, we show the potential of apMNKQ2, a DNA aptamer targeting MNK1, to therapeutically target CSCs and reduce PDAC tumor burden in patient-derived xenografts (PDXs). PDX cell lines and in vivo mouse models were used to assess the effects of apMNKQ2 on cell viability, apoptosis, cell cycle progression, migration, epithelial-to-mesenchymal transition (EMT) and CSC properties. Functional CSC targeting was validated through clonogenic and self-renewal assays as well as extreme limiting dilution analysis. Systemic administration of free apMNKQ2 was tested for biodistribution, pharmacokinetics, toxicity, and antitumor efficacy at escalating doses. apMNKQ2 downregulated MNK1 and anti-apoptotic proteins (MCL1, XIAP), impaired cell proliferation, induced apoptosis, and disrupted cell cycle progression in PDX PDAC cells. Importantly, apMNKQ2 also inhibited migration, mesenchymal properties and angiogenesis in vitro, and lung colonization in vivo. Notably, apMNKQ2 strongly targeted PDAC CSCs, reducing CD24, CD133, CXCR4 and ALDH expression, clonogenicity and in vivo tumor initiation over 600-fold. Free apMNKQ2 (without transfection agents) entered PDAC cells efficiently, retaining anti-CSC activity. Systemic delivery of free apMNKQ2 accumulated in tumors, was well tolerated up to 400 mg/kg and showed no toxicity. Importantly, 10 mg/kg of apMNKQ2 produced strong antitumor effects in PDX models. Increasing the dose 20-fold enhanced tumor uptake but not efficacy, suggesting a therapeutic plateau at 10 mg/kg. MNK1 plays a central role in PDAC progression and CSC maintenance. apMNKQ2 is a potent anti-MNK1 DNA aptamer with robust preclinical activity, including CSC-targeting and anti-invasive effects. Its low toxicity, systemic bioavailability, and efficacy at low doses support further development as a novel therapeutic strategy for PDAC.

Western diet and associated production of secondary bile acids (BAs) have been linked to the development of sporadic colorectal cancer (CRC). Despite observational studies showing that secondary BAs produced by 7α-dehydroxylating (7αDH+) gut bacteria are increased in CRC, a causal proof of their tumour-promoting effects is lacking. Investigate the causal role of BAs produced by 7αDH+ gut bacteria in CRC. We performed feeding studies in a porcine model of CRC combined with multi-omics analyses and gnotobiotic mouse models colonised with 7αDH+ bacteria or a genetically modified strain to demonstrate causality. Western diet exacerbated the CRC phenotype in APC 1311/+ pigs. This was accompanied by increased levels of the secondary BA deoxycholic acid (DCA) and higher colonic epithelial cell proliferation. The latter was counteracted by the BA-scavenging drug colestyramine. Metagenomic analysis across multiple human cohorts revealed higher occurrence of bai (BA inducible) operons from Clostridium scindens and close relatives in faeces of patients with CRC. Addition of these specific 7αDH+ bacteria (C. scindens/Extibacter muris) to defined communities of gut bacteria led to DCA production and increased colon tumour burden in mouse models of chemically or genetically induced CRC. A mutant strain of Faecalicatena contorta lacking 7αDH caused fewer colonic tumours in azoxymethane/dextran sodium sulfate treated mice and triggered less epithelial cell proliferation in human colon organoids compared with wild-type F. contorta. This work provides functional evidence for the causal role of secondary BAs produced by gut bacteria through 7αDH in CRC under adverse dietary conditions, opening avenues for future preventive strategies.

Spatial transcriptomics (ST) assays are transforming our understanding of tumor heterogeneity, but their high cost limits their application in large-scale biomarker discovery. Here, we present "Path2Space," a deep-learning model that predicts spatial gene expression directly from histopathology slides. Trained on extensive breast cancer ST data, Path2Space robustly predicts the spatial expression of thousands of genes, outperforming 21 established methods. Charting the tumor microenvironment (TME) of 976 breast cancer TCGA (The Cancer Genome Atlas) tumors, it accurately infers cell-type abundances and identifies three spatially defined breast cancer subgroups with distinct survival outcomes. Notably, the derived low-cost spatial TME landscapes enable more accurate predictions of patient response to chemotherapy and trastuzumab compared with costly conventional bulk-sequencing-based biomarkers. Path2Space thus offers a scalable, fast, and cost-effective alternative to molecular assays. It opens avenues for large cohort treatment biomarker discovery and translationally relevant insights into tumor biology, with potential applicability across many cancer indications.

Non-small-cell lung cancer (NSCLC), the predominant type of lung cancer, is characterized by high invasiveness and significant mortality. Despite its clinical impact, the molecular mechanisms driving its pathogenesis and progression remain poorly understood. This study demonstrates that TMED9 is overexpressed in NSCLC and showed using multiple independent sample sets that its expression level is significantly associated with poor patient prognosis. Gain- and loss-of-function experiments revealed that TMED9 promotes proliferation, invasion, and migration of NSCLC cells in vitro and significantly accelerates tumor growth and metastasis in vivo. Mechanistically, TMED9 interacts with ATG9A and recruits USP5 to facilitate the deubiquitination and stabilization of ATG9A, thereby activating autophagy and driving malignant progression. Notably, genetic depletion of TMED9 enhances the sensitivity of NSCLC cells to osimertinib. Collectively, these findings identify the TMED9-USP5-ATG9A signaling axis as a critical driver of NSCLC malignancy, highlighting TMED9 as a promising therapeutic target.

WNK2 is a known tumor suppressor in a set of solid tumors, including glioblastoma multiforme and pancreatic ductal adenocarcinoma. WNK2 functions are largely unknown beyond the kinase domain (KD)-dependent inhibition of ERK1/2 signaling. Waldenström's Macroglobulinemia (WM) is an indolent, yet incurable, B cell lymphoma characterized by highly recurrent MYD88 (MYD88MUT) and CXCR4 (CXCR4MUT) mutations that trigger sustained NF-κB and ERK1/2 signaling. Although not expressed in healthy B cells, WNK2 is a top dysregulated gene in MYD88MUT WM. To study WNK2 regulation and signaling, we performed multi-omics analyses, including bulk and PacBio Iso-Seq RNA-Seq and methylome, in 264 untreated WM and functional studies in cell lines and primary WM cells. Aberrant expression of WNK2 emerged as a near universal feature of early-stage WM and a hallmark of plasma cell-like MYD88MUT WM. We identified novel isoforms that carried a shared aberrant splicing event, either contained or lacked the KD and were highly expressed by the tumor cells, unlike the canonical full-length isoforms. Functionally, WNK2/S-NK1, the most expressed of the KD-lacking isoforms in WM, triggered a pro-inflammatory cascade that activated ERK1/2 and NF-κB signaling. We observed a similar, cancer-specific upregulation of WNK2 in a set of solid tumors, including the highly aggressive cholangiocarcinoma. Our findings reveal an undocumented oncogenic function for WNK2 driven by novel, cancer-specific isoforms and provide a framework for its further investigation as a determinant of disease progression in MYD88MUT WM and a novel therapeutic target in hematological and solid tumor oncology.

Gastric acid-suppressive medications, particularly proton pump inhibitors (PPIs), are commonly used in patients with alcohol-associated liver disease (ALD) to prevent and manage upper gastrointestinal bleeding, gastroesophageal reflux disease, and non-steroidal anti-inflammatory/aspirin-induced gastroesophageal damage. By inhibiting the gastric H⁺/K⁺-ATPase, PPIs suppress acid secretion and impair bacterial killing, thereby promoting gut dysbiosis that disrupts barrier integrity and enhances bacterial translocation, ultimately exacerbating liver injury. PPIs are frequently co-administered with antibiotics for indications such as gastrointestinal bleeding, Spontaneous Bacterial Peritonitis (SBP), other infections, or hepatic encephalopathy prophylaxis, but the consequences of this combined therapy on gut microbial ecology and disease outcomes remain unclear. Our study addresses this gap by showing how PPI use, alone or with antibiotics, reshapes the gut microbiome and aggravates liver disease progression. In previous studies, we showed that PPIs promote dysbiosis and ALD progression in mice and humans by facilitating intestinal expansion and hepatic translocation of Gram-positive Enterococcus. Fecal cytolysin, an Enterococcus faecalis exotoxin that induces hepatocyte death, predicts mortality in patients with alcohol-associated hepatitis (AH). In this study, we have examined the mechanism by which PPIs alone and in combination with non-absorbable antibiotics targeting Gram-positive bacteria influence ALD, as well as the disease mechanisms associated with cytolytic Enterococcus faecalis and the development of therapeutic strategies. In mice, alcohol administration during gastric acid suppression promoted expansion of Gram-positive taxa, including cytolysin-producing Enterococcus. Similarly, PPI use in patients with AH was associated with increased fecal Enterococcus and higher 30-d mortality, underscoring the translational relevance of our findings. Unexpectedly, treatment of acid-suppressed mice with non-absorbable antibiotics designed to suppress Gram-positive bacteria worsened ethanol-induced steatohepatitis: while Enterococcus abundance decreased, Streptococcus and other potentially pathogenic taxa expanded, leading to increased bacterial translocation and aggravated liver injury. In patients with cirrhosis or metabolic dysfunction-associated steatotic liver disease (MASLD), PPIs did not promote Enterococcus expansion, indicating etiology-dependent microbiome responses. Finally, we identified dipalmitoylphosphatidylcholine and Caspase-1 inhibitor as in vitro and in vivo modulators of cytolysin activity, highlighting potential therapeutic avenues. Collectively, our study demonstrates how PPIs and non-absorbable antibiotics targeting Gram-positive bacteria interact with the gut microbiome to drive ALD, underscoring the need for careful therapeutic management.

Epithelial-to-mesenchymal transition (EMT) is a dynamic process during which cells lose their epithelial characteristics and acquire mesenchymal traits. In cancer, EMT is closely associated with tumor initiation, progression, invasion, metastasis, and therapy resistance. Rather than being a binary state switch, EMT encompasses a spectrum of tumor states with distinct functional properties. However, the transcription factors (TFs) that govern transitions between these EMT states remain poorly defined. Here, using multi-omic approaches combining single-cell RNA-seq and single-cell ATAC-seq, we delineate the transcriptomic and chromatin landscapes of distinct EMT states in a mouse model of skin squamous cell carcinoma (SCC). Through CRISPR/Cas9-mediated loss-of-function studies coupled with in vitro and in vivo functional assays, we identify TFs regulating specific EMT states. Klf5 and Pitx1 control the early stages of EMT and are essential for metastasis formation. In contrast, Nfatc1 and Creb3l1 act at later stages of EMT. Similar EMT states and regulatory patterns are found in mouse pancreatic adenocarcinoma and human cancers. Altogether, our study defines the transcriptional and chromatin landscape controlling EMT progression in mouse skin SCC, identifies EMT state-specific TFs and highlights their essential roles in regulating metastasis.

Diabetes mellitus remains a global health challenge, as current insulin-based therapies merely control blood glucose without restoring islet β cell function, leaving patients dependent on lifelong medication and vulnerable to hypoglycemia. Antioxidant drugs hold promise for islet β cell repair but are limited by poor gastrointestinal transportation and insufficient pancreatic targetability. Here, we initially identify ferroptosis-associated oxidative stress as a key cause of islet β cell death and develop six stiffness-gradient nanoparticles by embedding bent oleic acid into ordered 1,2-distearoyl-sn-glycero-3-phosphoethanolamine to enhance drug bioavailability. Nanoparticles with intermediate stiffness optimize membrane wrapping and minimize energetic cost, enhancing intestinal M cell transcytosis and macrophage-mediated hitchhiking, thereby increasing pancreatic curcumin accumulation by ~4.5-fold. In diabetic models, this formulation suppresses ferroptosis-associated oxidative stress, promotes in situ islet β cell repair, and restores insulin homeostasis and autonomous glycemic control, maintaining normoglycemia without hypoglycemia even after treatment cessation. This study represents a patient-friendly oral nanotherapy that outperforms insulin therapy in long-standing diabetes management.

Stem cell-derived glucagon-(α) and insulin-producing (β) cells allow to engineer in vitro biomimetics of islet of Langerhans, the micro-organ controlling glycemia; however, a knowledge gap in the mechanism by which human stem cell-derived α and β cells are specified persists. Mouse studies postulated that Aristaless Related homeobox (Arx) and Paired box 4 (Pax4) transcription factors cross-inhibit each other in endocrine progenitors to promote α/β fate allocation, respectively. To test this model in human, we combine lineage labelling with single-cell multiomic analysis in our newly generated ARXCFP/CFP; PAX4mCherry/mCherry knock-in induced pluripotent stem cell reporter line. Lineage tracing, proteomic and gene regulatory network analysis and potency assays reveal a human specific regulation of α/β cell fate allocation. Pharmacological perturbations previously proposed to trigger α-to-β transdifferentiation or identified by our gene regulatory network lead to enhanced endocrine induction and directed α/β cell fate. Studying mechanisms of endocrinogenesis and fate segregation enables the engineering of islets in vitro, and has broader implications for cell-replacement therapy, disease modelling and drug screening.

The pancreatic tissue is composite and plastic. The epigenetic makeup of different cell types is poorly characterized due to the lack of protocols that enable efficient recovery of epithelial cells. Here, we improve the canonical pipeline for single-cell analysis, optimizing the harvesting of nuclei from hard-to-dissociate tissues. We provide an in-depth mapping of the murine pancreas through paired RNA/ATAC multiomic single-nucleus sequencing, documenting the transcriptomic and chromatin accessibility profiles of every cell in the parenchyma. This enables a superior examination of the exocrine fraction of the pancreas. We described endocrine-exocrine interaction in silico and validated pro-proliferative signals in vitro. At the same time, we assessed pancreatic cellular heterogeneity in a holistic manner. We showed that pancreatic acinar cells can acquire multiple phenotypic states that lead to the functional diversification of pancreatic acini. In particular, we identified ultra-specialized ZG16HIGH cells in vivo and associated them with superior protein synthesis by tracking the incorporation of a synthetic amino acid. Leveraging the pairing of gene expression and chromatin opening, we studied the epigenetic elements that dictate acinar cell specialization and how they are affected by tissue repair after inflammation. We found that a subset of acinar cells shows enhanced sensitivity to inflammatory cues that determine extensive and persistent chromatin opening. The dataset is an open-source framework to interrogate the molecular histology of the murine pancreas.

A key feature of the intestinal immune system is balancing pathogen defense with antigen-specific tolerance to commensal bacteria. Here, using conditional deletion models, we identified the transcription factor BHLHE40 as a central regulator of group 3 innate lymphoid cell (ILC3)- and RORγt⁺ antigen-presenting cell (APC)-dependent mucosal immunity. In ILC3s, BHLHE40 drove transcription of cytokine effector programs and maintenance of mucosal immunity. Cytokine TL1A stimulation and inflammation induced Bhlhe40 expression, amplifying these programs through epigenetic modulation of chromatin accessibility at effector loci. Bhlhe40 was also highly expressed in RORγt⁺ APCs, where it was required for the generation of antigen-specific Tregs. In parallel, BHLHE40 integrated microbial cues to promote expression of the co-stimulatory molecule OX40L by ILC3s, further promoting antigen-specific Treg induction. Together, these findings define Bhlhe40 as a coordinated regulator of barrier immunity and support a model in which ILC3s and RORγt⁺ APCs act in concert to shape antigen-specific intestinal immunity.

The liver is known to play a pivotal role in modulating blood glucose homeostasis through intrahepatic glucose metabolism. Here, we reveal a unique mechanism by which fatty liver exacerbates hyperglycemia through remote communication from hepatocytes to intestinal stem cells (ISCs), independent of enhanced intrahepatic gluconeogenesis. Mechanistically, hepatocyte-derived alkaline phosphatase (ALP) targets α2δ-1 in ISCs to promote the membrane translocation of Cav1.2. This process triggers increased intracellular calcium levels, which subsequently activates the calcineurin/NFATC2 signaling axis, thereby inhibiting SOX21 expression. Then, decreased expression of SOX21 downregulated bone morphogenetic protein 7 (BMP7), ultimately hindering ISCs differentiation into intestinal L-cells. Consequently, the levels of hypoglycemic enteroendocrine hormones secreted by L-cells are decreased, thereby promoting hyperglycemia. Therapeutically, inhibiting ALP synthesis in fatty liver independently reduces blood glucose and synergistically enhances the hypoglycemic effect of metformin. Our study highlights the role of liver-gut communication in regulating the fate of ISC differentiation and blood glucose homeostasis.

Chronic stress (CS) exacerbates anxiety and hyperglycemia, emerging as a key risk factor for type 2 diabetes, yet the mechanism remains unclear. Here, we found that CS induces hyperglycemia and enhanced amygdaloid astrocytic senescence in mice. The amygdaloid astrocytic senescence was mediated by the reduction of hexokinase 2 (HK2) driven by pre-B cell leukemia homeobox transcription factor 1 (PBX1). The astrocytic Hk2 deletion mice and amygdala-specific astrocytic Hk2 knockdown mice both display anxiety-like behaviors and hyperglycemia. The reduction of HK2 in astrocytes reduces L-serine synthesis and decreases the supply to neurons for the generation of D-serine by disrupting the astrocyte-neuron serine shuttle. Reduced neuronal D-serine level in the amygdala impaired the balance of sympathetic and parasympathetic amygdala-pancreas projections, leading to hyperglycemia. L-serine supplementation or dasatinib/quercetin administration to eliminate senescent cells alleviates both CS-induced neurobehaviors and peripheral hyperglycemia. Together, these findings reveal that HK2 in amygdaloid astrocytes mediates CS-induced neurobehaviors and hyperglycemia.

Viral proteases are critical targets in antiviral drug development due to their essential role in the viral lifecycle and high conservation across viral strains. This review summarizes the recent progress of antiviral drugs based on protease structures, with a focus on the structural characteristics and inhibitor design strategies targeting Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), Enterovirus 71 (EV71), Hepatitis C Virus (HCV) and Human Immunodeficiency Virus (HIV). Based on the three-dimensional structure of proteases, various types of protease inhibitors such as peptidomimetic, non-peptidic, and natural product-derived inhibitors are discussed. Given the critical challenge of drug resistance in antiviral therapies, we address the mechanisms of protease inhibitor resistance and explore advanced strategies, including allosteric inhibition, dual-target inhibition, and Proteolysis-Targeting Chimeras (PROTAC) to overcome drug resistance. This review also explores the cutting-edge applications of artificial intelligence (AI) for viral protease-based antiviral drug discovery, providing valuable insights for the rational design and development of antiviral drugs.

Drugs targeting genes supported by human genetic evidence are more likely to succeed in clinical trials. While previous approaches have benchmarked individual methods such as genome-wide association studies (GWAS), rare variant burden testing, and quantitative trait locus (QTL)-informed Mendelian randomization, it remains unclear how best to integrate these signals for drug target discovery. We compared gene-prioritization strategies across 30 complex traits, evaluating their ability to recover approved drug targets compiled into lenient and moderate gold-standard sets from six curated databases. Gene-level association scores from GWAS, expression QTL, protein QTL, and exome-based analyses were integrated using five unsupervised approaches. Predictive performance was assessed with area under the receiver operating characteristic curve (AUROC) and enrichment-based statistics. Across traits, GWAS alone ranked known drug targets on average ∼652 ranks (3.42%) above random expectation, and the minimum-rank-based integration strategy further improved performance by approximately ∼558 positions (2.93%), achieving the best AUROC in 23 of 30 traits. Genetic correlation and drug target overlap across trait pairs showed a significant positive association ([Formula: see text]). Cross-trait analyses further revealed that prioritization scores derived from related diseases could at times equal or even surpass a trait's own performance. For instance, coronary artery disease data improved the prediction of stroke targets ([Formula: see text]), while inflammatory bowel disease data enhanced the prioritization of chronic kidney disease targets ([Formula: see text]). These results demonstrate that using the strongest signal from complementary genetic prioritization methods, combined with information from genetically related traits, systematically strengthens drug target identification across complex diseases.

Neuron-tumor communication is emerging as a distinct layer of tumor-host interaction beyond conventional stromal, vascular, and immune regulation. Recent studies show that malignant cells can detect neuronal activity and convert neural signals into growth-promoting cellular responses. In the brain, glioma cells can become electrically integrated into neuronal circuits, where glutamatergic input drives depolarization, calcium influx, and downstream signaling. In extracranial tumors, neural influence appears more heterogeneous, involving spatially organized neurochemical niches, receptor-enriched cancer-nerve contacts ('pseudo-synapses'), autonomic pathways, and injury-associated neuroimmune remodeling. These findings raise important questions about how neural input regulates tumor cell state, metabolism, immune tone, and therapeutic adaptation. This review evaluates the evidence linking neural activity to cancer progression across anatomical contexts and outlines the experimental standards needed to distinguish structured neuron-tumor interfaces from broader neural effects within the tumor microenvironment.

Tumor-seeking fluorescent dyes enable precise lesion localization by recognizing overexpressed receptors, providing a critical adjunctive technology for cancer histopathology. However, tumor heterogeneity and the poor understanding of targeting mechanisms limit their efficacy. Here we engineer CETIF6a, a click chemistry-compatible heptamethine cyanine dye, for multi-cancer targeting, intraoperative histopathology, and proteome-wide target identification. CETIF6a demonstrates margin delineation across multiple cancer types ( > 90% concordance with H&E staining). Quantitative proteomics reveals that the dye targets 5-15 times more tumor-specific proteins than in paracancerous tissues. Synergistic pan-cancer targeting is achieved through 491 conserved tumor-enriched proteins involved in ribosomal, proteasomal, and metabolic pathways, effectively overcoming heterogeneity. Mechanistic studies confirm that CETIF6a emits bright fluorescence upon covalent binding to targets via nucleophilic substitution at cysteine thiol residues within hydrophobic cavities. The modifiable scaffold of CETIF6a supports both intraoperative tumor diagnosis and functional targets profiling, providing a foundation for systematic probe optimization.